68762 Uruguay Investment Climate Assessment January 2008 Finance and Private Sector Unit Poverty Reduction and Economic Management Unit Latin America and the Caribbean Region Document of the World Bank REPUBLIC OF URUGUAY - FISCAL YEAR (January 31 – December 31) CURRENCY EQUIVALENTS (Exchange Rate Effective as of September 20, 2007) Currency Unit = Uruguay Pesos US$1 = 23.1500 UYU ABBREVIATIONS AND ACRONYMS ADME Electricity Market Administration (Administración del Mercado Eléctrico) ANCAP National Administration of Combustibles, Alcohol, and Portland (Administración Nacional de Combustibles, Alcohol, y Portland) ENH Encuesta Nacional de Hogares ENHA Encuesta Nacional de Hogares Ampliada GDP Gross Domestic Product GNI Gross national income IC Investment Climate ICA Investment Climate Assessment ICS Investment Climate Survey INIA National Institute for Agricultural Research (Instituto Nacional de Investigación Agropecuaria) ISO International Organization for Standardization ITU International Telecommunications Union LAC Latin America and the Caribbean MIEM Ministry of Industry, Energy and Mining (Ministerio de Industria, Energía y Minería) RER Real Exchange Rate SEDLAC Socio-economic Database for Latin America and the Caribbean SMEs Small and Medium Enterprises TFP Total Factor Productivity URSEA Energy and Water Services Regulatory Unit (Unidad Reguladora de Servicios de Energía y Agua) UTE National Administration of Electricity Generation and Transmission (Administración Nacional de Usinas, y Trasmisiones Eléctricas) WDR World Development Report WEF World Economic Forum Vice President : Pamela Cox Country Director : Pedro Alba Sector Director : Marcelo Giugale Sector Manager : Lily L. Chu Task Manager : Esperanza Lasagabaster ii URUGUAY INVESTMENT CLIMATE ASSESSMENT EXECUTIVE SUMMARY ....................................................................................................... V CHAPTER 1. INTRODUCTION ......................................................................................... 1 1.1. WHAT IS THE INVESTMENT CLIMATE AND WHY DOES IT MATTER? ...................................... 1 1.2. OVERVIEW OF THE ECONOMY: ACHIEVEMENTS AND CHALLENGES ...................................... 1 1.3. WHY DOES URUGUAY NEED AN INVESTMENT CLIMATE ASSESSMENT? ................................. 4 CHAPTER 2. OVERALL INVESTMENT CLIMATE FINDINGS .................................... 6 2.1. INTRODUCTION ................................................................................................................................ 6 2.2. REGULATION AND GOVERNANCE ................................................................................................ 7 2.3. INFRASTRUCTURE........................................................................................................................... 15 2.4. TECHNOLOGY AND INNOVATION............................................................................................... 22 2.5. SKILLS DEVELOPMENT AND LABOR MARKETS ......................................................................... 24 2.6. FINANCE.......................................................................................................................................... 25 2.7. WHAT IS THE IMPACT OF THE INVESTMENT CLIMATE ON FIRM PRODUCTIVITY?............... 27 2.8. CONCLUSION .................................................................................................................................. 29 CHAPTER 3. ARE URUGUAYAN ENTERPRISES FINANCIALLY CONSTRAINED? 31 3.1 INTRODUCTION .............................................................................................................................. 31 3.2. ACCESS TO FINANCE FROM THE DEMAND SIDE ...................................................................... 32 3.3 SUPPLY SIDE CONSTRAINTS ......................................................................................................... 35 3.4 CONCLUSIONS AND POLICY IMPLICATIONS ............................................................................... 42 CHAPTER 4. LABOR MARKETS, SKILL DEVELOPMENT, AND PRODUCTIVITY IN URUGUAY 46 INTRODUCTION .............................................................................................................................. 46 4.1. 46 4.2. RECENT LABOR MARKET TRENDS .............................................................................................. 47 4.3. LABOR MARKET REGULATIONS AND RELATIONS .................................................................... 49 4.4. DEVELOPING THE SKILLS FOR THE FUTURE ............................................................................. 56 4.5. CONCLUSION .................................................................................................................................. 64 CHAPTER 5. CONCLUSION ............................................................................................. 66 ANNEXES ANNEX 1: SAMPLING METHODOLOGY ................................................................................................... 68 ANNEX 2A: PRODUCTIVITY ANALYSIS .................................................................................................... 70 ANNEX 2B: DETERMINANTS OF BRIBE PAYMENTS IN URUGUAY ....................................................... 84 ANNEX 2C: SELECTED INTERNATIONAL INDICATORS4.4. ........................................................... 85 ANNEX 3: SELECTED INFORMATION ON URUGUAY‘S FINANCIAL SECTOR AND INTERNATIONAL COMPARISONS ........................................................................................................................................... 87 ANNEX 4: INDICATORS AND ECONOMETRIC ANALYSIS ON LABOR MARKETS AND SKILLS ........... 90 REFERENCES ................................................................................................................................................ 98 iii Acknowledgements This report was prepared by a team led by Esperanza Lasagabaster and composed of Thomas Haven; Mario Guadamillas, Mari Luz Cortes, and Paula Cobas (finance); Maria Beatriz Orlando, Paola Aritomi, Georgina Pizzolitto, and Leopoldo Laborda (labor and skills); Ana Maria Oviedo (productivity analysis); Andrés Pizarro (energy and transport); Eloy Vidal (telecommunications); and Micky Ananth (editorial support). The team is grateful for comments received from Jose L. Guasch, James Parks, Jose Guilherme Reis, and Emily Sinnott. Peer reviewers were William Maloney, Mary Hallward-Driemeier, and Daniel Oks. The team benefited from insights provided in meetings with a range of stakeholders in Uruguay, including: the Cámara de Industrias del Uruguay, Unión de Exportadores del Uruguay, Instituto Nacional de Estadística, Department of Economics of the University of the Republic, Cámara Nacional de Comercio y Servicios del Uruguay, Cámara de Industrias del Uruguay, Olivera & Delpiazzo (Abogados), and the Ministry of Economy and Finance, including its Unidad de Apoyo al Desarrollo y la Inversión en el Sector Privado. The team is grateful to the Foro Consultivo del Uruguay that conducted the focus groups and to the workers and entrepreneurs that participate in them. The findings and views expressed here are exclusively those of the World Bank and do not represent the views of the Government of Uruguay. iv EXECUTIVE SUMMARY Why Does the Investment Climate Matter? The health of Uruguay’s investment climate is a key driver of growth and prosperity. The investment climate is a broad concept that encompasses the skills of the labor force, firms‘ technological capabilities, their access to credit, the quality and quantity of infrastructure, and the strength of national institutions and regulatory policies. The productivity and competitiveness of a given country is the result of aggregating the individual performance of firms. Adequate conditions for individual firms to invest, compete, and succeed in a globalized environment will lead to aggregate country growth and competitiveness. As the macroeconomic environment strengthens in Uruguay, investment climate/microeconomic constraints are gaining prominence. The economy has bounced back from the deep economic crisis of 1999– Figure A: Components of Growth in Output per Worker 2002 and continues to enjoy a strong 1980–2003 recovery. Proactive measures are 5 Contribution to output per worker growth (%), 1980-2003 necessary to maintain high growth rates TFP and break with the modest historical Education 4 4.1 Physical Capital growth pattern. Over the past 25 years, Output per Worker 3.5 1.6 3 GDP per capita (in constant US dollars) has grown at an average annual rate of 2.3 2.3 2.3 0.6 2 1.1 0.7 percent. In contrast, other small, 1.2 1.5 Latin 0.6 open economies enjoyed remarkable 0.3 1 0.3 1.9 America 0.2 0.5 Uruguay growth during the same period. Ireland 1.0 0.4 0.7 0.6 0.8 0.1 0.3 0 -0.1 was able to nearly triple its GPD per Australia Finland Chile Ireland Thailand -0.6 -0.5 Australia Finland Chile Ireland Thailand capita, and Chile nearly doubled it. The -1 -1.1 -0.7 performance of physical capital investments and low Total Factor -2 Productivity (TFP) growth in Uruguay Note: The and physical capital to output per worker growth.contribution ofin the education, data show the average annual percentage point The numbers TFP, help explain this growth performance. white squares next to the circle icons denote the compound annual growth rate From 1980 to 2003, Uruguay of output per worker. experienced an average annual decrease Source: Updated dataset from Bosworth and Collins (2003). in TFP of 0.7 percentage points and the contribution of physical capital to growth was -0.1 percent. In turn, TFP is affected by the investment climate including technology adoption. As a small economy, Uruguay could also improve its integration in the global economy to enhance its growth prospects. The sum of its exports and imports is close to 60 percent of GDP compared to 100 percent or far more for other small economies (e.g., Costa Rica, Ireland, Lithuania, and Latvia). The economic recovery does not appear to be significantly expanding jobs in the formal sector. During the last decade, the Uruguayan labor market has been characterized by moderately high unemployment and informality levels, and volatile earnings. In general, it has followed the economic cycle with deteriorations during recessions but more modest gains during recovery periods. Recent improvements have taken place, but more could be done to reduce informality and unemployment. A number of structural factors are affecting these downward rigidities, as explored in this report. Shifting the policy agenda toward investment climate factors can help sustain the recent rapid growth and expand the pace of formal job creation. v This investment climate assessment (ICA) aims to help Uruguay identify and craft solutions for its microeconomic constraints to growth and employment generation. Multiple and complementary sources are used to analyze the investment climate. The report benchmarks Uruguay‘s performance against survey data from comparator countries within and beyond Latin America (mainly upper middle income countries) to identify the shortcomings and reform priorities.1 The primary data source for the report is an investment climate survey undertaken in 2006 that covers 617 small, medium, and large Uruguayan firms in a range of sectors, including food, clothing, chemicals, retail, and others. The survey data yields subjective indicators of firm perceptions about key obstacles as well as objective data for econometric analyses of firm productivity determinants. Data from the investment climate survey is complemented by other cross-country databases such as the World Competitiveness Report and the Cost of Doing Business. In addition, the report uses Uruguay‘s Expanded National Household Survey (2006), which includes a specific and comprehensive module on training.2 Complementary data sources allow for a richer analysis of the labor market and the impact of investment on training. In an innovative approach, focus groups with workers and entrepreneurs were conducted in April 2007 to understand better the functioning of the labor market and the dynamics of labor relations and identify opportunities for building a constructive dialogue. Workers are critical partners and stakeholders in the investment climate and listening to their voices is important. The Foro Consultivo Económico Social del MERCOSUR led the focus group, and the World Bank participated as an independent observer. Overview of Key Findings The analysis conducted confirms the importance of the investment climate for enhancing productivity in Uruguay. The country fares well in public integrity, but its overall performance relative to an international peer group shows room for improvement.3 The picture is complex since the analysis does not point to one or two key binding constraints, rather it points to multiple investment climate factors that are affecting productivity and have not kept up with international trends. Thus, a comprehensive competitiveness agenda is called for. Financial stability has been restored in Uruguay and it appears that there is liquidity in the system, but external financing of Uruguayan firms—essential to their future growth--remains low by international standards. ―Red tape‖ is widespread. Firms would benefit from more investment in training and technology, and new challenges are surfacing in the infrastructure area, especially in energy. Harmonious labor relations would favor employment generation in the formal sector. Besides investment climate factors, other factors within the firm—such as risk aversion and a more entrepreneurial culture—also appear to be affecting growth. Increased global integration could benefit the country. Opening external markets would allow Uruguayan firms to take advantage of economies of scale and achieve higher productivity (Casacuberta and Gandelman 2007). In short, improving global integration and investment climate factors would make the domestic market more dynamic. According to the World Economic Forum, Uruguay ranks 71 out of 80 countries on intensity of local competition. Improvements in the 1 The comparator group largely comprised 10 upper middle income countries and one high income country (Ireland). 2 The survey was conducted by the Instituto Nacional de Estadística de Uruguay. 3 In Latin America and excluding Chile, its investment climate compares more favorably, but the worldwide competitiveness of some of these countries has slipped over the years which points to the need for a broader international perspective. vi investment climate are important to enhance firm productivity and create the conditions to support Uruguay‘s integration in the global economy. Addressing these challenges requires social partnerships involving the government, employers, trade unions, and civil associations. The progress attained on the macroeconomic front in recent years reflects an underlying consensus that sound macroeconomic management is vital for growth and poverty reduction. A similar consensus across social partners would help strengthen labor relations and address other investment climate challenges, laying more solid foundations for continued high growth and employment generation in the formal sector. The findings of the focus groups are promising, indicating that both workers and entrepreneurs value good labor relations and share a common concern that the economic recovery is not generating sufficient employment in the formal sector. The focus groups also identified areas where entrepreneurs and workers were not yet sharing a common understanding. Although social institutions are unique to each country, international experiences can be illustrative. Ireland was immersed in a major economic crisis in 1987, when the Irish Program for National Recovery was negotiated as the first of five social partnerships agreements (O‘Connell, 1999, and Baccaro and Simoni, 2004). Since then, Ireland has experienced rapid growth and an overall social recovery. Social partnerships agreements have also been important to the development of national competitiveness agendas in northern European countries, e.g. Finland. Some reform options for consideration in a national competitiveness agenda are presented below. Labor regulations and skills: Strong social partnerships and labor relations are vital to achieve more flexible labor regulations, while maintaining adequate social protection. Entrepreneurs pointed to labor regulations as one of the top investment climate obstacles. The quantitative analysis based on the household survey also identified that some regulations, in particular Wage Councils provisions, could be increasing the likelihood of informality in some sectors. Providing information to Wage Councils on the impact of regulations on informality and adapting regulations to the needs of specific sectors is vital. Internationally, Uruguay ranks high on rigidity of working hours. Introducing a more flexible regime for working hours would also be welfare enhancing, especially for women that are prone to informality and unemployment. Evidence from other countries in Latin America corroborates the importance