POL AND PUBLIC FINANCE REVIEW JUNE 2021 POL AND PUBLIC FINANCE REVIEW JUNE 2021 © 2021 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they rep- resent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowl- edge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “World Bank. 2021. Poland Public Finance Review. © World Bank.” All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover design and typesetting: Wojciech Wolocznik, Cambridge, United Kingdom Interior design: Piotr Ruczynski, London, United Kingdom The cutoff date for the data used in this report was June 2, 2021. Contents ix Acknowledgements x Acronyms xiii Executive Summary xiii Poland rebuilt fiscal buffers in the years prior to the COVID-19 crisis xiv Fiscal consolidation is needed over the medium term, but should not undermine the economic recovery xv Structural tax policy and tax administration reforms can help to mobilize additional resources xvii Tax administration reforms helped to improve compliance xvii Public spending savings potential is limited given important spending needs and budget rigidities xviii EU funds can help relieve budget pressures, and can be used strategically to strengthen resilience while advancing green priorities xxiv Notes 1 CHAPTER 1 Fiscal assessment 2 Fiscal Trends 12 Fiscal Framework 15 Fiscal Policy Cyclicality 18 Public Debt Sustainability 22 Fiscal Consolidation 27 Notes 27 References 30 ANNEX 1.1 Key economic indicators 33 CHAPTER 2 Tax assessment 34 Overview of the Tax System 36 Diagnostics by Instrument 60 Policy Recommendations 64 Conclusions 66 Notes 69 References iii 73 CHAPTER 3 General government expenditure 74 Overview 76 Social protection expenditure 80 Expenditure on public services 82 Growth-enhancing expenditure 84 Performance and efficiency 86 Climate-related expenditure 88 Public employee compensation expenditure 90 Expenditure decentralization 91 Outlook, main challenges, recommendations 96 Notes 97 References 99 ANNEX 3.1 PEP2040 financial framework defined in the state budget in multi- year planning 101 CHAPTER 4 Public Procurement: Saving and Greening 102 Introduction 103 Public procurement market in Poland 104 Green public procurement 105 Strategic Procurement Analysis 108 Purchasing System Review 111 Strategy Package 112 Potential savings scenarios 118 Greening public procurement strategy 119 Policy choices for savings in public procurement 121 Proposed future work 122 Notes 122 References 124 ANNEX 4.1 Methodology 129 CHAPTER 5 Enhancing Public Financial Management and Fiscal Transparency — Progress and Persistent Challenges 130 Introduction 131 The Budget System 132 Public Sector Accounting 133 Fiscal Transparency 134 Drivers of Institutional Reform 136 Reform Options and Recommendations 139 Notes 140 References iv Contents BOXES BOXES BOXES 3 Box 1.1 Poland’s fiscal consolidation in 40 Box 2.1 Equity and poverty analysis of the wake of the Global Financial Crisis current fiscal system and announced tax 2010 – 15 reforms 10 Box 1.2 Poland’s COVID-19 emergency 54 Box 2.2 The EU Emission Trading Scheme package (ETS) 14 Box 1.3 Stability and Growth Pact fiscal 56 Box 2.3 Simulating the Effect of Carbon rules and national fiscal rules in Poland Taxes 17 Box 1.4 Fiscal policy cyclicality assessment FIGURES FIGURES FIGURES 3 Figure 1.1 General government revenues 13 Figure 1.18 Stability and Growth Pact 3 Figure 1.2 General government tax preventive arm procedure revenues 16 Figure 1.19 Output gaps and fiscal 4 Figure 1.3 Contributions to tax-to-GDP ratio impulses by country change by tax type 17 Figure B.1.4.1 Output gap estimates for Poland 4 Figure 1.4 VAT revenues and final consumption 17 Figure B.1.4.3 Estimated output gaps in Poland 5 Figure 1.5 Changes in general government expenditure 17 Figure B.1.4.2 Output gap estimates for the euro area 6 Figure 1.6 Contributions to changes in general government expenditure 17 Figure B.1.4.4 Estimated output gaps in ECA 6 Figure 1.7 Poland GDP growth 18 Figure 1.20 Poland debt relative to EU and 7 Figure 1.8 Poland GDP growth composition ECA averages 7 Figure 1.9 Quarterly GDP path 18 Figure 1.21 Poland debt ratio is below EU 8 Figure 1.10 Lost output relative to average in 2019 pre-COVID 5-year trend 18 Figure 1.22 Debt dynamics 8 Figure 1.11 COVID-19 fiscal policy 19 Figure 1.23 Contributions to changes in responses by country debt-to-GDP ratio 9 Figure 1.12 Fiscal responses and outlook 19 Figure 1.24 Composition of general 10 Figure B1.2.1 Anti-crisis and financial government debt Shields in 2020 19 Figure 1.25 Composition of general 10 Figure B1.2.2 Support to enterprises government debt 11 Figure 1.13 Change in general government 20 Figure 1.26 Fan chart evolution of debt-to- spending in 2020 GDP ratio 11 Figure 1.14 General government balance 20 Figure 1.27 Projected fiscal balance and 11 Figure 1.15 General government primary government debt in Poland balance 20 Figure 1.28 Cyclically adjusted primary 11 Figure 1.16 General government balance balances by government level 21 Figure 1.29 Fiscal sustainability gaps in 12 Figure 1.17 Higher tax and social 2021 contributions helped narrow headline 21 Figure 1.30 Fiscal sustainability gaps in deficits 2021, by condition Contents v 22 Figure 1.31 Output loss two years after 59 Figure 2.18 Analysis of Tax Expenditures fiscal consolidation (2015) 22 Figure 1.32 Total expenditure multipliers 74 Figure 3.1 General government 22 Figure 1.33 Capital expenditure multipliers expenditure by country, 2019 22 Figure 1.34 Fiscal multipliers in Poland 74 Figure 3.2 General government during recessions and expansions expenditure in selected groups of countries 26 Figure 1.35 Fiscal impulse over 2020-22 74 Figure 3.3 Structure of general government expenditure in Poland 26 Figure 1.36 Spending consolidation composition 74 Figure 3.4 Structure of general government expenditure in some countries 34 Figure 2.1 Total Tax Revenue and Selected 75 Figure 3.5 Functional breakdown of general Characteristics of the Tax System government expenditure in 2019 38 Figure 2.2 PIT Rates by Income Tax Base 75 Figure 3.6 Change in the functional (General System) breakdown of general government 41 Figure B2.1.1 Average joint PIT and HIC expenditure, 2015 – 19 burden by PIT income base percentiles: 75 Figure 3.7 Structure of highly rigid current scheme and “New Deal” proposals’ expenses in Poland simulation. 75 Figure 3.8 Rigid expenses in OECD 42 Figure 2.3 Tax Wedges by Wage Level in countries, 2019 OECD Countries 76 Figure 3.9 2020 change in general govern- 43 Figure 2.4 Fiscal Burden by Annual Gross ment expenditure by type: EU countries Income 77 Figure 3.10 Social transfers, 2019 or latest 44 Figure 2.5 Implicit and Effective CIT Rates 77 Figure 3.11 Changes in social protection in Selected EU Countries expenditure 46 Figure 2.6 Public Support to R&D in Poland 77 Figure 3.12 Gini coefficient of equalized and other OECD Countries disposable income 47 Figure 2.7 Foregone Revenue (in Zl) from 77 Figure 3.13 Households at risk of poverty Selected CIT Tax Expenditures and social exclusion 48 Figure 2.8 CIT Rates and CIT Revenue 78 Figure 3.14 Family and children benefits 48 Figure 2.9 Evolution of CIT Gap 79 Figure 3.15 Gross and net position by type of household 49 Figure 2.10 Revenue by Type of Indirect Taxes by country (2019) 79 Figure 3.16 Net position by type of household 50 Figure 2.11 VAT Rate in Selected EU 80 Figure 3.17 Replacement rates in the EU in Countries 2050 51 Figure 2.12 VAT Gap and VAT Revenue 81 Figure 3.18 Satisfaction and confidence 52 Figure 2.13 Excise Tax Revenue across public services in Poland 53 Figure 2.14 Environmental Taxes in Poland 81 Figure 3.19 Satisfaction and confidence and Selected EU Countries across public services: Poland vs. OECD, 54 Figure 2.15 Implicit Tax Rate on Energy 2019 and Energy Intensity in Selected EU 81 Figure 3.20 Self-perceived health, 2019 Countries 81 Figure 3.21 Practicing doctors, 2018 or 55 Figure 2.16 Carbon Price Gap and Carbon latest Intensity in Selected OECD Countries 82 Figure 3.22 General government capital 56 Figure B2.3.1 Simulated Effect of an expenditure Increase in Carbon Taxes Under Alternative 83 Figure 3.23 Local government investment Scenarios expenditure 58 Figure 2.17 Implicit Tax Rates on Property 84 Figure 3.24 Efficiency gap in infrastructure (2017) spending: EU countries vi Contents 84 Figure 3.25 Structure of growth-enhancing 110 Figure 4.11 Distribution of used tender government expenditure selection method by markets 85 Figure 3.26 Efficiency gap in health and 110 Figure 4.12 Market entry rate by CPV education spending: EU countries division 86 Figure 3.27 Performance in economic 111 Figure 4.13 Spending volume by supplier transformation priorities market share category 88 Figure 3.28 General government wage bill 111 Figure 4.14 Percent of contracts bidding 88 Figure 3.29 Age structure in public above estimated price by market administration and business 112 Figure 4.15 Relative price distribution, 89 Figure 3.30 Average monthly wage and trimmed salary: growth 112 Figure 4.16 Total savings associated with 89 Figure 3.31 Average monthly wage and each intervention, 2010 – 20 salary: level 114 Figure 4.17 Impact of buyer’s effectiveness 90 Figure 3.32 Estimated public administra- on relative prices tion wage premium/gap, by age group 114 Figure 4.18 Efficiency in bid evaluation and 90 Figure 3.33 Estimated public award had an impact on relative prices administration wage premium/gap, by education level 115 Figure 4.19 Effects of procurement methods on relative prices 93 Figure 3.34 Projection of expenditure related to public pensions 115 Figure 4.20 Annual spending structure by procurement method 102 Figure 4.1 Poland General Government Procurement 115 Figure 4.21 Seasonality and relative prices 106 Figure 4.2 Distribution of contracts by 116 Figure 4.22 Impact of bidder turnout on procurement category, Poland 2010 – 2020 relative prices 107 Figure 4.3 Distribution of average share by 117 Figure 4.23 Effects of share of SME procurement category, Poland 2010 – 2020 bidders on relative prices 107 Figure 4.4 Spending structure, Poland 117 Figure 4.24 Effects of buyer-level spending 2010 – 2020 concentration on relative prices 107 Figure 4.5 Number of contracts by year, 118 Figure 4.25 Impact of market Poland concentration on relative prices 107 Figure 4.6 Spending volume over time, 118 Figure 4.26 Local suppliers and relative Poland 2010-2020 prices in public procurement 108 Figure 4.7 Supply positioning 131 Figure 5.1 Public Financial Management Cycle 109 Figure 4.8 Average relative price 134 Figure 5.2 Open Budget Survey 109 Figure 4.9 Average relative price, by region 2019 — Fiscal Transparency 110 Figure 4.10 Distribution of average relative price by supplier TABLES TABLES TABLES xix Table ES.1 Policy Recommendations 25 Table 1.4 State-contingent expenditure Summary multipliers capital expenditure 21 Table 1.1 Debt sustainability analysis 30 Table A.1 1 Key economic indicators and 23 Table 1.2 Expenditure multipliers outlook 24 Table 1.3 State-contingent expenditure 31 Table A.1 2 General Government Finances multipliers total expenditure Contents vii 35 Table 2.1 Tax Revenue by Type of Source 106 Table 4.1 Summary of main groups of and Level of Government, 2007 – 2019 explanatory factors used in the analysis 37 Table 2.2 General and Self-Employed PIT 113 Table 4.2 Summary of savings Schemes interventions 40 Table B2.1.1 Poverty among households 120 Table 4.3 Policies and strategies to save with children for disposable income with money on public procurement—Directly and without child tax credits impacted by government policies and 65 Table 2.3 Main Recent Tax Reforms strategies 121 Table 4.4 Policies and strategies to save 85 Table 3.1 Poland’s position among the EU money on public procurement — Indirectly member states by government capacity and impacted by government policies and performance indicator strategies 93 Table 3.2 Age-related general government expenditure 125 Table A4.1.1 Overview of indicators used for price modelling, Poland 94 Table 3.3 Main findings and recommendations viii Contents Acknowledgements This report was produced by a core team comprised of: Cristina Savescu (Task Team Leader), Nicolò Dalvit, Reena Badiani-Magnusson, Andrzej Halesiak, Michal Tulwin, Barbara Ziolkowska, Alberto Leyton, Iwona Warzecha, Mihaly Fazekas, Nora Regos, Hasan Dudu, Jan Witajewski-Baltvilks, Karolina Goraus-Tańska, Collette Mari Wheeler, Sergiy Kasyanenko and Agnieszka Boratyńska; valuable contributions were pro- vided by researchers at the Institute for Structural Research. Excellent research support was also pro- vided by Michal Tulwin, Franz Ullrich Ruch, Damien Matthias Valentin Boucher, Julia Renee Roseman Norfleet, while Leah Laboy, Mismake Galatis and Maria Hazel Macadangdang provided administrative support. Managerial guidance and direction were provided by Gallina Andronova Vincelette (Country Director, EU Member States), Jasmin Chakeri (Practice Manager, Macroeconomics, Trade & Investment), Enrique Blanco Armas (Acting Practice Manager, Macroeconomics, Trade & Investment), Marcus Heinz (Resident Representative, Poland and Baltic Countries), Emilia Skrok (Program Leader) and Reena Badiani- Magnusson (Program Leader). The main authors and contributors to the report were: Executive Summary was produced by Cristina Savescu. Chapter 1 was written by Cristina Savescu and Collette Mari Wheeler with contributions from Sergiy Kasyanenko. Chapter 2 was written by Nicolò Dalvit with contributions from Michal Tulwin (green fiscal instru- ments), Hasan Dudu, and Jan Witajewski-Baltvilks (carbon tax). Chapter 3 was written by Andrzej Halesiak with contributions from Michal Tulwin, Karolina Goraus- Tańska, Irina Capita, and Massimo Mastruzzi (BOOST). Chapter 4 was written by Mihaly Fazekas and Barbara Ziolkowska with contributions from Nora Regos. Chapter 5 was written by Iwona Warzecha and Alberto Leyton. The authors are grateful for invaluable comments provided by peer reviewers: Alexandre Borges de Oliveira (Lead Procurement Specialist), Ewa Korczyc (Assistant to the President), Juan Carlos Serrano-Machorro (Senior Financial Management Specialist), as well as William Steven Clark, Ivailo Izvorski, David Palmer, and Elena Glinskaya. The authors are also grateful to Łukasz Czernicki, Chief Economist at the Ministry of Finance of Poland who has coordinated comments prepared by the various departments at the Ministry of Finance. Special thanks also go to Jasmin Chakeri (Practice Manager in the Macroeconomics, Trade, and Investment Global Practice) for her support and guidance. Finally, the team wishes to thank Gallina Andronova Vincelette (Country Director, EU Member States) of the World Bank for her valuable advice and guidance. Euny Hong edited the report, Wojciech Wolocznik typeset the report, Filip Kochan provided communica- tions support, and Agnieszka Boratyńska supported the team throughout the process. ix Acronyms AP Aspirational Peers (South Korea, Germany, and US) AQLI Air Quality Life Index ARWU Academic Ranking of World Universities BGK Bank Gospodarstwa Krajowego BOS Bank for Environmental Protection CASE Center for Social and Economic Research CEE Central and Eastern Europe CEE10 EU member states from Central and Eastern Europe CenEA Center for Economic Analysis CIT Corporate Income Tax CO2 Carbon Dioxide CSR Country Specific Recommendation EBITDA Earnings Before Interest, Taxes, Depreciation, and Amortization EC European Commission ECA Europe and Central Asia EEA European Environment Agency EMDE Emerging market and developing economies ESA European System of Accounts ETS Emissions Trading System EU European Unition EU27 EU (current members) EU28 EU (current members plus UK) EUPACK European Public Administration Country Knowledge FDI Foreign Direct Investment FUS Social Insurance Fund (Fundusz Ubezpieczeń Społecznych) GDP Gross Domestic Product GFC Global Financial Crisis GFCF Gross Fixed Capital Formation GHG Greenhouse Gas Emission GUS Statistics Poland (Główny Urząd Statystyczny) IBS Institute for Structural Research IFS Institute for Fiscal Studies ILO International Labour Organization IMF International Monetary Fund IP Intellectual Property KAS National Revenue Administration MoED Ministry of Economic Development MoF Ministry of Finance NBP National Bank of Poland NDC Notional Defined Contributions NFOSiGW National Fund for Environmental Protection and Water Management NFZ Polish National Health Fund NRA National Revenue Authority NRP National Reform Program x Poland Public Finance Review NRRP National Recovery and Resilience Program OECD Organisation for Economic Co-operation and Development OFE Open Pension Funds (Otwarte Fundusze Emerytalne) PEP2040 Energy Policy of Poland until 2040 PFIC Passive Foreign Investment Company PIE Polish Economic Institute PISA Program for International Student Assessment PIT Personal Income Tax PLC Tax on Civil Law Transactions PM 2.5 Fine particulate matter R&D Research and Development RDC Research and Development Center RRF Recovery and Resilience Facility SCD Systematic Country Diagnostic SEZ Special Economic Zone SGI Sustainable Governance Indicators SME Small and Medium-sized Enterprise SP Structural Peers (Chile, Czech Republic, Hungary, Romania, Mexico) SPM Split Payment Mechanism SPV Special Purpose Vehicle SSC Social Security Contribution TECM Tax on the Extraction of Certain Minerals U.K. United Kingdom VAT Value Added Tax VET Vocational Education and Training WEF World Economic Forum WHO World Health Organization ZEW Leibniz Centre for European Economic Research Zl Polish Zloty ZUS Social Insurance Institution xi EXECUTIVE SUMMARY Poland’s public finance management was pru- The Poland Finance Review examines the impli- dent in the years preceding the COVID-19 pan- cations of these trends for fiscal management and demic and facilitated a swift, substantial, and recommends measures to help support the green, frontloaded policy response to the crisis. Poland resilient, and inclusive agenda. built fiscal buffers in the years before the COVID-19 pandemic through spending containment and im- Strong institutions are critical to support these proved tax collections, aided by robust growth. This shifts in fiscal policy. High-quality and effective created the fiscal space needed to respond to the institutions are critical to design and implement COVID-19 pandemic, which affected lives and live- fiscal policies in a strategic and accountable man- lihoods and triggered Poland’s first annual output ner. Improved consistency of public policies, com- contraction in 30 years. The fiscal deficit and debt mitment to achieving strategic objectives, coordina- level deteriorated markedly in the wake of the cri- tion across levels of government, and cooperation sis. Even though the pace of fiscal adjustment is ex- with other stakeholders are key to strengthening pected to ensure public debt sustainability, debt is institutions in general, but particularly in relation projected to stabilize at a higher level than expect- to the use of public resources. Enhancing budget ed before the crisis commenced. institutions and an improved tax system are also critical to ensure high-quality public finances that Coming out of the crisis, Poland could make support Poland in achieving its development goals. better use of public finances as a powerful tool The Poland Finance Review examines challenges to foster green, resilient, and inclusive devel- in Public Financial Management and fiscal trans- opment. Once the economic recovery is on a solid parency in Poland, focusing on the budgeting and footing, revenue and expenditure measures will be accounting systems, as well as on fiscal transpar- needed to rebuild fiscal space. As part of this pro- ency practices, and recommends measures to ad- cess, fiscal policy could be reoriented to better pri- dress them. oritize growth-enhancing investments and to sup- port a just transition to a low-emission economy, which will require significant public and private Poland rebuilt fiscal buffers in the resources. The fiscal, economic, and social costs of years prior to the COVID-19 crisis such a transition need to be weighed against the costs incurred in the business-as-usual scenario.1 Poland replenished its fiscal buffers in the years Meanwhile, demographic changes are a source of leading up to the COVID-19 pandemic, but new fiscal pressure in the longer term and will likely social spending initiatives contributed to a dete- strain the health care and pension systems, posing rioration in the structural deficit. Fiscal consoli- challenges for fiscal sustainability and inclusion. dation measures, improved revenue collection, com- xiii bined with robust GDP growth, helped to rebuild omies and the euro area. Moreover, projections for fiscal buffers that had eroded in the wake of the both the primary balance and cyclically-adjusted global financial crisis and the euro area debt cri- primary balance suggest that the planned pace of sis. Sustained efforts to rein in spending and in- fiscal adjustment is swift enough to quickly stabi- creased tax revenues helped to cut Poland’s fiscal lize debt, albeit at higher levels relative to pre-pan- deficit by half from 2010 – 14, with a brief interrup- demic trends. In its Convergence Program for 2021, tion during the European debt crisis on account of the government of Poland expects a decline in pri- weak growth. Robust growth and stronger fiscal in- mary spending over the 2021 – 24 period of close to stitutions, with adherence to the European Union 5 percentage points of GDP. Premature fiscal con- fiscal framework, strengthened fiscal sustainabil- solidation is a near-term risk. ity. In the three years leading up to the COVID-19 pandemic, despite strong revenue mobilization There are several sources of public spending pres- and output growth, Poland breached limits on gov- sure over the medium term, including those that ernment spending as set by the EU preventive arm, relate to environmental sustainability. Meeting on account of new spending initiatives (including the 2030 greenhouse gas emission (GHG) reduction a new child benefit program in 2016), a reduction target requires significant investments to transition in the national pension age at the end of 2017, and to a lower-emission economy, particularly in the en- additional pension benefits. ergy sector, but also transport, buildings, and agri- culture. Large funding gaps remain, however. For Activating the general escape clause of EU fis- Poland, lower emissions require, among others: diver- cal rules provided EU member states, including sifying its energy mix away from coal, while ensur- Poland, with the budgetary flexibility to deal ing an inclusive and just transition; greening trans- with the COVID-19 crisis. Poland’s national stabi- portation to contain fast-rising transport emissions; lizing expenditure rule was also suspended in 2020, and improving energy efficiency, including in the while in 2021 a modified escape clause of the stabi- residential sector. For example, for the 2018 – 30 pe- lizing expenditure rule provided scope for additional riod, the investment gap for the energy transition in spending. The increased flexibility allowed Poland to Poland for at least a 40 percent reduction in emis- accommodate an extraordinary budgetary response sions was estimated at €20 billion out of a total €70 to the COVID-19 crisis over the 2020 – 21 period, in- billion in generation and network investment needs cluding a large fiscal support package. The tempo- (PKEE E&Y Business Advisory 2018). While Poland rary departure from the adjustment path toward stands to receive allocations under the EU budget for the medium-term objective of a structural deficit of Climate and Environment, it remains the case that 1 percent of GDP could create additional consolida- EU ETS allocations, funds from the Modernization tion pressure when the general escape clause is de- Fund, Just Transition Mechanism funds, and nation- activated. Nonetheless, the departure was necessary al co-financing funds will need to be sizeable. The to cushion the impact of the crisis and strengthen EU Recovery and Resilience Facility can help to ac- the recovery, and it does not currently pose a threat celerate the transition toward climate neutrality, as to public finance sustainability in the medium term. at least 37 percent of expenditure is to be earmarked for climate investments and reforms. An accelerated transition to a low-carbon economy could pose risks, Fiscal consolidation is needed over including fiscal risks. The State Treasury might buy the medium term, but should not coal-fired power units from utility companies, in or- undermine the economic recovery der to allow power companies to access financing for investments in natural gas and renewable energy. The fiscal consolidation that is needed to sta- Such stranded assets need to be carefully managed. bilize debt is relatively modest and the policy Meanwhile, severance and early retirement condi- mix should not threaten the economic recov- tions for coal miners will increase liabilities. Such a ery. The size of the adjustment needed in Poland transition also risks diminished energy affordabili- is modest relative to other Central European econ- ty (World Bank Poland SCD 2017). xiv Poland Public Finance Review Demographic changes in Poland over the me- Structural tax policy and tax dium- to long-term represent another source administration reforms can help to of spending pressure. The rollback in the retire- mobilize additional resources ment age could impact the adequacy of future pen- sion benefits, increasing the incidence of minimum To achieve ambitious climate change commit- pension and potentially the associated fiscal costs. ments, a broader package promoting a green While gross public pension expenditure is project- transition could be considered, while support- ed to be broadly stable over the medium term, there ing workers and households in the industries are risks of increased spending pressure. The de- and regions most affected by this transition. mographic component is projected to significant- Decarbonization and climate policies can affect ly impact gross public pension expenditure as the public finances significantly, while alignment with number of beneficiaries relative to that of poten- the new Energy Tax Directive under preparation tial contributors rises (3.7 percentage points of GDP will be critical. Environmental revenue represent- increase over the 2019 – 30 period). The old age de- ed 2.64 percent of GDP in 2019, of which 87.5 per- pendency ratio is projected to increase by nearly cent came from taxes on energy products, includ- 10 percentage points by 2030 to 38.9 percent, as the ing excise duties on energy products, permit fees post-war baby-boom generation continues to reach from the EU Emission Trading System (ETS), and retirement age. This is estimated to be largely offset a wide range of environmental concession and us- by the impact of reform-related factors2 that posi- age fees. The latter are levied primarily on energy tively impact pension expenditures (EC Joint Paper and fuel production activities. Estimates of the ef- on Pensions 2019). Sharp declines in replacement fective tax rate on energy, however, are low relative rates on account of the notional defined contribu- to those in most other EU countries, as the Polish tion (NDC) system and its interactions with the ex- tax system does not perform well in pricing car- pected rise in longevity could pose risks, howev- bon emissions at efficient levels. er. Minimum pension guarantees and other social benefits for pensioners with pension benefits below The government could consider these measures, the minimum pension could result in higher-than among others: estimated old-age expenditures. Furthermore, an adverse macroeconomic structural shock could in- • reversing its decision on the 2016 coal and gas ex- crease the gross public pension expenditure by 0.4 cise exemptions, in which case revenue could be percentage points of GDP by 2030.3 used to finance tax credits for energy efficiency and clean energy production Spending on health is expected to increase over the medium term, in part due to aging but also • introducing a carbon dioxide-related vehicle tax due to efforts to promote resilience in the health care system. The government has committed to in- • broadening the base and increasing the carbon crease public financing of health by more than 1 per- tax to increase effectiveness of carbon-pricing centage point of GDP to 6.2 percent of GDP by 2024, signaling, while lowering social contributions of and an integrated healthcare strategy is critical to en- employers and/or increasing transfers to poor- sure the highest efficiency of this additional spend- er households ing. Legislated wage adjustments for health person- nel and reforms to ensure overall improvement of • phasing out fossil fuel subsidies accessibility of health care services are expected to have direct budgetary impacts (2019 – 23). Health care Poland’s tax system generates less revenue-to-GDP expenditure is expected to increase by 1.4 percent- and direct taxes have a smaller share. Tax revenue, age points of GDP between 2019 and 2030 in the ref- including mandatory Social Security Contributions erence scenario. Long-term care costs are projected (SSC), amounted to 35.2 percent of GDP in 2019 com- to rise by around 0.3 percentage points of GDP each pared to an EU-27 average of 40.1 percent. Personal by 2030 (EC 20214). Income Tax (PIT) and Corporate Income Tax (CIT) Executive Summary xv statutory rates are relatively low, and there are sev- sive PIT scheme, in lieu of the 19 percent flat rate eral special treatments and exemptions applicable to and lump-sum options; a fixed dividend tax cred- certain subgroups of taxpayers. it could compensate for the effects of the double taxation on the profits of incorporated businesses. Overall, the combined PIT and SSC system is one of the least progressive in Europe while impos- • introducing targeted measures to reduce a relatively ing a relatively high tax wedge on low incomes. high labor informality to broaden the PIT tax base Due to various PIT special treatments, the top stat- utory PIT rate applies to less than 5 percent of PIT Tax policy measures aimed at increasing the pro- taxpayers. Furthermore, some special treatments gressivity of the tax system are currently under and exemptions are likely to distort the choices of discussion, but reform design should be budget households and firms and raise some concerns in neutral in light of the mounting pressures on terms of equity. Additionally, part of the poten- public finances described in detail above. The gov- tial PIT tax base remains untaxed and represents ernment of Poland is currently discussing a set of re- an important untapped source of public funding. form proposals (the Polish Deal), including tax pol- It also incentivizes the shift to atypical work ar- icy changes: (1) large increase in the tax allowance; rangements to the detriment of regular labor con- (2) increase in the first PIT rate threshold; and (3) tracts, raising concerns in terms of both horizon- large reduction in deductible health insurance con- tal and vertical equity. tribution (HIC) from PIT tax and from changing the rules of health insurance contribution’s calculation The government could consider the following key for self-employed. The first two would benefit all tax- policy recommendations to address the issues payers, with the largest relative impact on low earn- identified: ers, while the changes related to the HIC would in- crease effective PIT rates. • increasing the progressivity of the general PIT and SSC schemes by introducing a higher basic Special tax treatments and exemptions distort PIT allowance and redesigning its phasing-out taxpayers’ choices and may affect horizontal rate; and financing a reduction in SSCs for low-in- and vertical equity. Besides the foregone revenue, come earners via an increase in SSCs on high-in- some of these preferential treatments raise efficien- come salaried earners and the self-employed cy and equity concerns. Special tax treatments and optional schemes are applicable to PIT, CIT, and a • harmonizing social security and health contri- list of items is subject to reduced VAT rates and VAT bution treatments for standard labor contracts, exemptions. The current CIT system is designed to atypical contracts, and self-employment, accom- provide fiscal support to small firms but features panied by the introduction of an SSC allowance limited incentives to support firms’ growth. While on low incomes the preferential CIT treatment of small firms sup- ports businesses, it can distort firms’ growth deci- • redesigning health contributions to be propor- sions and limit their potential to create jobs. The tional to the PIT tax liability, increasing progres- VAT exemption on financial services can generate sivity and equity important distortions in the relative prices of fi- nancial inputs and services, but there are practi- • removing the optional joint taxation regime, as cal difficulties in measuring the value added of fi- positive tax jointness raises concerns in terms of nancial services. both its efficiency and equity To reduce distortions in the tax system and in- • taxing the income earned by self-employed work- crease horizontal and vertical equity, the gov- ers and capital income under the general progres- ernment could consider the following: xvi Poland Public Finance Review • systematically evaluating tax expenditures to im- Public spending savings potential prove effectiveness of tax policy is limited given important spending needs and budget rigidities • refocusing the system of tax incentives, includ- ing for CIT, in favor of greater direct support A relatively high share of rigid expenditures to investments to sustain firms’ growth and job affects the composition of fiscal consolidations. creation Budget rigidities have increased over the past de- cade by almost 2 percentage points to 75 percent of spending. Tax administration reforms helped to improve compliance Flagship family support government programs, supplementary pensions, and demography-re- Poland has made important advances in increas- lated changes carry a substantial fiscal cost. The ing VAT compliance in recent years, with a posi- Family 500+ program, introduced in 2016, reached a tive impact on CIT compliance as well. Since 2016, cost of 1.8 percent of GDP in 2020. This program is the government has adopted a series of measures in addition to other components of the family sup- to combat VAT evasion, increase VAT revenues, and port programs and benefits. These programs sig- reduce unfair competition from fraudulent busi- nificantly reduce the extreme poverty rate among ness practices. In 2016, the government reduced households with children. Except for family allow- the threshold for cash transactions between firms. ance with supplements, other family support pro- In 2018, it introduced penalties for VAT underes- grams are characterized by limited progressivity. timation together with an optional Split Payment Furthermore, supplementary annual pensions cost Mechanism (SPM) on business-to-business trans- of the first 2 years program of 13th pension was Zl actions. The latter measures became mandatory 22.5 billion (MFLSP information), or 0.4 percent of in November 2019 in sectors that were considered GDP annually. A proposed 14th pension expected to particularly prone to VAT evasion. Furthermore, start by end-2021 could cost an additional 0.37 per- an Online Cash Registry System was introduced in cent of GDP. The design of the old-age pension sys- 2019, and a “white list” containing information on tem based on a notional defined contribution theo- registered VAT taxpayers was introduced. retically ensures long-term stabilization of pension expenditure in relation to GDP, but it is likely come On the tax administration side, the government at the cost of a dramatic drop in the replacement could continue its efforts to fight tax evasion, in- rate. The cumulative effect of demography-related cluding by the following means: changes is expected to increase government spend- ing by about 2 percentage points of GDP by 2030. • contrasting wealth and profits offshoring through enhanced cooperation at the EU level Renewed emphasis on increasing the efficiency, targeting and equity impact of public spending • limiting tax competition within the EU and re- would support fiscal consolidation without en- inforcing cross-border information-sharing ini- dangering the recovery. With regard to that the tiatives government could consider the following: • introducing PFIC rules to tax capital income from • increasing the efficiency of the pro-family pol- offshore funds on a current basis, as well as re- icy instruments quiring residents to report the value of their for- eign portfolio of assets to limit tax avoidance of • increasing the retirement age and the gradual taxation on capital income from foreign assets, alignment of the male and female statutory re- should also be considered by the authorities. tirement ages Executive Summary xvii • introducing longevity adjustment of retirement • buying a few items in bulk, which is the most age promising strategy to save money in procurement, especially for high-volume, low-complexity items • protecting growth-enhancing expenditures during fiscal consolidation • introducing measures to ensure fewer monopo- lies and oligopolies at the market and buyer level • implementing an integrated health care strategy, allocating resources to primary care, prevention, • increasing the success rate of procurement pro- and e-health services cesses • capping the mandatory seniority premium • improving the efficiency of bid evaluation and contract award • limiting automatic raises in salary with tenure Public procurement is a powerful tool to achieve EU funds can help relieve budget the economic, social development, and envi- pressures, and can be used ronmental goals and its efficiency is critical to strategically to strengthen resilience achieving these goals. While green public procure- ment (GPP) may have started out as an “alternative” while advancing green priorities procurement approach, it is now recognized as an Poland is expected to receive €170.7 billion from essential element of modern procurement systems. the EU over the 2021 – 2027 period, representing A strategic procurement analysis unveiled oppor- 3.7 percent of the projected GDP, most in form tunities to save an estimated 4.9 percent of annual of grants. The three main programs under which public procurement spending through measures that EU funds are expected to be made available are the do not require new laws or modifications to existing multi-annual financial framework (MFF) 2021 – 27 of laws and regulations. Rather, the measures reflect the Cohesion Policy (CP) (€75 billion), the Recovery a combination of policies and strategies that could and Resilience Facility (RRF) (Poland could access generate savings in public procurement. up to €58.1 billion, of which €23.9 billion in grants), and the Agricultural Policy (AP) (€32.2 billion). The For an effective green procurement strategy, EU funds can be used to finance government policies the government could consider incorporating and investments. The RRF will finance the reforms good practices along the six pillars included in and investment programs included in the National World Bank’s Green Public Procurement Pillars Recovery and Resilience Plan (NRRP). This extraor- and Good Practices Framework. The GPP pil- dinary one-off financing provides an opportunity to lars and good practices framework proposed by carry out significant reforms and investment pro- the World Bank is built around six pillars that al- grams without putting pressure on government fi- low for a customized, non-linear yet balanced ap- nances since EU funds are budget neutral. The RRF proach: 1) building the business case; 2) enabling is designed to frontload disbursements to facilitate the framework; 3) market engagement; 4) profes- the economic recovery from the COVID-19 crisis. sionalization; 5) implementation tools; and 6) mon- itoring tools. Poland’s National Recovery and Resilience Plan (NRRP), the basis for receiving RRF funds, is com- Furthermore, the government could consider the prised of both structural reforms and invest- following measures to achieve savings: ment plans. It is comprised of five components: 1) resilience and competitiveness of the economy; 2) • increasing competition to help drive down rela- green energy and reducing energy consumption; 2) tive prices paid in public procurement, both in the digital transition; 4) efficiency, availability and qual- form of the procedure used and bidder turnout ity of health care; and 5) green, intelligent mobili- xviii Poland Public Finance Review ty. The plan’s main objectives are to restore the de- ery plan after the crisis. The budget system reforms velopment potential of the economy lost because of for a modern program budgeting system (including the COVID-19 pandemic; to support the building of ICT solutions) and increased fiscal transparency are sustainable competitiveness of the economy; and needed to: ensure the long-term stability of the public to increase the standards of living in the long-term. finance system; increase the efficiency and transpar- ency of public assets/spending; and manage current The NRRP components seek to address identi- and anticipated fiscal pressures due to population ag- fied development challenges that relate to the ing, social spending, and the green and digital agenda. six pillars of the EU RRF instrument. These chal- lenges include productivity and the ability to create The government could consider the following high-quality jobs in the conditions of a transforming measures, among others: economy; unfavorable demographic trends and the supply of labor resources; investment climate and • further aligning Polish public sector Generally Ac- level of private investments; independence from cepted Accounting Principles (GAAP) with Interna- coal and transforming key sectors of the economy tional Public Sector Accounting Standards (IPSAS) to a low-emission model; digital transformation of the economy; insufficient quality and limited access • simplifying and standardizing Polish public sec- to health services as well as the ability of the health tor accounting regulations across all public sec- care system to quickly respond to epidemic threats; tor entities the condition of infrastructure, structure and safe- ty of transport serving a competitive, green econo- • developing aggregated financial information to my and intelligent mobility; concentration of devel- gain a better understanding of the government’s opment and climate issues, loss of growth potential overall financial position and fiscal risks than that and low resistance to crisis phenomena in the spatial reported by the current consolidated cash-based dimension; and ensuring sustainable public financ- budget execution es. To implement the NRRP by August 2026, Poland plans to spend the entire amount of available grants • supplementing the current cash-based budget (€23,858 billion) and apply for €12.12 billion in prefer- execution reports with accruals-based informa- ential loans to be allocated primarily to projects re- tion, including assets and liabilities lated to the climate transformation and digitization. • revising to broaden the coverage and definition Strengthening fiscal transparency and continued of public sector entities modernization of Public Financial Management is critical for implementing the country’s recov- • introducing an independent fiscal council Table ES.1 Policy Recommendations Summary Expected Expected Expected Expected fiscal output social environmental Recommendations Timing impact impact impact impact Supporting Fiscal Sustainability Increase retirement age, gradually; align MT + + Uncertain (bet- Neutral male and female statutory retirement ter econom- ages; and ultimately introduce longevity ad- ic situation in justment of retirement age. retirement, but other as- pects are also important, e.g., people preferences) Executive Summary xix Expected Expected Expected Expected fiscal output social environmental Recommendations Timing impact impact impact impact The establishment and institutionaliza- MT + + + Neutral tion of an independent fiscal council could help to achieve greater transparency and accountability. Enact structural fiscal reforms to support the MT Uncertain. Uncertain. Uncertain. + green transition and environmental sustain- Depends on Depends on Depends on ability, including removing coal and gas ex- the package the package the pack- cise exemptions, with revenues generated of measures of measures age of meas- used to finance tax credits for energy efficien- adopted adopted ures adopted. cy and clean energy production; introducing Could be neg- a CO2-related vehicle tax; broaden the base ative without and increase the carbon tax to increase effec- compensating tiveness of carbon pricing signaling, while low- measures ering social contributions of employers and/ or increasing transfers to poorer households; phase out fossil fuel subsidies. Further strengthen the public expenditure MT + Neutral + Neutral review processes, especially through imple- menting effective follow-up mechanisms. Tax Policy and Efficiency Increase the basic PIT allowance. ST/MT - + + Neutral Increase progressivity in SSC contributions. ST/MT Either nega- + + Neutral tive or neutral depending on the package of measures adopted Harmonize SSC and health contributions. ST/MT Positive or Uncertain. Uncertain. Neutral neutral de- Depends on Depends on pending on the package the pack- the package of measures age of meas- of measures adopted ures. Risk of adopted negative ef- fects without compensat- ing measures adopted Single PIT system for labor income, ST + - + Neutral self-employment income and capital income. MT Uncertain Uncertain + Neutral Remove joint PIT scheme. ST/MT + + + Neutral Capital income from foreign sources ST/MT + Uncertain + Neutral Shift from size-based CIT preferences to di- ST/MT Depends on + Uncertain Positive if in- rect tax support to investment. the package vestments of measures are mostly adopted green Replace the “Estonian CIT” with direct tax ST + + Neutral Neutral support to investment. MT Neutral or + Neutral Neutral positive xx Poland Public Finance Review Expected Expected Expected Expected fiscal output social environmental Recommendations Timing impact impact impact impact Remove coal and gas excise exemptions. ST + - Depends on + the pack- age of meas- ures adopt- ed. Negative without com- pensating measures. MT + Uncertain Uncertain + Reform local tax sharing to reduce the vol- ST/MT Neutral Neutral Uncertain Neutral atility of local budgets. Increase local revenue autonomy ST/MT + Uncertain Uncertain Neutral Conduct comprehensive reviews to in- ST + + Neutral Neutral crease the efficiency of the pro-fami- or nega- ly policy instruments and free funds for tive (higher growth-enhancing projects. growth under brown econo- my model) MT + + Neutral Neutral or positive (fast- er transi- tion to green economy) Suspend the 14th pension and redirect ST Neutral Neutral + Neutral the resources released to targeted aid for those most at risk of poverty and exclusion (in-depth review of the poverty/inclusion situation of the single-person households). Maximize the return on spending in edu- ST Neutral Neutral + Neutral cation through optimal allocation of pub- lic resources to skills formation at different MT + + + Neutral stages of the life cycle. Focus on pre- school and adult education as well as on- the-job training. Adopt an integrated healthcare strategy to ST - Neutral Depends on Neutral address pandemic-related and structur- the package al issues and to allocate additional pub- of measures lic health spending. The strategy could give adopted more prominence to primary care, preven- tion, and e-health services MT Neutral or Neutral or Depends on Neutral positive (if positive the package the expenses (if the ex- of measures translate into penditure adopted better health) translates into great- er economic activity) Adjust the remuneration system in pub- ST Neutral or Neutral or Neutral Neutral lic administration to make it attractive to positive positive young, well-educated people. Executive Summary xxi Expected Expected Expected Expected fiscal output social environmental Recommendations Timing impact impact impact impact Leverage digitalization to optimize employ- MT + + Neutral Neutral ment in administration and public services. Consider limiting automatic salary increas- es with tenure. Use the National Recovery and Resilience ST Neutral + Neutral Negative Plan to ensure consistency of various de- (higher velopment programs and to increase growth under growth-enhancing expenditure in the brown econo- short-term. my model) MT + + Neutral or Positive (fast- positive er transi- tion to green economy) Conduct an independent public investment ST + Neutral + Neutral management assessment to identify key inefficiency sources. MT + + + Neutral Greening Public Procurement Build the business case for GPP by launch- MT Uncertain Uncertain Uncertain + ing GPP communication campaigns, iden- tify and support GPP champions; identify entry points for implementation; mobilize financial resources Strengthen the GPP-enabling framework by MT Uncertain Uncertain Uncertain + building a sound institutional framework; well-identified policy space; a clear legal basis. Prioritize and develop an implemen- tation plan. Green innovation market engagement as- MT Uncertain Uncertain Uncertain + sessing market readiness; provide notic- es to the market with sufficient lead time; engage in market consultations to design realistic tenders; build supplier capacity; promote SME participation; and conduct innovation procurements. Build professional competencies for GPP MT + Uncertain Uncertain + assessing the gaps in these competen- cies; assess training needs; ensure buy- in from procurers for green procurement; develop manuals and toolkits for GPP, pi- lot for green tenders, and foster peer learn- ing and networking to ensure profession- alization. Driving organizational change is also critical. Build GPP implementation and monitor- MT + Uncertain Uncertain + ing tools. Promoting Strategic Public Procurement Move contracts in any non-open procedure ST Zl 12.27 bn + Not Not to fully competitive procedures. measured measured xxii Poland Public Finance Review Expected Expected Expected Expected fiscal output social environmental Recommendations Timing impact impact impact impact Reduce the use of framework agreements ST Zl 8.06 bn + Not Not or improve their scope and design. measured measured Move paper-based contracts to electron- ST Zl 4.41 bn + Not Not ic auctions. measured measured Increase advertisement periods: Eliminate ST Zl 7.25 bn + Not Not short advertisements (1-39 days) by in- measured measured creasing them to more than 40 days. Smooth spending throughout the last quar- ST Zl 5.4 bn + Not Not ter of the budget year by reallocating some measured measured of contracts to a nearby cheaper month. Move tenders bundling heterogeneous ST Zl 0.15 bn + Not Not products to less diverse tenders (bunding measured measured two products only). Shorten decision periods by use of elec- ST Zl 6.67 bn + Not Not tronic or simplified processes and by in- measured measured creasing capacity Move contracts from the lower half success ST Zl 2.51 bn + Not Not rate organizations (less than 99 percent) measured measured to highest success rate organizations (100 percent). Move 0-2 bidder contracts to 3 bidder ST Zl 67.9 bn + Not Not contracts measured measured Move contracts from high-spending con- ST Zl 13.19 bn + Not Not centration buyers (2.97-100 percent) to measured measured lower-spending concentration buyers (1.7- 2.9 percent) Move contracts in high-concentration mar- ST Zl 0.44 bn + Not Not kets (0.26-100 percent) to lower-concen- measured measured tration markets (0.11-0.259 percent) Increase market share of local suppliers ST Zl 0.32 bn + Not Not (i.e., same state). measured measured Increase share of SME participants. ST Zl 0.65 bn + Not Not measured measured Move contracts supplied by highly-special- ST Zl 3.3 bn + Not Not ized suppliers (operating on 1 market) to measured measured less-specialized suppliers (operating on 2 markets. Enhancing Public Financial Management and Fiscal Transparency Prepare and publish a comprehensive road ST Neutral + Neutral Neutral map and implementation plan for sequenc- ing the interconnected SBR and PSAR in- cluding further alignment with IPSAS and preparation of consolidated financial statements. Executive Summary xxiii Expected Expected Expected Expected fiscal output social environmental Recommendations Timing impact impact impact impact Supplement budget execution reports on a ST + + Neutral Neutral cash basis with additional information on an accrual basis, including on assets and liabilities. Complete the development and introduc- ST/MT + + Neutral Neutral tion of the SCoA and the integration of fis- cal and financial reporting across the PFM cycle. Carry out an assessment of ICT capabil- ST/MT - + Neutral Neutral ities and requirements to prepare medi- um-term strategy, including ICT solutions, to enable collection and consolidation of financial information and other PFM pro- cesses at the central and decentralized level. Establish a permanent and comprehensive ST/MT - + Neutral Neutral training program for government staff with PFM related functions. Establish a baseline and gap analysis of ST/MT Neutral + Neutral Neutral existing PFM practices and carry out peri- odic evaluations of progress. Actions to strengthen budget and fiscal transparency practices Improve budget transparency – (online ST Neutral + Neutral Neutral mid-year budget implementation and Citizens Budget, a more detailed mac- ro-economic forecast). Improve public participation in the budget ST Neutral + Neutral Neutral process Make budget oversight more effective ST Neutral + Neutral Neutral (at the budget proposal, in-year reporting stage) Note: ST: short term, MT: medium term Notes 1. World Bank. 2017. Poland — Systematic Country Diag- period. Minimum pension expenditure is expected to nostic: Toward a Strategic, Effective, and Accountable rise by 2.3 percent of GDP by 2070. State (English). Washington, D.C.: World Bank Group. 3. The 2021 Ageing Report (EC 2021) defines the adverse 2. In the case of Poland, reduction in the benefit ratio, cov- structural scenario as a stronger cyclical downturn in erage ratio, and labor market are estimated to reduce the lagged recovery scenario and that the growth po- gross public pension expenditure-to-GDP. The coverage tential will be lower over the next decade and poten- ratio will contribute to a 1.5 percentage point of GDP de- tial output growth will thus be permanently lower than cline over the 2019 – 2030 period, while the benefit ra- under the baseline scenario. tio will contribute 1.4 percentage points over the same 4. The 2021 Ageing Report. European Commission, 2021 xxiv Poland Public Finance Review CHAPTER 1 FISCAL ASSESSMENT FISCAL TRENDS the implementation of a stabilizing expenditure rule3 in planning the 2014 budget expenditures — which be- Prudent macroeconomic policies and effective ab- came binding starting in 2015 — has further contrib- sorption of EU investment funds contributed to uted to the consolidation, and expenditures above the inclusive growth and poverty reduction. Poland re- limit set by the rule had to be matched with new reve- corded almost 30 years of uninterrupted growth until nue measures. The government managed a sizeable fis- 2020 and has been one of the fastest-growing econo- cal consolidation, and the deficit declined from more mies in Europe. The economy expanded at a sustained than 7.3 percent of GDP in 2009 – 2010 to 2.6 percent pace, averaging 3.6 percent over the 2010 – 19 period, of GDP by 2015, and further to a historical low deficit despite the euro debt crisis. Poland continued its in- of 0.2 percent of GDP by 2018, with a slight uptick in come convergence, reaching more than 70 percent of 2019 to a deficit of 0.7 percent of GDP. the average EU GDP per capita by 2015.1 This strong growth performance and convergence was support- Strong cyclical factors, in conjunction with efforts ed by Poland’s record of prudent macroeconomic and to improve tax compliance and combat tax fraud fiscal policies, anchored by inflation targeting, a flex- and abuse since 2016, helped support revenues. The ible exchange rate, and a sound fiscal framework sup- efforts to improve tax compliance include the introduc- ported robust and resilient growth. A sound financial tion in 2016 of the Standard Audit File–Tax (SAF).4 This sector and better access to long-term credit have also improved electronic data reporting and data processing, supported sustained growth. The global COVID-19 cri- starting with large enterprises, and was rolled out grad- sis has affected Poland as well, triggering the first out- ually to the other enterprises. Furthermore, the modifi- put contraction in 30 years, affecting lives, livelihoods, cation of the rules for accounting VAT on intra-Commu- economic activity, and public finances markedly. nity purchases of motor fuels (so-called fuel packages) came into force in August 2016, including amendments to the VAT for intra-EU purchases, and excise and con- Poland has built fiscal buffers cession provisions for trading in liquid fuels. The en- for most of the decade prior to ergy package introduced in September of the same the COVID-19 crisis year further supported revenues from VAT, excises on motor fuels and fuel fee revenues, by closing loop- Poland built fiscal buffers in the years preceding holes to expand the legal fuel market. Similar efforts the COVID-19 pandemic, aided by robust growth, in tax compliance resulted in increased excise duties expenditure consolidation, and efforts to increase on tobacco products. Furthermore, the General Anti- tax collections. Poland started rebuilding fiscal buf- Avoidance Rule (GARR) became effective in mid-2016, fers in the wake of the 2007 – 08 Global Financial Crisis and in March 2017 the National Revenue Administration (GFC), after its deficit shot up to 7.3 percent of GDP on was founded, by merging the tax administration, fiscal average over the 2009 – 10 period. Poland was the only control, and Customs service. These measures, together EU member that did not experience a recession.2 The with strong economic growth supported an increase in Council of the European Union triggered the Excessive general government revenues of 1.2 percentage points Deficit Procedure (EDP) in 2009 for Poland, recom- of GDP over the 2016 – 17 period. Of this figure, it is esti- mending that Poland bring its general government mated that the compliance effect represented close to deficit below 3 percent of GDP by 2012, which was 0.9 percentage points of GDP (Figure 1.1).5 subsequently extended to mid-2015. In June 2015 the EU Council closed EDP based on the 2014 deficit figure Improved tax administration in conjunction with corrected for the pension reform cost. A temporary strong growth resulted in higher tax revenues. expenditure rule was proposed to lower the structural General government tax revenues increased by 2.7 per- deficit and stabilize it to a level consistent with meet- centage points between 2015 and 2019 to the high- ing the Medium-Term Objectives (MTO) of a deficit of est level since 1999. Higher direct and indirect taxes 1 percent of GDP, while the VAT rates were increased contributed to this increase. Higher wages and em- starting in 2011. Social contributions were hiked by ployment supported personal income tax revenues, 2 percentage points starting in 2012 (Box 1.1). Notably, which contributed 0.7 percentage points of GDP, or 2 Poland Public Finance Review Figure 1.1 General government revenues Figure 1.2 General government tax revenues 45 36 40 32 28 Percent of GDP 35 24 Percent of GDP 30 20 25 16 20 12 8 15 4 10 0 5 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Individual income taxes SSC Tax Revenues (excluding SSC) Total SCC Corporate income taxes Taxes on G&S Non-Tax Revenues Other taxes Source: Eurostat. Source: OECD. Box 1.1 Poland’s fiscal consolidation in the wake of the Global Financial Crisis 2010 – 15 Poland achieved fiscal consolidation in the wake of the On the revenue side, several structural revenue meas- 2007 – 08 Global Financial Crisis through a policy mix of ures were introduced, albeit with a limited impact on spending and revenue measures aiming to lower the defi- tax-to-GDP ratio of less than 0.7 percentage point. The cit to levels consistent with the MTOs. The pace of consoli- tax thresholds and the lump sum of deductible costs dation slowed in 2013. Poland remained under the exces- for PIT were frozen at the 2009 level, and changes were sive deficit procedure from 2009 until mid-2015. also made to the child relief schemes in early 2013. The On the expenditure side, containment measures helped government introduced in 2011 a temporary increase in reduce expenditure by 3.3 percentage points of GDP and the three VAT rates to 23 percent, 8 percent, and 5 per- included the following: cent until 2013 (which then became permanent), while broadening the base for the lowest VAT rate to several • Gradually raising the retirement age for both women and basic food products. The government has since 2011 men and equalizing it at 67 starting in 2013, as well reduced rights to deduct VAT on purchases of person- as increasing retirement age for uniformed services; al vehicles. In 2014, the government limited access to • Introducing a disciplinary expenditure rule to limit the partial VAT refunds on purchases for housing purpos- increase in discretionary expenditures and new fixed ex- es. The catalog of goods to which the reverse charge penditures to 1 percent annually in real terms starting mechanism applies was extended in 2013, and the VAT in 2011 (replaced by the expenditure stabilizing rule ef- rate on non-universal postal services was increased to fective at the end of 2013); 23 percent in April 2013. The limit below which taxpay- • Taking debt-servicing reduction measures, through the ers are authorized to keep a cash register was cut by 50 transfer of assets from open insurance funds to the percent to Zl20,000 in 2013. The excise duty for ciga- Social Insurance Institution and consolidation of liquid- rettes has been increased annually since 2010, while ity management of public finance sector units; the excise tax on oil and fuel fees was hiked in 2012. • Freezing the remuneration fund in the state budget units The excise duty relief for biocomponents was removed at the level of the previous year, starting in 2011; in mid-2011, and a fee on the use of selected natural re- • Banning the adoption of legal acts provisions by the sources was introduced in April 2012. Since November Council of Ministers that result in a rise in expenditures 2013, the government applied the excise tax on natural or a decline in revenues in the public finance sector units gas used for heating purposes, excluding households. In when Poland is under the excessive deficit procedure; 2013, it introduced the system of CO2 emission allow- ances, and in 2014 it increased excise duties on ethyl • Adjusting the fiscal provisions for the local government alcohol. The internet discount was eliminated gradually, units (balanced current accounts and individual debt and the 50 percent of tax-deductible costs due to copy- limits); rights and related rights was reduced in January 2013. • Cutting early retirement benefits; The share of pension contributions transferred to the • Introducing an income criterion for the one-off child- Open Pension Fund (OFE) was lowered by 5 percentage birth allowance, and; points to 2.3 percent of the base as of May 2011, and • Reducing sickness benefits for uniformed services, the disability contribution was increased by 2 percent- judges, and prosecutors, to be covered by the gener- age points in February 2012. In 2014, taxation of limit- al scheme (as of 2014). ed joint-stock partnership was introduced. Source:  based on information from Convergence Program Updates 2014 – 2015, and the Multi-Annual State Financial Plans. Fiscal assessment 3 almost 25 percent to the increase in tax revenues over collections contributed nearly 1 percentage point of this period. Meanwhile, the increase in corporate in- GDP to higher tax revenues over the 2015 – 19 peri- come tax revenues was a more modest 0.4 percent- od, accounting for more than a third of the overall age points of GDP, even though there was a significant increase. Higher VAT revenues were supportedby increase in nominal terms over the 2016 – 2018 peri- higher consumption, higher VAT rates, improved tax od following the introduction of GAAR (Figure 1.3). compliance through measures aimed at sealing the Measures to improve tax compliance and to fight ag- VAT, including the split payment mechanism (SPM), gressive tax planning have also had a positive impact and the so-called fuel and energy packages, which on CIT and PIT revenues. For example, it is estimated boosted indirect tax revenues. VAT revenues grew that the compliance effect for CIT averaged 0.2 per- faster than consumption (Figure 1.4). The estimat- cent of GDP over the 2017 – 18 period.6 The effect of ed compliance effect for VAT is around 0.3 percent the introduction of the flat PIT for small businesses of GDP on average for the 2017 – 19 period.7 in terms of the Standard Audit File for Tax (Jednolity Plik Kontrolny, or JPK) is estimated at 0.1 percent of Figure 1.4 VAT revenues and final GDP. A strong labor market also boosted social secu- consumption rity contributions, which increased by 0.8 percent- 400 age points, accounting for nearly 30 percent of the 350 increase in tax revenue. Measures implemented by 300 ZUS (Social Security Institution), such as the intro- Index, 2000=100 duction of the so-called e-Contribution (e-Składka), 250 electronic sick leave, and the increase in the effec- 200 tiveness of enforcement of receivables also support- 150 ed social insurance contributions. 