Republic of Fiji Systematic Country Diagnostic P160757 June 12, 2017 WORLD BANK GROUP Contents 1. Poverty reduction and shared prosperity .............................................................. 15 Poverty ....................................................................................................................... 15 Shared prosperity ....................................................................................................... 21 Drivers of poverty reduction and shared prosperity .................................................. 23 2. Determinants of economic growth ......................................................................... 25 Patterns of growth ...................................................................................................... 25 Drivers of growth ....................................................................................................... 26 Financing of growth ................................................................................................... 34 Drivers of volatility.................................................................................................... 36 3. Building on opportunities and strengths ............................................................... 39 Government’s vision .................................................................................................. 39 Pacific Possible ......................................................................................................... 40 Achieving Fiji’s vision .............................................................................................. 42 4. Pathway I. Stronger growth .................................................................................. 43 Why is investment low?............................................................................................. 43 Why are exports weak? .............................................................................................. 51 Why is employment low? .......................................................................................... 55 Policies to accelerate inclusive growth ...................................................................... 59 5. Pathway II. Better access to services by all ........................................................... 61 Economic mobility and inequality in opportunities................................................... 61 Education ................................................................................................................... 62 Health ......................................................................................................................... 66 Housing ...................................................................................................................... 69 Connective infrastructure........................................................................................... 71 Financial inclusion ..................................................................................................... 72 Gender inequality and violence against women ........................................................ 75 Policies to improve access to services for all ............................................................ 77 6. Pathway III. Building resilience ............................................................................. 79 Climate resilience ...................................................................................................... 79 Fiscal resilience ......................................................................................................... 83 Protecting the vulnerable ........................................................................................... 89 Policies to build resilience ......................................................................................... 92 7. Priorities ................................................................................................................... 94 Policy priorities .......................................................................................................... 94 Cross-cutting priority: Strengthening institutional capacity of the public sector ...... 97 Knowledge gaps....................................................................................................... 101 2 Boxes Box 2.1 Fiji's growth history ......................................................................................................... 25 Box 2.2 Land tenure in Fiji............................................................................................................ 29 Box 3.1 Fiji’s ambition to be a regional hub ................................................................................. 39 Box 4.1 Promoting SME in Fiji ..................................................................................................... 45 Box 5.1 Causes of NCDs in Fiji .................................................................................................... 66 Box 5.2 An overview of health care system in Fiji ....................................................................... 67 Box 5.3 Costs of remittances in Fiji .............................................................................................. 74 Box 5.4 Uneven progress on gender equity................................................................................... 76 Box 6.1 The Fiji National Provident Fund .................................................................................... 85 Figures Figure 1.1 Poverty rates, $3.10 a day poverty line, 2014 or latest ................................................ 15 Figure 1.2 Recent poverty rates in Fiji .......................................................................................... 15 Figure 1.3. Changes in poverty (US$3.10/day) ............................................................................. 16 Figure 1.4 Economic mobility ....................................................................................................... 16 Figure 1.5 Poverty maps ................................................................................................................ 20 Figure 1.6 Growth incidence curves, 2002-13 .............................................................................. 22 Figure 1.7 Remittance receipts by consumption quintile: percentage of population..................... 23 Figure 1.8 Changes in poverty rate, 2002–13................................................................................ 24 Figure 2.1 Economic growth, 1970-2015 ...................................................................................... 25 Figure 2.2 Decomposition of growth by sector ............................................................................. 27 Figure 2.3 Evolution of employment by subsector, 1990-2011 .................................................... 31 Figure 2.4 Contribution to growth by demand, 1970-2016 ........................................................... 32 Figure 2.5 Investment to GDP ratio .............................................................................................. 33 Figure 2.6 Trade balances, 1991-2014 .......................................................................................... 33 Figure 2.7 Growth accounting, 1970-2015 .................................................................................... 34 Figure 2.8 Saving and investment ................................................................................................. 35 Figure 2.9 Saving by sector ........................................................................................................... 35 Figure 2.10 Financing of growth ................................................................................................... 35 Figure 2.11 Annual aid flows, average 2006–15 ........................................................................... 36 Figure 2.12 Changes in terms of trade........................................................................................... 38 Figure 2.13 Real effective exchange rate ...................................................................................... 38 Figure 4.2 Most important constraints reported by firms in Enterprise Survey ............................ 46 Figure 4.3 Doing Business in Fiji, 2012 vs. 2017 ......................................................................... 48 Figure 4.4 Cross-country comparison of main indicators of access to finance ............................. 49 Figure 4.5 Goods export diagnostics ............................................................................................. 51 Figure 4.6 Product space map of Fiji’s exports ............................................................................. 52 Figure 4.7 Service export diagnostics ........................................................................................... 53 Figure 4.8 Trading across borders ................................................................................................. 55 Figure 4.9 Number of jobs created from 2000 to 2014 ................................................................. 56 Figure 4.10 Population by age and sex .......................................................................................... 57 Figure 4.11 Schooling in population ............................................................................................. 57 3 Figure 5.1 Economic Mobility ...................................................................................................... 61 Figure 5.2 Human Opportunity Index ........................................................................................... 62 Figure 5.3 Education attainment of boys and girls aged 15 or less in 2008 .................................. 63 Figure 5.4 Composition of education spending ............................................................................. 64 Figure 5.5 Level of education and health spending ....................................................................... 64 Figure 5.6 Decomposition of Inequality in Opportunity ............................................................... 65 Figure 5.7 Beneficiaries of government education budget, 2015 .................................................. 66 Figure 5.8 Beneficiaries of government education related subsidies, 2015 .................................. 66 Figure 5.9 Distribution of health-care benefit: public sector......................................................... 68 Figure 5.10 Distribution of benefit of Public Rental Board subsidies ........................................... 71 Figure 5.11 Access to financial services by bottom 40 percent .................................................... 73 Figure 5.12 Account at a financial institution by gender .............................................................. 74 Figure 6.1 Building exposure in Fiji (in 2010 US$) ...................................................................... 80 Figure 6.2 Average annual losses due to cyclones and earthquakes (in 2010 US$)...................... 80 Figure 6.3 Sources of rural household income .............................................................................. 82 Figure 6.4 Breakdown of agribusiness incomes ............................................................................ 82 Figure 6.5 Tax revenue, average 2010-14 ..................................................................................... 84 Figure 6.6 General government balance ........................................................................................ 84 Figure 6.7 Evolution of spending .................................................................................................. 84 Figure 6.8 Operating spending ...................................................................................................... 84 Figure 6.9 Stochastic projections of the deficit and debt, 2016–21............................................... 88 Figure 6.10 Sources of post disaster financing .............................................................................. 89 Figure 6.11 Coverage of cash transfers in selected countries........................................................ 90 Figure 6.12 Adequacy of cash transfers in selected countries ....................................................... 90 Figure 6.13 Expenditure of Cyclone Winston top-up payments ................................................... 91 Figure 7.1 Indicators of governance, selected countries, 2015...................................................... 97 Tables Table 1.1 Profile of the poor ............................................................................................. 18 Table 1.2 Poverty by Division .......................................................................................... 19 Table 1.3 Gini coefficient measure of inequality.............................................................. 21 Table 2.1 Sugarcane productivity in selected countries.................................................... 28 Table 4.1 Obstacles to growth as reported by firms here ................................................. 44 Table 4.2 Efficiency of land administration ..................................................................... 50 Table 4.3 Quality of land administration .......................................................................... 50 Table 4.4 Minimum wages and paid leave ....................................................................... 58 Table 4.5 Wage Regulation Orders: Minimum wage and core conditions ....................... 59 Table 5.1 Government housing subsidies, 2016/17 .......................................................... 70 Table 5.2 Financial inclusion by division ......................................................................... 73 Table 6.1 Planning authorities .......................................................................................... 82 Table 6.2 State owned enterprises, 2015 .......................................................................... 87 Table 6.3 Major interventions after Cyclone Winston...................................................... 92 4 Acknowledgements This report was prepared by Mizuho Kida, Imogen Halstead, Deva De Silva, Daniel Street, and Paul Barbour, with contributions from Gayatri Acharya, Donna Andrews, Maude Archambault, Tijen Arin, Natasha Beschorner, Jesse Doyle, Fook Chuan Eng, Simone Esler, Robert Gilfoyle, Dao Harrison, Deni Jordi, Andres Garcia, Wayne Irava, Oleksiy Ivaschenko. Susan Lynette Ivatts, Dhawal Jhamb, Miguel Angel Jorge, Kamleshwar Khelawan, Jonathon Kirkby, Ying Li, Brenna Moore, Isabel Neto, Dina Nicholas, Kofi Nouve, Anthony Obeyesekere, John Perrottet, Jim Reichert, Benjamin Sikuri, Venkatesh Sundararaman, Anuja Utz, John Vivian, Binh Thanh Vu, Anis Wan, and Judy Yang. The team would like to thank the peer reviewers, Gallina Andronova Vincelette (Practice Manager, GMF12), Julio Revilla (Lead Economist, GMF02 and SCD advisory group), and Jennifer Keller (Lead Economist, GMFDR). The team is also grateful for comments received on earlier drafts from colleagues from across the World Bank Group including … The team worked under the guidance of Victoria Kwakwa (Regional Vice President), Michel Kerf (Country Director), Ndiame Diop (Practice Manager, MFM), Salman Zaidi (Practice Manager, POV), Thomas James Jacobs (Country Manager, IFC), and Robert Utz (Program Leader, CMU). 5 Abbreviations ADB Asian Development Bank DFAT Australia Department of Foreign Affairs and Trade EU European Union FDI Foreign Direct Investment FNPF Fiji National Provident Fund FSC Fiji Sugar Corporation GDP Gross Domestic Product HIES Household Income and Expenditure Survey ICT Information and Communications Technology ILO International Labor Organization IMF International Monetary Fund MDG Millennium Development Goal NCD Non-Communicable Diseases NGO Non-Government Organization NPL Nonperforming Loan OECD Organization for Economic Cooperation and Development PPP Purchasing Power Parity SCD Systematic Country Diagnostic SME Small and Medium Sized Enterprises SOE State Owned Enterprise TFP Total Factor Productivity UN United Nations UNDP United Nations Development Program VAT Value Added Tax WDI World Development Indicators WHO World Health Organization WTO World Trade Organization 6 Executive Summary Fiji is a small island nation in the South Pacific Ocean with population of 870,000. It has an area of 18,000 square kilometers spread over 330 islands, of which about 110 are inhabited. Most of the population lives on two large islands, Viti Levu and Vanua Levu. Fiji is also one of the remotest countries in the world—New Zealand is 2,000km away, Australia 3,000, and the United States 5,000. Nevertheless, Fiji has one of the most sophisticated economies among the Pacific Islands. The economy is the second largest in the Pacific after Papua New Guinea, and it is the most industrially advanced, with substantial services and manufacturing sectors. It is a hub for re-exports to the rest of the Pacific. It is also somewhat unusual in the Pacific in that it has developed a major tourism industry, which now attracts over 750,000 tourists a year and contributes about 38 percent of GDP. Fiji has not, however, realized its full economic potential. Since independence from the United Kingdom in 1970, real GDP growth has averaged 2.8 percent a year, or 1.6 percent per capita. One problem has been the frequency of natural disasters. Fiji is on a tropical cyclone belt, and on average one cyclone passes through Fijian waters each year. Last year’s Cyclone Winston caused damage of F$2 billion, or 25 percent of GDP. With climate change, losses of 2.5 percent of GDP are expected every year and losses of 20 percent every 50 years. Another problem has been intermittent political crisis. Since independence in 1970, the country has experienced thee coups d’état, in 1987, 2000, and 2006. There are two main ethnic groups in the country—the indigenous majority, iTaukei, and Indo-Fijians, the descendants of indentured laborers brought from India by the British to work on sugarcane plantations—and political leaders in the past tended to exploit ethnicity to increase political support. The 2014 election was a turning point for Fiji in many respects. First, the current government was elected with an outright majority and strong cross-ethnic support. Second, the election saw a genuine political debate, with citizens offered a choice between different visions for the future (Frankel 2014). Third, the election, which was declared free and fair by international observers, paved the way for Fiji’s re-engagement with development partners and created a better environment for private investment. Fiji was re-admitted to the Commonwealth and Pacific Islands Forums, and Australia and New Zealand lifted sanctions. The World Bank Board approved the new Country Engagement Note in June 2015, together with a US$50 million investment loan for a road project —the first IBRD lending in 23 years. Fiji can build on its relatively strong institutions to deliver faster growth and shared prosperity. It has a reputation for a disciplined and well-educated civil service and pays for about 80 percent of its spending out of tax revenue. While other Pacific Island countries such as Papua New Guinea and the Solomon Islands have relied on capacity supplementation, Fiji has recruited domestic and international experts to the public service 7 on performance contracts and paid for them without donor support. According to traditional global indicators of governance, Fiji ranks relatively high on political stability and control of corruption. It has also shown it can develop and put in place significant reforms quickly when necessary, and adopt appropriate policies for the country. Poverty and shared prosperity Extreme poverty is rare in Fiji. According to the World Bank’s measures of poverty, 2.3 percent of the population live in extreme poverty (less than US$1.40 a day), and 15.1 percent live in poverty (less than US$3.10 a day)—among the lowest rates in the Pacific but similar to those in other upper-middle-income countries. According to national estimates of poverty, 2.5 percent of the population live below the food poverty line (an equivalent of extreme poverty in Fiji), and as much as 34 percent live below the national basic needs poverty line. The latter, thus, roughly corresponds to the bottom 40 percent in Fiji. Economic growth in the past decade has also been inclusive. Between 2002 and 20013, the real per capita consumption of the bottom 40 percent grew faster than that of the average household, and this was true in both rural and urban areas. The success of the bottom 40 percent emerged as a result of particularly high consumption growth in the very bottom of the distribution. Growth, however, has been slow. Since independence, it has averaged 2.8 percent a year, or 1.6 percent a year per capita. Low growth has been largely attributed to low levels of investment, exports, and skill-intensity of jobs. Since independence, structural shifts from agriculture to industries and services have failed to make much impact on aggregate productivity, implying that the reallocation of workers and capital from agriculture to industry and services did not make them more productive. Over the same period, average total factor productivity growth was also zero. Growth has also been volatile, especially in earlier decades, when agriculture was a larger share of GDP. Nevertheless, poverty has declined. How did this happen despite low growth and low productivity gains? First, the declining importance of agriculture in the incomes of the bottom 40 percent reduced the volatility of their incomes. Second, urbanization diversified their sources of income. Third, there was an expansion of low-skilled jobs in traditional services sector such as transport and hospitality. This helped absorb workers from the declining sugar and garment industries. Finally, remittances have risen, further protecting the poor from domestic shocks. The government plans to build on its achievements in poverty reduction and shared prosperity. The national development plan is to double real per capita income by 2035. This is to be achieved by boosting private and public investment, exploiting new export opportunities, and creating higher productivity jobs. Alongside strong growth, reducing inequality and redressing rural-urban imbalances are seen as paramount. Fiji also wants to 8 become a regional hub, for example in aviation, shipping, and information and communication technology (ICT). The World Bank’s report Pacific Possible sets out several ideas for accelerating growth in Fiji by 2040. Areas of opportunities include tourism, migration, fisheries, deep sea mining, and the knowledge economy. For example, in tourism, Fiji can expand the high-end resorts market, and capitalize on aging in Australia and New Zealand by developing long-stay facilities for retirees. In migration, Fiji can seek to establish a caregiver program with Australia and New Zealand that would recruit Fijians to provide residential care to the elderly. The development of ICT has the potential to generate sizeable economic dividends—through productivity gains, skilled employment, and expansion of knowledge-intensive industries—especially given Fiji’s diaspora of skilled entrepreneurs. While Fiji has great potential, success cannot be taken for granted. Achieving the government’s goal of doubling per capita income by 2035 will require annual GDP growth of about 5 percent. This is substantially higher than Fiji has managed in the last four and a half decades. It is also higher than the estimated long term potential growth rate of 3.8 percent. Taking advantage of the opportunities set out above and achieving higher growth will require concerted efforts in a number of interrelated areas. This Strategic Country Diagnostic (SCD) identifies three pathways, as well as cross- cutting issues i. Stronger growth ii. Better access to services by all iii. Building resilience. These three pathways should be supported by cross-cutting efforts to improve governance, that is, to improve policy and the institutional capacity of the public sector to accelerate progress toward the twin goals. Pathways to poverty reduction and shared prosperity The first pathway is generating stronger growth. Growth in the past has been inclusive, but low. To accelerate progress toward the twin goals, therefore, Fiji needs to accelerate growth while making sure that it remains inclusive. At present, there are several obstacles in the pathway to stronger growth. They include low investment, weak exports, and low- productivity jobs. • Low investment. Political instability has been the constraint most often cited by firms. Together with crimes, theft, and disorder, which were important to small firms, this suggest that investors’ concerns about their ability to appropriate the returns to investment have been a major cause of low investment. Firms of all sizes also identified low productivity of labor as a constraint, but the concern was more 9 frequently voiced among larger firms, foreign-owned firms, and exporting firms, suggesting that low productivity of labor is a potentially binding constraint on foreign direct investment (FDI). Frequent changes in policies also increase uncertainty and compliance costs. • Weak exports. The main challenge is to diversify into higher value exports, but Fiji’s exports—dominated by food and raw materials—have shown little sign of diversification. Fiji’s fixed exchange rate and rising remittances do not appear to be significant causes of non-competitiveness. But the costs of trade are higher for Fiji than for leading export nations in East Asia. • Low productivity jobs. On the supply side, despite favorable demographic developments, the emigration of skilled labor has accelerated. On the demand side, the long standing issue of finding and retaining skilled workers may have discouraged the creation of skill-intensive businesses. Labor market mismatch also plays a role. Cumbersome and frequent changes in labor regulations may also be raising compliance costs, especially for small businesses. To clear the pathway, Fiji could revamp its growth model. • Low investment. Traditionally, the government has encouraged private investment either by investing itself or by providing an elaborate system of incentives. Instead, the government could accelerate structural reforms to address a broader competitiveness agenda that would help reduce red tape and time spent dealing with officials. To reduce policy uncertainty and compliance costs, the government could undertake more careful analysis of the costs and benefits of proposed policies before they are introduced and systematic monitoring of their impacts after they are introduced, both in consultation with stakeholders. • Weak exports. Strengthening exports requires changing the traditional model of relying on trade preferences, which have failed to create lasting competitiveness. Instead, the government could continue to invest in connective infrastructure. Although there is not much Fiji can do about its smallness and remoteness, it can reduce the red tape that is making trade across its borders costlier. • Low productivity jobs. To create more productive jobs, these reforms to increase private investment and exports should help. At the same time, the government could encourage hiring by reviewing and simplifying the complex system of labor market regulations and employment incentives, again in consultation with stakeholders. The second pathway is better access to services for all. Although poverty and inequality have fallen, visible signs of inequality are rising, especially in urban areas. With accelerating migration from rural to urban areas, urban poverty has risen, and visible signs of inequalities are increasing, especially with the expansion of squatter settlements. In 10 addition to closing the remaining rural-urban gaps in infrastructure and essential public services, the government thus faces the challenge of improving services for the growing urban population, in education, health, infrastructure, and housing. • Education. Achievements in education have been remarkable. Early education has been successfully introduced throughout the country. Most students complete school, and achievement does not vary much with gender or with the circumstances of the household the child is born into. • Health. Outcomes have been less impressive in health. Although 75 percent of the population report having access to health services, the poor have limited access to higher quality services. Health care costs are also high relative to incomes for all income levels, and aggregate health outcomes are poor. • Infrastructure. Although the majority of the population has access to improved water and sanitation, there remains a large rural-urban divide in access to electricity, reflecting the challenge of difficult terrain, sparsely populated areas, and widely scattered islands. Access to the Internet is also unevenly distributed across the country, with almost no households having access in the Eastern and the Northern Divisions. The high costs of services such as broadband may also be putting the bottom 40 percent at a disadvantage. • Housing. In urban areas, housing shortages are becoming extreme. In addition, as Fiji is susceptible to cyclones and flooding, more resilient homes are a priority. Challenges include high building costs, outdated urban planning, the limited reach of housing finance, public housing agencies with conflicting objectives, and inadequately targeted housing subsidies. Several things can be done to improve services to the poor. • Education. In education, where progress has been relatively good, the government could help maintain quality by targeting spending to rural and remote schools that are costly to run. It could also continue to address the high dropout rates in secondary school by continuing with the “Matua” program, which helps people who have dropped out of school to return. This is important because of the importance of education in employment outcomes. • Health. In health, the government could increase spending on programs that are better targeted at the bottom 40 percent. It could also continue to explore opportunities to involve the private sector, investigating further the international evidence on the cost of hospital service delivery under private and public provision and carefully monitoring the results of its planned project to introduce private management of a hospital, particularly the fiscal costs of subsidizing services for poor patients. 