of flexible working hours to facilitate the participation of women in the labor force. According to the investment climate survey, entrepreneurs highlighted severance payments as the most important obstacle to hiring and firing. One option for consideration would be to enhance the unemployment insurance scheme, which is a superior social protection instrument, and make the severance payment scheme more flexible to reduce its costs on employment generation. This matter merits further investigation. A quantitative analysis of the firm and household surveys indicates that training has a positive impact on workers’ earnings and firms’ productivity. Yet, investment in training by Uruguayan firms is modest relative to comparator countries. The gap is most acute among SMEs. Only 8 percent of workers receive training among small firms versus 36 percent among large firms. Training policies should tackle market failures and support the unemployed, as well as enhance the productivity of employed workers. Instruments to reach the two groups might be different. Several recommendations emerge from the analysis: (i) reduced duplication among agencies supporting training programs and more participatory training, involving firms and unions in the design of training programs; (ii) higher emphasis on training programs that provide transferable skills versus vii non-transferable skills; and (iii) provision of public incentives to foster training by firms, especially targeted at SMEs and combined with other tools to promote firm productivity such as technology transfer. Technology: The analysis shows that training and technology are complementary. Firms using new technologies in their production process are more likely to invest in training. The capacity to transfer and adapt technology in private companies could be improved. Overall, investment of Uruguayan firms in innovation is low. About 36 percent of firms invested in innovation during the 2001-2003 period, according to the DICYT. Addressing these technology and innovation issues has already become a policy priority. The government has developed a strategy comprising four key pillars. The first envisions a strengthening of the policy and institutional framework. The second focuses on investments in human capital and high-quality research teams, and the third supports the productive sector, inter alia stimulating private sector demand for innovation, developing and strengthening technology transfer institutions for priority sectors, and fostering overall linkages between supply and demand for research and technology transfer. The fourth pillar supports pro-poor innovation. International agencies are supporting the government in these efforts. Long-term sustainability of innovation policies will be vital to their success due to lags in innovation policy returns. In the short-term, programs to promote technology transfer, especially to SMEs could go a long way. Enhancing access to finance: Access to financial markets by Uruguayan firms is limited. The low use of bank credit by Uruguayan firms is notable, accounting for as little as 6 percent of working capital and less for asset financing. Small and medium enterprises (SMEs) tend to be more financially constrained. The survey suggests that the limited use of external financing is due to both demand and supply constraints. Perhaps the low demand responds to the fact that some enterprises remain indebted or have become risk-averse after the crisis. As the investment climate survey indicates, despite notable improvements, macroeconomic instability remains among the three highest concerns of entrepreneurs. If macroeconomic stability and growth continue, demand for credit is likely to increase. For those that demand but do not receive credit, lack of collateral appears to be the main constraint. Uruguayan enterprises have to pledge more collateral as a percentage of their loans than enterprises in many comparator countries, possibly reflecting the low level of collateral recovery in Uruguay and shortcomings in the legal protection of creditors‘ rights. Figure B.1 Fixed Assets Financing Figure B.2 Working Capital Financing 90.0 80 82 73 15 7 4.9 5.2 3.8 0.5 2.6 0.7 0.3 0.8 0 0.1 1.6 0.4 0.6 0.8 0.0 0 s k s ew s k s k r k on s s s bt s ng or r g er er er e an nd cto an St ank d nk an an rm me ar in ien de h th m ni ct b eb -b ie eb se rn ba Ot sh n- sto r se o O b /fr /f r ea st al at ea at No e ew e u al i ly cu rm at ily at iv St N N ed r/C ed N Pr m iv r/ m in fo n lie Fa Pr lie fo ta Fa In ai pp Re pp In et Su R Su Source: World Bank Investment Climate Survey Source: World Bank Investment Climate Survey viii According to the econometric analysis, being incorporated and audited increases the probability of accessing credit, most likely because it enhances the transparency of information. Nonetheless, Uruguay still shows room for improvement with respect to auditing and corporate governance practices. Uruguay‘s capital markets play a minor role in investment financing. Pension fund administrators are rapidly accumulating funds, but these are largely invested in government securities. Some of the weaknesses that limit enterprises‘ access to bank credit also constrain their access to the capital markets: investor‘s rights and corporate governance both show room for improvement. In addition, the enforcement capacity of the regulator could be strengthened. The government has undertaken a series of measures to foster financial sector stability and improve the enabling environment for expanding access to finance in the wake of the crisis. Most important among these are the draft Insolvency and Corporate Reorganization Bill, the draft Capital Markets Law, and a plan for enhancing market transparency. Parliament is considering a new Insolvency and Corporate Reorganization Bill which is critical to facilitate the exit of non-viable firms and the financial restructuring of those with a potential. The authorities are drafting a new Capital Markets law that would strengthen investors‘ protection, corporate governance, and the powers of the regulator. A number of initiatives to enhance the reliability of information are in progress inter alia the development of new accounting and auditing procedures. These above reforms deserve to be supported and accelerated. An active engagement and dialogue with market agents would facilitate the design and implementation of these reforms and the development of additional initiatives to further financial sector growth. Some of the aforementioned reforms will apply to a limited market segment. Additional reforms may need to be implemented for SMEs, which form the bulk of enterprises. For example, simplified accounting rules and a corporate code of conduct could be developed for non-listed enterprises. Is there a role for pro-active market development policies? Recovering market confidence and building the necessary enabling environment will take time, and there is scope for transitory pro- market Government interventions to improve access to financial services and promote the development of capital markets. Thus, the Corporación Nacional de Desarrollo (CND) could play a role in facilitating SME finance and spurring market development. A new management team has been put in place and has formulated a new strategy for the institution. Notwithstanding these encouraging improvements, its overall corporate governance and accountability need to be strengthened to mitigate political risks and a re-emergence of past problems when the CND was utilized as a tool to rescue companies in financial distress. Regulation and governance. Uruguay ranks well on integrity, but its performance on other governance fronts—in particular regulatory and institutional quality—shows scope for improvement. Relatively few Uruguayan firms perceive bribes to public officials as necessary or feel compelled to make informal payments. Public integrity is a valuable public good to be preserved and further enhanced. Its higher public integrity could make the country an attractive place for new foreign direct investment within its regional location. Evidence from the investment climate survey and other cross-country data sources all point in the same direction: there is room to enhance regulation and government effectiveness and reduce red tape. ―Red tape‖ encourages informality by increasing the costs and reducing the benefits of participating in the formal economy. Several initiatives to ease business regulations and enhance the ix quality of responsible institutions are under implementation. Broadly speaking, tax and customs administration reforms are generally proceeding well, but others reforms could benefit from far greater impetus. A widespread culture of business friendly regulations deserves to be more actively promoted. International experience shows that efforts to establish a business friendly environment are most successful when jointly promoted by a public-private commission or council. The private sector can also play a useful role in promoting reform sustainability by participating in public-private councils that regularly monitor outcome indicators such as costs, time, and requirements to obtain permits. While the costs of theft and crime are lower than in most other high middle income countries in Latin America, there is scope for improvement. The econometric analysis indicates that reducing crime and theft would enhance firm productivity. According to focus groups with entrepreneurs, theft losses are owed to internal as well as external theft. Improving labor relations could help reduce internal security problems. Addressing external crime and theft will also be easier than in other countries in the region since it does not suffer from organized crime. A more detailed diagnostic will be necessary to develop a comprehensive strategy, but international experience suggests that one of the most effective entry-points for crime prevention is the municipal level and that social prevention measures can also be effective. Infrastructure: Overall infrastructure development in Uruguay is relatively strong, but new challenges are surfacing, especially in the energy sector. These challenges will grow further as demand expands and new needs emerge from the export sector. Uruguay ranks relatively well in terms of access to energy, road density, port services, and fixed and mobile lines. Efficiency, however, differs markedly across infrastructure sectors and public services. The country is highly dependent on energy imports and facing increasing challenges to meet the growing energy demand. Proactive measures can be taken now to ensure that the country is prepared to meet new demands and shortcomings in the future. The emerging forestry industry and its by-products, for example, will generate new pressures on the existing transport infrastructure. The railway system holds promise for the forest sector but this will require a joint road and rail strategy and a restructuring of the National Railway Administration (AFE). If the supporting infrastructure is developed, Uruguay could possibly develop as a logistics hub. More could also be done to encourage new telecommunication services such as broadband internet, which is important to support the transformation to a knowledge economy and the emerging IT-related industry. The expansion of infrastructure services could greatly benefit from further promotion of competition and participation of new players. Where Uruguay has allowed some competition, it has achieved significant gains in efficiency, e.g., in ports and long-distance telecommunication services, and in coverage, e.g. in mobile services. The three distinct roles of policy definition, regulation, and operations could be more clearly differentiated in many infrastructure sectors to foster competition. Similarly, improving the autonomy and capacity of regulatory bodies for energy and telecommunications could encourage competition, efficiency, and the entry of new market participants. An improved regulatory framework in the energy sector would create the conditions for the entry of new participants and the expansion of local generation capacity, including in renewable sources that have now become more competitive in light of high oil prices. Together with this, on- going public efforts to promote energy efficiency and savings deserve to be further promoted. x CHAPTER 1. INTRODUCTION Abstract: This chapter outlines why it is important to focus on the investment climate in Uruguay. First, the key role of microeconomic determinants of growth is noted. Uruguay’s key economic challenges with respect to growth, total factor productivity, export dynamism, and investment levels are then highlighted. After brief discussions of macroeconomic stability and trade integration issues, the chapter highlights the critical role of microeconomic/investment climate factors in catalyzing future growth. The chapter concludes by noting how an investment climate assessment can help identify and craft solutions for Uruguay’s microeconomic challenges to growth by drawing on a firm-level survey in Uruguay and benchmarking results with relevant countries in the region and elsewhere. 1.1. WHAT IS THE INVESTMENT CLIMATE AND WHY DOES IT MATTER? 1. Sound macroeconomic policies are necessary but not sufficient to ensure faster economic growth. The microeconomic foundations of economic development also need to be taken into account, as they strongly affect competitiveness and growth. These microeconomic determinants of growth encompass a wide range of factors, including the skills of the labor force, firms‘ technological capabilities, their access to credit, the quality and quantity of infrastructure, and the strength and quality of national institutions and regulatory policies. These determinants constitute the business environment or investment climate in which firms operate. The productivity and competitiveness of a given country is the result of aggregating the individual performance of firms. Adequate conditions for individual firms to compete and succeed in a globalized environment are a necessary input to aggregate country growth and competitiveness. 2. There is ample international evidence on the importance of microeconomic determinants for growth. Over 1960–2000 the bulk of the differences in growth between countries (45–90 percent) is accounted for by differences in total factor productivity (TFP), the part of growth that cannot be explained by the accumulation of physical or human capital (Bosworth and Collins, 2003). Prescott (1998) argues that differences in TFP are attributed to resistance to the adoption of new technologies and to the efficient use of existing operating technologies; the latter is in turn affected by the institutional and policy arrangements a society employs, or the ―investment climate‖. The World Bank‘s 2005 World Development Report (WDR) entitled A Better Investment Climate for Everyone (World Bank, 2004a) also linked the investment climate to total factor productivity growth and positioned the investment climate as a key pillar of economic growth, job creation, and poverty alleviation. 1.2. OVERVIEW OF THE ECONOMY: ACHIEVEMENTS AND CHALLENGES 3. The Uruguayan economy has bounced back from the deep economic crisis of 1999- 2002 and continues to enjoy a strong recovery. Real GDP growth in 2004 and 2005 was 11.8 percent and 6.6 percent, respectively, and reached 7.0 percent in 2006. Macroeconomic risks have diminished. External debt as a percentage of GDP fell from a peak of 98 percent in 2003 to an estimated 55 percent in 2006, and inflation decreased from 25.9 percent in 2002 to 6.4 percent in 2006. Strong macroeconomic performance owes much to buoyant exports and a very positive response of domestic aggregate demand. The government‘s adoption of a flexible exchange rate and its subsequent depreciation together with a surge in the price of key agricultural exports increased Uruguay‘s competitiveness. While these have provided a temporary boost, longer-term 1 competitiveness will also depend on other microeconomic attributes. Foreign investment has increased. Particularly important are large- Figure 1.1: Evolution of GDP per Capita scale projects in the wood pulp sector which 250 have begun to materialize. Foreign investment is also having a dynamic effect in the rural, 225 Australia forestry and tourism sectors. 