100 50 Figure 1.3 Contributions to tax-to-GDP ratio 0 change by tax type 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 3.0 Taxes on goods and services VAT Revenues Final consumption 2.7 2.5 Other taxes Sources: World Bank; OECD. Taxes on property 2.0 Taxes on payroll and workforce Spending containment efforts contributed to the Social security fiscal consolidation in the wake of the GFC and Percentage points 1.5 contributions the euro area debt crisis, with investment bear- Taxes on income, 1.0 profits and capital ing the brunt. Spending by the general government gains declined by 2.7 percentage points of GDP to an aver- 0.5 Tax-to-GDP age of 41.5 percent of GDP for the 2015 – 19 period, rel- 0.6 p.p. change ative to the 2000 – 14 period (Box 1.1). After a marked increase in the spending-to-GDP ratio in the pre-ac- 0 cession years and in the wake of the GFC and the euro area debt crisis (when general government spending -0.5 grew on average by 8.5 percent per year), spending growth slowed markedly as the government started -1.0 2011–15 2015–19 fiscal consolidation, averaging 2.1 percent annually Sources: World Bank; OECD. over the 2012 – 15 period. A large part of the adjust- ment was done through investment, which fell close Policy changes for indirect taxes and measures to to 30 percent in nominal terms between 2011 and 2013, improve VAT compliance combined with robust in part because of budget rigidities. The share of cap- consumption boosted indirect tax revenues. VAT ital expenditure declined by 1.3 percentage points 4 Poland Public Finance Review of GDP between 2010 and 2015, while the share of continued on the revenue side. Social expendi- compensation of employees declined by 0.7 percent- tures rose faster than GDP on account of the social age points of GDP over this period. This is due to the policies introduced by the government. The govern- slowdown in the growth in compensation as a result ment’s key social policies introduced by the govern- of freezing the wage bill for most central govern- ment included the child benefit (Family 500+ pro- ment institutions. gram), in force since April 2016, as well as lowering again the statutory retirement age since October 2017 The strengthening of the fiscal framework to 60 years for women and 65 years for men, which through the introduction of the stabilizing ex- carried additional costs estimated at 0.5 – 0.6 percent penditure rule limited growth of public expendi- of GDP. The child benefit was introduced in 2016, with ture in the subsequent years. The rule, introduced the full yearly cost effect felt in 2017. This contrib- in 2013 and applied for the first time to the 2015 budget uted to an increase in social benefits between 2015 law, also accounts for the estimated effect of discre- and 2019 of 1 percentage point of GDP (Figure 1.5). tionary tax and SSC revenue measures when setting Furthermore, the Family 500+ contributed to the the expenditure limit. The subsequent consolidation 1 percentage point of GDP increase in current trans- was concentrated on administration costs and on so- fers received by households and non-profit institu- cial expenditure. tions serving households (NPISH) in 2016 (to 17.4 per- cent of GDP). Transition between EU MFF affects absorption of EU structural funds and public investment. Figure 1.5 Changes in general government In 2016 public investment slumped significantly by expenditure 1.2 percentage points of GDP to 3.3 percent of GDP, but 2007 42.9 recovered in 2017. The decline in 2016 was brought Capital exp. 1.5 about partly by terminating the use of the funds from Compensation 42.9 0.6 the EU’s Financial Perspective for 2007 – 13, and part- Social benefits 0.5 ly by the transition to the EU’s 2014 – 20 MFF, both of Other (net) 0.3 which affected the absorption of EU structural funds. 2010 45.8 In particular, the very low use of the EU funds by Capital exp. 1.3 local governments contributed to a 0.8 percentage Compensation 0.7 Interests 0.7 point decline in the public investment-to-GDP ratio. Intermediate cons. 0.6 The same year, central government investment de- Subsidies 0.4 clined on account of lower infrastructure expendi- Other (net) 0.4 tures incurred by the PKP PLK S.A. (the Polish rail- 2015 41.7 way infrastructure managing company), which were Social benefits 1.0 partly offset by the growth in infrastructural expen- Interests 0.4 ditures incurred by the National Road Fund under Other (net) 0.5 2019 41.8 the National Road Construction Program for 2014 – 23. In 2017 general government investment rose 0.4 per- 39 40 41 42 43 44 45 46 47 centage points of GDP to 3.9 percent of GDP, as local Percent of GDP, percentage points government investments surged on the increased Sources: World Bank calculations; Eurostat. use of funds from the EU’s Financial Perspective for the years 2014 – 20. Increased investment in railway High budget rigidity affects the quality and com- infrastructure by the Polish Railway Lines (Polskie position of fiscal consolidation. The share of rigid Linie Kolejowe, or PLK) contributed to the increase expenditures has increased over the past decade by in the central government investment. almost 2 percentage points, to 75 percent of spend- ing. Highly rigid spending (compensation of employ- Since 2016 the consolidation was slower than pre- ees, social transfers, and interest payments) accounts viously envisaged as new social programs were for 69 percent of total spending, while an additional introduced, although progress on consolidation 6 percent of total spending has a moderate degree of Fiscal assessment 5 Figure 1.6 Contributions to changes in in open market operations, and its holdings of gov- general government expenditure ernment and government-guaranteed securities in- 2.5 creased to 8 percent of general government debt by Capital expenditures 2020. Low and declining financing costs had con- 2.0 Social benefits tributed to the increase in the fiscal space and prior 1.5 Intermediate to the crisis there were no concerns with fiscal sus- consumption tainability, with both medium- and long-term risks 1.0 Compensation assessed as low.8 Percentaga points 0.5 of employees Interests 0 Other -0.5 Total The COVID-19 pandemic has affected -1.0 lives, livelihoods, economic activity, -1.5 and public finances markedly -2.0 The global COVID-19 pandemic caused the largest global recession since World War II, also affecting -2.5 2008–11 2012–15 2016–19 Poland. Poland suffered its first output contraction Sources: World Bank calculations; Eurostat. since 1991, with lives and livelihoods lost. Economic activity in Europe and Central Asia (ECA) is estimat- rigidity, including economic subsidies and other trans- ed to have contracted 2.1 percent in 2020 in the wake fers. Declining interest rates helped reduce the cost of disruptions related to the COVID-19 pandemic. of servicing the public debt, creating additional fiscal The pandemic is expected to erase at least five years space. Interest payments declined by 0.7 percentage of per capita income gains in about a fifth of the re- points of GDP between 2010 and 2015 (Figure 1.5), as gion’s economies and increase the poverty headcount. debt burden declined. Interest expense also declined Economies with strong trade or financial links to the following the 2014 debt reduction, the one-off trans- euro area and those heavily dependent on services fer and subsequent redemption of government debt and tourism have been hit the hardest in the region. securities (of close to 9 percent of GDP) from Open Even though the well-diversified Polish economy was Pension Funds to ZUS. By 2019, the cost of servicing one of the economies least affected by the COVID-19 the general government debt dropped to 1.4 percent pandemic in the Europe and Central Asia region, its of GDP, compared with 2.7 percent of GDP in 2012. GDP declined by 2.7 percent in 2020 (Figure 1.7), caus- ing job losses (ILO 2021; World Bank 2021a). Cyclical factors and improved tax administration contributed to primary surpluses over the 2017 – 19 Figure 1.7 Poland GDP growth period. Strong economic growth and a strong labor 8 market boosted tax revenues and social security con- 6 tributions. This, coupled with improvements in tax compliance, resulted in primary surpluses averaging 4 0.7 percent of GDP over the 2017 – 2019 period. The cor- 2 ollary was a marked decline in general government Percent 0 debt of almost 9 percentage points, from 54.6 per- cent of GDP in 2016, to 45.7 percent of GDP by 2019. -2 General government debt held by non-residents fell -4 below 20 percent of GDP by 2019. -6 Strong fundamentals and favorable global in- -8 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 terest rates have contributed to a declining ef- fective interest rate. The decline has been more Source: GUS. pronounced since early 2020. The NBP has engaged Note: Orange line denotes the COVID-19 crisis. 6 Poland Public Finance Review Pandemic-related restrictions, heightened un- rebounding 7.5 percent in Q3.9 Poland’s sizeable and certainty, and negative confidence effects damp- swift policy response, both monetary and fiscal, and ened Poland’s household consumption and invest- an easing in restrictions in Q3 contributed to the re- ment. Household consumption contracted by 3.0 covery in economic sentiment and output. percent while investment dropped sharply by 8.4 per- cent year-on-year in 2020 (Figure 1.8). Government A second wave of the pandemic and targeted re- spending to reduce pandemic impacts contributed strictions affected output in Q4 2020. Partial ad- to the 3.2 percent increase in public consumption, aptation by some economic agents to operating un- while public investment remained stable. Disruptions der restrictions, a narrower scope of the restrictions to international trade and transport and lower de- imposed, more resilient economic sentiment among mand from some key EU partners caused a mild 0.5 both consumers and businesses, and recovering ex- decline in exports. Lower domestic expenditure re- ternal demand limited the output decline to 0.7 per- sulted in a 2.6 percent decline in imports. As a re- cent quarter-on-quarter (Figure 1.9). Even though the sult, net exports contributed 1 percentage point to infection rates increased markedly during the sec- growth in 2020. ond wave of COVID-19, private consumption spend- ing declined at a slower pace than during the first Figure 1.8 Poland GDP growth composition wave. This was in part due to narrower restrictions, 14 but also due to pent-up demand that was partly sup- ported by higher wage growth. Investment activity, 12 on the other hand, continued to suffer on account 10 of pandemic restrictions, depressed business sen- 8 timent, a high degree of uncertainty regarding the duration and intensity of the pandemic, low capaci- Percentage points 6 ty utilization, and expectations of lower level of eco- 4 nomic activity. 2 0 -2 Figure 1.9 Quarterly GDP path -4 110 Index of real GDP in US$, 2019Q4=100 -6 105 -8 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 100 Gov. cons. Exports GFCF 95 Inventories Private cons. Imports 90 Statistical disc. GDP Source: GUS. 85 Subsequent waves of the COVID-19 pandemic 80 and restrictions imposed to contain them have 2016Q2 2016Q3 2016Q4 2017Q1 2017Q2 2017Q3 2017Q4 2018Q1 2018Q2 2018Q3 2018Q4 2019Q1 2019Q2 2019Q3 2019Q4 2020Q1 2020Q2 2020Q3 2020Q4 2021Q1 2021Q2 2021Q3 2021Q4 2016Q1 affected the pace of economic recovery and may continue to do so. As the pandemic intensified and strained the health sector, Poland and other nations Actual GDP Pre-Covid 5-year GDP trend imposed far-reaching restrictions on mobility and Sources: GUS; World Bank. economic activity. These restrictions affected both demand and supply, with global value chains coming Due to a resurgence of COVID-19 infections and under pressure. As a result, Poland’s GDP plunged the current pace of vaccinations, the econom- 9 percent quarter-on-quarter in Q2 of 2020, before ic recovery is projected to be above potential in Fiscal assessment 7 2021. The outlook remains uncertain, however, and A substantial and front-loaded growth could be weaker than envisioned if the pan- stimulus package was launched demic takes longer than expected to fade, if exter- in response to the COVID-19 crisis nal financing conditions tighten further, or if geo- political tensions escalate again. Even though output The government swiftly launched a substantial rebounded and is expected to reach 2019 levels by and front-loaded stimulus package and accom- end-2021, a significant output loss compared with modative monetary policy to mitigate the health, a non-pandemic counterfactual remains, with out- social and economic impact. The announced emer- put by 2023 still 8 percent lower than in a non-pan- gency package amounted to 6.4 percent of GDP in demic counterfactual (Figure 1.10). Poland, 3.1 percentage points higher than the EU av- erage, as of December 2020.10 Meanwhile, liquidity Figure 1.10 Lost output relative to pre-COVID support announced by the government to the tune of 5-year trend 5.0 percent of GDP was comparatively lower relative 130 to the EU average (Figure 1.11, Figure 1.12). The size- able economic package has helped support the re- covery of economic agents’ confidence and helped Index of real GDP, 2019=100 110 preserve employment, even though the number of registered unemployed rose by 180,000 by December 90 2020, year-on-year, while the average paid employ- ment was 67,000 lower over the same period. 70 Fiscal space narrowed considerably in the wake of 50 the COVID-19 crisis. The large fiscal stimulus togeth- er with the decline in economic activity brought about 30 an increase in general government deficit (ESA 2010) to an estimated 7 percent of GDP in 2020 (Figure 1.8). 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 The central government deficit rose sharply to an es- Revised GDP Pre-Covid 5-year GDP trend timated 8 percent of GDP in 2020, while both the lo- Sources: GUS; World Bank. cal government and the Social Insurance Fund saw Figure 1.11 COVID-19 fiscal policy responses by country 35 30 25 Percent of GDP 20 15 10 5 0 Romania Slovak Republic Finland Bulgaria Portugal Spain Estonia EU Belgium Hungary Cyprus Czech Republic Netherlands Lithuania Italy Malta Luxembourg Sweden France Croatia Norway Denmark Ireland Poland Greece Slovenia Latvia Germany Austria United Kingdom Spending and Revenue measures Liquidity Support Source: IMF. Note: EU = European Union. Figure includes all measures announced through March 17, 2021. 8 Poland Public Finance Review Figure 1.12 Fiscal responses and outlook A. Composition of fiscal response to COVID-19 B. Poland growth prospects and levels in Poland and the euro area relative to pre-pandemic projections 30 30 6 4 25 25 2 20 Percentage points 20 Percent of GDP 0 Percent 15 15 -2 10 10 -4 5 5 -6 0 0 -8 Euro area Poland 2020 2021 2022 2023 Discretionary measures Contingent liabilities Level relative to pre-pandemic Equity injections Total fiscal (RHS) C. Cumulative change in cyclically adjusted D. Output gap and fiscal impulse prospects primary balance in ECA and Poland 0 6 6 Percentage points of potential GDP -1 4 4 Percentage points of GDP 2 2 -2 Percent of potential 0 0 -3 -2 -2 -4 -4 -4 -5 -6 -6 2020 2021 2022 2020 2021 2022 -6 ECA Poland 2019 2020 2021 2022 2023 2024 2025 2026 Output gap range Output gap median Poland ECA Euro area Fiscal impulse (RHS) Sources: IMF; World Bank. Note: ECA = Europe and Central Asia. B.-D. Shaded areas indicate forecasts. A. Figure shows total measures either planned or under consideration as of March 17, 2021, as a share of 2019 nominal GDP. Sample includes 18 advanced economies. B. Projections are latest World Bank forecasts included in the June 2021 Global Economic Prospects report. Bars denote latest forecast; dashes correspond to the percent difference between the latest projected levels of GDP and those in the January 2020 Global Economic Prospects report. C.D. Aggregates are calculated using 2019 real GDP weights at average 2010-19 prices and market exchange rates. C. Projections are International Monetary Fund calculations included in the April 2021 World Economic Outlook report. D. Fiscal impulse defined as change in the cyclically adjusted primary balance (CAPB) from previous year. Decline in the CAPB indicates fiscal consolidation; increase in the CAPB indicates fiscal expansion. Fiscal impulse projections are International Monetary Fund calculations included in the April 2021 World Economic Outlook report. Fiscal assessment 9 an improvement in their balances to 0.2 percent of increasing by 0.6 percentage points of GDP to 41.7 per- GDP and 0.9 percent of GDP, respectively (Figure 1.10). cent of GDP in 2020. Meanwhile, general government Despite the important contraction in output, govern- spending rose sharply, by 18 percent year-on-year in ment revenues increased by an estimated 2.9 percent 2020, on account of the sizeable fiscal package, com- in 2020, albeit a marked deceleration from an average pared with an average 7.7 percent annual increase over 9.3 percent annual increase over the 2017 – 2019 period. the period 2017 – 2019. Spending increased 6.9 percent- Government revenues were stronger than expected, age points of GDP to reach 48.7 percent of GDP in 2020. Box 1.2 Poland’s COVID-19 emergency package The support measures implemented in 2020 impact- Fund (PFR), ZUS social insurance contribution exemp- ed the fiscal deficit by an estimated 5.2 percent of GDP, tions, wage subsidies, and loans to micro-enterprises. Its with an additional 1.5 percent of GDP expected in 2021. aim was to protect the labor market and provide compa- The economic packages consisted of increased unem- nies with financial liquidity. New credit guarantees and ployment benefits, wage subsidies, self-employed bene- micro-loans for entrepreneurs estimated at Zl 74 billion fits, ZUS social insurance contribution exemptions, loan (3.3 percent of GDP) were also approved. As of December guarantees (BGK Liquidity Guarantee Fund), forgivable 2020, the Polish Development Fund has disbursed close loans to micro-enterprises (Zl 5,000), liquidity loans to to Zl60 billion (2.6 percent of GDP) of its liquidity pro- enterprises (PFR), with up to 75 percent eligible for for- gram, benefitting more than 340,000 micro, small and giveness subject to employment conditions. The eco- medium enterprises (Figure B1.2.2). nomic packages also included additional funding for the Poland has extended some measures into 2021 as a sec- health care sector in the amount of 0.24 percent of GDP ond and then a third wave of the pandemic affected the (Figure B1.2.1). The fiscal measures in 2020 were com- recovery. In response to the second wave of the pandem- prised primarily of PFR liquidity loans write-offs for micro, ic, measures were targeted primarily to the most affected small and medium enterprises (MSMEs) (about 34 per- sectors, and including write-offs of repayable loans for cent of fiscal measures). Wage subsidies, SIC exemp- SMEs subject to conditions, extensions of liquidity loans, tions, and benefits to self-employed individuals amount- guarantees, wage subsidies and ZUS social insurance ed to about a third of the fiscal package. contribution exemptions, benefits for the self-employed A large part of the stimulus emergency support was and subsidized loans, amounting to an estimated 1.5 per- disbursed in Q2 2020. This included grants under the cent of GDP. A worsening of pandemic trend may lead to “Financial Shield” financed by the Polish Development a further extension and broadening of relief measures. Figure B1.2.1 Anti-crisis and financial Figure B1.2.2 Support to enterprises Shields in 2020 Zl 0.197 bn 6 Wage subsidies ARP: Zl 0.197 bn 6 Wage loans Micro for micro subsidies support from the Industry ARP: Development Agency 5 enterprises Zl 70.08 bn Micro loans for micro support from the Industry Benefits enterprises for self Development Agency PFR: 5 Zl 70.08 bn employed Zl 32.30 bn Financial PFR: 4 Benefits for self Shields Exemption employed from SSC ZUS: Financial of GDP Zl 32.30 bn 4 exemption from the obligation Shields Allowance Exemptionfor parents from SSC ZUS: to pay ZUS contributions of GDP 3 exemption from the postponement obligation / installment Support Addit. Unempl. Allowance benefits for parents to payofZUS contributions to enterprises Percent payment contributions to 3 postponement / installment ZUS Support Tourism vouchers Addit. Unempl. benefits Zl 202.45 to bn enterprises Percent payment of contributions to 2 PFR liquidity loans ZUS Tourism vouchers Zl 202.45 bn 2 Zl 30.39 bn BGK credit guarantees PFR liquidity loans Zl 69.48 bn Ministry of Development, Zl 30.39 bn 1 Capital inject. BGK credit PFR, ARP guarantees Labor & Technology: BGK: Zl 69.48 bn Ministry Development, of for co-financing job protection Loan 1 ARP support Capital inject. PFR, ARP fromLabor & Technology: the Employee Benefits BGK: guarantees co-financing for Guarantee job protection Fund Loan 0 Health care funding ARP support from the Employee Benefits guarantees co-financing of employee Fiscal Guarantees Other Health care funding Guarantee Fund salaries from the European 0 measures and loans Social co-financing of Fund employee Fiscal Guarantees Other salaries from the European measures and loans Source: Chancellery Social Fund of the Prime Minister. Sources: Convergence Program; IMF; World Bank. Note: Figures reflect data as of March 2021. Source: based on information from Convergence Program Updates 2014 – 2015, and the Multi-annual State Financial Plans. 10 Poland Public Finance Review Many of the measures are one-off and have been Figure 1.14 General government balance largely withdrawn in 2021. Reflecting the compo- 50 8 sition of the COVID-19 support package, subsidies were the largest driver of the increase in general Percent of GDP Percent of GDP 45 6 government spending in 2020. Subsidy increases 40 4 accounted for nearly 45 percent of the increase in government spending, or 3.3 percentage points of 35 2 GDP, followed by social benefits, which accounted 30 0 for a quarter of the increase (1.5 percentage points 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 of GDP) (Figure 1.13). The general government defi- cit is expected to decline to 5.8 percent of GDP in 2021 as the economic recovery strengthens, and as Expenditure Revenues Deficit temporary stimulus measures are being withdrawn. Source: Eurostat. Figure 1.13 Change in general government Figure 1.15 General government spending in 2020 primary balance 2019 41.8 2 Subsidies 3.3 0 Percent of GDP Social benefits & transfers in kind 1.5 -2 Capital transfers 0.8 -4 Compensation of employees 0.6 Other current transfers 0.3 -6 Gross capital formation 0.2 -8 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Other (net) 0.2 2020 48.7 38 40 42 44 46 48 50 52 Primary Balance Percentage points of GDP General Government Balance Sources: World Bank analysis based on Eurostat data. Source: GUS. Due to the large deficit, general government debt Figure 1.16 General government balance rose by 11.9 percentage points to 57.5 percent of by government level GDP in 2020 (Figure 1.14). This is the highest debt 3 level recorded in almost three decades. The State 2 Treasury debt reached 47.2 percent of GDP, up 4.8 per- 1 centage points, while public debt following the Public 0 Percent of GDP -1 Finance Act (PFA) definition reached 47.8 percent of -2 GDP, still well below the 55 and 60 percent of GDP -3 thresholds defined in the PFA and the Constitution. -4 An important part of the stimulus has been provid- -5 ed through the Polish Development Fund (PFR) and -6 Bank Gospodarstwa Krajowego (BGK)11, which have -7 -8 issued bonds worth 7.1 percent of GDP in 2020. 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 The National Bank of Poland’s accommodative Central government deficit monetary policy contributed to lowering financ- Local government deficit ing costs. The NBP, like other central banks in ECA, Social security funds balance has responded to the economic and financial mar- General government deficit ket shocks induced by the COVID-19 pandemic by Source: Eurostat. Fiscal assessment 11 cutting short-term policy rates to a level close to FISCAL FRAMEWORK their effective lower bounds. The NBP cut its refer- ence rate from 1.5 percent to 0.1 percent; launched Despite a strong regulatory framework, fiscal per- a quantitative easing program; and lowered the re- formance deteriorated even prior to the COVID-19 serve requirement from 3.5 percent to 0.5 percent. crisis. After the fiscal position improved consider- To stabilize financial markets and support activity, ably over the last decade, with the fiscal deficit nar- NBP employed an asset purchase program amount- rowing and the stock of public debt remaining below ing to nearly 4.6 percent of GDP. This, in conjunction 60 percent of GDP, Poland’s fiscal performance wors- with spillovers from accommodative monetary pol- ened in both 2017 and 2019. Over this period, Poland’s icies in advanced economies, has contributed to fi- compliance with the requirements under the pre- nancial market stability. ventive arm of the EU Stability and Growth Pact has been unsatisfactory under European Commission Figure 1.17 Higher tax and social guidelines. The reforms implemented to raise reve- contributions helped narrow headline deficits nue through increased tax compliance and expand- 55 ed excise taxation have proven insufficient to offset additional social spending, especially with respect to 50 pension and child benefits (IMF 2021). The European 45 Commission’s assessments of Poland’s fiscal position Percent of GDP relative to budgetary objectives have concluded that 40 Poland’s structural balance breaches the MTO by an 35 increasing margin (European Commission 2018). In 2020, the European Commission found that there was 30 significant deviation from the adjustment required to comply with the MTO both in terms of change in 25 structural balance and annual growth of net public 20 expenditure (European Commission 2020). 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 To strengthen adherence to the EU framework, Tax Revenues Other revenues the Council of the European Union has for- Expenditures mally provided recommendations to Poland Source: Eurostat. as part of its yearly round of country-specific recommendations. Under the preventive arm of The decline in Treasury bond yields reduced the the SGP, it has been recommended that Poland limit costs of public debt servicing, thus providing the nominal growth of government primary expen- additional room for fiscal stimulus. The NBP bal- diture (net of discretionary revenue measures and ance sheet has expanded by 5 percent of GDP be- one-offs) to 4.2 percent in 2019, consistent with a tween February and June 2020, and an additional structural adjustment of 0.6 percent of GDP. Poland 3 percent of GDP between July and December 2020, did not meet this benchmark, however, resulting in as the NBP intervened on the foreign exchange a deviation of 2 percent of GDP. The structural defi- market. Increased liquidity provision to banks cit deteriorated by 0.8 percentage points of potential meant that the balance sheet expansion was sig- GDP, implying a significant deviation of 1.4 percent nificantly more than the asset purchases. Domestic of GDP in 2019 from the recommended 0.6 percent banks have also increased their government-issued of GDP structural adjustment. The primary balance bond holdings. The expectations are that the NBP deteriorated by 0.5 percentage points to 0.7 per- will maintain an accommodative monetary poli- cent of GDP, while the structural primary balance cy stance through further currency interventions increased by 1.1 percentage points to 1.5 percent. and asset-purchase programs. This meant a significant deviation from the recom- 12 Poland Public Finance Review Figure 1.18 Stability and Growth Pact preventive arm procedure YES MTO met ex post? NO Determine require adjustment Change in structural balance according to “matrix” 1 and Consider exceptions (e.g. in the case of structural reforms) Assess compliance using relevant indicators No YES I. Change in structural balance appropriate? 2 procedure and launched II. Expenditure growth appropriate? NO Assess whether deviation is significant NO Deviation quantitatively significant for I or II? 3 and Qualitative overall assessment negative YES Launch significant procedure in particular with recommendation for nudger adjustment 4 Asses whether “effective action” has been taken 5 (on basis of various indicators and methods) NO Sanction: 6 Deposit, Interest-bearing Source: Adapted from Deutsche Bundesbank Monthly Report June 2017. mended fiscal adjustment path consistent with the the net primary government expenditure — when MTO in 2019.12 Poland was recommended in July corrected for one-off expenses — grew 9.8 percent 2019 to ensure that the nominal growth rate of net in 2019 (European Commission 2020). In June 2020, primary government expenditure does not exceed however, as the COVID-19 virus was spreading rap- 4.4 percent in 2020, to take further steps to improve idly, the Council of the European Union pivoted its the efficiency of public spending; and to ensure the focus and emphasized the need to effectively address sustainability of the pension system (Council of the the COVID-19 crisis, sustain the economy, and sup- European Union 2019). As assessed by the European port the near-term recovery — in line with the gen- Commission in 2020, Poland did not comply with eral escape clause of the Stability and Growth Pact the recommended cap of expenditure growth, as (Council of the European Union 2020). Fiscal assessment 13 Due to the expected severe impact of the COVID-19 getary constraints. Furthermore, the full flexibil- pandemic, the Stability and Growth Pact general ity foreseen under EU State Aid rules was afforded, escape clause was activated in March 2020. This particularly through the adoption of the Temporary allows for a temporary departure from the recom- Framework in March 2020 and subsequent exten- mended fiscal adjustment path towards the MTO, sions, to allow increased public support to research, provided it does not endanger fiscal sustainability testing and production of products relevant to fight in the medium term. The general escape clause does the COVID-19, to protect jobs and to further support not suspend the procedures of the SGP, but rather the economy, to provide targeted public support to allows the member state to depart from the bud- otherwise viable companies that have entered into getary requirements. The general escape clause has financial difficulty as a result of the coronavirus remained active in 2021 – 22. Meanwhile, the national outbreak, and allow recapitalization and subordi- escape clause was also triggered to suspend bud- nated debt measures. Box 1.3 Stability and Growth Pact fiscal rules and national fiscal rules in Poland Poland’s fiscal framework is based on a set of rules in- a structural deficit of 1 percent of GDP — slightly higher cluded in the national Public Fiance Act, the Constitution, than the 0.5 percent of GDP threshold used as a bench- and the EU Stability and Growth Pact: mark for other EU member states. A debt rule at the public sector level; The national fiscal rules consist of a 60 percent of GDP ratio ceilling for public debt (local definition that deviates A stabilizing expenditure rule (Public Finance Act) set- from EDP debt) inscribed in the Constitution of Poland and ting an upper limit for the state budget expenditure set a safety threshold of 55 percent of GDP as inscribed in for the next year (covering approximately 90 percent of Poland’s Public Finance Act, both of which form a strong the general government expenditures); framework for the conduct of sustainable fiscal policy. The Individual debt limits at the level of local governments; and national limit on the growth of public expenditures — or the Stabilizing Expenditure Rule, in force since 2015 — is Abiding by the reference values for the general govern- aligned with the MTO and provides further guidance on ment nominal deficit (3 percent of GDP) and the gen- the sound management of public finances. The national eral government debt (60 percent of GDP), and attain- expenditure rule was amended in December 2015, in- ing the MTO. ter alia replacing actual inflation with the inflation target The agreements on the monetary union have established followed by the National Bank of Poland (2.5 percent) in the fiscal rules, with the Maastricht Treaty of 1992 intro- the formula to calculate expenditures. Given the lower in- ducing the requirement to comply with government deficit flation economic environment, the amendment de fac- and debt ratios of 3 percent and 60 percent of GDP, re- to allowed higher expenditures than under the previous spectively, as conditions for participation. The 1997 the version of the rule. The ammendment also allowed for Stability and Growth Pact (SGP) introduced additional increased expenditure in the event of one-off and tem- requirements, which have been modified and amended porary revenue measures. over time. In 2012, the fiscal compact was adopted, in- The Public Finance Act (Article 182) requires an ex post cluding the decision that the SGP requirements were, in assessment of compliance with the rule to be included in principle, to be anchored in national law with provisions the report on the execution of the budget law that is sub- of binding force and a permanent character. (See Treaty mitted to the Parliament and the Supreme Audit Office on Stability, Coordination and Governance in the EMU, by May 31st and discussed in the Supreme Audit Office Title III, Fiscal Compact.) report on budget implementation. As an EU member, Poland is required to adhere to the The Stabilizing Expenditure Rule, however, excludes EU EU Stability and Growth Pact. Under this framework, funds, which is an important source of revenue for the Poland must comply with rules limiting the general gov- government and is expected to grow over the next few ernment deficit and debt at 3 percent and 60 percent years as part of the EU Multiannual Financial Framework of GDP, respectively. Furthermore, Poland has agreed (MFF) 2021 – 27 and the Next Generation EU (NGEU) pro- with the European Commission on an MTO, targeting gram supporting post-pandemic recovery. 14 Poland Public Finance Review FISCAL POLICY Although fiscal policy presented a drag on growth in CYCLICALITY Poland, authorities partly mitigated the adverse im- pact on household consumption by expanding social protection, including a benefit for families with chil- Fiscal policy was largely procyclical dren through the Family 500+ program. In ECA, the in the decade leading up to the stance was similar but reflected the need for procy- COVID-19 pandemic clical fiscal tightening in the region’s energy export- ers amid the oil price plunge of 2014 – 16 (Figure 1.19.c; Poland did not manage to “graduate” from pro- Stocker et al. 2018; Wheeler et al. 2020). cyclical fiscal policy, but its stance largely mir- rored that of the euro area. Several economies in Fiscal policy pivoted from procyclical tightening Europe, including Poland, were unable to escape pro- to procyclical expansion by 2019 as the economic cyclicality in the decade prior to the COVID-19 cri- recovery gathered momentum in the period lead- sis. For Poland, much like other EMDEs, procyclical ing up to the pandemic—aided by a strong rebound fiscal policy is a longstanding issue. Of the EMDEs in investment amid fresh EU funding. Rising ca- that graduated from procyclicality, stronger institu- pacity constraints and core inflation suggest several tions—including robust legal and investment frame- Central European economies, including Poland, were works, minimal corruption, and reliable governance— expanding at a pace above potential output (Figure appeared to be a factor (Frankel, Vegh, and Vuletin 1.19.d). Nevertheless, the shift toward expansionary 2013). In Poland, although fiscal frameworks have procyclical fiscal policy was also observed in the euro strengthened over the past decade, several indica- area. In ECA, however, fiscal policy turned decided- tors of governance—including accountability, polit- ly countercyclical as macroeconomic policy frame- ical stability, rule of law—have eroded, despite the works strengthened in some large economies such ECA average either remaining stable or making im- as the Russian Federation. provements in these areas.13 The dampening effect of procyclical fiscal tight- Fiscal policy amplified, rather than smoothed, ening was partly offset by sizable EU structural business cycles—expanding during economic funds in the decade leading up to the pandemic. upturns and contracting during economic down- Despite substantial external headwinds—including turns.14 Although usually procyclical fiscal policy is the European debt crisis and escalating geopoliti- more characteristic of EMDE fiscal policy, the euro cal and trade tensions—underlying drivers of long- area was also forced to conduct severe procyclical term growth strengthened in Poland over 2010 – 19. fiscal consolidation in the aftermath of the GFC and Accelerating investment—helped by EU funding and European debt crisis (Figure 1.19.a and Figure 1.19.b; inward FDI—further deepened global value chain Riera-Crichton, Vegh, Vuletin 2016; Cugnasca and integration and corresponded to a period of rapid Rother 2015). TFP and labor productivity growth, sharply contrast- ing the productivity slowdown in the euro area and Following procyclical fiscal consolidation in the ECA (Figure 1.19.e; Dieppe 2020). Per capita income wake of the European debt crisis, temporarily convergence with the euro area gained momentum waning EU structural funds and still-tight fiscal and debt dynamics improved on the back of robust policy weighed on growth in Poland over 2015 – 16. growth (Kindberg-Hanlon and Okou 2020). Fiscal assessment 15 Figure 1.19 Output gaps and fiscal impulses by country A. Output gap and fiscal impulse in Poland B. Output gap and fiscal impulse in the euro area 4 2 2 2 Percent of potential GDP Percent of potential GDP Percentage points Percentage points 2 1 1 1 0 0 0 0 -2 -1 -1 -1 -4 -2 -2 -2 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Fiscal impulse (RHS) Output gap Fiscal impulse (RHS) Output gap C. Output gap and fiscal impulse in ECA D. Core inflation and capacity utilization 2 4 in Central Europe and Poland 6 90 Percent, year-on-year Percent of potential GDP 1 2 4 85 Percentage points Percent 2 80 0 0 0 75 -2 70 -1 -2 2015Q1 2015Q3 2016Q1 2016Q3 2017Q1 2017Q3 2018Q1 2018Q3 2019Q1 2019Q3 2020Q1 2020Q3 2021Q1 -2 -4 Core inflation in CE 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Core inflation in Poland Capacity utilization in CE (RHS) Fiscal impulse (RHS) Output gap Capacity utilization in Poland (RHS) E. Poland labor productivity growth composition 10 10 5 5 Percentage points 0 0 Percent -5 -5 -10 -10 -15 -15 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Capital contribution Human capital contribution TFP Labor productivity (RHS) Sources: Haver Analytics; IMF; Penn World Tables; World Bank. Note: Estimated using data up to 2020Q4 CE = Central Europe, ECA = Europe and Central Asia. A.-B. Fiscal impulse defined as change in the cyclically adjusted primary balance (CAPB) from previous year. Decline in the CAPB indicates fiscal consolidation; increase in the CAPB indicates fiscal expansion. B.-D. Aggregates calculated using 2019 real U.S. dollar GDP weights at average 2010-19 prices and market exchange rates. D. Sample includes Hungary, Poland, and Romania. Last observation is 2021Q1. E. TFP = total factor productivity. Labor productivity defined as output per worker in 2017 U.S. dollars. 16 Poland Public Finance Review Box 1.4 Fiscal policy cyclicality assessment The assessment of fiscal cyclicality is sensitive to the which incorporates more economic information than measurement of output gaps. In order to assess the cy- univariate filters, is the technique used in this analysis. clicality of fiscal policy, an estimate of the output gap is needed to gauge the business-cycle phase. Measuring Output gap estimates vary widely for European economies, output gap at the national level is complex since it is including for Poland. Nevertheless, estimates are largely an unobserved variable. Nevertheless, output gaps can in the same direction — implying that the assessment of be estimated using a range of methods. Univariate fil- fiscal policy cyclicality is broadly consistent, regardless of ters — which typically decompose quarterly output series the underlying assumptions (Figures B.1.4.1 – 4). There into trend and cycle components — are often expanded are important exceptions, however, such as in 2017, for into multivariate filters that include inflation, unemploy- which the direction of cyclicality cannot be easily deter- ment, various financial indicators, and commodity pric- mined since output gap estimates range from -1.1 percent es. Filtering techniques distinguish short-run deviations to 1.1 percent of potential GDP for Poland and from -0.6 of output from trends, which is most relevant for the as- percent to 0.8 percent of potential GDP for the euro area. sessment of fiscal policy cyclicality. A multivariate filter, Figure B.1.4.1 Output gap estimates Figure B.1.4.2 Output gap estimates for Poland for the euro area 6 6 6 6 Percent of potential GDP Percent of potential GDP Percentage points 4 4 4 4 Percentage points 2 2 2 2 0 0 0 0 -2 -2 -2 -2 -4 -4 -4 -4 -6 -6 -6 -6 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Output gap Fiscal impulse (RHS) Output gap Fiscal impulse (RHS) Figure B.1.4.3 Estimated output gaps Figure B.1.4.4 Estimated output gaps in Poland in ECA 6 6 6 6 Percent of potential GDP Percent of potential GDP 4 4 4 Percentage points Percentage points 4 2 2 2 2 0 0 0 0 -2 -2 -2 -4 -4 -2 -4 -6 -6 -4 -6 2019 2020 2021 2022 2019 2020 2021 2022 Output gap Fiscal impulse (RHS) Output gap Fiscal impulse (RHS) Sources: IMF; World Bank Note: ECA = Europe and Central Asia. B.1.4.1 – 4. Shaded area indicates forecasts. Fiscal impulse defined as change in the cyclically adjusted primary balance (CAPB) from previous year. Decline in the CAPB indicates fiscal consolidation; increase in the CAPB indicates fiscal expansion. Fiscal impulse projections are International Monetary Fund calculations included in the April 2021 World Economic Outlook report. Error bars indicate confidence interval around one standard deviation. B.1.4.2, B.1.4.4 Aggregates are calculated using GDP weights at average 2010-19 prices and market exchange rates. Fiscal assessment 17 PUBLIC DEBT Figure 1.21 Poland debt ratio is below SUSTAINABILITY EU average in 2019 200 Percent of GDP Recent debt dynamics 150 100 Poland’s general government debt level has been be- low the EU Growth and Stability Pact norms and be- 50 low the EU and ECA averages. Large deficits averag- 0 ing 5.5 percent of GDP contributed to the increase in the Estonia Bulgaria Czech Republic Sweden Romania Lithuania Latvia Malta Poland Slovak Republic Netherlands Ireland Finland Germany Hungary Slovenia Austria Croatia EU27 Cyprus Spain Belgium France Portugal Italy Greece Denmark Luxembourg debt-to-GDP ratio following the GFC and the euro area debt crisis, with debt rising 9.8 percentage points of GDP between 2008 and 2013. Weak growth and stalled revenue collections stemming from the euro area debt Source: Eurostat. crisis widened the deficit in 2013, subsequently push- ing debt up to 56.4 percent of GDP. Fiscal consolidation Strong growth accompanied by primary surplus- in the context of sustained growth and a one-off debt es averaging 0.7 percent of GDP over the 2017 – 2019 transfer in 2014 helped bring Poland’s general govern- period allowed further significant debt reduction ment debt down markedly (Figure 1.20). Following for- (Figure 1.22). The growth effect is estimated at 2.4 per- mal adoption of the Stabilizing Expenditure Rule in centage points over the 2017 – 2019 period, while the 2015, Poland successfully realigned government reve- inflation effect amounted to an additional 1 percent- nues with spending, which helped bring the fiscal defi- age point of GDP per year on average (Figure 1.23). cit within the EU limit of 3 percent and contributed Over this period there was a marked decline in gen- to downward pressure to public debt. These measures eral government debt of almost 9 percentage points, contributed to the sustainability of its public financ- from 54.6 percent of GDP in 2016, to 45.6 percent of es through a decrease in public debt, to 52 percent of GDP by 2019. External general government debt fell GDP on average over the 2010 – 19 period —above the below 20 percent of GDP by 2019. At end-2019, the ECA average of 44 percent, but lower than the EU av- public debt-to-GDP ratio stood at 45.6 percent, well erage of 71 percent (Figure 1.21). below the EU27 average of 77.6 percent of GDP. Figure 1.20 Poland debt relative to EU Figure 1.22 Debt dynamics and ECA averages 20 60 Percentage points, percent of GDP 70 15 50 60 10 40 Percent of GDP Percent 50 5 30 40 0 20 -5 10 30 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 -10 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 EU 2010–19 average Poland ECA 2010–19 average Deficit Contr. nominal GDP gr. Sources: IMF; World Bank. Stock-flow adjustments Change in gross debt Note: EU = euro area; ECA = Europe and Central Asia. Gross debt [RHS] Aggregates calculated as simple averages. Source: Debt Sustainability Monitor, EC. 18 Poland Public Finance Review Figure 1.23 Contributions to changes 2014 (Figure 1.24). Exchange rate risks have there- in debt-to-GDP ratio fore declined, and hedging can help mitigate the 15 residual risks. The share of debt held by non-res- idents remains relatively elevated at 41.4 percent Percentage points, percent of GDP 10 of total debt (18.9 percent of GDP), and it remains a source of vulnerability, although it has declined 5 from 57.1 percent of general government debt in 2014 (29 percent of GDP) (Figure 1.25). The coun- 0 try’s public debt management strategy assumes that the share of foreign currency debt in total -5 debt will remain at less than 30 percent and ex- ternal debt in total debt around 40 percent over -10 the medium term. 2013 2014 2015 2016 2017 2018 2019 2020 Primary balance Interest expenditure Figure 1.24 Composition of general Growth effect Inflation effect government debt XR effect—linked to ir Stock-flow adjustment 60 Changes in the ratio 50 Source: Debt Sustainability Monitor, EC. Percent of GDP 40 The general government debt increased from 30 45.6 percent of GDP in 2019 to 57.5 percent of 20 GDP in 2020. A large fiscal deficit of 7 percent of 10 GDP, the accumulation of financial assets, FX ef- fects,15 and the 2.7 percent decline in GDP were 0 among the factors contributing to this expansion 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 of debt in 2020. Off-budget funds were used in 2020 on a much larger scale than in the past. At the Denominated in local currency end of 2020, the value of liabilities due to the is- Denominated in foreign currency sue of own-debt securities by BGK (for the COVID Source: GUS. fund) and PFR (Polish Development Fund Financial Shield) amounted to 7 percent of GDP. In this way, Figure 1.25 Composition of general fiscal rules that refer to the local definition of pub- government debt lic debt were circumvented. Over 90 percent of the 60 general government debt is issued by the central 50 government, which according to the EU rules also Percent of GDP includes off-budget SPVs. 40 30 The debt composition is not a source of signifi- 20 cant risks. With an average maturity of outstand- 10 ing State Treasury debt estimated at 4.7 years, and the share of short-term government debt at around 0 1 percent of the total debt, the rollover and refi- 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 nancing risks are low. The share of debt denom- inated in foreign currency declined to 28.4 per- Domestic debt (held by residents) cent of general government debt by the end of 2019 External debt (held by non-residents) (12.9 percent of GDP), from 35.3 percent of debt in Source: GUS. Fiscal assessment 19 Public debt is sustainable Figure 1.27 Projected fiscal balance amid planned fiscal adjustment and government debt in Poland 60 Public debt is expected to stabilize at higher levels than in the pre-COVID period. The envisioned pace 55 of economic recovery and fiscal adjustment in Poland 50 will not be sufficient to return fiscal deficits or gov- Percent of GDP ernment debt to pre-pandemic projections over the 45 medium term (Figure 1.26). Nevertheless, public debt in Poland is projected to stabilize at a sustainable rate, partly thanks to wide fiscal space prior to the pandem- 5 ic and relatively benign external financing conditions, 0 as well as to the size of the planned fiscal adjustment (IMF 2021). Fiscal sustainability gap estimates, how- -5 ever, are sensitive to sharp reassessments of growth Fiscal balance Government debt or sudden shifts in financial market conditions. 2019 2025 Sources: IMF; World Bank. Figure 1.26 Fan chart evolution of Note: 2025 projections are IMF calculations included debt-to-GDP ratio in the April 2021 World Economic Outlook report. 70 Figure 1.28 Cyclically adjusted 65 primary balances 60 0 Percent of GDP 55 50 -1 45 Percent of potential GDP 40 -2 35 -3 30 2019 2020 2021 2022 2023 2024 2025 2026 -4 10th–25th 25th–50th 50th–75th 75 –90 th th Baseline -5 Source: World Bank DSA calculations. -6 2020 2021 2022 2023 2024 2025 2026 Fiscal adjustment needs are relatively modest in Poland. A baseline estimate of the fiscal sustainabil- Poland ECA Euro area ity gap—assuming current benign financing condi- Sources: IMF; World Bank. tions and GDP growth projections—suggests Poland Note: Shaded area indicates projections. Aggregates calculated using GDP weights at average 2010-19 prices and market exchange needs to reduce its primary deficit by 1.2 percentage rates. Projections are IMF calculations included in the April 2021 points of GDP to stabilize its gross government debt as World Economic Outlook report. of 2021.16 The size of the adjustment needed in Poland is modest relative to the ECA average, as well as to Public debt to GDP ratio other Central European economies and the euro area is sensitive to interest payments (Figure 1.27). Moreover, projections for both the pri- and growth assumptions mary balance and cyclically adjusted primary balance suggest that the planned pace of fiscal adjustment is A sudden reassessment of growth or shift in mar- swift enough to quickly stabilize debt, albeit at high- ket conditions could substantially increase fiscal er levels relative to pre-pandemic trends (Figure 1.28). adjustment needs. Fiscal sustainability gap estimates 20 Poland Public Finance Review are sensitive to sharp reassessments of growth or For instance, one standard deviation below median sudden shifts in financial market conditions. Worse- growth and above the median nominal interest rate than-expected growth or tighter-than-anticipated could trigger a substantial rise in interest payments financing conditions—triggered perhaps by an in- as a share of GDP, which would require a primary bal- tensification of the pandemic or a sudden shift in in- ance adjustment of 9.8 percentage points of GDP to vestor sentiment—could result in far higher adjust- stabilize debt in Poland—larger than the average for ment needs than projected in the baseline scenario. other economies (Figures 1.29 – 1.30). Figure 1.29 Fiscal sustainability gaps Figure 1.30 Fiscal sustainability gaps in 2021 in 2021, by condition Poland Other CE Euro area Poland Euro area Other CE 0 0 -2 Percent of GDP -1 -4 Percent of GDP -6 -2 -8 -10 -3 -12 Stressed conditions Current conditions Sources: Kose et al. (2017); World Bank. -4 Note: CE = Central Europe. Aggregates calculated using GDP weights Sources: Kose et al. (2017); World Bank. at average 2010-19 prices and market exchange rates. Error bars Note: CE = Central Europe. Aggregates calculated using GDP weights denote interquartile range. Under stressed conditions, growth is one- at average 2010-19 prices and market exchange rates. Error bars standard deviation below and interest rates one-standard deviation denote interquartile range. above the country-specific median. Table 1.1 Debt sustainability analysis Type of shock Magnitude of the shock Impact Primary balance shock Deterioration in the primary balance by Public debt would increase to 56.9 percent 1 p.p. in 2022 and 2023. in 2026, 2.8 p.p. higher than in the baseline. Gross financing needs would be proportional- ly larger in 2022 and 2023. Growth shock A decline in GDP growth by about 2.5 p.p. in Public debt would be at 60.6 percent of GDP both 2022 and 2023 relative to the base- in 2026, 6.5 p.p. higher than in the baseline. line, along with a parallel decline in inflation In all years gross financing needs would be and a deterioration in the primary balance. higher. Interest rate shock A permanent 500 bps increase in the nomi- Public debt stabilizes at around 58 percent of nal interest rate beginning in 2022 GDP. Elevated financing needs in 2024-2026. Exchange rate shock A 22 percent nominal exchange rate depre- Practically no impact on either the public debt ciation in 2022, calibrated to emulate the over the medium term or on gross financing maximum historical movement of the ex- needs. This resilience reflects the predomi- change rate over the last 10 years. nance of public debt in local currency. Combined shock A combination of lower GDP growth, a larger Debt/GDP ratio stabilizes above 68 percent primary deficit, higher interest rates, and an of GDP. exchange rate shock. Gross financing needs remain significantly higher than in the baseline scenario. Fiscal assessment 21 FISCAL CONSOLIDATION Fiscal multiplier estimates for Poland suggest fiscal policy has a significant impact on output. Using the local projections model as specified over a Fiscal adjustment imposes output two-year horizon, output losses from fiscal consoli- costs that vary with the policy mix dation are estimated to be particularly pronounced and business-cycle phase in ECA (Figure 1.31). There is evidence, however, that the size of output losses can depend on the composi- Fiscal consolidation is associated with short-term tion of the fiscal adjustment; in OECD countries, for output losses, but the magnitude depends on busi- instance, spending-based fiscal adjustments—such ness-cycle positions and the policy mix. For many as from the expiration of fiscal support measures— EMDEs, fiscal adjustment is often needed to ensure ac- are estimated to generate smaller declines in output cess to market financing and to restore medium-term relative to tax-based fiscal adjustment (Alesina and debt sustainability. There is broad consensus, howev- Ardagna 2013; Alesina, Favero, and Giavazzi 2015). er, that fiscal consolidation is contractionary in the short term.17 Moreover, the business cycle can ampli- Fiscal spending multipliers are sensitive to the fy the output losses from fiscal adjustment, as fiscal type of expenditures and business-cycle phase. multipliers tend to be larger during recessions (Riera- Government spending multipliers for Poland are Crichton, Vegh, Vuletin 2016; Jordà and Taylor 2015). estimated using a local projection model and range Figure 1.31 Output loss two years after fiscal Figure 1.33 Capital expenditure multipliers consolidation 1.6 fiscal multiplier 1.0 1.2 * Cumulative 0.8 * * 0.5 Percentage points * 0.4 * 0 * 0 -0.5 Euro area EU Spain Poland Romania Hungary -1.0 On impact After 2 years Sources: David and Leigh (2018); Eurostat; World Bank. -1.5 LAC ECA SAR EAP SSA MNA Notes: “*” indicate significance at 10% significance level; EU and euro area refers to aggregates for the European Union and euro Sources: David and Leigh (2018); Eurostat; World Bank. area, respectively. Figure 1.32 Total expenditure multipliers Figure 1.34 Fiscal multipliers in Poland during recessions and expansions 0.6 * Cumulative fiscal multiplier * 2.5 Long run cumulative 0.5 * 2.0 * fiscal multiplier 0.4 * * 1.5 0.3 1.0 * * * * * 0.5 * 0.2 * 0 0.1 * -0.5 -1.0 0 Output Unemployment Output Unemployment -0.1 Total spending Capital spending Euro area EU Spain Poland Romania Hungary Expansions Recessions On impact After 2 years Sources: David and Leigh (2018); Eurostat; World Bank. Notes: “*” indicate significance at 10% significance level; EU and Sources: David and Leigh (2018); Eurostat; World Bank. euro area refers to aggregates for the European Union and euro Notes: “*” indicate significance at 10% significance level; EU and area, respectively. euro area refers to aggregates for the European Union and euro Bars show cumulative fiscal multipliers 2 years after the fiscal shock. area, respectively. Output refers to expansions/recessions identified as real GDP above/ Bars are cumulative output losses following fiscal adjustment. below its HP trend; Unemployment refers to expansions/recessions Lines indicate 95 percent confidence intervals. identified as the unemployment rate below/above its HP trend. 22 Poland Public Finance Review from 0.3 in the short term to 0.5 in the long term— Generation EU funds (Figure 1.33). Consistent with on par with other Central European economies the literature, both measures of spending multi- and larger than in advanced EU countries (Figure pliers for Poland are larger during recessions, with 1.32; Table 1.2). Capital spending multipliers, how- capital expenditure multipliers being particular- ever, are nearly twice as large as spending multi- ly sizeable during periods of weak labor markets pliers for total government expenditures—hav- when the unemployment rate is above its trend ing important implications for the upcoming Next level (Figure 1.34; Table 1.3).18 Table 1.2 Expenditure multipliers Expenditure multipliers total expenditure 1 quarter 2 quarters 3 quarters 1 year 2 years 3 years 0.256* 0.286* 0.350* 0.397* 0.543* 0.390* Poland (0.12) (0.12) (0.12) (0.13) (0.14) (0.12) 0.018 0.158 0.236 0.275 0.423* 0.174 EU (0.23) (0.25) (0.27) (0.27) (0.25) (0.14) 0.333* 0.328* 0.321* 0.324* 0.611* 0.626* Hungary (0.10) (0.11) (0.11) (0.12) (0.19) (0.21) 0.310 0.452* 0.481* 0.446* 0.567* 0.362* Romania (0.19) (0.24) (0.25) (0.23) (0.12) (0.10) 0.496* 0.463* 0.408* 0.379* 0.306* -0.125 Croatia (0.18) (0.15) (0.13) (0.13) (0.09) (0.10) -0.003 0.165 0.271 0.328 0.452* 0.187 Euro Area (0.24) (0.25) (0.27) (0.27) (0.27) (0.14) 0.275 0.413 0.480 0.481 0.945* 0.541* Italy (0.24) (0.30) (0.34) (0.33) (0.32) (0.15) 0.074 0.093 0.119 0.128 0.159* 0.152* Spain (0.10) (0.09) (0.09) (0.09) (0.07) (0.04) 0.033 0.020 0.057 0.114 0.170 0.106 France (0.18) (0.17) (0.17) (0.16) (0.14) (0.08) 0.101 0.116 0.118* 0.127* 0.178* 0.091 Greece (0.06) (0.07) (0.07) (0.07) (0.10) (0.09) Expenditure multipliers capital expenditure 1 quarter 2 quarters 3 quarters 1 year 2 years 3 years 0.921* 0.907* 0.933* 0.916* 0.918* 0.756* Poland (0.19) (0.19) (0.20) (0.21) (0.23) (0.25) 0.405 0.617 0.708 0.727 0.842 0.444 EU (0.90) (0.84) (0.83) (0.81) (0.88) (0.67) 0.366* 0.474* 0.580* 0.663* 1.418* 1.748* Hungary (0.16) (0.19) (0.22) (0.25) (0.42) (0.94) 0.498 0.520 0.453 0.460 0.707* 0.366* Romania (0.35) (0.40) (0.40) (0.38) (0.25) (0.19) 0.575* 0.638* 0.644* 0.640* 0.493* 0.119 Croatia (0.24) (0.25) (0.23) (0.22) (0.25) (0.21) 0.130 0.374 0.453 0.505 0.451 0.230 Euro Area (0.96) (0.88) (0.86) (0.86) (0.86) (0.62) -0.490 -0.441 -0.376 -0.370 -0.074 0.431 Italy (0.38) (0.44) (0.49) (0.51) (0.63) (0.71) 0.168 0.253 0.389 0.482* 0.608* 0.590* Spain (0.31) (0.27) (0.25) (0.22) (0.16) (0.14) 0.211 0.216 0.262 0.388 0.777 0.432 France (0.69) (0.65) (0.64) (0.65) (0.49) (0.39) 0.758* 0.787* 0.698* 0.860* 1.722* 1.137* Greece (0.37) (0.39) (0.39) (0.43) (0.53) (0.62) Standard errors in parentheses * p<0.1 Fiscal assessment 23 Table 1.3 State-contingent expenditure multipliers total expenditure 1 quarter 2 quarters 3 quarters 1 year 2 years 3 years 0.279* 0.352* 0.452* 0.533* 0.658* 0.533* Expansions (0.16) (0.16) (0.17) (0.16) (0.14) (0.12) Poland (1) 0.383* 0.436* 0.478* 0.576* 1.087* 0.561* Recessions (0.20) (0.18) (0.19) (0.21) (0.32) (0.32) -0.463 -0.488 -0.505 -0.512 -0.318 -0.361 Expansions (0.30) (0.35) (0.42) (0.51) (0.31) (0.22) Poland (2) 0.368* 0.450* 0.566* 0.649* 0.980* 0.687* Recessions (0.11) (0.12) (0.14) (0.17) (0.27) (0.26) 0.358* 0.362* 0.332* 0.292* 0.301 0.071 Expansions (0.14) (0.15) (0.16) (0.17) (0.41) (0.36) Hungary (1) 0.377* 0.466* 0.572* 0.726* 0.808* 1.418* Recessions (0.17) (0.16) (0.17) (0.19) (0.16) (0.65) 0.461* 0.430* 0.385* 0.356 0.311 0.095 Expansions (0.20) (0.20) (0.20) (0.22) (0.35) (0.23) Hungary (2) 0.399* 0.449* 0.477* 0.476* 0.809* 0.940* Recessions (0.15) (0.17) (0.20) (0.23) (0.24) (0.32) 0.141 0.174 0.167 0.169 0.928* 0.484* Expansions (0.22) (0.33) (0.42) (0.41) (0.40) (0.20) Italy (1) -0.087 0.219 0.443 0.670* 1.038* 0.547* Recessions (0.36) (0.35) (0.31) (0.33) (0.32) (0.22) -0.180 -0.241 -0.229 -0.154 0.568* 0.490* Expansions (0.26) (0.38) (0.44) (0.40) (0.33) (0.16) Italy (2) 0.852* 1.074* 1.024* 1.067* 1.379* 0.641* Recessions (0.37) (0.41) (0.36) (0.38) (0.47) (0.30) 0.035 -0.011 0.047 0.040 0.070 0.095 Expansions (0.17) (0.15) (0.13) (0.11) (0.09) (0.07) Spain (1) 0.157 0.204 0.234* 0.257* 0.277* 0.181* Recessions (0.12) (0.13) (0.14) (0.15) (0.15) (0.09) 0.631* 0.468* 0.436* 0.368* 0.177* 0.0893* Expansions (0.16) (0.17) (0.16) (0.14) (0.09) (0.05) Spain (2) 0.052 0.092 0.125 0.148 0.210* 0.178* Recessions (0.12) (0.12) (0.12) (0.13) (0.11) (0.07) Note: Standard errors in parentheses * p<0.1 (1) recessions are identified using GDP deviations from trends estimated with the HP filter; (2) recessions identified when the unemployment rates exceed trend unemployment, estimated using the HP filter. 