11 • Infrastructure. In infrastructure, there is also greater scope for private sector participation. The government has already pursued this option in ports and electricity. Making further progress would require updating and strengthening the existing framework for public private partnerships, formulating new regulatory standards, reviewing regulated prices and competition policy, and building regulatory capacity in relevant government agencies. • Housing. In housing, the government could help expand access to affordable housing in several ways. For example, to reduce building costs, the housing authorities could pilot smaller, standardized houses that could be produced on a large scale with prefabricated materials through a public-private partnership. To extend mortgage loans to low income households, the government could work with prospective mortgage lenders to better target mortgage-linked subsidies. The third pathway is building resilience. Fiji is one of the countries in the world that is most affected by natural disasters such as cyclones, floods, earthquakes, and tsunamis. As a result, it incurs average annual losses of about 2 percent of GDP. These losses add to environmental pressures, constrain productive investments, and delay improvements in service delivery. And because of a lack of savings, credit, and insurance, the poor suffer most from the shocks. To respond to the shocks, the government must maintain fiscal flexibility and buffers. However, rising expenditure and frequent disasters have eroded fiscal space, while rising costs of subsidies and wage bills have increased expenditure rigidities. In addition, there are contingent risks related to state-owned enterprises (SOEs). Several things can be done to build resilience. • Climate resilience. To strengthen urban resilience, the government could assess the disaster resilience of critical public infrastructure and cost the needed upgrade. It could also update the urban plan to ensure risk sensitive land use, review the building code to ensure it is appropriate for Fiji, and better enforce both of them. To strengthen rural resilience, the government could review the consistency of sectoral policies in the resource incentive sectors, such as agriculture, mining, tourism, and fisheries. It could strengthen enforcement of existing regulations and safeguards to ensure environmental sustainability, and ensure that zoning for coastal areas, river banks, and wetlands is sensitive to risks. Finally, it could explore ways to engage citizens to improve warnings and responses to disasters. • Fiscal resilience. Improving fiscal resilience involves expanding the tax base by consolidating the extensive system of tax exemptions and incentives. On the expenditure side, it involves better targeting of subsidies to create fiscal space and reduce rigidities. Contingent liabilities need to be monitored and reduced. To insure against catastrophic risk, the government could further explore alternative ex ante financing instruments. 12 • Social protection. Not everyone can be protected ex ante. There must also be in place a system to protect the most vulnerable after disasters. After Cyclone Winston, the government experimented with cash transfers based on the existing social protection system. The most effective intervention was allowing people to withdraw some of their savings from the Fiji National Provident Fund to rebuild their houses. However, the Provident Fund only covers the formal sector, so most of the poor were not helped by this measure. Cross-cutting theme: Improve policy and institutional capacity to achieve shared prosperity Fiji has a disciplined and well educated civil service. It has shown it can develop and put in place significant reforms quickly when necessary, and work with development partners to adopt appropriate policies for the country. However, it continues to face a number of cross-cutting governance challenges. Improvements could be made in at least four areas. First, the legal and regulatory framework could be updated to encourage investment and private participation in service delivery. Second, a better coordination across government agencies under more unified strategy would facilitate effective policy implementation and better service delivery. Third, the public administration should involve more delegation and accountability for decisions. Fourth, improving transparency and participation in policy formulation by businesses, citizens, and community service organizations would support more cooperative and inclusive social and economic outcomes. Priorities This report has identified many constraints and many possible policy solutions. But not everything can be done at once: priorities need to be selected. In selecting the priorities, this SCD uses three main criteria. The first is whether a policy is a precondition for making progress in pursuing others. The second is whether a policy is expected to have positive spillovers across different domains (e.g., growth, equity, resilience). The third is whether a policy is feasible, that is, it can be implemented in the medium term given cost, capacity, and political feasibility for action. The resulting priorities are shown in Table E.1. 13 Table E.1. Priorities 1. Improving the business • Reduce policy uncertainty environment • Reduce red tape and the complexity of incentives • Create a stronger legal and regulatory framework for private sector participation • Communicate better and consult more with businesses and civil society 2. Investing in urban • Improve the quality of urban housing and essential public resilience infrastructure • Update the urban plan to ensure risk sensitive land use • Update the building code to ensure it is appropriate for Fiji, and enforced • Exploring ways to expand insurance coverage 3. Safeguarding fiscal • Develop medium term fiscal consolidation plan sustainability • Expand revenue base by reviewing the system of tax exemptions and incentives • Reduce rigidities in expenditure by consolidating fragmented subsidy programs • Better monitor and manage contingent liabilities and catastrophic risks 4. Expanding access to quality • Increase spending on health services targeted to the bottom 40 healthcare percent • Make extra effort to serve rural areas and remote islands • Partner with the private sector to serve the poor • Monitor the cost of hospital service delivery under private and public provision 5. Expanding access to • Continue public investment and encourage private connective infrastructure participation in infrastructure to enhance economy-wide competitiveness • Make extra effort to serve rural areas and remote islands 14 1. Poverty reduction and shared prosperity Poverty Fiji is an upper-middle-income country with a correspondingly low level of extreme poverty. In 2013, the rate of extreme poverty was 2.3 percent according to the international poverty line of US$1.90 a day and 15.1 percent at US$3.10 a day (both in 2011 purchasing power parity, or PPP). Although this is lower than in most countries in the region (Figure 1.1), it is broadly in line with what would be expected given Fiji’s per capita income. Based on the national consumption based poverty lines, the rate of “food” poverty (which is considered extreme poverty in Fiji) was 2.5 percent, and the rate of “basic needs” poverty was 34 percent. 1 The group of basic-needs poor therefore roughly corresponds to the bottom 40 percent in Fiji. The rate of poverty has been declining (Figure 1.2). Between 2002 and 2013, according to the consumption based “basic needs” poverty line, the rate of poverty fell from 40 to 34 percent. Between 2002 and 2008, urban poverty fell from 35 to 26 percent, while rural poverty remained stubbornly high, at about 44 percent. Between 2008 and 2013, however, rural poverty fell faster, from 44 to 38 percent, while urban poverty increased, from 26 to 30 percent. The increase may be partially explained by rural to urban migration, but this is unlikely to be the full story. Even if all those who migrated in 2008–13 were poor, this would not fully account for the increase in the headcount of the urban poor, suggesting that there has been a genuine rise in poverty in urban areas. Figure 1.1 Poverty rates, $3.10 a day Figure 1.2 Recent poverty rates in Fiji poverty line, 2014 or latest 50 40 40 30 30 20 20 10 10 0 0 2002/3 2008/9 2013/14 $3.10/day $1.90/day Consumption basic needs poverty Income basic needs poverty Source: WDI and Fiji Bureau of Statistics. 1 The food poverty line is estimated by calculating the cost of meeting a given caloric energy requirement in Fiji. Basic needs poverty adds to this the cost of obtaining other basic items consumed by households close to the food poverty line. 15 Poverty reduction has been less pronounced than in other countries. The small decrease in the overall poverty rate in Fiji stands out when compared with the experience of faster growing East Asia, where poverty rates have declined at a much faster pace (though from a considerably higher base) (Figure 1.3). Estimates of socio-economic mobility based on synthetic panels also show that, between 2002 and 2008, 6 percent of the population moved out of poverty and 4 percent fell into poverty, while between 2008 and 2013, 4 percent of people moved out of poverty and 7 percent fell into poverty (Figure 1.4).2 Figure 1.3. Changes in poverty (US$3.10/day) Lao PDR Indonesia Philippines Cambodia China 2002 Fiji 2014 Vietnam Malaysia Thailand 0 10 20 30 40 50 60 70 80 Source: WDI. Note. Years are for 2002 or 2014 or nearest years available. Figure 1.4 Economic mobility 2002/03–2008/09 2008/09–2013/14 100 100 80 45 80 53 47 52 54 51 60 0 60 0 8 4 0 0 6 0 7 40 1 40 5 4 20 46 47 42 20 43 46 38 0 0 Rural Urban National Rural Urban National Chronically Poor Escaped Poverty Chronically Poor Escaped Poverty Fell into Poverty Not Poor Fell into Poverty Not Poor Source: World Bank staff estimates based on the World Bank poverty harmonized database and HIES. 2 The analysis of economic mobility in and out of poverty is based on synthetic panel data constructed as in Dang et al. (2014). This technique is applied to three household surveys in the absence of panel data. The technique relies on strong and weak assumptions to generate upper- and lower-bound estimates of economic mobility. The upper-bound estimates are likely to overstate the extent of economic mobility, while the lower- bound estimates always understate mobility, except in the extreme case that prediction errors are perfectly positively autocorrelated. The true estimate of the extent of mobility is generally between the upper-bound and lower-bound estimates. Lower-bound estimates are reported here, which are conservative estimates of economic mobility. 16 There are substantial differences between the poor and the non-poor in their human capital, employment status, and household characteristics. A profile of the poor according to the consumption based basic needs national poverty line is presented in Table 1.1. • Female headed households are less likely to be poor. However, there is heterogeneity among these households, for example, between those with access to remittances from spouses working overseas and those who are headed by single women. • The rate of poverty among households headed by 60–64 year olds is more than twice that of households headed by 25–29 year olds. Household composition is also noticeably different between the poor and non-poor, with the poor living in larger households, with more dependents. • On average, heads of households who are poor have less schooling than the average household. However, perhaps related to improving conditions in rural areas, education of the heads of households is no longer strongly correlated with poverty. • Households are less likely to be poor when the head is employed. But because heads of poor households are rarely unemployed (less than 1 percent in the sample), the activity type of household heads is a better indicator of poverty than their employment status. The poor are more likely to be self-employed or engaged in unpaid or community work. Data on the sector of employment are not yet available from the latest survey. In the 2008 survey, poor households were more likely to be employed in agriculture, but, given the faster decline in poverty in rural areas and the marked increase in poverty in urban areas, the pattern observed in 2008 is likely to have evolved. 17 Table 1.1 Profile of the poor Poverty Headcount Rate Distribution of the Poor Distribution of Population 2002/03 2008/09 2013/14 2002/03 2008/09 2013/14 2002/03 2008/09 2013/14 Gender of the household head Male 38.8 35.5 33.6 87.0 88.7 83.9 89.2 88.0 85.1 Female 47.8 33.0 36.7 13.0 11.3 16.1 10.8 12.0 14.9 Age of the household head 15-19 0.0 0.0 11.2 0.0 0.0 0.1 0.1 0.0 0.2 20-24 27.9 13.9 22.9 1.0 0.3 0.8 1.4 0.7 1.3 25-29 31.5 31.1 20.0 3.9 4.1 2.7 5.0 4.7 4.6 30-34 35.9 27.4 25.7 8.9 6.8 6.3 9.9 8.7 8.4 35-39 35.3 32.4 28.9 13.8 11.5 9.6 15.6 12.4 11.3 40-44 40.2 33.8 32.8 16.9 13.6 11.6 16.8 14.2 12.1 45-49 38.1 37.4 35.0 13.3 17.5 15.0 13.9 16.5 14.6 50-54 41.2 31.1 37.7 12.4 11.1 15.8 12.0 12.6 14.3 55-59 36.7 36.5 34.4 8.0 11.7 11.7 8.7 11.3 11.6 60-64 50.8 40.2 41.0 8.4 7.4 10.8 6.6 6.5 8.9 65+ 52.1 45.6 41.3 13.3 16.0 15.5 10.2 12.4 12.7 Education of the household head None 51.9 45.1 32.6 9.6 4.0 3.4 7.3 3.2 3.5 Primary 50.1 51.8 46.1 22.2 21.7 22.9 17.6 14.8 16.9 Secondary 40.6 37.4 35.9 64.9 69.0 66.8 63.5 64.9 63.4 Postsecondary 11.7 22.8 56.5 3.4 1.4 0.1 11.6 2.2 0.0 Cert. or dip. 10.9 18.1 3.6 5.2 11.6 9.8 Higher 2.2 8.5 0.2 1.6 3.3 6.3 Ethnicity of the household head Fijian 41.7 37.1 38.9 57.3 62.6 69.8 54.7 59.4 61.1 Indian 37.8 33.7 26.7 38.9 33.2 27.5 41.0 34.8 35.0 Other 34.0 25.3 23.4 3.7 4.2 2.7 4.4 5.8 4.0 Household size 1 10.2 2.2 0.5 0.1 0.1 0.0 0.5 0.9 0.9 2 14.1 11.7 7.4 1.3 1.7 1.2 3.8 5.2 5.6 3 19.9 19.6 13.9 4.5 6.1 4.6 9.0 10.9 11.3 4 25.1 23.9 22.1 10.4 11.2 11.4 16.4 16.5 17.6 5 36.7 30.6 32.4 18.7 15.9 16.5 20.3 18.4 17.3 6 41.9 40.4 38.6 17.8 18.2 18.3 16.9 15.9 16.1 7 46.5 42.6 43.5 14.9 14.6 14.6 12.7 12.1 11.5 8+ 63.0 56.2 57.4 32.3 32.2 33.3 20.4 20.1 19.8 Presence of children None <15 yr. 27.0 24.2 22.2 13.9 16.5 17.4 20.4 24.0 26.7 Some <15 yr. 43.1 38.7 38.4 86.1 83.5 82.6 79.6 76.0 73.3 Presence of older household members None 65+ 38.0 32.7 31.8 79.2 74.4 73.9 82.8 80.1 79.2 Some 65+ 48.3 45.4 42.8 20.8 25.6 26.1 17.2 19.9 20.8 Total 39.8 35.2 34.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: World Bank staff estimates based on HIES. 18 There are also differences in the prevalence of poverty across space. Poverty rates are highest in the Northern Division and lowest in the Central Division (Table 1.2). However, with migration from rural areas to urban areas, the spatial disparity in rates of poverty is declining. Consistent with the rising rate of poverty in urban areas, the rate of poverty has significantly increased in the Central Division, which includes the capital, Suva. The poverty map from 2008 highlighted pockets of deep poverty in the Central Division, including in squatter settlements (Figure 1.5). Although the poverty rate overall remains low in the Central Division, more than a third of the poor now live in this area. Meanwhile, the rate of poverty has declined in the less populated Northern and Western Divisions. In particular, the Western Division, which includes the sugar belt, Nadi, and many tourist hotels, experienced the fastest decline in poverty. Importantly, though, a quarter of the poor still live in the Northern Division, where the poverty rate is close to 50 percent. Access to services and connectivity also differs across space (more on this in Chapter 5). Table 1.2 Poverty by Division Poverty Headcount Rate Distribution of the Poor Distribution of Population 2002/03 2008/09 2013/14 2002/03 2008/09 2013/14 2002/03 2008/09 2013/14 Urban 34.5 26.2 29.9 39.1 36.8 44.7 45.1 49.4 50.8 Rural 44.1 44.0 38.3 60.9 63.2 55.3 54.9 50.6 49.2 Geographical division Central 28.9 23.4 28.6 29.8 26.4 35.7 41.0 39.8 42.5 Eastern 35.0 33.0 37.1 5.1 5.1 5.1 5.7 5.4 4.7 Northern 57.6 53.5 49.2 24.8 26.3 23.0 17.1 17.3 15.9 Western 44.4 39.7 33.4 40.4 42.3 36.3 36.1 37.5 36.9 Total 39.8 35.2 34.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: World Bank staff estimates based on HIES. 19 Figure 1.5 Poverty maps Poverty headcount rate at the Tikina3 level Distribution of the poor at the Tikina level as a proportion of total poor Source: World Bank (2011) based on 2007 census and 2008/09 HIES. 3 Fiji is divided administratively into four divisions, which are further subdivided into fourteen provinces. A Tikina is a subunit of a province. 20 Shared prosperity Poverty reduction has been paired with shared prosperity.4 Between 2002 and 2013, the per capita consumption of the bottom 40 percent grew faster, at 1.2 percent a year, than that of the population median, which grew at 0.6 percent a year (Figure 1.6). This pattern of inclusive growth was observed in both inter-survey periods (2002–08 and 2008–13) and in both urban and rural areas. The success of the bottom 40 percent emerged as a result of particularly high consumption growth in the very bottom of the distribution, which also reduced extreme poverty. Inequality has also declined in recent years. The Gini coefficient decreased to 38.3 in 2013, from 42.8 in 2008 and 39.6 in 2002.5 The increase in the Gini coefficient between 2002 and 2008 reflected the rise in inequality between rural and urban areas and within urban areas (Table 1.3). The decrease between 2008 and 2013 reflects reductions in inequality both between rural and urban areas and within urban areas. Table 1.3 Gini coefficient measure of inequality 2002/03 2008/09 2013/14 Total 39.6 42.8 38.3 - Urban 38.9 43.1 38.2 - Rural 36.6 33.7 32.8 Urban / rural breakdown Within-group inequality 18.7 19.9 18.2 Between-group inequality 11.3 16.7 12.4 Overlap 9.7 6.3 7.7 Source: World Bank staff estimates based on HIES. 4 Following the approach used to construct internationally comparable shared prosperity statistics as published in the World Development Indicators (WDI), shared prosperity is analyzed here using mean per- capita consumption measured at 2011 PPP, and so growth is adjusted for inflation. The underlying consumption aggregate makes no adjustment for spatial differences in the cost of living. 5 This Gini coefficient is based on nominal per capita consumption (with no adjustment for spatial differences in prices) and is the inequality statistic reported in the World Bank’s poverty and shared prosperity databases. It is slightly higher than an alternative Gini statistic reported in poverty assessments for Fiji, which are based on a consumption distribution which adjusts for spatial differences in food prices across rural and urban areas (but not for any differences in non-food prices), and reflects an adult-equalized aggregate, which assumes a child aged under 15 has half the consumption needs of an adult. Both inequality measures have traced similar trajectories over time. The poverty assessments report a Gini coefficient of 38.2 in 2002, 41.1 in 2008, and 36.4 in 2013. 21 Figure 1.6 Growth incidence curves, 2002-13 Period 1 Period 2 Total Period (2002/03 - 2008/09) (2008/09 – 2013/14) (2002/03 - 2013/14) 10 10 10 Total 5 5 5 0 0 0 -5 -5 -5 -10 -10 -10 Average annualized growth (percent) 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 10 10 10 Urban 5 5 5 0 0 0 -5 -5 -5 -10 -10 -10 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 10 10 10 Rural 5 5 5 0 0 0 -5 -5 -5 -10 -10 -10 0 20 40 60 80 1000 20 40 60 80 100 0 20 40 60 80 100 Percentile in the distribution Source: World Bank staff estimates based on HIES. Note: The Growth Incidence Curve plots the annualized growth rate of per capita consumption for every percentile of the consumption distribution between two points in time. Period 1 corresponds to the first inter-household survey period (between 2002/3 and 2008/9), and Period 2 corresponds to the second inter-household survey period (between 2008/9 and 20013/14). Total period corresponds to the decade between the 2002/3 and 2013/14 surveys. 22 Drivers of poverty reduction and shared prosperity Contributions of labor income and remittances Employment and labor income have played an important role in poverty reduction, though the nature of this role has changed over time. Estimates of socio-economic mobility based on synthetic panels show that 76 percent of people who moved out of poverty between 2002 and 2008 were living in households headed by wage earners, compared to 66 percent of those who remained poor. Between 2008 and 2013, 48 percent of people who moved out of poverty were living in households headed by someone self- employed, compared to 42 percent among those who remained poor.6 Remittances have made an important contribution to welfare across all quintiles. They were received by 24 percent of people in the bottom quintile and 31 percent of people in top quintile (Figure 1.7). In all quintiles, more people receive overseas remittances than domestic remittances, but the proportion receiving overseas remittances increases with income. About 28 percent of people in the top quintile receive overseas remittances, while only 18 percent of those in the bottom quintile do. In contrast, 5 percent of the top quintile receive domestic remittances, compared to 9 percent of the bottom quintile. Figure 1.7 Remittance receipts by consumption quintile: percentage of population International remittances Domestic remittances 30 30 Percentage of population Percentage of population 25 25 20 20 15 15 (%) (%) 10 10 5 5 0 0 1 2 3 4 5 1 2 3 4 5 (Lowest) (Lowest) Consumption quintile Consumption quintile 2002/03 2008/09 2013/14 2002/03 2008/09 2013/14 Source: World Bank staff estimates, Fiji Bureau of Statistics. Contribution of growth and inequality Economic growth was the main driver of poverty reduction. Specifically, growth in per capita consumption accounted for 62 percent of the poverty reduction between 2002 and 6 Lower-bound estimates using synthetic panels constructed as in Dang et al. (2011). The patterns were similar for the upper-bound estimates. 23 2013 (Figure 1.8).7 The remaining 38 percent was explained by a decline in inequality. The inter-survey periods, however, tell a more complex story. Before 2008, growth was strongly poverty reducing, but the accompanying increases in inequality reduced growth’s poverty reducing impact. After 2008, the growth was negative, which increased poverty; however, inequality declined as welfare converged, and the reduction in inequality explained much of the poverty reduction that took place in this period. Figure 1.8 Changes in poverty rate, 2002–13 6.0 4.0 2.0 3.9 2.5 0.0 -2.0 -3.7 -5.4 -4.0 -7.0 -2.3 -6.0 -8.0 2002-08 2008-13 2002-13 growth redistribution Source: World Bank staff estimates based on the World Bank poverty harmonized database and HIES. 7 The estimates are based on a Datt-Ravallion decomposition of changes in poverty between 2002 and 2013, and in the sub-periods 2002-08 and 2008-13. Total change in poverty in each period is decomposed into two key components: contribution of higher mean consumption, holding constant the distribution of consumption (“growth”), and contribution of shifts in distribution of consumption, holding constant the level of mean consumption (“redistribution”). See Datt and Ravallion (1992) for details. 24 2. Determinants of economic growth Patterns of growth Since independence in 1970, Fiji has undergone difficult structural changes, both economic and political (Box 2.1). As a result, its growth has been low and volatile (Figure 2.1). Between 1970 and 2015, real GDP grew at an average of 2.8 percent a year. Per capita income grew at an average of 1.3 percent a year. When viewed against this background, achieving a reduction in extreme poverty has been a creditable performance. Figure 2.1 Economic growth, 1970-2015 15 Stabilization Stagnation Growth Post-independence boom and bust interrupted by and the acceleration 10 the second coups global financial crisis 5 0 -5 1987 2000 2006 coups coups coups -10 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Source: ADB (2014). Box 2.1 Fiji's growth history Fiji’s post-independence growth history can be divided into four periods. Boom and bust (1970‒87). The first period was characterized by a post-independence boom followed by a secular growth decline. The annual growth rate averaged 9 percent in the first 3 years, helped by favorable sugar prices, growing tourism, and large public investment. Like many developing countries after independence, Fiji pursued import substitution. But growth became more volatile and began to decline, to 2 percent in the late 1970s and 1.5 percent in the early 1980s, reflecting, inter alia, bad weather, falling sugar prices, and an overvalued exchange rate. Weaker growth meant lower government revenue, but the government borrowed rather than reigning in public expenditure. In 1986, faced with increasing levels of foreign debt, the government began to shift its strategy from import substitution to export promotion. The first coup d’état in 1987 caused large output losses ( -7 percent) and large declines in investment (-25 percent). Inflation shot up, investment fell further, and there was an exodus of skilled Indo-Fijian workers, of 5,000–6,000 a year (Duncan and Nakagawa 2014). Stabilization interrupted by the second coup (1988–2000). The economy rebounded strongly in 1988 –89, and the average growth rate recovered to about 3 percent in the following decade. By the mid-1990s, inflation eased to below 1 percent. But investment failed to recover because of lack of confidence (IMF 1995). Growth was interrupted by the second coup d’état i n 2000. During the following 18 months, around 25 1,000 people left each month—most of them professionals and business owners. The government used generous economic incentives, especially tax related incentives, to encourage private investment, but failed to gain traction. Stagnation (2001‒10). The third period was characterized by economic stagnation. The emigration, loss of confidence, and strained international relations were exacerbated by the third coup in 2006. Tourism was hit first, as visitors from Australia and New Zealand (more than half of the arrivals) declined as their respective governments issued travel warnings. Several large developments of tourist resorts were abandoned or scaled back. At the same time, the garment industry was hit by the expiration of preferential trade agreements. Many foreign owned garment factories were closed and employment dropped. The global financial crisis that began in 2008 affected Fiji mainly through reduced exports and remittances. Output contracted in 2009 by 1.4 percent. Tourism avoided the worst, however, thanks to recovering arrivals from Australia and New Zealand. Declining oil prices following the crisis also softened the blow to the real sector. Growth acceleration (2011‒present). The most recent period has seen an acceleration in growth interrupted by bad weather. Growth recovered from the global financial crisis, reaching about 3 percent in 2010‒11 (the highest rate since the 2006 coup). Cyclone Evan hit in 2012, one of the worst in Fiji’s history, slowing growth by damaging sugar output (-7.1 percent) and tourist arrivals (-2.5 percent). Growth recovered to about 5 percent in 2013‒14, supported by tax cuts, low interest rates, and increased government spending. Following the democratic election in September 2014, growth was 4.7 percent in 2015, spurred by rising confidence, stronger visitor arrivals, and private and public investment. In 2016, growth was again disrupted, by Cyclone Winston, the biggest in Fiji’s history, but is expected to recover to 3–4 percent in the next few years. Drivers of growth Decomposition by sector Since independence, services have been the main drivers of growth. Between 1970 and 2015, real value added in services grew on average by 3.6 percent per year, compared to 2.5 percent in industry and 1.3 percent in agriculture. As a result, services ’ share in GDP expanded from 52 to 71 percent, while agriculture’s declined from 25 to 11 percent and industry’s from 22 to 19 percent. This means that over the last 45 years, roughly 65 percent of total output growth came from services ( 26 Figure 2.2 Decomposition of growth by sector ). 