200 Chile Finland Index (1980=100) Ireland 4. The challenge is to build on these 175 LAC gainsand break away from the modest Uruguay growth rates of the past. Over the past 25 150 years, GDP per capita (in constant 2000 US 125 dollars) has grown at an average annual rate of 0.7 percent (Figure 1.1). In contrast, other 100 economies with similar structures (small, 75 open, and/or based on adding value to 80 82 84 86 88 90 92 94 96 98 00 02 04 primary products)—such as Australia, Chile, 19 19 19 19 19 19 19 19 19 19 20 20 20 Finland, and Ireland—enjoyed remarkable Note: Ireland’s GDP per capita reached index 300 (where 1980=100) in growth during the same period driven to a 2005. Actual 2005 GDPChile capita values in 2000 constant US dollars are: Australia $ 22,422; per $5,747; Finland $25,591; Ireland $29,295; large extent by exports. Ireland was able to LAC average $4,036; and Uruguay $6,268. nearly triple its GDP per capita, and Chile Source: World Bank, World Development Indicators nearly doubled it. Australia and Finland, in particular, have been very successful at adding value to their rich natural resource base. In addition, towards the late part of the last century, Finland was able to develop a very dynamic high tech sector.1 5. Modest TFP growth and Figure 1.2: Components of Growth in Output per Worker, 1980–2003 investment held back Uruguay’s 5 growth potential for a long time. Average annual growth in output per Contribution to output per worker growth (%), 1980-2003 TFP Education 4 worker in Uruguay was actually negative 4.1 Physical Capital Output per Worker 0.5 percent over the period 1980–2003 3.5 1.6 3 (Figure.1.2). The difference between growth in GDP per capita and output 2.3 2.3 2.3 0.6 2 per worker is due to an increase in the 1.5 1.2 1.1 Latin America 1 labor force relative to the population 0.6 0.5 0.3 0.3 1.9 size, a trend that will come to a close 0.2 Uruguay 1.0 0.4 0.7 0.6 0.8 Australia Finland Chile 0 Ireland 0.1 Thailand with the aging of demographics. Similar 0.3 -0.1 Australia Finland Chile Ireland Thailand -0.6 to other countries in the figure, -0.7 -0.5 -1 -1.1 education made a small positive -2 contribution to growth in output per Note: The data show the average annual percentage point contribution of TFP, worker in Uruguay (an annual average of education, and physical capital to output per worker growth. The numbers in the 0.3 percentage points). However, the white squares next to the circle icons denote the compound annual growth rate of output per worker. contribution of physical capital to growth Source: Updated dataset from Bosworth and Collins (2003). fell well short of that in other countries: -0.1 percentage points in Uruguay versus between 0.6 and 1.9 percentage points in the comparator countries. TFP growth—or lack thereof—made up the difference. Uruguay 1In 1990, high-tech exports constituted less than 7 percent of Finnish exports but had risen to 30 percent by 2000 (World Bank, 2001). 2 experienced an average annual decrease in TFP of 0.7 percentage points, which begs the question of what microeconomic factors dampened TFP. In contrast, strong positive TFP growth in the comparator countries accounted for roughly half of total output growth. 6. While private and foreign investments have increased since the Figure 1.3: Savings and Investment Rates (1990–2005 average) crisis, Uruguay experienced modest investment rates for an extended period. 35 Figure 1.3 highlights the modest savings and 30 investment rates in Uruguay, averaging 25 roughly 15 percent between 1990 and 2005. % of GDP 20 In contrast, countries like Mexico and Brazil 15 averaged close to 20 percent, and investment 10 and savings rates in Thailand were over 30 5 percent. Modest investment rates reflected 0 low demand for investment. Hausmann, o il ay d nd a nd le a lia az ic in n ic Rodriguez-Clare, and Rodrik (2005) argue hi ra gu la a la ex R nt Br C nl Ire ai st ru ge ta M Fi Th Au U os Ar that a low supply of savings was not a key C Gross fixed capital formation Gross domestic savings constraint for Uruguay. They note that the Note: Investment refers to Gross Fixed Capital Formation and Savings country had not fully used its available refers to Gross Domestic Savings. sources of external savings since the current Source: World Bank, World Development Indicators (2007). account deficit barely averaged 1.4 percent of GDP during the 1990s and the country enjoyed an investment-grade credit rating. Hence, Hausmann et al. conclude that the modest investment ratios reflected low demand for investment. Also, the modest demand for investment could possibly reflect an insufficient entrepreneurial culture. 7. Three factors affected investment demand: low macroeconomic stability, trade integration, and microeconomic/investment climate constraints. The combination of crises (the 1982 debt crisis and the 2002 financial crisis) and high real exchange rate volatility in between painted a picture of macroeconomic instability that dampened investment. As noted earlier, Uruguay‘s macroeconomic stability has improved considerably since the 1999–2002 recession and crisis.2 Insufficient trade integration also affected investment. The sum of Uruguay‘s exports and imports is close to 60 percent of GDP compared to 100 percent or far more for other small economies, e.g., Costa Rica, Ireland, Lithuania, and Latvia (Figure 1.4). 8. Trade and integration policies can help provide an impulse to growth through at least four channels: by reducing Uruguay‘s vulnerability to regional volatility and fluctuations in terms of trade; by facilitating access to cost-effective and high quality and technology intermediate and capital goods; by improving access to Uruguay‘s products in foreign markets—including those where access is currently limited by agricultural protectionism; and by allowing domestic companies to achieve economies of scale and thus enhance their productivity. 2 In between the two crises which entailed large drops in output, Uruguay was subject to exchange rate fluctuations within its two powerful neighbors, Brazil and Argentina, resulting in high real exchange rate (RER) volatility. In fact, a ranking of 73 countries in terms of volatility of their 5-year RER in the period between 1980 and 2000 shows Uruguay to have the 8th most volatile RER. A volatile RER increases the risk and lowers investment in the tradable sector for any expected level of the RER. See Hausmann, Rodriguez-Clare, and Rodrik (2005) for more details and also Hausmann, Panizza and Rigobon (2004a and b) for a study of the negative two-way relationship between real exchange rate volatility and openness. 3 9. There is scope for deepening Figure 1.4: Exports and Imports of Goods and Services integration within MERCOSUR and (% of GDP), 2004 possibly beyond MERCOSUR. The government‘s strategy is to continue 90 international trade integration beyond 80 Exports MERCOSUR. Uruguay is exploring 70 Imports the possibility of signing Bilateral Free 60 Trade Agreements outside the % of GDP 50 MERCOSUR and has attempted to get 40 these considered within MERCOSUR. 30 In January 2007, Uruguay signed a 20 Trade and Investment Framework 10 Agreement (TIFA) with the USA that 0 will allow both countries to make y a o le l ia nd a nd . a zi progress in several areas of mutual ep a ni ic tin ic hi tv a gu la la ex R ua R Br C La en ai Ire ru ta M k th Th g a interest, such as intellectual property U os Li Ar ov C Sl rights, trade in services and government Source: World Bank, World Development Indicators, 2007. procurement, and which may facilitate an eventual FTA negotiation in the future. 10. Lastly, microeconomic/investment climate constraints also affect firm investment and productivity and thus growth. As Uruguay‘s macroeconomic environment has recovered investment climate constraints have gained prominence. According to the World Bank‘s Investment Climate Survey (discussed at length in chapter 2), firms considered the top obstacles affecting their business to be competition from the informal economy, tax rates, macroeconomic instability, labor regulations, and access to finance. With the exception of macroeconomic instability, all of theses obstacles are microeconomic issues. Similarly, according to the Global Competitiveness Report 2006–2007 survey of business executives, the top three most problematic factors for doing business are: 1) access to financing; 2) inefficient government bureaucracy; and 3) restrictive labor relations. These factors are all key elements of the investment climate. 1.3. WHY DOES URUGUAY NEED AN INVESTMENT CLIMATE ASSESSMENT? 11. Uruguay needs an investment climate assessment (ICA) to identify and craft solutions for its binding microeconomic constraints to growth. The preceding section noted Uruguay‘s improved macroeconomic stability but also highlighted the need to increase TFP to sustain high growth rates. To boost exports, and maintain high growth, it is vital that Uruguay successfully encourages firms to invest, adopt technology, and innovate. To accomplish those goals, a friendly investment climate is critical, and an ICA provides actionable policy recommendations to make substantial steps in the right direction. 12. An ICA relies on firm survey data to benchmark investment climate conditions across comparable countries. The main focus of an ICA is on microeconomic factors constraining firms‘ operations, such as weaknesses in governance, infrastructure, finance, skills, and technology acquisition, and linking them to firm-level productivity. In the case of Uruguay, these factors were measured through a survey of roughly 600 firms. The sample consisted of a range of small, medium, and large firms in the food, clothing, chemicals, and retail sectors, as well as a residual category for other types of firms. Annex I contains details about the survey data and methodology. Similar surveys have been undertaken by the World Bank in other countries in Latin America and 4 throughout the world, providing a rich pool of data for cross-country comparisons. This ICA will benchmark Uruguay‘s performance against survey data from comparator countries to identify the shortcomings and reform priorities. Furthermore, the ICA will likely be the first in a series within Uruguay, providing a baseline to track the impact of policy changes. 5 CHAPTER 2. OVERALL INVESTMENT CLIMATE FINDINGS Abstract: This chapter mines the investment climate survey and complementary sources and confirms the importance of the investment climate for productivity in Uruguay. The impact varies according to the investment climate factor. Uruguay fares relatively well on aspects concerning public integrity. However, on other governance fronts (especially on regulatory and institutional quality), access to finance, and technology, there is substantial scope for improvement. Upgrading workers’ skills will enhance productivity, but tensions in labor relations are not supportive of the investment climate and employment generation in the formal sector. In addition, new gaps are surfacing in infrastructure. Addressing these challenges will be crucial to promote sustainable growth and job creation and to create the conditions for Uruguay’s integration in the global economy. Uruguay ranks 71 out of 80 countries on intensity of local competition, the local market is small, and opening external markets will allow Uruguayan firms to benefit from economies of scale and achieve higher productivity. 2.1. INTRODUCTION 1. Using firm-level data from the investment climate survey and other cross-country data sources, such as the World Bank‘s Doing Business Initiative, the World Economic Forum and Uruguay‘s household survey, this chapter identifies key microeconomic Figure 2.1: List of Comparator Countries areas for improvement in Uruguay. The investment climate is T hailand ( 2 0 0 3 ) 3 ,13 7 C o st a R ica ( 2 0 0 5) 4 ,8 58 disaggregated in five areas: (i) M aur it ius ( 2 0 0 5) 5,12 9 regulation and governance; (ii) A r g ent ina ( 2 0 0 6 ) B r az il ( 2 0 0 3 ) 5,4 58 5,717 infrastructure; (iii) technology and U r ug uay ( 2 0 0 6 ) 6 ,0 0 7 innovation; and (iv) skills and labor M exico ( 2 0 0 6 ) 8 ,0 6 6 markets. In each area, Uruguay is Lat via ( 2 0 0 5) 8 ,550 Lit huania ( 2 0 0 5) 8 ,6 10 benchmarked to relevant comparator C hile( 2 0 0 3 ) 8 ,8 6 4 countries, mainly high middle income Slo vak R ep ub lic ( 2 0 0 5) Ir eland ( 2 0 0 5 10 ,158 52 ,4 4 0 countries in Latin America and the 0 10 ,0 0 0 2 0 ,0 0 0 3 0 ,0 0 0 4 0 ,0 0 0 50 ,0 0 0 6 0 ,0 0 0 rest of the world (Figure 2.1). Select GD P p er cap it a f o r 2 0 0 6 variations within Uruguay according year next to to firm size, economic sectors, and Note: The Climate Survey the country is the year in which the Investment was conducted. export characteristics are also presented. The investment climate is a broad subject and reforms in some areas will have more of an impact than in others. The impact of investment climate reforms depends on both a country‘s performance relative to best practice in each area and the impact of potential improvements in that area on firm productivity. We complement the basic benchmarking with an econometric analysis of the impact of different investment climate attributes on firm productivity and workers‘ earnings. 2. Firm perceptions provide an initial weighting of the importance of different obstacles but need to be interpreted with caution. Perceptions may be biased by recent events reported in the media, and they may also reflect specific cultural and socioeconomic backgrounds. For instance, managers of firms that concentrate on local as opposed to national or international markets may lack the necessary benchmarks to judge the severity of the problems existing in their cities or provinces, and 6 compare them to national or international best practices. Bearing these caveats in mind, the survey indicates that Uruguayan Figure 2.2: Uruguay. Percentage of Firms that Consider Issue to be entrepreneurs perceive their most 1 of Top 3 Obstacles significant obstacles to be the Competition in informal economy 58% practices of competitors in the Tax rate 53% informal sector, tax rates, Macroeconomic instability Access to finance 25% 34% macroeconomic instability, access Labor regulation 23% to finance, and labor regulations Crime Political instability 14% 21% (Figure 2.2). While Uruguay has Electricity connection 14% made substantial improvements in Inadquate education Corruption 12% 10% regaining macroeconomic stability Tax administration 9% (Chapter 1), the perceptions of Trade regulation 8% Court system 4% entrepreneurs may still be affected Transportation Access to land 3% 2% by the deep crisis of 2002 and the Permit & License 2% repeated macroeconomic shocks experienced in previous decades. Source: Investment Climate Survey, 2006 A longer period of macroeconomic stability will be necessary to regain the full confidence of entrepreneurs. Perceptions on other investment climate issues are further analyzed below by investigating other objective indicators from the survey. 3. The remaining of the chapter is structured as follows: Sections 2.2, 2.3, 2.4. 2.5, and 2.6 analyze regulation and governance, infrastructure, innovation, labor markets and skills, and access to finance, respectively. Section 2.7 summarizes the results of the firm productivity analysis and Section 2.8 concludes. Chapters III and IV follow with a more in-depth analysis of access to finance and labor markets, labor relations and skills development given the importance of these two areas for the overall investment climate and employment generation and the important challenges the country faces in these areas. 2.2. REGULATION AND GOVERNANCE 4. Uruguay ranks well on integrity, but its performance in other governance fronts-- informality, regulations, and security--shows scope for improvement. Competition from the informal sector is a major concern for firms, and the econometric analysis found that there is potential for productivity gains if regulatory compliance was improved. Potential drivers of informality include business and labor regulations and high payroll taxes. Even though Uruguay generally scores better on measures of red tape than many Latin American comparators, improvements are needed to catch up with countries outside the region and Chile. Although there is no organized crime in Uruguay, losses from theft and security costs reduce firm productivity. 7 Uruguay Fares Well on Integrity 5. Uruguay fares well on integrity measures. Only 10 percent of firms in the investment climate survey considered corruption to be one of the three most important obstacles facing their establishment (Figure 2.2). The investment climate survey asked firms whether informal payments were requested for licenses, permits, utility connections, and in other interactions with public officials. Uruguay scored relatively well on these measures compared to other high middle income countries within Latin American and around the world (Figure 2.3). A Probit analysis was used to estimate the relationship between various aspects of the investment climate and firm characteristics, and the probability of having to pay bribes to public officials. The econometric analysis was conducted for Uruguay data alone and for Uruguay data pooled with data from its peer countries in order to compare Uruguay‘s relative performance. Interesting results emerge. First, firms that depict lower regulatory compliance and that have requested public services during the last two years tended to be more exposed to bribe payments. Second, our econometric results indicate that Uruguayan firms tend to be less exposed to bribe payments compared to most of their peer countries, especially Argentina, Brazil, Latvia, Lithuania and Slovakia. See Annex 2B for detailed econometric results. 