24 Poland Public Finance Review Table 1.4 State-contingent expenditure multipliers capital expenditure 1 quarter 2 quarters 3 quarters 1 year 2 years 3 years 1.002* 1.002* 1.022* 0.999* 0.838* 0.521* Expansions (0.27) (0.26) (0.29) (0.30) (0.30) (0.26) Poland (1) 0.869* 0.802* 0.764* 0.806* 1.258* 1.061 Recessions (0.24) (0.26) (0.27) (0.26) (0.40) (0.72) -0.158 -0.681 -0.890* -0.858 -0.548 -0.276 Expansions (0.50) (0.55) (0.54) (0.52) (0.41) (0.21) Poland (2) 1.452* 1.543* 1.651* 1.536* 2.095* 1.558* Recessions (0.27) (0.25) (0.26) (0.24) (0.46) (0.59) 0.347 0.529* 0.646* 0.660 1.499* 2.692 Expansions (0.22) (0.28) (0.34) (0.41) (0.80) (3.98) Hungary (1) 0.627* 0.443 0.310 0.162 -0.443 -4.410 Recessions (0.27) (0.30) (0.34) (0.41) (0.98) (11.50) 0.516* 0.632* 0.740* 0.803* 1.045* 0.196 Expansions (0.26) (0.31) (0.39) (0.46) (0.59) (0.41) Hungary (2) 0.911* 1.133* 1.301* 1.324* 2.794* 5.918 Recessions (0.44) (0.45) (0.45) (0.44) (0.72) (6.14) 0.438* 0.491* 0.534* 0.558* 0.183 -0.322 Expansions (0.26) (0.25) (0.24) (0.27) (0.44) (0.34) Croatia (1) 0.996* 1.006* 0.851* 0.663* 0.384* 0.269 Recessions (0.25) (0.20) (0.18) (0.15) (0.11) (0.17) 0.790* 0.918* 0.954* 0.971* 1.233* 0.710* Expansions (0.42) (0.40) (0.47) (0.43) (0.28) (0.16) Croatia (2) 0.681* 0.733* 0.694* 0.642* 0.469 -0.032 Recessions (0.28) (0.31) (0.27) (0.30) (0.32) (0.24) 0.494* 0.548* 0.670* 0.760* 0.705* 0.484* Expansions (0.22) (0.20) (0.20) (0.19) (0.16) (0.15) Spain (1) 0.448 0.552 0.669 0.875 1.400* 1.533* Recessions (0.83) (0.76) (0.71) (0.76) (0.82) (0.77) 0.794* 0.831* 0.915* 0.933* 0.705* 0.536* Expansions (0.22) (0.22) (0.27) (0.28) (0.16) (0.11) Spain (2) -0.552 -0.304 -0.207 -0.084 0.308 0.544* Recessions (0.51) (0.46) (0.42) (0.40) (0.34) (0.26) Note: Standard errors in parentheses * p<0.1 (1) recessions are identified using GDP deviations from trend estimated with the HP filter; (2) recessions identified when the unemployment rates exceed trend unemployment, estimated using the HP filter. Fiscal assessment 25 Poland can put in place policy Figure 1.35 Fiscal impulse over 2020-22 measures to bolster a robust 0 recovery -1 Percentage points The fiscal impulse is projected to weigh more heav- -2 ily on growth in Poland than in other European economies in the near term, reflecting differenc- -3 es in the pace of policy adjustment (Figure 1.35). In a downside scenario of economic growth, especial- -4 ly in light of the recent deterioration in pandemic -5 trends, policymakers could consider postponing fis- Poland Euro area ECA cal consolidation efforts to prevent additional out- Sources: IMF; World Bank. put losses (World Bank 2021a, 2021b; Guénette and Note: Fiscal impulse defined as change in the cyclically adjusted Yamazaki 2021). primary balance (CAPB) from previous year. Decline in the CAPB indicates fiscal consolidation; increase in the CAPB indicates fiscal expansion. Once recovery from the COVID-19 pandemic is on Figure 1.36 Spending consolidation solid footing, Poland needs to restore fiscal space composition in the context of additional expenditure pres- sures arising from the aging population and the 8 decarbonization process, while ensuring greater efficiency of its fiscal system. Poland has consider- 6 able social and infrastructure spending needs, which are critical in the context of its aging population and Percentage points 4 the country’s commitments to decarbonization. The government also needs to protect capital spending to 2 sustain medium-term growth. Revenue and expendi- ture policies must be consistent with the deficit reduc- 0 tion path mandated by the Stability and Growth Pact. -2 Although Poland has adequate fiscal space within its policy framework to confront adverse shocks -4 2020 2021 2022 2023 2024 in the near term, expenditures will eventually need to be realigned with revenues to replenish Interests Compensation buffers. Enhancing spending efficiency could help Social transfers Subsides and capital transf. ease budget pressures and yield growth dividends. GFCF Other (net) These efforts can be complemented by measures that Total improve domestic revenue mobilization, including Sources: IMF; World Bank. those that make the tax structure more efficient. Note: Poland Convergence Program 2021. Harnessing sizable EU funds. Poland benefits from tive. Although the funds are expected to be distrib- sizable EU funding, including the upcoming Next uted over 2022 – 23, limited near-term absorption is Generation EU (NGEU) recovery package—which likely to prolong the full execution of these funds. could amount to about 4.5 percent of 2020 GDP in To this end, bolstering administrative capacity could additional funding for Poland over the next five or improve absorption rates and accelerate investment. so years (IMF 2021). The boost to growth from the NGEU funding could help offset the drag from fiscal Bolstering the underlying drivers of long-term consolidation, especially given that fiscal spending growth. To support a robust, resilient, and sustain- multipliers are higher for capital spending (relative able recovery, policy makers could leverage spending to total spending) and when output gaps are nega- reductions with parallel structural reforms that lift po- 26 Poland Public Finance Review tential growth (Eyraud, Gaspar, and Poghosyan 2017; of ambitious, but not unprecedented, reforms target- Anderson, Hunt, and Snudden 2014). Some of the ad- ing physical and human capital investment, as well as verse effects of the pandemic on potential output over measures that strengthen governance and enhance the the next decade could be stemmed by a combination business climate (Kilic Celik, Kose, and Ohnsorge 2020). Notes 1. Poland GDP per capita, PPP (current international $) 15. Appreciation of the zloty value of debt denominated in reached USD26,862 in 2015, compared with USD38,216 foreign currency for the EU. 16. Similar to Kose et al. (2017), fiscal adjustment needed is C  ߝ N Jߝ , where b = stock of gross gen- calculated as: >             HߝC 2. The sharp increase in the deficit in 2009 was the result of the crisis but also to a fiscal stimulus of nearly 2 per- eral government debt (as percent of GDP), r = net gen- cent of GDP, comprising reductions in personal income eral government interest payments (as share of general tax, higher public investment, and social transfers. government gross debt stock), g = average annual GDP 3. The expenditure stabilization rule introduced in the growth during the past 10 years, and     N Jߝ = primary bal- Public Finance Act imposed a limit on expenditure as ance (percent of GDP) in the previous year. follows: E(t)=E(t-1)*(Y’+C)*CPI+ε; where the limit E(t) 17. A key challenge, however, is that estimation requires is linked to the previous year’s limit E(t-1) and trend the identification of exogenous fiscal adjustment ep- growth (Y’ is average GDP growth in 8 years with MoF isodes—namely, fiscal responses that are orthogonal projection for t-1 and t, corrected for a factor depend- to cyclical fluctuations in real activity. Fiscal consoli- ing on deficit and debt (C) and then multiplied by NBP dation episodes in this box are identified using the cy- inflation target (CPI), plus an adjustment parameter ε, clically-adjusted primary balance (CAPB) as a share of which depends on revenue measures. Currently it also potential GDP—a common identification approach. An depends on the cumulative deviation of the general gov- alternative approach, pioneered by Romer and Romer ernment deficit from MTO (+/- 6%). (2010), uses narrative history to identify fiscal adjust- 4. Act of 10 September 2015 on amending the Tax Ordi- ment episodes driven by policy actions related to long- nance Act and other acts (Journal of Laws item 1649 term fiscal sustainability considerations. Refer to Escol- as amended). ano et al. (2014) for a survey of the relevant literature. 5. Source: Ministry of Finance; European Commission, 18. State-dependent fiscal multipliers are estimated us- Study and Reports on the VAT Gap in the EU-28 Mem- ing two methods to identify periods of economic slack: ber States: 2017 Final Report. (1) a negative cumulative output gap exceeding 1 per- 6. Poland Convergence Program 2019. cent of GDP two quarters prior to fiscal expansion, and 7. CASE Report for the European Commission, Study and (2) the employment rate exceeding its trend level, esti- Reports on the VAT Gap in the EU-28 Member States: mated using the Hodrick-Prescott (HP) filter. Consistent 2018 Final Report; and estimates of the Ministry of Fi- with the literature, total and capital spending multipli- nance for 2018. ers are larger in recessions, with capital expenditure 8. Debt Sustainability Monitor, European Commission. multipliers being particularly large during the periods 9. GDP on a seasonally adjusted, calendar adjusted basis. of weak labor markets when the unemployment rate is 10. Covid-19 anti-crisis measures regularly updated at: above its trend level. imf.org/en/Topics/imf-and-covid19/ Policy-Responses-to-COVID-19#P. 11. The Public Finance Act does not consider PFR and BGK public sector entities. References 12. European Commission, Assessment of the 2020 Con- vergence Program for Poland, May 2020. Auerbach, A., and Y. Gorodnichenko. 2012. “Measuring 13. WGI https://info.worldbank.org/governance/wgi/ the Output Responses to Fiscal Policy.” American Econ. Home/Reports; Kaufmann, Kraay and Mastruzzi 2010. Journal: Economic Policy 4 (2): 1 – 27. 14. Throughout this figure, procyclical fiscal policy occurs ———. 2013. “Fiscal Multipliers in Recession and Expansion.” when the fiscal impulse (measured as the negative an- Fiscal Policy after the Financial Crisis, edited by A. Alesina nual difference between the cyclically adjusted primary and F. Giavazzi, 63 – 98. Chicago: University Chicago Press. balance as a percent of potential GDP) and output gaps Reena Badiani-Magnusson, Karolina Goraus-Tańska. are either both positive or both negative. Countercycli- 2020. Redistributive effects of taxes and transfers in cal fiscal policy occurs when the fiscal impulse is positive Poland. Background paper for Public Finance Review (negative) when the output gap is negative (positive). © World Bank. Fiscal assessment 27 Blanchard, O., and R. Perotti. 2002. “An Empirical Charac- Escolano, J., L. Jaramillo, C. Mulas-Granados, and G. Terrier. terization of the Dynamic Effects of Changes in Gov- 2014. “How Much Is a Lot? Historical Evidence on the ernment Spending and Taxes on Output.” Quarterly Size of Fiscal Adjustments.” IMF Working Paper 14/179. Journal of Economics 117 (4): 1329 – 68. Washington, DC: International Monetary Fund. Guenette, J.D., and T. 2021. “Projecting the Economic Con- European Commission. 2018. Assessment of the 2018 Conver- sequences of the COVID-19 Pandemic.” Policy Research gence Program for Poland. Brussels: European Commission. Working Paper 9589. Washington, DC: World Bank. European Commission. 2020. Assessment of the 2020 Conver- Jordà, Ó. 2005. “Estimation and Inference of Impulse gence Program for Poland. Brussels: European Commission. Responses by Local Projections.” American Economic Eyraud, L., V. Gaspar, and T. Poghosyan. 2017. “Fiscal Politics Review, 95 (1): 161 – 82. in the Euro Area.” IMF Working Paper 17/18. Washington, Plagborg-Møller, M., and C. Wolf. 2021. “Local Projections DC: International Monetary Fund. and VARs Estimate the Same Impulse Responses.” Econo- Frankel, J. A., C. A. Vegh, and G. Vuletin. 2013. “On gradu- metrica 89 (2): 955 – 980. ation from fiscal procyclicality.” Journal of Development Alesina, A., C. Favero, and F. Giavazzi. 2015. “The output ef- Economics 100 (1): 32 – 47. fect of fiscal consolidation plans.” Journal of International Guénette, J.D., and T. Yamazaki. 2021. “Projecting the Economic Economics 96: S19 – S42. Consequences of the COVID-19 Pandemic.” Policy Research Alesina, A., and S. Ardagna. 2013. “The Design of Fiscal Working Paper 9589. Washington, DC: World Bank. Adjustments.” Tax Policy and the Economy 27 (1): 19 – 68. Huidrom, R., M. A. Kose, J. J. Lim, and F. L. Ohnsorge. 2016. Anderson, D., B. Hunt, and S. Snudden. 2014. “Fiscal consol- “Do Fiscal Multipliers Depend on Fiscal Positions?” Policy idation in the euro area: How much pain can structural Research Working Paper 7724. Washington, DC: World Bank. reforms ease?” Journal of Policy Modeling 36 (5): 785 – 799. ILO (International Labour Organization). 2021. ILO Monitor: Auerbach, A., and Y. Gorodnichenko. 2012. “Measuring COVID-19 and the World of Work, seventh edition. Geneva: the Output Responses to Fiscal Policy.” American Econ. International Labour Organization. Journal: Economic Policy 4 (2): 1 – 27. IMF (International Monetary Fund). 2021. “Republic of Auerbach, A., and Y. Gorodnichenko. 2013. “Fiscal Multipliers Poland: Staff Report for the 2020 Article IV Consultation.” in Recession and Expansion.” In Fiscal Policy after the IMF Country Report 21/35. Washington, DC: International Financial Crisis, edited by A. Alesina and F. Giavazzi, 63 – 98. Monetary Fund. Chicago: University Chicago Press. Jordà, Ó. 2005. “Estimation and Inference of Impulse Blanchard, O., and R. Perotti. 2002. “An Empirical Charac- Responses by Local Projections.” American Economic terization of the Dynamic Effects of Changes in Gov- Review, 95 (1): 161 – 82. ernment Spending and Taxes on Output.” Quarterly Jordà, Ò., and A. Taylor. 2015. “The Time for Austerity: Journal of Economics 117 (4): 1329 – 68. Estimating the Average Treatment Effect of Fiscal Council of the European Union. 2019. Recommendation for Policy.” Economic Journal 126: 219 – 55. a Council Recommendation on the 2019 National Reform Kaufmann, D., A. Kraay, and M. Mastruzzi. 2010. “The Program of Poland and delivering a Council opinion on the Worldwide Governance Indicators: Methodology and 2019 Convergence Program of Poland. Brussels: Council Analytical Issues.”. Policy Research Working Paper 5430. of the European Union. Washington, DC: World Bank. Council of the European Union. 2020. Recommendation for Kilic Celik, S., M. A. Kose, and F. Ohnsorge. 2020. “Subdued a Council Recommendation on the 2020 National Reform Potential Growth : Sources and Remedies.” Policy Research Program of Poland and delivering a Council opinion on the Working Paper 9177. Washington, DC: World Bank. 2020 Convergence Program of Poland. Brussels: Council Kindberg-Hanlon, G., and C. Okou. 2020. “Productivity of the European Union. Convergence : Is Anyone Catching Up?” Policy Research Cugnasca, A., and P. Rother. 2015. “Fiscal Multipliers During Working Paper 9378. Washington, DC: World Bank. Consolidation: Evidence from the European Union.” Vox Kose, M. Ayhan, Sergio Kurlat, Franziska Ohnsorge, and CEPR Policy Portal. December 5, 2015. Naotaka Sugawara. 2017. “A Cross-Country Database https://voxeu.org/article/fiscal-multipliers-during- of Fiscal Space.” Policy Research Working Paper 8157. consolidation-evidence-european-union. Washington, DC: World Bank. Deutsche Bundesbank 2017. Monthly Report June 2017. Plagborg-Møller, M., and C. Wolf. 2021. “Local Projections Dieppe, A., ed. 2020. Global Productivity: Trends, Drivers, and and VARs Estimate the Same Impulse Responses.” Policies. World Bank: Washington, DC. Econometrica 89 (2): 955 – 980. Dieppe, A., B. van Roye, and R. Legrand. 2018. The BEAR Ramey, V., and S. Zubairy. 2018. “Government Spending Mul- toolbox. https://www.ecb.europa.eu/pub/research/ tipliers in Good Times and in Bad: Evidence from US His- working-papers/html/bear-toolbox.en.html torical Data.” Journal of Political Economy 126 (2): 850 – 901. 28 Poland Public Finance Review Riera-Crichton, D., C. A. Vegh, and G. Vuletin. 2015. “Procy- Upturn, but for How Long? January. Washington, DC: clical and countercyclical fiscal multipliers: Evidence World Bank. from OECD countries.” Journal of International Money World Bank. 2021a. Europe and Central Asia Economic Update: and Finance 52: 15 – 31. Data, Digitalization, and Governance. March. Washington, Stocker, M., J. Baffes, Y. Some, D. Vorisek, and C. M. Wheeler. DC: World Bank. 2018. “The 2014 – 16 Oil Price Collapse in Retrospect: World Bank. 2021b. Global Economic Prospects, January 2021. Sources and Implications.” Policy Research Working Paper Washington, DC: World Bank. 8419. Washington, DC: World Bank. Ramey, V., and S. Zubairy. 2018. “Government Spending Wheeler, C. M., J. Baffes, A. Kabundi, G. Kindberg- Hanlon, Multipliers in Good Times and in Bad: Evidence from P. S. Nagle, and F. Ohnsorge. 2020. “Adding Fuel to the US Historical Data,” Journal of Political Economy 126 (2): Fire: Cheap Oil During the COVID-19 Pandemic.” Policy 850 – 901. Research Working Paper 9320. Washington, DC: World Bank. Poland National Reform Programs 2015 – 2019. World Bank. 2018. Global Economic Prospects: Broad-Based Poland Convergence Programs 2015 – 2019. Fiscal assessment 29 ANNEX 1.1 KEY ECONOMIC INDICATORS Table A.1 1 Key economic indicators and outlook Annual percent change unless indicated otherwise 2018 2019 2020 e 2021 f 2022 f 2023 f Real GDP growth, at constant market prices 5.4 4.7 -2.7 3.8 4.5 3.9 Private Consumption 4.5 3.9 -3.1 4.0 4.1 3.3 Government Consumption 3.5 6.5 4.4 3.5 3.4 4.6 Gross Fixed Capital Investment 9.4 6.1 -9.6 6.0 7.1 8.8 Exports, Goods and Services 6.9 5.2 -0.2 5.4 5.9 5.5 Imports, Goods and Services 7.4 3.0 -1.9 6.3 6.1 6.8 Real GDP growth, at constant factor prices 5.3 4.6 -2.8 3.8 4.4 3.9 Agriculture -9.1 -0.8 -3.0 4.4 1.0 1.0 Industry 7.0 2.1 -1.0 3.1 3.1 2.8 Services 5.0 6.1 -3.7 4.2 5.3 4.5 Inflation (Consumer Price Index) 1.6 2.3 3.4 3.4 3.0 3.0 Current Account Balance (% of GDP) -1.3 0.5 3.5 2.1 1.6 0.5 Net Foreign Direct Investment (% of GDP) -2.6 -1.6 -1.4 -1.2 -1.0 -1.0 Fiscal Balance (% of GDP) -0.2 -0.7 -7.0 -6.8 -4.3 -3.7 Debt (% of GDP) 48.8 45.6 57.5 59.5 59.0 58.4 Primary Balance (% of GDP) 1.2 0.7 -5.7 -5.7 -3.0 -2.5 Source: World Bank. Notes: e= estímate; f= forecast. The cutoff date for the data used in this report was June 2, 2021. 30 Poland Public Finance Review Table A.1 2 General Government Finances Percent of GDP 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Revenues 38.4 39.1 39.4 38.8 39.0 39.1 38.7 39.8 41.3 41.1 41.7 Taxes 20.4 20.6 20.0 19.6 19.8 19.9 20.5 21.1 21.9 21.7 22.0 PIT 4.3 4.3 4.5 4.5 4.6 4.7 4.8 5.0 5.3 5.3 n.a CIT 1.9 2.0 2.1 1.8 1.8 1.8 1.8 1.9 2.1 2.2 n.a. VAT 7.6 7.8 7.2 7.1 7.2 7.0 7.2 7.8 8.1 7.9 8.0 Excise 0.2 0.2 0.3 0.3 0.4 0.5 0.4 0.3 0.2 0.2 n.a. Other taxes 6.3 6.2 6.1 5.9 5.8 5.9 6.2 6.2 6.1 6.1 n.a. Social contributions 11.8 12.2 13.1 13.4 13.3 13.5 13.8 13.9 14.1 14.2 14.7 Other revenues 6.2 6.4 6.2 5.7 5.9 5.7 4.4 4.8 5.3 5.2 5.0 Capital revenues 1.3 1.7 1.3 1.0 1.2 1.1 0.6 0.7 1.2 1.1 1.1 Sales of goods and services 2.7 2.5 2.5 2.5 2.6 2.5 2.4 2.4 2.4 2.3 2.1 Other current revenue 2.2 2.2 2.4 2.2 2.2 2.1 1.4 1.7 1.8 1.8 1.8 Expenditure 45.8 44.1 43.1 43.0 42.6 41.7 41.1 41.3 41.5 41.8 48.7 Expense 40.1 38.2 38.5 39.0 38.2 37.3 37.9 37.5 36.9 37.5 44.1 Compensation of employees 11.1 10.7 10.5 10.6 10.6 10.4 10.4 10.2 10.1 10.3 10.9 Use of goods and services 6.3 5.7 5.7 5.7 5.7 5.7 5.5 5.5 5.7 5.7 5.9 Interest 2.5 2.5 2.7 2.5 2.0 1.8 1.7 1.6 1.4 1.4 1.3 Subsidies 0.9 0.8 0.7 0.7 0.7 0.5 0.5 0.5 0.4 0.5 3.8 Social benefits 16.5 15.7 15.9 16.5 16.4 16.2 17.1 17.0 16.7 17.2 18.7 Other expense 2.8 2.9 3.0 3.0 2.9 2.8 2.6 2.7 2.5 2.4 3.5 Other current expenditure 2.2 2.2 2.3 2.4 2.1 2.2 2.0 1.9 2.1 2.0 2.3 Capital transfers 0.6 0.7 0.7 0.6 0.8 0.6 0.6 0.9 0.4 0.4 1.2 Net acquisition 5.7 5.8 4.6 4.0 4.4 4.4 3.2 3.8 4.7 4.3 4.5 of nonfinancial assets Gross operating balance -1.7 0.9 0.8 -0.3 0.7 1.8 0.8 2.3 4.4 3.6 -2.4 Net lending/borrowing -7.4 -5.0 -3.8 -4.2 -3.6 -2.6 -2.4 -1.5 -0.2 -0.7 -7.0 Sources: Eurostat, European Commission. Annex 1.1 31 CHAPTER 2 TAX ASSESSMENT OVERVIEW EU-27 average of 40.1 percent.2 In the same year, indi- OF THE TAX SYSTEM rect taxes represented 39.9 percent of total tax reve- nue — 5.7 percentage points above the EU-27 average. At 32 percent and 19 percent, respectively, the top statu- Poland’s tax system relies heavily on indirect tax- tory rates on personal and corporate income in Poland es and generates relatively low revenue compared are low compared to the corresponding EU-27 average to the EU average.1 Consolidating the budget from rates of 38.8 percent and 21.5 percent (bottom left pan- all levels of government, total tax revenue (including el Figure 2.1). On the other hand, the standard 23 per- mandatory social security contributions) amounted to cent value added tax (VAT) rate in Poland is 1.5 percent- 35.2 percent of GDP in 2019 (Table 2.1) compared to an age points higher than the EU member state average. Figure 2.1 Total Tax Revenue and Selected Characteristics of the Tax System A. Tax revenue B. Tax revenue by source Ireland 50 Romania Bulgaria Lithuania 45 Latvia Malta Estonia 40 United Kingdom Cyprus Slovak Republic 35 Spain Percent of Total Portugal 30 Poland Czech Republic Hungary 25 Slovenia Croatia Netherlands 20 Greece Luxembourg Germany 15 EU27 Italy Finland 10 Austria Sweden Belgium 5 Denmark France 0 0 10 20 30 40 50 Indirect Direct SSC Percent of GDP Poland EU27 Germany Excluding SSC Including SCC Czech Republic Estonia Hungary C. Statutory rates D. Share of inactives in population aged 15 – 64 (2019) 50 Poland 40 EU 30 Germany Percent 20 Czech Rep. Estonia 10 Hungary 0 CIT PIT VAT 0 10 20 30 40 50 Percent of Total Poland EU27 Germany Czech Republic Estonia Hungary Total Female Male Sources: European Commission, OECD, and ILO. 34 Poland Public Finance Review Table 2.1 Tax Revenue by Type of Source and Level of Government, 2007 – 2019 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 As percent of GDP Indirect Taxes 14.5 14.5 12.9 13.9 13.9 13.1 12.9 13.0 13.1 13.6 14.0 14.2 14.0 VAT 8.2 7.9 7.3 7.6 7.8 7.1 7.0 7.1 7.0 7.2 7.8 8.1 8.0 Taxes and Duties on Imports 0.5 0.4 0.3 0.3 0.3 0.4 0.5 0.5 0.6 0.6 0.4 0.4 0.4 Excluding VAT Taxes on Products, Except VAT and 4.4 4.7 3.9 4.5 4.3 4.1 4.0 3.8 3.8 4.0 4.0 4.0 3.8 Import Duties Other Taxes 1.4 1.5 1.5 1.5 1.5 1.5 1.4 1.5 1.7 1.8 1.7 1.7 1.9 on Production Direct Taxes 8.3 8.4 7.2 6.7 6.7 7.0 6.7 6.8 6.9 7.1 7.3 7.8 8.0 Personal Income 5.2 5.3 4.5 4.3 4.3 4.5 4.5 4.6 4.7 4.8 5.0 5.3 5.3 Taxes Corporate Income 2.7 2.7 2.2 1.9 2.0 2.1 1.8 1.7 1.8 1.8 1.9 2.1 2.2 Taxes Other 0.4 0.4 0.4 0.4 0.4 0.4 0.5 0.5 0.4 0.5 0.4 0.4 0.4 Social 11.9 11.3 11.2 10.9 11.2 12.1 12.3 12.2 12.5 12.8 12.9 13.1 13.2 Contributions Employers’ 4.6 4.5 4.5 4.6 4.5 4.8 4.8 4.9 4.9 5.0 5.1 5.1 5.1 Households’ 7.3 6.8 6.7 6.3 6.7 7.2 7.5 7.3 7.6 7.8 7.8 8.0 8.1 Capital Transfers 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Total 34.6 34.1 31.2 31.4 31.8 32.0 31.9 31.9 32.4 33.4 34.1 35.1 35.2 As percent of Total Central 52.4 53.2 51.1 52.6 52.3 49.7 48.6 48.7 48.4 48.6 49.2 49.5 49.3 Government Local Government 13.3 13.5 13.1 12.6 12.3 12.6 12.6 13.0 12.9 12.8 12.7 12.7 12.6 Social Security 34.1 32.9 35.5 34.4 35.0 37.3 38.4 37.9 38.3 38.0 37.6 37.2 37.6 Funds EU Institutions 0.3 0.4 0.3 0.4 0.4 0.4 0.4 0.4 0.5 0.5 0.5 0.5 0.6 Source: European Commission. Overall, the Polish tax system shows limited pro- weakly progressive. SSC weighs heavily (around gressivity. First, indirect taxes have limited redis- 28.5 percentage points) and proportionately on the tributive potential. Thus, the high reliance on these tax wedge of employees across the income distribu- taxes to finance the state budget constrains the tax tion. Third, several of the current PIT and SSC spe- system’s overall progressivity. Second, the combined cial treatments and exemptions provide tax reliefs system of personal income tax (PIT) and social se- primarily to high-income earners, giving rise to ef- curity contributions (SSC) on labor income is only ficiency and equity concerns. Tax assessment 35 Local governments receive a relatively high share Labor force participation in Poland remains rel- of tax revenue, but this is mostly raised via the atively low compared to peer countries, especial- shared PIT and CIT taxation. Local governments in ly among women, with negative consequences on Poland include municipalities (gminy), counties (powi- the total PIT base. In 2019, only 70.85 percent of the aty), and regions (województwa).3 In 2019, these three working-age population participated in the labor levels of government received 12.6 percent of the to- market, 2.44 percentage points below the EU aver- tal consolidated tax revenue raised in the same year age and far from the 78.57 percent seen in Germany (Table 2.1), a level above the EU average of 10.2 per- (Figure 2.10). On this aspect, Poland also tends to per- cent. In 2018, around 74 percent of this revenue came form worse than peer economies such as the Czech from shared PIT and corporate income tax (CIT) tax- Republic and Estonia (bottom right panel Figure ation, both of which are managed by the central gov- 2.1). Labor force participation is particularly low for ernment. This left little room for local governments to women. The gender participation gap in 2019 stood independently adjust their revenue if needed. at 13.8 percentage points compared to 8.4 percent- age points in Germany and an EU average of 10.9 per- The main characteristics of the Polish tax sys- centage points. This high gender gap is both a driver tem are common among Central and Eastern and a sign of gender-based inequities and can mag- European (CEE) countries. These countries usu- nify gender inequality both outside and within the ally rely on indirect taxes as the main source of fi- household. Additionally, the low participation of nancing for the government budget and raise lower women in the labor market depresses aggregate la- tax revenue on average than in Western EU mem- bor supply, with negative consequences on the to- ber states. In 2019, the level of total tax revenue rel- tal PIT base. Tackling gender disparities in the labor ative to GDP in Poland was close to that in the Czech market remains critical to address this issue. Republic (36.1 percent), Hungary (36.5 percent) and Slovenia (37.4 percent) and above that in other CEE countries (top left panel in Figure 2.1). Similarly, while in 2019 indirect taxes as a share of total tax revenue in Poland were 5.7 percentage points above DIAGNOSTICS the EU-27 average, they were in line with most of BY INSTRUMENT Poland’s CEE peers (top right panel Figure 2.1). Personal Income Tax A large part of the potential tax base remains un- and Social Contributions taxed and is an important untapped source of public funding. According to estimates from the Overview of the System European Commission (2017), in 2013 Poland had the highest rate of undeclared labor in the EU.4 The The combined PIT and SSC system in Poland is report estimated that undeclared labor produced only mildly progressive and features several spe- 27.3 percent of the gross value added in the private cial treatments applicable to specific sources of sector, compared to an EU-28 weighted average of income. Revenue from personal income taxes (PIT) 14.3 percent. ILO (2018) estimates for 2016 show that amounted to 5.3 percent of GDP in 2019, equivalent 20.1 percent (15.9 percent outside agriculture) of jobs to 66 percent of the total revenue collected from di- were informal, 53.3 percent of which were offered by rect taxes. In line with the specificities of the Polish formal firms.5 Tax avoidance and evasion of CIT and tax system, the size of PIT revenue relative to GDP VAT have been declining in recent years, but remain is small compared to the EU-27 average of 9.6 per- sizable.6 In 2018, they were responsible for a loss of cent but in line or above that seen in several CEE tax revenue estimated at 33 percent and 9.3 percent peers. Social security contributions (SSC), exclud- of the respective theoretical tax bases. International ing health contributions, were equivalent to 13.2 per- profits, offshoring, and the underreporting of foreign cent of GDP in 2019, in line with the EU-27 average, wealth also contribute to the drain in the potential but below that of most Western European countries. tax base for PIT, CIT, inheritance, and property taxes.7 The Polish PIT system includes a general progressive 36 Poland Public Finance Review scheme that is mandatory for income earned under commuters) and a solidarity levy was introduced to a labor contract; an optional flat rate scheme for the finance a new disability fund. The basic tax allow- self-employed; and several separate treatments for ance was also adjusted. The current phaseout design income earned from a list of specific sources. As a of the general PIT allowance came about in 2017, re- rule, PIT in Poland is filed individually. However, placing the previous universal flat allowance. This couples who have been married for the whole fiscal reform increased the tax benefits for low earners year can opt to be taxed jointly.8 and decreased them for high earners, increasing the progressivity of the system. General Scheme and Labor Income As of January 2021, the general progressive PIT scheme consists of two income-dependent tax The Polish PIT system was introduced in 1992 rates and a progressive general tax allowance and was gradually reformed over the last two de- (Table 2.2). The general scheme applies to the in- cades. Between 1998 and 2008, the system featured come earned under labor contracts and civil law three income brackets with tax rates of 19 percent, contracts as well as to income from pensions.9 The 30 percent, and 40 percent. In 2009, the number of progressive scheme also applies to self-employed income brackets was reduced to two with margin- workers who do not opt for the alternative flat rate al rates of 18 percent and 32 percent. The 18 percent scheme. The taxable income of employed workers rate was further decreased to 17 percent in October is defined as the gross wage earned minus social 2019. In the same year, the amount of tax-deductible contributions (excluding health contributions) paid costs was increased from Zl 1,335 (Zl 1,668 for com- by the employee. Up to Zl 3,000 of deductible costs, muters) to the current level of Zl 3,000 (Zl 3,600 for increased to Zl 3,600 if the worker commutes, can Table 2.2 General and Self-Employed PIT Schemes General Scheme Self-Employment Base Gross wage minus social contributions Revenue minus social security (excluding health contributions) paid by employee contributions Deductible Actual costs related to Zl 3,000 (Zl 3,600 if commuting) Costs self-employment Tax Rates Above Below Rate Choose between First Bracket Zl 0 Zl 85,528 17 percent flat 19 percent rate or general scheme (PIT first rate of Second Bracket Zl 85,528 - 32 percent 0 percent not available for young self-employed even if they choose Workers up to Workers up to 26 years old receive a tax revenue deduction salaried regime) 26 years old offsetting income tax on their income up to Zl 85,528. Yearly Tax Allowance Formula • Zl 1,360 for people with tax base of 8,000 and less • Zl 1,360 minus the value calculated as 834.88* (tax base – 8,000)/5,000, for people with tax base higher than 8,000 and up to 13,000 Cannot be claimed if the taxpayer • Zl 525.12 for tax base higher than 13,000 and up to 85,528 chooses the flat 19 percent option • Zl 525.12 minus the value calculated as 525.12* (tax base – 85,528)/41,472, for people with tax base higher than 85,528 and up to 127,000 Source: Bachas et al. (2020) and statutory rates. Tax assessment 37 be deducted from the worker’s taxable income.10 allowance. The taxable base for the self-employed is Taxable income below Zl 85,528 is taxed at a basic computed as total revenue minus social security con- rate of 17 percent, while taxable income above the tributions. Workers can fully deduct the costs that Zl 85,528 threshold is taxed at a basic rate of 32 per- they incur when performing their activity. The re- cent. The tax threshold is doubled for couples opt- duced 0 percent rate on income earned by individu- ing for joint taxation. Since August 1, 2019, the nor- als up to 26 years old is not available to the self-em- mally taxable income below Zl 85,528 earned by ployed, irrespective of the selected taxation regime. workers up to 26 years old is not taxed.11 On top of Natural persons who conduct a micro business ac- these two marginal rates, the general PIT scheme tivity can also apply for a system of simplified tax features a progressive tax allowance, which can- forms. Individuals opting for the simplified forms cels the tax liability for individuals earning up to Zl can either be subject to a tax on registered revenue 8,000. The allowance is phased out rapidly between (before the deduction of costs) or to a fixed tax rate Zl 8,000 and Zl 13,000 and more slowly thereafter, depending on the type of their business activities. disappearing above Zl 127,000. The system results To be eligible for the simplified tax forms, taxpay- in a mildly progressive PIT schedule (Figure 2.2), ers need to earn revenue below a limit that as of with an average tax rate that rapidly increases with January 2021 is set at Euro 2 million. incomes up to Zl 13,000, flattening out between Zl 20,000 and Zl 85,528 and slowly increasing again thereafter. This structure results in a mildly pro- Childcare Tax Credit gressive PIT. Since January 2019, a new 4 percent solidarity levy applies on high incomes (from any Besides the basic tax allowance and the deduct- source) above Zl 1 million. ibility of costs as described above, individual tax- payers with dependent children have the right Figure 2.2 PIT Rates by Income Tax Base to a childcare tax credit. The amount of the credit (General System) per child increases based on the number of depen- 40 dent children. As of 2020, the monthly tax credit is equal to Zl 92.67 for the first and second child, Zl Rate Percent 30 166.67 for the third child, and Zl 225 for the fourth and 20 above child. The childcare tax allowance for the first 10 child is available only to parents earning less than Zl 112,000 annually. The bulk of the public support to 0 0 16 32 48 64 80 96 112 128 144 160 families with children in Poland is nonetheless pro- vided via direct cash benefits, which equal Zl 500 per Income (Zl, thousands) month for each child up to age 18.12 As a result, the Average Rate Marginal Rate size of the childcare tax credit is relatively low. The Source: Statutory rates. benefit on the first child is currently equivalent to Notes: The average annual (gross) wage in 2018 was equal to 3.6 percent of the median monthly net income. As a Zl 55,271 (OECD). The x-axis corresponds to the PIT tax base. comparison, the monthly benefit paid in Germany Tax rates are for income earned on labor by a worker older than 26 years old. Deductions of health insurance contributions for each of the first two children amounts €204, or and tax-deductible costs are not considered. 10.4 percent of the median monthly net income. Self-Employed Flat Scheme Capital Income and Other Specific Treatments Since 2004, self-employed individuals can choose one of four schemes: the general progressive Income earned from capital and other specific scheme, a flat 19 percent rate, a lump-sum tax, sources is taxed separately and is subject to a spe- or a fixed tax rate. The self-employed who opt for cific tax treatment. In general, capital income is the flat tax cannot benefit from the progressive tax taxed separately at a flat rate of 19 percent. This in- 38 Poland Public Finance Review cludes income from dividends, interest on savings, Guaranteed Benefits Fund, which guarantees employ- gains from capital funds, gains from the sale of se- ees’ benefits in the case of their employer’s insolven- curities, the selling of private properties and capi- cy and amount to 0.1 percent of the total gross salary. tal gains gained abroad. Individuals may choose to subject their earned rental income to either the gen- Since 1999, when the Polish health financing sys- eral progressive system or a flat 8.5 percent rate for tem was transformed into an insurance-budget- income up to Zl 100,000 and 12.5 percent above this ary system, employees and the self-employed must threshold.13 Finally, income earned in competitions contribute to the public health insurance fund. and gambling is taxed at a flat 10 percent rate. For workers under a labor contract, these amount to 9 percent of the assessment base, which is equal to Under the current Polish tax law, residents are re- an individual’s gross income reduced by the amount quired to pay tax on the receipt of dividends, inter- of social security contributions paid by the employee. est, and realized capital gains from foreign assets. Small farmers and their household members, unem- However, the law does not tax gains from foreign as- ployed workers, and recipients of social assistance are sets held indirectly through an intermediary offshore not required to pay contributions to the health fund. fund. Taxes are paid only when returns are paid out to These are instead covered by the state budget. Part of Polish resident investments. This exception favors tax the contribution corresponding to 7.75 percent of the avoidance of capital income from foreign sources, al- assessment base is deducted from the employee’s PIT lowing residents to legally avoid paying capital income liability while the remaining share, equal to 1.25 per- taxes by investing their savings in an offshore fund. cent of the assessment base, is not deductible. The self-employed are also required to contrib- Social Security and Public Health ute to the social security system. Contributions Contributions are calculated using a theoretical assessment base. This theoretical base is 60 percent of the predicted Employers and employees are required to pay man- average wage in the enterprise sector. Any additional datory social security contributions, which are contribution up to 250 percent of the average wage is used to finance several social insurance and as- voluntary. The self-employed who either have reve- sistance programs. Employees working under a la- nue below 50 percent of the minimum wage (or are bor contract must pay 11.26 percent of their total gross in the first six months of their activities) are not re- salary to the pension and disability insurance funds, quired to contribute. Additional specific treatments while employers need to contribute for an additional apply to the self-employed in small firms and the 16.26 percent. Neither employee nor employer pension self-employed earning less than Zl 120,000.14 Health and disability contributions are required on incomes contributions for the self-employed are computed above a maximum salary cap that is adjusted every using a theoretical assessment base equal to 75 per- year. In 2021, the cap is set at Zl 157,770. The employ- cent of the average wage in the enterprise sector in ee pays an additional 2.45 percent contribution to the the fourth quarter of the previous year. sickness insurance fund. The employer must also pay contributions to the accident insurance fund, Labor Specific rules also apply to individuals working Fund, Solidarity Fund for Persons with Disabilities under a civil law contract. Employers and employ- and Employee Guaranteed Benefits Fund. The acci- ees who stipulate a commission contract pay mandato- dent insurance contribution comes to 1.67 percent ry social contributions on an assessment base at least for firms with up to nine employees and varies be- equal to the minimum wage, while additional contribu- tween 0.67 percent and 3.33 percent, depending on tions are only voluntary.15 Health contributions follow the sector of activity for such firms. Contributions to the same rule applied to workers employed under a la- the Labor Fund and Solidarity Fund for Persons with bor contract. Workers employed under a results-based Disability sum up to 2.45 percent of the total gross sala- contract, on the other hand, are not subject to any man- ry. Finally, employers must contribute to the Employee datory social security or health contribution. Tax assessment 39 Main Issues with the Current System tion of the tax burden for salaried employees. With the SSC rate flat up to Zl 157,770 and decreasing there- Limited Progressivity of PIT and SSC after, the incidence of SSCs on the labor tax wedge strongly limits the degree of progressivity (Figure 2.3). Overall, the Polish PIT and SSC system exhibit a These two factors combine to produce a very limit- limited degree of progressivity (Box 2.1). Several ed progressivity in the tax burden on labor income factors contribute to this outcome. First, the basic (Figure 2.4). Data from the OECD for 2019 shows that progressive scheme features only two marginal tax in Poland, the differential in the total (including SSC) rates, with a top rate well below the average in EU average tax wedge on labor income between work- countries, and a relatively limited tax allowance that ers earning 67 percent and 167 percent of the average concentrates progressivity in PIT at very low incomes. wage was equal to 1.1 percentage points, the second Second, SSCs represent a disproportionately big frac- lowest among European OECD countries (Figure 2.3).16 Box 2.1 Equity and poverty analysis of current fiscal system and announced tax reforms As part of the present review, the team assessed the dex by 3.6 and 3.7 percentage points, respectively. This redistributive effects of the current Polish tax system equity enhancing effect is however driven solely by direct and of the tax reforms that are part of the proposed transfers. The CEQ assessment, in fact, finds the current Polish “New Deal” program. The analysis is based on the system of direct taxes to be close to neutral, while indirect Commitment to Equity (CEQ) modelling framework, cali- taxes are found to be regressive, increasing the Gini coeffi- brated using the 2019 Polish Household Budget Survey cient from 0.33 to 0.34 and the two poverty measures from (PHBS). The CEQ approach provides a comprehensive 2.9 to 3.4 percent and from 3.2 to 4.3 percent, respectively. assessment of the impact of fiscal policy on inequality and poverty, accounting for both the effect of direct and Equity assessment of child tax credits indirect taxes and subsidies and for the effect of public spending on education and health.a Child tax credits are found to be mildly progressive and contribute to the reduction in the poverty rate among households with children. Categorizing households based Redistributive performance of the current system on their pre-family-transfers disposable income, around Overall, the Polish fiscal system is found to reduce income 43 percent of child credits resources go to households inequality and poverty, but this result is driven by exist- in the first four quartiles of the income distribution, while ing expenditure programs rather than by taxes. The pen- 34 percent go to those in the top four quartiles. The pol- sion system represents the biggest equalizing force in the icy contributes to reducing the extreme poverty rate in current system, reducing the Gini coefficient from 0.54 to households with 1 child, 2 children, and 3 or more chil- 0.4 and the absolute and national extreme poverty index dren by 0.1, 0.3, and 2.1 percentage points (Table B2.1.1). from 24 percent to 7 percent and from 25 percent to 8 per- Overall, the program represents, on average, 1 percent, cent, respectively.b The combined effect of direct taxes and 2 percent, and 3.7 percent of the disposable income of transfers and indirect taxes further decrease the Gini co- these three types of households, covering a total of 6.6 efficient by 0.06 and the absolute and national poverty in- million children in the country. Table B2.1.1 Poverty among households with children for disposable income with and without child tax credits Extreme poverty among households with 1 child 2 children 3+ children Disposable income minus family support programsc 4.0% 5.7% 13.9% + Child tax credit 3.9% 5.4% 11.8% Source: Poland CEQ model. 40 Poland Public Finance Review Redistributive effect of PIT and HIC reforms The results of the analysis show that the proposed reform under the proposed Polish “New Deal” will reduce the average joint PIT and HIC burden over most of the income distribution (Figure B2.1.1). The benefits of On May 15, 2021, the ruling coalition announced a large the reform are expected to be larger for taxpayers between reform program that the government plans to implement the 2nd and 8th decile of the income distribution, while an in the coming months. The program, known as the Polish increase in the average PIT/HIC burden is expected in the “New Deal”, contains major reforms of the PIT and HIC sys- top decile and below the fifth percentile.d Assessing the tem. These include: an increase in the PIT tax allowance reform by household type shows that retirees are, on av- to guarantee a zero PIT rate on income up to Zl 30,000; erage, predicted to benefit the most thanks to their over- an increase in the first PIT threshold from Zl 85,528 to Zl representation around the middle of the income distribu- 120,000; the abolition of HIC deductibility from the tax- tion. Most of the effect of the proposed reform is expected payers’ PIT liability and the homogenization of HIC across to come from the increase in the tax allowance and from sources of income. The government also announced that the changes to the HIC system, with the increase in the a compensating tax credit will be provided to workers un- PIT threshold affecting a more limited number of taxpay- der a labor code or self-employment contract earning ers. The simulation estimates that the overall reform will monthly gross wages between Zl 5,701 and Zl 11,141, reduce the total PIT revenue by Zl 21.29 billion.e which will offset any increase in PIT for these taxpayers. Figure B2.1.1 Average joint PIT and HIC burden by PIT income base percentiles: current scheme and “New Deal” proposals’ simulation. 35 30 25 20 Percent 15 10 5 0 -5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 71 75 80 85 90 95 100 Percentiles of the PIT Income Base Current PIT+HIC Tax allowance of PLN 30,000 (1) No HIC deduction (2) PIT threshold of PLN 120,000 (3) (1) + (2) + (3) + tax credit for middle class Source: Own preparation based on Poland CEQ 2020 a https://commitmentoequity.org/ b The absolute poverty rate is computed using the $5.5 PPP headcount index while the national extreme poverty rate is based on the minimum existence level as defined by the Polish Institute for Labor and Social Studies (IPiSS). c Family support programs include: the child tax credits; the family allowance with its supplements; the Family 500+ program for every child; the Good Start program. d While the reform will increase the average tax burden on the bottom decile of the income distribution, this increase will be small in absolute terms and is estimated in the order of Zl 5 per month on average. e This is likely to be an upper estimate of the reduction in PIT revenue due to the underrepresentation of high-income households in the PHBS used in the CEQ analysis. Several special treatments applicable to specific The optional joint PIT scheme also allows high-in- sources of income or under specific circumstanc- come earners to reduce their average income tax es are a third factor reducing the progressivity rate. The effective tax rate on high-income earners of the Polish PIT and SSC system. The favorable is thus often substantially lower than those present- PIT and SSC treatments of self-employment income ed in Figure 2.1. As pointed out in Bachas et al. (2020), and the flat rate on capital income disproportionately for example, only 4.8 percent of Polish PIT taxpayers benefit high-income earners who earn a large share have part of their income subject to the maximum of their income from capital and self-employment. 32 percent PIT rate. Tax assessment 41 Figure 2.3 Tax Wedges by Wage Level in OECD Countries A. Total Tax Wedge on Labor Income by Country B. Tax Wedge on Labor Income by Country (no SSC) Hungary Hungary Poland Poland Latvia Latvia Slovak Republic Slovak Republic Czech Republic Czech Republic Lithuania Lithuania Germany Slovenia Slovenia Denmark Austria Norway Estonia France Spain Spain Denmark Estonia Norway United Kingdom Portugal Portugal Greece Greece Sweden Austria United Kingdom Germany Netherlands Finland Finland Sweden Italy Belgium Belgium Luxembourg France Italy Luxembourg Ireland Ireland Netherlands 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Percent Percent Gap 67% of Average Wage 167% of Average Wage Source: OECD. Notes: Tax wedges are expressed as percentages of gross wage earnings. Data is for 2019. Atypical Contracts, Labor Market Duality cal contracts of this sort give minimal protection to and Equity. the employed person and provide limited contribu- tions to the employee’s pension fund. Besides its lack of progressivity, observers have pointed out that the current PIT and SSC system in- The use of atypical work arrangements as a sub- centivizes the overuse of alternative (or atypical) stitute to regular labor contracts thus raises work arrangements in the labor market (World concerns in terms of both horizontal and ver- Bank 2020). The use of atypical contracts exacer- tical equity. From a horizontal equity perspective, bates the stark duality seen in the Polish labor mar- using atypical contracts for what are effectively em- ket, where in 2017 around one quarter of employees ployer-employee relations often means that work- worked under a temporary labor contract — the sec- ers with similar qualifications performing similar ond-highest share in the EU. In the same year, 1.2 mil- jobs can work under very different conditions and lion people worked exclusively under a civil law con- be subject to different tax wedges. Experiences tract and around 170 thousand worked outside the from other EU countries suggest that a widening agricultural sector as self-employed in firms serv- duality in the labor market tends to amplify dis- ing mainly or only a single client (World Bank 2020).17 parities between incumbent/older workers and workers new to the labor market or the recently From the perspective of the employer/contrac- unemployed.18 The social cost of this duality is ex- tor, the specific SSC treatment of these two atypi- acerbated when low-paying jobs under atypical cal work arrangements makes them particularly contracts stop being steppingstone opportunities advantageous compared to standard temporary for younger individuals and become the norm in or open-ended labor contracts. Mandatory con- their early working years. A similar equilibrium in tributions on these contacts tend in fact to be con- the labor market can have negative consequences siderably lower compared to the contributions re- on the human capital of young workers and affect quired on standard labor contracts, thus reducing important personal choices, with long-lasting con- labor costs (Figure 2.4). On the other hand, atypi- sequences on their well-being.19 42 Poland Public Finance Review Figure 2.4 Fiscal Burden by Annual Gross Income A. Average Fiscal Burden on Gross Adjusted Income B. Average Fiscal Burden on Gross Adjusted Income by Percentiles of Income (2017) by Source of Income (2017) 60 100 50 80 40 60 Percent Percent 30 40 20 20 10 0 0 0 20 40 60 80 100 120 140 160 180 200 3 9 12 13 15 18 21 24 27 28 32 36 40 46 52 59 68 81 106 220 Adjusted Gross Income (PLN, thousands) Adjusted Gross Income (PLN, thousands) Total Employed Self-Employed PIT ZUS NFZ Civil Contract Source: Polish Ministry of Finance. Notes: ZUS: social security contributions. NFZ: health contributions. Gross adjusted income equals: revenue — (tax deductible costs + the amount of tax-exempt revenue) + (social and health insurance contributions paid by the payer or the state budget + tax-deductible costs for employment contracts and civil law contracts + social security contributions included in the costs of economic activity). Part of the unequal treatment associated with ty that characterizes atypical contracts and with an different contracts is not immediately visible in inadequate level of retirement contribution to the a worker’s net paycheck and will only material- workers retirement plan. It is nonetheless import- ize upon the employee’s retirement. If present be- ant to note that employers may deem a standard havioral biases are strong, by excessively discount- contract unprofitable, given its more onerous con- ing the value of contributing to their retirement, ditions. Any attempt to solve this issue should thus workers may accept jobs benefiting from limited SSC compensate for its possible perverse effect on ag- contribution without receiving compensatory wage gregate employment and on the employment of low- differentials. Furthermore, the low wages usually as- skilled/low-wage workers. sociated with these work arrangements are unlike- ly to allow workers under these contracts to make large enough additional, voluntary contributions to Corporate Income Taxes their pension fund to compensate for their low in- comes after retirement. Their low expected retire- Overview of the System ment pensions thus raise concerns that the current overuse of atypical work arrangements might entail General System substantial social costs in the future. This calls for a solution that could guarantee good retirement con- The current corporate tax system in Poland fea- ditions for these workers. tures a basic 19 percent flat rate. Small firms, which get preferential tax treatment, are required The fact that these atypical arrangements are to pay a flat 9 percent rate on their incomes from increasingly used as substitutes for standard sources other than capital gains. Small firms are de- labor contracts for low-paying jobs also raises fined as enterprises that either just started their ac- concerns in terms of vertical equity. This practice tivity or earned less than €2 million in revenue (in tends to amplify inequalities among salaried work- zloty equivalent) in the previous tax year. Companies ers. The overuse of atypical contracts means that, making losses in one year can carry them forward besides providing lower wages, low-paying jobs are for five consecutive years and deduct them from increasingly associated with the limited job securi- their taxable income.20 Tax assessment 43 The statutory CIT rate has been gradually reduced “Estonian” CIT over the past 25 years and is currently among the lowest in the EU, at 2.5 percentage points below Starting from the January 1, 2021 CIT, eligible the EU27 (simple) average. Similarly, standardized taxpayers can opt for a so-called “Estonian CIT” estimates of the effective tax rate on large firms (right scheme.22 The new optional scheme allows firms panel Figure 2.5), which consider tax credits and al- to delay CIT payments until profits are paid out lowances, show that Polish firms pay a lower CIT to shareholders, thus freeing up resources that (16.9 percent effective rate in 2019) on their profits could be reinvested into a business’s growth.23 The relative to the EU27 average (19.7 percent rate).21 The “Estonian CIT” is available only to limited and pub- difference is particularly big when comparing Poland lic liability companies with total annual gross rev- to the two biggest EU economies. The Polish effective enue not exceeding 100 million Zl and listing only rate on big firms in 2019 was estimated to be 12.3 and natural persons as shareholders. Several addition- 16.8 percentage points lower than the effective rates al criteria further limit the eligibility for the new in Germany and France, respectively. On the other scheme. The government projects that the new poli- hand, a low level of effective CIT rate is common in cy will help boost investment from small and medi- CEE countries: Hungary, Estonia, and Lithuania all um businesses. Importantly, however, the eligibili- feature effective CIT rates below the estimated rate ty threshold adds a new size-based distortion to the in Poland. A similar picture emerges when consider- Polish CIT system, distorting choices and creating ing the implicit CIT rate (left panel Figure 2.5), calcu- uncertainty for firms around the revenue cutoff.24 lated as the ratio between total CIT tax revenue and In addition, the number of eligibility criteria and theoretical CIT tax base. Once again, Hungary and complexity of the new legislation might discourage the Baltic states are among the EU countries with taxpayers from benefiting from the new tax scheme, the lowest implicit CIT rate. While above the level in thus limiting its effective benefits for the econo- these peers, the implicit CIT rate in Poland remains my.25 The recent introduction of an “Estonian CIT” well below that of most other EU countries. As of in Latvia also suggests that this policy can exert a 2018, the implicit rate was estimated at 13.7 percent, high cost on the public budget, at least in the short compared to 20.3 percent in Germany, 22.7 percent run. After the introduction of an “Estonian CIT” in the Czech Republic, and 35.4 percent in France. scheme — covering all corporations in the coun- Figure 2.5 Implicit and Effective CIT Rates in Selected EU Countries A. Implicit CIT Rate by Country (2018) B. Effective CIT Rate for Large Firms by Country (2019) Lithuania Croatia Latvia Lithuania Croatia Estonia Estonia Poland Netherlands Czech Republic Poland Latvia Slovenia Finland Slovak Republic Austria Sweden Denmark Finland Slovak Republic EU27 Slovenia Denmark Germany Netherlands Sweden Austria Czech Republic Germany France France 0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40 Percent Percent Source: European Commission. Note: Effective rates are computed for large non-financial firms, with average assets and funding mix. 44 Poland Public Finance Review try — in Latvia on January 1, 2018, the country’s This expansion in R&D tax allowances had an im- CIT revenue relative to GDP fell from 1.06 percent portant effect on the estimated cost of R&D ex- in 2018 to 0.16 percent in 2019 — a 85 percent drop penditures for Polish firms (left panel Figure 2.6). compared to the previous year. OECD estimates show that in 2020, the implied tax subsidy rate on a euro spent on R&D was equal to 22 percent for both large firms and SMEs (without Thin capitalization rule RDC status), while there was no estimated implied subsidy in 2015. The expansion of R&D tax allowanc- On January 1, 2018, the government introduced es seems important for creating an innovator-friend- a thin capitalization rule imposing a limit to ly business environment. While no rigorous assess- the amount of tax-deductible interest rate pay- ment of these recent changes has so far been carried ments. The rule is meant to be a disincentive for out in Poland, evaluations of similar policies in other firms’ over-indebtment and prevent the use of debt countries usually find important positive effects on financing (over equity) to reduce CIT payments. Tax- both total R&D expenditures and on various prox- deductible interest costs are currently limited to ies of R&D output.29 Recent evidence has also shown 3 million Zl plus 30 percent of the EBITDA. positive spillovers across firms, which are likely to magnify the positive effect of these policies on ag- gregate innovation.30 On the other hand, it is worth R&D Tax Incentives mentioning that the use of these new tax allowances so far has been relatively limited in Poland, accord- The Polish tax system provides tax relief for ing to data from the Polish Ministry of Development. firms investing in Research and Development Only 951 CIT taxpayers in 2018 and 1277 in 2019 ex- (R&D) activities. R&D tax incentives are provid- ploited the scheme. While this number is increasing, ed through a tax allowance based on R&D-related it is important that the government work to mini- costs.26 Firms can deduct 100 percent of their R&D mize any bureaucratic complexity or uncertainty in- expenses from their tax base, with no upper limit volved in the process of applying for and obtaining to the amount of R&D expenditures that can be de- these incentives. The government should also en- ducted. The rate of relief for the qualifying costs is sure that firms are fully aware of the benefits that raised to 150 percent for firms with Research and they are entitled to receive. Development Center (RDC) status.27 Unused allow- ances during a given tax year can be carried for- On January 1, 2019, an innovation support tax ward for up to six years and they can be refunded scheme, named Innovation Box (IB), was intro- if the firm is a start-up. duced. The IB introduces a preferential 5 percent tax rate applicable to qualifying income derived from R&D tax allowances were introduced in 2016 and intellectual property (IP) rights. To benefit from they have since been gradually expanded. The tax the tax incentive, a firm needs to conduct R&D ac- relief rate for labor expenses increased from 30 per- tivities related to the IP item registered in the new cent in 2016, to 50 percent in 2017, and to 100 per- scheme. The firm is also required to separately re- cent in 2018. The rate on other R&D expenses for port in its accounting books any income related to SMEs was raised from 30 percent in 2016, to 50 per- the IP rights. Firms are entitled to the preferential cent in 2017 and 100 percent in 2018 while the rate tax rate until the IP right expires. Recent research for big firms increased from 10 percent in 2016, to on similar types of tax incentives in other coun- 30 percent in 2017 and 100 percent in 2018.28 The car- tries shows preliminary evidence of a positive effect ry-forward period for unused allowances was ex- on both the quality and quantity of R&D output.31 tended from 3 to 6 years in 2017. Prior to January 1, When considering the role that IP boxes can play in 2016, R&D support was limited to a system of accel- stimulating innovation, it is important to consider erated depreciation of machinery, equipment and that, contrary to R&D allowances, IP boxes are in- buildings used in R&D. come-based tax incentives. Firms thus benefit from Tax assessment 45 Figure 2.6 Public Support to R&D in Poland and other OECD Countries A. Implied Tax Subsidy Rate on R&D B. R&D Public Support C. R&D Public Support Expenditures by Country (2020) in Poland by Country (2018) Slovak Republic 2000 France France Hungary 2001 Netherlands Portugal 2002 Italy Spain Portugal Lithuania 2003 Slovenia Greece 2004 Austria Ireland 2005 Belgium Poland Ireland 2006 Czech Republic Slovenia Czech Republic 2007 Poland Germany 2008 Spain Germany Hungary 2009 Greece Austria Malta 2010 Belgium Slovak Republic Netherlands 2011 Estonia Italy 2012 Lithuania Sweden 2013 Denmark Romania Romania 2014 Sweden Croatia Denmark 2015 Croatia Latvia Latvia 2016 Luxemburg Estonia 2017 Bulgaria Bulgaria 2018 Luxemburg 0 0.1 0.2 0.3 0.4 Malta 0 0.03 0.06 0.09 0.12 0.15 Percent of GDP -0.2 0 0.2 0.4 0.6 Percent of GDP 1 — B-index Direct Government Financing Direct Government Financing R&D tax incentives Large firm SME R&D tax incentives Subnational R&D tax incentives Source: OECD. Notes: Implied tax subsidy is defined as (1 – B-index). For more information on the OECD B-index see OECD (2020). Estimates of implied subsidies are for profitable firms. Data on total R&D tax incentives for France and Belgium in the third panel are from 2017. IP boxes only if their research is successful and pro- to R&D was entirely provided via direct financing of duces new revenue. While this can create additional R&D expenditures. The 2016 R&D reform introduced incentives to invest only in high-quality innovations, indirect financing through R&D tax incentives and it might also reduce the relative benefits from in- the amount was increased thereafter. As of 2018, tax vesting in highly innovative but risky research. The incentives still represented only 12.2 percent of the IP box also requires paying R&D costs in advance total R&D support. In the same year, the amount of and is thus likely to favor bigger firms–which have R&D support relative to GDP in Poland was above the easier access to finance–over smaller firms, which median among EU OECD countries but remained at might be unable to pay the upfront costs of risky around one third of the level in France, the EU OECD investments in innovation. Given these consider- country with the highest relative public spending on ations and recognizing the risk of firms simply re- R&D. Total public R&D support was nonetheless above locating IP rights from abroad to enjoy the favorable that provided in most CEE countries except for the treatment granted under the IP box, the government Czech Republic and Hungary. Support provided via should consider carrying out a rigorous assessment R&D tax allowances remains low relative to GDP, as of net benefits of this policy.32 compared to most EU OECD countries. According to the most recent OECD data, total pub- lic support in R&D as a share of GDP has been on an Special economic zones (SEZs) upward trend in Poland over recent years (Figure 2.6), reaching 0.123 percent of GDP in 2018, the last Until the end of June 2018, business activities op- year with available data. Until 2015, public support erating in 14 Special Economic Zones (SEZs) and 46 Poland Public Finance Review in possession of an officially-issued SEZ business Figure 2.7 Foregone Revenue (in Zl) permit had the right to a preferential tax treatment. from Selected CIT Tax Expenditures This included a partial exemption from CIT based 4 on the amount of investment made in SEZs and on the total wage bill for newly created local jobs. The 3 reform implemented in 2018 essentially eliminated Billions of Zl the SEZs system. Businesses can now apply for the 2 tax incentives independently of their location. An investment project’s right to the tax exemption is 1 evaluated based on a series of quality criteria, but preference is still given to investments in less devel- 0 oped regions. Tax credits earned under the previ- 2014 2015 2016 2017 2018 ous SEZs system can be used until 2026. SEZs were first introduced in Poland in 1994 and were aimed SEZs Reduced Rate R&D Credits primarily at attracting technology transfers, busi- Sources: PIE (2020) and OECD. ness activities, and foreign direct investments (FDI) to less-developed regions. Past studies have usual- CIT Tax Gap and Trend in CIT Revenue ly found SEZs to be successful at attracting foreign investments, but their effect on employment and CIT revenue as a share of GDP decreased until 2014, wages remains ambiguous.33 Since the 2018 reform but has since been on an upward trend, with its of the SEZ system, tax incentives are no longer re- share in GDP increasing by 0.38 percentage points stricted to certain regions. There may thus be more between 2014 and 2019 (Figure 2.8). This increase efficient and less administratively-demanding poli- happened while the government introduced a series cies that would provide similar broad-based support of new CIT tax incentives that had a (likely) direct neg- for investment. The government should consider an ative impact on CIT revenue. This negative pressure in-depth assessment of the cost and benefits of the on tax revenue was at least partially offset by a con- current SEZ system relative to alternative invest- temporaneous improvement in CIT compliance. This ment support schemes. was a result of a series of measures adopted by the Polish authorities in 2014 and aimed at reducing tax evasion, which focused primarily on increasing VAT Foregone Revenue from CIT Credits compliance and included tighter regulations on busi- and Allowances ness-to-business payments and a reduction in the limit on cash payments. These measures were successful at The set of CIT allowances and preferential tax stimulating the emergence of shadow transactions.35 treatments currently in place represents a source of foregone tax revenue (Figure 2.7). The Polish The estimated CIT gap has decreased in recent Economic Institute (PIE 2020) estimates that the com- years but remains sizable (Figure 2.9). Estimates bined revenue cost of the SEZs CIT benefits and the from the Polish Economic Institute (PIE 2020), sug- 15 percent preferential CIT rate for small business- gest that the CIT gap, measured as the difference be- es amounted to Zl 3.3 billion in 2018 or 7.5 percent of tween estimated potential CIT revenue and effective the total CIT revenue.34 As of 2018, the R&D tax al- revenue being collected, has decreased between 2014 lowances accounted for an additional Zl 318 million and 2018. The drop in the CIT gap since 2014 mir- (0.7 percent of CIT revenue). While the direct reve- rored the contemporaneous increase in CIT tax rev- nue cost of the preferential CIT treatment for small enue. While the total CIT gap is estimated to have firms was relatively small in 2018 (Zl 0.6 billion), it decreased since 2014, the estimated gap induced by is likely to be substantially higher in the subsequent profit shifting to foreign destinations has increased years, due to the reduction of the preferential rate from 0.13 percent of GDP in 2014 to 0.15 percent in from 15 percent to 9 percent implemented in 2019. 