27 Figure 2.2 Decomposition of growth by sector Percentage points Agriculture 0.2 Industry 0.5 Services etc. 2.1 0 0.5 1 1.5 2 2.5 Source: WDI and World Bank staff estimates. Note: Sectoral growth rates are weighted by sectoral shares. Percentage points add up to total GDP growth rate for the period. Within services, tourism-related industries expanded the fastest. Specifically, wholesale and retail trade, restaurants and accommodation, and transport and communications recorded the highest growth. The government sector also expanded strongly, with the increasing developmental role of the government after independence. In contrast, the share of agriculture in GDP has declined steadily, largely reflecting the decline of sugar. The agriculture sector largely consists of sugar, coconut, tobacco, and some emerging export crops such as kava, ginger, spices, and tropical fruits. There is also some livestock and subsistence farming. The decline in agriculture started soon after independence (World Bank 1977). At independence, sugar accounted for 13.5 percent of GDP, 27 percent of employment, and 34 percent of exports. Today, it accounts for only 1.2 percent of GDP, 4 percent of employment, and 5 percent of exports. The sugar sector is heavily dependent on preferential access to the European Union (EU). And the erosion of the preference has driven the sector’s decline. Under the Sugar Protocol (1975–2008), the EU purchased a specified quantity of sugar from Fiji each year at guaranteed prices that were several times higher than world prices. In 1996, the World Trade Organization (WTO) ruled that the EU sugar policy did not comply with international trade rules. Since then, the Sugar Protocol has therefore been gradually phased out and the guaranteed prices have been declining. Today, under the bilateral trade agreement with the EU, Fiji continues to enjoy quota-free, duty-free access to the EU market and benefits from the premium of EU prices over world prices. However, in October 2017, the preference is finally coming to an end. The EU will eliminate production quotas within the EU, which will bring the prices in the EU closer to the world prices. Fiji’s land tenure system (Box 2.2) is often blamed for the decline of sugar, but several other forces are also responsible. In fact, declining investment in sugar largely reflects a rational response by farmers to the elimination of preferences. 28 • Low productivity by growers, the high cost of cane farming, and mill inefficiencies combine to make the industry very cost inefficient (Oxfam 2005). “A significant level of inefficiency exists at the farm level” (Reddy and Yanagida 2000; see also Narayan 2004; Haszler et al. 2010). The majority of commercial sugarcane farmers in Fiji are operating at 20-30 percent below the efficiency frontier (Haszler et al. 2010). The productivity of cane farmers in Fiji is also well below that of some of the other sugar producers (Table 2.1). Table 2.1 Sugarcane productivity in selected countries Productivity Productivity Tons of (Tons/acre) Differential Cane/Ton of (Fiji – Country) Sugar Fiji (1997-2001) 19.6 9.6 Mauritius (1997-98) 31.9 12.3 9.2 Queensland (1999-2001) 35.4 15.8 13.9 Hawaii (1997-99) 43.8 24.2 7.9 Louisiana State, USA (1999) 31.3 11.7 11 Source: Reddy (2003). • The costs of cane harvesting are high. Cane is cut and loaded by hand onto portable carts, which are then transported to collection points on a rail network, or loaded onto trucks for cartage by road to mills. Often, labor is hired for this purpose, but during the cutting season cane cutters are scarce and expensive according to farmers, even though the work is arduous and cutters’ annual income is generally well below any definition of the family poverty line (Government of Fiji 2006, quoted in Mahadevan 2008). • Fiji’s sugar mills are also inefficient. All four mills, which are owned by the Fiji Sugar Corporation (FSC), have been operating at loss in most years. The mills are old, overstaffed, and expensive to run and maintain. They have frequent break downs, creating problems for farmers: “there is little incentive for farmers to improve their efficiency if factors not within their control are going to have an adverse impact on their profit” (Mahadevan 2008). Although the government has spent millions of dollars restructuring the sugar industry and has guaranteed loans raised by the FSC to upgrade transport and milling infrastructure, there has been little improvement in the efficiency of the mills. This year, one of the mills is being closed, because the FSC could not afford to “throw good money after bad.”8 • With migration from rural to urban areas, the younger generation is moving away from the sugar sector and looking for opportunities in the cities with more pay or less hard labor. As a result, “[m]ost of Fiji's 12,000 sugar cane farmers are over 50 8 FSC chairman, Mr. Vishnu Mohan, quoted in the Fiji Sun, at http://fijisun.com.fj/2017/03/28/penang-mill-will-be-closed-for-good/. 29 years old.” “Definitely the sugar industry is in trouble because when this generation’s time goes away in about five or ten years’ time and they become old, then what will happen is that there will be nobody to work in the farm. There is lots of land left over here which is being unused.”9 • Although it is sometimes said that 200,000 people are dependent on the sugar industry, that is probably no longer the case. A better estimate is 90,000 or fewer.10 Box 2.2 Land tenure in Fiji Land in Fiji is mostly held as customary land. About 87 percent of country’s total area is iTaukei land— owned or reserved for indigenous Fijians. Most of this land is collectively owned by the traditional clans and cannot be sold. However, it can be leased for 30 years for agricultural use and up to 99 years for residential, commercial, and industrial uses. The iTaukei Land Trust Board currently manages 35,586 leases, most of them for agricultural and residential use. Some have argued that customary tenure harms growth and that it should be replaced by individual freehold tenure. However, given Pacific people’s attachment to customary ownership, others suggest that the best way to create secure land tenure is to create a system of long-term leases within the customary system (Duncan and Nakagawa 2012). Fiji’s long-term leases have supported a large sugar industry as well as a large and growing tourism industry. Unfortunately, the system came into disrepute in the wake of the 2000 coup because of the politicization of the renewal of land leases to sugarcane farmers, who are mostly Indo-Fijians. Before the May 1999 election, 90 percent of the expiring 30-year leases (negotiated under the 1969 Agricultural Landlord and Tenant Act) had been renewed. After the election, however, the renewal of leases became “political football” between the new government, which was dominated by the Indo-Fijian labor party and which threatened land reform and landowners, who retaliated by not renewing leases (McCarthy 2007). Is customary land tenure a binding constraint on growth and shared prosperity in Fiji? The dispute and the ensuing decline in the sugar industry have often been blamed on the lease system. Declining investment by leasehold farmers could, however, reflect a rational response to the looming expiration of the EU sugar subsidy, low global sugar prices, and opportunities in other crops. Moreover, the main constraints on investment may not be the nature of land tenure per se, but political instability, poorly enforced contracts, and inefficient dispute resolution. For example, a study has shown that, other things being equal, economic incentives for investment under leasehold and freehold tenure are equivalent if contracts are well enforced. “If the land remains productive, the lease will be renewed indefinitely and therefore its economic value will be the same as it would be under individualized freehold. If the rights under leasehold and freehold are equally enforceable under law there is essentially very little, if any, 9 The General Secretary of Fiji Cane Growers Association, quoted at http://www.radionz.co.nz/international/pacific-news/277011/fiji-sugar-industry-in-trouble-without-young- people. 10 According to the Fiji Bureau of Statistics, there were 14,804 registered farmers in 2014, of whom 12,632 were active. In 2013, there were also 3,725 people working in sugar processing. In addition, there are up to 2,000 employees in the FSC at peak season (FSC Annual Report 2015). In total, this makes 18,357 workers directly employed in the industry. Other people are, of course, dependent on the incomes of these workers. The dependency ratio in 2015 (the ratio of the population aged 0-14 or 65 or older to the population aged 15- 64) was 0.5285. Using this ratio gives an estimate of the population dependent on the sugar industry of 28,059 (18,357 × [1+0.5285]). Alternatively, one can use the typical household size in Fiji. According to the 2014 census, the median household size in Fiji was 4.9. Counting all other family members as dependents gives an estimate of the population dependent on the sugar industry of 89,949 (18,357 × 4.9). 30 difference in value between the two” (Lightfoot 2005, p. 24). Even if the lease is nonrenewable, if it is sufficiently long, the economic incentives for investment are much the same. For example, at a 10 percent discount rate, a 30-year lease is worth about 94 percent freehold value (Lightfoot 2005, p. 24). Customary land rights could play an important role in shared prosperity, but evidence is missing in Fiji. In some countries, customary land tenure appears to help reduce poverty and increase shared prosperity (e.g., by providing secure livelihood or social protection to rural communities), but in others, it has a negative impact (e.g., by exacerbating gender disadvantages in asset holdings, access to finance, or intra-household bargaining power). There appears to be no empirical research on the impact of the land tenure system on poverty and shared prosperity in Fiji. Manufacturing began to shrink in the early 2000s with the erosion of trade preferences. Its decline largely reflects the decline of sugar processing and the loss of preferences by the garment sector. The manufacturing sector in Fiji was built during the years of import substitution in the 1970s and grew rapidly in the late 1980s and the 1990s with development of the export-oriented garment industry. Growth of the garment industry was, in turn, driven by preferential trade agreements with Australia and New Zealand (the South Pacific Regional Trade and Economic Cooperation Agreement) and with the United States (the Multi Fiber Agreement). The government encouraged FDI in the garment industry with tax free factories. Since 2000, the garment industry has rapidly declined as trade preferences and tax concessions have been phased out. With continuing political instability, and ensuing declines in investment and outflows of entrepreneurs, the industry continued to struggle. The loss of competitiveness against cheaper and more productive manufacturing workers in Asia also hurt the industry. Services were the main drivers of job creation, absorbing labor from agriculture and industry. Between 1990 and 2011 (years for which there are comparable data on employment by sector), services accounted for 79 percent of growth in employment. Reliance on services for job creation has only intensified in recent years. During the 1990s, services accounted for 72 percent of growth in employment, and in the new millennium, 96 percent. The creation of jobs in services has contributed to shared prosperity. Detailed data are not available, but it appears that expansion of low skilled jobs in traditional services (e.g., transport, wholesale, retail, and restaurants) played an important role, helping absorb many of the low skilled workers from agriculture and the garment industry (Figure 2.3). By contrast, the expansion of skilled jobs in modern services (e.g., information technology, finance, insurance, and real estate) played a less important role. 31 Figure 2.3 Evolution of employment by subsector, 1990-2011 100% 80% Traditional services 60% Public administration 40% Modern services 20% Industry 0% Agriculture 1990 1996 1998 2000 2004 2006 2009 2011 Source: Fiji Bureau of Statistics. Decomposition by components of aggregate demand On the expenditure side, consumption was the main driver of growth. Public and private consumption together explained over 80 percent of total output growth between 1970 and 2015. Private consumption explained 74 percent while public consumption explained 17 percent (Figure 2.4). Private consumption grew strongly in the post- independence boom, helped by increasing employment from rapidly expanding tourism and construction sectors as well as rising public sector employment and wages. It continued to grow strongly in the late 1980s and the 1990s, likely reflecting growing employment in the garment sector. Consumption remained resilient during the period of slow growth beset by the coups, food and commodity price shocks, and the global financial crisis, because it was supported by rising remittances from the diaspora. Growing remittances, by helping the poor diversify their sources of income, were likely key to poverty reduction during this period. 32 Figure 2.4 Contribution to growth by demand, 1970-2016 Private Consumption 1.9 Government Consumption 0.4 Investment 0.4 Net Exports -0.2 -0.5 0.0 0.5 1.0 1.5 2.0 By sub-period (percentage point) Period 1 Period 2 Period 3 Period 4 1970-1987 1988-2000 1970-2015 2001-2010 2011-2015 Private Consumption 2.4 3.0 1.2 3.0 Government Consumption 0.8 0.5 0.1 0.9 Investment 0.5 1.2 0.7 2.0 Net Exports 0.2 -0.9 -0.4 -0.4 Annual GDP Growth 3.6 2.9 1.6 4.3 Note: Expenditure growth rates are weighted by their average shares in total expenditure evaluated at the mid-point between beginning and end of the period. Source: WDI and World Bank staff estimates. Investment contributed less to growth, accounting for 17 percent of the total between 1970 and 2015 (Figure 2.4). Its contribution was also more erratic across sub-periods, reflecting turbulent economic and political history. After the first coup in 1986, it rebounded strongly as the garment industry took advantage of tax incentives and preferential access to Australia, New Zealand, and the United States. With the expiration of preferences and tax concessions, precipitated by another coup in 2000, FDI fell to nearly zero. In the past five years, however, investment has contributed strongly to growth because of renewed political stability and strong public investment in infrastructure. Fiji needs more investment to accelerate growth. In the last five years, its investment rate was 18 percent, lower than its peers (Figure 2.5). The Commission of Growth and Development (2008) concluded that an investment rate of 25 percent of GDP or higher was common among the high growth countries of post-World War II. Empirical literature also tends to confirm that investment has determined how fast economies can grow. 33 Figure 2.5 Investment to GDP ratio Average 2011-15 30 20 28.5 25.0 24.7 23.3 10 18.4 0 High growth Global Middle Small states Fiji economies benchmark* income Source: WDI; *Growth Commission (2008). Net exports failed to contribute positively to growth. From 1970 to 2015, they reduced growth by 0.8 percentage points on average. The negative contribution persisted in each sub-period, suggesting a structural issue rather than a cyclical one (Figure 2.4). Fiji’s trade deficit is large and has widened over time. The weak external performance reflects a growing deficit in merchandize trade, likely driven by erosion of traditional exports such as sugar and garments, as well as rising import demand from growing consumption and investment. Trade in services fared better, because of tourism and Fiji’s position as a transport hub (Box 3.1). Visitor arrivals have grown substantially in recent years—mostly from Australia and New Zealand but increasingly from China and India—helped by more regular air services. Growing surpluses in trade in services, however, has not been enough to offset the widening gap between exports and imports of goods (Figure 2.6). Figure 2.6 Trade balances, 1991-2014 Percent of GDP 30 20 10 0 -10 -20 Goods balance -30 Service balance -40 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Source: WDI. 34 Decomposition by factors of production Growth accounting finds that there has been little productivity growth. The structural shifts in the economy are often associated with increases in aggregate productivity, as both capital and labor move from lower to higher productivity sectors attracted by higher returns. However, the result from growth accounting suggests that this was not the case for Fiji. In the last four and a half decades, average growth in total factor productivity (TFP) was zero. Instead, economic growth has relied heavily on the accumulation of factors of production (Figure 2.7). Figure 2.7 Growth accounting, 1970-2015 Contribution of different factors of production to economic growth, percentage points 8 1.1 6 0.9 0.8 4.0 0.7 4 0.2 2.3 2.7 1.8 2.4 0.8 2 1.5 1.7 2.4 2.4 1.0 1.4 0 0.0 -0.4 -0.9 -2 1970- 1970- 1988- 2001- 2011- 2015 1987 2000 2010 2015 Total Factor Productivity Labor Capital Stock Human Capital Source: WDI and World Bank staff estimates. Looking at growth accounting by sub-period, the story that emerges is consistent with the foregoing discussion. During 1970–87, TFP growth was at its lowest (Figure 2.7). Following the dismantling of the import substitution policies, it turned positive during 1988–2000, but it fell into negative territory again during 2001–10, perhaps reflecting increasing political instability and the shocks from the global financial crisis. Notably, this period also saw the lowest contribution of human capital to growth, possibly as a result of the emigration of skilled workers and entrepreneurs. In the most recent period (2011–15), however, TFP growth has turned positive and has been the main driver of growth. Financing of growth Fiji has traditionally run a current account deficit, which widened in the late 1990s because of a rapid decline in gross domestic saving (Figure 2.8). The decline in saving was, in turn, driven by a decline in private saving, possibly reflecting the growing political instability and the emigration of high income earners during the period (Figure 2.9). 35 Figure 2.8 Saving and investment Figure 2.9 Saving by sector Percent of GDP Percent of GDP 40 40 Gross domestic savings 30 Private 30 Public 20 20 10 Saving 10 0 -10 0 Current account balance -20 -10 1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015 Source: WDI and World Bank staff calculations. The widening gap between saving and investment has been largely financed by non- debt-creating flows (Figure 2.10). FDI inflows reached their highest in the last decade, averaging 9 percent of GDP and financing 53 percent of total domestic investment. Remittance inflows also began to rise strongly since 2000, averaging 5 percent of GDP in the last decade, and financing 7 percent of domestic private consumption. Figure 2.10 Financing of growth US$ million 500 400 300 200 100 0 -100 FDI (net) Remittance (gross) Portfolio flow (net) office flow (net) 2000 2005 2010 2015 Source: IMF Balance of Payment Database. Portfolio flows, consisting of equity, bond, and cross-border bank lending, have not been an important source of financing. Cross-border bank lending is minimal. Bond flows are also small, as local issuance is largely limited to the government and there is virtually no secondary market trading of these bonds. Equity flows are also small given the small size of Fiji’s South Pacific Stock Exchange with little or no trading. 36 Similarly, official flows have played a limited role in financing growth. Between 2006 and 2015, Fiji received about 1 percent of GDP in aid each year, compared to an average of 3 percent for upper-middle-income countries, 9 percent for small states, and 20 percent for Pacific Island countries (Figure 2.11). External borrowing also remains relatively low. External debt at the end of 2016 was 14 percent of GDP, consisting of 46 percent bilateral, 33 percent commercial, and 21 percent multilateral debt. Figure 2.11 Annual aid flows, average 2006–15 20 20 16 percent of GDP 15 percent of GNI 10 10 9 9 8 5 4 4 3 2 1 1 0 Pacific East Asia Small Middle Upper Fiji and States Income Middle Pacific Income Source: WDI. Drivers of volatility One of the key features of Fiji’s growth has been its volatility. The swings in annual growth ranged from -7 to 13 percent (Figure 2.1). Furthermore, the volatility is higher than in peer countries, even the small states and Pacific Island countries. Volatility is harmful for growth as well as poverty reduction. It not only reduces output in the short run but could also reduce potential growth in the long run (Aizenman and Pinto 2004). Frequent shocks can also weaken shared prosperity as the poor often suffer disproportionately during downturns, for example, because they have less diversified assets, less secure jobs, or less access to insurance. The decline in volatility, which may have resulted from the decline of agriculture and the rise of service, may therefore help explain Fiji’s success in reducing poverty. Empirical analyses of the determinants of growth volatility in Fiji support this hypothesis. Moring and Williams (2000) investigate output growth in Fiji during 1975– 98, using the error correction framework. They found that in the short run more than half of the volatility in output was explained by the volatility in sugarcane production (a proxy for more general agricultural shocks). The outsized influence of agriculture is surprising— after all, it accounted for less than 15 percent of GDP in the mid-1970s even if sugar 37 manufacturing is included. The magnitude of the volatility of sugar production, possible flow-on effects to the rest of the economy, and the fact that sugarcane production is a proxy for a broader weather-related shocks likely explain the result. Extending the sample period to 2015 shows that the influence of agriculture on growth has been roughly halved. Instead, annual fluctuations in visitor arrivals (not included in the original study by Moring and Williams 2000) emerges as the main driver of volatility. Because the volatility of visitor arrivals is less extreme, the shift from agriculture to tourism has reduced aggregate volatility. The same empirical approach shows that shocks in major trading partners have an immediate but smaller impact. On average, a 1 percentage point increase in the real growth rate of trading partners increases Fiji’s growth rate in the short run by around 1/10th of a percentage point. Although the external sector plays an important role, terms of trade and exchange rate shocks do not appear to explain volatility. Because Fiji is an open economy and its nominal exchange rate is pegged, large terms of trade stocks are expected to cause pronounced swings in domestic income. The terms of trade have moved quite sharply at times in response to shifts in the world prices of Fiji’s major commodity exports (Figure 2.12). However, according to the analyses, they have not had a significant influence on volatility. Similarly, the real effective exchange rate has had a limited influence. Indeed, it has been very stable in Fiji, with the exception of occasional sharp devaluations (Figure 2.13). The analyses suggest that the primary channels of influence of international shocks have been output linkages through trade and investment—agriculture, tourism, and more generally through trading partner growth—rather than prices. The influence of domestic policies on volatility appears to have been limited. The effect of monetary policy on output may have been weak because the pegged exchange rate constrains the Reserve Bank of Fiji’s capacity to conduct independent monetary policy. There is some evidence of a negative relationship between real short-term interest rates and economic growth (as expected), but it is not the dominant influence on volatility. Similarly, fiscal policy does not appear to have a strong influence on volatility despite sizeable swings in the government’s fiscal balance. However, political uncertainty in the 1980s and 2000s is likely to have contributed to volatility, although its impact is harder to quantify. 38 Figure 2.12 Changes in terms of trade Figure 2.13 Real effective exchange rate Percent Index (2010=100) 30 180 20 160 10 140 0 120 -10 100 -20 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 Source: WDI. While the analyses focused on explaining volatility, they also show the importance of Fiji’s trading partners’ growth for Fiji’s long run growth. The error correction model used in the analyses is useful for estimating both short-run and long-run influences on output growth in Fiji. The results show that there was a stable relationship between Fiji’s output and that of trading partners, with an estimated long-run elasticity of 0.44, meaning that a 1 percentage point increase in trading partners’ income has led to a 0.44 percentage point increase in Fiji’s output. Adjustment was relatively quick. Thus, the result points to the possible gains to be made from building closer trade and investment linkages with the faster-growing economies of Asia. How did Fiji reduce poverty despite low growth and low productivity gains? The analyses in Chapter 1 showed Fiji has managed to reduce poverty and that growth has been inclusive. In both urban and rural areas, the per capita consumption of the bottom 40 percent grew faster than the average, and especially strong growth at the very bottom of the distribution helped reduce extreme poverty. How did Fiji manage this? The analyses in this chapter provides several clues. First, the declining importance of agriculture in the incomes of the bottom 40 percent reduced the volatility of their incomes. Second, urbanization diversified their sources of income. Both of these factors may have helped them maintain growth in their consumption. Third, there was an expansion of low-skilled jobs in traditional services sector such as transport and hospitality. This helped absorb workers from the declining sugar and garment industries. Finally, remittances have risen, further protecting the poor from domestic shocks. 39 3. Building on opportunities and strengths Government’s vision The government aims to double Fiji’s per capita income by 2035. It is in the process of finalizing a National Development Plan for 201635 that sets out a vision for growth and shared prosperity. The plan has drawn on consultations with people around the country, including civil society and business leaders, to capture their collective aspirations for what a better future looks like. It establishes a set of 20-year targets for achieving the vision. The goal of doubling real per capita income by 2035 requires annual GDP growth of about 5 percent or about 3.5 percent per capita. Other targets include reducing poverty and inequality, redressing rural-urban gaps, lowering public debt, and creating more jobs. The government has two strategies for meeting these targets. The first is to establish the fundamentals of sustainable economic growth. This encompasses the development of sustainable cities, reinvigorating the rural economy, protecting the environment, expanding trade, and supporting the expansion of tourism and ICT. Recognizing that long-run growth will require more than business as usual, it envisions transforming Fiji into a regional hub, leveraging its location, connectivity, and human capital. The second strategy is to focus on essential services. This encompasses improving access to education, health, sanitation, and housing. Efforts under both strategies are to be supported by improvements in the policy and institutional environment. This encompasses implementing appropriate macroeconomic policies and accelerating structural reforms. Box 3.1 Fiji’s ambition to be a regional hub Fiji’s vision to become the regional hub is part of its long-run strategy. A clear picture of what a successful hub might look like and in which specific activities is yet to fully emerge, but it could involve the following elements. Finance. Prime Minister Bainimarama has announced a plan to develop Fiji into a Pacific financial center. The country's fledgling South Pacific Stock Exchange has just 18 listed companies, including local telecommunications and food companies. To raise the profile of the exchange, a 10 percent corporate tax rate for companies that list there has been promised. Aviation. Fiji’s central location in the South Pacific makes it an ideal hub for flights to other Pacific islands. Fiji Airways, the national airline, serves 48 destinations in 13 countries around the Pacific. Destinations include Australia, Hong Kong SAR China, New Zealand, Singapore, and the United States, as well as Kiribati, Samoa, Solomon Islands, Tonga, Tuvalu, and Vanuatu. Fiji’s main international airport, in Nadi, is being upgraded. Tourism. Because of the better connectivity, Fiji receives almost 40 percent of all visitors to the Pacific, positioning it as the ideal tourism hub of the region. To increase the economic benefit of tourism, the government is looking to achieve higher yield per visitor, increase the spread of tourism across the country, as well as maintaining visitor arrivals from the traditional markets of Australia and New Zealand. 