6. Courts also received Figure 2.3: Relatively Few Uruguayan Firms Perceive Bribes to relatively high marks on integrity Public Officials as Necessary or have been Exposed measures: they are considered to to Informal Payments be fair, affordable, and able to enforce decisions, albeit slow. 0.800 0 .7 6 Only 4 percent of firms in the 0.700 investment climate survey 0.600 considered courts to be one of the 0 .4 7 2 0 .4 7 8 Percent of Firms top three obstacles to their 0.500 0 .3 6 7 business. On measures of fairness, 0.400 0 .3 0 9 affordability, and enforceability of 0.300 decisions, Uruguayan courts 0.200 0 .16 0 .19 2 0 .12 8 received top marks vis-à-vis the 0 .0 9 2 0 .0 9 8 comparator countries. For 0.100 0 .0 6 1 instance, 54 percent of firms in 0.000 Uruguay considered courts to be l y ia ia a le o nd zi ia a s ic ua ic tv tin hi ak iu ra an la R ex La g C rit B fair, impartial, and uncorrupted, ov en uv Ire ru ta M au Sl rg os th U M A Li C compared to 16 percent in Note. A firm is considered to be exposed to gifts or bribes if the firm agrees that Argentina and 27 percent in informal payments are necessary to obtain government contracts, or if the firm Mexico and also ahead of other has had to make informal payments to obtain electricity connection, telephone Eastern European countries connection, construction permit, import license, operating license, or if it had to make an informal payment whenever inspected by tax officials (Figure 2.4). Only 7 percent of Uruguayan firms, however, Source: Investment Climate Surveys, World Bank considered courts to be quick when handling business disputes between private parties, compared with 15 percent in Lithuania and 17 percent in Mexico (Figure 2.5). 8 Figure 2.4: Percentage of firms that Consider the Figure 2.5: Percentage of Firms that Consider that Court System to Be Fair, Impartial, & Uncorrupted Courts are Quick Argentina 2.1 Ireland 56.6 Uruguay 53.8 Uruguay 7.3 Slovakia 33.5 Slovakia 7.7 Latvia 32.7 Latvia 9.4 Lithuania 27.4 Mexico 26.7 Lithuania 14.9 Argentina 15.8 Mexico 16.8 Percentage of Firms 0 25 S ource: World Bank Investment Climate Surveys (2006) Source: World Bank Investment Climate Surveys (2006) Informality remains a Concern and “Red Tape� is High 7. Informality remains a concern in Uruguay. When measures of informality in the investment climate survey—including percentage of firms that formally registered when starting operations, percentage of total annual sales declared for tax purposes, and percentage of total workforce declared for tax purposes—are benchmarked with other countries, informality in Uruguay is slightly smaller than in the Latin American comparators such as Brazil, Argentina, and Mexico, but higher than in Chile and Costa Rica. The same measures showed that Uruguay has slightly more informality than the Eastern European comparators such as Latvia, Lithuania, and the Slovak Republic. Perry et al. (2007), which investigate informality in Latin America and the Caribbean, also find that the share of informal workers in Uruguay is higher than in Chile and Costa Rica, but lower than in Brazil and Mexico (Figure 2.6). 8. Uruguay‘s household survey also confirms Figure 2.6: Share of Informal Workers, 2003 relatively high informality rates, hovering between 22 and 42 percent of the labor force depending Brazil 55.0% on the definition. Definition 1 includes all salaried workers in small firms, non-professional self- Mexico 54.1% employed workers, and zero-income workers. Argentina 44.8% Definition 2 includes workers who do not have 43.6% the right to receive pensions when they retire. Uruguay According to the first definition, informality was Costa Rica 41.4% 42 percent in 2005, compared with 26 percent Chile 37.0% using the second definition. While having workers engaged in the informal economy is superior to 0% Percent of the Labor Force 60% having them idle, informality is likely to impose Note: An individual is considered an informal worker if (s)he productivity costs. There may be economies of is: (i) unskilled self-employed, (ii) salaried worker in a small firm, or (iii) zero-income worker. scale arising from employment in larger firms, Source: Gasparini and Tornarolli (2007). and informality can discourage or impede the growth of firms by preventing their access to credit and good markets and public services. The analysis below confirms the positive correlation between formality (as measured by higher regulatory compliance) and labor productivity. 9 9. Improved business and labor regulations and the quality of public institutions could help reduce the level of informality in Uruguay. According to Perry et al (2007), informality tends to increase with the quantity of labor and product markets regulations, and to decrease with the quality of governance—for example, the prevalence of the rule of law, the level of democratic accountability, and the quality of public institutions. Recent initiatives in Latin America to reduce red tape and introduce simplified tax and registration systems for micro-firms have led to statistically significant increases in the number of new formally registered firms. In particular, some recent estimates suggest that the simplification of entry regulation leads more former high-ability wage workers to open formal businesses (Perry et al., 2007). 10. Evidence from the investment climate survey and other cross-country data sources all point in the same direction: Uruguay presents a number of regulatory shortcomings and there is scope for reducing red tape and improving government effectiveness.1 According to Kaufmann el al., the government is substantially more effective in Uruguay than in Brazil, Argentina and Mexico but performs below its Eastern European comparators and Chile (Figure 2.7). Performance indicators on overall regulatory quality are slightly weaker (Figure 2.8). The investment climate also found that managers in Uruguay have to spend 9 percent of their time dealing with government regulations, an amount higher than in Thailand, Ireland and the Eastern European comparators, although better than in most other high middle income countries in Latin America, except Chile. 11. Historically, the tax system could have encouraged informality. The new, more streamlined tax code and on-going administration reforms are helping to reduce the cost of enterprise compliance. Uruguay‘s tax administration and tax policy was not very efficient. Enterprises had to comply with a sizable number of taxes, many of which generated low revenue but imposed an additional administrative burden on enterprises.2 A new tax code became effective in July 2007 establishing a simpler, more equitable and efficient system. The key change to the system is the introduction of dual personal income taxes, combining a progressive tax schedule for labor income with a low flat tax rate on capital income (12 percent). The reform institutes a personal income tax for all labor and capital income, broadening the base and improving horizontal equity in the system.3 It also simplifies the tax structure, reducing by half the number of taxes and thereby eliminating a number of highly distortive, low-revenue generating taxes. It is expected to be revenue neutral. 12. The government has launched in tandem a program to modernize the tax administration agency (Dirección General de Impuestos). This will reduce the cost of enterprise compliance and increase the agency‘s capacity to enforce compliance with the law, helping expand the base and generating a more level-playing field for enterprises. An expansion of the tax base could also generate space for some reductions in the value added tax in the future, which seems on the high side compared to international indicators (Annex 2.C). A modernization of the social security administration is underway, too. Payroll taxes, however, remain about 3 percentage points above the average of Latin American countries. Total labor taxes represent 40.5% of the gross wage in Uruguay, compared with 28% on average in Latin America and 35% in the OECD (Annex 2.C).4 An important pension reform was implemented in 1996, but the system 1 Other country data sources include the Ease of Doing Business, Kaufmann et al, and World Economic Forum. 2 There were 12 different direct taxes and 18 indirect taxes, with 20 taxes each accounting for less than 1 percent of revenue. Just five taxes, three consumption taxes (VAT, IMESI and COFIS) and two income taxes—the wage tax (IRP) and the income tax rate for corporations and individuals (IRIC)—provide over 85 percent of tax revenues. 3 The dual income tax system pioneered by the Nordic countries in the 1990s, responds to the plight of small open economies that are unable to trace non-domestic sourced income in the face of increased capital mobility across borders. For these countries, a low flat capital tax reduces the risk that taxpayers will have the incentive to move capital abroad. 4 See World Bank (2005a) and Packard (2001) for additional details. 10 remains relative generous compared to most other countries in Latin America, especially in view of its aging demographics. Recent reforms to eliminate payroll tax exemptions for certain sectors constitute a positive change that will generate a more level playing field. The labor market analysis also suggests that some labor regulations could have contributed to informality (Chapter IV). Figure 2.7. Government Effectiveness Figure 2.8 Regulatory Quality (Scale from -2.5 to 2.5) (Scale from -2.5 to 2.5) Ireland 1.6 Ireland 1.6 Chile 1.4 Chile 1.3 Slovakia 1.2 Slovakia 1.0 Lithuania 1.1 Lithuania 0.9 Latvia 1.0 S Africa 0.8 Costa 0.6 Latvia 0.7 S Africa 0.6 Mauritius 0.6 Thailand 0.4 Uruguay 0.5 Mexico 0.3 Mauritius 0.32 Thailand 0.4 Uruguay 0.26 Costa Rica 0.3 Brazil 0.1 Mexico -0.01 Argentina -0.6 Brazil -0.1 Argentina -0.3 -1.8 0.0 1.8 -1.8 0.0 1.8 Source: Kaufman et al. 2006 Source: Kaufman et al. 2006 13. Gaps are found in areas concerning procedures for registering property, enforcing contracts, and protecting investors (Table 2.1). These increase the cost of doing business and hinder access to finance by enterprises. A number of measures to address them are in progress or being considered as further discussed in Chapter 3 on Access to Finance. Starting a business also appears to be particularly burdensome—most likely contributing to informality (Table 2.1). Reducing the cost of business registration would encourage more firms to participate in the formal economy. Table 2.1: Selected Ease of Doing Business Rankings, 2007 Thailand Latvia Chile Mexico Uruguay Argentina Brazil Doing Business 15 22 33 44 98 109 122 Starting a Business 36 30 39 75 151 114 122 Closing a Business 44 64 98 23 37 65 131 Registering Property 20 85 34 71 139 96 110 Enforcing Contracts 26 3 64 83 91 47 106 Protecting Investors 33 51 33 33 83 98 64 Trading Across Borders 50 19 43 76 125 107 93 Employing Workers 49 96 68 134 72 147 119 Dealing with Licenses 12 82 58 21 137 165 107 Source: Doing Business, 2007 (www.doingbusiness.org). 14. Shortcomings exist in the responsible government institutions. All commercial entities (sociedades comerciales) need to register their constitutional contract in the Public Commercial Registry (Registro Público de Comercio), and all public limited companies (sociedades anónimas) need to also be authorized by the Internal Auditor of the Nation (Auditoría Interna de la Nación or AIN). These two institutions have limited capacity to properly attend investors, generating delays estimated by experts to be at least two months. Furthermore, the regulatory framework requires documentation that is no 11 longer relevant. Faced with these hurdles, the market has developed alternative ways to register a business that are more expeditious but still require subsequent modifications.5 15. The government has undertaken a number of promising reforms to facilitate starting a business. In 2005, the tax and social security registration procedures in Montevideo were merged in a single window, resulting in significant time savings. The single window is currently being rolled out to other areas of the country. In addition, a private sector development unit (Unidad de Apoyo al Sector Privado) has been created at the Ministry of Economy and Finance. The department‘s mission is to serve investors and to provide them with all relevant information, including information about procedures and government incentive programs. A guide to ―start a business procedures‖ has been created and is available on the Ministry website. The department is exploring further changes to facilitate registration. Plans to automate the enterprise registry are underway. 16. The following actions could also be effective in making it easier to start a business in Uruguay. First, public agencies would benefit from more adequately trained staff easing congestion. Second, revising procedures would help eliminate duplication of information and obsolete requirements, and a move toward a single window would provide more accessibility and clarity to the investor. Third, minimum capital requirements, which are 183 percent of per capita income (compared with a regional average of 18 percent) could be reevaluated. Fourth, procedures at the municipal level could be simplified. Lessons can be learned from reforms elsewhere in Latin America, in particular Mexico and Peru which have launched several initiatives to facilitate ―starting a business‖ (Box 2.2). An important lesson is the role that the private sector can play in promoting reform sustainability by monitoring key performance indicators over time. 17. Enhancing the effectiveness of agencies that will support the global integration of Uruguay is also vital. On trade clearance, Uruguay generally performs better than most of its peer comparators in Latin America but falls behind when compared to other rapidly growing export economies outside the region (Figures 2.9 and 2.10). It takes an average of 3.7 days to clear export customs in Uruguay, compared with 1.5 days in Thailand, 2 days in Latvia, and 6 days in Argentina. Results for the average number of days to clear import customs are roughly similar: 7.3 days in Uruguay versus 3.3 days in Latvia, 4.6 days in Thailand. 18. While Uruguay boasts state-of-the-art customs information technology systems, outdated procedures and not very efficient human resources management prevail. These are critical impediments to more efficient and transparent customs operations. Furthermore, paper-based processing is still common, thereby increasing processing times and associated costs as well as opportunities for illegal practices. Recent World Bank studies have found that outdated human management practices within customs constitute the single overriding impediment to the overall efficiency and transparency of the customs function.6 To overcome these deficiencies, the 5 First, many businesses start as a ―society in the process of formation‖. After registering with the tax collection agencies, the firm can start operations in a maximum of three days and finalize the formal constitution process over time. The founding partners, however, assume responsibility until the corporate structure has been formalized. A second accepted practice is the purchase of ―standard societies‖. Several firms undertake all of the needed procedures for creating ―standard societies‖, which can be applied to a wide range of commercial activities. These ―standard societies‖ are sold for about US$1,000 to investors, who can then start their activities within two days. In most cases, these ―standard societies‖ require subsequent modifications to be adapted to the specific needs and requirements of owners and shareholders (level of capital and board structure) and to operate in a specific activity, for example permits by local governments, environmental authorizations, and licenses for economic zones. 