2018. As of 2018, the CIT gap is estimated to have re- Tax assessment 47 mained considerable, equal to 33 percent of the es- pliance and further increased, at least temporarily, timated potential CIT revenue.36 While it is too early the CIT gap. Taken together these estimates suggest to have estimates for 2020, the disruptions from the that there remain ample margins to improve CIT col- COVID-19 pandemic might have decreased tax com- lection efficiency and limit harmful tax competition.37 Figure 2.8 CIT Rates and CIT Revenue A. Implicit CIT Rate B. Statutory CIT Rates 20 20 18 16 Percent 16 Percent 12 14 12 8 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 10 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Top Rate Reduced Rate C. Effective CIT Rate D. CIT Revenues (Percent of GDP) 18.0 4 17.5 3 Percent Percent 17.0 2 16.5 1 16.0 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Source: European Commission. Figure 2.9 Evolution of CIT Gap A. CIT Gap B. CIT Gap and CIT Revenues 35 2.15 55 Percent of Theoretical CIT Revenues 30 2.05 50 25 Percent of GDP Billions of Zl 1.95 45 20 15 1.85 40 10 1.75 35 5 0 1.65 30 2014 2015 2016 2017 2018 2014 2015 2016 2017 2018 Domestic Gap External Gap CIT Revenues (left) CIT Gap (right) Sources: PIE (2020) and European Commission. Notes: Estimates for the external CIT gap account only for profit shifting to Belgium, Cyprus, Ireland, Luxembourg, Malta, the Netherlands, and Switzerland. 48 Poland Public Finance Review Discussion relatively limited profitability and facing a high risk of exiting the market. At the same time, evidence The current CIT system is designed to provide fiscal on job creation shows that a firm’s age rather than support to small firms but features limited incen- its size is the key observable predictor of its abili- tives to support firms’ growth. Firms that are small ty to create new jobs.41 With the current CIT being enough (total revenue below 2 million Euros) to be eli- skewed in support of small incumbent firms with gible for the 9 percent preferential rate enjoy substan- potentially low expected growth, the authorities tial tax benefits relative to firms that are just above the should assess the distortionary effects of current eligibility threshold. Since January 2021, firms with to- preferential policies and evaluate the possible bene- tal revenue below Zl 100 million are also eligible for the fits of an alternative system–one based primarily on optional “Estonian” CIT scheme, allowing them to de- a direct support to investments in business growth. lay CIT payments until profits are paid to shareholders. While the preferential CIT treatment for small Indirect Taxes firms supports less-robust businesses, it can dis- tort firms’ growth decisions and limit their job-cre- Indirect taxes are Poland’s main source of tax reve- ation potential. For firms growing from below to nue. In 2019, the revenue raised via indirect taxation above the size cutoff, the average tax rate on their prof- was equivalent to 14 percent of GDP, 0.3 percentage its goes from 9 percent to 18 percent. This produces a points above the EU-27 average. VAT represented spike in the marginal tax rate on profits around the the main component of indirect taxation, account- preferential size threshold. As (profitable) small firms ing for 8 percent of GDP (Figure 2.10). Import du- consider the benefits of growing, this discontinuity in ties, non-VAT taxes on products, and taxes on pro- the marginal tax rate reduces the net marginal bene- duction accounted for 0.4 percent, 3.8 percent, and fit from doing so.38 Some firms that would otherwise 1.9 percent of GDP, respectively. find growing (staying) above the threshold to be op- timal, might thus decide to remain small (downsize) Figure 2.10 Revenue by Type of Indirect to enjoy the preferential tax treatment.39 Taxes by country (2019) Germany Besides its effect on the growth dynamics of in- Lithuania Czech Republic cumbent firms, the current system is likely to Slovak Republic distort the returns on new business opportuni- Netherlands ties as well. A size-based policy would change the EU-27 Slovenia relative expected benefits of opening businesses op- Poland portunities with larger optimal sizes (e.g., businesses Austria Finland with relatively high returns to scale), and thus higher Latvia expected growth, in favor of firms with a lower po- Estonia Denmark tential size. Recent evidence suggests that this chan- France nel can be important and that disincentivizing the Hungary Croatia creation of new high growth firms could have long- Sweden term consequences on job creation.40 0 5 10 15 20 25 Percent of GDP The government should reconsider the trade-off between supporting firms with higher growth VAT Import Duties Other on Products Other on Productions potential and providing support to small incum- bents. The current size-based preferential treat- Source: European Commission. Notes: Other taxes on products (ESA D.214) include excise and ment provides support to small firms and seems consumption taxes. Other taxes on production (ESA D.29) include primarily aimed at supporting fragile firms with taxes on licenses, permits, and taxes on pollution. Tax assessment 49 VAT fective VAT rate, which is estimated at 12.1 percent, a value close to the EU median. Poland first introduced The VAT tax is the biggest source of tax revenue VAT in 1993 with a standard rate of 22 percent and for the central government. The standard VAT rate a single 7 percent reduced rate. In 2011, these rates is set at 23 percent, relatively high compared to most were increased to 23 percent and 8 percent, respec- of its peer countries (Figure 2.11). The Polish VAT tively, and the new 5 percent preferential rate was system also features two reduced rates of 8 percent introduced. These increases were initially adopted and 5 percent. The 8 percent rate applies to a list of as a temporary fiscal consolidation tool but were items covering mostly food products, medical sup- eventually kept on. Further adjustments to the list plies, restaurants and hotels, and social housing sup- of goods subject to the reduced rates were carried plies. The 5 percent reduced rate applies to the sup- out in 2018 and 2019. ply of certain printed books and basic food products such as bread, dairy products, and meat. A special The VAT gap in Poland decreased considerably VAT tax treatment applies to farmers and small SMEs between 2015 and 2019, reaching a level close to and — as in the rest of Europe — financial services the EU average (2.12). Until recently, compliance are exempted from VAT.42 This system of preferential to VAT taxes was low, generating important revenue rates and exemptions substantially reduces the ef- losses for the state budget. Since 2016, the govern- ment has adopted a series of measures to contrast Figure 2.11 VAT Rate in Selected this phenomenon, increase VAT revenue, and re- EU Countries duce unfair competition form fraudulent business A. Top VAT Rate in 2020 practices. As a first measure to reduce VAT tax eva- Germany sion, in 2016 the government reduced the threshold France France for cash transaction between firms from €15,000 to Netherlands Austria Germany Slovak Republic Zl 15,000 Zl, or about €3,500. In 2018, penalties for Austria Estonia Slovakunderestimation VAT Republic were introduced together with Netherlands Latvia Latvia Slovenia Split Payment Mechanism (SPM) on an optional Czech Republic Poland business-to-business transactions. Under the SPM, Lithuania Finland Slovenia a buyer’s Czech payment to the supplier is split in two: The Republic Poland Estonia Finland Sweden is paid on the supplier’s basic account, net amount Sweden while the VAT amount is deposited on a separate ded- Lithuania Denmark Hungary Hungary icated bank account. On November 1, 2019, the SPM Denmark 0 5 10 15 20 25 30 became mandatory 0 2 in4 sectors 6 that 8 were 10 considered 12 14 16 Percent particularly prone to VAT evasion.43 Percent Evidence on SPM policies in other countries suggests that they B. Effective VAT Rate in 2018 are effective at increasing compliance and reduc- France ing the VAT gap.44 However, estimates also point to Netherlands Germany the potential increase in administrative and com- Austria pliance costs for businesses, which depending on Slovak Republic Latvia the specific design and scope of the SPM system Slovenia Poland could outweigh the benefits in terms of VAT gap Finland reduction.45 It will thus be important in the future Czech Republic Estonia to carry out a rigorous ex-post analysis of the net Sweden benefits of the Polish SPM system, accounting for Lithuania Hungary its overall welfare and distributional effects. Two Denmark additional measures were adopted in 2019. Starting 15 20 25 30 0 2 4 6 8 10 12 14 16 from May 1, the government put in place an Online Percent Percent Cash Registry System, which automatically records Source: European Commission. any registered transaction into a central database. 50 Poland Public Finance Review Figure 2.12 VAT Gap and VAT Revenue A. VAT Gap B. VAT Revenue, Poland 25 8.5 20 Percent of Tax Base 8.0 Percent of GDP 15 10 7.5 5 7.0 0 2014 2015 2016 2017 2018 2019 6.5 EU Poland 2012 2013 2014 2015 2016 2017 2018 2019 Source: European Commission. Finally, on September 1, a “white list” containing production inputs in that it does not alter the relative information on registered VAT taxpayers was in- cost of inputs and thus favors production efficiency. troduced. Starting from January 1, 2020, taxpayers In practice, the actual degree of efficiency associated risk sanctions if they make payments to a supplier with a system of VAT rates will depend, among oth- that is not registered in the “white list”. er things, on the design of preferential rates and ex- emptions and on the resulting compliance and admin- Tax data suggests that the recent measures tak- istrative costs.46 In terms of equity, conclusions on en by the authorities were successful at boosting whether VAT is regressive largely depend on whether VAT tax compliance. According to estimates from the analysis considers savings as deferred consump- CASE (2020), as of 2015 Poland had one of the high- tion, in which case it would attempt to account for est VAT gaps in the EU, measured as the difference the relative VAT burden over the lifecycle of differ- between the VAT total tax liability (the potential VAT ent individuals. In the absence of data on lifetime revenue) and the VAT that is actually being collect- income and consumption, researchers often suggest ed. The VAT tax gap equaled 24.7 percent of the total that measuring regressivity based on expenditures VAT tax liability, well above the EU average of 13 per- rather than income provides a more accurate, albeit cent. Between 2015 and 2019, Poland experienced a imperfect, picture of the distributional consequences strong reduction in the VAT tax gap, a decline that of VAT.47 Using this approach, VAT systems are usu- outpaced that of the aggregate EU level. The VAT gap ally found to be either mildly regressive, neutral, or declined to 20.4 percent in 2016, 14.3 percent in 2017, mildly progressive. These differences are often ex- 9.9 percent in 2018, and 9.7 percent in 2019, just above plained by the specific design of reduced VAT rates the EU average of 9.6 percent. In parallel to this de- and VAT exemptions. Regarding the specific features crease in tax evasion, total VAT revenue relative to of the Polish system, a recent analysis by Thomas GDP saw a sharp increase, going from 7 percent in (2020) on a subset of OECD countries finds the VAT 2015 to 8 percent in 2019. system in Poland to be mildly progressive when as- sessed based on an expenditure-based methodolo- gy.48 Nonetheless, even when VAT systems manage Efficiency and Equity to reduce inequalities, their degree of progressivi- ty is usually limited compared to standard progres- Efficiency and equity arguments play a role in de- sive income tax systems. Disproportionately relying termining the optimal use of VAT relative to di- on indirect taxation, while possibly reducing tax in- rect taxation. It is often argued that a uniform VAT duced distortions, would thus put a constraint on the is in general more efficient than a direct taxation of level of redistribution that the tax system can effec- Tax assessment 51 tively achieve. The current relatively high reliance retail price of a cigarette pack, just below the EU av- on VAT and other indirect taxes to finance the Polish erage of 80.3 percent. Several other taxes on products state budget can thus raise some concerns when it have either been introduced or expanded in recent comes to the overall equity of the Polish tax system. years. A new levy on novel tobacco products and liq- uids used in electronic cigarettes was introduced on July 1, 2020. A new sugar tax on beverages with add- Tax on Civil Law Transactions (PCC) ed sugar/sweeteners, caffeine and taurine, as well as an additional levy on alcohol beverages sold in bottles Assets that are sold by a non-professional sell- of up to 300ml, will also be introduced on January 1, er are subject to a 2 percent Tax on Civil Law 2021. Energy excises were expanded to cover coal, lig- Transactions. The tax paid by the buyer is based nite, and coke products in 2012 and have been grad- either on the purchasing price or on a higher value ually increased over the years for fuels. defined by an expert assessment. The PCC tax ap- plies to the sale of real property other than non-ag- Poland continues to grant coal and gas excise ricultural land. Assets sold by entities carrying out exemptions to energy-intensive industries sub- businesses and non-agricultural land are instead ject to the European Emission Trading Scheme subject to the standard VAT rate. without an explicit requirement to meet strict efficiency standards, a practice that violates EU laws.50 On November 27, 2019, the European Tax on Financial Assets (“bank tax”) Commission referred Poland to the European Court of Justice for failing to comply with EU directives. A levy on financial assets was introduced in 2016, While exemptions on energy tax incentives tend to with a monthly tax rate of 0.0366 percent on to- favor high-emissions activity, tax exemptions on tal assets above Zl 4 billion (Zl 2 billion for insur- cars and vehicles excises apply mostly to low-emis- ance companies). The new levy has since raised a se- sions hydrogen, hybrid, and electric cars. ries of concerns from European institutions, which creates worry that the tax could undermine the fi- Total revenue from excise duties as a share of nancial sector’s competitiveness and stability.49 The GDP has declined since 2005 (Figure 2.13). The com- government should take the concerns raised by these bined effects of the recent duty adjustments and of a institutions into account and reassess the desirabil- decline in tobacco consumption has left the total tax ity of the current “bank tax”. Figure 2.13 Excise Tax Revenue 2.4 Excise and Other Taxes on Products 2.0 Indirect taxes in Poland include a set of excise taxes levied on alcoholic beverages, tobacco, energy prod- 1.6 Percent of GDP ucts and electricity, cars, and gambling. In 2019, ex- 1.2 cises on energy products represented the biggest source of excise tax revenue (1.97 percent of GDP), followed 0.8 by excises on tobacco (0.91 percent of GDP) and alco- hol (0.54 percent of GDP). The most recent available 0.4 data on excises on cars and gambling show that in 2018, they accounted for an additional 0.23 percent of GDP. 0.0 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 Excise duties on tobacco products have gradual- ly increased over the last 30 years. According to Alchool Tobacco Energy WHO data, the total tax burden on tobacco products Cars Gambling now represents around 79.8 percent of the average Source: European Commission. 52 Poland Public Finance Review revenue from duties on tobacco products relatively Since 2016, businesses engaged in prospecting, stable over the past years (as a share of GDP. Excises exploring, and extracting hydrocarbons need to on cars and gambling show a similar pattern, while pay an environmental concession and usage fee the revenue from alcohol duties has been decreasing as determined in the Geological and Mining Act. relative to GDP. Until 2005, this decline was more than The concession fee is a one-off payment based on compensated by an increase in the revenue from en- the size of the concession area that is to be explored. ergy products and electricity, leading to an increase The usage fee is based on the volume of hydrocar- in the total tax revenue from excise duties. This trend bons extracted during the exploitation of the con- reverted in 2005, driven by a mild decline in revenue cession. The revenue from these environmental fees from energy products and a continued declined in rev- is used mostly to finance local governments, which enue from alcohol products, leading to an aggregate receive 90 percent of the collected income, with the decline in the share of excise revenue relative to GDP. remaining 10 percent going to the National Fund for Environmental Protection and Water Management. Royalties and extraction fees are levied on ex- traction and exploration activities of minerals. Since 2012, the Tax on the Extraction of Certain Environmental Taxes and Permits Minerals (TECM) has been levied on the extraction of copper and silver. Contrary to standard royal- Total revenue from environmental taxes in Poland, ty-based taxes, the TECM is based on the gross val- defined as those levied on any tax base with a prov- ue of the extracted minerals and is payable regard- en negative effect on the environment, accounted for less of the profitability of the extractive operations. 2.64 percent of GDP in 2019, above the EU-27 average Since January 1, 2020, TECM has been expanded to of 2.37 percent (Figure 2.14). The total revenue from cover the extraction of crude oil and natural gas. these taxes has increased considerably between 1996 On the same date, the obligation to pay the special and 2005 and has since remained stable relative to GDP. hydrocarbon tax (introduced with the 2014 Special As of 2019, most of the revenue from environmental Hydrocarbon Tax Act) took effect. The special hy- taxes comes from taxes on energy products, which drocarbon tax is levied on the profits from the ex- represent 87.5 percent of the total. Environmental tax- traction of hydrocarbons and can thus be viewed as es in Poland include excise duties on energy products, a tax on the rent from natural resources. permit fees from the EU Emission Trading System Figure 2.14 Environmental Taxes in Poland and Selected EU Countries A. Environmental Taxes Revenue by Country in 2019 B. Environmental Taxes Revenue in Poiand Netherlands 3.0 Slovenia Denmark 2.8 Latvia Finland Poland 2.6 Slovak Republic Percent of GDP EU27 2.4 France Austria 2.2 Sweden Czech Rpublic Lithuania 2.0 Germany 0 1 2 3 4 1.8 Percent of GDP 1.6 Energy Pollution/Resources 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 Transport Source: Eurostat. Tax assessment 53 Box 2.2 The EU Emission Trading Scheme (ETS) The EU Emission Trading Scheme (ETS) is a cap-and- and growth, should be used by Member States for cli- trade system designed to incentivize the reduction mate- and energy-related purposes. For Poland, how- in CO2 emissions to meet the EU’s climate goals. It ever, the revenues from auctions are not earmarked sets legally binding caps on CO2 emissions covering to specific projects. (European Commission 2017). power and heat generating plants, energy intensive Between 2013 and 2020, Poland received free al- industry sectors, and aviation. Close to 48 percent location of emission allowances to the power gen- of Poland’s GHG emissions were covered by the EU- eration sector to support transformation of the en- wide Emissions Trading System (ETS), including most ergy sector. Poland was among eight countries that emissions from electricity generation and industry. received free allowances. These allowances saved The ETS auctions helped generate sizeable reve- Poland an estimated 4.65 billion euros that was ear- nues: Poland received 0.5 percent of GDP in reve- marked to finance modernization of the energy sec- nues from auctioning allowances in 2019. The EU tor in accordance with the National Investment Plan ETS Directive provides that at least 50 percent of auc- across five areas: infrastructure retrofitting, infrastruc- tion revenues, including all revenues generated from ture modernization, clean technologies, diversification allowances distributed for the purposes of solidarity of the energy mix, and diversification of supply sources. (ETS) and a wide range of environmental concession equivalent), is estimated to be low relative to the and usage fees, which are levied primarily on energy European average of 247.1. The large tax base on and fuel production activities (Box 2.2). which energy taxes are levied contributes to the discrepancy between total revenue and implicit While the tax revenue from energy taxes in rate. According to the most recent available data Poland is high relative to the EU-27 level, es- from the European Environment Agency, in 2017 timates of the effective tax rate on energy are the Polish economy remained one of the most ener- low compared to other EU countries (Figure gy intensive in the EU, with an estimated intensity 2.15). The implicit energy tax (amounting to 166.9), (calculated as the ratio between gross inland ener- calculated as the ratio between energy tax reve- gy consumption and real GDP) 20 percent higher nue and final energy consumption (in tons of oil than the EU-28 level. Figure 2.15 Implicit Tax Rate on Energy and Energy Intensity in Selected EU Countries A. Implicit Tax Rate on Energy by country in 2019 B. Relative Energy Intensity by country in 2017 Denmark Finland Netherlands Estonia France Czech Republic EU27 Slovak Republic Slovenia Sweden Sweden Croatia Germany Poland Estonia Slovenia Croatia France Austria Netherlands Slovak Republic Latvia Poland Lithuania Finland EU28 Czech Republic Germany Latvia Austria Lithuania Denmark 0 100 200 300 400 0 40 80 120 160 200 Euros per ton of oil equivalent Relative to EU level (=100) Source: Eurostat, EEA. Note: The implicit tax rate is computed as the ratio between energy tax revenue and final energy consumption (in tons of oil equivalent). Energy intensity is calculated as the ratio between gross inland energy consumption and real GDP. 54 Poland Public Finance Review The mining and electricity production sectors be too optimistic. Overall, there seems to be ample are subject to particularly low implicit tax rates margin to develop a more efficient taxation of carbon on energy. The electricity sector recently received emissions in Poland (Box 2.3). Working in this direc- a transitional free allocation of EU ETS emission tion is particularly crucial given the level of carbon allowances and currently benefits from several ex- intensity shown by the Polish economy (right pan- emptions from energy taxes on the fuels used to gen- el in Figure 2.16). In 2018, emissions per capita were erate electricity, and on the electricity used in cer- in fact 29 percent higher and emissions per euro of tain industrial processes (e.g. chemical reduction). GDP were 193 percent higher than the EU-27 average. As a result, energy tax revenue is raised primarily via excise duties on motor fuel and other fuel fees. The Polish tax system includes several energy tax expenditures schemes. While some of these policies The low implicit taxation on energy contributes are meant to support the transition toward greener to the weak performance of the Polish tax sys- energy sources, large resources remain devoted to tem in pricing carbon emissions at efficient levels, reducing energy cost for households and energy-in- according to OECD estimates (left panel in Figure 2.16). tensive sectors52. Several of these tax expenditure As of 2015, an estimated 85 percent of emissions was schemes effectively reduce the cost of carbon-inten- priced at less than €60 per ton of CO2, a price equal to sive energy sources, thus contributing to the country’s a mid-point estimate for the carbon cost in 2020. This estimated high carbon price gap.53 Tax schemes re- was the second-worst performance among EU OECD ducing the cost of fossil fuels and of other carbon-in- countries. A similar pattern is seen when considering tensive energy sources are currently in place in the the carbon price gap at €60, calculated as the cumulat- energy, agricultural, and transport sectors as well as ed unpriced social cost of CO2 emissions.51 These esti- being available directly to households. Overall, ac- mates were derived for 2015, prior to the introduction cording to the European Commission (2020), in 2018 on January 1, 2016 of the coal and gas excise exemption tax expenditure on energy — regardless of their envi- for energy-intensive industries and thus they might ronmental impact — amounted to 0.1 percent of GDP. Figure 2.16 Carbon Price Gap and Carbon Intensity in Selected OECD Countries A. Carbon Price Gap in 2015 B. Carbon Intensity in 2018 Luxemburg 18 900 Tons Per Million of 2015 Euros of GDP Norway Slovenia 16 800 Italy France 14 700 Netherlands Denmark 12 600 Tons per capita Spain United Kingdom 10 500 Dinland Austria 8 400 Portugal Greece 6 300 Sweden Belgium 4 200 Latvia Hungary 2 100 Germany Slovak Republic 0 0 Czech Republic Latvia Slovak Republic Slovenia Sweden Czech Republic Poland Germany EU27 France Denmark Finland Croatia Estonia Lithuania Austria Netherlands Latvia Poland Estonia 0 20 40 60 80 100 Percent Price Gap at 60 EUR Per Capita (left) Share Priced Below 60 EUR Relative to GDP (right) Sources: OECD, EEA. Notes: Blue columns show the share of emissions priced below €60 per ton. Red columns show the carbon price gap at €60 per ton as defined in OECD (2018). Tax assessment 55 Box 2.3 Simulating the Effect of Carbon Taxes Carbon taxes — and carbon pricing in general — can be ef- aligned with the high EU prices scenario considered in fective in promoting a green transition by efficiently in- PEP2040. Under this alternative climate policy, a carbon centivizing a decarbonization of the economy. The clear tax is applied to all sectors of the economy and its lev- economic rationale for carbon pricing as an instrument to el is increased from Zl 113.4/tCO2 emission in 2021 to internalize environmental externalities makes it a prom- Zl 227/tCO2 by 2030. inent green fiscal instrument. At the same time, while a Relative to the baseline, a high carbon tax scenario sees carbon tax is likely to reduce carbon emissions, its im- a stronger decrease in carbon-intensive energy sources pact on total tax revenue is less clear. By reducing the and a higher increase in carbon tax revenue. The increase demand for fossil fuels, thus reducing the revenue from in revenue from the carbon tax is only partially offset by a the taxation of these products, a carbon tax can have a drop in revenue from excises and VAT caused by a reduc- negative long-run effect on the state budget. tion in the consumption of carbon-intensive products and As part of the present review, a dynamic CGE model of services. Under a generalized high carbon tax, the simulat- the Polish economy was used to simulate the impact of ed carbon tax revenue relative to GDP in 2030 is 0.93 per- an increase in carbon taxes over the next 10 years. The centage points higher than in the baseline scenario, while model was used to simulate two scenarios (Figure B2.3.1). the revenue from excise taxes and VAT is 0.13 and 0.08 The first — or baseline — is consistent with the current com- percentage points lower, respectively. The higher carbon mitments of the Polish government as outlined in Energy tax is also effective at reducing the use of carbon-inten- Policy for Poland (PEP2040) and sees an increase in the sive energy sources. Over a 10-year horizon, the model tax on carbon in the power, paper, chemical and metal- under the high carbon tax scenario projects an addition- lurgical sectors from Zl 74.76/tCO2 emission in 2021 al drop in the demand for electricity and gas, coal, and to Zl 126/tCO2 by 2030. The second — or high tax sce- coke and refined petroleum of 5, 12.5, and 7.1 percentage nario — assumes a more ambitious climate policy and is points, respectively, relative to the baseline (Figure B2.3.1). Figure B2.3.1 Simulated Effect of an Increase in Carbon Taxes Under Alternative Scenarios A. Size of CO2 Tax in Alternative Scenarios B. Demand for Energy Commodities per Unit of Output 250 Electricity Coke and and Gas Coal Refine Petroleum 0 Change between 2021 and 2030 200 150 -20 Zl / tCO2 100 -40 50 -60 0 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 -80 Baseline High CO2 tax Baseline High CO2 tax Source: World Bank staff calculations As the Polish economy remains highly energy- force workers to relocate between sectors. The cost and carbon-intensive, the government should of these adjustments might disproportionately fall develop plans for a significant green transition. on those low-skilled workers who currently work in Phasing out state support for polluting industries in polluting industries and in regions where polluting favor of the green economy can benefit the Polish sectors represent a major source of jobs, requiring economy in the medium and long run. In the short- the government to provide adequate support to lim- term, it might nonetheless cause disruptions and it the social costs of this transition. 56 Poland Public Finance Review While concerns over the social costs of the green Subnational Revenue transition are understandable, it is recommended and Intergovernmental Transfers that the government develop a coherent set of measures in this area. The structural transforma- The 2003 Act on Local Government Revenue es- tions required to cope with the ongoing climate cri- tablishes that local government can benefit from sis and the current push for a Green Deal at the EU three sources of revenue: own revenue, which in- level leave no room to further postpone a major scale clude local taxes, fees and charges; shared taxes and up of green policies in Poland. Deferring plans for a revenue from assets; general subsidies from the cen- green transition would risk leaving Poland lagging tral government budget; and targeted grants used to other countries in sectors that will be increasingly perform central government functions delegated to relevant in future economies.54 In general, the me- local governments. Considering these three sources, dium- and long-term social and environmental costs the revenue available to local governments amount- of deferring actions in this area are likely to far out- ed to 14.1 percent of GDP in 2019 relative to an EU-27 weigh the initial adjustment costs associated with a average of 10.9 percent. In 2018, local governments’ green transition. own resources represented 41 percent of their total revenue, compared to the EU average of 53 percent, indicating a relatively low autonomy in local reve- Trade Taxes nue, with the remaining 59 percent coming from central government transfers.55 With Poland being part of the European cus- toms union, tariffs and custom duties are set at the European level and are common among Property Taxes the Union’s members. In 2018, import tariffs and custom duties collected in Poland equaled 0.4 per- Taxes on immovable properties are levied at the cent of GDP. 80 percent of the duties collected are municipal level and in 2019 amounted to 1.7 per- transferred to the EU budget, while the remain- cent of GDP and around 29 percent of local gov- ing 20 percent contributes to the Polish central ernments’ own resources. Tax rates are set by the state budget, officially to cover collection costs. As local municipal council, with maximum applicable of 2018, the weighted average tariff rate in the EU rates defined at the national level, and differ accord- was equal to 3 percent, a level below the estimat- ing to the type of immovable property being taxed. ed world average of 4.17 percent. Property taxes are levied on lands, buildings, and con- struction structures. The tax base is defined by either the surface or value of the property. Construction Policy Recommendations structures are taxed based on the initial market val- ue of construction, while lands and buildings are tax- The government should consider reverting its de- es based on their surface. The definition of the tax cision on the 2016 coal and gas excise exemptions. base on buildings thus differs from those adopted The tax revenue recovered from the abolition of these in several other EU countries where property taxes exemptions could be used to finance tax credits for are levied based on a building’s value. energy efficiency and clean energy production. This reallocation of resources would help transition the Buildings are taxed based on their surface, with Polish economy away from carbon-intensive energy different rates applying depending on the activ- sources and toward more dynamic sectors and activ- ities they are used for. Residential buildings are ities with higher future growth potential. This mea- usually taxed at the lower rate (maximum rate of sure could be part of a broader package promoting a Zl 0.75 /m2), while commercial buildings are taxed green transition of the economy, while supporting at the highest rate (maximum rate of Zl 23.13 /m2). workers and households in industries and regions Numerous tax exemptions are granted to proper- that will be most affected by this transition. ties owned or used by socially relevant organizations, Tax assessment 57 such as universities and nursery schools. The implicit Several other taxes and fees contribute to the rev- recurrent tax rate on property was estimated at 1.09 enue of local governments. These include a tax on percent in 2017 (Barrios et al. (2019)), a level above means of transportations weighing more than 3.5 tons. those applied in most other EU countries (Figure 2.17). The tax is applicable in every commune in the country and tax rates are set by the municipal council. Other Figure 2.17 Implicit Tax Rates taxes can be levied at the discretion of the local coun- on Property (2017) cil. These include advertising fees, local market fees, France dog fees and fees collected on people visiting the local Poland jurisdiction for tourist, leisure, or training purposes. Denmark Netherlands 90 percent of the environmental concession and us- Sweden age fees goes to the local authorities that have jurisdic- Estonia Finland tion over the environmental resource being exploit- Slovenia ed. 60 percent of the total fee goes to the commune, Slovak Republic Czech Republic 15 percent to the county, and 15 percent to the province. Germany Latvia Lithuania Austria Shared Taxation 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Percent Local governments receive a share of the of the Source: Barrios et al. (2019). PIT and CIT levied on natural and legal persons residing or operating in their jurisdiction. The Separate taxes are levied on agricultural land law states that 39.34 percent of PIT revenue should (agricultural tax) and on the ownership or reg- go to the municipal government, 10.25 percent to ular use of forests (forestry tax), which are ex- counties, and 1.60 percent to regions. Similarly, empted from real estate taxation. Agricultural and 6.71 percent of CIT revenue should go to the mu- forestry taxes are calculated on standardized mea- nicipal government, 1.4 percent to counties, and sures of the output value of these properties. In the 14.75 percent to regions. Local revenue from shared case of agricultural land, the tax base is computed taxation represented 3.31 percent of GDP in 2018, using a standardized hectare unit, which is based equivalent to 74 percent of the total subnational on the actual surface, adjusted for different classes government’s own revenue. While these shared of arable lands, local economic conditions, and lo- taxes are officially considered to be a source of lo- cal climate. Tax rates applicable to a square meter cal revenue, the local authority has no power to de- of agricultural or forest land are considerably low- termine their amount, thus limiting the fiscal au- er than those applied on a square meter of land sub- tonomy of local governments. Besides limiting local ject to the real estate tax. fiscal autonomy, the high reliance of local govern- ments on CIT — a highly volatile source of tax rev- enue — makes local budgets highly dependent on Other Local Taxes, Fees and Charges local economic conditions, increasing the uncer- tainty surrounding local budget planning. Inheritance taxes are levied by local governments. Individuals are required to declare to the tax author- ity the amount of donations they received above a Analysis of Tax Expenditures tax-exemption limit, which for close family mem- bers is currently set at Zl 9,637. Any donations above Tax expenditures (TEs) in Poland mobilize large the tax-exemption threshold are taxed at rates that public resources and are provided primarily via vary between 3 percent and 12 percent, depending reduced VAT rates and CIT and PIT exemption, on the total amount received and the relationship credits, and allowances (Figure 2.18). The most re- between the donor and the receiver of the donation. cent detailed data available from the Polish Ministry 58 Poland Public Finance Review of Finance shows that in 2015, the Polish tax system Figure 2.18 Analysis of Tax Expenditures (2015) included 528 TE schemes, accounting for total fore- A. Tax Expenditure by Type of Tax gone tax revenue equivalent to 5.04 percent of GDP. Property VAT TEs accounted for the largest share of tax expen- VAT ditures (2.56 percent of GDP) and primarily took the Excises form of reduced VAT rates on specific goods and ser- PIT vices. PIT and CIT TEs amounted to another 1.18 per- CIT cent and 0.82 percent of GDP, respectively. Both PIT 0 0.5 1.0 1.5 2.0 2.5 3.0 and CIT TEs were implemented primarily via tax ex- Percent of GDP emptions and tax credits or allowances, with an ad- ditional 17 percent of total PIT TEs provided via re- B. Types of Tax Expenditures duced PIT rates. Reduced rate Exemptions Deductions/Credits The declared purpose of TEs is primarily to sup- port households, promote economic growth and 0 0.5 1.0 1.5 2.0 2.5 3.0 support specific priority sectors of the econo- Percent of GDP my. As of 2015, around 42.5 percent of TE resourc- CIT Other PIT VAT es were meant to help households in need of social and economic support. TE policies pursued this goal C. Tax Expenditures by Objective primarily via the use of favorable PIT treatments Support Subgroups in the Population and reduced VAT on social goods and services. In Environment Increase Demand 2015, another 0.93 percent of GDP of TE resourc- Priority Sector es served to promote economic growth. Most of Promote Investment these resources were provided via CIT TEs, with 0 0.5 1.0 1.5 TEs amounting to around 0.27 percent of GDP that Percent of GDP were meant to attract or stimulate investments. A CIT Other PIT VAT large share of TE resources was used with the ex- plicit purpose of supporting a specific sector of activities that were deemed as a priority by the D. Tax Expenditures by Function government. Total TE support to priority sectors Education/Science/Culture/Sport accounted for 1.43 percent of GDP in 2015 and was Other Transport/Environment provided primarily via reduced VAT rates applied Family/Social to the goods and services produced by the priori- Support to the Economy ty sector or activity. Health Agriculture As pointed out in the diagnostics provided in 0 0.5 1.0 1.5 2.0 2.5 the rest of this chapter, some of the TE schemes Percent of GDP currently in place raise equity and efficiency CIT Other PIT VAT concerns. There seems thus to be room to ratio- nalize the use of TEs in the Polish tax system and E. Tax Expenditures by Recepients recover some of the large resources they current- Other ly mobilize. Some possible interventions in this Businesses area have been outlined in this chapter. In gen- Households eral, a systematic review and a rigorous evalua- 0 0.5 1.0 1.5 2.0 2.5 3.0 tion of the current TE schemes could provide the Percent of GDP groundwork for future improvements of the Polish tax system and could produce considerable gains CIT Other PIT VAT for the Polish economy. Source: Polish Ministry of Finance. Tax assessment 59 POLICY 3. a progressive solidarity transfer on the contri- RECOMMENDATIONS butions paid on high-incomes. Increasing the progressivity of the tax wedge PIT and SCC and decreasing the tax burden on low incomes seem particularly relevant considering the high Increasing Progressivity degree of labor informality and relatively low of the Standard PIT Scheme labor force participation in Poland. The issue of informality is usually more salient for low-skilled Given the limited progressivity of the current sys- workers, who tend to work in low-paying jobs at tem, the government should consider reallocating higher risk of informality.58 At the same time, low- some of the tax burden away from low-income in- skilled workers are usually found to have higher dividuals. This can be done by intervening on both the elasticities of labor supply.59 Reallocating the tax PIT and SCC schedules. Increasing the size of tax-de- burden away from low incomes can thus have pos- ductible costs; increasing the basic tax allowance and itive effects on both these dimensions, favoring the redesigning its phaseout; and revising the stepwise sys- participation of low-skilled workers in the formal tem of marginal tax rates are viable options to reor- labor market. Increasing participation in the for- ganize the current PIT system. The analysis in Bachas mal economy would at the same time increase the et al. (2020) provides a detailed assessment of these total PIT tax base and would favor a more general- alternatives and suggests that redesigning the tax al- ized reduction in PIT. lowance might be one of the most effective ways to ob- tain a more progressive PIT in Poland. More specifical- The issues of informality and low labor force par- ly, the report proposes to increase the size of the basic ticipation remain important in Poland. While esti- tax allowance and reduce its phaseout rate.56 mates of the size of labor informality vary depending on the methodological approach being used, Poland Given the relatively low contribution of PIT to is usually found to have the largest informal econo- the total tax wedge, a meaningful improvement my among EU countries. According to estimates from in progressivity would also require an adjust- the European Commission (2017), in 2013 Poland had ment of the current SSCs schedule and a reduc- the highest rate of undeclared labor in the EU. The tion of SSC on low incomes. The government could report estimated that 27.3 percent of the gross value consider achieving this by introducing a social insur- added in the private sector was being produced by ance allowance that reduces social insurance contri- undeclared labor, compared to an EU-28 weighted butions on low incomes at constant expected replace- average of 14.3 percent and to the 16.9 percent and ment rates. The equilibrium of the pension system 21.3 percent in the Czech Republic and in Estonia, re- could be guaranteed by covering the costs of this spectively. Similarly, ILO (2018) estimates for 2016 policy with a transfer from the contributions paid show that 20.1 percent (15.9 percent outside agricul- on high-incomes.57 In practice, these additional re- ture) of jobs were informal, compared to an EU-27 sources could come from: average of 16.8 percent (14.6 percent outside agricul- ture).60 Besides reducing labor informality, increas- 1. an increase in the threshold under which incomes ing labor force participation remains an important are exempt from pension contributions (current- goal for Poland considering its relatively low partic- ly set at Zl 157,770); ipation rate compared to the EU-27 average and to many of its peers (Figure 2.1). As of 2019, the share 2. a harmonization of the social insurance contri- of inactive population in Poland stood at 29.2 per- butions system applicable to salaried employ- cent, 2.5 percentage points above the EU-27 average ees, self-employed and atypical labor contracts and 6.1 and 8.1 percentage points above the level in (see below); and the Czech Republic and Estonia, respectively. 60 Poland Public Finance Review Harmonize SSC and Health Contributions negative consequences on their probability to grow be- Treatments yond what can be feasibly achieved under an unincor- porated organizational form with negative effects on It is recommended that the government consider their long-term growth prospects.62 The structure of harmonizing the social security and health contribu- the current system is thus likely to at least mitigate and tion treatments for standard labor contracts, atypical possibly revert the negative effect on business growth contracts, and self-employment. A recent analysis of increasing PIT on the high-income self-employed. jointly carried out by the World Bank, the European The fact that under the current tax law, self-employed Commission, and the Polish Ministry of Finance workers can already opt for paying PIT under the pro- (World Bank 2020) assessed the impact of harmo- gressive system, an option that is advantageous for nizing the SSC system across typical and atypical low/medium income self-employed (earning less than contracts. The effectiveness of this policy change 100,000 Zl), suggests that the recommended reform might, however, come at the cost of an increase in would raise taxes primarily on high-income self-em- income inequality, as it would impose a higher tax ployed currently opting for the 19 percent flat rate or wedge on civil law contracts, which in turn are of- lump-sum options. Accordingly, the reform is likely ten used in low paying jobs.61 To mitigate its nega- to both increase PIT progressivity and raise addition- tive distributional effects, the harmonization would al tax revenue. These additional resources could then need be accompanied by the introduction of an SSC be used to finance part of the proposed increase in the allowance on low incomes. The introduction of the basic PIT allowance outlined in Bachas et al. (2020). allowance would help offset the effect of the harmo- nization on the cost of low-wage atypical contracts The government could also consider going a step and at the same time reduce the labor cost of stan- further and subject both net capital and labor in- dard contracts, making them more advantageous for come to the general progressive scheme, elimi- employers. The proposed SSC allowance would also nating the separate taxation of capital income. help increase the overall equity and progressivity Under this alternative, capital income and labor in- of the Polish PIT system. come — both from employment and self-employ- ment — would be taxed under the same progressive scheme. This harmonized system should be accom- PIT rate on self-employed panied by a fixed dividend tax credit compensating and capital income for the effects of the double taxation on the profits of incorporated businesses. Without this compen- The government should consider subjecting the in- sating tax credit, double taxation would increase the come earned from self-employed workers to the effective tax rate on profits earned from incorporat- general progressive scheme, thus removing the ed businesses above the tax rate on other sources of 19 percent flat rate and lump-sum options. Two income. Depending on the design of the dividend tax main arguments support this recommendation. First, credit, this system could achieve tax neutrality be- the current 19 percent flat rate reduces the degree of tween incorporated and unincorporated businesses, progressivity in the system, reducing the average tax thus reducing distortions in the organizational choic- rate on high-income self-employed earners. Second, a es of businesses. Under the proposed comprehensive flat 19 percent rate on self-employment is advantageous scheme, net capital gains and losses should be cal- relative to the double taxation on profits and capital culated separately before adding them to the other income for shareholders of incorporated businesses. sources of income. Capital losses should be deducted This lower tax rate on unincorporated businesses in- only against capital gains and any resulting net cap- creases the return on investment on small firms, but ital losses could be carried forward to offset future it also reduces their incentives to incorporate. While taxable capital gains. The net capital gains obtained this favorable treatment might push small business- from this separate calculation would then be added es to grow marginally bigger, it is also likely to have to labor income and taxed under the same regime. Tax assessment 61 A comprehensive income tax scheme would in- is that, if secondary earners tend to have a higher crease both the progressivity and horizontal eq- elasticity of labor supply, they should be taxed at uity of the PIT system, reducing the number of pref- lower marginal rates to maximize aggregate labor erential treatments available to high-income earners supply.66 Compared to an individual PIT system, for and ensuring that taxpayers earning similar incomes a same total household income, a joint system im- pay similar average tax rates.63 The capital taxation lit- poses a higher marginal tax rate on the secondary erature has long debated the optimality of a compre- earner and a lower tax rate on the primary earner.67 hensive PIT scheme, where labor and capital income As it imposes a higher (lower) marginal tax rate on are taxed under the same system. A recent contribu- individuals with higher (lower) labor supply elas- tion in Saez, Stancheva (2018) summarizes the main ticity, positive jointness is thus likely to reduce ag- theoretical results on the subject and shows that the gregate employment. Supporters of joint taxation desirability of a comprehensive PIT scheme depends usually claim that this system guarantees a more eq- crucially on the importance of equity concerns in the uitable distribution of the tax burden across house- design of the PIT system, as well as on taxpayers’ abil- holds. This argument is based on a ‘unitary’ view ity to shift their income from labor to capital income of the household, where household members per- bases. First, as equity concerns become more import- fectly share their income and jointly decide how to ant, the optimal system will tend towards a comprehen- spend it. Accordingly, horizontal equity would re- sive PIT scheme, even when this is less efficient than quire households earning the same income to pay a separate taxation of labor and capital. Second, even the same level of taxes. This view however ignores in the absence of equity concerns, a comprehensive the inequities that arise within households and the scheme can be optimal when taxpayers can easily shift perverse effects that these can have on gender in- between income sources. By altering the relative cost equality and gender gaps in the labor market. As of earning income from different income bases, a sepa- they disincentivize the participation of secondary rate taxation of capital and labor can distort taxpayers’ earners to the labor market and reduce their labor optimal choices. This is notably the case for incorpo- supply, joint systems tend in fact to magnify initial rated and unincorporated (pass-through) businesses, differences within the household and further po- where, as already mentioned above, the differential larize the relative bargaining power within couples. tax treatments can induce the owner to inefficient- ly alter the organizational structure of the activity. 64 These issues are particularly sensitive consid- ering that women are disproportionately rep- resented among secondary earners. The joint PIT Joint PIT Scheme scheme is likely to sustain the large gender partic- ipation gap currently observed in Poland, which is The government should consider removing the op- in turn one of the main drivers of the comparative- tional joint PIT declaration scheme. Poland current- ly low labor force participation in the country. As of ly gives couples the option to jointly declare annual in- 2019, the gap in gender participation between men come. A couple opting for the joint declaration is taxed and women stood at 13.8 percentage points, 2.9 per- under the progressive scheme on the total income of centage points above the EU-27 average and 7.6 per- its two members, doubling the level of the marginal centage points above the level in Estonia. rate threshold and basic allowance. In a progressive PIT system, jointly declaring income introduces some As noted in Bachas et al. (2020), the optional joint degree of what the tax literature defines as positive tax system in Poland also represents a simple way for jointness: The marginal tax rate on an individual’s in- high-income earners to cut their average tax rate, come increases with the income of her or his partner. reducing the number of taxpayers effectively subject to the 32 percent marginal PIT rate, and thus limiting Positive tax jointness raises concerns in terms the effective progressivity of the Polish PIT system. of both efficiency and equity. From an efficiency Considering these efficiency and equity arguments, perspective, optimal taxation theory usually finds it is recommended that the government considers re- positive jointness to be inefficient.65 The intuition moving the optional joint declaration scheme. 62 Poland Public Finance Review Capital Income from Foreign Sources The government should consider having one sin- gle entity collecting both PIT and SSC. Currently Poland should introduce passive foreign invest- two separate entities collect PIT and SSC, which in ment company (PFIC) rules to tax capital income the case of labor income are both collected as a de- from offshore funds on a current basis and re- duction from wages.68 To increase collection effi- quire residents to report the value of their for- ciency and reduce collection cost, the government eign portfolio of assets. International tax evasion could consider having only one agency collecting all and avoidance of capital income taxes represent a ma- deductions from wages. jor drain of public resources and undermine the pro- gressivity and fairness of tax systems. As mentioned above, the tax treatment of capital income from off- CIT shore funds in the current Polish tax system favors tax avoidance by allowing residents to legally avoid In designing a CIT system that sustains firms’ paying capital income taxes by investing their savings growth and job creation, the authorities could in an offshore fund. Countries may counter this tax consider refocusing their system of tax incen- avoidance strategy by introducing rules — so-called tives in favor of greater direct support to in- PFIC rules — that would tax resident individuals on vestments. The government could consider the a current basis, taxing their pro rata share of inter- possible negative and unintended effects that the est, dividends, and capital gains realized by offshore current preferential CIT treatment for small busi- funds in which they invest. Poland should consider nesses can have on their incentives to grow big. At introducing a similar set of rules in its tax code. The constant tax revenue, the preferential rate can dis- authorities should also consider requiring residents tort firms’ growth choices, reducing the incentives to report the total value of their foreign portfolio of to invest in business growth (above the preferen- assets. This new reporting requirement would pro- tial cutoff size) and to incorporate business activ- vide tax authorities with insights into the scale of ities, and can negatively affect the relative gains offshore tax avoidance on capital income from res- from starting-up businesses with high growth po- idents using offshore portfolio investment vehicles. tential. If the government’s objective is to promote Finally, to counteract international tax evasion of job creation through business growth, it could con- capital income taxes, Poland’s tax administration, in sider refocusing its system of CIT tax preferences coordination with EU authorities and other member to support businesses with higher growth potential, countries, should step up its efforts to pursue infor- especially at an early stage in their activity. While mation-exchange agreements with other countries. a system of age-based incentives can be problem- atic and hard to efficiently design, devoting more resources to policies supporting business growth Additional Recommendations. (e.g., via accelerated depreciation schemes) can be a step in this direction.69 The government should reconsider the sharp age eligibility threshold for the PIT exemption avail- If the government is committed to a size-based able to taxpayers up to 26 years old. The current CIT, it should consider less distortionary alter- threshold induces a sharp increase in the labor cost natives to the current system of tax preferenc- at 26 years of age and can distort labor supply and es. An alternative scheme based either on a prof- demand choices. A system that increases the PIT li- it eligibility threshold and two marginal tax rates ability more smoothly with age would reduce these above and below the cutoff, or on a tapered tax rate distortions while providing support to youth em- around the threshold, could be a more efficient way ployment. A formal assessment of the impact of the to implement size-based CIT tax preferences. This current system on the employability of individuals alternative would eliminate the current discon- around the 26 years old threshold could help inform tinuity in the average tax rate and thus mitigate policymakers on the possible distortionary effects some of the distortions associated with the pres- the current policy. ent CIT schedule. Tax assessment 63 The government should consider replacing the ments. In particular, the report proposes to calculate “Estonian CIT” scheme with a permanent acceler- around half of the local CIT share as a moving average ated depreciation of capital costs. If the introduc- of local CIT collection over a window of five years70. tion of the “Estonian CIT” is aimed at encouraging In the longer term, the government could consid- investment in productive capital, the government er a shift from a shared taxation of CIT to a shared should consider alternative approaches that can taxation of VAT, a less volatile source of tax revenue. achieve this objective more efficiently and at a lower cost for the public budget. Replacing the “Estonian The government could consider increasing the CIT” with a permanent accelerated depreciation (e.g., revenue autonomy of local governments. As point- full depreciation in the first year of investment) of ed out in World Bank (2018), this could be done in sev- capital costs is a more efficient alternative that the eral ways. First, the government could give local gov- government could consider in the future. ernments the possibility to adjust the local PIT tax rates to meet their budget needs via a “piggyback” PIT tax. Second, the government could reform the Indirect Taxes property tax, reforming the assessment methodology to better reflect the actual value of land and real es- The government should consider reversing its de- tate properties. Finally, the government could allow cision on the 2016 coal and gas excise exemptions. local governments to introduce a local business tax. The tax revenue recovered from the abolition of these exemptions could be used to finance tax credits for energy efficiency and clean energy production. This reallocation of resources would help transition the CONCLUSIONS Polish economy away from carbon-intensive energy sources and toward more dynamic sectors and activ- This review suggests that the Polish tax system ities with higher future growth potential. This mea- shows a limited degree of progressivity. The tax sure could be part of a broader package promoting a system generates a relatively low revenue compared green transition of the economy, while supporting to the EU average, while disproportionately relying workers and households in industries and regions on revenue from indirect taxes. Consumption is thus that will be most affected by this transition. subject to a relative high taxation, while CIT and PIT cover a comparatively small share of the state’s bud- get. On the aggregate, this low level of direct taxa- Subnational Taxation tion, together with SSC contributions that are in line with the EU average, results in a relatively low aver- When planning reforms of PIT and CIT, the gov- age tax wedge on labor income. However, the current ernment needs to consider their effect on the rev- PIT and SSC design fails to produce any notable level enue of local governments. Local governments re- of income tax progressivity, making the Polish PIT/ ceive 51.19 percent and 22.86 percent of the total PIT SSC system one of the least progressive in Europe and and CIT revenue collected in their jurisdiction, respec- imposing a relatively high tax wedge on low incomes. tively, and these resources represent an important The limited degree of progressivity in the tax system share of their budget. Changes in PIT and CIT rates is exacerbated by a comparatively high tax on con- and tax bases therefore need to be offset with target- sumption. While the Polish government introduced ed inter-governmental transfers or other compensa- two reduced VAT rates on basic products, VAT is usu- tory measures to support local governments’ budgets. ally found to be at best only mildly progressive. The low degree of progressivity in the PIT system and the The government should consider reforming tax high share of revenue collected from VAT thus result sharing to reduce the volatility of local budgets. in a system that imposes a high tax burden on low-in- As proposed in World Bank (2018), the government come households. It is thus important that the gov- should consider measures that could smooth out the ernment considers reforming its the current tax sys- volatility in the CIT revenue going to local govern- tem to increase its progressivity and equity. 