40 Shipping. Fiji is well placed to serve as a transshipment hub for smaller Pacific countries. At present, however Fiji’s ports are struggling to cope with local demand. Fiji’s Ship Owners and Agency Association chairman, Jeffrey Lin, has said Fiji was “unattractive” as a hub because of inefficient cargo handling, cargo disruptions, and increasing costs. “When I talk about expanding the port facility, I am speaking about how disruption to cargo loading and unloading adds to costs,” Lin said. “Cruise liners have priority of berth and this means that if a cargo vessel is being unloaded or loaded, it has to vacate the berth for the cruise liner.” Mr. Lin says that this puts Fiji at a competitive disadvantage relative to countries such as Samoa and Papua New Guinea (Chaundary 2015). International organizations. Prime Minister Bainimarama has expressed Fiji’s desire to take a leading role in multilateral forums to advance the interests of the Pacific nations (Delaibatiki 2015). The country already hosts the Pacific Islands Forum Secretariat, the Pacific Islands Development Forum Secretariat, the Pacific Community, and the regional offices of the EU, the World Bank, the United Nations, and the Asian Development Bank (ADB). This year, it will hold the President of COP 23. In 2020, it will host the annual meeting of the ADB. Re-exports. Permanent Secretary for Industry and Trade Shaheen Ali has urged the motor industry to capitalize on Fiji as a hub rather than just opening a showroom in Fiji: “We need to capitalize on our many advantages as a country, like having better skills, better infrastructure and better connectivity in the region . . . Some Fijian companies are exporting to our island neighbors. What is limiting Asco Motors from doing the same, and becoming the supplier of Toyota Vehicles, not just in Fiji, but in the region?” (Ministry of Industry, Trade and Tourism 2013). ICT. Fiji is already a hub for international submarine cables, with three landing in Fiji: one from Tonga, one from Vanuatu, and the Southern Cross Cable, which connects Sydney to Honolulu with a spur to Suva. Fiji liberalized its telecoms market in 2008, which resulted in substantial increases in coverage and access and the lowest wholesale Internet prices in the region. Fiji has the region’s only 4G mobile network. Basic mobile coverage is nearly universal, at 89.5 percent. Fiji has an emerging ICT-enabled services and business process outsourcing industry. It won the 2014 European Outsourcing Association’s Offshoring Destination of the Year Award, which acknowledges operators that have most successfully serviced European outsourcing markets. Education and training. Fiji is already a hub for tertiary education in the Pacific Islands. It hosts the main campus of the region’s only transnational institute of higher learning, the University of the South Pacific and the Fiji National University, which includes the only medical school in the Pacific. The University of the South Pacific is owned by the governments of 12 Pacific Island countries and has campuses in each of them. The Pacific Theological College and the Fiji Institute of Technology excel in training leaders and professionals of the region. Pacific Possible Pacific Possible (World Bank 2017) charts five opportunities that could enable small Pacific Island countries to accelerate shared prosperity by 2040. These opportunities are identified by taking into account the countries’ common physical and geographical challenges. The five opportunities are tourism, fisheries, labor mobility, deep sea mining, and the knowledge economy. Tourism. The most important regional opportunities for tourism are the Chinese visitor market, regional cruise ships, the high-end resorts market, and capitalizing on population 41 aging in origin markets by developing long-stay opportunities for retirees. Given Fiji’s small size and limited capacity to serve mass markets, boosting spending per tourist is crucial. By improving health care to international standards, Fiji would be better placed to take advantage of the long-stay retiree market. By exploiting these opportunities, Fiji could gain 500,000 additional international visitors each year, an additional 57,000 jobs, and an additional US$190 million in government revenue by 2040. Fisheries. The seas surrounding the Pacific Island countries are endowed with valuable tuna resources. Some 34 percent of the world’s annual tuna catch is supplied from the exclusive economic zones of Pacific Island countries. Better management of these resources and stronger regional cooperation (e.g., to prevent overfishing) could create major benefits for Pacific Island countries. But about 90 percent of the expected benefits would be captured by Kiribati, Nauru, Tuvalu, Papua New Guinea, the Solomon Islands, and the Federated States of Micronesia. Fiji could expect to gain only US$6.5 million a year. Labor mobility. For many small Pacific Island countries, the most important opportunities for labor mobility are expansion of low- and medium-skilled migration to Australia, New Zealand, and the Republic of Korea. For Fiji, along with a few others with a relatively skilled labor force, the expansion of skilled migrant visas could be further explored. Given the already large number of skilled caregivers overseas, Fiji could, for example, explore establishing a formal caregiver program that would regularly recruit skilled Fijians to provide residential care to the elderly in Australia and New Zealand. By exploiting these opportunities, Fiji could expect to gain additional seasonal employment of 3,800 to 7,600 by 2040. However, concerns about the economic impact of brain drain and the social impact of absentee spouses and parents need to be addressed. Deep sea mining. Most Pacific Island countries have few subsoil resources that can be extracted onshore. But there may be undersea resources in their large exclusive economic zones. Their proximity to the tectonic ring of fire provides the right conditions for the buildup of mineral deposits. With improvements in sea-bed mining technologies, the costs of extraction are declining. Fiji, in particular, has a number of opportunities. Within its exclusive economic zone, sulfide resources have been identified. Exploration licenses have been issued by the government. However, it is said that the resources could not be commercialized until the 2030s. But there is little independent research, so it is impossible to estimate the impact of these opportunities on growth, employment, or government revenue. The report recommends adopting a precautionary principle in the management of deep sea minerals. Knowledge economy. The knowledge economy encompasses education and training, ICT infrastructure, and innovation systems. In the region, Fiji is best placed to take advantage of opportunities in these areas, given its high levels of human capital, supporting ICT infrastructure, and its diaspora of skilled workers and entrepreneurs. By boosting 42 investment in these areas, Fiji could generate sizeable economic dividends, creating an additional 30,000 jobs and 10 percent higher per capita income by 2040. Achieving Fiji’s vision While Fiji has great potential, success cannot be taken for granted. To meet the ambitious target of doubling per capita income by 2035, Fiji must substantially increase its growth rate. In the short to medium term, adding capital and labor can stimulate growth. The investment to GDP ratio, which is below 20 percent of GDP in 2015, is targeted to rise to 25 percent by 2035 in the National Development Plan. At the same time, there is potential to increase employment, given that a portion of Fiji’s labor force is unemployed or under-employed. But per capita income growth in the longer run can only be sustained through productivity growth, for example, by attracting FDI, increasing exports, and creating more jobs for highly skilled workers. To achieve further poverty reduction and shared prosperity, Fiji faces a new set of challenges. The poverty reduction agenda now largely revolves around reducing rising urban poverty. At the same time, the gaps in access to essential services, especially in remote areas, need to be addressed to reduce chronic rural poverty. Enhancing the prosperity of the bottom 40 percent rests in large part on the implementation of reforms to transform the economy to a more private sector and productivity led growth path, and creating better, more productive jobs that moves people out of the informal sector and permanently out of poverty. In addition, Fiji needs to ensure that development is sustainable. This will require adapting to and mitigating climate change, reducing the impact of exogenous shocks by building buffers and increasing flexibility, and protecting the most vulnerable through well-targeted social protection. The SCD sets out three pathways to achieving the vision by taking hold of the opportunities: • Accelerating inclusive growth • Ensuring access to services by all • Building resilience. 43 4. Pathway I. Stronger growth The analysis in the preceding chapters shows that growth has been inclusive, but low. To accelerate progress toward the twin goals, therefore, Fiji needs to accelerate growth while making sure that it remains inclusive. The analysis in Chapter 2 points to three key constraints on growth: low investment, weak exports, and low productivity of jobs (see Drivers of growth). Why is investment low? Fiji’s investment rate, at 18 percent of GDP, is unlikely to be enough to achieve the government’s indicative growth target (see Chapter 3). A rule of thumb, admittedly approximate, indicates that achieving a long-run GDP growth rate of 5 percent requires gross capital formation of at least 25 percent of GDP, of which 50 percent or more should be private. The low investment rate could have many causes, and the problem is to identify the most important. This matters, because not all potential constraints can be addressed at the same time. To find the most important constraints, both perception data and objectives measures are considered. Perception data The most direct approach to identifying the binding constraints to private investment is to ask the firms.11 There have been two firm surveys in Fiji, the SME Business Survey (2003) and the Enterprise Survey (2009), but unfortunately nothing more recent. The top five constraints that the firms identified in these surveys are summarized in 11 Can perception data provide useful information on the main constraints on the private sector and be used to prioritize reforms? Clarke (2010) uses a natural experiment in South Africa to investigate these questions. When the World Bank’s 2007-08 Enterprise Survey was being carried out, a major electricity crisis hit South Africa. The crisis resulted in many more managers saying that power was a serious constraint on enterprise operations—the share rose from about 10 percent of managers before the crisis to close to 50 percent after the crisis. But it also resulted in greater concern about most other areas of the investment climate—including taxation, regulation, and other kinds of infrastructure unrelated to the crisis. This suggests that managers do not fully compartmentalize their responses. Moreover, the changes were large enough to suggest that cross- time comparisons of perception are difficult. 44 Table 4.1. 45 Table 4.1 Obstacles to growth as reported by firms here Enterprise Survey (2009) SME Survey (2003) Large firms Medium firms Small firms Political instability Political instability Political instability Finding, retaining staff Labor regulations Access to finance Tax rates Productivity of staff Inadequately educated Labor regulations Crime, theft, and disorder Lack of customers workforce Access to land Inadequately educated Labor regulations Government stability workforce Crime, theft, and disorder Practices of the informal Practices of the informal Inefficiencies in the sector sector production process Source: World Bank Group. Political stability was identified as a top constraint in both surveys, perhaps not surprisingly because the surveys were done three years after coups. Concerns about crime, theft, and disorder also ranked high among the concerns of small firms. These results suggest that concerns about the appropriability of returns to investment have been a major cause of low investment in Fiji. Labor regulation, lack of skills, and low productivity of labor were also among the top concerns for firms of all sizes. Productivity complaints increased with firm size: the Enterprise Survey found that this was among the top constraints for 9 percent of small firms, 20 percent of medium-size firms, and 26 percent of large firms. The SME Business Survey found that productivity was among the top constraints for 30 percent of micro firms and 95 percent of medium-size firms. As the complaints were more frequently voiced by larger firms, foreign-owned firms, and exporting firms, the quality of labor may have been an important impediment to attracting FDI. Access to finance did not appear to be a major issue. • The SME Business Survey paid particular attention to the survey respondents’ ability to access capital. The findings were striking. Fiji was the only country in the region where working capital was not among the most common problems. The proportion of Fiji firms relying on friends and family savings was also the lowest in the region. Commercial bank loans were provided to 22 percent of firms at startup. About 65 percent of the respondents received loans, overdrafts, or revolving facilities from banks; 32 percent of the respondents did not apply for the loan; and only 4 percent of those had applied did not get a loan. • The Enterprise Survey also did not find access to finance to be a constraint for either small or for large firms (Figure 4.1). For medium-size firms, however, 14 percent reported access to finance as a major constraint. A closer look at the survey data 46 suggests that to some extent this reflected greater demand for investment finance among the medium-size firms.12 More information is needed on the efficiency and equity of finance to cater for startups, small businesses, and businesses owned by women. The government has identified promotion of SMEs as one of its goals and is seeking ways to strengthen the support framework (Error! Reference source not found.). Box 4.1 Promoting SME in Fiji The government wants to create a more consistent and coherent policy to support SMEs. It provides a considerable amount of support to SMEs through tax incentives and small grants, but it is delivered through various sectoral ministries and agencies without an overall understanding of the full package of assistance, whether there is any duplication, or whether it is effective. Some assistance, for example, is provided by Ministry of Industry, Trade and Transport through small grants to entrepreneurs; some through the Ministry of Agriculture to the agriculture sector; some to women entrepreneurs through other programs. The government’s plan is to create a coordinating agency for SMEs that will plan, develop, and deliver assistance to SMEs. It would coordinate with sectoral agencies and be guided by a council chaired by the Prime Minister. The government will also review the assistance currently available to SMEs and develop an SME master plan. Part of this process will involve more research on SMEs in Fiji, including developing a database of SMEs in Fiji. To the extent that data exists already, it is scattered across different agencies. A committee chaired by the Deputy Governor of the Reserve Bank is developing legislation for the SME agency. IFC is helping review the draft SME law and the terms of reference for the agency. Neither access to land nor government regulations featured among the most pressing constraints (Figure 4.1). However, given the uncertainty regarding perception data and their vintage, these issues are investigated below with objective data. 12 About 56 percent of the firms had access to bank loans or lines of credit; 56 percent also used bank loans to finance, on average, 35 percent of their investment; 46 percent of the respondents said they did not need a bank loan to finance their investment; the medium-size firms reported using equity to finance on average 12 percent of investment. These figures together suggest that the medium-size firms that already use bank loans wanted to increase the size of the loans to finance investment but found it difficult. Excess demand was also reflected in the required collateral as proportion of loans which were higher than those required for loans to smaller firms. 47 Figure 4.1 Most important constraints reported by firms in Enterprise Survey Percent of firms surveyed (number of employees in parentheses) Small (5-19) Political instability Tax rates Crime, theft and disorder Labor regulations Practices of the informal sector Customs and trade regulations Corruption Access to finance Electricity Transportation Tax administration Inadequately educated workforce Courts Business licensing and permits Access to land 0 10 20 30 40 50 Medium (20-99) Political instability Access to finance Labor regulations Inadequately educated workforce Practices of the informal sector Crime, theft and disorder Electricity Business licensing and permits Tax rates Access to land Transportation Tax administration Customs and trade regulations Courts Corruption 0 10 20 30 40 50 Large (100+) Political instability Labor regulations Inadequately educated workforce Access to land Crime, theft and disorder Electricity Customs and trade regulations Tax rates Transportation Tax administration Practices of the informal sector Courts Corruption Business licensing and permits Access to finance 0 10 20 30 40 50 Source: Enterprise Survey (2009). 48 Cross-country benchmarking and other objective measures The second approach to identifying constraints on investment is to look at objective data, including benchmarking Fiji against its peers. Fiji’s investment climate has not kept up with reforms in other countries (Figure 4.2). Fiji’s best rank in Doing Business was 62, achieved in 2010. Since then, it has dropped every year and its rank in 2017 is 97.13 The drop was mostly because other countries were getting better, not because Fiji was getting worse. But there were some exceptions. For example, starting a business now takes 11 steps instead of 8 in 2010; registering a property takes 69 days instead of 68 in 2010; and paying taxes takes 38 payments and 247 hours a year instead of 33 payments and 212 hours in 2010. Last year, Fiji decided to close the credit bureau, which according to Doing Business made it more difficult to access finance, and this decision has pushed Fiji’s ranking on getting credit down from 78 to 157. Frequent changes in policies also increases uncertainty and compliance costs . Frequent changes in the tax system do not help create an investment-friendly environment. Investors are attracted by simple, stable, and predictable tax systems that are administered efficiently and transparently. However, between 2014 and 2017, the tax system in Fiji has undergone 32 reforms.14 These changes have mostly increased the complexity of the tax system. More generally, businesses find it difficult to predict the evolution of public policy in Fiji. Many reforms in business policy in recent years were made with limited public consultation and have added to a sense of uncertainty among the business community (ADB 2012). 13 Country rankings in 2010 and 2017 are not strictly comparable because of differences in the number of countries studied in the report (183 economies in the 2010 report; 190 economies in the 2017 report). For a comparison over time, the report provides an alternative measure, the Distance to Frontier (DTF), which shows the absolute distance to the best performance on each Doing Business indicator. Somewhat confusingly, the higher the DTF score the better the performance: the DTF score ranges from 1000 to 0, with 1000 representing the best performance and 0 the worst performance. Based on the DTF, between 2010 and 2017, Fijis performance deteriorated from 68 to 60. 14 The changes include introduction of new taxes, introduction or elimination of exemptions and incentives, and legislative changes. Changes in tax rates or customs and import charges were not counted. 49 Figure 4.2 Doing Business in Fiji, 2012 vs. 2017 a. Fiji has slipped in the ranking… b. …and distance to frontier has increased DTF Compared to 2010 Ranking Compared to 2012 2012 2017 2012 2017 Starting a Business Starting a Business 100 0 Dealing with Dealing with Resolving Insolvency 80 Resolving Insolvency Construction Permits Construction Permits 50 60 40 100 Enforcing Contracts Getting Electricity Enforcing Contracts Getting Electricity 20 150 0 200 Trading Across Borders Registering Property Trading Across Registering Property Borders Paying Taxes Getting Credit Paying Taxes Getting Credit Protecting Minority Protecting Minority Investors Investors Note: DTF = Distance to Frontier. DTF ranges from 100 to 0, with 100 representing the best performance. Closer to the center is worse performance. Where dark blue is visible, performance has worsened. c. Most economies made starting business easier…but Fiji made it harder Startup procedures Time (days) 0 5 10 0 10 20 30 40 50 OECD OECD East Asia & East Asia & Pacific Pacific Pacific Pacific Islands Islands Fiji Fiji 2012 2017 2012 2017 Source: World Bank (2017). The recent banking sector indicators show general improvements in access to credit. Credit to the private sector has increased at double digit rates since 2013 (Reserve Bank of Fiji 2016). It grew fastest for construction, mining, agriculture, transport, and private individuals. Together these sectors accounted for two-thirds of the growth. Credit for manufacturing and services also grew strongly, albeit at slower rates. The sector-wide composition of loans and advances of the commercial banks in 2016 shows that total business advances accounted for 62 percent of total loan assets, households for 28 percent, and the public sector for 9 percent. On December 31, 2016, the ratios of interest margin to gross income and of trading income to total income were 54.6 percent and 17.8, respectively, largely unchanged since 2011. The combined interest spread for commercial banks was 3.6 percent, a slight decline from 4 percent in 2011. Nonperforming loans continue to be low, at 1 percent of total assets. Cross-country indicators of access to finance also place Fiji ahead of its neighbors, but access could be rationed among fewer firms. In particular, domestic credit to the private sector as a share of GDP, the main aggregate indicator of access to finance, is higher in Fiji than in its peers (Figure 4.3a). In addition, outstanding loans with commercial banks 50 as a share of GDP, another indicator of access to finance, is higher in Fiji, suggesting that Fiji is more successful in getting credit to businesses (Figure 4.3b). However, there is some evidence that access to finance is concentrated among fewer businesses. Loan accounts per 1,000 adults are 178, similar to the average for the region but lower than for upper-middle- income countries (Figure 4.3c). This suggests that a fewer businesses obtain loans from commercial banks. Indeed, compared with other developing countries, Fiji provides less access to commercial banks loans for a given amount of credit (Figure 4.3d). Figure 4.3 Cross-country comparison of main indicators of access to finance a. Domestic credit to private sector b. Deposits and loans with commercial banks Percent of GDP Percent of GDP 100 80 84 80 63 65 60 53 60 40 40 20 20 0 0 Fiji East Asia Upper Small Pacific Fiji East Asia Upper Small Pacific & Pacific middle states Islands & Pacific middle states Islands income income Domestic credit to private sector Outstanding deposits with commercial banks Domestic credit to private sector by banks Outstanding loans with commercial banks c. Deposits and loans with commercial banks d. Less access to bank loans for the same Per 1,000 adults amount of credit 1,500 1000 1,389 Bank borrowers per 1,000 adults, 2015 800 1,000 600 500 400 178 - 200 Fiji East Asia Upper Small Pacific & Pacific middle states Islands income 0 Deposit accounts with commercial banks 0 50 100 150 Loan accounts with commercial banks Bank credit to the private sector/ GDP Source: IMF Financial Assessment Surveys 2017 and World Bank staff estimates. Few cross-country indicators exist for access to land for commercial use. But Doing Business provides some quantitative and qualitative information. According to these indicators, access to land does not stand out as especially problematic in Fiji. Based on cost and the number of procedures, Fiji’s land administration is relatively efficient: Fiji outperforms the average country in the Pacific, in East Asia, and even in the Organization for Economic Cooperation and Development (OECD) (Table 4.2). Based on time, Fiji falls 51 behind East Asia and the OECD, although it is well ahead of the Pacific. Specifically, in Fiji, it takes 69 days to complete a transfer of title, with the final settlement at the Registrar of Titles Office taking up to 60 days, compared to 4.5 days in Australia and just 1 day in New Zealand. Based on quality of land administration, Fiji also does not do badly: it is well ahead of the average in the Pacific and in East Asia and only a little behind the OECD (Table 4.3). One difficulty in Fiji, however, is transparency: statistics about land transactions are not made available to the public, while the procedures involved in property transactions, the service standards, and the fee schedule for land registry services are not easily accessible. There are also no mechanisms in place for filing a complaint. Table 4.2 Efficiency of land administration Fiji East Asia Pacific OECD Procedures (number) 4 5 5 5 Time (days) 69 50 124 22 Cost (% of property value) 3 4 5 4 Quality of the land administration index (0–30) 19.5 16.6 8.5 22.7 Source: World Bank (2017). Only high-income OECD countries are included. Table 4.3 Quality of land administration Fiji East Asia Pacific OECD Quality of the land administration index 20 16 9 23 Reliability 4 4 2 7 Transparency 2 3 2 4 Coverage 8 4 2 6 Dispute Resolution 6 6 4 6 Equal Access 0.0 -0.1 -0.3 0.0 Source: World Bank (2017). Only high-income OECD countries are included. Major changes in the customary land tenure system are unlikely given the political sensitivity. In addition to political risk, distributional consequences of land reform are not well studied, although poverty is becoming more prevalent among iTaukei households, who are the communal land owners. The government’s approach is to make the lease system more efficient. For example, the government has set up a land bank through which land owners can allow government to on-lease their land at market rates. The government has also extended the maximum lease period to 99 years for businesses. There is much that can be improved in the performance of land administration in Fiji even without a wholesale change in the customary land tenure system. 52 Why are exports weak? As a small, open economy, Fiji is well aware that a strong export performance is a prerequisite for reaching robust, sustained, and shared growth. But its export performance has been weaker than its peers’. Strong export performance does not simply mean high export growth: increased diversification from low value added to higher value added exports would reduce export volatility and raise the growth enhancing potential of trade. Even on this score, Fiji has not done so well. Its export basket has shown little sign of diversification. Analysis of disaggregated trade data shows that food and raw materials still dominate Fiji’s merchandise exports (Figure 4.4a). The technology content of the exports has decreased slightly, with the disappearance of garments exports (Figure 4.4b). There is a slight decline in concentration of exports but this may simply reflect the decline of sugar (Figure 4.4c). More importantly, the current structure of exports and its underlying capability set does not provide a good basis for a future rapid structural transformation. The “product space” depicted in Figure 4.5 describes the network of relatedness between globally traded products. Red and blue markers plot what Fiji current exports, both traditionally (“classics”) and more recently (“emerging”). Most of Fiji’s exports are in a very disconnected part of the product space. According to the theory, this suggests that the set of inputs and skills used to produce current exports (“the capability set”) could not easily translate into other areas of specialization. It would involve very long jumps. Figure 4.4 Goods export diagnostics a. Sectoral composition b. Technology composition 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 2000 2002 2004 2006 2008 2010 2012 2014 2005 2007 2009 2012 2014 Ores and metals Manufactures Food exports Agricultural raw materials Primary Products Resourced Based Low tech Medium tech High tech c. Largest products exported d. Herfindahl indexes 100% 0.5 0.2 80% 0.4 60% 0.3 0.1 40% 0.2 20% 0.1 0% 0.0 0.0 1990 1993 2001 2004 2007 2010 2013 1990 1993 1996 1999 2002 2005 2008 2011 2014 Twenty largest Ten largest Five largest Sugarcane Export Partners Import Partners Export goods (RHS) Source: UN Comtrade database and World Bank staff estimates. Note: Technology classification of export is based on Lall (2000). Herfindahl Index ranges from 1/N to 1, where N is the number of trading partners. The higher the value, the more concentrated (less diversified). 53 Figure 4.5 Product space map of Fiji’s exports Classics Emerging Source: Hidalgo et al (2007), UN Comtrade database, and World Bank staff estimates. Note: Both “classics” and “emerging” are products that Fiji exports with revealed comparative advantage (RCA), that is, the product’s value share in Fiji’s export exceeds the product’s value share in global trade. “Classics” are products that Fiji exports with RCA since 1998. “Emerging” are products that Fiji exports with RCA in the last five years but not earlier. 