6 Discretionary recruitment procedures and outdated staffing profiles undermine the efficiency and transparency of customs operations. At present, job profiles for customs posts are not well developed; staff are contracted through 12 government launched in 2006 a program to modernize the customs administration agency (Dirección General de Aduanas) with a strong emphasis on a gradual and comprehensive human resources reform.7 Reforms of the Instituto Uruguay XXI, agency responsible for promoting exports and foreign investment, are also underway. These merit to be accelerated. Box 2.1: Starting Business Reforms in Mexico and Peru Mexico’s Federal Regulatory Improvement Commission (COFEMER) works with municipalities to simplify business opening procedures. The Rapid Business Opening System (Sistema de Apertura Rápida de Empresas, SARE) was created by COFEMER and issued at the federal government level in 2002. It allowed businesses to complete two federal procedures—obtaining a fiscal identification number and name registration—in one business day (with the exception of business operations with environmental implications). Building on this mechanism, some 85 municipalities (representing 36 percent of Mexican GDP) had implemented the SARE by early 2006, reducing the time to start up a business from over 30 days to 1–2 days. The SARE is also considered an important tool in developing compatible systems at the state and local levels. However, the SARE has been only partially implemented in some locations. Moreover, even with the SARE, the entrepreneur still has to register separately with labor and social security authorities. In Peru, obtaining an Operating License used to be the most bureaucratic procedure when starting a business, accounting for 62 percent of the total time. In January 2006, with assistance from the World Bank‘s International Finance Corporation, the municipality of Lima launched a new process that reduced the total time needed from 60 to 3 days. In addition, the number of inspections was reduced from five to just one multipurpose inspection; average visits by business owners to the municipality were slashed from 11 to 2; and costs fell by over 50 percent for small firms. In the first nine months after the reform was introduced, the municipality registered over 8,314 firms, more than in the previous 7 years combined. Five key lessons were learned along the way: Awareness building. To promote reforms at the sub-national level, it is necessary to get the support of the local media to build awareness of the importance of reducing bureaucratic barriers. Municipal officers also toured areas where informal businesses operated where they could observe that, despite existing legislation, the informal sector operated without even minimal safety conditions. Proposal. The reform proposal must include clear messages supported by performance benchmarks, such as outcome and impact indicators, to guarantee a clear understanding of the results that will be achieved with the simplification reform. After the diagnostic analysis, the team in Lima was able to predict the reduction in days, costs, and number of visits by entrepreneurs as well as the increase in the number of firms that were going to register. Those estimates were used by municipal authorities to highlight their leadership and political vision through newspaper headlines. Implementation. To implement the new procedures, municipal personnel need to be technically trained and highly motivated and committed. The reforms required a new working paradigm breaking barriers among various departments to improve the flow of information in the municipality. Sustainability. The private sector can play a useful role in promoting reform sustainability. For example, the formation of a public-private Advisory Council that regularly monitors outcome indicators such as costs, time and requirements to obtain a license, and impact indicators such as number of firms registered. Source: Rada, Kristtian and Ursula Blotte (2007) and World Bank (2007b). 19. Several initiatives to ease business regulations and enhance the effectiveness of public institutions are under implementation. Tax administration, social security administration and customs administration have received great attention. Other reforms could benefit from greater momentum. inequitable processes or are performing work for which they are not qualified; and the prevalence of ―moonlighting‖ among customs officials not only increases the potential for conflicts of interest but also results in shortened work schedules, further undercutting the overall efficiency and quality of service delivery. 7 World Bank (2007b). Uruguay: Institutions Building Technical Assistance Project. Project Appraisal Document. Report No: 39027-UY. 13 Figure 2.9: Average Days to Clear Export Figure 2.10: Average Days to Clear Import Customs Customs Thailand 1.5 Lithuania 2.0 Latvia 2.0 Latvia 3.3 Lithuania 2.5 Ireland 3.3 Ireland 3.0 Slovak Republic 3.9 Costa Rica 3.4 Thailand 4.6 Uruguay 3.7 Mauritius 5.5 Mauritius 4.4 Costa Rica 6.0 Mexico 5.0 Uruguay 7.3 Chile 5.2 Chile 8.1 Slovak Republic 5.8 Mexico 9.3 Argentina 6.1 Argentina 9.6 Brazil Brazil 13.8 8.4 Número de días Núm ero de días Source: World Bank Investment Climate Survey (2006) Source: World Bank Investment Climate Survey (2006) Improving Security 20. While the costs of theft and Figure 2.11: Uruguay. Reducing Crime would Increase security in Uruguay are lower than in Productivity most other high middle income countries in Latin America, there is still Losses from crime scope for improvement (Figure 2.11). Argentina 2.46 On average, Uruguayan firms lose Brazil 2.22 about 1.47 percent of annual sales due M exico Full sample 2.14 2.14 to crime, i.e., losses due to theft plus Latvia 2.03 the cost of security. The analysis on the Costa Rica Slovakia 1.76 1.71 impact of investment climate attributes South Africa 1.52 on firms presented later in this chapter Uruguay 1.47 M auritius 1.35 will also confirm that improvements in Lithuania 1.25 security would increase the productivity Chile 1.25 of Uruguayan firms. The findings from Ireland 0.63 the focus groups conducted with 0.00 0.50 1.00 1.50 2.00 2.50 3.00 Percentage of annual sales entrepreneurs indicate that losses due to theft are owed to internal as well Note: Losses from crime includes losses due to theft and expenditures on security external theft. Enhancing labor Source: Investment Climate Survey (2006) relations (Chapter 4) and risk management controls could help reduce internal security problems. Addressing external crime will be easier in Uruguay than in other countries in the region since it does not respond to organized crime. Data from the World Economic Forum, for example, indicates that organized crime is the lowest among the comparator countries and that the homicide rate is low, especially compared to Brazil, Mexico, and Cost Rica. 21. While a more detailed diagnostic will be necessary to develop a strategy to reduce crime, experience from Latin America and elsewhere suggests that one of the most effective entry-points for crime prevention is the local or municipal level. An effective strategy should be based on comprehensive safety audits or crime diagnostics and combine a three-tiered approach: (1) judicial/policing reform—ensuring that order, fairness, and access to due process is maintained in the day-to-day activities of the community and reducing the fear of crime; (2) social prevention—targeted multi-agency programs that address the causes and associated risk factors of crime and violence; and 14 (3) situational prevention—measures that reduce the opportunities for particular crime and violence problems through urban spatial interventions such as the design of public spaces, housing, lighting, public transport, recreational spaces, and parks. 2.3. INFRASTRUCTURE 22. Overall infrastructure development in Uruguay is currently adequate, but challenges are starting to emerge. These challenges will become more important as demand, especially from the export sector, continue to expand. Broadly speaking, Uruguay ranks well in terms of access to energy, road density, port services, and fixed and mobile lines. Efficiency, however, differs markedly across infrastructure sectors and public services. Also, Uruguay has not kept pace with its regional neighbors and other peer comparators outside the region with regard to new telecommunication services such as broadband internet, Figure 2.12: High Fixed Line Penetration in Uruguay which is important to support the transformation to a knowledge Ireland 496 Costa Rica 316 economy and the emerging IT-related Uruguay 291 industry. Additional gaps may surface in Mauritius 287 other infrastructure areas. It is vital that Latvia 273 Lithuania 239 Uruguay prepares itself to address new Slovakia 232 demand needs. The emerging forestry Brazil 230 industry and its by-products, for Argentina Chile 227 206 example, will generate new pressures in Mexico 174 the existing transport infrastructure, and Thailand 107 developing the supporting infrastructure Telephone Mainlines per 1000 People could enable, Uruguay to develop as a Source: World Bank, World Development Indicators 2006 logistics hub. 23. Infrastructure sectors could benefit from more competition. Where Uruguay has allowed some competition, it has achieved significant gains in efficiency, e.g., in ports and long-distance telecommunication services, and in coverage, e.g., in mobile services. The lack of a clear differentiation between the three independent roles of policy definition, regulation, and operations prevalent in nearly all infrastructure sectors does not foster competition. Similarly, greater autonomy and capacity of regulatory bodies is vital to increase competition, efficiency, and the entry of new market participants. The remainder of this section briefly assesses the performance of Uruguay‘s telecommunications, transport, and energy sectors, and their impact on the investment climate. See World Bank (2007b) for a detail review of the transport and energy sectors. Telecommunications 24. Fixed line telephone penetration is high, and recent growth in mobile telephony has been substantial, assisted in part by healthy competition in the sector since the market opening. Uruguay has a tele-density of 291 fixed lines per 1000 inhabitants, higher than all of the comparator countries except Costa Rica and Ireland (Figure 2.12). According to Administración Nacional de Telecomunicaciones (ANTEL), the incumbent, almost all households in the country have a fixed telephone. 25. The telecommunications sector was partially opened to competition in 2001. Prior to the reform, all telecommunication services in Uruguay were provided by a state-owned monopoly, ANTEL. In 2001, a law liberalized cellular, international and data services and created the Communications Services Regulatory Unit (URSEC). The result is that ANTEL has a monopoly on 15 fixed public switched services (local, rural and domestic long-distance telephone services), while other services, including international long distance, mobile, and data services are opened for competition. Greater competition has led to rapid growth in the mobile telephony, especially during the last two years with the entry of a third operator, AM Wireless Uruguay (Figure 2.13).8 26. By contrast to the high mobile and fixed Figure 2.14: Uruguay. Competition Accelerates line tele-density, broadband penetration is Growth of Mobile Telephony relatively low, due in part to regulatory 70 limitations on competition from cable 60 companies. According to the investment Fixed Lines per 100 inhab. 50 Mobile climate survey data, businesses in Uruguay have 40 far fewer broadband internet connections than 30 businesses in neighboring Argentina (50 percent 20 in Uruguay versus 80 percent in Argentina). 10 Uruguayan firms are also less prone to use email 0 or web communication with clients and 1990 1992 1994 1996 1998 2000 2002 2004 2006 suppliers (Figure 2.14). Further evidence on Uruguay‘s lower broadband penetration relative Source: Internal Telecommunications Union to its peers comes from the International Telecommunication Union (ITU) (Figure 2.15). Broadband is usually given through DSL technology, making ANTEL a de facto monopoly since they own the copper lines to the households/businesses, even though there are several Internet Service Providers in the country that only sell dial-up access. Cable TV Figure 2.13 Use of Email or Web Communication with companies could become competitors Clients and Suppliers (Percent of Firms) in the broadband market, but the Mexico 46 regulator has not issued the licenses Thailand 59 required to provide internet access Latvia 73 through cable modems. The reluctance Uruguay 78 Ireland 89 from the regulator to provide these Lithuania 89 licenses (that could threaten the Slovak Republic 91 Mauritius 92 dominance of state-owned ANTEL in Brazil 92 the broadband market) could suggest a 94 Chile Costa Rica 96 possible conflict of interest between the Argentina 97 regulator and the operator. While broadband prices are difficult to Percentage of Firms compare across countries because data Source: Investment Climate Survey (2006) rates offered in each country and by each operator are different, Figure 2.16 attempts such a comparison. From this figure it appears that prices in Uruguay are a somewhat higher than most of the comparators. Moreover, due to minimal competition, ANTEL does not have an incentive to develop an aggressive strategy for increasing broadband penetration. 27. Less than adequate broadband connections could hinder the expansion of Information Technology Enabled Services and Business Process Off-shoring industries. Increasing the 8 The other two operators are ANCEL and AVIATAR. In 1994, ANCEL began operations as a subsidiary of ANTEL, competing with AVIATAR—a private company that had a ―shared-risk‖ contract with ANTEL by which the private agent operated and managed the mobile business, while the state-owned company remained an owner of the spectrum and received a portion of all revenues. This complicated scheme continued until 2002, when AVIATAR was awarded a frequency of its own and terminated the shared-risk scheme with ANTEL 16 penetration of broadband services in businesses would make them more competitive, but it would also enhance the IT Enabled Services industry in the country, which has started to flourish during the last few years. Uruguay has also witnessed substantial growth in the number of contact centers and Business Process Off-shoring companies being installed in the country. There are ample opportunities for companies in the international outsourcing and off-shoring market, but a solid telecommunications infrastructure and competitive prices are necessary to facilitate the industry‘s development, along with a skilled labor pool. Figure 2.15: Low Broadband Penetration in Figure 2.16: Comparison of Broad Band Prices in Uruguay Compared to other Economies Uruguay (Monthly Prices in US$) Ireland 78 Chile 40 Lithuania 68 Chile 45 Uruguay 28 Colombia 29 Brazil 28 Slovakia 26 Argentina 24 Argentina 27 Latvia 22 Uruguay 19 Colom bia 25 Mexico 18 Costa Rica 19 Costa Rica 11 US$ Thailand 1 Broadband Users/1000 inhabitants Note: Chile, Colombia, Brazil, Uruguay and Costa Rica: 200/256kbps downstream. Argentina: 512kbps downstream (does Source: International Telecommunication not offer 200kbps anymore). Union Source: Company websites 28. A stronger regulatory framework is vital to promote greater competition. A few European countries that have retained state-owned operators have created strong and independent regulatory authorities to prevent capture of the regulator by politically strong and financially dominant state-owned incumbent firms. In Uruguay, the industry could also benefit from a clearer distinction between policy and regulation. It would be desirable to transform URSEC into an autonomous entity, while policy matters could rest with the Planning and Budget Office (or a Telecommunications Ministry as is the case in most countries in the region). Inadequate funding has held back URSEC‘s development, especially its technical resources. URSEC could be financed by the industry, through regulatory fees, which would provide a more stable funding base. In sum, greater autonomy would enhance URSEC‘s impartiality and would level the playing field to increase competition. 29. Competition in broadband deserves to be actively pursued. With respect to broadband penetration, two complementary suggestions are made. First, an unbundling of the local loop for broadband purposes could be implemented. In this way, ANTEL would allow existing ISPs and other operators to use its copper line infrastructure to give broadband access to customers. In this kind of schemes, it is very important to define the access price right: it needs to be low enough to attract ISPs to provide the service at a retail level but high enough to encourage ANTEL to offer the service and expand its own access network. Second, cable operators, which are a natural alternative to the fixed networks, could be allowed to offer internet services introducing some competitive dynamics in the market and providing more alternatives to users. 