64 Poland Public Finance Review The Polish tax system also includes several spe- ments aimed at improving their efficiency and eq- cial treatments and exemptions granted to spe- uity could provide important gains for the econo- cific taxpayers and taxed items. Besides their ef- my and free up additional resources. These could fect on the state budget, some of these preferential then be used to finance programs that can more treatments raise efficiency and equity concerns, effectively support the most vulnerable segments often disproportionately benefiting high-income of the population and stimulate firms’ growth and earners. A systematic revision of these special treat- job creation. Table 2.3 Main Recent Tax Reforms Personal Income Taxes and SSC 2020 New rules were introduced reducing the amount of SSC for self-employed earning less than Zl 120,000.* Introduction of a solidarity levy of 4 percent on income above Zl 1 million. Introduction of a tax exemption for labor income below Zl 85,558 earned by individuals up to 26 years old 2019 under a labor contract. Reduction of the lowest PIT rate from 18 percent to 17 percent. 2017 Progressive tax allowance replaced the universal flat allowance previously in place. 2009 System based on three income brackets is replaced by two income brackets taxed at 18 percent and 32 percent. 2004 Introduction of the optional flat 19 percent rate regime for self-employment income. Corporate Income Tax 2021 Introduction of the so-called “Estonian” special CIT scheme. The preferential CIT rate for small businesses is reduced from 15 percent to 9 percent. 2019 Introduction of an exit tax on the unrealized profits in relation to the moving of assets to another country. Introduction of a minimum taxation on commercial property (malls or office buildings). The tax is part of the 2018 corporate income tax. 2017 A preferential CIT rate of 15 percent is introduced for small businesses. 2004 The CIT rate is reduced from 27 percent to 19 percent. VAT 2020 Revision of the list of goods subject to the reduced VAT rates. Introduction of a list of VAT registered taxpayers (“White list”). Introduction of the Online Cash Registry. 2019 Mandatory Split Payment Mechanism is made mandatory for sectors considered particularly vulnerable to VAT evasion. The VAT rate on a set of products is revised from 8 percent to 23 percent. 2018 Optional Split Payment Mechanism is introduced. 2017 Introduction of penalties for the underestimation of VAT liability. 2016 Threshold for business-to-business cash transaction is reduced from €15,000 to Zl 15,000 (around €3,500). Standard VAT rate is increased from 22 percent to 23 percent and reduced rate is increased 2011 from 7 percent to 8 percent. A third reduced VAT rate of 5 percent is introduced for a list of specific items. 1993 The VAT tax is introduced. Tax assessment 65 Other Indirect Taxes: on products and production Introduction of a new tax on beverages with added sugar/sweeteners, caffeine, or taurine. 2021 Introduction of an additional levy on alcohol beverages sold in bottles of up to 300ml. Excise duty on novel tobacco products and liquids used in electronic cigarettes is introduced. 2020 The special hydrocarbons tax levied on profits derived from the extraction of gas and oil is introduced. The tax on the extraction of minerals is extended to gas and oil. 2019 Reduction of the rate on copper and silver extraction. The environmental usage fee on businesses engaging in prospecting, exploring, and extracting 2016 hydrocarbons is introduced. Excise energy tax become applicable to coal, lignite, and coke. 2012 A new tax on the extraction of silver and copper is introduced. Property and Other Local Taxes Introduction of a minimum taxation on commercial property (malls or office buildings). 2018 Part of the corporate income tax. The environmental usage fee on businesses engaging in prospecting, exploring, and extracting 2016 hydrocarbons is introduced. Note: As of February 2020, SSC due by self-employed earning less than Zl 120 000 may be calculated from 50 percent of self-employed income using standard rates that apply to employed workers. If this method is used, the SSC liability cannot be lower than the amount of SSC calculated from 30 percent of minimum wage. If the SSC liability is higher than the SSC calculated from 60 percent of average wage (standard SSC value for self-employed), then standard SSC rules apply. Notes 1. See Table 2.3 for a list of the main tax reform recently timates. The ILO estimate for the share of informal implemented in Poland. employment is in line with estimates in Williams et 2. On top of tax revenue, as of 2019 the Polish state bud- al. (2017), which find that in 2013 20.8 percent of to- get benefited from additional non-tax resources in the tal labor inputs in the private sector in Poland was form of property income, market output, output for undeclared. own final use, and other transfers received by the gov- 6. See estimates in PIE (2020) and CASE (2020). ernment for a total of 5.2 percent of GDP and an addi- 7. Tørsløv et al. (2020) estimate that in 2017 because of tional 0.85 percent in non-mandatory SSC. These sourc- profit shifting Poland lost tax revenue equivalent to es of non-tax revenue consist of the following ESA 2010 8 percent of the total CIT it collected, with around 90 categories: P.11, P.12, P.131, D.39r, D4r, D.7r, D.92r, D.99r. percent of these profits being shifted towards other 3. There are currently 2478 municipalities, 314 counties EU countries. The European Commission (2019) esti- and 16 regions in Poland. mated that in 2016 Poland lost €1.35 billion in tax reve- 4. Estimating the size of the informal economy is not an nue due to tax evasion from international wealth off- easy task. Alternative methodologies and different defi- shoring, defined as the underreporting of holdings nitions of informality exist, and the estimated size of held by the residents of a given country in other tax the informal sector will depend on the specific defini- jurisdictions. tion and methodology adopted. What most studies on 8. Some additional conditions need to be met to be eligi- informality in the EU seem to agree upon is that Po- ble to the joint taxation regime. For example, none of land has one of the biggest informal/shadow economies the individuals in the couple should conduct business among EU member states. For additional estimates of activities taxed under a scheme different than the gen- the size of the shadow economy in EU countries see a eral progressive scheme. recent analysis by the IMF in Kelmanson et al. (2019). 9. Civil law contracts include contracts of results and com- 5. See ILO (2018) for a discussion of the methodology mission contracts. Deductible costs cannot be deduct- and definition of informality underlying these es- ed from the income earned from pensions. 66 Poland Public Finance Review 10. This cost deduction does not apply to civil law contracts. 23. Profits taxed under the new scheme are subject to spe- In these cases, the deduction is equal to either the actu- cial rates of 25 percent (15 percent for small firms). Spe- al costs incurred in their activity or 20 percent of their cific deductions and simplified reporting rules apply contract value (increased to 50 percent if the contract under the new scheme, balancing these higher rates. involve the transfer of copyright material). Addition- Some deductions available to firms under the standard al special provisions apply to commuting workers with scheme, such as the IP box, do not apply to CIT taxpay- multiple employment relationships. ers under the new scheme. 11. This deduction does not apply to income earned under 24. The fact that the new optional scheme is available only a Contract of Result or to self-employed workers. to businesses that operate under certain ownership 12. See the chapter on fiscal expenditures in this report for structures, introduces yet another distortion, possi- more details. bly altering firms’ choices of their optimal corporate 13. Individuals opting for the flat rate system cannot deduct structure. any relevant expense from their taxable rental income. 25. The complexity of tax legislations and the number of 14. As of February 2020, SSC due by self-employed earning eligibility criteria might be particularly discouraging less than Zl 120,000 may be calculated from 50 percent for small businesses and could thus create additional of self-employed income using standard rates that ap- distortions. Understanding a new legislation and re-op- ply to employed workers. If this method is used, the SSC timizing a business plan accordingly represents a fixed liability cannot be lower than the amount of SSC cal- cost that small businesses might be more reluctant to culated from 30 percent of minimum wage. If the SSC pay given the relatively limited (compared to bigger liability is higher than the SSC calculated from 60 per- firms) benefits that they could enjoy. cent of average wage (standard SSC value for self-em- 26. These are defined as: employees’ wages and social con- ployed), then standard SSC rules apply. tributions; purchase of commodities and raw materi- 15. This provision applies only to a person that stipulates als; expertise, research, and opinions bought from sci- multiple contracts at the same time. If the person works entific units; payments for use of research equipment; under a single commission contract the standard social amortization of intangible assets and fixed assets, ex- contributions applies. cluding passenger cars, buildings, and constructions; 16. Including social security contributions decreases the costs of obtaining IP protection. differential in tax wedges from 1.3 to 1.1, due to the 27. Firms with revenue larger than Zl 2.5 million can ap- regressive nature of SSCs. The very low differential ply for RDC status. wedge from PIT is a direct consequence of the very 2 8. The definition of an SME follows the one used by the Eu- fast phase-out of the tax allowance at low-income ropean Commission. A firm is thus defined as an SME levels. Consequently, most of the progressivity in the if it has less than 250 workers and has either an annu- Polish PIT schedule is concentrated between Zl 0 and al turnover not exceeding €50 million or an annual to- Zl 13,000 with a very slow increase in the average tax tal balance sheet not exceeding €43 million. rate thereafter. A sizeable wedge differential would 29. Several authors have analyzed the effect of R&D tax thus be evident only if considering workers earning credits on innovation in eligible firms. Research on the well below 67 percent of the average wage and corre- topics has considered both the effect of R&D credits on sponding to low percentiles of the income distribution. the inputs of the R&D process (e.g. R&D expenditures) 17. Civil law contracts include commission contracts and and on (proxies of) the output of the R&D process (e.g. contracts of results that are often used as alternatives number of patent applications). Recent examples in- to temporary labor contracts. See World Bank (2020). clude evidence on input effectiveness in the UK (Gu- 18. See for example Garibaldi, Taddei (2013) for a discus- ceri, Liu (2019)) and in several EU countries (Appelt et sion of labor market duality in Italy and its implica- al. (2020)) and evidence on the number of new patent tions on the labor market conditions of the youth. applications in the UK (Dechezleprêtre et al. (2020)). See 19. See for example Lopes (2020) for evidence on the neg- Dechezleprêtre et al. (2020) for evidence on the positive ative effect of job insecurity on fertility decisions. spillover effects of an expansion of R&D credits in the 0. A limit of 5 million Zl or 50 percent of losses applies to 2 UK on firms that perform similar research activities. the amount deductible in each year. 3 0. See Dechezleprêtre et al. (2020) for evidence on the pos- 21. See ZEW (2019) for more details on the methodology itive spillover effects of an expansion of R&D credits in used to derive these estimates. the UK on firms that perform similar research activities. 22. The “Estonian” option is so-called because its design was 31. Davies et al. (2020) use data from European Patent Of- based on the default CIT regime in Estonia. While the fice and provide evidence of a positive effect of IP boxes new optional scheme in Poland and the default scheme on both the number of new patent applications and on in Estonia share a common rationale, the two systems the average probability of success of patent applications, differ in several important ways. which the authors interpret as a proxy of patents’ qual- Tax assessment 67 ity. The authors also suggest that to limit the risk of big to justify size-based preferential policies for businesses. firms simply relocating IP rights to low-tax jurisdictions, 42. The exemption applies to firms with annual turnover IP boxes should include a clause conditioning the eligibil- below €47,010. ity to the incentives on the local development of IP rights. 43. These sectors are listed in Annex 15 to the VAT act. 32. Given that various form of public support to R&D in Po- 44. An ex-post evaluation of a mandatory SPM system in land, it is important to note that an analysis of the effec- Italy applicable to payments by public authorities to tiveness of the IP box would need to consider the over- their suppliers found it to be successful at increasing all framework of tax and non-tax support. The analysis VAT compliance (Carfora et al. (2017), Deloitte (2017)). should also consider the risk of potential overlap be- In the first year after its introduction, the system is esti- tween these different support measures, which could mated to have increased VAT revenue by 2.2 billion Eu- result in an inefficiently high level of support ros, equal to around 30 percent of the total VAT paid via 33. See Jensen, Winiarczyk (2014) for an assessment of the SPM by public authorities in that year. SEZs in Poland. While the study finds a positive effect 45. See the analysis commissioned by the European Com- of SEZs on FDI, it does not find any significant effect on mission in Deloitte (2017) for a discussion of these points. employment and wages. The authors also suggest that It is worth pointing out that the existing evidence on SEZs should be used as temporary policies and that for the relative costs and benefits of SPM policies has so them to be effective in the medium-to-long run they far focused on a comparison between reduction in VAT should be supported by more comprehensive local de- gap on one side and increase in compliance and admin- velopment policies. Other authors have analyzed the istrative costs on the other. It has however overlooked impact of Chinese SEZs, finding positive impacts on the effects of SPM policies on efficiency and welfare, ig- FDI and productivity (see for example (Wang 2013)). noring, for example, any efficiency gain arising from 34. These estimates account only for the direct foregone a reduction in fraudulent anticompetitive behaviors revenue associated with these measures and do not from businesses avoiding VAT payments. consider the equilibrium effect that these measures 46. See for example Abramovsky et al. (2017) for a discus- can have on the CIT tax base. sion on the efficiency and equity of special VAT rates 35. See PIE (2020). and exemptions. 36. See IMF (2018) for a review of the limitations of the top- 47. See Crossley et al. (2009) and Thomas (2020) for a dis- down approach used to estimate the CIT gap in PIE (2020). cussion on this point. 37. Promoting international and EU cooperation on the is- 48. The assessment carried out in this report shows that on sue of profit offshoring remains key to limit tax compe- the aggregate the Polish system of indirect taxes, which tition and reduce CIT revenue losses from profit shift- includes both VAT and other indirect taxes, is regressive. ing. See the discussion in De Mooij et al. (2021). 49. The European Central Bank has criticized the new tax 38. When size cannot be fully controlled, this discontinu- for its potential negative effects on the provision of ity can also create uncertainty as firms will be unsure credits to the economy and on the financial stability about their eligibility to the preferential scheme. of the banking system. The European Commission has 39. Besides creating a jump in the average CIT rate, the current also raised concerns that the tax might distort competi- system alters the relative expected return on investments tion in the financial sector, due to the existence of sev- for firms above and below the eligibility threshold. Rela- eral exemptions applicable to state banks, investment tive to a common CIT rate between 9 percent and 18 per- funds, and pension funds. cent, under the common scheme, net returns on invest- 0. The tax exemption was introduced on January 1, 2016 in 5 ments are higher for small firms and smaller for big firms. contrast with European Council directive 2003/96/EC. This additional effect is likely to incentivize the growth 51. See OECD (2018) for more details on how the carbon of small firms up to the eligibility thresholds and disin- price gap is calculated. centivize the growth of firms above the revenue cutoff. 52. Tax expenditures aimed at fostering the green transition 40. Recent evidence has shown the importance of ex-an- of the economy include: i) exemptions from personal in- te firms’ types in explaining the growth dynamics of come tax (PIT) on subsidies for thermal modernization businesses (Sterk et al. (2021)) and have studied how provided through the “Clean Air” program and deduc- entrepreneurs’ decisions to start-up high growth firms tions on the cost of thermal modernization from taxable change with the relative profitability of different start- income; ii) investment relief from agricultural tax (local up opportunities (Sedlácek, Sterk (2017)). tax) on investment in equipment used for production of 41. See Haltiwanger et al. (2013) for evidence from the US. natural energy; iii) preferential 8% VAT rate on thermal Recent evidence based on firm-level data has shown the modernization of buildings covered by the social hous- lack of foundation for the commonly held belief that small ing program; and iv) exemptions from excise tax on elec- firms are the engine of job creation. The belief that size, tricity produced from renewable energy sources and on rather than age, is the key observable characteristic defin- passenger electric vehicles and hydrogen-powered ve- ing the job creation potential of firms has often been cited hicles and hybrid vehicles (up to 2000cm3). 68 Poland Public Finance Review 53. These tax expenditures include: i) reimbursements of dress one of the main critiques that a system subjecting the excise tax for fuel oil (diesel) consumed for agri- self-employed income and capital income to different cultural purposes; ii) preferential 8% VAT rates in mar- tax rates would face. Self-employment usually requires itime and domestic airline transport of passengers; and the use of both capital and labor. Subjecting self-em- iii) exemption of coal and coke products consumed in ployment income to the employed income scheme thus residential sector from energy tax. implies that the capital component of self-employment 54. Poland is already lagging most other EU countries in income is considered for tax purposes as labor income. terms of its energy transition. It currently ranks second- Harmonizing the tax regime across these three sourc- to-last among EU countries in the Energy Transition In- es of income would address this possible critique. dex published annually by the World Economic Forum 64. See Saez et al. (2012) for a discussion of income shifting (2019). See PIE (2020) for a discussion on this aspect. and of the response of different tax bases to the struc- 55. Data is from the analytical project “Division of Powers” ture of the tax system. commissioned by the European Committee of the Regions. 65. Works in this literature usually find that some (lim- 56. Any policy reforming the current PIT system should also ited) degree of negative jointness maximizes welfare. consider its impact on the resources of local governments. See for example the analysis in Kleven et al. (2009) and Local governments receive 51.19 percent of the total PIT Gayle, Shepard (2019). revenue collected in their jurisdiction and these resourc- 66. Relative to an individual system with no tax jointness, es represent an important share of their budget. Changes negative jointness achieves this goal more efficiently in PIT rates will therefore have to be offset with target- by counteracting the negative income effect of having ed inter-governmental transfers or other compensato- a higher-income partner on the labor supply decisions ry measures to support local governments’ budgets. of secondary earners. 57. The proposed reform would thus imply the introduction 67. The marginal tax rate on both household members will be of a transfer from the contributions paid by high-in- equal to the rate applicable to a single individual earning come earners towards low-income earners’ retirement their average income. If the two members do not earn accounts. Given the low expected replacement rates es- the same income and the system is (strictly) progres- timated for the current Polish pension system, these sive, this rate will be higher for the secondary earner transfers would be necessary to make sure that the re- and lower for the primary earner compared to the mar- duction in SSC for low-income earners does not trans- ginal rate they would get if they were taxed separately. late into a reduction in their future pension transfers. 68. Currently the National Revenue Administration (KAS) is 58. See World Bank (2019) for a discussion of informali- charged with the collection of PIT while ZUS collects SSC. ty and of a worker’s characteristics usually associated 69. An age-based system of tax preferences, where the eli- with a higher risk of informality. gibility to certain tax incentives is restricted to new or 59. See Attanasio et al. (2018) for recent evidence on the young businesses can result in artificial business churn- heterogeneity in labor supply elasticities for women. ing, with existing firms attempting to qualify for the tax 60. See ILO (2018) for a discussion of the methodology and reliefs available to new firms. To gain eligibility to the definition of informality underlying these estimates. tax incentives, business owners can for example either The ILO estimate for the share of informal employment close an existing business and invest the freed-up cap- is in line with estimates in Williams et al. (2017), which ital in a newly incorporated/established firm or decide find that in 2013 20.8 percent of total labor inputs in the to spin off certain business activities into a new busi- private sector in Poland was undeclared. ness entity. 61. The main reform scenarios considered in World Bank 70. World Bank (2018) contains a detailed discussion of this (2020) assume that the additional revenue from the re- recommendation. form is used to finance a uniform decrease in the PIT rate. 62. Examples of studies considering the effect of reducing the difference in tax treatments between incorporated and unincorporated businesses include Carroll, Joul- References faian (1997) and Chen et al. (2018) . Studies on the subject Abramovsky, Laura and Phillips, David and Warwick, Ross. usually find that policies harmonizing these tax treat- 2017. “Redistribution, efficiency and the design of VAT: ments have important positive effects on the number a review of the theory and literature.” IFS Briefing Note of unincorporated businesses deciding to incorporate, BN212. on their future growth prospects and on their contri- Appelt, , Silvia, Bajgar, Matej, Crisculo, Chiara, Galindo- bution to net job creation. Some evidence on the effect Rueda, Fernando. 2020. “The effects of R&D tax in- of decreasing PIT on entrepreneurs on the growth of centives and their role in the innovation policy mix: small firms is provided in Carroll et al. (2000). Findings from the OECD microBeRD project, 2016 – 19.” 63. Imposing a common progressive tax scheme for em- OECD Science, Technology and Industry Policy Papers, No. ployed, self-employed and capital income would also ad- 92, OECD Publishing, Paris. Tax assessment 69 Attanasio, O., Levell, P. Low, H. and Sanchez-Marcos, V. Gayle, George-Levi and Shepard, Andrew. 2019. “Optimal 2018. “Aggregatung Elasticities: Intensive and Extensive Taxation, Marriage, Home Production, and Family Margins of Women’s Labor Supply.” Econometrica, Vol. 86, Labor Supply.” Econometrica, Vol. 87, No. 1 (January, No. 6, 2049 – 2082. 2019), 291 – 326. Bachas, P., Boratyński, J., Goraus-Tańska, K., Klejdysz, J. Guceri, Irem, Liu, Li. 2019. “Effectiveness of Fiscal Incen- Lewczuk, E., Savard, L. 2020. “Increasing the Progressivity tives for R&D: Quasi-Experimental Evidence.” Amer- of Personal Income Taxation in Poland.” World Bank Report. ican Economic Journal: Economic Policy, 11(1): 266 – 291. Barrios, Salvador and Denis Céline and Ivaškaitė-Tamošiūnė, Haltiwanger, John, Jarmin, Ron S., Miranda, Javier. 2013. Viginta and Reut, Andriana and Vázquez Torres, Estefanía. “Who Creates Jobs? Small Versus Large Versus Young.” 2019. “Housing taxation: a new database for Europe.” The Review of Economics and Statistics, 95(2): 347 – 361. JRC Working Papers on Taxation and Structural Reforms Harding, Michelle and Marten, Melanie. 2018. “Statutory No 08/2019. tax rates on dividends, interests and capital gains: the Carfora, Alfonso and Marigliani, Massimiliano and Pisani, debt equity bias at the personal level.” OECD Taxation Stefano and Spingola, Andrea. 2017. “Gli effetti del- Working Papers No. 34. lo split payment sulla compliance IVA.” Argomenti di Huizinga, Harry. 2002. “A European VAT on Financial Ser- Discussione, Agenzia delle Entrate, N. 01/2017. vices?” Economic Policy, October 2002. Carroll, Robert, Holtz-Eakin, Douglas, Rider, Mark, Rosen, IFS. 2011. “A retrospective evaluation of elements of the EU Harvey S. 2000. “Personal Income Taxes and the Growth VAT system.” European Commission, TAXUD/2010/DE/328. of Small Firms.” NBER Working Paper 7980. ILO. 2018. “Overview of the informal economy in Poland.” Carroll, Robert, Joulfaian, David. 1997. “Taxes and Corporate ILO Factsheet. Choice of Organizational Form.” U.S. Department of the ILO. 2018. “Women and men in the Infromal Economy: a Treasury Office of Tax Analysys Working Paper 73. Statistcal Picture (Third Edition).” International Labour CASE. 2020. “Study and Reports on the VAT Gap in the EU- Office — Geneva. 28 Member States.” European Commission, Directorate- IMF. 2018. “Estimating the Corporate Income Tax Gap: The General for Taxation and Customs Union Report. RA-GAP Methodology.” Technical Notes and Manuals. Chen, Daphine, Qi, Shi, Schlagenhauf, Don. 2018. “Corpo- Jensen, Camilla and Winiarczyk, Marcin. 2014. “Special Eco- rate Income Tax, Legal Form of Organization, and Em- nomic Zones — 20 years later.” CASE Network Studies & ployment.” American Economic Journal: Macroeconomics, Analyses No. 467. 10(4): 270 – 304. Kelmanson, Ben and Kirabaeva, Koralai and Medina, Crossley, Thomas F. and Phillips, David and Wakefield, Leandro and Mircheva, Borislava Mircheva and Weiss, Matthew. 2009. “Value added tax.” The IFS Green Budget Jason. 2019. “Explaining the Shadow Economy in Europe: 2009, Institute for Fiscal Studies. Size, Causes and Policy Options.” IMF Working Papers Davies, Ronald B. and Kogler, Dieter F. and Hynes, Ryan. WP/19/278. 2020. “Patent Boxes and the Success Rate of Applications.” Kleven, Henric Jacobsen and Kreiner, Klaus Thustrup and CESIFO Working Papers, n. 8375. Saez, Emmanuel. 2009. “The Optimal Income Taxation De Mooij, Ruud and Alexander Klemm, and Victoria Perry of Couples.” Econometrica, Vol. 77, No. 2 (March, 2009), and eds. 2021. “Corporate Income Taxes Under Pressure: 537 – 560. Why Reform is Needed and How it Could be Designed.” Lopes, Marta C. 2020. “Job Insecurity and Fertility Decision.” Washington, DC: International Monetary Fund. Working Paper. Dechezleprêtre, Antoine, Einiö, Elias, Martin, Ralf, Nguyen, OECD. 2018. “Effective Carbon Rates 2018: Pricing Carbon Kieu-Trang, Van Reenen, John. 2020. “Do tax incen- Emissions Through Taxes and Emissions Trading.” OECD tives increase firm innovation? An RD Design for R&D.” Publishing, Paris. Working Paper. OECD. 2020. “OECD R&D tax incentives database, 2020 edi- Deloitte. 2017. “Analysis of the impact of the split payment tion.” OECD Publishing, Paris. mechanism as an alternative VAT collection method.” OECD. 2019. “Pensions at a Glance 2019: OECD and G20 European Commission, Technical Report. Indicators.” OECD Publishing, Paris. European Commission. 2017. “Analysis of the use of Auction PIE. 2020. “The CIT gap in Poland in 2014 – 2018.” Polish Revenues by the Member States.” Directorate-General Economic Institute Report. for Climate Action. PIE. 2020. “Time for Decarbonisation.” Polish Economic Fenochietto, Ricardo and Pessino, Carola. 2013. “Under- Institute Report. standing Countries’ Tax Effort .” IMF Working Paper Polish Ministry of Health and WHO. 2010. “Global Adult WP/13/244. Tobacco Survey Poland 2009 – 2010.” Garibaldi, Pietro and Taddei, Filippo. 2013. “Italy: A dual Saez, Emmanuel and Slemrod, Joel and Giertz, Seth H. labour market in transition: Country case study on la- 2012. “The Elasticity of Taxable Income with Respect bour market segmentation.” ILO Employment Working to Marginal Tax Rates: a Critical Review.” Journal of Paper No. 144. Economic Literature 50:1, 3 – 50. 70 Poland Public Finance Review Saez, Emmanuel and Stancheva, Stefanie. 2018. “A simpler Williams, C. C., Bejakovic, P., Mikulic, D., Franic, J., Kedir, A, theory of optimal capital taxation.” Journal of Public and Horodnic, I. A. 2017. “An evaluation of the scale of Economics Volume 162, Pages 120 – 142. undeclared work in the European Union and its struc- Sedlácek, Petr and Sterk, Vincent. 2017. “The Growth Potential tural determinants: estimates using the Labour Input of Startups over the Business Cycle.” American Economic Method.” European Commission, Directorate-General for Review, 107(10): 3182 – 3210. Employment, Social Affairs and Inclusion. Sterk, Vincent, Sedláček, Petr and Pugsley Benjamin. 2021. World Bank. 2020. “Economic and fiscal effects of reducing “The Nature of Firm Growth.” American Economic Review, labor market segmentation in Poland: Results based on Forthcoming. the Polish Applied General Equilibrium (PAGE) model.” Thomas, Alastair. 2020. “Reassessing the regressivity of World Bank, European Commission and Polish Ministry of the VAT.” OECD Taxation Working Papers, OECD Paris. Finance Technical Paper. Tørsløv, T., Wier, L. and Zucman, G. 2020. “The Missing World Bank. 2018. “Fiscal Capacity.” Fairness and Efficiency: Profits of Nations.” Dataset. Reforming Fiscal Equalization in Poland (RAS). Trinomics. 2020. “Energy subsidies, Energy costs, taxes and World Bank. 2019. “Growing in the Shadow: Challanges the impact of government interventions on investments: of Informality.” Chapter 3, Global Economic Prospects, final report.” Directorate General for Energy (European January 2019. Commission). World Economic Forum . 2019. “Fostering Effective Energy VAT Expert Group. 2020. “VAT treatment of financial ser- Transition: 2019 edition.” Insight Report. vices — a reflection on possible options for review.” Zee, Howell. 2004. “Financial Services and the Value- European Commission, VEG NO 089, 29 April 2020. Added Tax.” In Taxing the Financial Sector : Concepts, Vellutini, C., Casamatta, G., Bousquet, L. and Poniatowski, Issues, and Practice. USA: INTERNATIONAL MONETARY G. 2019. “Estimating International Tax Evasion by FUND. Individuals.” European Commission, Taxation Papers, ZEW. 2019. “Effective tax levels using the Devereux/Griffith Working Paper n. 76 – 2019. methodology.” Project for the EU Commission TAXUD/2019/ Wang, Jin. 2013. “The economic impact of Special Economic DE/312. Zones: Evidence from Chinese municipalities.” Journal of Development Economics Volume 101, Pages 133 – 147. Tax assessment 71 CHAPTER 3 GENERAL GOVERNMENT EXPENDITURE OVERVIEW about a third lower than the OECD average. Social pro- tection spending accounted for 40 percent of the total Poland’s government expenditure as a share spending in 2019, a level comparable to the EU aver- of GDP is close to the OECD average, but the age, and higher than that of regional peers (Figure 3.5). share of rigid spending has been rising Meanwhile, social spending has recorded the largest in- crease in its share in the 2015 – 2019 period (Figure 3.6). Poland’s general government expenditure level is similar to that of countries at a similar level of Figure 3.3 Structure of general government development. General government expenditure av- expenditure in Poland eraged around 43 percent of gross domestic product 60 (GDP) over the decade prior to the COVID-19 crisis, 50 Percent of total close to the 44 percent OECD average (Figure 3.1 and 40 Figure 3.2). The expenditure structure differs from that 30 of the OECD average, as the share of social and capi- 20 tal expenditure in total expenditure exceeds that in 10 other OECD countries. Meanwhile, the share of cur- 0 rent expenditure on public services is lower (Figure 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 3.3 and 3.4). Some public services are seemingly un- derfunded, with health care spending as a share of Social protection Capital expenditure GDP 1.9 percentage points lower than the OECD aver- Interests Public services age. Similarly, environmental protection spending is Sources: World Bank analysis; OECD. Figure 3.1 General government expenditure Figure 3.4 Structure of general by country, 2019 government expenditure in some countries Mexico United States Korea Korea Czech Republic Chile Poland Iceland Germany Australia Hungary Israel OECD Average Hungary Structural peers Czech Republic Aspirational peers Latvia Estonia 0 10 20 30 40 50 Slovak Republic OECD Average Percent of GDP Lithuania Ireland Source: OECD. United Kingdom Netherlands Belgium Figure 3.2 General government expenditure Slovenia in selected groups of countries Norway Poland 52 Sweden Portugal 50 Spain Austria Percent of GDP 48 Japan 46 France Luxembourg 44 Italy Denmark 42 Germany 40 Finland 38 0 20 40 60 80 100 Percent of total 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 EU27 Poland CEE10 Social protection Capital expenditure OECD Average Interests Public services Source: OECD. Sources: World Bank analysis; OECD. 74 Poland Public Finance Review Figure 3.5 Functional breakdown of general Figure 3.6 Change in the functional government expenditure in 2019 breakdown of general government 100 expenditure, 2015 – 19 Social protection Education 4 Social protection Recreation, culture Education 80 and religion 3 Percent of total expenditure Recreation, culture Health and religion Housing and 2 Health Percentage points of total 60 community Housing and amenities 1 community Environmental amenities 40 protection Environmental 0 Economic affairs protection Public order -1 Economic affairs and safety Public order 20 Defence and safety -2 General public Defence services General public 0 -3 EU27 CEE10 Poland services Source: Eurostat. -4 EU27 CEE10 Poland Source: Eurostat. A large share of expenditure is rigid, reducing the scope for discretionary policy on the spend- ing side. The share of rigid expenditure has increased Figure 3.8 Rigid expenses in OECD countries, over the past decade by 2 percentage points to 75 per- 2019 cent of spending by 2019, with Poland ranking in Belgium the middle of OECD countries (Figure 3.7 and 3.8). France Greece Highly rigid expenditure — which includes com- Spain pensation of employees, social transfers and inter- Portugal Germany est payments — accounts for 69 percent of all expens- Italy Austria es. Medium-rigid expenditure (economic subsidies Luxembourg and other current transfers) accounts for an addi- Japan Lithuania tional 6 percent. Among the rigid expenditure, the Slovak Republic Netherlands importance of social transfers has been increasing, Slovenia more than offsetting the decline in interest payments. Denmark Poland OECD average United States Figure 3.7 Structure of highly rigid expenses Ireland Switzerland in Poland Czech Republic Norway 45 Finland Percent of total expenditure 40 Latvia 35 United Kingdom Korea 30 Estonia 25 Chile 20 Sweden Australia 15 Israel 10 Iceland 5 0 20 40 60 80 100 0 Percent of total 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Medium* High** Compensation of employes Source: World Bank analysis; OECD. Social benefits Interests Note: *subsidies and other current transfers Source: World Bank analysis; OECD. ** compensation, social benefits and interests General government expenditure 75 In response to the COVID-19 pandemic, government General government spending is expected to re- expenditure increased by 6.9 percentage points of main high in 2021 and decline gradually by 2024. GDP to 48.7 percent in 2020. Fiscal buffers that had The government projects expenditure of 48.4 percent been built during the years leading up to the COVID-19 of GDP in 2021, declining to 43.5 percent of GDP by crisis allowed for a relatively large fiscal response: 2024 (2021 Convergence Program). In early November above-the-line measures were estimated at 6.4 per- 2020, the authorities announced additional support cent of GDP, larger than in most other EU countries. following the second wave of the pandemic, effective In addition to an increase in pandemic-related health as of late 2020, with eligibility limited to companies spending, crisis-related spending focused on preserv- operating in the sectors most impacted by the partial ing jobs and incomes. Expenditure support measures lockdown. Also extended were wage subsidies, social included among others: extended childcare allowances security exemptions, and other existing measures, for parents; support for the economic downtime or re- as well as some incentives for firms to reorient their duced working time; financial support for the costs of lines of business. Although the expiry of temporary employees’ salaries and social insurance contributions measures together with a resumption in growth are due in the case of micro-, small-, and medium-sized expected to contribute to the decline in spending to enterprises with employees; financial support for the 45.4 percent of GDP in 2022, and to 43.5 percent in operating costs of business activity of the self-em- 2024, spending will be 1.7 percentage points higher ployed; increasing monthly subsidies for the salaries compared with the pre-pandemic level. of employees with disabilities; and downtime benefits for persons employed under civil law agreements and Social programs introduced in recent years con- self-employed persons. Compared to other countries tinue, except for the 14th pension, which is sched- (Figure 3.9), aid in Poland focused more on companies uled as a one-off measure for 2021. In the period (subsidies, capital transfers) than on social transfers. 2021 – 24, the government assumes the continuation of the payment of child-rearing benefit in the amount Figure 3.9 2020 change in general of Zl 500 per month1 and the so-called 13th pension government expenditure by type: EU countries (2021 Convergence Program)2. Family and care ben- Greece Greece efits and payments from the Alimony Fund are ex- Spain Spain pected to amount to Zl 12 billion annually. Meanwhile, Malta Malta Austria Austria payment of the so-called 14th pension (Zl 10.6 billion) Lithuania Lithuania Slovenia is scheduled as a one-off measure in 2021. Italy Italy Croatia Croatia Belgium EU27 EU27 Poland France France Cyprus Cyprus Bulgaria Bulgaria Netherlands Netherlands SOCIAL PROTECTION Estonia Estonia Czech Republic Romania EXPENDITURE Germany Germany Portugal Portugal Hungary Hungary The share of social spending that is not Slovak Republic Luxembourg Luxembourg means-tested has risen, affecting spending Latvia Latvia efficiency Denmark Denmark Ireland Ireland Finland Finland Sweden Social protection expenditure increased to -2 0 2 4 6 8 10 12 14 16.7 percent of GDP (Figure 3.10), narrowing the Percent of GDP gap with the EU average. Social transfers are 3.5 per- Social benefits Subsidies centage points higher than the average for the CEE Compensation of employees countries that are EU members. The increase in social Capital transfers Gross capital formation expenditures was accompanied by a change in their Other net Total structure: expenditures on families with children Source: World Bank analysis; Eurostat. increased markedly, while most other transfers de- 76 Poland Public Finance Review Figure 3.10 Social transfers, 2019 or latest for children from disadvantaged backgrounds. Korea Income inequality in Poland, as measured by the Hungary Gini coefficient (28.1), is lower than in the CEE10 and Czech Rep. the EU average (Figure 3.12). Moreover, the share of Poland Germany people at risk of poverty and social exclusion has OECD - Average declined markedly to 18.2 percent in 2019, below the Aspirational peers EU average of 21.4 percent. New programs (Family 0 5 10 15 20 25 500+, Good Start) also contributed to a significant Percent of GDP reduction in the level of child poverty (Figure 3.13), Source: OECD. especially in the case of families with parents with Note: Includes cash and non-cash. lower educational attainment. With the introduc- tion of the Family 500+ benefit, the reduction in clined, in part due to a favorable labor market and to a poverty is attributable equally to changes in dis- tightening of the benefit-granting system (Figure 3.11). tribution and to income growth. Both relative and Starting in 2019, along with the introduction of new subjective poverty have been reduced by the pro- benefits and higher indexation of pensions, old-age re- gram (Paradowski 2020)5 . lated expenditure also increased. Poland’s benefits for families and children amount to close to 3 percent of Figure 3.12 Gini coefficient of equalized GDP, one of the highest in OECD countries. Meanwhile, disposable income old-age related expenditure (close to 10 percent of GDP) is now at levels comparable with that of coun- a 2015 After social tries further along in the aging process. Poland also transfers a 2019 has one of the lowest retirement ages in the OECD; in October 2017, the retirement age was rolled back to 2015 60 years for women and 65 for men, increasing gener- Before social transfers al government expenditure (0.5 – 0.6 percent of GDP3) 2019 and affecting the labor market.4 0 10 20 30 40 50 60 Figure 3.11 Changes in social protection Scale from 0 to 100 expenditure Poland CEE10 EU27 1.5 Source: Eurostat. 1.0 1 Figure 3.13 Households at risk of poverty and Percent of GDP 0.5 social exclusion 28 0.0 0 26 Percent of GDP -0.5 24 -1.0 -1 22 2009–2012 2013–2015 2016–2019 20 Old-age Sickness and disability 18 Survivors Family and children 16 Unemployment Other Total 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: OECD. Note: In relation to the previous period. Total households: EU Poland Higher social transfers to families contribute to Households with dependent children: EU Poland poverty reduction and social inclusion, as well as to greater accumulation of human capital Source: Eurostat. General government expenditure 77 Social benefits could be contained, with limited There are several fiscal instruments to support impact on achieving equity objectives, through families with a large overall cost relative to re- better targeting. Less than 5 percent of social pro- sults (Brzezinski, 2019).9 These include tax breaks, tection benefits are currently means-tested (Eurostat, family benefits, and the Family “500+” benefit. Family 20206). Until 2016, noncontributory family benefits benefits, which were the main source of financial and social assistance were low compared with oth- support for families with children until 2016, include: er high-income countries. For many low-income households this translated into a low level of redis- 1. regular financial benefits for low-income fami- tribution and a negative cash position after taxes and lies, i.e., family allowance and its supplements; transfers (Poland SCD 2017). In its original form, the 2. benefits for families of disabled children related Family 500+ program, introduced in April 2016, re- to care responsibilities (nursing allowance, nurs- duced coverage gaps and increased social expendi- ing benefit, special care allowance); ture, applied on top of previously existing programs. 3. one-off payments after the birth of the child; and Since mid-2019, the income criterion for the pro- 4. parental benefit for parents not entitled to usual gram has been removed, nearly doubling the pro- maternity/parental leave and benefit. gram’s cost and transforming it into a general child support benefit easy to distribute, but very costly The 500+ program was introduced as an overlapping and not well-targeted towards the poor. The bottom benefit (not affecting eligibility to any other) in an 20 percent receive only 11.7 percent of all annual ex- already complex social assistance and family sup- penditures on the program. Families in the upper port system. Moreover, there were other important half of the income distribution receive 58 percent changes in recent years that also increased financial of the Family 500+ program funds, while the poor- support available to families. Child tax credit val- er half of the population receives 42 percent of the ues were increased, and so were the family benefit funds (Krol et al., 20217). Expanded benefit eligibili- amounts and eligibility thresholds. Additionally, the ty resulted in a larger share of the families and chil- Good Start program was implemented, providing a dren benefits that is not means-tested (Figure 3.14) financial transfer to all parents of school-age chil- and contributed to reducing the efficiency of the dren at the beginning of school year (Goraus et al. overall social assistance system. On the other hand, forthcoming). The redesign of the full set of policies eliminating extreme child poverty could be achieved for families with children, taking a comprehensive with 12.4 percent of the total annual program cost, approach, could increase their effectiveness while according to a study analyzing the original version considering their implications for labor force par- of the program (Brzezinski, 2019)8. ticipation. An additional benefit for all pensioners was introduced in 2019, the so-called 13th retirement Figure 3.14 Family and children benefits pension. Yet another benefit (the 14th pension) equal 100 to the minimum pension was paid late 2021, bene- fitting an estimated 87 percent of pensioners, while 80 an additional 8 percent of pensioners will receive a reduced benefit.10 Percent of total 60 40 The introduction of the new child benefits has significantly changed the relative situation of 20 households with and without children. The in- troduction of additional child benefits has signifi- 0 cantly improved the net position of households 2014 2015 2016 2017 2018 2019 2020 with children towards the budget, which was not Non means-tested Means-tested the case with families without children (Figure 3.15 Sources: Eurostat; 2019, 2020 World Bank estimates. and 3.16). In the case of low-income single parent 78 Poland Public Finance Review families, the initial change (in 2016/2017) was fun- Figure 3.15 Gross and net position damental — going from net payers to net beneficia- by type of household ries. This was because from the very beginning, the Single person, 167%, no child benefit was payable as of the first child. On the oth- Single person, 100%, no child er hand, with a constant value of the benefit and Single person, 67%, no child increasing average earning, the relative impact of Single person, 67%, 2 children Married couple, 100%/0%, 2 children the benefit on the position of the poorest house- Married couple, 100%/67%, 2 children holds gradually diminishes. The situation is dif- Married couple, 100%/100%, 2 children ferent for higher-income households that became Married couple, 100%/67%, no child beneficiaries of 500+ for the first child only from -30 -20 -10 0 10 20 30 2019. In this case, the full benefit of the 500+ pro- Percent of total labor cost gram was not felt until 2020 (the first full year of 500+ for all children). SSC + income tax SSC + income tax - cash benefits The risk of poverty and exclusion for other vul- Sources: World Bank analysis; OECD. nerable groups remains and there are consider- Note: 67, 100 and 167 percent stand for income level of the able spatial differences. Despite generally positive adult family members (as percent of average earning). trends regarding the risk of poverty and social ex- clusion, challenges remain for people in single-per- Figure 3.16 Net position son households. In this group, the percentage of by type of household people at risk of poverty and exclusion is twice 30 the average and has increased since 2015. The prob- 20 lem affects people whether under or over the age Percent of total labor cost of 65. Spatially, the differentiation of the risk of 10 poverty and social exclusion is still high. In 2019, 0 the highest share of those at risk is found in east- -10 ern and northeastern Poland. The lowest share is in Warsaw and southwestern Poland. The main so- -20 cial exclusion challenges are now related to provid- -30 ing support to single and socially-excluded adults -40 and improving the housing situation of the poor. 2015 2016 2017 2018 2019 The design of the old-age pension system (based Single person, 67%, 2 children on a notional defined contribution) theoretical- Married couple, 100% / 67%, 2 children Married couple, 100% / 67%, no child ly ensures long-term stabilization of pension ex- penditure in relation to GDP (EC, 202111), but it Sources: World Bank analysis; OECD. Note: Net position = income tax + SSC – cash benefits will come at the cost of a dramatic drop in the 67, 100 and 167 percent stand for income level of the replacement rate (Figure 3.17). Women, for whom adult family members (as percent of average earning) in 2017 the retirement age was rolled back to 60, will be in a particularly difficult situation. The re- pensions rather than old-age pensions to maximize placement rate drop means that official forecasts of their lifetime pension income (Poland SCD 2017). pension expenditure could be underestimated. The In a dozen years’ time, this underestimation could actual expenditure could be greater due to 1) the be as high as 2 – 3 percent of GDP a year (Tyrowicz, budgetary subsidy necessary to cover the difference 202013). Additionally, old-age related expenditure between the actual and the minimum guaranteed will also increase due to social benefits for pension- pension benefit12; and 2) the risk that with declin- ers with very low pension benefits who will not ac- ing replacement rates workers may claim disability quire minimum pension rights. General government expenditure 79 Figure 3.17 Replacement rates in the EU in 2050 60 20 50 10 percent of the last salary Ratio — first pension as Percentage points 40 0 30 -10 20 -20 10 -30 0 -40 Latvia Slovak Republic Slovenia Sweden Czech Republic Romania Belgium Poland Germany EU27 Spain Ireland France Bulgaria Denmark Italy Finland Malta Croatia Estonia Lithuania Netherland Austria Cyprus Hungary Luxembourg Portugal Latvia Level (left) Change to 2019 (right) Sources: EC, Aging report 2021. Combined, the tax system and social transfers EXPENDITURE led to a visible reduction in inequality. The Polish fiscal system is characterized by progres- ON PUBLIC SERVICES sive and inequality-reducing transfers (including both pensions and non-pension transfers), close to Current expenditure on public services is neutral direct taxes, and regressive indirect tax- relatively low es. The inequality-reducing capacity of the fiscal system has increased in recent years. Compared Current spending on the provision of public ser- to other countries, Poland’s fiscal system has rel- vices is low compared with the OECD average, atively large inequality-reducing effects. (World across most key spending categories. In recent Bank, 202114). years, spending on public services has been below 20 percent of GDP. The largest gaps are in spend- In 2020, pandemic-related disruptions to the ing on health, which reached 4.6 percent of GDP in economy — including rising unemployment and 2019 — almost 2 percentage points below the OECD strong industry-specific performance dispari- average — even though total health care spending ties — increased household income differences, has increased sixfold since the 1990s. The latter re- with the Gini coefficient reaching its highest mains one of the lowest in the EU as a share of GDP or level since 2015. Given the uneven pace of recov- spending per capita. During the pandemic, structural ery in individual parts of the economy, this situa- weaknesses of public services became more apparent tion may persist this year. The situation of poorer partially related to underfunding, e.g., COVID testing households is also adversely affected by relatively per population has been one of the lowest in the EU. high inflation, subsiding the real effects of salary in- creases and leading to a decline in the real value of System-level policies promote equity in both ed- social transfers not subject to indexation. Since no ucation and healthcare but in practice, access to major changes are foreseen in social transfers, the services as well as satisfaction differ markedly planned amendments in personal income tax and (Figure 3.18 and 3.19). In the case of health services, social security contributions might drive changes satisfaction rates are below the OECD average and to income equality in the coming years. many people face restrictions in accessing services 80 Poland Public Finance Review (OECD, 2020a15). Poland scores well in education, on While access to high quality health services is an both satisfaction and attainments, as measured by important determinant of inclusion and human the position in the PISA ranking. capital building, Poland still faces challenges. Despite initiatives like the introduction of free access Figure 3.18 Satisfaction and confidence to a range of medicines for people aged 75+ and inte- across public services in Poland grated care programs for selected population groups, access to care still depends on the place of residence, 80 education, or wealth level (Figure 3.20). The health Health 60 Education care system is understaffed (Figure 3.21) and suffers 2 care system from inefficiencies. Despite a visible shortening of the median waiting times for some procedures, they remain relatively high (OECD, 2020b16). Pandemic- 20 induced delays in diagnosis and rapid population ag- ing will put even more pressure on the health care 0 system in the coming years. Although pressure for Figure 3.20 Self-perceived health, 2019 Poorest a Very good or good Local Judical 20% a Bad or very bad police system Richest Very good or good 20% Bad or very bad 0 20 40 60 80 2009 2019 Percent of 16+ population Source: OECD; Gallup World Poll. Source: Eurostat Note: * satisfaction with health care and education, confidence with judicial system and local police Figure 3.21 Practicing doctors, 2018 or latest Figure 3.19 Satisfaction and confidence across public services: Poland vs. OECD, 2019 Austria Norway 80 Lithuania Switzerland Germany Health 60 Education Sweden 2 care system Denmark Czech Republic 40 Spain Italy Iceland Australia 20 Netherlands OECD Average Estonia 0 Hungary New Zealand Latvia Israel Slovenia France Belgium UK Local Judical Canada police system US Japan Mexico Korea Poland Colombia Poland OECD Average 0 1 2 3 4 5 6 Source: OECD; based on Gallup World Poll. Per 1,000 population Note: * satisfaction with health care and education, confidence with judicial system and local police Source: OECD. General government expenditure 81 efficiency and quality is in place in individual hospi- in 2012 – 13 and has not recovered to its 2010 – 11 peak tals, it is still lacking at higher levels (voivodeships, (Figure 3.22). A similar drop in investments was re- NFZ, and ministry). Experience from other countries corded in 2016 – 17. With the public capital stock to GDP clearly shows that a more holistic (regional or na- ratio21 one of the lowest in the EU and infrastructure tional) approach is needed to achieve high efficien- gaps existing in many areas (IMF 202022), there is sig- cy and quality of care (Poland SCD 2017). nificant scope to boost public investment by further tapping EU funds, which could in turn crowd-in ad- The impact of recent education reform on educa- ditional private investments. tional attainments is yet to be assessed; however, there is room for improvement in developing com- Figure 3.22 General government petencies needed in a modern economy and soci- capital expenditure ety. Challenges to developing such competencies as the 7 ability to cooperate, creativity, etc. have been well-doc- 6 umented, including in government studies.17 Poland’s Percent of GDP 5 schools were ill-prepared for online learning, which 4 increases the risk of a divergence in learning outcomes, 3 to the detriment of rural areas and small towns. Even 2 before the COVID-19 crisis, learning outcomes in rural 1 areas were falling behind urban areas by the equiva- 0 lent of one year of schooling. Comparisons with oth- 2007 2007 2008 2008 2009 2009 2010 2010 2011 2011 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 er countries also show that there are still gaps in early (below 3 years) childhood education enrollment, and Investments Capital transfers the availability of preschool education facilities varies EU funds* across regions. Also, the percentage of vocational ed- Source: Eurostat. ucation and training (VET) students enrolled in com- Note: * funds disbursed in a given year under infrastructure programs (due to the method of settlement, they are only an bined school- and work-based programs is low (OECD, approximation of actual expenditure in a given year) 2019b18). Only two Polish universities have been rated as among the best 500 in the world (ARWU, 202019). New EU funds, including the Recovery and Resil- In a rapidly changing labor market — accelerated by ience Facility (RRF), provide an opportunity for the pandemic — the low participation of adults in life- higher public investment in the coming years. long learning is a challenge (Eurostat, 202120). Closing For 2021 – 22, the government assumes a temporary these gaps does not require an increase in spending increase in investment expenditure to 4.8 percent of as such, but a more precise targeting. GDP, and then a return to the 2019 level (4.3 percent of GDP)23. Projects co-financed from EU funds, includ- ing cohesion funds and the crisis-related Recovery and Resilience Facility, will help to maintain capital GROWTH-ENHANCING expenditure above 4 percent of GDP. In its National Recovery and Resilience Plan (NRRP), Poland applied EXPENDITURE for all available grants (€23.9 billion) and €12.1 billion of preferential loans24. Full utilization of the potential Growth-enhancing expenditures are that exists in public investments by crowding-in pri- relatively high and volatile vate sector investments requires a stable and predict- able regulatory and tax environment for companies. Public investment expenditure is relatively high On the other hand, a tight labor market and capaci- compared to other OECD countries, but the public ty constraints in selected sectors, such as construc- capital stock remains one of the lowest in the EU. tion-related manufacturing branches, could limit the Public investment dropped by a cumulative 25 percent effective absorption of the EU funds. 