54 Services exports have fared better than goods exports but have also failed to improve sophistication. An analysis of disaggregated trade data on services shows that Fiji’s service export sophistication has failed to keep pace with other East Asian countries, suggesting that Fiji has not maximized the growth potential of its services export (Figure 4.6).15 Figure 4.6 Service export diagnostics a. Composition of service export b. Export and re-export 100% 1.6 US$ billion 80% 1.2 60% 0.8 40% 20% 0.4 0% 0.0 2005 2007 2009 2011 2013 2005 2008 2011 2014 Travel Transport Insurance and financial Computer & communications Exports Re_export c. Re-export by destination d. Sophistication of service export 600 15000 $US million 500 14000 EXPY for Service Exports 400 300 13000 200 12000 100 11000 0 2005 2008 2011 2014 6.5 7 7.5 8 8.5 9 Areas, nes Bunkers Australia Log of GDP per capita China Japan Singapore China India Sri_Lanka Indonesia Malaysia Fiji USA Pacific Islands Source: UN Comtrade database and World Bank staff estimates. Fiji, like many other small developing states, maintains a fixed exchange rate. The question of whether the fixed exchange rate has contributed to its weak export performance has been debated (e.g., IMF 2013). Under fixed exchange rates, the real exchange rate usually becomes too high over time as domestic inflation exceeds inflation in trading partners. In Fiji, in particular, frequent natural disasters and terms of trade shocks can quickly lead to inflation and reduce competitiveness. While a study by the Reserve Bank 15 To formalize the notion of sophistication, a measure of export sophistication (EXPY) is constructed using the framework developed in Hausmann, Hwang and Rodrik (2007). Roughly speaking, EXPY is a weighted average of the per capita income levels of countries that are competitively exporting similar services. An increase in EXPY results if Fiji is increasingly exporting those services that are also being exported by higher income countries, which is taken to imply that Fiji’s service export is becoming more sophisticated, or becoming more growth enhancing. 55 of Fiji (2016) has shown that the most recent devaluation, in 2009, was associated with a marked improvement in the trade balance that year, a more recent study by the IMF found it difficult to establish the impact of nominal devaluations in the long run (Gottschalk et al. 2016). Fiji has enjoyed growing remittances, raising the question of the Dutch disease. Such an exogenous inflow of foreign exchange, in the absence of central bank interventions, can lead to an appreciation of the currency, a deterioration of competitiveness, and a fall in net exports. Yet there is no empirical evidence that remittances in Fiji have led to real appreciation in the long run (Prakash and Mala 2016). Furthermore, in the short run, there is some evidence of depreciation of the real exchange rate after a positive shock in remittances, which dissipates by gradual appreciation over the following few years. The rationale, according to the authors, is that remittances are channeled to productive investment and boost domestic capacity, which puts little to no pressure on the domestic exchange rate to appreciate. Waqabaca (2000) also argues that Fiji has a reasonably well- developed financial system that helps channel household funds into investment opportunities. Fiji’s policy makers have recognized that FDI can play a positive role in promoting export diversification and sophistication. It can help by creating foreign demand, enabling deeper understanding of foreign preferences, and creating spillovers that raise quality standards. It can also play a positive role in export promotion, as foreign companies can help integrate Fiji into the global economy by easing access to foreign markets and including local enterprises in global production chains. But attracting FDI has been difficult, and Fiji has attracted much less than its comparators. A recent analysis of FDI, including a survey of would-be investors, identified that the existing tax incentives are not effective and a broader competitiveness agenda is required. A recent review of the investment law and associated legislation has identified a number of shortcomings. The arbitration law does not provide for the recognition of overseas arbitral awards. There are impediments to the repatriation of dividends and profits. And, in contrast to standard international practice, confiscation of the assets of an investor is possible, not only if the investors commit crimes but also if they breach conditions of their investment certificate. The costs of trade are high in Fiji. The small size of shipments and long distance from the markets will always put Fiji at a cost disadvantage against leading exporters of East Asia. The frequent natural disasters also increase the costs of trade and insurance premiums. But Fiji can improve the quality of trade and transport related services. There has been an improvement in Fiji’s ranking in the Logistic Performance Index—from 144 in 2010 to 136 in 2016. The improvement in performance has been driven by improvements in customs clearance (e.g., speed, simplicity and predictability of formalities), trade and transport related infrastructure (e.g., ports, railroads, roads, information technology), and ability to track and trace consignments. However, the country has suffered setbacks in ease 56 of arranging cost competitive international shipments and the timeliness of shipments in reaching destinations. At a ranking of 136, Fiji is still well behind its peers in upper-middle- income countries. There are also several non-infrastructure factors that currently constrain trade. Trading across borders in Fiji is made costlier by red tape—money and time spent on border compliance is greater in Fiji than in its peers, although they do not stand out as exceptionally high in a global context (Figure 4.7). Observers also suggest that frequent changes in import tariffs, customs duties, and price controls affect the costs and uncertainty of trade. These changes are often determined without transparency and are poorly communicated to traders. Figure 4.7 Trading across borders a. Time to export and import b. Time to export and import Boarder compliance in hours Documentary compliance in hours 250 180 200 Export Export Import 120 Import 150 100 60 50 56 56 42 34 0 0 Fiji Top Quintile Quintile Quintile Bottom Fiji Top Quintile Quintile Quintile Bottom quintile 2 3 4 quintile quintile 2 3 4 quintile (best) (worst) (best) (worst) c. Cost to export and import d. Cost to export and import Border compliance in US$ Documentary compliance in US$ 1200 180 Export Export 800 Import 120 Import 400 60 76 317 320 58 0 0 Fiji Top Quintile Quintile Quintile Bottom Fiji Top Quintile Quintile Quintile Bottom quintile 2 3 4 quintile quintile 2 3 4 quintile (best) (worst) (best) (worst) Source: World Bank (2017). Why is employment low? With official unemployment of about 8 percent, one of the main concerns in the country is underemployment. The question is whether the problems are on the demand side, on the supply side, or in a mismatch of supply and demand. 57 Are there problems of demand? Job growth has taken place in relatively low wage sectors. Employment has grown at 2 percent a year since 2000, in line with what one could expect given the low economic growth (with a growth elasticity of employment of 0.7). Average earnings have increased a little and job growth has taken place in relatively low wage sectors (Figure 4.8). In the last decade and a half, the sectors that increased employment the most were retail and hospitality, which created about 12,500 additional jobs, and construction, which added about 8,500 jobs. Over 8,000 jobs in manufacturing were lost following the 2000 coup and the expiration of the various trade preferences. Among higher wage sectors, the only significant areas of growth were financial services (2,400 jobs) and the public administration (6,700 jobs). Figure 4.8 Number of jobs created from 2000 to 2014 100% Public services 80% Business & Financial services Transport 60% Retail & hospitality 40% Construction Utilities 20% Manufacturing 0% Mining Agriculture -20% Source: World Bank staff estimates based on Fiji Bureau of Statistics employment statistics. At the same time, firms are finding it difficult to find and retain skilled workers. Both labor regulation and inadequately educated workforce ranked high among the most important constraints. Complaints about labor regulation were higher among the smaller firms, while complaints about lack of skills and productivity of workers increased with size. The Enterprise Survey (2009) found that 9 percent of the small firms, 20 percent of the medium-size firms, and 26 percent of the large firms surveyed complained about “inadequately educated workforce.” The SME Business Survey (2003) found that 30 percent of the micro firms and 95 percent of the medium-size firms found productivity of staff as the most important constraints. The complaint was more frequently voiced among foreign-owned firms, exporting firms, and those operating in the manufacturing sector. There is anecdotal evidence that vacancy rates in skilled positions are being filled with delays and difficulties. While information on the vacancy rates in the private sector is not readily available, information for the public sector suggests that vacancies for doctors, engineers, IT specialists, and other skilled and technical positions are filled with difficulty. Reflecting shortages of specialized skills, a recently completed compensation survey—a benchmarking of salary levels of more than 140 civil service positions against comparable 58 positions in the private sector—suggested larger gaps among accountants, engineers, statisticians, surveyors, and medical professionals, where private sector counterparts typically command much higher salaries. Are there problems of supply? The supply side of the labor market has benefited from positive trends in demography, human capital, and the labor force. The share of working age population is rising (Figure 4.9), the rate of labor force participation is rising, and the educational attainment of the population is rising (Figure 4.10). Figure 4.9 Population by age and sex Figure 4.10 Schooling in population Age 100 Tertiary complete 75+ Tertiary incomplete Female 65-69 80 Male 55-59 Secondary complete 45-49 60 35-39 Secondary incomplete 25-29 40 15-19 Primary complete 5-9 20 Primary incomplete No education -10.0 -5.0 0.0 5.0 10.0 0 Share of population 1985 1990 1995 2000 2005 2010 Source: WDI and Barro-Lee Educational Attainment Data. Set against these positive trends, Fiji has experienced strong emigration over the past quarter century, particularly among skilled workers. It has one of the largest stocks of highly educated migrant workers in OECD countries. While they have contributed to growing remittances, the emigration of highly educated and skilled workers represents a brain drain, subtracting from current and future productivity growth in the country. There is some international evidence that rising remittances may reduce labor force participation, for example by raising reservation wages and requiring women to care for the extended family of migrant spouses. At the same time, the emigration of highly educated workers could encourage young people to get more education in the hope of finding better jobs abroad. There is, however, little empirical evidence on the general equilibrium impact of emigration, remittances, and labor market outcomes in Fiji. Are there problems of mismatch? Labor market mismatch—the gaps between skills demanded and skills supplied— could result in lower labor productivity. In Fiji, there is some evidence that there is a mismatch between the career aspirations of young people and human resource needs (Nilan 59 et al. 2006; ADB 2015). Also, school leavers are said to be returning to their villages and settlements without jobs because they had been focused on securing white collar, but could not get them (Cavu et al. 2009; ADB 2015). In addition, a survey of employers by the Fiji National University during 2013–14 revealed that many were dissatisfied with skills and attitudes of job seekers. For example, employers in the agriculture sector said that most graduates were theory-based and lacked practical experience, while those in the hospitality sector said that jobseekers lacked housekeeping and cooking skills (ADB 2015). Excessive or cumbersome regulations could increase labor market frictions and mismatch. Well-designed regulations play a key role in protecting equity and promoting efficiency in labor markets, but firm level surveys in Fiji identified labor market regulations as among the most important constraints (Figure 4.1). Smaller firms, in particular, were more likely to find it difficult or costly to comply with these regulations. A recent study by the ADB (2015) finds that Fiji has relatively stringent labor market regulations compared with other Pacific Island countries (Table 4.4). Table 4.4 Minimum wages and paid leave US$/hour and days/year Fiji Solomon Islands Samoa Australia New Zealand Adult minimum wage 1.00 0.56 1.01 13.80 11.03 Sick, personal, or career leave 10 10 10 5 Bereavement or compassionate leave 3 2 3 Annual leave 10 15 10 28 20 Public holidays 10 11 12 ~12 11 Source: ADB (2015). The national minimum wage was raised from F$2.00 to F$2.32 an hour in July 2015, and there is a campaign to raise it to F$4.00 an hour. In addition to the national minimum wage, there are many other, industry-specific minimum wages and conditions (Table 4.5). The government also provides (and frequently modifies) employment incentives aimed at promoting employment among certain groups. For example, this year it has introduced the following deductions: 200 percent for salaries and wages paid to first- time employees for the first 12 months of employment (previously 150 percent); 200 percent for salaries and wages paid to students; 150 percent for employees’ education fees; and 300 percent for salaries and wages paid to disabled persons—all “subject to certain conditions.” While they may be well intended, the detailed nature of the regulations, and their seemingly frequent and sudden changes, raise costs of compliance and increase policy uncertainty. 60 Table 4.5 Wage Regulation Orders: Minimum wage and core conditions Hourly Weekly Industry Other minimum terms and conditions rates (F$) hours Building, Civil and Overtime, meal allowance, bereavement leave, 1.51–2.49 45 Electrical Engineering subsistence allowance, attendance money Rest day, split shifts, overtime, night shift, meal Hotel and Catering 1.36–1.57 48 allowance, bereavement leave Garment Industry 0.90–1.07 45 Overtime, meal allowance, bereavement leave Shift work, rest day, overtime, meal allowance, Manufacturing 1.51 48 bereavement leave, night allowance Rest day, out-station allowance, meal allowance, Mining and Quarrying 1.34–2.17 45–48 overtime, bereavement leave Overtime, meal allowance, other allowance, Printing Trades 1.30–1.57 45–48 bereavement leave Rest day, subsistence allowance, meal allowance, Sawmilling and Logging 1.74–2.59 45–48 overtime, bereavement leave Overtime, bereavement leave, meal allowance, Security Service 1.20 45–48 transportation provided Wholesale and Retail Overtime, meal allowance, subsistence allowance, 1.36–1.72 45–48 Trade bereavement leave Split shift, rostered day off, overtime, subsistence Road Transportation 1.20–1.70 48 allowance, meal allowance, bereavement leave Source: ADB (2015). Policies to accelerate growth To achieve stronger growth, Fiji needs to revamp its growth model. To strengthen investment, traditionally, the government has relied on investing by itself or by providing an elaborate system of incentives. Instead, the government could accelerate structural reforms. To reduce policy uncertainty and compliance costs, the government could undertake more careful analysis of the costs and benefits of proposed policies before they are introduced and systematic monitoring of their impacts after they are introduced, both in consultation with stakeholders. Access to land did not feature as an important constraint in the firm surveys but existing literature often cites its negative impact on investment. Cross-country comparison of costs and efficiency of land administration suggests that much that can be improved in the performance of land administration in Fiji even without a wholesale change in the customary land tenure system. Strengthening exports requires changing the traditional model of relying on trade preferences, which have failed to create lasting competitiveness. Instead, the government could continue to invest in transport and telecommunications infrastructure. Although there is not much Fiji can do about its size and distance from the market, which places it at disadvantage vis-à-vis leading exporters in East Asia, it can reduce the red tape that is 61 making trade across its borders costlier. To attract for FDI, the government could accelerate structural reforms to reduce costs and improve the business climate as discussed above. In addition, it could review and consolidate the complex system of exemptions and incentives, which is making the tax system in Fiji less not more attractive to investors. Further, the government could amend the current investment law to bring it in line with good practice, while recognizing national policy objectives. To create more productive jobs, these reforms to increase private investment and exports should help. At the same time, the government could encourage hiring by reviewing and simplifying the complex system of labor market regulations and employment incentives, again in consultation with stakeholders. 62 5. Pathway II. Better access to services by all Economic mobility and inequality in opportunities Although poverty and inequality have been falling, signs of inequality are rising, which could become a source of tension and threaten economic and social stability. Urban poverty and visible signs of inequality are increasing with the expansion of squatter settlements. Meanwhile, there is still a gap in access to some essential public services between rural and urban areas. Economic mobility also appears to be weakening: the analysis of economic mobility suggested that about 6 percent of the poor were upwardly mobile between 2002 and 2008, and about 4 percent between 2008 and 2016 (Figure 5.1). This could intensify discontent and social tensions. Furthermore, inequality in opportunity in access to services, where they exist, could widen inequalities and reduce economic mobility in the future (Figure 5.2). Figure 5.1 Economic Mobility Intra-generational mobility, 2002-08 100% 80% 60% 40% 20% 0% No Education Tertiary Indo Fijian Household Size 6+ Secondary Wage I-Taukei Female head Employer 1 2 3 4 5 Married 25-34 25-44 45-54 55-64 Rural Population Urban Age of Head Primary Education Male head Self-Employed Not Married Unpaid Intra-generational mobility, 2008-13 100% 80% 60% 40% 20% 0% Secondary Female head Household Size I-Taukei Wage 6+ Tertiary Employer Married 2 4 No Education Rural Urban 1 3 5 Indo Fijian Not Married 25-34 25-44 45-54 55-64 Self-Employed Unpaid Population Age of Head Male head Primary Education Chronic Downward Upward Non-poor Source: World Bank staff estimates based on the World Bank poverty harmonized database and HIES. Note: Results shown are the lower-bound estimates, which are conservative estimates of economic mobility. See also Chapter 1, Figure 1.4 and the associated footnote. 63 Figure 5.2 Human Opportunity Index Source: World Bank staff estimates based on HIES 2013/14. Note: The Human Opportunity Index measures how individual circumstances such as gender, region, and education of household head affect a child’s access to basic opportunities such as water, education, electricity, and sanitation. It is an estimate of how far a society is from universal access and how equitably access is distributed across individuals. The opportunities are examined among those aged 15 and less to reduce selection bias and to determine if children have similar chances and access to services in life. Education Fiji enjoys almost universal primary education. The net primary enrolment rate is 96.8 percent, well above the Pacific average of 86.4 percent. The net enrolment rate in secondary education is 80.3 percent, the highest in the Pacific. Broad access to education has helped reduce the youth literacy rate to 99.5 percent, which is above the average for upper-middle- income countries and for East Asia and the Pacific. There is little inequality in opportunities to access basic education (Figure 5.2). Early education is established throughout the country among children in both rural and urban areas. There is very little difference in educational attainment, and virtually all children between the ages of 6 and 12 are enrolled (Figure 5.3). 64 Figure 5.3 Education attainment of boys and girls aged 15 or less in 2008 100 4 4 4 4 18 19 21 20 80 60 40 38 36 39 40 6 5 5 5 20 32 33 34 32 0 Boys Girls Boys Girls Rural Urban No education Pre-school Primary not completed Completed primary Completed lower secondary Source: World Bank staff estimates based on HIES 2008/09. The quality of education appears to have been improving. Fiji does not participate in internationally comparable studies such as the Program for International Student Assessment (PISA). But available within-country indicators of education quality suggest improvement. For example, between 2000 and 2012, the survival rate to the last grade of primary school in Fiji increased for boys and girls to 98 and 96 percent, respectively, from 82 and 91 percent. Between 2011 and 2015, the basic literacy and numeracy of school children in years 4, 6, 8 also improved. Finally, between 2010 and 2014, student teacher ratios improved from 27 to 25 in primary schools and from 16 to 14 in secondary schools. One of the remaining challenges is reducing the school dropout rate. Educational enrollment begins to fall in the upper secondary and tertiary education age groups. Inequality begins to emerge in the lower secondary completion rate (Figure 5.2). Careful examination reveals that differences in completion rates is explained by the education of the household head (Figure 5.6). The income of the households, their urban or rural location, and the gender of the child also contribute to the differences. In rural and maritime schools, long travel time, poor transport infrastructure, and lack of reliable supply of water and electricity for schools contribute to regional differences in completion rates. Ensuring equal opportunity for education is an important policy objective of the government. Public spending on education has increased significantly in recent years, rising from F$252 million in 2011 to F$467 million in 2015. In addition, there are a number of subsidies programs, including free buses, school meals, and scholarships and student loans for tertiary education. As a share of total government expenditure, spending on education has remained around 15 percent. As a share of GDP, it has risen from 6 to 8 percent, and spending per capita for the population under age 25 has more than doubled from F$618 to F$1,253. 65 The composition of education spending is similar to that of other countries in the region (Figure 5.4). Although Fiji appears to spend less on secondary education, this is misleading because it spends more on education overall (Figure 5.5) so that its spending on secondary education as a share of GDP is similar to that of other countries. The larger share of “unallocated” in Fiji possibly reflects the higher overhead of serving a smaller, dispersed population. Figure 5.4 Composition of education Figure 5.5 Level of education and health spending spending Percent of total education spending Percent of GDP 40 6.0 5.8 4.8 4.84.6 30 4.4 4.2 4.2 4.1 3.9 4.0 3.7 20 10 2.0 0 East Asia Small Middle Upper Fiji 0.0 and States Income Middle East Asia Small Middle Upper Fiji Pacific Income and States Income Middle Pacific Income Primary share Secondary share Tertiary share Unallocated Education Health Source: WDI. Public spending on education has been pro-poor. The benefits have generally been progressively distributed (Figure 5.7 and Figure 5.8). Although equally available to all income groups, free education, school meals, and transport subsidies have benefited the poor more. University grants and scholarship schemes are less pro-poor because students who attend universities are mostly from richer households. The government has implemented specific initiatives to reduce dropout rates and improve rural schools. To reduce dropout rates, the government has started the “Matua” program, providing continuing education for students who have dropped out of schools for various reasons—such as living in a rural area or getting pregnant—to complete formal schooling. To increase access to education in remote areas, the government has introduced distance learning and flexible learning in nine primary and eight secondary schools. The programs help both students and teachers get access to curriculum materials and experienced teachers. To improve the quality of teaching throughout the country, the government is providing an allowance to encourage teachers to go to rural and maritime areas. 66 Figure Figure 1.5.6 Decomposition Shapley of Inequality Decompositions in Opportunity of the Dissimilarity Index Source: World Bank staff estimates based on HEIS 2013/14. Note: The results show the Shapley decomposition of the dissimilarity indexes shown in Figure 5.2. Each color block represents the marginal contribution of a child’s circumstance (e.g., gender, region, and education of household head) to the dissimilarity in the child’s access to basic opportunities (e.g., education and electricity). 