17 Transport 30. Access to transport infrastructure in Uruguay is high. The road and port sectors Table 2.2: Density and Network Coverage record good levels of service quality and of Road and Railway Networks Density Network coverage efficiency, but the railway sector is not by population by land area competitive. Density and coverage indicators for Country (Km./1,000 people) (Km/1,000 Sq. Km) the road and railway networks are the highest in Roads Railways Roads Railways Latin America, with most of the country‘s Argentina 6.1 0.9 77.5 11.4 population within a short distance of Brazil 10.3 0.2 201.8 3.4 transportation infrastructure. The road network Chile 5.3 0.7 104.9 13.2 provides good accessibility to practically anywhere Colombia 2.7 0.1 99.2 2.8 in the country (Table 2.2). Furthermore, only 4 Mexico 3.4 0.2 168.3 9.0 percent of firms in the investment climate survey Paraguay 5.7 0.0 72.5 0.4 considered transportation to be one of the top 3 Peru 3.1 0.1 60.8 1.3 obstacles to their business. The investment climate also indicates that transportation costs and Uruguay 21.4 0.9 401.4 17.0 losses from breakage and spoilage as a percentage Indonesia 1.5 0.0 162.8 2.8 of shipment value are low in Uruguay compared Malaysia 2.9 0.1 199.8 4.9 with countries in the region (Figure 2.17).9 Philippines 1.0 0.1 111.9 7.9 Thailand 2.7 0.0 673.3 1.6 31. Roads are relatively well-maintained and Source: World Bank staff safe in Uruguay. Uruguay has a total road network of 70,732 km, of which 7,743 km are paved. The small proportion of paved roads out of the total network length reflects the fact that the level and composition of traffic do not warrant a large extension of asphalt roads in the regions outside Montevideo. The share of paved roads in good condition is higher in Uruguay than in neighboring countries (Table 2.3). Roads in Uruguay are also safe: the total rate of road fatalities per 100,000 of population and 10,000 vehicles is among the lowest in the region.10 32. The government’s Table 2.3 : Primary Network Condition, 2001-2003 approach to the road network has been National Percentage National paved road network condition innovative, resourceful and Country (primary) road network (Km) paved Good Fair Bad has achieved significant Argentina 38,484 81% 38% 57% 5% results. It has mainly focused Brazil 92,038 63% 28% 59% 14% on managing roads rather than Colombia 16,528 72% 59% 36% 5% constructing, rehabilitating Peru 17,101 50% 36% 50% 14% and maintaining them. The Uruguay 8,767 89% 77% 17% 6% government has used a wide Source: National road agencies, latest data available range of management techniques including classic concessions, a quasi-public concession, long-term rehabilitation and maintenance contracts, routine maintenance via micro-enterprises, and rehabilitation and maintenance directly executed by the Ministry of Transport and Public Works. Activities have been adjusted to respond to user demand and roads have been managed as commercially oriented assets, 9Transportation costs for the last shipment of the firm‘s main product to its primary destination are 3, 4.5, and 6.8 percent of shipment value for Uruguay, Argentina, and Mexico, respectively. 10 World Report on Road Traffic Injury Prevention, World Bank and World Health Organization, 2004. 18 both of which have helped to focus resources and obtain comparatively better results than in neighboring countries. 33. By contrast, rail transport in Uruguay is not competitive, and railway labor Figure 2.17: Breakage or Spoilage Losses for Export productivity is relatively low. Cargo traffic Shipments density (ton-km/ network in service) is the lowest Uruguay 0.19 among the railway operations analyzed in Table 2.4. The network‘s modest condition restricts Mauritius 0.32 usage to 1600km out of 3000km. Labor Argentina 0.33 productivity—measured by the number of tons carried per employee—within the National Chile 0.35 Railway Administration (AFE) has improved Costa Rica 0.49 slightly due to railway sector reforms in 2002 but falls significantly below that of private rail Mexico 0.86 operators in other Latin American countries. 0 1 AFE‘s revenues barely cover half of its Percentage of Shipm ent Value operational costs. Considering a joint road and Source: World Bank Investment Climate Survey (2006) rail strategy would be desirable, especially to manage the transport of forest products. 34. The extensive forest planting in the 1990s is expected to generate a surge in forest products-- about 2.4 million tons of freight traffic or 50 percent of current freight traffic in the next five years. If this wood is transported by road, the number of trucks expected on the main routes would probably triple, accelerating the damage to pavement and requiring additional investments estimated at US$600 million to strengthen road pavement. Alternatively, if 30 percent of the wood was transported by a truck/rail combination, an estimated US$150 million and US$40 million would be required for the road and rail networks, respectively. The main north-south (Montevideo-Rivera) rail line is already being rehabilitated and holds substantial promise for the transport of forest products. Further studies are necessary to determine the most viable and efficient combinations of road-rail transport throughout the country and necessary investments to meet Uruguay‘s growing trade requirements. The railway sector holds promise but will require a substantial restructuring of AFE to raise its efficiency levels. Otherwise, investments in the sector will not yield returns. A joint-venture between AFE and a private operator could be a suitable framework to facilitate such a restructuring. Table 2.4: Railway Cargo Traffic and Productivity, 2007 Ton-km per km of Ton-km per Ton-km per Ton-km Average distance of Country Operator Table 2.5: Container Trafficin Montevideo Port operational railway locomotive in wagon in service (millions) cargo traffic (Km) (millions) service (millions) (thousands) Annual Growth Year TEUs (%) FEPSA 1,357 458 530 37 680 Argentina 1990 64,286 Mesopotámico 844 618 401 34 622 2000 287,298 35 Bolivia FCO 682 605 548 69 514 2001 301,641 5 Uruguay AFE 331 251 2051 222 278 [1] With a 1611 km network. [2] Considering 8 locomotives. 2002 292,962 -3 Kohon, Background333,871 Source: World Bank (2007), Uruguay Development Policy Review, Jorge 2003 Report. 14 2004 424,791 27 2005 454,531 7 19 35. Ports operations in Montevideo are Source: Montevideo Port webpage; competitive and have grown considerably http://www.anp.com.uy/montevideo/default.asp since the 2002 reform. In 2005, the port of Montevideo handled 454,531 TEU (Twenty-foot Equivalent Units) (Table 2.5). Overall, the number of TEUs has increased seven-fold since the 2002 reform, which resulted in intra-port competition, a decline in port tariffs, and significant new investments from the private sector. The average cost of container transfer in Montevideo is substantially lower than non-concessioned ports in Peru, Ecuador, and Central America.11 The port of Montevideo is fairly competitive relative to other privately operated ports in Latin America. For instance, the cost per TEU in Montevideo is less than in Buenos Aires (Argentina) and Valparaiso (Chile) but slightly more than in Cartagena (Colombia). (See Universidad Politécnica de Valencia, 2003.) 36. Efficiency gains attained through the improved used of the port infrastructure are reaching their limit and the port‘s expansion has become necessary. A project to expand the TCP terminal has been approved which will help to solve the envisioned bottlenecks.12 The port‘s expansion will accommodate vessels of 6000 TEUs compared to the current vessels of 3000-4000 TEUs, resulting in economies of scale and an eventual reduction in users‘ costs. In addition, it will allow the port of Montevideo to increase its activities as a distribution center for the region and serve as a mini- logistics hub. With this expansion, the development of a deep water port in the Atlantic could be put on hold. Technical studies could be conducted to identify potential sites. Market developments over the medium term, however, will determine whether a second expansion of Montevideo‘s port or the construction of a new port will be more suitable. Private financing should be pursued for either alternative. For further details on potential port development see World Bank (2007c).13 Figure 2.18: Losses Due to Power Outages (% of Sales) Energy Uruguay 1.2 Slovak Republic 1.2 37. Electricity services in Uruguay are Lithuania 1.4 Thailand close to reaching universal coverage. Latvia 1.5 1.5 Electricity coverage is already 94 percent. Argentina 1.8 UTE, the state-owned utility with monopoly Ireland Chile 2.0 2.1 in transmission and distribution, expects to Brazil 2.5 provide solar home systems to the dispersed Mexico 2.5 Costa Rica 3.0 rural population still lacking commercial Mauritius 4.0 services in the areas where the extension of Percentage of Sales the conventional distribution grid is not Source: World Bank Investment Climate Survey (2006) economic. If this plan is successful, Uruguay could reach universal coverage in a few years. 11 See Drewry (2006), Drewry Annual Container Market Review and Forecast 2006/07. 12 This project will involve private sector investments of more than US$70 million, the largest investment in the sector in more than 80 years. 13 World Bank (2007c). Eficiencia en Infraestructura Productiva y Provisión de Servicios en Uruguay: Sectores de Transporte y Electricidad. Washington. D.C. 20 38. Uruguay‘s electricity service is also reliable with few power outages but connection Figure 2.19: Wait Days for an Electricity Connection waits are long. According to the investment Slovak Republic 6.0 climate survey, firms in Uruguay experienced Chile 10.4 losses due to power outages equivalent to 1.2 Mexico 14.5 Mauritius 23.0 percent of sales, lower than neighboring Brazil 25.6 countries and all comparator countries outside Thailand 26.4 the region (Figure 2.18). Connection services in Ireland Lithuania 28.0 30.2 Uruguay, however, are slow. The average wait Latvia 31.0 to get an electricity connection was nearly three Uruguay 43.4 and four times that of Mexico and Chile, Argentina 46.0 Costa Rica 48.6 respectively. Only Argentina and Costa Rica Days had longer waits. (Figure 2.19). Electric tariffs Source: World Bank Investment Climate Survey (2006) in Uruguay do not seem higher than in other countries in the region (Figure 2.20). That said, there is limited information on the cost of generation, transmission, and distribution and thus inadequate clarity on how efficient services are. 39. Uruguay is highly dependent on energy imports and is facing increasing challenges to meet the growing energy demand in a reliable and efficient way. Local energy sources mainly comprising hydroelectricity and non-conventional Figure 2.20: UTE Tariffs compared with Average Tariffs of renewable energies are satisfying only 30 REIC enterprises percent of national consumption. Electric generation capacity has barely COMERCIALES 50.000 kWh/mes changed in recent years and is not COMERCIALES adequate to meet demand, while 2.000 kWh/mes Uruguay is facing challenges with its INDUSTRIALES 5.000.000 kWh/mes traditional supply sources in the region. INDUSTRIALES UTE 500.000 kWh/mes Recent experience with unreliable Promedio CIER INDUSTRIALES 5000 supply from Argentina due to energy kWh/mes problems in that country has RESIDENCIALES 1600 kWh/mes underscored the need for other sources RESIDENCIALES 200 kWh/mes of energy. 0 20 40 60 80 100 120 140 160 US$/MWh 40. A comprehensive energy Note: The REIC average excludes Argentina and Venezuela strategy is needed to enhance the Source: Regional Energy Integration Commission, REIC reliability of supply. The following elements are proposed for consideration in the development of such a strategy. Expanding the local generation and transmission capacity and furthering renewable energy sources would be desirable. Among others, biomass, wind power and small hydropower generators have now become more competitive in light of the high oil prices. There is a potential for additional energy savings through more efficient use, and on-going public initiatives to encourage a more rational and efficient use should be further promoted. Finally, an improved regulatory framework is crucial to foster the development of the sector, facilitate the expansion of local generation capacity, promote competition in the wholesale market when the right conditions arise, and facilitate regional integration. The introduction of regulatory accounting would provide greater transparency in tariff setting and allow the identification of new opportunities for cost reductions. Setting tariffs for the use of transmission networks would facilitate the development of renewable sources by the private sector. 21 2.4. TECHNOLOGY AND INNOVATION 41. Investment of Uruguayan firms in innovation is modest. Improving Figure 2.21: Uruguay - Firms’ Expenditures on Innovation Activities product quality and broadening the 138.5 current market were the primary reasons Acquisition of capital goods 91.6 for Uruguayan firms to conduct Design 8.1 11.3 innovation activities.14 The innovation Acquisition of software 6.8 9.8 surveys conducted by the Directorate of 17.2 Internal R&D Innovation, Science, and Technology 6.1 6.3 for Development (DICYT) during the Technology transfer or consulting 3.8 periods 2003-2001 and 2000-1998 Acquisition of hardware 3.8 8.0 indicate that 36 and 33 percent of firms Management 3.3 5.7 2000 2003 undertook innovation activities, Training 5.4 2.5 respectively. Of the universe that 4.7 External R&D engaged in innovation during 2001- 1.7 2003, 58 percent acquired capital goods $0 $5 $10 $15 $20 $25 $30 and 45 percent undertook some sort of US$ Millions Source: DICYT’s surveys on innovation for the periods 1998-2000 and worker training. Slightly over a third of 2001-2003 firms (38 percent) undertook internal research and development (R&D). Technology transfer and external R&D were the least common innovation activities. Indeed, total R&D investment in Uruguay (0.22 percent of GDP in 2002) is very low reflecting less than adequate private sector demand for knowledge and Table 2.6. Uruguay: Correlates of Innovation Inputs(1) technology and insufficient and volatile Quality Foreign R&D certification license investments by the public sector. The dummy dummy dummy country R&D investments fall short of the Exporter 0.083 0.045* -0.043** regional average (0.53 percent of GDP) and Foreign ownership 0.011 0.060** 0.003 the expenditures estimated for a country with its GDP per capita (0.7 percent of Training 0.184** 0.057** -0.007 GDP) Share of skilled production workers 0.011 0.057* -0.011 42. Large firms are more prone to 20-99 employees 0.150* 0.028 0.036* engage in innovation activities and to 100+ employees 0.392** 0.238*** 0.156*** conduct R&D. According to DICYT‘s Log of firm's age 0.023 -0.003 0.025*** survey, during the period 2001-2003, 79 Observations 359 544 343 percent of large firms undertook innovation activities compared to 50 percent of (pseudo) R-squared 0.15 0.23 0.20 medium firms, and 28 percent of small Notes: Reported coefficients are marginal effects in probit regressions. firms. Large firms are also more likely to Significant at 10%; ** significant at 5%; *** significant at 1% (1)Regression includes sectoral dummies undertake R&D as suggested by DICYT‘s Source: Investment Climate Survey, 2006 survey and subsequently confirmed by econometric estimates using the investment climate survey (Table 2.6). Firms investing in training are also more likely to invest in R&D. A similar pattern is found for foreign licensing, which can be an alternative way for Uruguayan firms to gain access to new technologies. The sector also matters. 14 According to the survey conducted by the Directorate of Innovation, Science, and Technology for Development DICYT‘s, these objectives were indicated by 67 percent and 56 percent, respectively, of firms engaging in innovation activities during the period 2001-2003, and were also the two most important objectives during the previous 3-year survey. 22 The econometric estimates confirm that firms in the chemical sector are more prone to invest in R&D, while firms in the garments and textiles sectors are less likely to do so. 43. Firms face a number of obstacles that drive down demand for innovation. According to DICYT‘s innovation survey, the largest obstacles to innovation were uncertainty about the economy and the reduced size of the market Figure 2.22: Uruguay - Obstacles to Innovation (2001-2003) (Figure 2.22). The first obstacle is not surprising since the survey was Uncertainty about the economy 52% 65% conducted shortly after the crisis, Reduced size of the market 52% 56% but the economy has improved Difficulties in accessing finance 33% 43% considerably since and confidence is 36% Risks associated with the innovation gradually recovering. The second 15% 34% obstacle reconfirms the importance Return on investment period 26% of trade integration. Access to Ease of imitation by third-parties 19% 23% finance and investment risks were Lack of trained personnel 21% 20% also highlighted by firms as key Scarce possibilities to cooperate with other firms 15% 19% obstacles. While these findings are Non-innovative firms 14% Innovative Firms Shortcomings in S&T public policies consistent with the innovation 24% 13% literature (e.g., Trajtenberg (2005) Lack of development of institutions related to S&T 16% and Rodrik (2004)), they seem to be 0% 10% 20% 30% 40% 50% 60% 70% more acute in Uruguay as discussed Percentage of Firms in the section below on access to Source. DICYT’s innovation survey 2001-2003. finance.15 Other hurdles listed by firms include lack of trained personnel, scarce possibilities to cooperate with other firms, and shortcomings in science and technology policies and institutions. Lack of trained personnel hampers the capacity of firms to absorb and adapt new technologies, and weaknesses in science and technology institutions limit the conduits available for technology transfer. They also constrain firms‘ possibilities to collaborate with external institutions in R&D activities. Finally, although not captured in the survey, it is widely recognized among leaders in the private and public sectors that the prevailing environment has not encouraged a high risk-taking culture or a business culture that fosters innovation. 