82 Poland Public Finance Review Local governments play an important role in Public investments appear to fulfill a dual eco- public investments. Local governments account- nomic and social development role. Many lagging ed for over 40 percent of investment expenditure regions have an above-average public capital stock to in the general government sector in recent years, GDP ratio. The impact of public investment on eco- with the notable exception of 2016 (Figure 3.23). nomic development in these areas may be negatively Depending on the year, these outlays accounted affected by constraints arising from geographical, en- for 1.4 to 3.1 percent of GDP, showing high volatili- vironmental and structural conditions, and/or chal- ty. Local governments play a particularly important lenges in preparation and selection of investment role in housing, recreation, and environment invest- projects. It may also reflect a preference for invest- ments. In terms of transport, education, and health, ments that improve the quality of life for the region’s they account for around half of the expenditure. citizens, rather than increase economic potential. Figure 3.23 Local government A strategic state reconciles growth with budget investment expenditure discipline; therefore, it is selective in investment 3.5 70 and focuses on the quality of public investment management. A strategic state targets public invest- 3.0 60 ment on sectors with high growth potential and pos- Percent of public investment 2.5 50 itive externalities such as transport infrastructure, Percent of GDP energy and broadband network, and R&D efforts. In 2.0 40 Poland, the largest amount of capital expenditure is 1.5 30 for transport infrastructure, but the lack of coordina- tion across all transport modes and the limited num- 1.0 20 ber of intermodal terminals resulted in the limited 0.5 10 complementarity of the different transport modes 0.0 0 and suboptimal use of existing infrastructure. There are also significant regional infrastructure dispari- 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 ties, e.g., limited connectivity between towns to re- Percent of GDP (left) gional centers in eastern and northwestern Poland Percent of total public investment (right) (Poland SCD 2017). A recent international study based Source: Eurostat. on Data Envelopment Analysis reveals the potential for improvement in public investment management Some reforms introduced in recent years — in- in Poland (IMF 202126, Figure 3.24). A comprehensive, cluding tax changes — affect local governments’ independent assessment would allow for the iden- own income and their capacity to invest, includ- tification of the main inefficiency sources. A long- ing in the greening agenda. Further changes in term life-cycle approach to managing and financ- PIT (e.g., increase in the tax-free amount, increase ing transport and ICT infrastructure would help to of the threshold from which 32 percent tax is paid) improve spending efficiency, prepare for the likely and CIT (new reliefs, increasing the scope of enti- phasing-out of EU financing, and reduce the over- ties that will be eligible to benefit from the so-called all infrastructure gap. Considering the role that local Estonian tax) planned for 2022, may intensify this governments play in public investment, the capaci- problem. According to initial estimates, local gov- ty of local actors to design and implement econom- ernments — which participate in the collected in- ic development initiatives in an inclusive manner come taxes — may lose Zl 10 – 11 billion as a result.25 ought to be strengthened. To compensate for the lost income, the government has announced a special investment subsidy, but Public investment in R&D and basic research is being earmarked, it will likely further reduce local low compared with other high-income countries. governments spending flexibility. Basic infrastructure in Poland continues to absorb General government expenditure 83 Figure 3.24 Efficiency gap in infrastructure Figure 3.25 Structure of growth-enhancing spending: EU countries government expenditure Romania 100 Italy Percent of total growth-enhancing expenditure Poland Bulgaria Greece 80 Slovak Republic Malta Croatia EU average Cyprus 60 Netherlands Slovenia Hungary Luxembourg 40 Spain Czech Republic Denmark Portugal 20 Latvia Belgium Austria France Germany 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Finland 0 10 20 30 40 50 60 Percent* Basic research R&D Source: IMF Fiscal Monitor – April 2021. Investment grants* GFCF* Note: * the difference between the country’s spending efficiency Source: Eurostat. and that of best performers Note: *excluding R&D and basic research expenses a significant share of public investment compared to crisis. Interest expense benefited also from the 2014 other high-income countries. Transport infrastruc- debt reduction, the one-off transfer and subsequent ture accounts for the largest share — approx. 40 per- write-down of government debt securities (worth cent — of the gross fixed capital formation. Poland approximately 9 percent of GDP) from open-ended continues to spend relatively less on R&D and basic pension funds to the Social Insurance Institution research and these expenditures represent a small (ZUS). In a low interest rate environment, even the share of growth-enhancing27 spending (Figure 3.25). significant increase in government debt due to the Although the final figures for 2020 are not yet avail- pandemic has not led to an increase in interest costs. able, there are many reasons to believe that the tar- The risk for the low-interest cost scenario could be gets for the increase in R&D spending envisaged the growing global inflationary pressure. in the Europe 2000 program and the Responsible Development Strategy have not been met28. Low financing costs and an important decline in the debt-to-GDP ratio have led to a decline in in- PERFORMANCE terest payments as a share of GDP. Lower inter- AND EFFICIENCY est payments have created additional fiscal space than would have been otherwise the case, with inter- Government performance and efficiency est payments declining to below 1.4 percent of GDP, have stagnated down from 2.7 percent of GDP in 2012. The cut in the National Bank of Poland reference rate to 1.5 percent Recent EU-wide studies29 place Poland in the mid- in March 2015 and a further reduction since February dle of rankings on public administration capaci- 2020, together with open market operations, has low- ty, measures of governance and service delivery ered financing costs for government debt during the and growth friendliness. Poland performs rela- 84 Poland Public Finance Review Table 3.1 Poland’s position among the EU member states by government capacity and performance indicator 1st quartile 2nd quartile 3rd quartile 4th quartile (best performers) (worst performers) • Strategic planning • Access to government • Transparency of • E-government users capacity information government • Barriers to public sector • SGIa implementation • Transparency • Voice and accountability innovation capacity International perception • Control of corruption • QOGb implementation • Societal consultation of corruption capacity • Gallup perception of • Use of evidence-based • Impartiality corruption • Trust in government instruments • Professionalism • Online services • Closedness • Online service • Pre-filled forms completion • Ease of doing business • Services to business • Inter-ministerial • Regulatory quality coordination • Rule of law Source: “A comparative overview of public administration characteristics and performance in EU-28” (EUPACK, 2018) Notes: a implementation capacity component of the Sustainable Governance Indicator (Bertelsmann Stiftung) b implementation capacity component of the Quality of Government institute Gothenburg expert survey tively well on issues related to the organization and the average for the EU, but still below the CEE av- processes of governance, while ranking lower on dig- erage (Figure 3.26). Both allocative and technical italization and services delivery, according to a com- efficiency of primary education were high in 2018, parative analysis of public administration character- because of continuous improvements over the ana- istics and performance (EUPACK, 201830, Table 3.1). lyzed period (2006 – 2018). The peak in efficiency in Overall performance of public expenditure in Poland secondary education was recorded in 2012 (Dinca did not improve significantly in the 2007 – 16 peri- 2021).35 It then declined in subsequent periods, al- od, in line with the EU-wide trend (Cepparulo and though it remained higher than in other EU coun- Mourre, 202031). There have been efficiency gains in tries.36 Technical efficiency of tertiary education the case of growth-friendly expenditure, such as pub- remained high throughout the period, while alloc- lic order and public services, as well as PISA-related ative efficiency has deteriorated over the last decade. education spending32. Meanwhile, efficiency losses were recorded in the case of capital expenditure, R&D Figure 3.26 Efficiency gap in health and and non-PISA related education expenses. A recent education spending: EU countries study conducted for OECD member states (Afonso Health et al., 202033) confirms that although the efficiency Education gap is narrowing over the long term, the process is 0 1 2 3 4 5 6 7 8 9 slow and the gap is still significant. Percenta Inefficiencies in spending on education are rel- Poland CEE Averageb EU Averagec atively small, greater in the case of health care. Min Max A recent assessment based on Data Envelopment Sources: IMF Fiscal Monitor – April 2021. Analysis (IMF, 202134) indicates inefficiencies in Notes: spending on education to be relatively small — small- a the difference between the country’s spending efficiency and that of best performers er than both the CEE and the EU average. In the case b average for the EU members from the CEE included in the analysis of healthcare, the efficiency gap is greater, above c average for the EU members included in the analysis General government expenditure 85 A performance assessment based on economic framework for spending reviews in 2015. Since then, transformation priorities indicates the need for several reviews have been carried out, covering, for faster changes in a post-pandemic world. The on- instance, expenditure on public administration and going global changes pose new challenges to govern- expenditure designed to support low-income fami- ments and require transformation not only of societies lies. They concluded with publicly available reports, and economies, but also of governments themselves. but there is no indication that recommendations In a recent study related to this transformation, the were used to inform the design of expenditure pol- World Economic Forum (WEF) tried to evaluate coun- icy (EC, 202038). Institutionalization and inclusion tries’ emerging priories as they embarked on a full in- of spending reviews into the budget process — en- tegration of social, environmental and institutional visioned in the 2021 Convergence Program — would targets into their economic systems (WEF, 202037). increase the likelihood of spending reviews to play Among 37 countries — mostly developed and emerg- their intended role. ing economies — Poland was ranked 33rd. It ranked worse on social and health care, human capital ac- cumulation, research and innovation, taxation, and facilitation of “markets of tomorrow” (Figure 3.27). CLIMATE-RELATED Figure 3.27 Performance in economic EXPENDITURE transformation priorities Spending needs to meet climate Public institutions commitments are significant Infrastructure Poland needs to step up its actions to meet more Taxation ambitious climate goals. While the EU’s medi- Human capital um-term emission reduction targets are becom- Labour market and ing even more ambitious, Poland faced challenges social protection Social and health in meeting its 2020 national targets. In 2019, the care share of renewable energy in gross final energy Long-term investments consumption increased by just 0.7 p.p. to 12.2 per- Competition and cent, well below the 15 percent aim. For primary en- anti-trust frameworks "Markets of ergy consumption, the assessment is hampered by tomorrow” the COVID crisis39. Research and innovation Diversity, equity and Despite the efforts made in the last 30 years, Poland inclusion still lags high-income European countries in terms 20 30 40 50 60 70 80 90 100 of the quality of environment. Poland is one of the Score 0–100a most greenhouse gas (GHG)-intensive economies in Poland CEE3b the EU, second only to Bulgaria. Poland is also one of Min Mean Max the most polluted countries in the EU. If pollution Source: “The Global Competitiveness Report – special edition were to improve to meet the WHO guidelines, resi- 2020,” WEF. dents in Warsaw would add 1.2 years to their life ex- Note: a The higher the score, the better performance pectancy (AQLI, 202040). Bad air quality translates to b CEE3 – average score for Czech Republic, Hungary and Slovakia premature death rates attributable to outdoor air pol- lution — PM2.5 — 55 percent higher compared to the EU Poland performs spending reviews, but their ef- average (OECD, 2020b41). Poor air quality causes nearly fective follow-up is assessed to be weak. Objectives a quarter of bronchitis cases among children, leading of spending reviews are to identify efficiency savings to over 200,000 cases every year. Pollution also dam- as well as to assess and compare the effectiveness of ages the economy, contributing to 8 percent of total various spending programs. Poland defined the legal lost workdays (WB, 201942). 86 Poland Public Finance Review General government spending on environmen- Environmental investments and reforms are tal protection declined in recent years and is an important part of the National Recovery and now about one-third lower than the EU aver- Resilience Plan (NRRP). Two of the five compo- age (Figure 3.5). In recent years, this expenditure nents of the NRRP are directly dedicated to the green has been at the level of 0.5 percent of GDP, below transformation. The objective of the Green Energy both the CEE (0.65 percent) and the EU-27 average and Energy Intensity Reduction component is to (0.8 percent). About one-third of this expenditure reduce the negative impact of the economy on the is investment expenditure. Over 60 percent of ex- environment while ensuring the country’s compet- penses were spent on waste management, while pol- itiveness and energy security. The Green, Intelligent lution abatement represented another 10 percent. Mobility component aims to develop a sustainable, Environmental protection investments by the gener- safe, and resilient transport system that adequate- al government were on average 0.15 percent of GDP ly serves the needs of the economy and society. A in the 2016 – 19 period. The scale of investments in total of €21.8 billion is allocated to these two com- recent years was significantly smaller than in pri- ponents, i.e., over 60 percent of the funds provided or periods (0.3 – 0.4 percent). Close to 90 percent of for in the NRRP. Given the nature of the expendi- general government investments on environmental ture, a significant part of the component 2 and 5 of protection were carried out at the subnational level, the NRRP funds will be allocated to institutions out- with budgetary units carrying out the remainder. side the general government sector, and within the general government to municipalities. Subsidies for the energy sector remain low and are skewed towards fossil fuels. Total energy sub- Poland is also a beneficiary of significant funds sidies (regardless of the environmental impact) ac- from the Just Transition Mechanism, a tool to sup- counted for 0.6 percent of GDP in 2018, below the port a fair transition towards climate-neutral EU average of 1.2 percent (EC, 202043). Of these, ap- economy. The Just Transition Mechanism is based proximately two-thirds are subsidies for fossil fu- on three pillars: Just Transition Fund, European els (coal mining industry restructuring, grants for Investment Bank loan facility and InvestEU Just thermal modernization projects or replacement of Transition scheme. Poland is set to become the big- high-emission heat sources with low-emission one) gest recipient of financing from the Just Transition and only one-third goes to renewable energy. Fund, amounting to €3.5 billion, which are expected to mobilize local funding to support national invest- Other environmental policy instruments include ment efforts towards a transition of country’s en- preferential and commercial loans for environ- ergy sector. Finally, as a part of the Cohesion Policy, mental protection and energy efficiency projects, 30 percent of the European Regional Development pro-ecological investments provided via public fi- Fund and 37 percent of the Cohesion Fund should nance institutions, and dedicated programs. The be allocated to activities related to climate (approx- National Fund for Environmental Protection and Water imately €17.7 billion).  Management (NFOŚiGW) and its regional counterparts (WFOŚiGW) are pivotal elements of the environmental Full implementation of the green agenda requires protection financing system in Poland, providing direct mobilization of different types of financial re- preferential financing (estimated at around 0.01 per- sources. According to the estimates contained in the cent of GDP in 2018) and pro-ecological loans in coop- Energy Policy of Poland until 2040 (PEP2040), the en- eration with the Bank for Environmental Protection ergy transformation of Poland, carried out in a social- (BOŚ, whose main shareholders is NFOŚiGW). Pro- ly acceptable manner, while guaranteeing energy se- ecological commercial investment and non-investment curity, maintaining competitiveness of the economy, loans granted by BOŚ in 2019 amounted to 0.1 percent and limiting environmental impact will require huge of GDP, financing projects related to development of investment outlays, approximately Zl 1,600 billion44; renewable energy sources, increasing energy efficien- Zl 867 – 890 billion in the fuel and energy sector, and cy in industry, circular economy, thermo-moderniza- Zl 745 billion in non-energy sectors (industry, house- tion, and improving air quality. holds, services, transport, and agriculture). While EU General government expenditure 87 funds will contribute significantly to transformation Figure 3.28 General government wage bill in the coming years (especially after the launch of the 11.0 NextGeneration EU package), there may be pressure 10.5 after 2025 for a significant increase in the commit- ment of national funds, including the budget. 10.0 Percent of GDP 9.5 While direct environmental spending is currently 9.0 low, Poland will need to use the full suite of fiscal policy instruments to reach its climate objectives. 8.5 The role of government in green transformation goes 8.0 beyond direct spending. It must incentivize changes 7.5 through tax incentives, pricing, regulation, and oth- 7.0 er tools at its disposal combined in a comprehensive 2011 2012 2013 2014 2015 2016 2017 2018 2019 environmental strategy. So far, some instruments support the energy transition (e.g., the feed-in tar- Poland CEE9 EU Germany iffs and feed-in premiums providing price support as Sources: IBS, Eurostat. well as grants to the development of renewable en- ergy capacities), but others support fossil fuels (e.g., The employment structure in the public sector tax expenditures supporting the utilization of fossil is skewed more towards the older age groups fuels in the energy sector, agricultural production, compared with the private sector (Figure 3.14). transport and by households). Seventeen percent of public administration employ- ees are within five years of retirement age, compared with 8.8 percent in the private sector. Less than 8 per- cent are younger than 30, compared with 18.5 percent PUBLIC EMPLOYEE in the private sector. Education and health care have a higher share of older workers at 18.5 and 19.7 per- COMPENSATION cent, respectively. Given the highly specialized skills EXPENDITURE in these sectors, an advanced age structure may re- sult in shortages of qualified personnel and could put pressure on some public service delivery (IBS, 2021). Public employee compensation as a share Poland already has one of the lowest numbers of prac- of GDP and employment relative to total ticing doctors per capita in the OECD. employment is similar to those of other OECD/EU countries Figure 3.29 Age structure in public administration and business The level of compensation as a share of GDP and 4 the share of public employment are both in line with that in peer countries. General government Percent of employment 3 employment accounts for 17 percent of total em- ployment in the economy45, while compensation ex- 2 penditure until 2020 was marginally above 10 per- cent of GDP, close to the EU average (Figure 3.28). 1 Employment in public administration is highly de- centralized, with close to 60 percent of all gener- al government employees working in the local gov- 0 16 20 24 28 32 36 40 44 48 52 56 60 64 68 ernment sub-sector. The education sector accounts Age for 30 percent of the general government wage bill, while the share of the public wage bill for health Public administration Business economy care and general public services are lower. Sources: IBS, Eurostat. 88 Poland Public Finance Review Figure 3.30 Average monthly wage Figure 3.31 Average monthly wage and salary: growth and salary: level 170 130 Percent of the national average 160 125 120 150 Index 2010 = 100 115 140 110 130 105 120 100 110 95 100 90 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Enterprise sector Enterprise sector Public administration and defence Public administration and defence Education Education Health care and social work activities Health care and social work activities Source: GUS data (excluding micro enterprises). Source: GUS data (excluding micro enterprises). Freezing the central government remunera- Over the last decade, the public tion fund was one of the expenditure contain- administration wage premium has ments measures in post-global financial crisis disappeared (GFC) fiscal consolidation. As a result, the pub- lic wage bill grew at a slower pace than GDP and Wage bill containment, used as a tool for consol- the wage bill to GDP ratio dropped by 1 percent- idating expenses, eliminated the wage premium age point from its 2010 peak to 10.1 percent in 2018. in the public administration46. While before the Increasing wage pressure in the economy prompt- global financial crisis, public sector employees en- ed the government to allow for faster growth of joyed a wage premium of 5 percent, in 2018 they faced wages in 2019 and 2020 (Figure 3.30 and 3.31), and a wage penalty of 3 percent compared to private sec- the level of compensation expenditure increased tor employees of the same age, education and occu- to 10.3 and 10.9 percent of GDP, respectively. In the pation (IBS, 2021).47 However, this overall indicator 2021 Convergence Program, the government proj- masks substantial heterogeneities. Older employees ects that the wage bill to GDP ratio will gradually in the public administration still benefit from a sig- decline to 10.5 percent of GDP in 2024. nificant wage premium (5 percent in 2018), while earnings of young workers are 12 percent lower than For some professions, salary levels appear inade- the wages of their peers in the private sector (Figure quate, including for selected categories of health 3.32). Furthermore, returns to education in the public care employees: nurses, midwives, and paramed- administration are lower than in the private sector; ics. The prevalence of self-employment among med- in 2018, tertiary educated public administration em- ical staff (physicians, but also other hospital staff ) ployees earned on average 10 percent less than similar obfuscates the true health care personnel expenses. private sector employees and public administration In education, the main challenge is the design of the workers with upper secondary education enjoyed a remuneration system, in which internships and ex- wage premium of 10 percent. An even larger premi- perience have little impact on earnings. This makes um — of 16 precent — was observed among employ- the teaching profession relatively unattractive, es- ees with elementary, lower secondary, or basic voca- pecially considering the rapid growth in wages in tional education (Figure 3.33). This situation is due to the rest of the economy. the design of the remuneration system, which pro- General government expenditure 89 Figure 3.32 Estimated public administration Figure 3.33 Estimated public administration wage premium/gap, by age group wage premium/gap, by education level 20 20 15 15 10 10 Percent 5 5 Percent 0 0 -5 -5 -10 -10 -15 2006 2008 2010 2012 2014 2016 2018 -15 2006 2008 2010 2012 2014 2016 2018 Employees with higher secondary education Age <30 Age 55+ Employees with tertiary education Sources: IBS; Structure of Earnings Survey. Sources: IBS; Structure of Earnings Survey. vides benefits such as one percentage point for each the central government — for which adequate statis- additional year of tenure starting from the fifth year, tics are available —indicate that the higher the level capped at 20 percent, and one-time bonuses for long of positions, the lower the share of women (OECD49). tenure48. These benefits contribute to a rising wage premium for this category of workers (IBS, 2021). The design of the remuneration system makes it difficult for the public administration to com- EXPENDITURE pete for talent with the private sector. While pub- DECENTRALIZATION lic administration remains an attractive workplace in case of lower-skill professions, it is unattractive Final general government expenditure is for those with high skills. Occupations in the pub- evenly split between three government levels. lic administration with the highest wage premia in- clude drivers, cleaners and helpers, and legal, social, The launch of significant family programs (Family and cultural professionals. For ICT professionals, life 500+, Good Start) distributed by municipalities science and health professionals, as well as manag- has led to an increase in the share of local gov- ers in the public administration, the wage gap rela- ernment spending in recent years. The final gov- tive to the private sector is negative. ernment expenditure is now evenly split between the three government levels: central, local, and so- In lagging regions, the public administration cial security funds. Since the split is not as even on remains an attractive employer. A particularly the revenue side, this is accompanied by large inter- large wage premium exists in Podkarpackie voivod- nal transfers, especially from the central budget to ship (14 percent). A public-private wage penalty is local authorities. With local governments controlling observed in three relatively high-developed regions, a third of all public expenditure, Poland has become including Mazowieckie, where the vast majority of one of the more decentralized countries in Europe. the central administration is located. Decentralization of decision-making over the use of funds, however, is much less advanced (EC, 201850). While women account for most of the employ- ment in the public sector, in managerial positions Experience in other countries shows that, in ad- they are underrepresented. In general, employ- dition to many benefits, decentralization of ex- ment in the public sector and public administration penditure can also lead to inefficiencies. Regular is attractive for women, who account for 60 percent evaluation of potential threats can lead to early mit- of the general government employment. But data for igation. Decentralization on its own is usually insuf- 90 Poland Public Finance Review ficient to improve the efficiency of public service de- The aging of the population and the project- livery. It must be accompanied by other conditions, ed shrinking labor supply are forcing a shift particularly revenue decentralization, which shows from “volumes to values.” Budgetary expen- positive and significant impacts on public service de- diture should contribute to this objective. This livery that are not observed with spending decentral- requires optimization of expenditure — switching ization alone. Spending decentralization may also lead from the wasteful and inefficient to those support- to an overlap in some government functions, poten- ing productivity growth. Since poverty and social tially creating waste. Expenditure assignments are exclusion have been significantly reduced in recent more often shaped by history and motivated by politi- years, Poland may re-focus toward growth-enhanc- cal and social dynamics than by efficiency (IDB, 201851). ing expenditure and underfunded public services. The former would ensure the sustainability of so- Consuming about one-third of total spending, so- cial policy also in the long term. The latter would ad- cial protection is the largest spending category dress systemic weaknesses that have become even for local governments, followed by education and more visible during the pandemic, prepare public economic affairs. Social protection represents the services for the aging process, and provide spill- largest category in overall local government expen- over effects to human capital. diture (29 percent of the total in 2020, i.e., 3.7 percent of GDP), followed by education expenditures (3.5 per- Further expansion of universal social programs cent of GDP) and expenditure in transport, commu- may crowd out growth-enhancing expenditure, nication, and other economic interventions (econom- leading to allocative inefficiencies from the per- ic affairs) accounting for 2.9 percent of GDP. Another spective of the intergenerational balance. Further major sub-national function is general public ser- improvements in poverty and inclusiveness should vices (administration), which accounted for nearly be achieved rather through well-targeted (specific 10 percent of total subnational spending. Spending groups, regions), means-tested social assistance. The on housing and community amenities accounted for use of pilot programs could also serve to improve around 3 percent of subnational expenditure. This the effectiveness of new transfer measures. Further function comprises various sub-sectors such as sup- progress in reducing poverty and social exclusion ply of potable water, public lighting, urban heating, can be also achieved through the tax and labor mar- housing (construction, renovation, and acquisition ket policies, provided that they encourage people to of land), and urban planning and facilities. On the choose employment forms, which provide security other end of the spectrum is spending on public or- against random events, and a decent pension in old der, safety and defense which accounted for only age through the (adequate) contributions to the so- 2 percent of subnational expenditure (0.2 percent cial security system. of GDP). This category includes mainly local and re- gional police services, fire protection, civil protec- Some expenditure on social benefits could be tion, and emergency services. reduced, without sacrificing equity objectives, through better targeting. Given the low effective- ness of some forms of social transfers in poverty reduction and social exclusion, a comprehensive OUTLOOK, review of some key policies would be desirable, especially those where multiple solutions — both MAIN CHALLENGES, on expenditure and tax side — overlap, leading to RECOMMENDATIONS a low overall effectiveness (e.g., pro-family policy). Due to wage pressure in the economy and the dis- Structural reforms will be needed to ensure appearance of the public administration wage pre- compliance with Stability and Growth Pact mium relative to business, in the short term the fiscal rules over the medium term and to wage bill does not appear to be a source of mean- meet socio-economic objectives ingful savings. Right now, the focus in this area General government expenditure 91 should be on existing shortcomings. Compared to portionately affected by the ongoing technological the business sector, employment in public admin- changes. This calls for intensified lifelong and on- istration is financially unattractive for young and the-job training. well-educated people. In turn, seniority is overpaid. The wage system in some public services (e.g., edu- Well-targeted health spending is also an import- cation) should allow for a greater performance-based ant element of investments in human capital. wage differentiation. Revision of the whole system Despite some progress, access to high-quality health and perhaps new systemic solutions are required in care still depends on individual circumstances such the health care (low remuneration in case of certain as socioeconomic status and location. Poland will categories of health care employment, reducing trans- need to improve access through greater and more parency widespread application of self-employment effective allocation of resources if it hopes to meet in the medical staff). In the medium-term, digitaliza- the needs of a growing elderly population. tion should become a source of some saving for the overall wage bill. To support a swift recovery and speed up the pace of structural changes (digitalization, automation, Investments in human capital should be a par- green transition, among others), a sustained level ticular priority. Identifying the optimal allo- of growth-enhancing expenditures would be re- cation of public resources to skills formation quired. Therefore, reducing this type of expenditure at different stages of the life cycle is crucial to should not be — as it used to be in the past — a vic- improving the quality of human capital, and tim of the fiscal consolidation process. Poland should be guided by the best available ev- idence on the returns to different interventions. In the years 2022 – 23, favorable conditions may New literature finds that if one accounts for all the arise for the start of the fiscal consolidation pro- components of human capital — quantity, quality, cess. Once the economic recovery is on a strong foot- and experience — the human capital explains much ing, the government should embark on a fiscal consol- more of the cross-country income differences than idation process. On the spending side, there could be previously estimated (IDB, 201852). Automation, ro- renewed emphasis on increasing the efficiency, tar- botization and digitalization — which are acceler- geting, and equity impact of public spending. ating as a consequence of the crisis — tend to in- crease the premiums to higher and better-quality Fiscal measures taken in the face of the pandem- education. Under such circumstances, the policy of ic were accompanied by a reduction in transpar- ensuring equal opportunities in access to building ency and accountability of fiscal accounts. This relevant skills and competencies becomes partic- is particularly true for the use of off-budget Special ularly important. Early-stage (pre-school) inter- Purpose Vehicles (SPVs), allowing the circumven- ventions are particularly efficient. They have high tion existing fiscal rules and ceilings and weaken- returns because they take full advantage of brain ing social control over public spending. While in sensitivity peaks and facilitate future learning, a crisis situations, the temporary use of SPVs may phenomenon called “dynamic complementarity” be justified, since amending a budget law to trig- (Cunha at el., 200653). In the first stages of life, the ger additional emergency spending may take some rates of return are much higher for interventions time. The authorities should strive to ensure full directed to disadvantaged children than to well-off transparency and accountability. The changes en- children. On-the-job training (OJT) is an important visaged in the NRRP — especially the inclusion of source of human capital, suggesting that policies special purpose funds in the stabilizing expendi- that influence OJT (for example, fighting informal- ture rule — are commendable. Poland could also ity and labor market dualism) can have a potential- consider establishing a fiscal council. If designed ly large impact on output per worker (IDB, 2018). in accordance with the principles for independent Given the high concentration of older workers in fiscal institutions, it would help to achieve greater easy-to-automate routine jobs, they may be dispro- transparency and accountability. 92 Poland Public Finance Review While gross public pension expenditure is pro- Figure 3.34 Projection of expenditure related jected to be broadly stable over the medium to public pensions term (Figure 3.34), there are risks of increased 2019 10.6 spending pressure. The rollback in the retirement 1- ∆ dependency ratio 3.7 age is expected to impact future pension bene- 1- ∆ benefit ratio 1.4 1- ∆ coverage ratio 1.5 fits adequacy, increasing the incidence of mini- 1- other (net) 0.4 mum pension and potentially the associated fis- 2030F 11.0 cal costs. While gross public pension expenditure ∆ dependency ratio 1.3 ∆ benefit ratio 2.0 is projected to be broadly stable over the medium ∆ coverage ratio 0.1 term, there are risks of increased spending pres- other (net) 0.1 sure. The demographic component is projected to 2040F 10.5 significantly impact gross public pension expen- 0 2 4 6 8 10 12 14 16 diture as the number of beneficiaries relative to Percent of GDP that of potential contributors rises (3.7 percent- Source: EC Aging report 2021. age points increase over the 2019 – 30 period). Old age dependency ratio is projected to increase by There are several sources of public spending pres- nearly 10 percentage points by 2030 to 38.9 per- sure over the medium term. In 2017, the government cent, as the post-war baby-boom generation con- pledged to increase the public share of health expendi- tinues to reach retirement age. This is estimated ture to 6 percent of GDP by 2024. This year, in its pro- to be largely offset by the impact of reform-relat- gram called Polish Deal, the governing coalition has ed factors54 that positively impact pension expen- committed to increase public financing of health to ditures (EC Joint paper on Pensions 2019). Sharp 6 percent of GDP by 2023 and 7 percent by 202756. An declines in replacement rates on account of the integrated healthcare strategy is critical to ensure the Notional Defined Contributions (NDC) system and highest efficiency of this additional spending. There its interactions with the expected rise in longevi- are also other sources of pressure on public spend- ty represent a source of risks, however. Minimum ing in the longer run, related to aging, long-term care pension guarantees and other social benefits for costs are set to rise by around 0.3 percentage points pensioners with pension benefits below the mini- of GDP by 2030 and another 0.4 percentage point by mum pension could result in higher than estimat- 2040 (EC 202157). The cumulative effect of demogra- ed old-age related expenditures. Furthermore, an phy-related changes is expected to increase govern- adverse macroeconomic structural shock could ment spending by about 2 percentage points of GDP increase the gross public pension expenditure by by 2030, and possibly by as much as 4 points when 0.4 percentage points of GDP by 2030.55 risk factors are taken into account (Table 3.2). Table 3.2 Age-related general government expenditure 2019 2030 2040 percent of GDP Public pensions 10.6 11.0 10.5 Health care 4.9 6.3 6.7 Long-term care 0.8 1.1 1.5 Education 3.8 3.8 3.5 Total 20.1 22.2 22.2 Total: risk scenario 20.1 23.1 24.0 Total: policy scenario linking retirement age to life expectancy 20.1 21.6 21.4 Source: Aging report 2021, EC. General government expenditure 93 The government could consider the following key • Maximize the return on spending in education policy recommendations to address the issues through optimal allocation of public resources to identified (see Table 3.3 for more details): skills formation at different stages of the life cy- cle. Focus on pre-school and adult education as • Targeted support to affected economic agents well as on-the-job training. and vulnerable groups, and support to the health care system should remain key priorities in the • Increase competitiveness of the public adminis- short term. tration remuneration system for young, well-ed- ucated people. Consider limiting automatic rais- • A comprehensive review of the pro-family sup- es in salary with tenure. port (benefits and tax reliefs) to optimize cost and free funds for growth-enhancing projects. • Use digitalization to optimize employment in ad- ministration and public services. • Increasing retirement age, gradual alignment of male and female statutory retirement ages, and • Seek fiscal coherence to support the green ultimately introduce longevity adjustment of re- transition. tirement age. • Further strengthen the public expenditure re- • In-depth review of the poverty/inclusion situa- view processes, especially through implement- tion of the single-person households. ing effective follow-up mechanisms. • An integrated healthcare strategy to address pan- • Conduct an independent public investment man- demic-related and structural issues and to effi- agement assessment to identify key inefficien- ciently use additional funds. cy sources. Table 3.3 Main findings and recommendations Main findings Recommendations Estimated net impact General • The pandemic had a significant impact • Further support could be provided in • Subsidies to GDP ratio is ex- on the economy and society. Fiscal the short-term, as there is fiscal space. pected to decline by 3.4 per- policy is a key tool for mitigating its ef- Targeted support to affected economic centage points of GDP in fects and support the recovery. agents and vulnerable groups and sup- 2021-22 as crisis measures • In the medium term, meeting formal port to the health care system remain are phased out. EU and national fiscal rules (once they key priorities in the short term. Better- are reactivated) and maintaining pub- targeted containment measures have lic finances sustainability will require a smaller budgetary implications. well-thought-out consolidation strategy • Growth-enhancing expenditures would that will not endanger the recovery. need to be protected during the fiscal • Spending pressure to support the consolidation. green transition will mount in the me- • Further strengthen the public ex- dium term. penditure review processes, especial- • Advances in poverty reduction and so- ly through implementing effective fol- cial inclusion coupled with positive ef- low-up mechanisms. fects from the recovery on the labor • Seek fiscal coherence to support the market could allow Poland to reori- green transition. ent spending toward growth-enhanc- ing expenditure and underfunded pub- lic services. • Lack of coherence of fiscal poli- cy instruments to support a green transition. 94 Poland Public Finance Review Main findings Recommendations Estimated net impact Performance and efficiency • The increasing universalization of so- • A comprehensive review of the • Potential savings: up to cial benefits gives rise to budgetary pro-family policy (benefits and tax re- 1.2-1.3 percentage points costs with limited additional benefits liefs) to optimize expenditure and free of GDP. Maintaining so- for poverty/inclusion. funds for growth-enhancing projects cial transfers at the level • The current level of expenditure (0.8 percent of GDP). of 2020 would translate in on benefits for families and chil- • Increase retirement age, gradual align- savings of 0.6 percentage dren ranks Poland among the OECD ment of male and female statutory re- points of GDP. leaders. tirement ages and ultimately introduce • Increase (current expendi- • Old-age expenditure is at similar levels longevity adjustment of retirement age ture): could be financed from as that of countries with much more (0.4-0.5 percent of GDP). higher health contributions. advanced aging processes. • In-depth review of the poverty/inclu- • Saving potential from the • Among single-person households, the sion situation of the single-person public wage bill is limited percentage of people at risk of pover- households. Use targeted rather than over the short-term. ty and exclusion is twice as high as the universal tools. • Public investment is expect- average. • Optimization of primary and second- ed to increase by 0.3-0.4 • Enrollment rates in early childhood ed- ary education expenses and shifting percentage points of GDP, ucation (below 3 years) remains low. the saved funds for life-long learning with additional increase in (with special focus on developing dig- R&D. • Participation in lifelong education and ital skills). training is low. • Further expansion of the childcare and • The competitiveness of both the mod- long-term care facilities. el (internship and experience have lit- tle impact on pay) and the level of wag- • An integrated healthcare strategy es in education are low. Given the high needed to address pandemic-relat- rate of wage growth in the economy ed and structural issues and to allo- this may lead to negative selection. cate additional public health spending. The strategy could give more promi- • Healthcare expenses are low, and the nence to primary care, prevention, and pandemic has revealed numerous sys- e-health services. temic weaknesses. • Reform could also address issues re- • The overall level of public wage to GDP lated to an aging population (e.g., ex- ratio is similar to that of other OECD cess capacity in the hospital sector is countries. accompanied by shortfalls in the provi- • Capital stock level is relatively low, sion of long-term care). and the EU funding could help support • Limiting automatic raises in salary with public investment. tenure should be also considered. • The 2020 targets for the level of gov- • Use the National Recovery and ernment spending on R&D/basic re- Resilience Plan to ensure consistency search are likely not met. of various development programs and • DEA analysis reveals relatively large in- to increase growth-enhancing expendi- efficiency in infrastructure spending. ture in the short-term. • An independent public investment management assessment would help identify key inefficiency sources. Transparency and accountability • Since its introduction, the expenditure • The establishment of an independ- rule was subject to attempts to sof- ent fiscal council could help to ten it. achieve greater transparency and • While in crisis situations, the tempo- accountability. rary use of special purpose vehicles • The inclusion of special purpose funds could be justified (amending a budget in the stabilizing expenditure rule. law to trigger additional emergen- cy spending may take several weeks). Care must be taken to prevent circum- venting public expenditure control pro- cedures or existing fiscal rules. General government expenditure 95 Notes 1. A fixed cost of around Zl 41 billion annually (1.7 percent 14. “Redistributive effects of taxes and transfers in Poland. of GDP in 2021 and then gradually falling to 1.4 percent Background paper for Public Finance Review,” R. Ba- of GDP in 2024). diani-Magnusson, K. Goraus-Tańska. The analysis was 2. Due to the increase of both the benefit and the number based on the CEQ model. of pensioners, the cost will go up from around Zl 13 bil- 15. “Citizen satisfaction with public services and institu- lion in 2021 to 14.1 billion in 2024 (it will remain more tions” in “Government at glance,” 2020. or less constant — at 0.5 — as a percentage of GDP). 16. “Waiting Times for Health Services,” OECD, 2020. 3. Based on ZUS estimates (as part of the so-called pen- 17. “Szkoła dla Innowatora: kształtowanie kompetencji sion system review). According to those estimates in proinnowacyjnych,” 2018. 2022-2025 the annual cost of restoration of a lower re- 18. “Education at a glance 2019: Poland,” 2019. tirement age equals to approx. Zl 18 billion, of which 19. 2020 edition of the Academic Ranking of World Uni- approximately 80 percent is due to increased expendi- versities. The University of Warsaw saw its position ture and 20 percent to lost income. improve, moving into the band of institutions ranked 4. According to Chłon-Dominczak, due to the rollback between 301st and 400th. Kraków’s Jagiellonian Univer- there are already 1 million people fewer on the labor sity, moved in the opposition direction, dropping into market and in the 2040s this loss will exceed 2 million the 401-500 band of rankings. people (“How the retirement age affects labor resourc- 20. Eurostat statistics on participation rate of adults in ed- es, labor force participation and pension system spend- ucation and training. ing” article in “The retirement age and the pension sys- 21. Based on IMF data: “Investment and Capital Stock Data- tem, the labor market and the economy,” 2021). set, 1960-2017.” 5. “Inequality, Poverty and Child Benefits: Evidence from 22. For the details see: “Infrastructure in Central, Eastern, a Natural Experiment”, LIS Working Paper Series No and Southeastern Europe - Benchmarking, Macroeco- 799, 2020. nomic Impact, and Policy Issues”, IMF, 2020. 6. Eurostat social protection statistics. 23. 2021 Convergence Program. 7. “Świadczenie wychowawcze po pięciu latach: 500 plus 24. The amount of RRF loans available to Poland is €34.2 ile?”, CenEA, 2021. It should be remembered that the billion. database of households surveyed by GUS - used by the 25. Based on the statements by representatives of the Min- authors for their calculations – does not fully reflect istry of Finance to the press (“Samorządy stracą na the structure of households in Poland; the top income Nowym Polskim Ładzie ok. 10-11 mld zł”, Gazeta Praw- deciles are underrepresented. na, 17.05.2021). 8. “Rodzina 500+ - ocena programu i propozycje zmian”, 26. “Fiscal Monitor – April 2021,” IMF. 2019. The estimates were for the original version of the 27. GFCF + basic research + R&D expenditure + investment program (when there was an income criterion for the grants. first child in the family). 28. In both documents it was assumed that total expendi- 9. „Rodzina 500+ - ocena programu i propozycje zmian”, tures would increase to 1.7 percent of GDP. According 2019. to the Responsible Development Strategy 0.9 p.p. was to 10. The top 5 percent with the highest pensions will not re- accrue to the government and higher education sector. ceive any 14th pension at all. 29. The indicators used in the study reflect the situation in 11. “Aging Report 2021,” European Commission 2015/2016. 12. Minimum pension is guaranteed for men and women 30. “A comparative overview of public administration char- with at least 25 and 20 contributory years respectively. acteristics and performance in EU-28,” EC. If the pension is below minimum level, then the pen- 31. “How & How Much? The Growth-Friendliness of Pub- sion is supplemented by the minimum pension guar- lic Spending Through the Lens”, EC. antee, which is financed from the state budget and gen- 32. Education expenditure on those levels of education eral revenue financing. A significant part of the group which are covered by PISA (OECD’s Program for Inter- of low-contribution payers, and thus future beneficia- national Student Assessment). ries of the minimum retirement pension subsidies, are 33. “Government Spending Efficiency, Measurement and those who run a business and generate high income. It Applications: a Cross-country Efficiency Dataset”, A. is difficult to consider such a situation as socially just. Afonso, J. Jalles, A. Venâncio, 2020. 13. Presentation from “The retirement age and the pension 34. “Fiscal Monitor – April 2021”. In education outputs are system, the labor market and the economy” mBank- test scores and net enrollment rates and input is public CASE seminar, 2020. education spending per student (the methodology was 96 Poland Public Finance Review described in Patrinos, H., and N. Angrist, 2018, “Global 52. “Better spendings for better lives - how Latin America Dataset on Education Quality”, World Bank Policy Re- and the Caribbean can do more with less,” IDB. search Working Paper 8592). In the health care output 53. Cunha, F., J. J. Heckman, L. Lochner, and D. V. Master- is life expectancy and input is total per capita health ov. 2006. “Interpreting the Evidence on Life Cycle Skill expenditure (methodology was described in Garcia-Es- Formation.” In E. Hanushek and F. Welch, eds., Hand- cribano, M., P. Juarros, and T. Mogues. Forthcoming book of the Economics of Education. Volume 1. “Patterns and Drivers of Health Spending Efficiency.”). 54. In the case of Poland, reduction in the benefit ratio, cov- 35. “Assessment of the European Union’s Educational Effi- erage ratio and labor market effects and are estimated ciency,” 2021. to reduce gross public pension expenditure-to-GDP. 36. These estimates do not capture the effects of the educa- The coverage ratio will contribute to a 1.5 percentage tion reform carried out in recent years. Since Septem- point of GDP decline over the 2019-2030 period, while ber 2017 primary school education was extended from the benefit ratio will contribute 1.4 percentage points 6 to 8 years, lower secondary schools - gimnazja - were over the same period. Minimum pension expenditure phased out, with the last classes completing this edu- is expected to rise by 2.3 percent of GDP by 2070. cation in school year 2018/2019. 55. The 2021 Aging Report defines the adverse structural 37. “The Global Competitiveness Report – special edition”, scenario as a stronger cyclical downturn in the lagged 2020. 11 priorities were taken into account in the eval- recovery scenario and that the growth potential will be uation. The assessment for each of them was based on lower over the next decade and potential output growth a set of various indicators. will thus be permanently lower than under the base- 38. “Budget System in Poland: Challenges & Ongoing Re- line scenario. forms”, Economic Brief 060, November 2020. 56. The main source of financing is to be increased reve- 39. Target for 2020 was 96.4 Mtoe. In 2019 the actual con- nues from the health contribution (harmonization of sumption was equal to 98.13 Mtoe. its amount and calculation method). 40. Air Quality Life Index – Europe Fact Sheet, 2020. 57. “The 2021 Aging Report” - European Commission, 2021. 41. “Health at a Glance: Europe 2020: State of Health in the EU Cycle”, OECD, 2020. 42. “Air Quality in Poland, what are the issues and what can be done?,” 2019. References 43. “Final Report Energy Subsidies,” 2020. A. Afonso, J. Jalles, A. Venâncio, “Government Spending 44. Annex 1 presents the amounts foreseen in PEP2040 fi- Efficiency, Measurement and Applications: a Cross-country nancial framework defined in the state budget multi- Efficiency Dataset”, 2020 year planning. M. Brzeziński, A. Chłoń-Domińczak, I.E. Kotowska, I. Magda, 45. In addition to employment in the general government M. Myck, M. Najsztub, J. Tyrowicz, “Rodzina 500+ — oce- sector, there is also employment in the public sector, na programu i propozycje zmian”, 2019. which additionally includes state-owned enterprises. Feng Y, Yu X, Chiu Y-H and Lin T-Y, “Energy efficiency and According to the ILO estimates, the public sector em- health efficiency of old and new EU member states”, 2020 ploys nearly 24 percent of all workers. A. Cepparulo and G. Mourre, “How & How Much? The Growth- 46. This section has been prepared based on inputs from the Institute of Structural Research. Friendliness of Public Spending Through the Lens”, 2020 47. “Public wage bill” – background note prepared for the A.Chłoń-Domińczak, “How the retirement age affects labor re- World Bank Poland Public Finance Review. sources, labor force participation and pension system spend- 48. In the form of seniority benefits, e.g., one percentage ing” article in “The retirement age and the pension system, point is added for each year of tenure starting from the labor market and the economy”, 2021 the fifth year (up to 20 percent). There are also extra Cunha, F., J. J. Heckman, L. Lochner, and D. V. Masterov awards for a long tenure (those who worked for 20, 25, “Interpreting the Evidence on Life Cycle Skill Formation.” 30, 35 and 40 years receive one-time bonus amounting in E. Hanushek and F. Welch, eds., Handbook of the to respectively 75, 100, 150, 200 and 300 percent of their Economics of Education. Volume 1, 2006 monthly wage.) M. S. Dinca, G. Dinca, M. L. Andronic, A. M. Pasztori, “Assess- 49. Government at a glance database. While in expert posi- ment of the European Union’s Educational Efficiency”, 2021 tions women account for 70 percent, in the case of se- EC, “A comparative overview of public administration charac- nior managerial positions for 50 percent. teristics and performance in EU-28”, 2018 50. “Public administration characteristics and performance EC, “Budget System in Poland: Challenges & Ongoing Reforms”, in EU-28: Poland”, S. Mazur, M. Możdżeń and M. Ora- Economic Brief 060, November 2020 mus, 2018. EC, “The 2021 Aging Report”, 2021 51. “Better spending for better lives - how Latin America EC, “Commission staff working document — Country Report and the Caribbean can do more with less,” 2018. Poland 2019”, 2019 General government expenditure 97 EC, “Commission staff working document — Country Report OECD, “Education at a glance 2019: Poland”, 2019 Poland 2020”, 2020 OECD, “Health at a Glance: Europe 2020 : State of Health in M. Halaskova, R. Halaskova, V. Prokop, “Evaluation of the EU Cycle”, 2020 Efficiency in Selected Areas of Public Services in European OECD, “Recommendation of the Council on Principles for Union Countries”, 2018 Independent Fiscal Institutions”, 2014 IBS, “Public wage bill”, 2021 (a document prepared for the OECD, “Waiting Times for Health Services”, 2020 World Bank) Poland’s Ministry of Finance, “Wieloletni Plan Finansowy IDB, “Better spendings for better lives — how Latin America Państwa na lata 2021 – 2024”, 2021 and the Caribbean can do more with less”, 2018 P. R. Paradowski, J. Wolszczak-Derlacz, E. Sierminska “In- IMF, “Investment and Capital Stock Dataset, 1960 – 2017”, 2020 equality, Poverty and Child Benefits: Evidence from a Natu- IMF, “Infrastructure in Central, Eastern, and Southeastern ral Experiment”, LIS Working Paper Series No 799, 2020 Europe — Benchmarking, Macroeconomic Impact, and WEF, “The Global Competitiveness Report — special edition”, Policy Issues”, 2020 2020 IMF, “2020 Article IV consultation — Poland. Staff report”, 2021 WiseEurope, “Navigating without a compass — Assessment of IMF, “Fiscal Monitor — April 2021”, 2021 draft Polish RRP released in February 2021.”, 2021 J. Jellema, N. Lustig, M. Trabelsi, Y. Zouhar, “Commitment to eq- J. Tyrowicz, “The retirement age and the pension system, the uity — public expenditure and inclusive growth survey”, 2021 labor market and the economy” presentation from the P. Lewandowski, I. Magda, J. Sawulski, “Will the Polish pen- mBank – CASE seminar, 2020 sion system go bankrupt?”, IBS Policy Paper, 2019 World Bank, “Fiscal Redistribution in the European Union” A. Król, M. Myck, M. Oczkowska, K. Trzciński, “Świadczenie  — background to “Growing United: Upgrading Europe’s wychowawcze po pięciu latach: 500 plus ile?”, CenEA, 2021 Convergence Machine”, 2018 M. Myck, K. Trzciński, “From Partial to Full Universality: World Bank, “Poland: Toward a Strategic, Effective, and The Family 500+ Program in Poland and its Labor Supply Accountable State. Systematic Country Diagnostic”, 2017 Implications”, CenEA, 2019 World Bank, “Redistributive effects of taxes and transfers in OECD, “Citizen satisfaction with public services and institu- Poland. Background paper for Public Finance Review”, R. tions” in “Government at glance”, 2020 Badiani-Magnusson, K. Goraus-Tańska, 2021 OECD, “Economic surveys: Poland 2020”, 2020 OECD, “State of Health in the EU Poland — Country Health Profile 2019”, 2019 98 Poland Public Finance Review ANNEX 3.1 PEP2040 FINANCIAL FRAMEWORK DEFINED IN THE STATE BUDGET IN MULTI-YEAR PLANNING (Zl million) 2016 2017 2018 – 2020 2021 – 2025 Development Expenditure of the State Budget according to the definition of the Development Expenditure Classification — DEC (consolidated, except for grants to local authorities) 47. Energy 17.96 16.86 52.24 87.06 48. Mineral deposit management 1068.91 3013.47 6123.57 10205.95 Total 1086.87 3030.34 6175.81 10293.02 Development Support Expenditure not covered by the DEC 47. Energy 51.45 56.91 162.55 270.91 48. Mineral deposit management 20.95 27.19 72.21 120.35 Total 72.40 84.11 234.76 391.26 Expenditure by Other Government and Self-government Units (where final expenditure is not shown, it is the amount of the grant from the State Budget to the unit) Low-Emission Transport Fund 0.00 0.00 857.30 4029.20 (from 10.2020 means under NFOŚiGW) Other general government and self-government units 13.59 152.92 249.77 416.28 Total 13.59 152.92 1107.07 4445.48 Cohesion Policy Expenditure and Co-Financing Energy 799.40 2474.56 12702.99 11262.44 R&D and entrepreneurship 217.66 145.56 1409.28 1249.47 Total 1017.06 2620.12 14112.27 12511.91 Expenditure under Other Foreign Instruments and Funds CEF (Connecting Europe Facility) 10.80 5.20 24.00 40.00 Norwegian Financial Mechanism, 137.48 207.46 517.41 862.35 EEA Financial Mechanism Total 148.28 212.66 541.41 902.35 TOTAL 2338.20 6100.15 22171.32 28544.02 Source: “Energy Policy of Poland until 2040”, Ministry of Climate and Environment Annex 3.1 99 CHAPTER 4 PUBLIC PROCUREMENT: SAVING AND GREENING INTRODUCTION Green public procurement (GPP) is a key compo- nent of Sustainable Public Procurement (SPP). It is defined as “a process whereby public authori- Public procurement is an integral part of effective ties seek to procure goods, services and works with public financial management and a high priority a reduced environmental impact throughout their for many countries. The last decade has seen a signif- life cycle when compared with goods, services and icant change in public procurement. Public procurers works with the same primary function that would no longer simply carry out an administrative, compli- otherwise be procured.”1 GPP constitutes an import- ance-based function; they now play an important role ant tool to promote the use of greener products and in spending taxpayer money for development strategy. services by the public authorities. GPP can also serve Public procurement is a critical tool for public service as a tool for broader development purposes: sustain- delivery, as it acquires the inputs — goods, construc- able development; helping governments achieve envi- tion works, and services — needed for delivering pub- ronmental policy goals through pollution reduction; lic services. Public procurement therefore materially improvements in resource efficiency; sustainable pro- contributes to the goals of government, be it through duction and consumption; reduced biodiversity loss; efficient spending that maximizes “the bang for the increased resilience; and addressing climate change. buck”; through timely availability of inputs required for public services; through quality inputs that can Sustainable public procurement and SME support, raise the level of satisfaction of users of public ser- along with value for money or cost savings, are vices; or by spending public money fairly and transpar- key public procurement goals in Poland. While sup- ently, which can increase public trust in government. porting SMEs through public procurement has long been a policy priority, supporting green or sustain- Public procurement constitutes a powerful tool able goals through public procurement has gained pol- for governments to achieve their economic, so- icy prominence in the last few years (Fazekas & Blum, cial development, and environmental goals and 2021; Rosell, 2021). There are potential synergies as its efficiency is crucial. Governments around the well as trade-offs between pursuing cost savings and world spend an estimated US$9.5 trillion on public strategic priorities (Adam et al, 2021). Understanding contracts every year — representing approximate- these trade-offs warrants further analysis. While the ly 12 percent of GDP in OECD countries and up to prevalence of SMEs in public procurement is high in 25 – 30 percent of GDP in developing countries. Poland Poland compared with other EU countries, the use of has spent between 11 and 14 percent of annual GDP green procurement criteria is among the lowest in the on procurement per year equivalent to Zl 24 billion EU, suggesting there is scope for policy to shift toward per year on average (Figure 4.1) (OECD, 2019). Faced a more sustainable public procurement (Rosell, 2021). with growing public scrutiny and limited budgets, governments are under pressure to spend in ways The increasing use of e-procurement systems that maximize outcomes for their citizens. around the world makes procurement data de- tails more readily available (OECD, 2016, 2017). The Figure 4.1 Poland General Government ready and real-time availability of government-wide, Procurement high granularity spending data (typically on the con- 15 tract or purchased item level) is increasingly available. 