67 Figure 5.7 Beneficiaries of government Figure 5.8 Beneficiaries of government education budget, 2015 education related subsidies, 2015 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Primary Secondary Tertiary Free busfare Freed University Education Education Education for children eucation scholarship q1 (lowest) q2 q3 q4 q5 (highest) q1 (lowest) q2 q3 q4 q5 (highest) Source: World Bank staff estimates based on HIES 2013/14. Health In health, Fiji has largely failed to meet the millennium development goals (MDGs). At the same time, as the country has become richer, non-communicable diseases (NCDs) have become widespread. About 82 percent of deaths in Fiji are due to NCDs, 10 percent to communicable diseases, and the remaining 8 percent to other causes (Box 5.1). Box 5.1 Causes of NCDs in Fiji Fiji has a high prevalence of NCDs, or “illnesses and injuries that are not communicable or infectious in nature” but are “related to life-style factors.” They encompass conditions such as “preventable blindness, asthma, environmental and inherited cancers, mental health disorders, as well as injuries, drowning, and other related accidents” (Ministry of Health and Medical Services 2014 a). In 2015, NCDs accounted for 84 percent of total deaths in Fiji, compared to an average of 72 percent in the East Asia and Pacific region and 76 percent in the upper-middle-income countries. The three most prevalent NCDs in Fiji are ischemic heart disease (hardening of the arteries), diabetes, and stroke. The leading risk factors for NCDs are obesity, poor diet, and high blood sugar (Global Burden of Disease Study 2013). According to the WHO’s NCD STEPS surveys, the frequency of risk factors increased between 2002 and 2011. For example, obesity has increased from 24 to 32 percent of the population surveyed, and those with high blood glucose from 20 to 30 percent. The Ministry of Health and Medical Services’ NCD Strategic Plan 201519 aims to continue its focus on prevention and to strengthen preventative care at grassroots levels. For example, the Plan discusses promoting awareness, counselling, and early detection through better training and appropriate tools at the levels of community health workers, nursing stations, health centers, and specialized outpatient services. The government is the primary provider of health care and access to health care is widespread. About 75 percent of the population report having access to health services 68 (Asante et al. 2017). The country has a good network of health facilities—nursing stations, health centers, and sub-divisional hospitals—that provide primary and secondary care (Box 5.2). Divisional and specialized hospitals provide a range of services at tertiary level. There are 1,726 inpatient beds in Fiji—a ratio of about 2 beds per 1000 people. Recent reforms to decentralize outpatient services from divisional hospitals to health centers is believed to have further increased access to services. Box 5.2 An overview of health care system in Fiji Primary health care services are delivered through a network of 98 nursing stations, 84 health centers, and 19 sub divisional hospitals (Roberts et al. 2011). Nursing stations are the first point of contact with the formal health system for rural people. Each of the country’s four administrative divisions (Central, Northern, Western, and Eastern) has at least 20 functional nursing stations. A nursing station is typically staffed by one registered nurse and caters for between 100 and 5,000 people. They deliver the most basic of health services, including maternal and child health care and family planning. Health centers, in contrast, are staffed by either a doctor or a nurse practitioner. The number of staff ranges from 2 to 20 and they usually serve between 3,500 people in rural areas and 10,000 in urban areas. Health centers provide comprehensive primary health care and also serve as the first level of referral for nursing stations. Secondary health care begins in sub divisional hospitals but more complex cases are referred to the three divisional and two specialized hospitals which provide secondary and tertiary-level care. There are about 130 private general practitioner clinics that provide services to complement those provided by the public sector. These clinics are largely day clinics and provide general outpatient services. Public spending on health has increased of late but falls well below the levels of Fiji’s peers. It has increased from F$150 million in 2011 to F$267 million in 2015 (3.9 percent of GDP)—lower as a share of GDP than the averages for East Asia and the Pacific (4.8 percent), small states (4.6 percent), and upper-middle-income countries (4.2 percent) (Figure 5.5). In terms of the composition of current spending, in 2014, hospitals received 64 percent of the total, and nursing stations and health centers—which are used more heavily by the bottom 40 percent—15 percent (Ministry of Health and Medical Services 2014b). In terms of the types of services provided, curative care accounted for 42 percent of current spending, while preventative care accounted for 33 percent. Spending on preventative care has expanded significantly in recent years, consistent with the government’s NCD strategy (Box 5.1). Within preventative care, information, education and counselling programs accounted for the largest share of the spending, followed by monitoring and early disease detection, all of which are important elements of the strategy. Public spending on health has been pro-poor. In total, the bottom 40 percent receives 43 percent of the benefits. Spending on nursing centers is the most pro-poor, with the poorest quintile receiving 61 percent of the benefits and the richest quintile only 2 percent (Figure 5.9). Unlike nursing stations, health centers and hospital outpatient and inpatient services appear to be used by all Fijians: the poorest and richest quintiles receive almost 69 equal benefits. Public spending on private hospitals and clinics account for less than 7 percent of spending. The distribution of the benefits from this spending is strongly pro-rich (Figure 5.9). The richest quintile received over 37 percent of the benefits of private general practitioner clinics and 41 percent of the benefits of private hospital outpatient services. The bottom 20 percent received less than 1 percent and 2 percent, respectively, of the benefits of these services. Private hospitals were not used at all by the poorest quintile. Figure 5.9 Distribution of health-care benefit: public sector 100% 80% 60% 40% 20% 0% Nursing Health Hospital Hospital GP/clinic Hospital Hospital station center outpatient inpatient outpatient inpatient Public sector Private sector q1 (lowest) q2 q3 q4 q5 (highest) Source: Asante et al. (2017). Although overall spending on health has been pro-poor, there is scope for increasing spending on health services that are better targeted at the bottom 40 percent . As mentioned above, 64 percent of government spending for health was on hospital services, compared to 15 percent on nursing stations and health centers. To expand coverage of health services and improve health outcomes, the government is considering contracting out health services to the private sector. The experience of other developing countries with such policies has been mixed. In some countries, the experience has been positive: the private sector has been more efficient and responsive to patient needs than the public sector because of market competition. In others, it involved trade-offs: outsourcing resulted in inequalities in access to health care, because the poor were unable to pay for private services. Additional research may be needed to provide sound policy advice in this area. Such research needs to be based on up to date information comparing the costs of service delivery under private and public hospitals. Information on the distribution of access to hospital services in the population by income and geography will also be needed to provide the baseline. 70 Housing Fiji has a housing crisis in urban areas, which is reflected in the growth of squatter settlements. The latest estimates suggest that 9 percent of the total population and 16 percent of urban population live in such settlements, most of them in greater Suva. The majority of the houses in these settlements are made of used timber and corrugated iron and are densely packed. They are vulnerable to fires, flooding, and landslides. Improper planning is a common feature of the settlements (Nemaia 2011). There are four main drivers of the growth of squatter settlements: • Accelerating migration from rural to urban area. Greater Suva accommodates about 244,000 people, or 57 percent of the total urban population, and is growing at 1.7 percent a year, which is more than double the national population growth rate. Urban infrastructure is unable to keep up with growing demand. • House price inflation. According to the Housing Authority, the lowest possible price for a two bedroom 60m2 house is F$82,000. This is too expensive for 65 percent of the urban population. The supply of homes at this price on the market is also scarce. Even the Housing Authority, with social obligations, only sets aside 5 percent of its lots and units in this price range. • High construction costs. The challenges include the use of expensive construction materials, a shortage of skilled labor, and a slow and complex process of obtaining land leases and construction permits. In addition, the annual cyclone season, which lasts from November to April, brings heavy rain and frequent floods, especially in Greater Suva, limiting the productive construction period. • The limited reach of housing finance. Commercial mortgage lenders typically lend F$100,000–200,000, thus serving only households in the top income quartile. The commercial mortgage rate is about 7 percent a year, which is reasonable relative to the annual inflation rate of around 4 percent, but lenders are not interested in going “down market” given the higher risk of serving lower income households. The government-owned Housing Authority provides mortgages to low- to middle- income households, but loan performance has been poor. While commercial mortgage lenders’ nonperforming loans have been around 3 percent of the total, the Housing Authority’s rate in 2011 was 39 percent. In 2015, the rate was down to 14 percent thanks to a government subsidy to allow the Authority to write off bad loans. Keenly aware of the needs, the government has been implementing numerous subsidy programs to expand access to adequate and affordable housing. In FY2016/17, the budget for public housing was F$25 million. The medium term projections show an increase to around F$30 million. In addition, the government has given loan guarantees and soft loans, including: a guarantee of a F$50 million loan from China Exim Bank to the 71 Housing Authority in 2010 for the development of 1,500 residential land lots; a guarantee of a F$20 million loan from China Exim Bank to the Public Rental Board in 2010 for the construction of 208 rental units; and a soft loan of F$25 million from the Reserve Bank of Fiji to the Housing Authority in 2010 for housing mortgage lending. The assessment of government spending on housing highlights two issues. • The first is insufficient targeting of public housing spending and subsidies. Around 40 percent of this year’s housing budget has been used to support the First Time Home Buyer program, which provides a subsidy of F$10,000 to buy land or F$5,000 to buy a home. This is a significant subsidy which is likely to benefit high- income households. Similarly, the F$500,000 transfer to the Housing Authority to write off its nonperforming loans was likely regressive (Table 5.1). • The second is the conflicting objectives of the two government agencies responsible for housing. The Housing Authority has a mandate to develop affordable homes and provide affordable home loans to low-income customers. The Public Rental Board has a mandate to provide affordable rental housing and subsidize rent for households with an annual income below F$10,000. As SOEs, both agencies are expected to make a profit. To meet its profitability objective, the Housing Authority has targeted households with an annual income of more than F$50,000. For instance, in its Valelevu Center project in which it is developing 900 lots, only 50 are for lower income segments. Similarly, the Public Rental Board’s efforts to increase profits are leading to an upward drift in the incomes of its target renters (Figure 5.10). Table 5.1 Government housing subsidies, 2016/17 De facto income targeting Program Share of housing subsids (%) High-mid income First-time home buyer 40 High-low income Housing Authority 22 Low income Squatter upgrading 23 Low income Lagi Lagi development 8 Low income Public rental Board 4 Special needs HART 2 Total 100 Source: Government of Fiji Budget Estimates 2016/17. 72 Figure 5.10 Distribution of benefit of Public Rental Board subsidies F$ 1,000 0 273 -42 273 273 273 273 -315 -583 -643 -778 -1,460 -1,000 -856 -917 -1,051 -2,000 -1,734 -3,000 q1 q2 q3 q4 q5 (lowest) (highest) Public Expenditure Household Expenditure, average rent paid Net benefit Source: World Bank staff estimates based on HIES 2013/14. Connective infrastructure Access to basic infrastructure in Fiji is good compared to that of its peers. The majority of the population has access to improved water and sanitation (see Figure 5.2). A relatively large share of the population also has access to the Internet. Far fewer Fijians, however, have access to electricity, reflecting the challenge of difficult terrain, sparsely populated areas, and widely scattered islands. Global evidence on social mobility shows that less connected municipalities and islands are often those with the highest levels of poverty. This was true in Fiji for the period 2002–08 when households living in urban areas were much more likely to have escaped poverty. During 2008–13, perhaps related to improving conditions in rural areas (or a deteriorating situation in urban areas), escaping poverty was more likely in rural areas.16 There are specific challenges of connectivity in more isolated rural areas. The analysis of the Human Opportunity Index shows that while inequality in access to improved water and sanitation across groups is small, the remaining inequality is explained by the households’ location (Figure 5.6). In access to electricity, the rural-urban divide is particularly acute, but household income also has a larger contribution to inequality, perhaps because wealthier households can afford generators in locations that lack electrical connections. Finally, access to the Internet is unevenly distributed across the country. 16 One caveat of this analysis is that it was assumed that households’ characteristics stay the same across the synthetic panel, which also included urban rural residence. If the households in question have migrated, this would show up as upward mobility within rural area rather than as a result of migration. 73 Almost no households have access to the Internet in the Eastern Division or the Northern Division. The high costs of services such as broadband may also be putting the bottom 40 percent at a disadvantage. With donor support, the government has rapidly increased investment in connective infrastructure. Capital transfers to the Fiji Road Authority and the Fiji Electricity Authority more than doubled between 2010 and 2016. Fiji has also invested in the trans- Pacific ICT cable connection. The efficiency of spending could be improved. With regards to roads, the backlog of investment is large, and the government has not made adequate provision for maintenances, repeating past mistakes. The execution rate is also low, in part because of bad weather. With regard to electricity, the government has sought to improve services by arranging private participation in the Fiji Electricity Authority for some time now, but it has not yet succeeded, in part because of lack of agreement on community service obligations. To boost ICT investment and expand related services and employment, the government has provided tax and customs incentives, including the development of free trade zones. There is scope for the private sector to play a greater role in infrastructure investment. Achieving this will require updating and strengthening the public-private partnership framework, formulating regulatory standards, reviewing the price-setting function of the Commerce Commission, reviewing and strengthening the competition policy framework, and building regulatory capacity in government agencies. Financial inclusion Fiji performs well in financial inclusion of the bottom 40 percent. It has the largest and most developed financial market among the Pacific Islands. There are 12 commercial bank branches per 100,000 households, which is above average in the East Asia and Pacific region but less than in Samoa or Tonga. The vast majority of the adult population report using financial services, and most of them use formal financial services. About 50 percent of the bottom 40 percent use formal financial services (Figure 5.11). This achievement is in part a result of government policy. Since 2011, the government has started transferring social welfare benefit payments to approximately 22,000 beneficiaries using electronic transfers to no-fee bank accounts. In addition to reducing processing costs, duplicates, and fraudulent claims, the initiative gave free bank accounts to previously unbanked welfare recipients. 74 Figure 5.11 Access to financial services by bottom 40 percent Fiji 43% 6% 12% 39% Samoa 21% 10% 17% 52% Solomon Islands 10%5% 39% 46% 0% 20% 40% 60% 80% 100% Banked Other formal Informal only Excluded Source: Reserve Bank of Fiji (2016). The challenge is to bridge the geographic and gender divides. About 40 percent of the population in the Eastern Division do not have access to formal financial services (Table 5.2). About 12 percent of rural administrative units do not have any formal financial access point. Gender disparities are also significant in Fiji: women are 16 percentage points less likely to have a bank account than men (Figure 5.12). Table 5.2 Financial inclusion by division Percent of population Central Western Northern Eastern Total Banked 67 58 51 41 60 Other formal 4 4 4 2 4 Informal only 5 5 25 16 9 Excluded 23 32 21 40 27 Total 100 100 100 100 100 Source: Reserve Bank of Fiji (2016). 75 Figure 5.12 Account at a financial institution by gender Percent of population aged 15+ 60% Total 39% 26% 68% Male 38% 32% 52% Female 40% 20% 0% 20% 40% 60% 80% Fiji Samoa Solomon Islands Source: Pacific Financial Inclusion Program (2016). The government invests strongly in financial literacy. Recognizing the importance of financial literacy to make progress on financial inclusion, since 2011, the government has implemented a number of initiatives to strengthen financial literacy among children and adults. These include the integration of financial education in the school curriculum from class 1 to form 6; the launch of financial literacy radio programs in English, Fijian, and Hindi; and adult literacy outreach programs to rural communities provided by licensed financial institutions. Reducing the costs of sending and receiving remittances could make a big difference to the bottom 40 percent. Remittances are important for entire distribution of income, but they account for a greater share of income for poorer households, especially in rural areas (Chapter 1). The costs of international remittances continue to be high (Box 5.3) and could rise because of tighter regulations against money laundering and the financing of terrorism (Alwazir et al. 2017). Domestic remittances require a convenient, secure mechanism that accommodates high-frequency, low-value transfers. At present, about 51 percent of people surveyed use the post office for domestic transfers, even though this is more expensive than using mobile money, which is used by only 3 percent of the people surveyed. Improving access to the Internet in rural areas could reduce the cost of receiving remittances. Box 5.3 Costs of remittances in Fiji Personal foreign remittances are a significant source of support for Fijians, and mostly originate from Australia, New Zealand, the United States, and Fijians serving as United Nations peacekeepers. This official figure may underestimate the actual volume as it is still common for Fijians to carry money home personally or to send it through friends and relatives. Official personal foreign remittances equate to nearly F$250 per person a year or F$100 a month per household. There are a number of remittance providers, including all the major banks and international remittances companies, such as MoneyGram and Western Union. The fees charged by these providers range from 9 to 35 percent, with an average of close to 18 percent. It can take anywhere from one hour to five days to receive remittances and there is no apparent 76 link between the cost and the speed of service. Low-volume transfer business is dominated by Post Fiji, which transfers approximately F$5 million per month domestically. The cost of a Post Fiji transfer starts at F$3 and increases according to the amount transferred. It is likely that the average cost falls between 3 and 7 percent of the value of the transfer. Source: Pacific Financial Inclusion Program (2009). Gender inequality and violence against women Women in Fiji have unequal access to economic opportunities and public services, and they suffer violence and abuse (Box 5.4). According to the survey by the Fiji Women’s Crisis Centre in 2011, 64 percent of women who have been involved in an intimate relationship reported experiencing physical or sexual violence by their husband or partner. The rate of non-partner violence against women or girls is also high, at 31 percent. A study found that 43 women are injured every day as a result of domestic violence, 1 of whom will become permanently disabled, 10 will lose consciousness, and 16 will need medical attention. According to one estimate, the direct and indirect costs of violence against women in Fiji is F$136 million, or 7 percent of GDP (DFAT 2015). Several services protect and support women who are subject to abuse, discrimination, or violence. These are primarily coordinated by the Fiji Women’s Crisis Center, which for nearly 30 years has been providing crisis counselling and referrals to legal, medical, and support services for women and children who are survivors of violence. Crisis and emergency shelters are concentrated in urban centers and are often run by civil service organizations, such as Medical Services Pacific and Empower Pacific, or faith-based organizations, such as the Salvation Army and Pacific Counselling and Social Services. Relevant public health services are available in some urban areas. But women in rural areas have limited access to these services. Women living in these areas, such as the Eastern Division, continue to have limited to relevant services. For example, police posts are harder to reach. Access to the formal justice system is also more difficult, so rural women frequently rely on traditional justice, which promotes reconciliation rather than protection or punishment. Women in rural areas are also less aware of the services that are available, and how to access them. However, there have been efforts to train rural health workers and police to provide coordinated services to victims of gender based violence. 77 Box 5.4 Uneven progress on gender equity The Gender Inequality Index of the United Nations Development Program (UNDP) reflects gender-based inequalities in three dimensions: reproductive health, empowerment, and economic activity. Fiji scores 0.418 in the 2014 index and ranks 87 of 188 countries, better, for example, Samoa (97) and Tonga (148). According to the World Economic Forum (2015), Fiji scores 0.65 in Gender Gap Index and ranks 121 of 145 countries. Its ranking has been declining since 2009. In terms of the sub-indexes, Fiji ranks the lowest (129) in women’s economic participation and opportunity. Reproductive health. According to the 2015 Pacific Regional MDGs Tracking Report, maternal mortality rates in Fiji are among the lowest in the Pacific at 19.1 per 100,000 live births. Almost 100 percent of deliveries in Fiji are attended by skilled attendants, reflecting broad coverage of public health services. Women in rural areas and outer islands are more isolated from maternal and infant health services, but the Ministry of Women, Children and Poverty Alleviation, has a food voucher program to encourage rural women to attend rural health clinics regularly for prenatal and postnatal assessments. Fiji’s total fertility rate declined from 3.1 in 2000 to 2.6 in 2013. However, the use of contraceptives was low at 38.4 percent in 2013. The fertility rate for women aged 15 –19 was 44 per 1,000 live births in 2013, suggesting that teenage pregnancy is a significant issue. Health. Women in Fiji have a longer life expectancy than men (73 years compared with 67 years), but both groups are increasingly susceptible to NCDs and related disabilities. According to the World Health Organization, obesity is a common issue: 41 percent of females were categorized as being obese as compared to 20 percent of men. The proportion of deaths among those aged 15–49 years due to NCDs increased for women from 65 percent to 72 percent between 1995 and 2010, while it remained the same for men, at 64 percent. Education. Fiji has effectively achieved gender parity in education at the primary level. In 2014, the net primary enrolment rate was 97 and 96 percent for boys and girls, respectively. Meanwhile, at the secondary level, the net enrolment rate is higher for females, at 88 percent, relative to 79 percent for males. Policies and programs to address literacy and gender disparity are specified in various policy documents of the Ministry of Education. The Ministry of Education’s Sector Strategic Development Plan 2012–2014 also targets an increase women’s leadership positions in schools. Labor force participation. Fiji has a relatively low rate of female labor force participation for its income. It is 47 percent, compared to 82 percent for men (ADB 2015). However, Fiji has relatively generous maternity leave. It is the only Pacific Island country offering three months’ leave. Women’s earnings remain significantly lower than men’s. Young women, in particular, have a high unemployment rate. Women make up 74 percent of unpaid home workers, and at least according to a 2005 analysis, work 26- 31 percent more hours than men in total due to their household responsibilities, despite working fewer hours in paid employment. The legal marital property regime in Fiji does include a recognition of non- monetary contributions to the household. Gender based violence. According to a survey conducted by the Fiji Women’s Crisis Centre in 2011, of women who have been involved in intimate relationships, 64 percent reported experiencing physical or sexual violence by a husband or an intimate partner. The rate of non-partner violence against women or girls is also high: 31 percent reported any experience of physical or sexual violence since the age of 15 by someone other than intimate partners or husbands. Voice and agency. Women’s participation in politics has been increasing. The Political Parties decree of 2013 required 5 percent of candidates to be women. In 2014, out of the 248 candidates approved by the Fijian Elections Office, 41 were women, representing 16 percent, an increase of 4 percentage points over the 2006 elections. After the 2014 elections, women made up 8 of 50 Members of Parliament. As a proportion of the total, this is an improvement on 2006, when they made up 8 of 71. The new Speaker of Parliament is a woman, and there are two female Ministers in the current cabinet. 78 Policies to improve access to services for all The several opportunities for improving access to services. In education, the government could help maintain quality by targeting spending to rural and remote schools that are costly to run. It could also continue to address the dropout rates in secondary education by continuing with the Matua program, which helps people who have dropped out of school to return, by expanding it to strategically located and relevant secondary schools. This matters because of the importance of education in employment outcomes. In health, the government could continue to explore opportunities to involve the private sector. The governments of many developing countries have implemented or are considering policies of contracting-out health services to the private sector, with mixed results. The proponents of this trend argue that the private sector has been more efficient and responsive to patient needs because of market competition. Others, however, have highlighted inequalities in access to health care resulting from the inability of the poor to pay for private services. The government could further investigate the cost of hospital service delivery under private and public provision and carefully monitor the results of its planned project to introduce private management of a hospital, particularly the fiscal costs of subsidizing services for poor patients. In infrastructure, there is also scope for the private sector to play a greater role. The government has already pursued private participation in ports and electricity. Continued public investment and more private participation in infrastructure provision could increase investment and economy-wide competitiveness. Encouraging greater private sector participation in infrastructure will require updating and strengthening the framework for public-private partnerships, formulating new regulatory standards, reviewing the price- setting function of the Commerce Commission, reviewing and strengthening competition policy, and building regulatory capacity within relevant government agencies. In housing, the government could improve access to housing in three ways. To reduce building costs, the housing authorities could pilot smaller, standardized houses that could be produced on a large scale with prefabricated materials through a public-private partnership. To extend mortgage services to low-income households, the government could work with prospective mortgage lenders to better target mortgage-linked subsidies. And to increase the supply of affordable rental housing, the Public Rental Board could shift from the building of rental units to management and maintenance of the units on behalf of private investors, while the government could provide capital subsidies to encourage the private supply. In gender-based violence, effective protection of women requires greater outreach and stronger awareness programs. This would reinforce women’s rights to get help and stop the violence. Frontline service providers need to be effectively trained to respond 79 sensitively so that women are encouraged to seek help. Emergency accommodation for women escaping violence needs to be expanded, in urban and, especially, rural areas. Frontline service providers should be better resourced to increase capacity. 80 6. Pathway III. Building resilience Climate resilience Fiji depends on natural resources for its critical industries. Tourism is the single largest industry in Fiji, accounting for 38 percent of GDP and 48 percent of total exports. While agriculture has declined in importance, subsistence agriculture, forestry, and fisheries remain as important sources of rural livelihood. Ensuring environmental sustainability is therefore important for inclusive growth. Climate change poses a major threat to sustainability of development. Natural disasters are the most tangible aspect of climate change. Each year, natural disasters cause an average loss of about 1.8 percent of GDP (EM-DAT 2016). With climate change, the annual average loss from cyclone alone is expected to increase to 2.5 percent of GDP and as much as a loss of 23 percent every 50 years (PCRAFI 2015). These losses add to fiscal pressures and constrain wealth accumulation, lowering potential growth. A study of the effects of 6,700 cyclones on long run economic growth found that they permanently lower income relative to the pre cyclone trend (Hsiang and Jina 2014). Natural disasters also harm shared prosperity because they hit hardest the bottom 40 percent, who tend to have less diversified assets, less secure jobs, and less access to insurance. How resilient are Fiji’s cities? Largely unplanned urbanization has exacerbated environmental challenges and the impact of disasters. Between the 1996 and 2014 censuses, the average annual growth rate of the urban population was 1.1 percent a year, compared to 0.1 percent for the rural population or 0.5 percent for the population as a whole. Limited urban planning and high poverty in urban areas has caused around 16 percent of urban dwellers to live in squatter settlements. In addition, limited enforcement of land use regulations and poor construction practices have led to the proliferation of settlements in increasingly steep slopes or along river protection areas in the peripheries of the main cities. The concentration of population and critical infrastructure in cities also increases the concentration of risk (Figure 6.1). Exposure is highest in Suva, at US$3.6 billion, Naitasiri, at US$2.7 billion, and Nadi, at US$ 2.3 billion. As a result of concentrated infrastructure and investments, the average annual economic losses due to cyclones are also highest in these areas (Figure 6.2). Suva had the highest average annual losses at US$9.6 million, while Naitasiri and Nadi had average annual losses of US$5.4 million and US$5.9 million, respectively. 