44. Quality certification can be a stepping stone in the introduction of new technologies for Uruguayan firms. Standards embody technology and act as channel for technology transfer for the firms that adopt them. Yet, the prevalence of quality certification is modest among Uruguayan firms. According to the investment climate, only 11 percent of firms in Uruguay have ISO certification. Quality certification is not only linked to size but also to exporting activity and foreign ownership. As entrepreneurs have expressed, quality certification is very often a requirement to enter foreign markets. 45. The capacity to transfer and adapt technology in private firms needs to be improved. This demands a marked increase in the number of technicians and researchers working in the private sector. DICYT‘s innovation survey for the period 2001-2003 found that only 1.9 percent of 15 Trajtenberg (2005) and Rodrik (2004), among many authors, argue that innovation activities entail a fundamental information asymmetry where the firm has more intimate knowledge of the innovation and capacity to develop it compared to the external financing agent, such as banks, who lack the capacity to verify it. Financial agents are likely to be skeptical of returns on innovation. The funding gap is less problematic when innovation is undertaken through investments in capital goods since they can be used as collateral. In fact, acquisition of capital goods constitutes the most important form of innovation for Uruguayan firms as noted earlier. Other types of innovation are subject to more serious financing constraints. 23 employees in the manufacturing industry are involved in R&D, of which only 0.16 percent are dedicated full time. It would also be effective to develop or strengthen institutions that can be dedicated to the identification of new technology sources for local firms and their promotion and adaptation to local needs. The Technological Laboratory of Uruguay (LATU), for example, is well- respected for its work related to product testing and certification, but has been much less successful in identifying sources of technology and adapting them. Furthermore, links between demand and supply of research and technological services deserve to be enhanced. In recent years, a few examples of successful public-private research partnerships have emerged in Uruguay, e.g., the programs promoted by the University of the Republic in pharmaceuticals and software testing and a few mixed projects by the National Institute for Agricultural Research (INIA). However, for the most part, these examples remain isolated cases. It is crucial to scale up cross-sectoral interactions in Uruguay and to make them more systematic and effective. This will entail an improvement in the incentives for researchers to work across sectors, adequate funding for mixed research projects, and stimulating private sector demand for knowledge services. In the short-run, much can be achieved by promoting technology transfer programs for SMEs in coordination with training programs as further discussed below. 46. The government has given high priority to innovation and has started to address shortcomings in the innovation system. The government‘s strategy comprises four key pillars. The first envisions a strengthening of the policy and institutional framework, which has already resulted in the creation of the National Research and Innovation Agency and an Innovation Observatory will be established shortly. The second focuses on investments in human capital and high-quality research teams, and the third supports the productive sector, inter alia stimulating private sector demand for innovation, developing and strengthening technology transfer institutions for priority sectors, and fostering overall linkages between supply and demand for research and technology transfer. The fourth pillar supports pro-poor innovation. International agencies are supporting the government in these efforts. Long-term sustainability and continuity of these policies will be vital to their success as highlighted by international experiences such as Finland and Ireland. See World Bank (2007b) for a further discussion on innovation challenges and the government‘s strategy. 2.5. SKILLS DEVELOPMENT AND LABOR MARKETS 47. Skills, labor relations and labor regulations are essential ingredients of the investment climate in Uruguay, and a sound investment climate is necessary for employment generation. Despite the quick economic rebound from the crisis, the expansion of formal employment has not kept the same dynamism. The level of informality and unemployment remain a concern. Labor groups with workers and entrepreneurs were conducted to hear first hand their concerns about labor market and labor relations and avenues for building a strong engagement. These views were complemented with the findings from the investment climate survey and the expanded household survey conducted in 2006 to analyze labor market constraints and the impact of skills, in particular training, on firms‘ productivity and workers‘ earnings.16 48. During the focus groups, workers and entrepreneurs shared an appreciation for healthy labor relations and both expressed concerns about slow employment creation and deficiencies in workers‘ skills (Boxes 4.2-4.4 in Chapter IV). Areas where there is not a common understanding also emerged, for example, on aspects concerning labor regulations. Developing strong labor relations is 16 See Annex IV for a discussion of the focus group methodology. 24 a national priority. Uruguay can draw on its history of tripartite dialogue to renew this collaboration. Successful international experiences on strong labor and employer relations such as Ireland and Finland could also offer some guidance. In these two cases, a strong partnership between workers, employers and the government has been the cornerstone of their national competitiveness strategies that have helped turned around their economies and create high-value jobs. 49. Strengthened labor relations would be the basis for achieving more flexible labor regulations that are balanced with adequate social protection. The ICA analysis indicates that rigidities in some labor regulations may have encouraged informality (Chapter IV). Wage Councils regulations may have increased the likelihood of informality in some sectors. Providing information to wage councils on the effects of regulations on informality and adjusting regulations to sectoral needs is important. Regulations on working hours appear restrictive by international standards. Enhancing the flexibility of working hours would be welfare enhancing especially for women since it would facilitate their participation in the formal labor market. In addition, severance payments are costly and constitute the most important obstacle to hiring and firing according to the investment climate survey. One option for consideration would be to enhance the unemployment insurance scheme, which is a superior social protection instrument, and make the severance payment scheme more flexible to reduce its costs on employment generation. This matter merits further investigation. 50. The quantitative analysis of the firm and household surveys indicates that training has a positive impact on workers‘ earnings and firm productivity and is complementary to technology (Chapter IV). Yet, Uruguayan firms underinvest in training relative to comparator countries. The gap is most acute among SMEs. Only 8 percent of workers receive training among small firms versus 36 percent among large firms. It is vital that training policies tackle market failures, supporting the integration of vulnerable groups in the labor market and enhancing the productivity of employed workers. Instruments to reach these two groups might be different. From the analysis, several recommendations emerge: (i) reduced duplication among agencies supporting training programs and more participatory training, involving firms and unions in the design of training programs; (ii) higher emphasis on training programs that provide transferable skills versus non-transferable skills; and (iii) provision of public incentives to foster training by firms, especially targeted at SMEs and combined with other tools to promote firm productivity such as technology transfer. Chapter 4 provides a detailed analysis of labor market and skill development challenges. 2.6. FINANCE 51. Enterprises’ access to financial markets is far more limited in Uruguay than in countries with similar characteristics. The IC survey indicates that, on average, retained earnings account for 82 percent of fixed assets and 73 percent of working capital financing, respectively, of enterprises (Figures 3.2-3.3 in Chapter III). Even though banks are adequately capitalized and there is liquidity in the system, bank financing accounts for less than 7 percent of working capital. SMEs tend to be more financially constrained. The survey suggests that the limited use of external financing is due to both low demand and supply constraints. Perhaps the low demand responds to the fact that some enterprises remain indebted or have become risk-averse after the crisis. As the survey indicated, despite notable improvements, macroeconomic instability remains one of the highest concerns of enterprises. For those that demand but do not receive credit, lack of collateral appears to be the main constraint. Uruguayan enterprises have to pledge more collateral as a percentage of their loans than enterprises in many comparator countries, which could be explained by the banks‘ perception that lending to enterprises is highly risky and/or the low level of collateral recovery in Uruguay in cases of creditors‘ default. Being incorporated and audited increases the 25 probability of accessing credit according to the survey. These findings are not surprising since incorporation and auditing increases the transparency of information, although auditing practices in Uruguay can be further enhanced. 52. Bank credit to Uruguayan enterprises is remarkably low. Banks have recovered from the 2002 financial crisis and are liquid and adequately capitalized, but credit for consumption seems to be recovering faster than credit to enterprises. This is not fully surprising given the much higher interest rates charged on consumer loans and the fact that it is easier to standardize consumer loans. SMEs risks are more opaque17. As observed in other post-crisis environments, banks may have also become more risk averse following the crisis. In addition, legal and institutional factors increase the risks of lending to enterprises. The mechanism for forced collection of collateral is not effective with lengthy judicial enforcement procedures, and registries for property and pledges are not yet fully computerized, hindering lenders from assessing the validity of the information. There is scope for improving the adequacy and reliability of financial information due to the diversity of accounting standards and limited number of audited enterprises. 53. Uruguay’s capital markets play a minor role in investment financing. Pension fund administrators (AFAPs) are rapidly accumulating funds (US$2.3 billion by end-2006), but these are largely invested in government issues. Some of the challenges that limit enterprises‘ access to bank credit also constraint their access to the capital markets: investor‘s rights are weak and corporate governance is modest. Improving the regulator‘s enforcement capacity would also be beneficial. 54. The government is taking important measures on both the demand and supply sides of the financial markets to improve enterprises‘ access to finance. Parliament is considering a new Insolvency and Corporate Reorganization Bill which is critical to facilitate the exit of non-viable firms and the financial restructuring of those with a potential. The authorities are drafting a new Capital Markets law that would strengthen investors‘ protection, corporate governance, and the powers of the regulator. A number of initiatives to enhance the reliability of information are in progress inter alia the development of new accounting and auditing procedures. Some of these measures, however, will apply to a limited market since the number of listed firms is likely to remain small and SMEs constitute the bulk of enterprises. Hence, special measures could be necessary to support SME financing. For example, special simplified accounting rules and a special corporate code of conduct could be developed for non-listed enterprises, and their compliance could be monitored by peer groups. Greater transparency could narrow information asymmetries and facilitate credit. Minimizing judicial intervention in the execution of guarantees would also help increase the recovery rates and demand for such high levels of collateral. 55. These regulatory measures will likely not suffice and more proactive market development policies, especially to support SMEs, could be considered. The government is contemplating using the Corporación Nacional de Desarrollo (CND) as a development agency to help develop the credit and capital markets through the structuring and managing of financial and guarantee funds. Recovering market confidence and building the necessary enabling environment will take time, and there is scope for transitory pro-market government interventions to improve access to financial services and promote the development of capital markets18. The CND, however, was sometimes employed in the past as an instrument to rescue companies in problems which affected its public image. A new management has been put in place and has developed a strategic plan. Among the new products, it is developing a partial guarantee scheme for SMEs. The securitization of SMEs assets holds promise 17 Discussion of these issues can be found in de la Torre, Gozzi, and Schmukler (2006). 18 De la Torre, Gozzi, and Schmukler (2006). 26 and could also be explored drawing from experiences in Latin America and around the globe. For these strategies to be effective, it will be desirable to strengthen the CND‘s corporate governance and enhance its accountability19, guarding the institution from political interference as experienced in the past. Access to finance challenges are further analyzed in Chapter III. 2.7. WHAT IS THE IMPACT OF THE INVESTMENT CLIMATE ON FIRM PRODUCTIVITY? 56. An econometric analysis was designed to yield concrete details about the effects of the investment climate on firm performance and workers‘ earnings. The econometric methodology of Escribano and Guasch (2006) was used as a starting point, but it was adapted by limiting the initial potential set of explanatory variables to those that are both objective and can be interpreted unambiguously. In addition, the regression analysis covered the Uruguay survey data alone as well as the Uruguay data pooled with survey data from its peer countries.20 The latter is likely to yield more efficient estimates because if investment climate conditions affect all the firms in a country in a similar fashion, then it is likely that individual analysis will not be able to capture the relevance of particular investment climate factors. Pooling the data also allow to conduct simulation exercises to estimate the productivity gains that Uruguayan firms could attain if investment climate conditions were to improve to a certain benchmark within the pool of countries. For additional details on the methodology, see Annex 2A. 57. The econometric analysis indicates that the investment climate impacts labor productivity of Uruguayan firms. In particular, regulatory compliance (as measured by the share of sales declared for tax purposes), theft and security costs, and technology use (as measured by use of website) are strongly correlated with labor productivity. (See column (3) in Table A.2.4 in Annex 2A.) Lower regulatory compliance has a negative effect on labor productivity because it is likely to capture the quality of regulations and of institutions that implement them. As presented earlier, there is scope for substantially reducing red tape, and the quality of responsible public institutions is often not strong enough. The costs of theft and security also negatively affect firm performance. The variable on access to finance has a positive but not significant coefficient in the regression with Uruguayan firms alone and is significant in the pooled regression. Interestingly, for the investment climate variables, most coefficients are not statistically different between the regressions with only Uruguayan firms and the regressions with the pooled data. This confirms to a large extent the assumption that, after controlling for the industry, size composition of the sample and other unobservable aspects, the effects of the investment climate on an ‗average‘ firm‘s productivity is the same across the countries in the sample (Table A.2.4 in Annex 2A). 58. Wages are positively correlated to training and negatively correlated to crime and the costs of security (See column 5 in Table A2.4 in Annex 2.A). Similar results are found with regards to training based on the analysis of the Uruguay household survey data (2006) in Chapter IV. Also, the estimates show a positive correlation between wages and specific firm characteristics, in particular foreign-ownership, exporting, and age. 19 Scott (forthcoming). 20 Thailand was not included due to lack of comparable questions on governance. 