14 Such datasets can potentially provide the much-need- ed efficiency metrics as well as identify the policy-rel- Percent of GDP 13 12 evant factors that can bring about increased efficiency. 11 10 The comprehensive and novel framework used in 9 this procurement analysis helps identify poten- 8 tial savings, as well as suggest policy interven- 2010 2011 2012 2013 2014 2015 2016 2017 2018 tions. This framework is used to understand pur- Source: OECDStat: https://stats.oecd.org chasing decisions and prices and identify potential 102 Poland Public Finance Review savings strategies across the whole public procure- would require policy or procedural changes, such ment system (De Oliveira et al, 2019). The framework, as increasing the length of advertising tenders or which has been applied in a wide range of countries, wider use of open procedures. Independently pric- combines traditional methods of strategic sourcing ing each intervention also aids prioritization. This with economic theories of auctions and data science. analysis discusses the strengths and weaknesses The analysis uses very detailed and comprehensive of such an approach and proposes improvements. administrative data on purchased goods, works, and services over a 10-year period, encompassing mil- lions of records. The data allows for developing an innovative framework of two key elements: PUBLIC PROCUREMENT 1) a procurement market overview using informa- MARKET IN POLAND tive visualizations; and 2) regression modelling of relative prices across Public procurement is critical for public service the whole procurement market. delivery, as it acquires the inputs — goods, con- struction works and services — needed for deliv- The proposed framework uses relative prices to ering services to communities. Public procurement do a government-wide assessment and identify therefore materially contributes to the goals of gov- the most impactful policies. Relative prices are de- ernment, be it through efficient spending that max- fined as the ratio of awarded to estimated contract imizes “the bang for the buck”; through timely avail- value of goods, works and services. Empirically, the ability of inputs required for public services; through framework employs both simple and complex quan- quality inputs that can raise the level of satisfaction titative methods, making use of both descriptive and of users of public services; or by spending public explanatory models. money fairly and transparently, which can increase public trust in government. The Polish government The analysis aims to contribute to the policy de- spent 11.2 percent of GDP on public procurement in bate in Poland based on robust data-driven in- 2017, slightly lower than the OECD average (12 per- sights. The methodology offers a reliable and specific cent) (OECD, 2019), and this increased to an estimat- tool for policy makers to understand public procure- ed 12.6 percent of GDP in 2019. Of this, 8.9 percent of ment markets and to identify factors driving cost GDP was the value of contracts awarded under the savings, either directly under policy influence (e.g., Public Procurement Law (PZP), while the remain- length of advertising tenders) or that which can be der was issued based on exceptions from the PZP.2 indirectly influenced (e.g., number of bids submit- For example, in 2019, public procurement amount- ted). The methodology feeds into day-to-day policy ing to an estimated 1.5 percent of GDP was for con- making, leading to recommendations typically fea- tracts below the minimum €30,000 EUR threshold, sible within existing legal frameworks by tweaking with an additional 2.6 percent of GDP awarded for the parameters of policy implementation. The meth- contracts that were given exemptions for other rea- odology is fully transparent and largely automated, sons.. The total value spent in the period of 2010 and so real-time policy advice can be provided based on 2020 is estimated at Zl 2,646 billion (US$ 697 billion simulating the impacts of distinct policy scenarios. in 10 years) (OECD, 2019; World Bank, 2021). The data-driven price modelling identifies im- The legal framework for public procurement in portant savings that do not require fundamen- Poland was adopted following entry to the EU in tally reconfiguring the composition of purchas- 2004, and changes introduced in 2008 sought to es or changing the regulatory framework. These increase the efficiency of public procurement. potential savings are estimated at 4.9 percent per The Public Procurement Law also implements rel- year (equivalent to Zl 24 billion in 10 years) and can evant EU directives (Classic Directive 2014/24/EU, be achieved through improved public procurement Sectoral Directive 2014/25/EU, Defense and Security processes and decisions. These projected savings Directive 2009/81/EC and Remedies Directives) and Public Procurement: Saving and Greening 103 is also bound by other acts of European law. In April able procurement approaches and will provide guid- 2008, incentives were introduced: Using procedures ance to public administration and contracting author- other than open tenders, where justified; application ities at central level. The four main objectives are: of non-price criteria, where desirable; application of green procurement; application of social clauses; pro- 1) professionalization of public procurement; motion of innovations; and support of the SMEs for 2) development of MSME potential and access to electronic procurement, though the impact of these public procurement incentives on procurement practices is assessed to 3) support to sustainable and innovative procure- have been limited. To increase efficiency, addition- ment; and al adjustments were introduced to the legal frame- 4) enhancement of health aspects in public procure- work, including making mandatory the use of non- ment including during COVID-19 pandemic. price criteria (since 2013); applying of certain social clauses (e.g., obligation to hire workers on an employ- ment contract); examining abnormally low prices; partitioning the procurement; electronic communi- cation (the implementation of the 2018 Directives); GREEN PUBLIC and direct payments to subcontractors in the event PROCUREMENT of unlawful non-payment by the general contractor. While green public procurement (GPP) may have A new Public Procurement Law (approved in started out as an “alternative” procurement ap- September 2019) came into force on January 1, proach, it is now recognized as an essential ele- 2021, replacing the previous act that had been ment of modern procurement systems. Countries in force since 2004. The law implements thresh- first started integrating environmental consider- olds and other regulations of the latest EU Directive ations into public procurement over twenty years 2014/25/EU, among other changes. The legal frame- ago, mainly in Europe and East Asia. work establishes one of the most open and transpar- ent public procurement systems in Europe. The key Poland has developed the legal framework for body responsible for regulating and overseeing pub- GPP, in particular by transposing the relevant lic procurement as well as providing public procure- EU regulations. The authority in charge of GPP pol- ment data is the Polish Public Procurement Office, icy is the Public Procurement Office (PPO). Since “Urząd Zamówień Publicznych” (UZP). 2007 PPO developed four Action Plans on Sustainable Public Procurement (APoSPP). The initiatives set The new Public Procurement Law mandates out in APoSPP contribute to achieving the goals set the preparation of four-year State Purchasing out in the Responsible Development Plan adopted in Strategies (SPR). The first SPR, which will be imple- 2016, which sets new directions for the state’s activ- mented over the 2021 – 25 period, makes a priority of ities and new incentives to ensure the Polish econo- ensuring that public procurement is instrumental in my’s stable development in the short and long term. achieving key strategic policy objectives — including It emphasizes the important role of public procure- supporting sustainable and innovative procurement ment in supporting responsible development aimed —and improve access of micro enterprises and SMEs at increasing the competitiveness of the economy, to the public procurement market. The Ministry of taking into account social responsibility and envi- Economic Development, Labor and Technology is ronmental issues. the custodian of the SPR. Since 2016, the PPO has been monitoring the ap- The SPR currently under development will set key plication of social and environmental clauses in strategic policy objectives of the Government. The public procurement. To encourage procuring enti- strategy will address key public procurement bottle- ties to support green procurement approaches, the necks and systemic issues affecting achievement of PPO translated and made available on its website the value for money including innovative and sustain- European Commission environmental criteria for 104 Poland Public Finance Review several product and services categories, including A. The analytical framework consists of 2 key dissemination of GPP guidance and good practices. components that work best together but can also As part of the annual reports on awarded contracts, be deployed independently: contracting authorities are required to indicate the i. public procurement market overview number and value of contracts in which one of the ii. price modelling green clauses was applied. The statistics on the appli- cation of these clauses clearly differ from the goals B. The main outcome variable used in the anal- set in the subsequent APoSPP. The last APoSPP for ysis is value for money, which is defined as 2017 – 20 assumed the use of environmental claus- the quality of goods, works, or services ob- es in 25 percent of procurements. Meanwhile, data tained for a given procurement price. Value shows that the environmental clauses were used in for money can improve because a given qual- only 1 percent of tenders (in terms of number of con- ity is achieved at a lower price or because a tracts) and 3 percent of value. Furthermore, in 2017, higher quality is achieved at a given price, or a study on the state of sustainable procurement in both. Increasingly, value for money is not mea- Poland and the use of social clauses and green crite- sured merely in terms of acquisition price and ria was carried out on behalf of the PPO. Telephone acquired quality, but more broadly applying surveys and an examination of almost 500 random- life-cycle costing approaches (Saussier & Yukins, ly-selected tender documents indicate that perfor- 2018). As measuring quality is difficult, most mance in both areas was weak. It also indicated that notably due to lack of comparable data across for green procurements, there is a strong need for widely differing sectors, the relative price of a further enhanced promotion of GPP benefits, prior- contract is used as the key dependent variable itization of green products and services, proactive for the analysis. The choice of the dependent stakeholder and market engagement including pro- variable rests on a set of important assumptions fessionalization efforts. that are described in Annex 4.1. Relative prices capture the discounts companies offer in com- parison to the reference price of the tender or auction (Coviello & Mariniello, 2014). Relative STRATEGIC prices are defined as PROCUREMENT Relative price = actual contract value / estimated lot value ANALYSIS C. The explanatory variables are selected to cap- ture all major phases and actors of the process The conceptual framework while also incorporating structural factors. Importantly, it includes policy-relevant factors The conceptual framework is comprehensive and that can be influenced more or less directly by utilizes data on all regulated public tenders, cap- policy interventions (Table 4.1, Annex 4.1 for turing detailed information on tendering pro- a more detailed description) (Fazekas & Blum, cedures, stakeholders, and tender results (De 2021). In order to help target policy decisions, Oliveira et al, 2019). The empirical analysis follows explanatory factors are organized into two main economic theory, exploring the impacts of auction groups: those that can be directly influenced design, market structure, and company character- through policy (e.g., length of advertising a ten- istics (e.g., size), as well as public administration der) and those that can only indirectly be influ- research on the effects of administrative capacities. enced by policy makers (e.g. number of bidders). Nevertheless, the identification of causal effects is Nevertheless, in tenders, these factors often ap- challenging, and descriptive and explanatory quan- pear in combinations. This gives rise to addition- titative models are utilized. A detailed methodolo- al variables, such as the locations of buyers and gy description and the key variables used are pre- the suppliers, which is often an important fac- sented in Appendix A. tor determining prices. Public Procurement: Saving and Greening 105 Table 4.1 Summary of main groups of explanatory factors used in the analysis Group Definition Examples Type Market The technological and competi- • Number of potential suppliers Structural characteristics tive structure of the market • Technological complexity Tender The conditions for bidders to • Length of advertising bids Factors that can be specifications participate in a tender • Financial conditions of participation influenced directly through policy Buyer Level of administrative capacity • Average time taken for evaluat- Factors that can be characteristics of the buying office or agency. ing bids influenced directly • Buyer type-specific regulations through policy Bidder/supplier Company productivity and • SME bidder and winner Factors that can be characteristics capacity. • Company headquarters location influenced indirectly Bidding Intermediate and final results • Number of bids submitted Factors that can be outcomes of the tendering process • Annual winner market share influenced indirectly (deciles) Source: World Bank. Data The analysis uses a rich dataset. The dataset com- frequent (10.77 percent and 8.73 percent). More than bines observations from Tenders Daily Electronic half of the works were related to civil engineering (TED)3 and the national Public Procurement Bulletin4, work (53.81 percent), close to one-fifth was general where relative price information was non-missing, construction work (18.7 percent). In terms of total covering the period of 2010 – 20.5,6 It contains data value of spending, works represent 34 percent of to- on 2,836,378 unique public procurement contracts, tal expenditure, services represent 24 percent, and awarded in 1,350,037 unique tenders, amounting to goods 22 percent (Figure 4.3). The spending struc- Zl 1,248 billion of public procurement spending during ture of the 10 largest markets is depicted in Figure 4.4. the period analyzed. This represents 48.5 percent of the total public procurement spending in Poland over Figure 4.2 Distribution of contracts by the past 10 years (OECD, 2019; World Bank, 2021).7 procurement category, Poland 2010 – 2020 Supplies Descriptive statistics and linear regression methods Services are deployed to measure how different factors affect 12.9% Works relative prices. Procurement method, seasonality, length of submission period, framework agreement, number of bidders, number of different items bought, supplier-buyer location, decision period length and 45.0% other bidding outcomes affect relative price. Close to 25.9% 45 percent of the contracts were for goods, 25.92 per- cent for services, and 12.93 percent for works (Figure 4.2). Most purchased goods (32.5 percent) were medical equipment and pharmaceutical products (28.83 per- cent). Services showed more diversity; however, train- Source: TED; national public procurement bulletins. ing and education services related markets were more Note: for 16 percent of contracts item type information was missing 106 Poland Public Finance Review Figure 4.3 Distribution of average share by Figure 4.4 Spending structure, procurement category, Poland 2010 – 2020 Poland 2010 – 2020 Supplies Services 2% 1% 2% Works 21.7% 3% 34.0% 3% 2% 27% 4% 24.2% Source: TED; national public procurement bulletins. 5% Note: for 16 percent of contracts item type information was missing The number of contracts during the analyzed 10% period follows a declining tendency after 2014, reaching a low point in 2017 with 122,793 award- ed contracts. There was an increase afterwards, and Works (complete; partial) & civil engineering work the number of contracts stagnated in 2018 and 2019 Construction work Pharmaceutical products (Figure 4.5). Site preparation work Health services Medical equipment Recreational, cultural and sporting services Meanwhile, a different trend is observed in the Engineering services Fuels spending volume (Figure 4.6). Public procurement Banking and investment services spending was significantly higher in 2011 and 2014 Source: TED; national public procurement bulletins. (close to national parliamentary elections), with the value significantly lower in other years (32 per- siderably lower, albeit still substantial (Figure 4.1). cent – 65.7 percent). The volatility in public procure- It confirms, however, that spending dropped con- ment spending as reported in budget statistics is con- siderably in the 2016 – 18 period. Figure 4.5 Number of contracts by year, Figure 4.6 Spending volume over time, Poland Poland 2010-2020 450 240 Number of contracts (thousands) 400 200 Public spend (billion Zl) 350 300 160 250 120 200 150 80 100 40 50 0 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Sources: TED; national public procurement bulletins. Source: TED; national public procurement bulletins. Note: 2020 partial data. Note: 2020 partial data. Public Procurement: Saving and Greening 107 PURCHASING SYSTEM is important to achieve intended results of procure- REVIEW ment. The goal is to devise a strategy with the low- est transaction cost possible, so that the price of the purchased goods or services does not end up higher The supply positioning matrix offers actionable in- than the cost to the government to buy it. A second sights to design procurement strategies and policies goal for an optimal strategy for routine categories is for improving results (Figure 4.7). to free up specialized procurement resources so that they have time to work on more complex, critical pro- Routine categories (Quadrant A) of low value and curement. Procurement of routine items in quadrant low risk require simple procurement procedures, A can be very time-consuming and demanding with- with efficiency as the main driver. In Poland, 135 out the right strategy. Although only 0.76 percent of categories of a total of 431 belong to this classification, the products purchased are in this quadrant, it can yet the total value of transactions is only Zl 1.7 bil- result in some small-value bidding processes that di- lion in ten years, or 0.07 percent of the total. Given vert resources from more complex procurement tasks. the high volume of transactions with simple require- ments, one strategy could be to employ automated Categories of high volume and low risk (Quad- transactions following built-in rules in an electron- rant B) could benefit from centralized procure- ic system. These types of categories are typically ex- ment strategies, such as Framework Agreements, pendable inputs to public services and timely delivery to ensure cost-effectiveness for the government. Figure 4.7 Supply positioning 20 Zl 1,350,000,000 (0.05%) Pipline transport services Zl 2,100,000,000,000 (86.4%) 19 31 categories 174 categories 1,851 contracts (0.05%) 1.628M contracts (40.83%) 18 C — Specialized D — Critical Low value, high risk High value, high risk Military aircrafts, missiles and spacecrafts 17 Cargo handling and storage services Rig-positioning services 16 Onshore and offshore services Explosives Mining equipment 15 Miscellaneous gardening equipment Hire of earthmoving equipment with operator Log average contract value 14 Construction work Radio broadcast transmission services Test and evaluation of military electronic systems Engineering services 13 Public utiliteis Forestry services Meter reading services Non-drinking water 12 Domestic services Furniture 11 Parts of foresty machinery Ink Medical equipments Training services 10 Pipeline-inspection Community healt services Higher education services Education and training services services 9 Non-drinking water Primary education services 8 Sewage, refuse, cleaning and environmental services Time-sharing services 7 6 Zl 1,740,000,000 (0.07%) Zl 324,000,000,000 (13.33%) 135 categories 91 categories 30,378 contracts (0.76%) 2.327M contracts (58.36%) 5 A — Routine B — Volume Low value, low risk High value, low risk 4 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Log total spending by market Number of contracts: 1 200,000 400,000 661,808 Average Source: World Bank; TED; national public procurement bulletins. 108 Poland Public Finance Review Given that the government is an important buyer on Critical categories (Quadrant D) that are charac- these markets because of the value it purchases as terized by high value and high complexity would well as the high number of suppliers, the government require the allocation of skilled resources. Given will benefit from a highly competitive approach to the inherent risks, due to size and complexity, it is procurement and by actively sourcing categories to recommended that the majority of buyers’ time fo- ensure buyers and end-users get the best deal. This cus on procuring and managing implementation of is the area in which the government should leverage these contracts. The highly technical nature of these its purchasing power to try and extract better value categories — e.g., construction work and health ser- and improve average quality. 91 market categories, vices — require dedicated review of price and suppli- 58.36 percent of contracts, and 13.33 percent of total er performance. In Poland, 2010 – 20, this quadrant procurement spending fall into this group. A central comprises 174 categories (40.37 percent of all cate- strategy for the procurement of items in Quadrant B, gories), adding up to 86.4 percent of the total award- even if implementation is decentralized, can help the ed value. This is where the procurement officers and government get more value for money overall. Frame- managers should focus their time and attention. work agreements added up to only a small fraction (0.62 percent) of all procurement during 2010 – 20 and Relative prices vary over time. They are relatively sta- the analysis shows that this represents only a very ble over the 2010 – 16 period, while after 2016, contract small percentage of all categories that could poten- values usually exceeded estimated bid prices on aver- tially be utilizing this type of procurement method. age by 4 – 8 percentage points (Figure 4.8). Average rel- Framework agreements were mostly used for medi- ative price is mostly constant across the country, rang- cal equipment. Consequently, there is ample room to ing between 0.9 – 0.94, only the central region operates utilize this tool for the markets where it is a good fit to with some overruns (1.04). Północno-Zachodni region improve results and maximize the bang for the buck. has the lowest average relative price (0.86) (Figure 4.9). The main guiding principle for specialized (Quad- Figure 4.8 Average relative price rant C) and low-value, high-risk categories that 1.10 Average relative price are volume-conscious is to improve ease of trans- 1.05 actions. Strategies to maximize efficiency for these categories include 1.00 0.95 1) making upfront effort on the design and specifi- 0.90 cations for the procurement activity toward long- 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 term contracts; and 2) ensuring stocks are available for a determined Sources: World Bank; TED; public procurement bulletins. period. Note: 2020 is an incomplete year. These categories comprise the most difficult pro- Figure 4.9 Average relative price, by region curements because the Government is not an at- 1.05 tractive buyer, but the products are specialized. Average relative price Thus, if the Government does not ease participation 1.00 of bidders, the best suppliers — such as manufactur- 0.95 ers — will simply not bid. The government may then end up purchasing from middlemen. A customized 0.90 strategy for these products that considers facilitat- ing participation, minimizing red tape, and paying 0.85 FS PL PL1 PL11 PL12 PL2 PL21 PL22 PL31 PL32 PL33 PL34 PL4 PL41 PL42 PL43 PL51 PL52 PL61 PL62 PL63 quickly will go a long way in contributing to sourc- ing from the better suppliers and maximizing results. Buyer region 0.05 percent of contracts and of total spending are Sources: World Bank; TED; public procurement bulletins. classified as specialized items or services. Note: 2020 is an incomplete year. Public Procurement: Saving and Greening 109 Figure 4.10 Distribution of average relative Figure 4.11 Distribution of used tender price by supplier selection method by markets 5 14 Frequency in 10,000s 12 4 10 3 8 Density 6 2 4 1 2 0 0 0 200 400 600 800 1,000 0 1 2 3 Market ID w_relp Source: World Bank. Lowest price MEAT Source: World Bank. Only 0.5 percent of suppliers have relative prices that are higher than 1.5 on average (Figure 4.10). Figure 4.12 Market entry rate by CPV division A consistently higher average contract value than 51 the estimated price suggests that the supplier may 75 73 enjoy a dominant market position, which it can use 16 to extract higher prices. If it occurs on a regular ba- 92 80 sis for particular buyers, they might need to consid- 31 er engaging further suppliers to break market dom- 42 76 inance and harness competition. 43 55 35 Lowest price criteria are usually used for vol- 63 71 ume products, while MEAT (Most Economically 19 Advantageous Tender) is more beneficial for spe- 44 85 cific, higher value products (Figure 4.11). The for- 14 98 mer selection method was used for 60.96 percent, 37 the latter for 39.04 percent of the contracts. This ra- 79 CPV division 32 tio appears also in most markets (e.g., telecommu- 3 nication services,) or even a more balanced, almost 18 48 equal distribution can be observed for some others 41 70 (e.g., medical equipment). In the categories of phar- 77 maceutical products, and civil engineering work, the 72 22 lowest price criteria was favored (65.54 – 69 percent), 50 while for training services, MEAT was more com- 90 60 monly used (61.5 percent). 45 39 15 The market entry rate varies by markets (Figure 34 38 4.12). The market entry rate is defined as the average 24 share of new companies (firms that have not partic- 66 9 ipated in procurement processes prior to the award 65 64 year). It is more difficult for new companies to enter a 30 large market, like civil engineering works, construc- 33 tion works, and pharmaceutical products (where av- 0 0.2 0.4 0.6 0.8 erage annual spending ranged between Zl 10 – 31 bil- Avg. market entry rate lion). The average entry rate for new companies was Sources: World Bank; TED; public procurement bulletins. 39 percent. This rate gradually increases for medi- Note: represents average share of new companies on a certain market. 110 Poland Public Finance Review um-sized markets with less specialized products (e.g., lent on the market of works for complete or part con- legal services, financial and insurance services, ed- struction and civil engineering work, site preparation ucation and training services). New companies in work, road transport services, engineering services, smaller markets, with a lower volume of products repair and maintenance services, computer equip- and services (community health services, school ment and supplies, pharmaceutical products, and catering, quilted textile), have the highest chance medical equipment (Figure 4.14). Over half of such (85.1 percent). In the future, adding purchased item contracts (25 percent and 29 percent) were awarded quantity to the obtainable variables could give more by public bodies and regional authorities. detailed insights on distribution of procurement packages, turnaround times, their share in the to- Figure 4.14 Percent of contracts bidding tal spending and number of contracts. above estimated price by market Following the ABC supplier analysis method- 1.7% 4.0% ology, 99.5 percent of suppliers had lower than 80 percent of market share — 0.83 percent on av- erage — accounting for 99 percent of total spend- 4.2% ing (Figure 4.13). These markets cover the most wide- ly-purchased goods, works and services, such as 1.5% construction works, pharmaceutical products, medi- 16.0% cal equipment, food suppliers, training, and business 1.3% and cleaning services. Of the total number of sup- 1.2% pliers, 0.02 percent had a more dominant position, concentrating 85.9 percent of market share. These included post and courier services, drilling services, pipeline services. Of the total number of suppliers, 1.6 percent of suppliers are monopolies, accounting on average for 99.6 percent of market share in mar- 11.9% kets such as non-drinking water, radio broadcast transmission services, and space transport services. Construction work Figure 4.13 Spending volume by supplier Computer equipment and supplies market share category Site preparation work Road transport services Repair and maintenance services 1,500,000 Pharmaceutical products Medical equipment Total awarded value (million Zl) 1,240,000 5,500 5,810 Engineering services Source: World Bank; TED; public procurement bulletins. 1,000,000 500,000 0 STRATEGY PACKAGE A (0–80%) B (80–95%) C (95–100%) Supplier category by market share Different strategies to achieve the savings poten- Source: World Bank; TED; public procurement bulletins. tial in public procurement are assessed using a relative price model. An explanatory model for rel- Bidding above the estimated price occurred in ative prices of goods, works, and services based on 43 percent of contracts, showing similar patterns the combined contract-level dataset collected from in each of the above-mentioned supplier category UZP and TED is used to identify and quantify savings groups by market shares. It was especially preva- potential throughout the whole public procurement Public Procurement: Saving and Greening 111 market while also appreciating the different strate- able factors and assuming only readily-achievable gies suggested by the above supply matrix. Outliers policy changes. Some of the considered changes do in relative prices were removed from the analysis: not require altering the existing regulatory frame- Relative price was bounded between 0.3 to 3 (Figure work, and instead rely on small tweaks to procure- 4.15). The sample also contained gross contract values, ment process design and implementation decisions, reflecting reduced and standard VAT (spikes at 1.08 e.g., month of spending, procedure type, or adver- and 1.23 in Figure 4.15), which is a large-scale error tisement period. Such savings potential is consid- in the dataset that cannot be precisely addressed. A erably below predictions obtained using unit prices, VAT error variable is included in the regression to signifying that using relative prices as the outcome control for the error when it is likely to occur, at rel- of the analysis is likely providing the lower bound ative prices of 1.08 and 1.23. of impacts from the proposed policy changes. Figure 4.15 Relative price distribution, Figure 4.16 Total savings associated with trimmed each intervention, 2010 – 20 4 140 Supplier specialisation 3 Bids received from smes Density 120 Supplier-buyer location 2 Market concentration 1 100 Buyer spending concentration Bidder number 0 Organisational quality: 0 1 2 3 80 avg. failed tenders % Billion Zl Filtered relative price Organisational quality: avg. decision making period Source: World Bank. 60 length (days) Note: Outliers (values below 0.3 and above 3 were removed; further Product bundling investigation of such data errors is recommended. 40 Month of spending Electronic auction Framework agreement 20 Advertisement period length POTENTIAL SAVINGS 0 Procedure type SCENARIOS Source: World Bank. The prices under alternative policy scenarios are The largest price impact can be achieved by im- estimated using advanced statistical modelling proving bidder participation: a Zl 67.9 billion de- techniques. This allows one to estimate how much crease in costs across the government is predicted the government would have paid for a particular if single bidding is cut by about half by increasing product had it used a more cost-effective procure- the bidder number to three bidders (Figure 4.16). ment strategy. The estimation is based on the price Importantly, the high prevalence of single bidding is modelling regressions that capture the associations not due to procedure type choice, as most such ten- between various policy relevant factors and relative ders followed an open procedure: 0.5 percent of con- prices. Suggested improvements to procurement pol- tracts with a single bid were direct contracts, 5 per- icy are presented in Table 4.2, and the estimated im- cent were contracts with single source method, while pact is presented in Figure 4.16.8 76 percent were awarded through an open procure- ment procedure. In a related policy intervention, de- Potential savings are estimated at about 5 per- creasing buyer spending concentration is similarly cent per year, or a cumulative Zl 24 billion for impactful: Zl 13 billion savings is predicted if half of the 10-year period analyzed using the underlying contracts are moved from high-spending concen- model (Figure 4.16). The total predicted savings are tration buyers (the top 3 highest deciles of buyers) achieved through only the most readily-influence- to lower-spending concentration buyers (7th decile). 112 Poland Public Finance Review Table 4.2 Summary of savings interventions procedure type move 70 percent of contracts in any non-open procedures to fully competitive procedures framework agreement reducing the use of framework agreements by 50 percent electronic auction Move 50 percent of paper-based contracts to electronic auctions advertisement period length increase advertisement periods: eliminate short advertisements Can be (1 – 39 days) by increasing them to more than 40 days directly influenced month of spending smooth spending between Sept and Dec by reallocating 50 percent of through contracts to a nearby cheaper month (Jan) policy product bundling moving 50 percent of more heterogeneous product bundling to the 2nd category (2 products) organizational quality: avg. deci- moving 70 percent of tenders in the medium and longest quintiles to sion making period length (days) the 2nd quintile organizational quality: avg. failed move all contracts from lower half success rate organizations (less tenders % than 99 percent) to highest success rate organizations (100 percent) bidder number moving 50 percent of 0 – 2 bidder contracts to 3 bidder contracts buyer spending concentration move 50 percent of contracts from high spending concentration buy- ers (the top 3 highest deciles of buyers) to lower-spending concentra- tion buyers (7th decile) Can be indirectly market concentration move 50 percent of contracts in high concentration markets (the top 3 influenced highest decile market) to lower concentration markets (4th highest decile) through policy supplier-buyer location increase market share of local suppliers (i.e., same state) by 50 percent share of bids received from SMEs increase share of SME participants by 50 percent supplier specialization moving 50 percent of contracts supplied by highly specialized suppli- ers (1&2 lowest deciles) to less specialized suppliers (3rd decile) Expanding the use of competitive procedure types is Agencies with a higher success rate of procurement predicted to have sizable savings potential too: Zl 12.3 processes saw lower relative prices, according to the billion savings could be achieved if 70 percent of regressions analysis (Figure 4.17). The success rate is contracts in non-open procedures are moved to ful- defined as the annual average of successful tenders, ly competitive procedures. Furthermore, improving which are procurement processes that resulted in a tender advertisement — eliminating short advertise- contract that was signed, as opposed to processes that ments (1 – 39 days) by increasing them to more than failed because no bidder turned out, wherein all bids 40 days — could contribute to a further Zl 8 billion were rejected or other reasons for failure. A contract price reduction. was awarded for 99 percent of all items or services put out for procurement. The government could save an estimated Zl 2.5 billion if agencies that underper- Factors that can be influenced form in this area were to increase their success rate. directly through policy This indicates that suppliers can offer discounts to more reliable buyers. This is also consistent with a Changes in policy parameters can generate im- broader explanation suggesting that higher success portant savings through changes in the prices rate agencies have higher capacity along dimensions contracted. Buyers with higher rates of success were that cannot be observed in the dataset; thus, success able to procure at lower contract values in 2010 – 20. rate can also be a proxy for administrative capacity. Public Procurement: Saving and Greening 113 Figure 4.17 Impact of buyer’s effectiveness In particular, “restricted” and “negotiated with on relative prices publication” are the methods that are likely to 0.995 render prices lower if bid evaluation and con- Predicted relative price tract award is faster. Even when controlling re- 0.990 sults for the number of bids received, which has a 0.985 material impact on the time to evaluate bids, this procurement method takes significantly longer 0.980 than all others. In addition, when bundling mul- 0.975 tiple items into one, single procurement process, 1 2 3 4 5 concluding bid evaluation and awarding individu- Avg.organisational success rate categories ally for each item might speed up the process over- Source: World Bank. all. The data indicates that on multiple-item pro- curement, one or more items may be held up until Procuring entities that are more efficient in the the evaluation of all items is completed. This could process of evaluating bids and awarding contracts be because of external clearances and approvals pay lower prices, on average, than those that take that are required for contract award and the back more time to complete these tasks (Figure 4.18). The and forth that would result if the award were to Government could save 0.25 percent (Zl 6.67 billion in be split up. But this strategy might be worth ex- ten years) if the process for bid evaluation and con- ploring on specific cases, as the impact on prices tract award is faster and less cumbersome. This is in is significant. line with estimates on a sample of other European countries, which find that procuring entities that are Assisting buyers that underperform on the task quicker in bid evaluation tend to attract more bidders of bid evaluation is another possible interven- (Fazekas, 2019). To achieve this, a key recommenda- tion that could yield savings. Results are non-lin- tion is to adopt electronic procedures to complete this ear across buyers even when comparing similar processing stage, as it can significantly cut times, par- complexity items procured using the same meth- ticularly when inter-agency approvals and clearances od. A target intervention that focuses only on the are needed. Another key recommendation is to allo- underperforming buyers will increase the effective- cate resources more strategically and beef up bid eval- ness of an assistance or training program. uation teams for complex, high-value procurement, or when a large number of bids is expected. Perhaps Unsurprisingly, the procurement method has an multi-agency task forces or committees for the most impact on prices: Any non-open procedures re- critical procurement can help to turn around bid eval- sulted in higher relative prices during 2010 – 20 uation faster and with higher-quality technical work. (Figure 4.19). The Polish Government could save an estimated 0.47 percent (Zl 12.27 billion per year) by Figure 4.18 Efficiency in bid evaluation and increasing the value procured through open proce- award had an impact on relative prices dures. Overall, contract complexity and complete- 0.990 ness are the key characteristics defining whether rule-bound decision-making in open auctions or Predicted relative price 0.988 less regulated direct contracting and negotiated pro- 0.986 cedures produce better results (Bajari, McMillan, & 0.984 Tadelis, 2009). Thus, the impacts crucially depend 0.982 on matching the right rules to the right tenders 0.980 and contracts (Parrado, Dahlström, & Lapuente, 0.978 2018). During 2010 – 20, already 76 percent of con- 1 2 3 4 5 tracts were awarded by open procedures, which Avg. decision making period length of buyer amounted to 63 percent of the total value procured (quintiles) in the period, therefore suggesting some room for Source: World Bank. improving efficiency. 114 Poland Public Finance Review Figure 4.19 Effects of procurement methods Poland has a balanced schedule of procurement on relative prices spending throughout the year. Poland procures a large share of contract value during the first quarter Negotiated without publication of the year (23.1 percent), a quarter that is typically characterized by lower procurement shares in other Negotiated countries; and even more so when framework agree- ments are a small share of public procurement. Most Negotiated with of the value is typically procured in the third quar- publication ter, with 30.2 percent of the total. Experiences from Open around the world show that the last quarter of the fis- cal year may face a substantial change in the profile of Restricted products purchased, and an increase in procurement of higher value items, such as vehicles and computers, Single source as governments rush to spend the available budget. In Least cost Poland, the government spends it mostly on civil en- gineering works, construction works, site prepara- Other tion work (summing up to 15 percent in that quarter), medical equipment (16.18 percent), and pharmaceuti- 0.95 1.00 1.05 1.10 1.15 cal products (11.96 percent). A relatively flat spending Predicted relative price volume in the last quarter of the year, coupled with Source: World Bank. markedly higher prices, suggest that bringing pro- curement forward in the year can generate savings. End-of-the-budget-year pressures to finalize pro- curement tenders raise prices. Public procure- Figure 4.21 Seasonality and relative prices ment during the September – December period is A. Predictive margins with 95% CIs typically more expensive Poland (Figure 4.20). The 0.995 Government of Poland could save an estimated 0.21 Predicted relative price percent (on average Zl 5.4 billion annually) with bet- 0.990 ter procurement planning that would avoid bunch- 0.985 ing up procurement at the last quarter of the year. 0.980 0.975 Figure 4.20 Annual spending structure 0.970 by procurement method 0.965 100 1 2 3 4 5 6 7 8 9 10 11 12 Month of purchase 80 Percent of spending B. Monthly total awarded value in billion Zl, 2010-20 60 200 Total spending (billion Zl) 40 160 20 120 80 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 40 0 Negotiated without publication 1 2 3 4 5 6 7 8 9 10 11 12 Negotiated with publication Negotiated Open Restricted Month of purchase Source: World Bank. Source: World Bank. Public Procurement: Saving and Greening 115 Factors that can be influenced The time it takes to pay an invoice is the number indirectly through policy one reason cited by bidders for not selling to gov- ernments. For that reason, shortening this process Changes in indirect policy parameters — such together with disclosing the data to show the time- as using electronic procurement methods — can liness of payments is recommended, as perceptions also affect prices. Procurement processes that had of payment delays can be hard to dispel and might a higher bidder turnout resulted in lower prices, and keep potentially good suppliers away from public the government of Poland could save an estimated procurement. 2.6 percent (Zl 67.9 billion over 2010 – 2020) if it re- ceived a higher number of bids per tender (Figure Encouraging SME bidders to participate in pub- 4.19). Competition for public contracts has the high- lic procurement can lower prices, according to est effect on prices among the procurement policies the analysis of TED data (Figure 4.22). Procurement and strategies that can be influenced indirectly. The notices available in the national electronic procure- average number of bids received is relatively low at ment system, the Public Procurement Bulletin, will 2.3. The use of electronic procurement can be a pow- be included in a subsequent version of the analysis erful tool, in addition to the benefits for bidder par- and might change the results. ticipation and internal processing. A study assessing the impact of e-procurement in Chile, for example, Figure 4.22 Impact of bidder turnout on suggests that the increase in the number of bidders relative prices and the corresponding decrease in bid prices is the A. Predictive margins with 95% CIs key driver in cost savings for the Chilean central 1.05 procurement agency (ChileCompra), amounting to 2.65 percent of total spending (Singer, Konstantinidis, Predicted relative price 1.00 Roubik, & Beffermann, 2009). 0.95 A mix of strategies, including training and out- 0.90 reach to suppliers, can help improve bidder turn- out. These include training programs, minimizing 0.85 the red tape for participation, shortening times 0.80 for payments, and customized market approach- 0–1 2 3 4 5 >6 es based on spend value and complexity. An out- Bidder number category reach effort to potential suppliers along with a train- ing offer on how to prepare responsive bids could B. Percent of volume of processes by number of bidders, contribute to new entrants to the public procure- 2010 – 20 ment market. Confidence on how to prepare a bid 100 18.9 that will not be rejected by the government can go a long way in motivating a supplier to try to sell to 80 5.7 the government. 7.9 9.8 60 11.7 Percent Furthermore, reducing documentation require- 42.1 ments for participation can help motivate poten- 40 tial suppliers to bid for public contracts. This is a frequent complaint from companies selling to gov- 20 ernments around the world as it raises the costs for participation and the risks for success. Electronic 0 procurement could help by eliminating requests 0–1 2 3 4 5 >6 for documents issued by the government, as the re- Number of biders quests could be automatically verified by integrat- Source: World Bank. ing government information systems. Note: in 4 percent of the contracts, bidder number was not indicated 116 Poland Public Finance Review Spending concentration at the buyer level result- ten years) if the largest monopolies and oligopolies ed in higher prices for the government. Even small were broken up and the supplier base was more di- differences on the level of concentration of award to versified (Figure 4.24). one or few bidders results in lower prices (Figure 4.23), and thus the Government of Poland could save an es- Figure 4.24 Effects of buyer-level spending timated 0.5 percent if buyers could further diversify concentration on relative prices their supplier base and break up monopolies and oli- A. Predictive margins with 95% CIs gopolies (Zl 13.2 billion annually, on average). 1.02 Predicted relative price Figure 4.23 Effects of share of SME bidders 1.00 on relative prices 0.98 A. Predictive margins with 95% CIs 0.96 1.03 Predicted relative price 1.02 0.94 1 2 3 4 5 6 7 8 9 10 1.01 Decile of buyer spending concentration 1.00 0.99 B. Awarded value and percent of market share of main supplier by buyer, 2010 – 20 0.98 0 0.8–50% 51–95% 100% 1.0 Max share spent on one supplier Share of SME bidders 0.8 B. Awarded value in billion Zl, by SME bidder share category, 2010 – 20 By PA-year: 0.6 SME bidder share category 96–100% 0.4 51–95% 0.2 0.8–50% 0 0% 0 5.0e+07 1.0e+08 1.5e+08 2.0e+08 Contract value, 2019 constant prices 0 5 10 15 20 25 Sources: World Bank; TED; public procurement bulletins. Spending structure (billion Zl) Note: On average, maximum share spent on one supplier was Source: World Bank. 27 percent The main supplier of a buyer was awarded on av- About 28.3 percent of the total value went to sup- erage 27.4 percent of the total value over a five- pliers with the highest market concentration. The year period. Whilst this does not apply to all govern- default approach of doing an open bidding will not be ment agencies, there are a few that might confront enough. When it comes to markets with higher levels monopolistic or oligopolistic powers. Consequently, of concentration of awards to one or a few suppliers, targeted interventions at a few buyers could bring sometimes a very large number of procurement pro- quick results in terms of prices. cesses for a certain product are awarded to the same supplier. A deeper review of such markets that in- Concentration at the level of markets also had a cludes scrutinizing the specifications, bid require- substantial impact on prices. Markets that were ments, and existence of patents will be required to im- dominated by one or a few suppliers showed higher prove results. And the potential of savings, even for a contract values during 2010 – 2020. Savings poten- small number of markets, shows that there is enough tial is estimated at 0.02 percent (Zl 0.44 billion over return on investment to justify the additional effort. Public Procurement: Saving and Greening 117 Local suppliers, or those located in the same Figure 4.26 Local suppliers and relative Department of the buyer, offered more econom- prices in public procurement ical prices on public procurement than outside A. Predictive margins with 95% CIs suppliers (Figure 4.25). Increased participation of 0.980 local suppliers in public procurement could gener- Predicted relative price 0.979 ate potential savings of 0.012 percent (Zl 0.32 billion 0.978 per year on average). Local suppliers were awarded 0.977 16.75 percent of the total value over the 2010 – 20 pe- 0.976 riod. Development and implementation of an out- 0.975 reach program to local suppliers, along with training 0.974 offers on how to identify and to prepare responsive 0 1 bids for public contracts, could raise the competi- Supplier-buyer location tiveness of local suppliers and, in turn, lower prices. B. Awarded value in billion Zl, by buyer-supplier location, Figure 4.25 Impact of market concentration 2010 – 20 on relative prices same regions A. Predictive margins with 95% CIs 0.990 different regions 0 100 200 300 400 500 0.985 Predicted relative price Total spending (billion Zl) Sources: World Bank; TED; public procurement bulletins. 0.980 0.975 0.970 GREENING PUBLIC PROCUREMENT 0.965 1 2 3 4 5 6 7 8 9 10 STRATEGY Decile of market concentration To construct an effective green procurement strat- B. Awarded value and percent of market share of main egy, the government could consider incorporat- supplier, 2010 – 20 ing good practices along the six pillars included 1.0 in World Bank’s Green Public Procurement Pillars and Good Practices Framework. This is built around share of supplier in total annual market six pillars that allow for a customized, non-linear yet 0.8 balanced approach: By market-year-supplier: 0.6 1) building the business case; 2) enabling framework; 0.4 3) market engagement; 4) professionalization; 5) implementation tools; and 0.2 6) monitoring tools. 0 For an effective GPP strategy, the government 0 1e+09 2e+09 3e+09 4e+09 5e+09 could consider a combination of good practic- Contract value, 2019 constant prices es that address Poland’s main design and imple- Sources: World Bank; TED; public procurement bulletins. mentation challenges. To overcome potential re- 118 Poland Public Finance Review sistance and set up realistic implementation plans, estimation of environmental benefits can support the government will need to build the business case the drive to ensure buy-in for GPP. Analyzing and for GPP. To this end, the government could consider effectively communicating the results and benefits launching GPP communication campaigns; identify of GPP are also powerful tools. and support GPP champions; identify entry points for implementation; and mobilize financial resourc- es. An enabling framework helps transforming the GPP from an ad hoc activity to a national priority. This presupposes a sound institutional framework; POLICY CHOICES FOR a well-identified policy space; a clear legal basis; pri- SAVINGS IN PUBLIC oritization and the development of an implementa- tion plan. Designing realistic tenders that attract a PROCUREMENT market response while driving green innovation mar- ket engagement is critical. The government could as- The strategic procurement analysis unveils op- sess market readiness; provide notices to the market portunities to save an estimated 4.9 percent of with sufficient lead time; engage in market consul- annual public procurement spending.9 These sav- tations to design realistic tenders; build supplier ca- ings could be achieved without introducing new laws pacity; promote SME participation; and conduct in- or modifying existing ones, via a combination of pol- novation procurements. icies and strategies on how to implement procure- ment and a phased implementation. Each of these Building professional competencies for GPP is an- policies could be implemented separately or in com- other critical element of a successful GPP strat- bination with other policies. egy. To achieve this, the government would need to assess the gaps in these competencies; assess train- Competition can help drive down relative prices ing needs; ensure buy-in from procurers for green paid in public procurement, both in the form of procurement; and develop manuals and toolkits for the procedure used and bidder turnout. Strategies GPP. Organizing trainings in GPP, setting up pilots for that could be designed and implemented systemical- green tenders, and fostering peer learning and net- ly include the use of open bidding instead of nego- working are effective ways to ensure professional- tiated procedures, and interventions to boost bid- ization. Driving organizational change is also critical. der turnout, both of which could benefit all buyers in the procurement system in Poland. To facilitate GPP and support its adoption at scale, it is necessary to provide implementation tools. The existence of monopolies and oligopolies at Meanwhile, monitoring tools are critical for show- the market and buyer level, the success rate of ing the benefits of GPP and identifying areas for procurement processes, and the efficiency of bid improvement. The GPP strategy would need to be evaluation and contract award all impact rela- supported at an institutional level and the needs as- tive prices. Addressing the issues of monopolies sessments would need to be coordinated. Framework and oligopolies at the buyer level as well as improv- agreements, catalogues, the development of envi- ing efficiency of bid evaluation could be very effec- ronmental criteria for priority categories, and en- tive in achieving savings when targeting underper- dorsement of eco-labelling schemes could also fa- forming buyers. These interventions represent some cilitate GPP. Life-cycle costing tools will help take of the low-hanging fruits. informed decisions. Joint procurements, leveraging e-procurement, and increasing the use of on-line Lowering end-of-the-year spending when pres- platforms can also foster adoption of GPP. For mon- sures to close the books at the end of the finan- itoring purposes, a monitoring plan, a baseline as- cial year are high represents an effective saving sessment, and identifying data collections methods strategy. About one quarter of annual procurement are critical. Selecting indicators and defining tar- contract value is awarded in the last three months of gets can help drive GPP, while tools that allow the the year when prices are considerably higher than Public Procurement: Saving and Greening 119 in earlier months. Shifting some of this expenditure The most impactful strategies and their corre- earlier in the year or allowing contracting process- sponding savings potential are summarized in es to span over financial years would unlock savings Tables 4.3 and 4.4. These strategies are grouped for the government. by elements that can be directly or indirectly influ- enced by policy.10 These savings predictions are de- Buying more diverse products (at least two dif- rived from the regression coefficients (i.e., the price ferent kinds) in the same tender could also be an sensitivity of each policy-relevant factor) and the effective strategy to save money. This is especially magnitude of the proposed change.11 true for high-volume, low-complexity items, which currently see their demand spread across many differ- Collecting data on itemized unit prices, units of ent buyers. In the sample, 91 percent of the contracts measurement, and quantities could help a more in- represented homogenous contracts in terms of prod- depth analysis of savings potential. An enhanced uct bundling, purchasing only one type of product. e-procurement system could consider collecting and analysing this type of data for informing policy. There Finally, an outreach program to government sup- are still some shortcomings related to the Polish pub- pliers could also help improve results. Private sector lic procurement data, even if it is among one of the companies as well as many high-performing public sec- best systems in Europe (Fazekas, 2017)12. Of central tor procurement systems work closely with suppliers importance for strategic sourcing analysis is the avail- to bring results to buyers. Therefore, easing participa- ability of unit prices, which is not readily available tion requirements based on a risk-management mod- in the case of Poland. el and training bidders in preparing responsive bids could also be an effective strategy in achieving savings. Table 4.3 Policies and strategies to save money on public procurement—Directly impacted by government policies and strategies Total savings Variable with impact in Poland Recommended policy or strategy on Contract value (billion 2019 Zl) 1) Encourage the use of fully competitive procedures by streamlining its process for evaluat- ing bids and awarding contracts; Procurement method 12.27 2) Carry out a survey with suppliers to gather insights into hidden costs to bid for govern- ment contracts. 1) Reduce the use of framework agreements where they do not result in sufficient savings Framework compared to one-off purchases; 8.06 agreement 2) Improve scope and design of framework agreements where they can bring about savings.a Time allowed for bid Grant more reasonable times for bidders to be able to prepare more responsive bids. 4.41 preparation Electronic auction 7.25 Expand the use of electronic procurement through UZP platform, adding buyers to the system Seasonality 5.4 Minimize procurement between the period of September and December. Number of items Consolidate items supplied by companies of similar markets, reducing the number of procure- 0.15 bundled on a process ment processes. Time spent on bid 1) Implement electronic processes for necessary paperwork to award contracts; evaluation and award 6.67 2) Strategically appoint resources for procedures with many bidders; and (days) 3) Simplify and standardize bid evaluation tasks. 1) Target intervention at relevant buyers to simplify bid requirements; Level of success of 2.51 2) Create training programs to bid evaluation committees of lowest performing quartile buyers; process 3) Develop standard bidding documents. a The use of framework agreements could lead to potential savings on certain markets, such as pharmaceutical products, while it can increase relative prices on other markets such as health services and post and courier services. 120 Poland Public Finance Review Table 4.4 Policies and strategies to save money on public procurement — Indirectly impacted by government policies and strategies Total savings Variable with impact in Poland Recommended policy or strategy on Contract value (billion 2019 Zl) 1) Offer training program to suppliers on how to bid for government contracts; 2) Minimize documentation required to bid for low complexity items; 3) Shorten time spent to pay invoices, Number of bidders 67.9 4) Create procurement strategies based on the supply positioning matrix (later discussed at this report); 5) Carry out a survey with suppliers to gather insights into hurdles to bid for public contracts. Monopolies and Deliver a training and advisory program for buyers with high levels of awards to one or few oligopolies at 13.19 suppliers. buyer-level Monopolies Develop targeted procurement strategies for products that show monopolistic or oligopolistic and oligopolies 0.44 characteristics in public procurement. at market level Supplier and buyer at 1) Offer automated and online alerts to sales opportunities for local companies; 0.32 the same location 2) Offer training programs to local companies in preparation of responsive bids. Encourage larger companies to bid by tailoring specifications for larger and more capable Share of SME bidders 0.65 companies. Supplier Procure related but different products in bundle making larger, more diversified companies 3.3 specialisation more likely to participate. PROPOSED FUTURE present synergies or trade-offs with prices. WORK Analyzing green and sustainable procurement requirements and scoring criteria on announce- ments could be done with the help of text-mining Relative price analysis can help identify a range of algorithms (Rosell, 2021). Such techniques could actionable policies to achieve savings with some make it possible to estimate the impact of green degree of confidence. Unit prices tend to be a more and sustainable procurement on market compe- reliable measure of value for money than relative tition and prices, for example. As SME participa- prices (e.g. De Oliveira et al, 2019; Kohler et al, 2015). tion is a strategic goal on its own, further analysis Hence, a targeted survey of unit prices in selected could look not only at the impact of SME partic- markets could help enhance the analysis. In addition, ipation on prices, but also help identify policies in the absence of reliable unit price information, one that could increase SME participation. Some SME- could also consider using alternative analytical tech- oriented policies could lower competition, there- niques and intervention types such as demand man- by increasing prices. Others, meanwhile, manage agement projects or value engineering (SAVE, 2020). to simultaneously increase prices and SME par- ticipation (Fazekas-Blum, 2021). Empirical analy- Public procurement in Poland is also expected sis could help identify policies that can simultane- to serve a range of strategic goals that could ously achieve multiple goals. Public Procurement: Saving and Greening 121 Notes 1. The definition adopted by the Task Force on Sustainable Bandiera, O., Prat, A., & Valletti, T. (2009). Active and Passive Public Procurement led by Switzerland (membership in- Waste in Government Spending: Evidence from a Policy cludes Switzerland, USA, UK, Norway, Philippines, Ar- Experiment. American Economic Review, 99(4), 1278 – 1308. gentina, Ghana, Mexico, China, Czech Republic, State Coviello, D., Guglielmo, A., & Spagnolo, G. (2015). The Effect of Sao Paolo (Brazil), UNEP, IISD, International Labor of Discretion on Procurement Performance (EIEF Working Organization (ILO), European Commission (DG-Envi- Paper No. 15/10). Rome. ronment) and International Council for Local Environ- Coviello, D., & Mariniello, M. (2014). Publicity require- mental Initiatives (ICLEI) and adopted in the context of ments in public procurement: Evidence from a regres- the Marrakech Process on Sustainable Production and sion discontinuity design. Journal of Public Econom- consumption led by UNEP and UN DESA. ics, 109, 76 – 100. 