81 Figure 6.1 Building exposure in Fiji (in 2010 US$) Figure 6.2 Average annual losses due to cyclones and earthquakes (in 2010 US$) Source: World Bank staff estimates based on. http://pcrafi.sopac.org/ 82 Only a small portion of disasters' economic losses are covered by insurance. Public infrastructure has limited insurance coverage, leading to delayed reconstruction efforts after disasters. For the private infrastructure, the commercial sector is said to be better insured. Residential property, however, is largely uninsured, possibly because of a lack of understanding or an expectation of government assistance in case of natural disasters. Moreover, many dwellings are not compliant with the building code, which prevents them from getting insurance. Because of expensive construction costs, people choose to build their dwellings themselves or to hire cheaper, unlicensed builders. Squatter settlements are excluded from insurance for the same reasons. As a result, most losses are absorbed by the government which predominantly finances post-disaster losses by increasing public debt. With tighter fiscal situations, this approach may no longer be sustainable in the future (see below). Building urban resilience requires additional government actions in three key areas. • Infrastructure. To insure against disaster risks, the government needs first to assess the disaster resilience of critical public infrastructure, develop upgrade plans, and cost them. Planned infrastructure investments and reconstruction projects should also aim to incorporate resilience and adaptation in the design. To expand private insurance, the government needs to encourage compliance to building standards and enforce appropriate land use. It needs also to continue explore ways to expand appropriate risk insurance products, for example, by subsidizing premiums for low income households or financing universal coverage financed by special taxes. • Urban planning. Modern urban planning is missing. The urban plan for the Suva, for example, was developed in 1978. The Fiji National Building Code was first commissioned in 1990, and there have been no significant updates since then. Fiji has long pursued ad hoc policies and initiatives to address post-disaster recovery for buildings, rather than a comprehensive update of building codes as a preventative measure. • Governance. Planning authority is fragmented across a number of agencies with varying jurisdictions and levels of authority, and a platform for coordination is lacking (Table 6.1). This leads not only to uncoordinated and unintegrated planning but also to poor maintenance and service delivery. Community response is not well utilized, either. There is strong capacity at community levels to respond to natural disasters (People’s Community Network and UN Habitat, 2016). Even within squatter settlements, there is generally some form of active community organization that can mobilize to meet needs during emergencies. However, community responses are usually weaker and insufficient in larger towns and cities. For example, In Lami (part of Greater Suva) there is no formal evacuation strategy. In February, when there was a tsunami warning in Suva, workers and residents took to cars to evacuate, causing massive traffic jams, even along the coast. 83 Table 6.1 Planning authorities Authorities Main roles Environment, Infrastructure and Coordinates urban planning, management and growth in urban Transport areas, and oversees the squatter settlement upgrading programs. Fiji Roads Authority, Water Authority of Main government agencies responsible for the management of Fiji, and Fiji Electricity Authority national and municipal infrastructure. Town councils Prepares planning but their authority usually does not extend to peri- urban areas outside of defined town boundaries. Local area advisory authorities Governs building and development of peri-urban areas. Central Board of Health Overseas the local area advisory authorities in their tasks. iTaukei Lands Trust Board Manages and administers native land holdings. Source: World Bank staff report. How resilient are Fiji’s rural areas? Fiji depends on natural resources to a greater extent than many other middle income countries. Tourism contributes 38 percent of GDP and depends on tropical rainforests, white sandy beaches, coral reefs, marine life, and clear blue waters. While agriculture has declined in importance, it contributes about 8.7 percent of GDP; forestry and fisheries contribute an additional 1.5 percent of GDP; and mining contributes about 0.8 percent of GDP. In total, over 50 percent of GDP is dependent on natural resources. Natural resources are important for the livelihood of the rural population. Even with increasing urbanization, about 48 percent of the population still lives in rural areas, and 38 percent of them live below the basic needs poverty line (Chapter 1). The rural bottom 40 percent derive 40 percent of their incomes from subsistence or commercial agriculture (Figure 6.3 and Figure 6.4). Even for the higher quintiles, subsistence agriculture remains an important source of livelihood, contributing around 10–15 percent of income (Figure 6.3). Figure 6.3 Sources of rural household Figure 6.4 Breakdown of agribusiness income incomes 100 100 7 4 4 4 2 5 3 3 4 5 80 44 39 39 36 36 80 60 60 66 64 59 56 18 21 23 66 15 29 40 40 23 25 25 26 21 29 20 24 20 15 17 18 10 17 16 15 10 12 12 0 0 11 11 8 1 2 3 4 5 1 2 3 4 5 (Lowest) (Highest) (Lowest) (Highest) Subsistence Agribusiness Fisheries Sugar Other crops Wages and salaries Other income Livestock Handicrafts Source: World Bank staff estimates based on HIES 2013/14. 84 In addition to contributing directly to rural livelihood, the natural resources increase the resilience of the rural community. Healthy coastal and nearshore marine ecosystems such as mangroves and coral reefs provide habitats for fish and protect rural communities from cyclones. Forests contribute to agriculture by controlling erosion and the flow of water and thus mitigate damage from floods. However, the use of some of these resources is not sustainable. Coastal fisheries are poorly regulated and overexploited. Illegal fishing is prominent in the high value coastal fisheries, such as bêche de mer. Given projected population growth and the decline of the coastal fisheries due to overfishing, it is likely that Fiji will not produce enough fish to meet its needs in the future. Native forests are being depleted because of deforestation driven primarily by conversion to commercial agricultural crops such as taro, kava, and ginger. They are also being degraded by unregulated logging and to a lesser extent by forest fires, firewood collection, and the spread of invasive species. Building rural resilience requires that the government develop an integrated set of policies to address the drivers of natural resource degradation. This requires (i) reviewing the consistency of sectoral policies, notably in agriculture, mining, tourism, and fisheries, to eliminate perverse incentives that lead to conversion or degradation of forests, including mangroves; (ii) strengthening the enforcement of regulations, such as the forest harvesting code; and promoting good agricultural practices to guide agricultural intensification; and (iii) expanding afforestation and reforestation programs. To safeguard coastal fisheries, the government needs to (iv) establish and enforce a community-based fishery management system, including for high-value nearshore resources such as bêche de mer. The focus on management of aquaculture needs to shift towards the control of environmental impacts rather that the promotion of expansion. The government should also invest in a system to effectively monitor trends. Fiscal resilience To rise to the development challenges identified and protect the vulnerable from frequent shocks, the government must build fiscal space. In recent years, however, rising expenditure and frequent natural disasters have eroded fiscal space. Resources available to the government are relatively high. Tax revenue is about 23 percent of GDP, higher than in peer countries (Figure 6.5). Nevertheless, the fiscal deficit has increased and averaged 4.5 percent of GDP in the last three years (Figure 6.6). 85 Figure 6.5 Tax revenue, average 2010-14 Figure 6.6 General government balance Percent of GDP Percent of GDP -0.5 -1.1 -2.2 23 -3.5 17 -4.2 -4.6 -4.7 14 13 11 -5.6 -5.9 Fiji Small World Upper East 2010 2011 2012 2013 2014 2015 2016p 2017p 2018p states middle Asia & income Pacific Source: World Bank staff estimates based on WDI and Government of Fiji Budget Estimates. Between 2011 and 2015, public expenditure doubled in nominal terms, from F$1.7 billion to F$3.3 billion, or from 27 percent to 35 percent of GDP (Figure 6.7). The increase was led by capital spending on infrastructure, which rose six-fold in nominal terms, followed by current spending on education and health, which rose by 164 and 148 percent, respectively—consistent with the government’s development strategy. Subsidies and transfers also expanded as a share of current expenditure, from around 21 percent to 23 percent (Figure 6.8). Figure 6.7 Evolution of spending Figure 6.8 Operating spending Percent of GDP Percent of Operating expenditure 21 90 21 19 18 20 17.0 18.0 17.7 15.2 14.1 20 20 21 70 18.6 2.2 2.1 21 2.6 2.7 2.5 2.8 20.6 23.1 23.1 21.4 22.0 23.5 50 14 14 15 11 12 30 44.1 42.0 41.0 44.1 44.6 7 8 8 40.7 5 10 2010 2011 2012 2013 2014 2015 2016p 2017p 2018p -10 2010 2011 2012 2013 2014 2015 Interest Pensions Operating expenditure Capital expenditure Subsidies & Transfers Wages & Salaries Source: Government of Fiji Budget Estimates. 86 Meanwhile, rigidities in expenditure have increased, which exacerbates the impact of external shocks. About 84 percent of current spending is in difficult to cut categories, such as interest, pensions, subsidies and transfers, and salaries and wage (Figure 6.8). Moreover, civil service compensation reviews, which are based on comparisons of civil servants’ salaries with private sector salaries, are expected to result in a large increase in civil service pay in 2017/18. Together with likely increases in spending pressures as the country prepares for an election in 2018, planned fiscal consolidation will be difficult. Building fiscal resilience also requires keeping government debt at a sustainable level. The debt was 46 percent of GDP at end-2016. While there is no immediate risk to debt sustainability, the recent increases in debt mean that measures need to be taken to keep it on a sustainable trajectory. To reduce currency risk, the government has largely relied on issuing domestic bonds. However, interest rates on domestic bonds are relatively expensive, and the Fiji National Provident Fund (FNPF) remains the main buyer, holding over 60 percent (Box 6.1). There is no secondary market for domestic government bonds, making price discovery harder. External borrowing represents a small share of the total and is largely tied to capital spending. The average cost of the external financing is increased by the cost of a global bond maturing in 2020—the only commercial external debt—which has a coupon rate of 6.7 percent, while multilateral and bilateral loans have much lower rates and those from the Exim Bank of China are fixed at 2 percent. The government plans to keep debt below 50 percent of GDP and to reduce exposure to exchange rate shocks by expanding domestic sources of funding, with a target ratio of domestic to external debt of 70:30. Box 6.1 The Fiji National Provident Fund The FNPF is the largest investor in Fiji and also the main holder of domestic government bonds. It is the only retirement fund that is mandated by law to collect pension contributions in both the public and private sectors. Contributions are 16 percent of salaries, with both employee and employer contributing 8 percent. The target investment return to the members is currently 8 percent a year. The board members of the FNPF are appointed by the Minister of Economy (Fiji National Provident Fund Decree 2011). In 2016, its portfolio size was about F$5 billion. It had a net contribution inflow of F$170 million and investment income of F$330 million (FNPF Annual Report 2016). The FNPF considers itself too large to invest only in Fiji and has been gradually diversifying internationally. In practice, its investment opportunities are limited as there are tight restrictions on overseas investment (IMF 2014). There are some direct commercial lending operations as well as substantial equity investment from the government’s privatization programs. The FNPF has held up to 80 percent of the government’s domestic bonds, but this has dropped to 60 percent, because of competing bids from insurance companies and, to a lesser extent, banks. The liquidity stemming from the easy monetary stance has resulted in lower interest rates in the past several years, with the 15-year bond recently being issued at around 7 percent. The FNPF also holds about a third of the government’s international bond issued in 2015 at 6.6 percent. 87 Keeping debt on a sustainable path requires fiscal adjustment. A fiscal sustainability analysis, focusing on the next five years, suggests that if current spending and growth trends were to continue, the deficit would widen to 6.2 percent of GDP and government debt would rise to 63.5 percent of GDP by 2021. If debt is to be kept below 50 percent of GDP, the government will have to make a significant fiscal adjustment in the next five years.17 Financing risks are expected to peak in 2020 due to the maturing of the US$200 million global bond. Building fiscal resilience also requires that the government manage contingent liabilities. One important source of such liabilities is the government’s portfolio of SOEs (Table 6.2). The total guaranteed debt of the SOEs (explicit contingent liabilities) was estimated at F$825 million, equivalent to 9 percent of GDP. The total contingent liabilities (explicit and implicit) of the government were estimated at F$2.8 billion, or 31 percent of GDP at the end of 2015 (2016/17 Budget Supplement), mostly represented by the pension liabilities of the FNPF. 17 The analysis draws on the standard intertemporal fiscal sustainability framework (e.g., Ley and Tran 2009). 88 Table 6.2 State owned enterprises, 2015 Notes Ownership Guarantee (June (%) 2014) F$ million Wholly owned commercial companies Airports Fiji Airports on remote islands subsidized by 100 government Fiji Broadcasting Two of six radio stations produced under 100 20.1 contract with government Fiji Development Bank Development bank 100 174.3 Fiji Hardwood Mahogany forestry 100 14.0 Fiji Public Trustee Estate and trust manager 100 Food Processors Food exporter 100 Post Fiji Has universal service obligation 100 Rewa Rice Rice miller and marketer Unit Trust of Fiji Intended to promote capital markets 100 Viti Corp Business training 100 Yaqara Pastoral Cattle breeding and farming 100 Commercial statutory authorities Biosecurity Authority 100 Civil Aviation Authority 100 Fiji Electricity Provides subsidized power to rural and 100 338.4 Authority remote areas Fiji Meats Industry Board Housing Authority 100 79.8 Public Rental Board 100 Water Authority of Fiji 100 Majority owned Companies Air Pacific Owns main airline, Fiji Airways 51 Air Terminal Services 51 Copra Millers Producer of coconut oil and copra meal 96 Fiji Pine Sawmill and chip plant operator 99.8 4.5 Fiji Ports Fiji National Provident Fund also owns 41 16.5 39% Fiji Sugar Corporation Monopoly sugar miller 68 156.2 Pacific Fishing Cannery 99.8 3.2 Company Minority owned companies Amalgamated Telecom Owns only landline company and one of 34.6 Holdings two mobile companies Pacific Forum Line Commercial shipping 23 Sources: Ministry of Finance, Economic and Fiscal Update: Supplement to the 2015 Budget Address (Tables 5.1 and 5.11), November 2014. (Note: the Economic and Fiscal Update for the 2016 budget does not update this information.); WTO (2016); websites of Fiji Ports. 89 To build resilience, the government must also manage catastrophic fiscal risks related to natural disasters. It is possible to make a rough estimate of these risks using the previously mentioned data on possible future GDP losses from cyclones (PCRAFI 2015) and adding a stochastic simulation to the fiscal sustainability analysis. 18 Using this approach, the fan chart in Figure 6.9 shows percentiles of the probability distribution of projected government debt. The 50th percentile forecast is depicted by the light-colored dotted line. In 2018, for instance, the 50th percentile value corresponds to a ratio of public debt to GDP of 63 percent. This means that there is a 50 percent chance that public debt is higher than 63 percent of GDP, instead of the 55 percent projected under the baseline of no natural disaster. The 75th percentile value suggests that there is a 25 percent chance that debt will be more than 66 percent of GDP. With the possibility of more than one natural disaster hitting the country, there is a 25 percent chance that debt will be higher than 97 percent of GDP by 2021. Figure 6.9 Stochastic projections of the deficit and debt, 2016–21 Deficit to GDP Debt to GDP 0% 120% 99% -5% 1% 95% 5% 100% 90% 10% 75% 25% -10% 50% 50% 80% 25% 75% 10% 90% 5% -15% 95% 60% 1% 99% 0% -20% 40% 2016 2017 2018 2019 2020 2021 2016 2017 2018 2019 2020 2021 Source: World Bank staff estimates. Several instruments area available for managing catastrophic risks. Figure 6.10 identifies some of them and indicates the speed with which they provide funds. Donor support post‐disaster is uncertain and depends on the generosity of the international community. The government can set aside reserves, but their amount is limited by other pressing needs. Contingent credit can provide governments with additional financial capacity in the aftermath of a disaster, but its amount is constrained by the borrowing capacity of the country. The costs of instruments also vary. Grants do not have a financial 18 The stochastic simulation creates 1,000 alternative scenarios for the deficit and debt based on random (“stochastic”) shocks to the GDP growth rate generated by natural disasters. To calibrate the shock distribution, we used the PCRAFI (2015, Annex 4, p.5, Table 2) estimates for cyclone risk and associated costs and modeled them using a lognormal distribution to reflect the main features of the implied distribution (e.g., a sharp lower-bound of zero, no sharp upper-bound, a single mode, and a positive skew). The analysis assumes that the realizations of natural disasters are independent over time. 90 cost but are often reallocated from other projects and may have an opportunity cost. Reserves are generally held in short‐term assets; their cost is the difference between the returns on long-term investments and the returns on short‐term investments. Unless they affect the credit rating of a government, the cost of emergency loans is reflected in the interest rate applied. Finally, governments have recently taken a closer look at instruments available in the financial markets such as traditional insurance, parametric insurance (e.g., catastrophic bonds). Some countries, such as Mexico, have recently ventured into the use of CAT Bonds to cover specific needs. Nevertheless, the use of parametric insurance remains a relatively expensive proposition for governments, and their use has remained limited. Figure 6.10 Sources of post disaster financing Relief phase Recovery phase Reconsruction phase (1-3 months) (3-9 months) (over 9 months) Post-disaster financing Donor assistance (relief) Budget reallocation Domestic credit External credit Donor assistance (reconstruction) Tax increase Ex-ante financing Budget contingencies Reserve fund Contingent debt facility Parametric industry CAT-Bonds Traditional insurance Source: World Bank (2015). Protecting the vulnerable Fiji has a relatively well developed social protection system. It consists of four main programs: Poverty Benefit Scheme, which is targeted at the bottom 10 percent; the Care and Protection Allowance, which is targeted at households with vulnerable children; the Social Pension Scheme, which is targeted at people over 65 who do not have public pensions; and food vouchers, which are provided to the recipients of the preceding programs. In terms of both coverage and adequacy, Fiji’s cash transfer programs performs reasonably well compared to those of other countries in the region (Figure 6.11 and Figure 6.12). 91 Figure 6.11 Coverage of cash transfers in selected countries Percent of population covered Sri Lanka Vietnam Pakistan Timor-Leste 8% Fiji 6% PNG 3% Thailand Nepal Tonga 0.5% 0% 5% 10% 15% 20% Source: World Bank Atlas of Social Protection Indicators of Resilience and Equity 2016. Figure 6.12 Adequacy of cash transfers in selected countries Benefits as percent of household consumption Mongolia Pakistan Fiji 6% Sri Lanka Vietnam Philippines Malaysia Nepal PNG 0% 0% 2% 4% 6% 8% 10% Source: World Bank Atlas of Social Protection Indicators of Resilience and Equity 2016. In the aftermath of Cyclone Winston, the government implemented its first climate contingent social protection program. Recognizing that natural disasters can harm poor families and reverse gains in poverty reduction, the government decided to give top-up payments to the beneficiaries of the main social protection programs. The 23,000 or so households covered by the Poverty Benefit Scheme were paid a lump sum of F$600; the 3,300 households covered by the Care and Protection Allowance and the 17,200 pensioners covered by the Social Pension Scheme received F$300. This was equivalent to three months of regular benefits. The total cost of the top-up payments was F$345 million. Prioritizing speed and efficiency, the government gave top-up payments to all beneficiaries whether or not they were in the cyclone-affected areas. Channeling the assistance through the existing safety net meant that it went to the poor and reached them efficiently. Three months after the cyclone, with the help of the World Bank, the government evaluated the impact of this initiative. The evaluation found a number of positive results. 92 First, the benefits reached the recipients in time to meet urgent needs: the top-up transfers were made one month after the cyclone hit and, within four weeks, the majority of the recipients had spent them. Second, most of the benefits were spent on essential items, with food and materials to repair dwelling being the most common items (Figure 6.13). Less than 1 percent was spent on kava, alcohol, or cigarettes. Third, the recipients recovered faster than those who are just above the benefit thresholds and who therefore did not receive social assistance. Three months after the cyclone, the beneficiaries of the top ups under the Poverty Benefit Scheme were more likely to report having recovered from the cyclone.19 Figure 6.13 Expenditure of Cyclone Winston top-up payments Percent of top-up assistance spent Food Repair Clothing/school supplies Medical Household goods/furniture/products Repay debts/saving/contribution Kava/alcohol/cigarettes 0% 10% 20% 30% Source: Fiji Bureau of Statistics; World Bank Cyclone Winston Impact Evaluation. More generally, the experience with the post-Winston interventions highlights a few key lessons. The first is the importance of savings as a buffer against shocks. Nine days after Cyclone Winston, the government allowed affected members of the FNPF to withdraw F$1,000 quickly and up to F$5,000 for approved housing repairs. This was subsequently extended to members who were not directly affected by the cyclone but who wished to help their family in affected areas. The facility was open for 60 days to the end of April 2016. More than 9,000 members were said to have withdrawn money during this period, and the total payments to have been F$260 million. Even among the low income households that were part of the impact evaluation survey, 15 percent reported benefitting from the special withdrawal allowance. This was by far the largest assistance the government managed to give affected households (Table 6.3). 19 The evaluation strategy used the regression discontinuity design in defining the treatment and control groups based on the Poverty Benefit Scheme eligibility (poverty score) threshold. The treatment group is the Poverty Benefit Scheme recipient households (25 percent below threshold) in affected areas that have also received the top-up benefit. The control group is households in affected areas that are not eligible for PBS, as they are above (but within 25 percent) the threshold, and therefore did not receive the top-up benefit.) 93 Table 6.3 Major interventions after Cyclone Winston Program Beneficiary households Duration Budget ($ million) Social welfare top-up payments 43,897 March-May 2016 19.9 Food Voucher Program 44,169 May-June 2016 4.6 Housing Program 24,664 April 2016- 87.0 Fiji National Provident Fund 170,000 March-April 2016 250.2 Total 344.7 Source: ADB (2016). The experience also reveals that ad hoc initiatives put in place post disaster can encounter unexpected difficulties. For example, the Home for Help initiative gave families with an annual income of less than F$50,000 grants of F$1,500 to repair partially damaged homes, F$3,000 for seriously damaged homes, and F$7,000 for destroyed homes. The initiative was expected to cost F$184 million by 2018, and the government has spent F$87 million in the first year. The initiative took longer than expected to be implemented. Eligible beneficiaries were required to enroll with the Ministry of Welfare and to provide a valid ID and other documents to confirm home location and eligibility for the program. Beneficiaries received e-cards issued by a mobile phone operator, preloaded with the approved funds, which could be used at any of ten hardware retailers to order from a list of pre-approved materials. Delivery of building materials, particularly to maritime areas, was subject to unexpected delayed. Challenges included shortage of building materials, difficulties delivering materials to maritime and remote areas, the reluctance of beneficiaries to accept partial deliveries, shortage of carpenters and other tradespeople, and inability to contact beneficiaries on their registered mobile phones. Where building materials have been delivered, rebuilding has also proved slow, because of the lack of carpenters and other skilled tradespersons, and because of beneficiaries had to pay labor costs. Policies to build resilience Several things can be done to build resilience. • Climate resilience. To strengthen urban resilience, the government could assess the disaster resilience of critical public infrastructure and cost the needed upgrade. It could also update the urban plan to ensure risk sensitive land use, review the building code to ensure it is appropriate for Fiji, and better enforce both of them. To strengthen rural resilience, the government could review the consistency of sectoral policies in the resource incentive sectors, such as agriculture, mining, tourism, and fisheries. It could strengthen enforcement of existing regulations and safeguards to ensure environmental sustainability, and ensure that zoning for 94 coastal areas, river banks, and wetlands is sensitive to risks. Finally, it could explore ways to engage citizens to improve warnings and responses to disasters. • Fiscal resilience. Improving fiscal resilience involves expanding the tax base by consolidating the extensive system of tax exemptions and incentives. On the expenditure side, it involves better targeting of subsidies to create fiscal space and reduce rigidities. Contingent liabilities need to be monitored and reduced. To insure against catastrophic risk, the government could further explore alternative ex ante financing instruments. • Social protection. Not everyone can be protected ex ante. There must also be in place a system to protect the most vulnerable after disasters. After Cyclone Winston, the government experimented with cash transfers based on the existing social protection system. One of the most successful interventions was allowing people to withdraw some of their savings from the FNPF to rebuild their houses. However, the FNPF only covers the formal sector, so most of the poor were not helped by this measure. 