2. The report of the President of the Public Procurement De Oliveira, Alexandre; Fabregas, Abdoulaye; & Fazekas, Office 2019. Mihaly. (2019): Strategic Sourcing 2.0: Improving Fis- 3. Available at https://ted.europa.eu/TED/main/ cal Efficiency Using Big Data. Conference paper: ”Pub- HomePage.do lic Procurement: Global Revolution IX” at the Univer- 4. Available at https://www.uzp.gov.pl/ sity of Nottingham, June 2019. 5. 2020 is an incomplete year Fazekas, M. (2017). Assessing the quality of government at 6. The dataset includes information on contract value, the regional level using public procurement data (Work- buyer and supplier identification and names, contract ing Papers No. WP 12/2017). European Commission, DG award date, tender dates (e.g. bid submission deadline), REGO, Brussels. procurement method, product classification and pro- Fazekas, M., & Blum, J. R. (2021). Improving public procurement curement categories. outcomes: review of tools and the state of evidence base. Policy 7. This represents 45.8 percent of all data Research Working Papers, World Bank, Washington, DC. 8. Please note that the total savings values are extrapolated Fazekas, M., & Kocsis, G. (2017). Uncovering High-Level from the regression analysis — which was conducted on a Corruption: Cross-National Corruption Proxies Using subset of the data — to the whole public procurement mar- Government Contracting Data. British Journal of Polit- ket. The total contract value captured in the regression mod- ical Science. els was calculated using micro-data and established its ratio Fazekas, M., & Tóth, B. (2016). Assessing the potential for de- to the total public procurement spending according to the tecting collusion in Swedish public procurement. Swedish OECD (2.12 percent on average for the 2010-2020 period). Competition Authority, Stockholm. 9. These savings would amount to Zl 24 billion over a ten- Fazekas, M., & Tóth, B. (2018). The extent and cost of cor- year period. ruption in transport infrastructure. New evidence from 10. It should be noted that the sum of the individual strat- Europe. Transportation Research Part A: Policy and Prac- egies will not add up to the exact total savings percent- tice, 113. https://doi.org/10.1016/j.tra.2018.03.021 age potential because of interactive effects among them, Fazekas, M., Tóth, I. J., & King, L. P. (2016). An Objective such as using more competitive procurement procedures Corruption Risk Index Using Public Procurement Data. will contribute to increasing the number of bidders. European Journal of Criminal Policy and Research, 22(3), 11. The model used to identify and quantify savings poten- 369 – 397. https://doi.org/10.1007/s10610-016-9308-z tial on public procurement explains 15.4 percent of the Fazekas, M. (2019). Single bidding and non-competitive ten- variance in relative prices, a robust result for this type dering procedures in EU Co-funded Projects. European of dataset. Commission, DG Regio, Brussels. 12. For detailed diagnostics see: https://opentender.eu/pl/ Kohler, J.C., Mitsakakis, N., Saadat, F. et al. (2015). Does Phar- dashboards/transparency. maceutical Pricing Transparency Matter? Examining Brazil’s Public Procurement System. Globalization and Health 11, 34, 2014. https://doi.org/10.1186/s12992-015- References 0118-8OECD (2019): Government at a Glance 2019, OECD Adam, I., Fazekas, M. and Zellmann, C. (2021). Open and Publishing, Paris, https://doi.org/10.1787/8ccf5c38-en. sustainable procurement. Towards deepened collabo- Parrado, S., Dahlström, C., & Lapuente, V. (2018). Mayors ration between reformers. GTI-WP/2021:01, Budapest: and Corruption in Spain: Same Rules, Different Out- Government Transparency Institute. comes. South European Society and Politics, 23(3), 303 – 322. Bajari, P., McMillan, R., & Tadelis, S. (2009). Auctions Ver- Rosell, J. (2021). Getting the green light on green public sus Negotiations in Procurement: An Empirical Anal- procurement: Macro and meso determinants. Journal ysis. Journal of Law, Economics, and Organization, 25(2), of Cleaner Production, Vol. 279, https://doi.org/10.1016/ 372 – 399. http://doi.org/10.1093/jleo/ewn002 j.jclepro.2020.123710 122 Poland Public Finance Review SAVE International (2020) Value Methodology Standard Singer, M., Konstantinidis, G., Roubik, E., & Beffermann, Reference. (Rev. 9/25/2020) Available online: E. (2009). Does e-procurement save the state money? https://www.value-eng.org/general/ Journal of Public Procurement, 9(1), 58 – 78. custom.asp?page=ValueStandards World Bank Data (2021): GDP (current LCU) 1960 – 2020, (accessed on 25/05/2021) https://data.worldbank.org/indicator/NY.GDP.MKTP.CN. Public Procurement: Saving and Greening 123 ANNEX 4.1 METHODOLOGY The analytical framework consists of 2 key com- Pri = αi + ꞵ1*X1i + ꞵ2 * X2i + ꞵ3 * X3i + εi ponents, which work best together but can also be deployed independently: Where Pri represents trimmed relative price for the ith item or service purchased; 1) Purchasing System Review: public procurement market overview, using informative visualizations; X1i stands for the set of predictors that can 2) Strategy Package: regression modelling of unit be directly influenced through policy for the ith item prices across government. or service purchased such as the choice of procure- ment method; First, the public procurement market overview or exploration component makes use of simple vi- X2i represents the set of predictors that sualizations to both set the stage for subsequent can be indirectly influenced through policy for the analysis and provide initial insights into cost driv- ith item or service purchased such as the number ers. A high-level overview is essential for identifying of bidders. the scope and boundaries of the analysis such as the amount of total spending, spending break-down by X3i denotes the set of control variables ac- product group (e.g. goods, or more detailed product counting for structural factors not amenable to poli- category such as vehicles) and geography (e.g. main cy intervention for the ith item or service purchased regions of spending), trends over time, and public such as the year of purchase. procurement share in the national budget. In addition, the overview component also begins to unpick major εi stands for the error term of the regres- cost drivers by allowing users to explore bivariate re- sion model. lationships in great depth. Most of the main explan- atory factors are incorporated already in this com- Employing relative prices as a proxy for value for ponent such as the impact of market concentration money requires making a set of assumption. First, on prices with interactive functions allowing users its validity as a price measure crucially rests on the to filter subsamples (e.g., high value tenders or a se- assumption that the estimated lot price is accurate- lected region). The analytical tools are kept simple in ly defined without bias or deliberate manipulation. this component predominantly restricted to simple Given that the methodology for contract value esti- visualizations and tables without explanatory mod- mation prior to launching a tender is regulated in els and predictions. Poland (Chapter 5 of the Polish Public Procurement Law), the price estimation methodology is assumed Second, the regression modelling of relative pric- to be consistent in theory. In practice, implementa- es directly builds on the first component by bring- tion may vary by procuring entity and type of product, ing together all major explanatory factors into nevertheless there is a strong practice of using past a single regression model. Such a comprehensive similar contract values as benchmarks. Second, as the model allows for system-wide price predictions and initial price estimate is based on an assessment of the simulating hypothetical scenarios. In particular, the market and key features of the purchase, it should following linear regression model for relative pric- account for quality and quantity differences when es of products, works or services is estimated at the setting the price. Nevertheless, prices in some mar- level of item, work or service purchased: kets are likely harder to estimate thus cross-market 124 Poland Public Finance Review comparisons may be biased. Third, estimated values based modelling may overestimate price impacts do not consider full life-cycle costs such as mainte- of policy interventions due to unobserved qual- nance costs borne by the user as they are predomi- ity differences. While this bias is likely less pro- nantly focused on purchase price. Fourth, contract nounced for relative prices-based modelling. This, award values do not account for cost changes due however, is more likely to suffer from biases of ref- to contract modification during the contract imple- erence prices orienting bids and reference prices mentation phase or changes in the specifications of being based on outdated historical prices. At any the delivered products (e.g. quality changes). If these rate, using relative prices rather than unit prices systematically vary across markets or buyers, esti- carry the advantage of being able to model price mates may be biased. effects across the whole market not only on stan- dardized goods and services. Data on unit prices for standardized goods and services, an alternative dependent variable, are The full list of variables used in the Polish analysis not readily available in public procurement are reported in Table 4.2. Each of these policy-rele- data in Poland. Experience with unit prices in vant variables derive from the literature using quali- other countries and comparisons of unit prices tative as well as quantitative methods while also being and relative prices suggest that these 2 alterna- supported by robust economic theory. Interpretation tive measures of prices only partially overlap (lin- and the relevant literature are discussed in the re- ear correlation coefficient around 0.5). Unit price- sults section below. Table A4.1.1 Overview of indicators used for price modelling, Poland Type Group Variable name Structural Market characteristics Market ID: reflecting product code (CPV code L2) VAT (23% and 8%) Supply type (G/W/S) Year of contract award (2010-2020)a Can be directly in- Tender specifications Month of bidding (January, February, etc.) fluenced through policy Electronic auctions (Y/N) Product method (open, restricted, etc) Advertisement period length (days) Decision making speed (days) Share of failed tenders % Framework agreement (Y/N) Bundled tenders (product diversity) Buyer characteristics Buyer type (ministry, independent agency, etc) Buyer sector (defence, education, health, etc) Buyer location (NUTS3 - subregions) Annex 4.1 125 Type Group Variable name Can be indirectly in- Bidder/supplier Buyer-supplier from the same region (Y/N) fluenced through characteristics policy Share of SME bidders Supplier specialization: number of markets the company supplies (deciles) Bidding outcomes Number of bidders Annual winner market share (deciles) Annual winner share in buyer spending (deciles) a 2020 is an incomplete year. Market or product group characteristics encom- istics can be directly influenced through policy in- pass those factors which describe the techno- tervention such as capacity building training, many logical and competitive structure of the mar- reforms may be challenging to implement such as ket where the items are purchased from. Most of giving more discretion to purchasing officials in the these variables are in effect structural givens from wake of public demands for accountability and im- the perspective of the analysis as they tend to be partiality. Among many others, procurement staff changing slowly or policy can only influence them capacity lowers prices across many contexts (Best, at high cost. Procurement markets can be defined Hjort, & Szakonyi, 2017), for example, in the US fed- as a combination of product codes and geographical eral bureaucracy, one standard deviation increase in codes (Fazekas & Tóth, 2016) which capture funda- competence decreases cost overruns by 29 percent mental differences, among others, by product com- (Decarolis, Giuffrida, Iossa, Mollisi, & Spagnolo, 2018); plexity and technological characteristics, geograph- or increasing procurement officers’ autonomy com- ical characteristics (e.g., remoteness), and market pared to their auditors reduces prices of standard concentration. goods by 9 percent in Punjab, Pakistan (Bandiera, Best, Khan, & Prat, 2019). Tender specifications capture all the conditions defining who can bid, under which conditions Bidder and supplier characteristics capture the and according to which assessment criteria. These key determinants of companies participating in variables can be typically directly influenced by pol- public procurement which determine their abil- icy makers without changes to the legal framework. ity to offer low prices such as productivity and Tender specifications variables are defined in the ten- capacity. As suppliers are selected through tenders der preparation and bidding stages of the procure- and results at least partially depend on the choices ment process. Among many, advertising tenders in a companies make, these characteristics can gener- widely used, free, online portal tends to lower pric- ally be only indirectly influenced by policy makers. es, such as in Italian public works tenders where the Among many factors, company location and size is effect size is 7 percent higher winning rebates (i.e. likely to influence prices while company risks such discounts compared to the reference price) (Coviello as tax haven registration or political connections & Mariniello, 2014); or longer term, fixed price con- also typically impact prices (Fazekas, Tóth, & King, tracts are more expensive in Russian sugar purchases 2016). For example, in India, roads built by political- while larger volumes lead to lower unit prices (Andrey ly connected contractors are on average 11 percent Yakovlev, Bashina, & Demidova, 2014). more expensive (Lehne, Shapiro, & Eynde, 2018). Buyer characteristics-related variables describe Bidding outcomes corresponding to the intermedi- the level of administrative capacity in the buying ate or final results of the tendering process natu- public entity such as a purchasing office or agen- rally influence unit prices contracted. These factors cy (Fazekas, 2017). While many of these character- can generally be influenced only indirectly through 126 Poland Public Finance Review policy given the important role played by bidder de- of bidders) could be considered. Alternative, more cisions. A widely cited factor, the number of bidders, advanced modelling strategies such as Multilevel plays a highly influential role in determining pric- Modelling incorporating different measurement lev- es under non-collusive conditions (Fazekas & Kocsis, els or Structural Equation Modelling incorporating 2017; Andrey Yakovlev et al., 2014). complex and often indirect relationships among ex- planatory factors hence represent a natural exten- For the sake of simplicity, we opted for a straight- sion of the OLS approach. forward ordinary least squares (OLS) estimation of the regression parameters. This is a computa- The adopted methodology carries the advantage of tionally efficient estimator able to handle models run exceptionally wide scope, while the identified cor- on millions of records while containing hundreds of relations are not necessarily causal. Nevertheless, explanatory variables. Including a battery of fixed ef- it combines strong, well-tested theories underpin- fects for structural factors such as market and year ning the choice of explanatory factors as well as a allows for building models with medium explanatory mixed methods approach, including case studies in power: 15.37 percent. In spite of the linear function- component 3, which together tentatively support a al form, the modelling framework is able to account causal interpretation. In addition, the explanatory for a range of nonlinear relationships by adopting a power of the models, lends support to the claim that logged unit price dependent variable and also look- omitted variable bias doesn’t plagues our estimations. ing at alternative formulations of the independent variables, for example taking deciles of market con- In addition, in-depth case studies could be added centration as a set of dummies rather than the con- to regression modelling results. This would allow tinuous scale. us to get a more precise understanding of the caus- al mechanism underpinning the regression parame- This modelling framework cannot fully account ters and also to look at additional cost drivers such as for a range of complex relationships between fac- product specifications. Such a detailed analysis draw- tors on multiple levels, and alternative model- ing on quantitative data but also on diverse qualita- ling strategies could be considered. Interactions of tive insights also considers the dependent variable market and tender level factors or explanatory fac- more comprehensively. It would look at maintenance tors influencing not only the dependent variable but and running costs on top of purchase price, taking also each other (e.g., advertising tenders not only in- the dependent variable definition closer to the ide- fluences prices directly but also through the number al, full life-cycle cost. CHAPTER 5 ENHANCING PUBLIC FINANCIAL MANAGEMENT AND FISCAL TRANSPARENCY  — PROGRESS AND PERSISTENT CHALLENGES INTRODUCTION coverage for the calculation of the budget deficit and public debt, with parallel reporting to EUROSTAT, using the European System of Accounts 2010 (ESA Effective implementation of fiscal policies and 2010). The lack of consolidated financial statements public expenditure programs requires a robust on an accrual basis makes it difficult to assess the institutional framework for public financial man- true and fair financial position (including all assets agement (PFM) and fiscal transparency. Adequate and liabilities) and performance even for the cen- planning and budgeting procedures are needed to tral government. The annual budget execution re- help allocate and track expenditures according to gov- port does not include a balance sheet, information ernment priorities within the applicable fiscal rules. on non-financial assets or bank balances although Moreover, accounting systems that allow the timely it includes some information on financial liabilities and accurate recording and control of spending pro- (e.g. amounts payable, debt and contingent liabilities cesses help provide information on the government’s resulting from issued guarantees) and receivables. fiscal and financial positions. Furthermore, report- Non-financial assets are reported in a separate re- ing procedures that follow applicable standards help port on state treasury assets issued by the General ensure accountability and transparency. These are Counsel of the Republic of Poland, but they are not all important elements of a modern PFM framework. included in the annual budget execution report, and are not subject to an external audit. Poland has updated and improved its PFM process- es and systems but continues to lag its European Accelerating the implementation of PFM reforms peers. A Budget System Reform (BSR) and a Public is critical for an effective implementation of na- Sector Accounting Reform (PSAR) were initiated in tional recovery plans. EU member states are prepar- 2016 with the preparation of comprehensive concep- ing to implement reforms and undertake investments tual frameworks that follow international best prac- financed under the temporary EU Next Generation tices. Yet, progress in the implementation of these Recovery and Resilience Facility to support the re- conceptual frameworks has been slow, with limited covery from the COVID-19 crisis. Adequate regula- adjustments made to the regulatory framework. A tions, processes, and tools are needed to time-bound action plan is needed for the implemen- tation. Some adjustments have been made to the reg- 1) support an effective, swift, and transparent allo- ulatory framework. The sustained reform plans are cation and utilization of the funding; and included in the recent Convergence Program and the 2) facilitate the monitoring of progress and report- four-year MoF Directions of Actions and Development ing of results according to established account- of the Ministry of Finance for 2021 – 24. ability and reporting standards. Existing gaps in institutional arrangements and systems might af- The slow implementation pace of budgeting and fect the effective implementation of Poland’s na- accounting reforms in Poland creates transpar- tional resilience and recovery plan. ency challenges and results in limits to account- ability mechanisms. Poland continues to have an The chapter reviews existing challenges in PFM incremental budgeting system, with cumbersome and fiscal transparency in Poland, focusing on and ineffective line-by-line appropriations, rising the budgeting and accounting systems and on fis- off-budget expenditures via extra-budgetary funds cal transparency practices. The review follows the and entities, a weak medium-term and performance principles and conceptual framework set out by the perspective, and lack of a systemic spending review World Bank Center for Financial Reporting Reform mechanism. Although the public sector Generally (CFFR) in the context of the regional Public Sector accepted accounting principles (GAAP) are accru- Accounting and Reporting Program (PULSAR),1 which al based, the application of public finance and pub- suggests an integrated view of the whole public fi- lic sector accounting laws and regulations result in nancial management cycle with budgeting, account- the use cash-based budget execution reports and the ing, and reporting procedures as the core elements use of cash-basis national methods with narrower of an effective PFM system (Figure 5.1). The chapter 130 Poland Public Finance Review Figure 5.1 Public Financial Management Cycle Products Policy Cycle Functions Financial Statement Public / Legislative Policy Financial Budgeting Scrutiny Formulation Planning Budgetary Reporting Preparation Cross-cutting Institutions & Human IT Elements Regulation Capacity System Statistical Reporting Authorization External Financial Audit Reporting Accounting Execution & Internal Control Disclosure Recognition Measurement PSA Environment Functions Source: World Bank Center for Financial Reporting. reviews progress in the implementation of the 2016 ing the required adjustments in the budget calendar, as Budget System Reform program and presents rec- well as a review of institutional roles and responsibil- ommendations on how to accelerate the reform ef- ities in the budget process for line ministries, the MoF fort, drawing on existing reports.2 and the CoM. The reform calls for the introduction of a new classification of budget revenues and expendi- tures, with a consistent chart of accounts and unifica- tion of the traditional budget classification and a per- THE BUDGET SYSTEM formance-based classification. Data collection systems would be organized to allow, among other things, the integration of budgetary reporting and financial re- The Budget System Reform3 initiated in 2016 was porting. At the same time, as part of the reform, the designed to address the main deficiencies and chal- use of expenditure reviews and other instruments to lenges in the public financial management systems support the efficiency of public funds spending will and practices. The key objective of the BSR was the be institutionalized. introduction of a budgetary system that supports the achievement of strategic goals and policy priorities, Progress in the implementation of the BSR has taking a long-term perspective, and in observance of been slower than anticipated. The overall imple- applicable fiscal rules. Such a system would allow for mentation plan has not been yet officially published. a more effective management of current and antici- On April 30, 2021, the Council of Minister adopted pated fiscal pressures stemming from population ag- Multiyear Financial Plan for the State for 2021 – 2024 ing, increasing social spending, and the implementa- (Convergence Program)4 which includes information tion of the green and digital agendas. The reform plans, on the continuation of the BSR. approved by the Council of Ministers (CoM) in 2016, entail the introduction of a medium-term budgetary A key priority for the government is the phased framework (MTBF) that would transform the budget development and implementation of a uniform into a more effective fiscal planning tool to reflect me- Standard Chart of Accounts (SCoA) by mid-2022 dium-term government priorities. They also integrate for the public finance sector integrated with the multi-annual and annual planning processes, includ- budget classification. After setting up adequate Enhancing Public Financial Management and Fiscal Transparency — Progress and Persistent Challenges 131 and effective governance arrangements and working PUBLIC SECTOR groups for the SCoA, some progress has been made in defining the conceptual structure of the SCoA for ACCOUNTING standardizing the economic, functional, and admin- istrative segments based on GFS/ESA to ensure con- In order for the government to assess more ac- sistency with regional and international reporting curately its overall financial position and the requirements for the general government and pub- fiscal risks, it is necessary to simplify and stan- lic sector. Next phases will include: dardize complex public sector accounting reg- ulations and increase the availability of aggre- 1) defining the structure of the new budget classi- gated information on an accrual basis. The Public fication integrated with the chart of accounts in Sector Accounting Reform (PSAR) is part of the mod- the form of a report with business assumptions; ernization of PFM practices and systems initiated in Poland in 2016. It is a critical instrument for sup- 2) devising an implementation plan for the new sys- porting implementation of fiscal policies, as well tem and IT tool in 2022; and as for enhancing fiscal and financial accountabili- ty in line with international best practices. The re- 3) piloting the new classification system in selected form considers alignment of public sector account- units in order to identify possible gaps in the new ing practices with the International Public Sector classification system, and areas for improvement; Standards (IPSAS). and in order to describe the results of testing the functioning of the designed solutions in 2023. Poland does not prepare or present consolidated whole-of-government or general government-lev- Some progress has been made in the implemen- el financial statements on an accrual basis that tation of the reform programs. The progress in- shows all assets and liabilities according to inter- cludes amendments to the Public Finance Act to in- national standards.5 Many of the issues highlighted corporate reforms in the budget process, and the in the 2015 IPSAS Gap Analysis report by the World introduction of multi-year forecast requirements in Bank’s Center for Financial Reporting Reform (CFRR) the 2019 MoF budget regulation. The regulation in- are still valid today. This includes the finding that troduces definitions of a “no-change” policy stance, Polish public sector accounting regulations were com- and baseline expenses amounts — indicative expen- plex, including sets of laws, regulations, sector-spe- diture amounts to two additional years after the bud- cific requirements, and separate budgetary reporting get year, which the MoF sets for the budget entities procedures. Poland’s regulations include applying the together with limits for the budget year. In parallel, accrual basis at entity level, but there are many ex- there are ongoing conceptual works on the MTBF in emptions. Moreover, information on an accrual ba- the Public Finance Act. Legislative solutions for the sis is generally not used in the decision-making pro- new classification system, as well as a new model cesses. The Accounting Act (AA) applies to both the of budget management that takes into account the corporate sector and to approximately 71,000 public MTBF should be developed by 2025. Other recom- sector entities, including State-Owned Enterprises mended actions including setting up an indepen- (SOEs). Nevertheless, approximately 68,000 of these dent fiscal council or deploying a change manage- entities have been granted exemptions from the main ment strategy have not been initiated. requirements related to presenting financial state- ments, consolidation, and auditing. The work on the institutionalization and the meth- odology of the public expenditure reviews contin- Although it uses accrual accounting at entity level, ued. In 2020, two pilot expenditure reviews related Poland continues to operate with cash-basis ac- to state budget sector and prisons employee salaries, and counting and budgetary reporting systems. Cash- support to employment and counteracting unemploy- basis accounting makes it difficult to accurately assess ment have been completed with OECD support and the government’s overall financial position and the their findings were used in the budget discussions. management of fiscal risks. On the other hand, it fa- 132 Poland Public Finance Review cilitates the shifting of spending and revenues across tional accounting standards. Transparency entails budget cycles. Moreover, the use of Special Purpose reliable and timely publication of information on Vehicles (e.g. extra-budgetary funds), which are not fiscal outcomes, fiscal forecasts, and assumptions included in the public sector as defined by the Public over the medium term, as well as quality informa- Finance Act, represent extra-budgetary operations, tion about past, present, and planned fiscal oper- increasing off-budget expenditures and thus weaken- ations. It also includes the processes, procedures, ing the effectiveness of the constitutional debt ceiling. and institutional arrangements for public financial management. Availability and accessibility of fiscal The MoF has undertaken several important ini- and financial information help relevant stakehold- tiatives in line with the recommendations of the ers in a country keep track of how public resourc- CFRR report. These include designing a standard- es are managed, and help governments not only to ized chart of accounts to be integrated with the bud- enhance their accountability but also to raise stake- get classification; preparing the phased implementa- holder awareness on the strategic choices and con- tion of consolidated financial statements; an approach straints affecting the public finances. towards increased alignment with IPSAS; updates to some of the regulations, including the preparation of Disclosure of Polish public finances has improved the balance sheet of the State Budget (covering mainly since 2017, on all dimensions of data availabil- financial assets and liabilities as the first steps toward ity, accessibility, and reusability. Poland now full accrual balance sheet for the central government) scores above the OECD average on the availability by the MoF to be issued for the first time in 2022 for and accessibility of open and useful government year 2021; inclusion of the tax revenues transactions data. Satisfaction and confidence across public ser- into the MoF accounts starting in 2018; and addition- vices, including the national government in Poland, al disclosure requirements for public sector entities. have also increased substantially between 2007 and 2018 (up to 43 percent), although remaining slight- The first phase of PSAR, which included the con- ly below the OECD average (45 percent). cept of the consolidation of government finan- cial statements, was completed in 2020. The con- Poland scores below the OECD average ratings on cept envisions a phased implementation including budget transparency, public participation, and methods of consolidation, reporting templates, and oversight, however. Budget information provid- preparation deadlines, as well as proposals for the ed by the Polish government is assessed as limited ICT solution and a cost-benefit analysis of the pro- in the Open Budget Survey (OBS) 2019, which ranks posed reforms. Apart from improving management Poland 32nd out of 117 countries with a transparency of the state assets and liabilities, the proposed re- score of 60 (out of 100), below the OECD average of form would also increase alignment and compliance 68 (Figure 5.2). Poland does not publish the budget of Polish accounting regulations with IPSAS, which mid-year review online and does not always produce is currently estimated at 68 percent.6 and publish the Citizens Budget in a timely manner. A more detailed macro-economic forecast, including projected interest rates for the upcoming budget year; estimates of total expenditures and revenues for the FISCAL TRANSPARENCY upcoming budget year; estimates on government bor- rowing and debt; and multi-year expenditure projec- tions could improve the comprehensiveness of the Fiscal transparency is a critical element to sup- pre-budget statement. The complexity of the public port effective and efficient fiscal and financial finance sector revenues and expenditures can be as- policy implementation and to drive PFM system sessed from the maps created by the NGO Fundacja reforms. Fiscal transparency should be understood Republikańska (website at https://www.mapawydat- as the ability of a government to produce and dis- kow.pl/), which presents an easier way for citizens to close fiscal and public financial information in the grasp the budget situation than the fragmented data scope, using a form and content that follow interna- provided on the website of the Ministry of Finance. Enhancing Public Financial Management and Fiscal Transparency — Progress and Persistent Challenges 133 Figure 5.2 Open Budget Survey 2019 — Fiscal Transparency Poland’s rankig: Global Average 32 of 117 countries OECD Average Bulgaria Germany Romania Ukraine Poland Slovak Republic Czech Republic Hungary 0 10 20 30 40 50 60 70 80 90 100 0 61 100 Insufficient Sufficient Source: International Budget Partnership. Comprehensive information on tax expenditures 1) the Legislature debates budget policy before the is not published as part of the budget execution Executive’s budget proposal is tabled and approves reports, while information on contingent lia- recommendations for the upcoming budget; bilities is mainly limited to issued guarantees. These are two important elements that could con- 2) the Legislature approves the Executive’s budget tribute to increased fiscal transparency. Making proposal before the start of the budget year; and information on tax expenditures available is a le- gal requirement in most of the EU member states; 3) a legislative committee examines in-year budget however, practices vary significantly. Disclosure of implementation and publishes its findings online. tax expenditures is critical to allow for a compre- hensive view of the country’s fiscal and budgetary The institutional framework supporting fiscal priorities and the actual allocation and use of pub- transparency can be strengthened further. Poland lic resources. Similarly, an accurate and compre- is currently the only EU member state without an hensive view of contingent liabilities is important, independent Fiscal Council. Fiscal councils operate as these could be a major source of fiscal distress as watchdogs for monitoring fiscal policy, assessing (including from SOEs, PPPs, subnational govern- macroeconomic and budgetary forecasts, overseeing ments, financial sector, and from special purpose compliance with fiscal rules, and analyzing long-term vehicles). An accurate assessment and disclosure of sustainability of public finances. Some of these func- these type of liabilities would allow stakeholders to tions mentioned are currently performed by different have a more complete picture of the country’s fiscal bodies in Poland, but this fragmentation of functions position and fiscal risks. weakens their impact (European Semester, 2018 – 208). Inclusive public participation is crucial for real- izing the positive outcomes associated with great- er budget transparency. Public participation in the budget process during both the approval (legislative) DRIVERS OF and implementation (executive) stages is lower in INSTITUTIONAL REFORM Poland than in other countries, with Poland scoring 24 (out of 100) on public participation, according to the The Poland Convergence Program9 and the MoF Principles of Public Participation in Fiscal Policies.7 strategic directions for 2021 – 24 recognize the Meanwhile, budget oversight is assessed as adequate, need to advance the implementation of BSR and including legislative oversight and audit oversight, PSAR to achieve their medium-term fiscal ob- with Poland scoring 78 and 95 out of 100, respectively. jectives. Ensuring stable public finance is a key Further improvements could be achieved if objective of the government, included in the four- 134 Poland Public Finance Review year strategic document Directions of Actions and starting from the national government accounting Development of the Ministry of Finance for 2021 – 24.10 standards and transforming them from cash to ac- The government aims to reduce general government crual basis, which can be challenging.13 Meanwhile, deficit and debt (ESA 2010 methodology), includ- IPSAS, as a set of accrual accounting standards, al- ing by developing a public finance consolidation lows for comprehensive debt and investment finan- strategy, that considers recommendations of the cial reporting on an accrual basis. This can substan- EU Council ECOFIN, EDP, and national fiscal rules, tially reduce the risk of systematic errors in the data as well as by monitoring progress in public finance used for preparing government finance statistics consolidation, in line with EU Council recommen- and informing policymaking.14 dations and ensuring compliance with the public finance security rules. To increase effectiveness of Through Directive 2011/85, the EU sought to im- public funds management, the government is con- prove economic governance in the wake of the sidering defining a new model of budget manage- global economic and financial crisis. Institutional ment, revising the budget calendar, and introduc- settings at national level can play an important role ing a medium-term budgetary framework. It is also in containing spending and avoid deficit biases; they planning the development and implementation of a include procedural rules of the budgetary process; unified, multidimensional budget revenues and ex- numerical fiscal rules guiding or constraining pol- penditures classification, integrated with the uni- icymakers’ discretion; independent fiscal bodies or fied chart of accounts, and strengthening assets and institutions in charge of providing inputs (e.g., fore- liabilities management. All these actions are fully cast, analysis); and formulating recommendations aligned with the objectives of the BSR and PSAR. in fiscal policy. The Directive contained a propos- al for a Council Directive on requirements for bud- EU law related to fiscal and public financial man- getary frameworks of the Member States and man- agement is an important external driver of PFM dated the application of the ESA 2010 standards.15 and fiscal transparency reforms in Poland and Lack of a unified EU public accounting standard other EU member states. ESA 2010 requires fiscal and the lack of common national standards are the reporting to EUROSTAT on an accrual basis for all main difficulties in implementing this Directive, EU countries. Ensuring uniform and comparable according to stakeholders from the member states. accrual-based accounting practices for all sectors Unfortunately, the project to create European Public of general government within the EU enhances the Sector Accounting Standards (EPSAS) for the EU did quality of the data on which the European System of not materialize, delaying public sector accounting Integrated Economic Accounts (ESA) is based, and reforms and implementation of IPSAS. consequently improves budget oversight and fiscal monitoring at the macro level to enable sound fis- The EU Recovery and Resilience Facility (RRF) cal policy decision-making.11 could also be an important PFM reform driver. Under the RRF, member states can access extraor- Article 126 of the Treaty on the Functioning of dinary financing to support their post-crisis recov- the European Union (TFEU), and Protocol No. 12 ery objectives, which might expose challenges re- on the Excessive Deficit Procedure annexed to lated to absorption and transparency. The Facility is the Treaty refer to fiscal targets to be observed centered around six priority pillars and specific al- by Member States.12 Moreover, the Protocol pro- location requirements. The priority pillars include vides relevant guidelines on fiscal and financial green transition; digital transformation; econom- reporting standards. In addition to setting the nu- ic cohesion, productivity, and competitiveness; so- merical deficit and debt fiscal rules, it establishes cial and territorial cohesion; health, economic, so- the requirement to follow the ESA 2010 definitions, cial, and institutional resilience; and policies for the which monitors financial flows on an accrual basis. next generation. At least 37 percent of funds are ear- Compiling ESA accounts, however, often entails marked for climate change, and at least 20 percent Enhancing Public Financial Management and Fiscal Transparency — Progress and Persistent Challenges 135 for digital transformation. RRF is a response to the REFORM OPTIONS AND negative impact of COVID-19 by helping the EU’s economies to become more resilient and fostering RECOMMENDATIONS the green and digital transitions. The MoF remains committed to the important BSR Modern budget and accounting systems are crit- reform and has established priorities for imple- ical for effective management and absorption of mentation. These priorities include a complete re- these extraordinary funds to ensure building form and unification of the chart of accounts (CoA) back better. The country’s budget and accounting and budget classification, and the introduction of the systems should provide tools (including ICT solu- MTBF. While the decision to prioritize these two re- tions) to allow for streamlined budgetary proce- forms is commendable and could yield results in ad- dures and cycles, producing relevant budgetary vancing the broader reform process, there are crit- and financial information in an orderly and timely ical technical aspects that need to be considered as manner, with sufficient level of detail and the pos- each of the reforms is linked to other important piec- sibility to tag and aggregate information using rele- es of the system that need to be reformed in parallel. vant criteria, including “green” and climate related. Examples of climate budgeting tagging methodolo- The reform process needs to be considered in an gies, climate finance reporting and climate expen- integrated manner. Introducing the new Standard diture reviews are provided in the 2021 World Bank Chart of Accounts (SCoA), for example, will require report: Climate Change Budget Tagging: A Review of improvements in organizational arrangements, in International Experience.16 Such budgetary and fi- accounting and financial reporting standards, and nancial information, including relevant tagging, in the accounting systems. The MTBF will require should be subject to audit and public disclosure adjustments to the budget calendars, identification following accounting and transparency standards. of programs, and the introduction of spending re- views. The IMF report highlights such requirements The RRF funds will be made available to the mem- in detail and provides specific technical recommen- ber states based on approval by the EC of nation- dations to advance the process. al recovery and resilience plans (NRRP) that in- clude a set of the reforms and public investment Further alignment of Polish public sector Gen- projects aligned with the six pillars with a re- erally Accepted Accounting Principles (GAAP) sults-based disbursement approach. To take full with IPSAS would contribute to increased finan- advantage of the RRF, the national public financial cial accountability, as well as improved manage- management framework should be capable of prop- ment of public resources and fiscal risks. Giv- erly allocating, executing, reporting, monitoring, en the complexity of the processes involved, full and evaluating the use of funds in respective pro- alignment can only be achieved over the medium- gram areas. Expenditure programs will require spe- to long-term and requires significant investments cific milestones and targets and procedures to mea- that need careful planning and sequencing. Experi- sure and verify progress and achievement of results ences from other countries underscore the impor- will need to be in place. Disclosure and transpar- tance of the commitment from senior management ency requirements will also need to be observed. and politicians to ensure buy-in and participation of Member states that do not have in place established key stakeholders and in widely disseminating infor- PFM arrangements capable of supporting program- mation on the benefits of investing in financial re- matic, multi-year and results-based budgeting and porting on an accrual basis. Developing a compre- accounting procedures, would have to develop inter- hensive program should be based on preparatory im arrangements to support RRF related programs. studies, including an assessment of actual public sec- Poland could use the opportunity of the RRF and tor accounting practices as well as an assessment of NRRP processes to further accelerate implementa- the institutional framework and any capacity con- tion of the BSR and PSAR thus avoiding the need for straints. In the short term, key stakeholders would such interim solutions. also benefit from further awareness raising and ca- 136 Poland Public Finance Review pacity building, including IPSAS training, peer learn- is both affordable and amply justified, considering ing from other EU countries that apply IPSAS-based the expected economic and fiscal benefits. standards, and policy-setting workshops to consid- er and stimulate thinking on these issues. The key recommendations of the 2015 IPSAS Gap Analysis report by CFRR still apply, a reflection In the short term, the current budget execution re- of the complex reform environment. Actions to ports on a cash basis should be supplemented with strengthen the Polish public sector GAAP, which may additional information on an accrual basis, includ- also be regarded as steps towards any eventual adop- ing on assets and liabilities. This is critical to better tion of an accrual basis IPSAS, include: inform policy and decision makers in their manage- ment of public resources and fiscal risks. A first step A) Simplifying and standardizing Polish public sector ac- would be to use the information already produced in counting regulations across all public sector entities. accordance with Polish public sector GAAP. Thereafter, This includes the design and implementation of improvements to Polish public sector GAAP, ideally a unified chart of accounts to facilitate the pro- in line with IPSAS, would help further bridge the in- vision of information to management, and bud- formation deficit and allow for incorporation of ESA getary/fiscal reporting. 2020 requirements for national methodology of fis- cal reporting instead of maintaining dual standards. B) Developing aggregated accrual financial informa- tion to achieve a better understanding of govern- Pending reform steps include completing the de- ment’s overall financial position and fiscal risks than velopment of the SCoA and the integration of fis- the one provided by current consolidated cash-based cal and financial reporting across the PFM cycle. budget execution reports. This can be implement- The SCoA needs to be fully developed at both the ad- ed as a phased process with consolidation to fol- ministrative and economic segments. The adminis- low the existing budgetary process with finan- trative segment needs to identify accountability lev- cial statements (together with comprehensive els at which individual and consolidated financial disclosures) produced at central and subnation- reports should be compiled, review and refine the al government levels. taxonomy of the Public Finance Sector, classify ap- propriate sub-types of entities, and define subsidi- C) Revising the coverage and definition of public sec- ary reference tables for reporting by other combina- tor entities. The Public Finance Act omitted cer- tions of public entities as required for statistical or tain public sector entities from the definition of legislative reporting. The economic segment of the a public sector entity. Polish public sector GAAP SCoA needs to be further aligned to the generally ac- should be reviewed and revised to include all ap- cepted concepts of revenue, expense, assets and lia- propriate entities and transactions. bilities; and apply the notions of relevance, materi- ality, and cost benefit. Integration of fiscal reporting The MoF could consider preparing and publish- should adopt a balanced view of different purposes ing a comprehensive road map and implemen- and uses including statistical reporting, budgetary, tation plan for sequencing the interconnected and financial, and ensure that all relevant user needs SBR and PSAR. It should be possible to do so based are equally represented and considered in the devel- on existing information, gap assessments, analytical opment of the SCoA.17 Furthermore, both financial reports, technical assistance support, etc. The road reporting and the development of the SCoA need to map and plans could be public documents subject to be supported with ICT tools and software applica- public consultation especially with the main stake- tions based on a detailed assessment of existing in- holders, parliament, politicians, and general pub- formation systems as well as on relevant internation- lic. The road map18 should include all main aspects al experience. The cost of EPSAS implementation is of the reform including: estimated between €26 and €28 million, given the ad- vanced level of the public sector accounting and the A) Current status: gap analysis, which has already ICT environment in Poland, a level of investment that been completed. Enhancing Public Financial Management and Fiscal Transparency — Progress and Persistent Challenges 137 B) Objectives: defining a clear rationale and expected data, stakeholder engagement, and further institu- outcomes. This is largely done already, although it tionalization of public participation in budget and could provide more specific results to each of the fiscal management issues. Immediate actions to be main stakeholder groups. One objective should taken could include: be increased fiscal transparency, including in- forming the public about public finance in an A) Improving budget transparency: publishing mid- accessible way. year budget implementation reports online, pro- ducing and publishing a Citizens Budget online, C) Approach: agreed-upon concepts for the reform, and a more detailed macro-economic forecast to which are prerequisite and influence the whole re- improve the comprehensiveness of the pre-bud- form path and implementation. For BSR, the vision get statement (multi-year expenditure projec- has been already published, while for PSAR, the tions with assumptions and indicators). concept was prepared but not published. Before the implementation, agreed-upon concepts must B) Improving public participation in the budget process: be included in the legislative framework. developing pilot mechanisms to exchange views from the public or NGOs in the budget process D) Organizational capacity: a capacity-building pro- during both the approval (legislative) and imple- gram for participants and actors; an IT tools de- mentation (executive). velopment plan should be prepared. C) Making budget oversight more effective: the parlia- E) Implementation processes: a detailed implementa- ment could debate budget policy before the bud- tion plan including timeframe, resources, prior- get proposal is submitted and approve recommen- ities, and sequencing should be prepared to im- dations for the upcoming budget, ensuring that plement agreed-upon approaches and concepts, a legislative committee examines reports on in- e.g., piloting, top-down approach, roll-out, etc. year budget implementation and publishes rec- ommendations online; consider setting up an in- F) Monitoring and evaluation: the process of reform dependent fiscal institution to further strengthen should be regularly monitored and transparent- budget oversight. ly evaluated for results (e.g., by parliamentary commission) allowing for modifications as need- D) Implementing and institutionalizing the Fiscal Council: ed and justified. In line with best practices from EU member states, Poland could benefit from setting up an indepen- G) Reform governance: a cross-cutting element of the dent fiscal council to support analytical and moni- reform implementation, which includes proper toring activities related to fiscal and budget policies. high-level support to the reform, strong leader- ship, dialogue with stakeholders, and ensuring Three additional actions could be considered by the sustainability of the reform to reach objectives. MoF to complement the short and medium-term plans already in place. These could serve as import- H) Communication and culture: another cross-cutting ant reform enablers and become a permanent support element that considers change management (ad- in the transition to an enhanced PFM system: aptation) strategy and the need to inform stake- holders and public about the implementation of A) Carrying out an assessment of ICT capabilities and the reform. requirements to prepare a medium-term strategy that includes ICT solutions to enable collection and consol- MoF could also consider taking immediate ac- idation of financial information, as well as other PFM tions to strengthen budget and fiscal transpar- processes at the central and decentralized level. Given ency practices. Based on the robust baseline of data the project’s strategic relevance, the hardware availability and openness, efforts could focus on pro- and software investments associated with its de- active dissemination of budget- and fiscal-related velopment could be supported through the NRRP. 138 Poland Public Finance Review B) Establishing a permanent and comprehensive training C) Establish a baseline and gap analysis of existing PFM program for government staff involved in PFM-related practices and carry out periodic evaluations of progress. functions. The transition to accrual-based account- Poland could consider using the Public Expenditure ing and adoption of international standards like and Financial Accountability (PEFA) framework for IPSAS will require extensive training and support a robust and systematic assessment and diagno- to public sector staff already used to pre-exist- sis of the entire budget cycle and other PFM areas. ing practices. Given that in Poland, International The PEFA methodology, which is an international- Financial Reporting Standards (IFRS) are required ly-accepted methodology to assess PFM practices, for all domestic companies whose securities trade could be a good option to further fine-tune the re- in a regulated market, there is an opportunity for form action plans and to monitor progress. knowledge transfer from the private to the public sector — despite differences in the IFRS and IPSAS. Notes 1. Stocktaking of Public Sector Accounting and Report- 8. https://ec.europa.eu/info/business-economy-euro/ ing Environment in PULSAR Beneficiary Countries, economic-and-fiscal-policy-coordination/ the World Bank 2020 eu-economic-governance-monitoring-prevention https://cfrr.worldbank.org/publications/stocktaking -correction/european-semester_en -public-sector-accounting-and-reporting-environment 9. https://www.gov.pl/web/finanse/wieloletni-plan -pulsar-beneficiary -finansowy-panstwa 2. The analysis draws on: S. De Clerck et.al., Poland — Prog- Every April, EU Member States (MS) are required to lay ress Report on Budget System Reform -Developing a Stan- out their fiscal plans for the next three years: “Stability dard Chart of Accounts (SCoA) and a Medium-term Bud- Programs” for Eurozone MS or “Convergence Programs” get Framework (MTBF), Technical Report, FAD, IMF, April for other MS. This exercise is based on economic gov- 2020. European Commission (2020), Review of the suit- ernance rules in the Stability and Growth Pact, which ability of the Council Directive 2011/85/EU on require- aim to prevent the emergence or exacerbation of fis- ments for budgetary frameworks of the Member States, cal difficulties. Staff Working Document SWD(2020) 211 final; European 10. https://www.gov.pl/web/finanse/strategia-mf Commission (2021), Supporting Public Administrations 11. Collection of information related to the potential impact, in EU Member States to Deliver Reforms and Prepare for including costs, of implementing accrual accounting in the Future, Staff Working Document SWD(2021)101; Eu- the public sector and technical analysis of the suitabil- ropean Commission (2020), Updated accounting matur- ity of individual IPSAS standards, Prepared for the Eu- ities of EU governments and EPSAS implementation costs, ropean Commission, PwC (2014) Paper by PwC on behalf of Eurostat. 12. https://ec.europa.eu/info/business-economy-euro/ 3. More details can be found at https://mf-arch2.mf.gov.pl/ economic-and-fiscal-policy-coordination/ c/document_library/get_file?uuid=5fee3505-4860-46c1 eu-economic-governance-monitoring-prevention -8539-2d4ad9349dd1&groupId=764034 -correction/stability-and-growth-pact/ https://mf-arch2.mf.gov.pl/c/document_library/ legal-basis-stability-and-growth-pact_en get_file?uuid=eaa75625-1b8b-40e2-a11f-b1e1aa083909 13. In many EU countries ESA accounts are derived from &groupId=764034 cash-basis public accounting systems, through a se- 4. https://www.gov.pl/web/finanse/wieloletni-plan- ries of “accruals adjustments”. These adjustments are finansowy-panstwa estimated on a macro basis, and as a result they are ap- 5. 2015 IPSAS Gap Analysis report by the World Bank CFRR proximations. Where there are no accruals accounts https://cfrr.worldbank.org/publications/comparison at the micro level, financial transactions and balance -polish-public-sector-gaap-ipsas sheets have to be derived from a mix of different sources, 6. EC 2020 “Updated accounting maturities of EU govern- leading to a “statistical discrepancy” between the defi- ments and EPSAS implementation cost.” https://ec.eu- cit compiled via non-financial accounts and the deficit ropa.eu/eurostat/documents/9101903/9700113/Updat- compiled via financial accounts. ed-accounting-maturities-and-EPSAS-implementation 14. Towards implementing harmonized public sector ac- -cost-June+2020.pdf counting standards in Member States, European Com- 7. http://www.fiscaltransparency.net/pp_principles/ mission Staff Working Document (2013) Enhancing Public Financial Management and Fiscal Transparency — Progress and Persistent Challenges 139 15. Ibid. European Commission (2021), Supporting Public Administra- 16. World Bank. 2021. Climate Change Budget Tagging: A tions in EU Member States to Deliver Reforms and Prepare Review of International Experience. Equitable Growth, for the Future, Staff Working Document SWD (2021)101. Finance and Institutions Insight European Commission (2020), Updated accounting matur- 17. S. De Clerck et.al. ities of EU governments and EPSAS implementation costs, 18. World Bank road map template can be found at Paper by PwC on behalf of Eurostat. https://cfrr.worldbank.org/index.php/publications/ European Commission Staff Working Document (2013), roadmap-public-sector-accounting-reform-good Towards implementing harmonized public sector account- -practice-template ing standards in Member States. OECD (2019) Public Governance Policy Papers: Open, Useful and Re-usable data (OURdata) Index: 2019 DOI: https://doi.org/10.1787/45f6de2d-en References World Bank, Centre for Financial Reporting Reform (May 2015), A Comparison of Polish Public Sector GAAP Ministry of Finance (July 2016), Assumptions of the Budget with IPSAS; https://cfrr.worldbank.org/publications/ System Reform, Warsaw, Poland. comparison-polish-public-sector-gaap-ipsas Ministry of Finance (October 2016, updated June 2017), World Bank (2021), PULSAR — Drivers of Public Sector Ac- Budget System Reform, Program Based Budgeting. Concept counting Reforms. of new division of the State Budget, Warsaw, Poland. World Bank (2018), PULSAR — Good Practice Template — To Ministry of Finance (April 2021), Convergence Program for Public Sector Accounting Reform — Roadmap. 2021 – 2024; World Bank (2020), PULSAR — Stocktaking of Public Sector https://www.gov.pl/web/finanse/wieloletni Accounting and Reporting Environment in PULSAR -plan-finansowy-panstwa Beneficiary Countries. https://cfrr.worldbank.org/pub- Ministry of Finance (2021), Directions of Actions and Devel- lications/stocktaking-public-sector-accounting-and-re- opment of the Ministry of Finance for 2021 – 2024. porting-environment-pulsar-beneficiary S. De Clerck et.al., Poland — Progress Report on Budget System World Bank. 2021. Climate Change Budget Tagging: A Review Reform — Developing a Standard Chart of Accounts (SCoA) of International Experience. Equitable Growth, Finance and and a Medium-term Budget Framework (MTBF), Technical Institutions Insight. Report, FAD, IMF, April 2020. Open Budget Survey 2019, https://www.internationalbudget. European Commission (2020), Review of the suitability of org/open-budget-survey/country-results/2019/poland the Council Directive 2011/85/EU on requirements for bud- Think Tank: Open Eyes Economy & Kolegium Gospodarki i getary frameworks of the Member States, Staff Working Administracji Publicznej Uniwersytetu Ekonomicznego Document SWD (2020) 211 final. w Krakowie, Alert Gospodarczy 14/2020. 140 Poland Public Finance Review