95 7. Priorities Policy priorities This report has identified many challenges and many possible policy solutions. But not everything can be done at once: priorities need to be selected. In selecting the priorities for shared prosperity in Fiji, this SCD uses three main criteria. The first is whether a policy is a precondition for making progress in pursuing others. The second is whether a policy is expected to have positive spillovers across different domains (e.g., growth, equity, resilience). The third is whether a policy is feasible, that is, it can be implemented in the medium term given cost, capacity, and political feasibility for action. The resulting priorities are shown in Table 7.1. The judgements behind the application of these criteria are set out in the narrative below. Table 7.1. Priorities 1. Improving the business • Reduce policy uncertainty environment • Reduce red tape and the complexity of incentives • Create a stronger legal and regulatory framework for private sector participation • Communicate better and consult more with businesses and civil society 2. Investing in urban • Improve the quality of urban housing and essential public resilience infrastructure • Update the urban plan to ensure risk sensitive land use • Update the building code to ensure it is appropriate for Fiji, and enforced • Explore ways to engage citizens to improve warnings and responses to disasters 3. Safeguarding fiscal • Develop medium term fiscal consolidation plan sustainability • Expand revenue base by reviewing the system of tax exemptions and incentives • Reduce rigidities in expenditure by consolidating fragmented subsidy programs • Better monitor and manage contingent liabilities and catastrophic risks 4. Expanding access to quality • Increase spending on health services targeted to the bottom 40 healthcare percent • Make extra effort to serve rural areas and remote islands • Partner with the private sector to serve the poor • Monitor the cost of hospital service delivery under private and public provision 5. Expanding access to • Continue public investment and encourage private participation connective infrastructure in infrastructure to enhance economy-wide competitiveness • Make extra effort to serve rural areas and remote islands 96 1. Improving the business environment. • Precondition. Fiji already has a sizeable private sector, but there is also considerable unexploited potential to expand it. Creating an efficient and up-to-date legal and regulatory framework and providing stable and predictable policy environment are preconditions for private sector growth. • Positive spillovers. Improving the business environment is expected to generate benefits across the three pathways. On growth, a sound legal and regulatory framework and a simpler and more stable tax system would encourage foreign and domestic investment. On equity, strengthening the framework for public-private partnerships, reviewing competition policy, and clarifying the community service obligations of SOEs would encourage private participation in delivery of public services. On resilience, better enforcement of safeguards and reduced fragmentation of policies in resource intensive sectors would improve environmental sustainability. • Feasibility. Improving the business environment is an ongoing effort for any country, including Fiji. Fiji has the capacity to introduce significant reform quickly when necessary and to work with development partners to adopt appropriate policies for the country. The declining Doing Business ranking has received media attention, which creates incentives for policy action. 2. Investing in urban resilience • Precondition. Natural disasters can wipe out years of investments in infrastructure and progress made in poverty reduction. Expansions of squatter settlements have increased disaster risk and potential costs. • Positive spillovers. Investing in urban resilience would have positive spillovers on growth, equity, and resilience. Adaptation of critical public infrastructure, modern urban planning, and risk-based land use would improve investment climate and access to insurance. Improving quality of urban housing would help reduce vulnerability of the urban poor. These improvements would reduce the costs of natural disasters, thereby improving fiscal resilience. • Feasibility. Fiji’s Green Growth Framework 2015 has made building climate resilience a national priority. The World Bank climate vulnerability assessment is expected to provide estimates for cost and capacity requirements, which would help Fiji access external financing and technical support. With frequent cyclones and floods during the rainy season each year, building climate resilience has strong political support across the country. 97 3. Safeguarding fiscal sustainability • Precondition. Without fiscal sustainability, the government would not be able to make investments to accelerate growth, expand and improve public services, and disaster-proof critical infrastructure, or protect the most vulnerable from shocks. • Positive spillovers. Safeguarding fiscal sustainability would have positive impact on growth, equity, and resilience. Expanding the revenue base by reviewing and simplifying the complex system of tax incentives would improve the investment climate. Reducing spending by consolidating the fragmented public subsidy programs would improve targeting of spending to the bottom 40 percent. Additional fiscal space would allow the government to invest in resilience, and respond to disasters with emergency spending and social protection. • Feasibility. After completing the two-year post-Winston reconstruction in this fiscal year, rebuilding fiscal space will be a challenge for the medium term. On revenues, reviewing the current system of investment incentives to ensure value for money would be the first step. On expenditures, ongoing effort in determining the right mix of public support and private initiatives in improving services such as transport, health, and housing would need to continue. The likelihood of actions in both is high, especially with appropriate technical assistance from the development partners. 4. Expanding access to quality healthcare • Precondition. Health is a value in itself, but it also influences economic opportunities and outcomes in terms of productivity, human capital, and employment. • Positive spillovers. Expanding access to quality health care would help Fiji to realize its potential as a health hub for the Pacific and thus attract health tourists and retirees. This would thus generate positive equity and growth impacts. • Feasibility. The government is embarking on the first PPP in healthcare in the coming year. It regards this an important test case for the future and, if it is successful, the government plans to do more. With appropriate technical assistance from the development partners, the likelihood of progress is high. 5. Expanding access to connective infrastructure • Precondition. Expanding connective infrastructure is a precondition for private- sector led growth and reducing inequality in opportunities. Connective infrastructure is also essential to ensure efficient disaster response and post-disaster reconstruction. • Positive spillovers. Continued public investment in connective infrastructure would encourage private investment and enhance economy-wide competitiveness. It 98 would also help address remaining gaps in access between rural and urban areas and reduce inequality of opportunities. • Feasibility. With tighter fiscal space, the government is increasingly exploring ways to improve efficiency through private participation. With appropriate technical assistance from the development partners, the likelihood of policy action is high. Cross-cutting priority: Strengthening institutional capacity of the public sector Some aspects of governance in Fiji are relatively good. For example, on standard global indicators of political stability and control of corruption, Fiji ranks at the 60th percentile or better. Other countries in the region, such as Papua New Guinea and Timor-Leste and China and Thailand do much worse on these indicators (Figure 7.1). Fiji also has a disciplined and well educated civil service. It has shown it can develop and put in place significant reforms quickly when necessary, and work with development partners to adopt appropriate policies for the country. Figure 7.1 Indicators of governance, selected countries, 2015 Percentile rank Political stability Control of corruption 0 20 40 60 80 0 20 40 60 80 Fiji Malaysia Solomon Islands Fiji Malaysia Solomon Islands Vietnam China Timor-Leste Thailand Papua New Guinea Vietnam Indonesia Indonesia China Timor-Leste Thailand Papua New Guinea Regulatory Quality Rule of law 0 20 40 60 80 0 20 40 60 80 Malaysia Malaysia Thailand Thailand Indonesia Vietnam China China Papua New Guinea Indonesia Fiji Fiji Vietnam Solomon Islands Timor-Leste Papua New Guinea Solomon Islands Timor-Leste Voice and accountablity Government Effectiveness 0 20 40 60 80 0 20 40 60 80 100 Solomon Islands Malaysia Indonesia China Timor-Leste Thailand Papua New Guinea Indonesia Fiji Vietnam Malaysia Fiji Thailand Papua New Guinea Vietnam Solomon Islands China Timor-Leste 75th to 90th percentile 50th to 75th percentile 30th to 50th percentile 0th to 30th percentile Source: World Bank World Wide Governance Indicators 2015. 99 As Fiji embarks upon the next phase of growth and development, however, the interplay between policies and governance will become even more important. The institutional environment will determine the direction of policy changes and whether they are properly implemented. Without strengthening of institutional arrangements and capacity, it will be difficult to deliver most of the priorities identified in this report. This section summarizes the four key cross-cutting governance challenges that emerge from the analysis. Legal and regulatory framework Modernization of the legal and regulatory framework is needed to ensure that it encourages investment by the private sector. The current framework does not reassure investors, and it raises costs. For example, a recent review of the investment law and associated legislation has identified a number of departures from international conventions including in areas of arbitration, confiscation, and repatriation of profits. Considerable red tape exists, and time spent dealing with officials is raising costs of doing businesses, especially for small firms. Labor market regulation could be simplified to reduce frictions and mismatch. Expanding private sector participation in public service delivery also requires modernizing the existing legal and regulatory framework. There have already been some achievements in involving the private sector in service delivery, but there is potential for further gains. The existing public-private partnership framework contains a number of provisions limiting foreign investment, and lacks clear guidelines for developing transparent PPP projects. Expanding private participation in public service delivery would also require reviewing the price-setting function of the Commerce Commission, reviewing and strengthening the competition policy framework, and building regulatory capacity in government agencies. Changes are also likely to be required within the institutional framework of government to reflect the evolving role toward a provider of public goods and an enabler of private sector participation. This will require the separation of regulation and service delivery, as well as the development of institutional capacity to monitor and manage the performance of external service providers. In some cases, the government may need to be more willing to give up some direct control. In general, the government is concerned about reducing the control of government agencies over the private sector for fear of losing control over the country’s development path. In the investment law, for example, the government fears that streamlining the foreign investment registration process or relaxing the confiscation clause will increase the potential for fraudulent investors to enter Fiji. It also fears that this could adversely affect both local businesses or employees, and that the public reaction to misdealing by foreign investors will result in the loss of confidence in the government more generally. But more 100 willing to give up some direct control maybe what is needed to encourage greater participation by domestic private sector and attract foreign investment. Building greater resilience will further require better enforcement of existing regulations and updating of the planning, risk monitoring, and management frameworks. Currently, many ministries have limited capacity to enforce regulations. This is a significant risk to government as these regulations are the safety nets that are designed to avoid disasters such as environmental pollution, adherence to building standards, and health and safety standards. For example, the Ministry of Local Government, Housing and Environment is limited in its ability to evaluation environmental impact assessments. Great resilience, including fiscal resilience, also requires enforcement of the existing legislation as well as updating them to meet the new challenges. For example, a significant risk to Fiji’s fiscal sustainability is the catastrophic risk from natural disasters. Assessment, monitoring and mitigation of this risk is a key action for government, as is the monitoring and management of other contingent risks such as the financial performance of SOEs. Coordination across government Coordination across government in relation to policies, incentives, and implementation can be improved. The lack of coordination could contribute to inconsistent policy, inefficiencies in resource use, and delays in policy implementation. For example, in housing, many agencies are involved, some with overlapping responsibilities, including the Housing Authority; the Public Rental Board; the iTaukei Land Trust Board; the Ministry of Local Government, Housing and Environment; town councils and rural authorities; and the Prime Minister’s informal Settlement Stakeholder Committee. Weak coordination and a unified strategy can also lead to difficulties in developing coherent policies. When there are multiple agencies involved in developing policy or designing incentives to achieve a specific policy outcome, it increases the risk that there will be fragmented or contradictory policy approaches. For example, in promoting SMEs the government provides a number of tax incentives and small grants, but it is delivered through various sectoral ministries and agencies without an overall understanding of the full package of assistance (see Error! Reference source not found. Error! Reference so urce not found.above). Systematic changes in coordination will be needed to facilitate to support integrated policy development across government. Conflicting mandates are also a challenge in some cases. For example, in housing, both Housing Authorities and Public Rental Boards are tasked to deliver the public policy objective of improved service delivery to low and middle income groups. At the same time, they are required improve their profitability and financial performance. They are achieving neither. More generally, the government needs to clarify SOEs’ mandates, establish guidelines for the identification, costing, contracting, and financing of community service obligations. At the same time, it should strengthen disclosure requirement and process of 101 selecting, appointing, evaluating the performance of these SOEs and monitor carefully whether the government assets and resources are achieving the policy objectives. Appropriate delegation and accountability for decisions Public administration reform, which has already begun, should improve delegation and accountability for performance. A government-driven program of civil service reforms began in 2015 following the adoption of a new constitution. While some progress has been made, there continues to be weaknesses in a planning, performance monitoring, performance management, and policy development. Changes in these areas are needed, including a greater coordination role for central agencies, to underpin greater institutional performance and efficiency. Improvements in leadership and management could improve institutional performance. In 2016, all permanent secretaries were recruited through an open, merit- based, international recruitment process and are now on performance-based contracts. However, there continue to be difficulties of delegation and management capacity across government. A competent leadership cohort below the level of permanent secretaries is, however, needed to allow senior officers, including Ministers and Permanent Secretaries, to focus on strategic tasks, rather than being distracted by day to day operations. A key accountability mechanism for permanent secretaries is a performance management framework, which is still under development. Capacity development in contract and regulatory management will be needed to support greater private participation in service delivery. As government seeks to improve service delivery through the use of innovative approaches and greater private participation, capacity development for civil servants will be required to support these new approaches. A significant improvement in contract management will be essential to ensure that contractual arrangements are beneficial for government and achieve cost and service effectiveness. While some agencies are already utilizing greater private participation, contract management arrangements remain nascent. Similarly, an improvement in civil service capacity, including the ability to better identify, monitor and mitigate risks, will be needed to underpin the development of more effective regulatory environments. Communication and consultation Enhancing transparency and participation in policy formulation is important to improve the functioning of the government. The government has taken steps to improve transparency and participation by stakeholders in policy formulation, but there is considerable room for improvement. For example, the closure of the credit bureau was made without any consultations. When significant policy changes, such as this, are made in the absence of stakeholder participation, this can create a loss of confidence in the private sector. Legal requirements for information disclosure are limited, but an Information Bill has been introduced to the Parliament to establish the right of access to information held 102 by the government and public agencies. This reform will require systemic and cultural change within government for successful implementation. Improving the mechanisms for feedback and complaints from the public supports a more cooperative and inclusive approach to service delivery. Encouraging clients to give voice to their experiences of the quality and effectiveness of services can support the development of a service-oriented culture. The existing mechanism for members of the public to lodge complaints about service delivery is through members of Parliament, most commonly through Ministers. The Prime Minister, in particular, receives a large volume of complaints through his office. While this mechanism does enable some individual complaints to be addressed, systemic issues are not identified or addressed, and nor are complaints prioritized based on seriousness or need. There are examples of some ministries that have duplicated complaints processing to prioritize complaints made through MPs rather than those coming directly through the ministry. Any improved complaints mechanism must be designed to incentivize complaints to be lodged through the established mechanism to maximize effectiveness. Knowledge gaps Although there are many analyses of the nature and causes of Fiji’s development challenges, this SCD has identified gaps in the analyses, some no doubt due to the World Bank’s long absence from the country. The gaps include: • Poverty profile and poverty map. Although this SCD has analyzed main developments in poverty, fully detailed poverty update is missing. An updated poverty map is also not yet available. Given the changes in nature and distribution of poverty identified in this report, these updates are urgently needed to gain deeper understanding of the drivers of recent changes in poverty and inform policy dialogue. • Growth, investment, and jobs. Fiji’s potential growth rate is important but essentially unknown, and estimating requires additional research. A breakdown of investment between private, government, and public enterprises has not been possible on a consistent basis due to changes in reporting conventions. On jobs, additional research is warranted to better understand the labor market dynamics and constraints: while the government undertakes regular labor force surveys, the empirical analysis of the survey data has been limited. • Analysis of productivity. Given data constraints, it has not been possible to fully understand the productivity challenges in Fiji. Further efforts are needed to link the macro information to the micro level data, such as firm and labor force surveys, to identify constraints to improving productivity. 103 • Firm level and informality surveys. The most recent firm-level survey is from 2008, which is too old to be a sound basis for rigorous policy analysis. A new Enterprise Survey would provide valuable information for a number of policy questions including innovation, investment, productivity, export, and job creation. Additionally, an informal sector survey is urgently needed to identify the drivers of informality, motivations and constraints facing firms and workers, and potential gains in productivity growth and job creation latent in the sector. • Migration, remittances, and poverty reduction. More analysis is needed to understand the costs and benefits of emigration. While emigration has contributed to growing remittances, the loss of highly educated and skilled workers represents a brain drain. There is some international evidence that rising remittances may also reduce labor force participation especially among women, but may also encourage education attainment among young people, especially girls, in the hope of finding better jobs abroad. There is, however, little empirical evidence on the general equilibrium impact of emigration, remittances, and labor market outcomes in Fiji. • Health. To expand coverage of health services and improve health outcomes, the government is considering contracting out health services to the private sector. The experience of other developing countries with such policies has been mixed. Additional research is needed to provide sound policy advice in this area. Such research should seek to provide up-to-date information on the costs of, and access to, health services under private and public provision in Fiji. • Land. There appears to be no empirical research on the impact of the land tenure system on poverty and shared prosperity in Fiji. In some countries, customary land tenure appears to be beneficial (e.g., by providing secure livelihood or social protection to rural communities), but in others, it appears to have a negative impact (e.g., by exacerbating gender disadvantages in asset holdings, access to finance, or intra-household bargaining power). New research could shed light on its impacts in Fiji. 104 References Aizenman, J. and B. Pinto (2004). Managing Volatility and Crises: A Practitioner's Guide Overview. NBER Working Paper, No. 10602. Alwazir, J., F. Jamaludin, L. Dongyeol, N. Sheridan, and P. Tumbarello (2017). Challenges in Correspondent Banking in the Small States of the Pacific, IMF Working Paper, WP17/90. Asante, A., W. Irava, S. Limwattananon, A. Hayen, J. Martins, L. Guinness, J. Ataguba, J. Price, S.Jan, A. Mills, and V. Wiseman (2017). Financing for Universal Health Coverage in Small Island States: Evidence from the Fiji Islands. BMJ Global Health 2017:2. Asian Development Bank (2012). Fiji Country Partnership Strategy 201418. Asian Development Bank (2014). Revitalizing The Fiji Economy. Asian Development Bank (2015). Fiji: Creating Quality Jobs—Employment Diagnostic Study. Asian Development Bank (2016). Tropical Cyclone Winston RecoveryA Desktop Assessment of Coordination, Implementation and Monitoring. Mimeo. Cavu, P., I. Tagicakiverata, S. Naisilisili, and V. Rabici. (2009). Education and Training Needs of Rural Communities: A Situational Analysis of Selected Communities in the 14 Provinces in Fiji. In International Handbook of Education for the Changing World of Work: Bridging Academic and Vocational Learning. Bonn: UNESCO–UNEVOC. Chaundary, F. (2015) “Fiji’s not a hub,” Fiji Times, October 14, 2015, available at http://www.fijitimes.com/story.aspx?id=325813 (accessed October 14, 2016). Clarke, G.(2010). “Are Managers’ Perceptions of Constraints to Growth Reliable? Evidence from a Natural Experiment in South Africa.” Retrieved from: https://mpra.ub.uni-muenchen.de/20098/1/MPRA_paper_20098.pdf Dang, H., P. Lanjouw, J. Luoto, and D. McKenzie (2011). “Using Repeated Cross-Sections to Explore Movements in and out of Poverty,” World Bank Policy Research Working Paper 5550. Datt, G. and M. Ravallion (1992). Growth and redistribution components of changes in poverty measures: A decomposition with applications to Brazil and India in the 1980s. Journal of Development Economics, 38 (2): 275-295. Delaibatiki, N. (2015) “PM: Fiji Natural Hub of Region,” Fiji Sun, August 30, 2015, available at http://fijisun.com.fj/2015/08/23/pm-fiji-natural-hub-of-region/#share (accessed October 14, 2016). 105 DFAT (2015). Pacific Women Shaping Pacific Development: First Annual Report 2012– 15. Duncan, R. and H. Nakagawa (2015). Obstacles to Economic Growth in Six Pacific Island Countries. Report prepared for the World Bank. Mimeo Fiji Sugar Corporation (2015). Annual Report 2015. Fiji National Provident Fund (2016). Annual Report 2016. Frankel (2014). Fiji’s Political and Economic Landscape in the Aftermath of the September 2014 Polls. Report prepared for the World Bank. Mimeo. Global Burden of Disease Study (2013). Country Profile: Fiji. Institute for Health Metrics and Evaluation (IHME). University of Washington. Gottschalk, J., C. Miller, L. Rauqeuqe, I. Wainiqolo, and Y. Yang (2016). The Real Exchange Rate: Assessment and Trade Impact in the Context of Fiji and Samoa. IMF Staff Working Paper WP/16/168. Growth Commission. (2008). The Growth Report: strategies for sustained growth and inclusive development. Washington: World Bank. Hidalgo, C., B. Klinger, A. Barabási, and R. Hausmann (2007). The Product Space Conditions the Development of Nations. Science 317: 482–487 Hsiang, S.M. and A.S. Jina (2014). The Causal Effect of Environmental Catastrophe on Long-Run Economic Growth: Evidence from 6,700 Cyclones. NBER Working Paper No. 20352. IMF. (1995). Fiji – Background Material. IMF Country Report No. 95/128. IMF (2013). Macroeconomic Issues in Small States and Implications for Fund Engagement. Policy Papers. IMF (2014). Fiji: A proposed roadmap for implementing accrual accounting in the government. Fiscal Affairs Department. Mimeo. International Finance Corporation. (2003). SME Business Survey: Fiji – Summary of Findings. International Finance Corporation. (2009). Fiji Country Profile. Enterprise Surveys. Lall, S. (2000). Export performance, technological upgrading and foreign direct investment strategies in the Asian newly industrializing economies: with special reference to 106 Singapore. Santiago, Chile: UN ECLAC, Division of Production, Productivity and Management, Unit of Investment and Corporate Strategies. Ley, E. and N. Tran (2009). Debt Sustainability Risk Analysis with Analytica. World Bank. Lightfoot, C. (2005). “Does customary land ownership make economic sense?” in J. Fingleton (ed). Privatising Land in the Pacific: A defense of customary tenures. Australian Institution. Mahadevan, R. (2008). “The less than sweet solution to Fiji's sugar industry problems.” International Development 21 (1):126-136. McCarthy, S. (2007). Political Instability in the Asia-Pacific: Lessons from the 2006 coups in Thailand and Fiji. Griffith Asia Institute Regional Outlook. Griffith University, Brisbane, Australia. Ministry of Health and Medical Services (2014a). Non-communicable Diseases Strategic Plan 2015-19. Ministry of Health and Medical Services (2014b). Fiji Health Accounts: National Health Expenditure 2011–14. Ministry of Industry, Trade and Tourism (2013). Motor industry urged to capitalize on Fiji as hub, available at http://www.mit.gov.fj/index.php/news/221-motor-industry-urged-to- capitalise-on-fiji-as-hub (accessed 14 October 2016). Mohan, V. (2016). “We have led the way in making Fiji the true hub of the region,” The Worldfolio, available at: http://www.theworldfolio.com/interviews/we-have-led-the-way- in-making-fiji-the-true-hub-of-the-region/3868/ (accessed 14 October 2016). Moring, S. and G. Williams (2000). Modelling Output Fluctuations in Fiji. Working Paper 2000/01. Reserve Bank of Fiji. Nemaia W. Koto (2011). Exploring Informal Urban Settlments in Suva, Fiji: Growth and Environmental Implications. Earthcaching Project, University of the South Pacific, Suva- 2011. Nilan, P., P. Cavu, I. Tagicakiverata, and E. Hazelman. (2006). White collar work: Career ambitions of Fiji final year school students. International Education Journal. pp. 895–905. Pacific Financial Inclusion Program (2009). Cutting costs on sending remittances to the Pacific. Retrieved from: http://www.pfip.org/newsroom/press-releases/2009-2/cutting- costs-sending-remittances-pacific/ Pacific Financial Inclusion Program (2016). Benchmarking financial inclusion in Fiji, Samoa, and Solomon Islands. 107 PCRAFI (2015). Country Note: FijiDisaster Risk Financing and Insurance. Washington, D.C.: World Bank. Prakash, K. and A. Mala (2016). “Is the Dutch disease effect valid in relation to remittances and the real exchange rate in Fiji?” Journal of the Asia Pacific Economy. 21 (4): 571-577. Reddy (2003). “Farm Productivity, Efficiency and Profitability in Fiji’s Sugar Industry.” Fijian Studies 1 (2): 225-241. Reserve Bank of Fiji (2016). The Impact of Exchange Rate Movements on Trade. Reserve Bank of Fiji (2016). Quarterly Review, December 2016. Reserve Bank of Fiji (2016). National Financial Inclusion Strategic Plan 2016–2020. UNESCO (2015). Education for All 2015 National Review: Fiji. Waqabaca, C. (2000). “Financial Sector Development and Reform in Fiji.” RBF Working paper 2000/05 Economics Department. Suva: Reserve Bank of Fiji. World Bank (1977). Economic Situation and Prospects of Fiji. World Bank Country Report No. 1296-FIJ. World Bank (2011). Poverty Trends, Profiles and Small Area Estimation (Poverty Maps) in Republic of Fiji (2003-2009). World Bank (2017). Pacific Possible. Washington, D.C.: World Bank. World Bank (2017). Doing Business 2017. Washington, D.C.: World Bank. World Bank (2017). World Development Report 2017: Governance and the Law. Washington, D.C.: World Bank. 108