27 59. The analysis suggests that substantial productivity gains are possible if the investment climate improves. To measure potential gains, three benchmarks were used: (i) the highest 75th percentile of the entire sample within the same industry and Figure 2.23: Total Labor Productivity Gains from Moving to 75th Percentile of the Sample size; (ii) the average in Chile; and (iii) the average in Mexico 32% Ireland.21 In the first Brazil 22.4 simulation, the benchmark is a Uruguay % 22% hypothetical firm located at the Costa Rica 20.6 75th percentile of firms within Mauritius 17.9 % the pooled sample. The % 17.4 Argentina difference between the % benchmark value and Latvia 10.8 % Uruguay‘s industry-size average Lithuania 10.2 % for each investment climate Slovakia 7% variable was calculated. All Chile 6.9% industry-size gains were then Ireland 3.2% averaged to obtain one number percentage gain per country. The calculations Crime & Security Costs Regulatory Compliance Access to Credit Internet Use Training indicate that Uruguayan firms Source: Authors‘ estimates based on Investment Climate Surveys, World Bank. could increase their labor productivity by 22 percent if the country reached the top 75th percentile for investment climate attributes related to regulation and governance, access to finance, human capital formation (as proxied by training), and technology (as proxied by web use) (See Figure 2.23). The largest gains could be achieved through improvements in technology use followed by regulatory compliance, reductions in crime and security, and enhanced human capital formation. Gains from access to finance would be more modest. 60. Overall, similar labor productivity gains were found with Chile as a benchmark (Figure A2.1B in Annex 2A). Using Ireland as a benchmark, the productivity gains would be even larger (29.7 percent). These econometric results highlight the potential gains from improvements in regulation and governance, training, and technology use. However, they might underestimate the importance of some investment climate attributes. Proxies within the regression may not have fully captured the desired investment climate characteristics. This is especially likely for access to finance, where evidence from other sources suggests that their importance in the econometric analysis is understated. Are Exporters Different? 61. Exports can be a key driver of growth, especially given Uruguay’s small domestic market. Numerous studies have found a positive correlation between exporting and firm productivity. Whether causality runs from higher productivity to exporting (i.e. self selection) or exporting leads to higher productivity (i.e. learning by exporting) is the subject of much debate in the literature.22 Evidence on the former seems to predominate, although they are not mutually 21 Due to differences in the industries covered in each country‘s survey, in (ii) and (iii) the overall country average was used as a benchmark, instead of taking averages by industry and size. 22 The self selection hypothesis is that exporters are more productive before they begin to export, and that being more productive is what encourages them to enter the export market. The ‗learning by exporting‘ hypothesis states that 28 exclusive.23 The literature also suggests that country size affects export premium, most likely because firms are able to better exploit economies of scale by exporting. 62. The econometric analysis also found that there is an exporter premium in Uruguay and exporters do not seem to be affected differently by the investment climate. A regression analysis found that, after controlling for several firm characteristics, exporters have 47 percent higher labor productivity and 45 percent higher average wage than non-exporters (Table A.2.5 in Annex IIA).2425 When estimating exporter premia by size, the econometric analysis found that the premia was more significant for small firms relative to medium firms.26 One possible explanation for this finding is that productivity differences between exporters and non-exporters arise from economies of scale, and scope and thus are mainly related to size. For small firms, only those that are highly productive relative to their peers are able to compete in export markets. Using data from Uruguay‘s Manufacturing Survey conducted by the Instituto Nacional de Estadística (INE) for 1988-1995, Casacuberta and Gandelman (2007) also find that exporters are more productive than non-exporters and that 80 percent of the output differential is due to differences in size. These findings suggest that exporting can allow Uruguayan firms to benefit from economies of scale and reiterate the importance of trade integration for Uruguay. A separate analysis of the investment climate survey investigated whether exporters suffer from investment climate conditions more or less than non- exporters. Results for exporters are not statistically different than the results for all firms, except for the audited variable in the labor productivity regression (Table A.2.7 in Annex IIA). 2.8. CONCLUSION 63. The analysis conducted confirms the importance of the investment climate for productivity in Uruguay. The impact varies according to the investment climate factor. Uruguay fares relatively well on aspects concerning integrity, a valuable public good that can make the country an attractive place for new foreign direct investment within its regional location. However, on other governance fronts (especially on regulatory and institutional quality), access to finance, and technology, there is substantial scope for improvement. Upgrading workers‘ skills will also enhance productivity. In addition, new vulnerabilities are surfacing in infrastructure, especially in energy. Addressing these challenges will be crucial to promote sustainable growth and create the conditions for Uruguay‘s integration in the global economy. Uruguay ranks 71 out of 80 countries on intensity of local competition,27 the local market is small, and opening external markets will allow Uruguayan firms to benefit from economies of scale and achieve higher productivity. 64. ―Red tape‖ is a problem and the effectiveness of institutions shows room for improvement. ―Red tape‖ has an adverse impact on productivity and reduces the benefits of participating in the formal economy. A number of regulatory reform initiatives are planned or under implementation, but many of these could benefit from greater impetus. A widespread culture of exporters start out with the same productivity as non-exporters, and that exposure to foreign (more competitive) markets puts pressure on firms to become more productive or be selected out. 23 Casaburta and Gandelmlan (2007) conduct a review of 54 studies that examine the productivity premium associated with exports and find that in 86 percent of the cases exporters are more productive than non-exporters and that in most of the studies there is some evidence of self-selection. Only 60 percent of the studies test the learning by exporting hypothesis, and the evidence on this is mixed. 24The ‗self-selection‘ versus ‗learning by exporting‘ hypotheses could not be tested due to the lack of time-series data since the investment climate survey was conducted in Uruguay for the first time, 25 When estimating exporter premia by size, we found that it was more significant. 26 Results for large firms were not robust due to the small sample size. 27 See World Economic Forum (2007). 29 business friendly regulations deserves to be more actively promoted. International experience shows that efforts to establish a business friendly environment are most successful when promoted by public-private commissions. Crime appears to be a smaller problem than in other countries in the region. Nevertheless, it remains a cost to firms and a special strategy to address this problem is necessary. Other country experiences indicate that the most effective entry-point for crime prevention is the local or municipal level. 65. Access to financial markets is limited in Uruguay. The government has undertaken a series of measures to foster financial sector stability and improve the enabling environment for expanding access to finance. Many of these reforms are still under implementation. It could take time before financial agents recover their confidence and the requisite enabling environment is in place. Thus, a strengthened CND could play a transitory role in facilitating SME finance and spurring market development (Chapter III). 66. Improving the capacity to transfer and adapt technology in private companies is important. Addressing these issues has become a policy priority, and the government‘s strategy to spur innovation contains elements to tackle these concerns. Long-term sustainability of these policies will be vital to their success. The results also indicate that training and technology are complementary. Training also enhances productivity but firms underinvest. A wide range of programs to support training are available. Efforts could be better integrated and streamlined with a focus on enhancing their impact and fostering more investment by firms, especially SMEs. Labor relations deserve to be strengthened. The focus groups with workers and entrepreneurs show promise indicating that they both value constructive relations. Future relations can build on traditional tripartite relationships (Chapter IV). 67. New gaps are emerging in the infrastructure sector that will only grow as market demand expands. The expansion of infrastructure will benefit from an enhanced regulatory framework, greater competition, and new market participants. In particular, it is critical to diversify and strengthen the reliability of energy supply sources, promote more energy efficiency, and enhance the regulatory framework to invite new market players. A joint road and rail development strategy is necessary to respond to the needs of the expanding forestry sector. The railway sector holds promise for the forest sector, but increasing its efficiency requires a substantial restructuring of AFE. More could also be done to encourage new telecommunication services such as broadband internet, which is important to support the transformation to a knowledge economy and the emerging IT-related industry. Introducing competition in this area will be vital to its expansion as observed in the mobile sector. 30 CHAPTER 3. ARE URUGUAYAN ENTERPRISES FINANCIALLY CONSTRAINED? Abstract: Financial stability has been restored. Yet, the modest use of external financing is remarkable, especially for SMEs. The problem stems from both demand and supply issues. Perhaps firms are still over indebted or have become very risk averse in the crisis aftermath. If macroeconomic stability and economic growth continue, demand for finance is likely to recover with time. Banks may have also become more- risk averse. In addition, legal and institutional factors increase the risks of bank lending to enterprises and do not foster capital market development. A series of measures to address legal and regulatory gaps are underway and deserve to be supported and accelerated, in particular the capital markets law, the draft Bankruptcy law, and the plan to enhance market transparency. An active engagement and dialogue with market agents would facilitate the design and implementation of these reforms and development of additional initiatives to further financial sector growth. It could take some time for the necessary institutional environment to be built and for the confidence of financial agents to be fully recovered. The CND could play a transitory role in facilitating SME finance and market development. It is vital to strengthen its corporate governance to avoid past problems. 3.1 INTRODUCTION 1. Following the devastating crisis of 2002, financial stability has been restored in Uruguay and overall, the banking system is well capitalized today. External credit to Uruguayan enterprises, however, continues to be more modest than in dynamic high middle income economies. Capital market financing to the productive sector is even more scant. In particular, small and medium enterprises, which constitute the bulk employment, largely operate on retained earnings. Expanding availability of external financing is crucial for enterprises to grow and compete. Figure 3.1. Firms with Credit or Overdraft Facility (percent of total) 31 2. Using the investment climate survey and other sources, this chapter 100 89.0 90.6 analyzes access to finance by enterprises 79.0 80.0 85.5 in Uruguay and factors constraining it.1 76.2 76.6 The investment climate survey indicates that 63.9 limited enterprise financing responds to 48.3 48.6 50.7 both demand and supply constraints. The chapter is structured as follows. Section I 23.6 presents an analysis of the main financing sources of enterprises and their demand for 0 external finance. Section II analyzes s au e constraints on access to finance from the Ire y ca ia Th zil th o a ge d a nd ic iu il in ua ic ni n tv Ri a Ch bl la rit la nt ex ua Br La ug pu ai a M supply side of the financial market and st Ur Re M Ar Li Co ak reviews public initiatives already underway ov Sl to overcome these constraints. Section III Source: World Bank Investment Climate Survey presents conclusions and recommendations and proposes additional policy interventions to expand finance with a special emphasis on SME finance. 3.2. ACCESS TO FINANCE FROM THE DEMAND SIDE 3. Broadly speaking, enterprises’ access to financial markets tends to be more limited in Uruguay than in other high middle income countries. A simple comparison of the share of firms that report having a credit line or overdraft facility across a group of countries comparable to Uruguay2, reveals that not very many Uruguayan firms make use of financial markets. Indeed, Uruguay is among the bottom three countries (with Mexico and the Baltic countries), with 64 percent of firms having either a line of credit or an overdraft facility by contrast with other small high middle income countries like Thailand (85 percent) or Chile (89 percent). These general trends are confirmed by regression estimates that are further discussed below. 4. The most important source of finance for firms is retained earnings with bank financing playing a minor role (Figures 3.2 and 3.3). Bank financing for working capital was 6.9 percent of total financing and less than 4 percent for fixed assets—substantially lower than levels available in most other comparator countries (Figures A.3.1-A.3.2 in Annex III). Suppliers‘ credits are relatively more important than bank financing, at least for working capital, a pattern that is common to other countries with modest access to bank credit. Equity financing was also modest. Figure 3.2 Fixed Assets Financing Figure 3.3 Working Capital Financing 1 This chapter benefited from other Bank studies and work including the Bank Financial Sector Assessment (FSA, June 2006), the Report on the Observance of Standards and Codes (ROSC) on Insolvency and Creditors´ Rights, the ROSC on Corporate Governance, the ROSC on Accounting and Auditing, the Western Hemisphere Credit Reporting Initiative (WHCRI) analysis of the credit reporting systems, the Western Hemisphere Payments and Securities Settlement Initiative (WHI) analysis of the payment systems, and the Uruguay Programmatic Reform Implementation Development Policy Loan (PRIDPL I). 2 Either because they are in the same income group as Uruguay, or because of similarities in the composition of economic activities. (In graph MUS = Mauritius, ZAF = South Africa). 32 90.0 80 82 73 15 7 4.9 5.2 3.8 0.5 2.6 0.7 0.3 0.8 0 0.1 1.6 0.4 0.6 0.8 0.0 0 k s k r s k s ew s s k ng er er on s s an bt s cto an d or an r g er e nd St ank nk an rm me ien ar in h m ni de b th eb ct eb se Ot -b ie rn ba n- sto r sh /fr se o O b ea /f r al at at No st ea u i ly e ew e rm iv al St ed cu r/C at ily at N N ed Pr m N in iv r/ fo m Fa lie in ta lie Pr In fo Fa pp Re a pp In et Su R Su Source: World Bank Investment Climate Survey Source: World Bank Investment Climate Survey 5. What are the main firm characteristics determining access to credit? Based on the investment climate survey data, a probit selection model was used to estimate the determinants of the probability of obtaining a loan, conditional on having demand for credit, i.e., the firm has credit, it applied and was turned down, or it did not apply for a reason other than ―I do not need credit‖ (Box 3.1). As many as 69 percent of Uruguayan firms did not apply for a loan and within this group 55 percent indicated that they did not need it (Figure 3.4). The estimation exercises were conducted for the Uruguayan firms alone as well as for the pooled data from Uruguay and its peer countries, where country and industry dummies were also included. A similar exercise was subsequently conducted to estimate the determinants of ―financing depth‖, a categorical variable that contains a count of how many financing products a firm reports. Box 3.1. Estimating the Determinants of Access to Credit A probit selection model is used to estimate the determinants of the probability of obtaining a loan, conditional on having demand for credit. The dependent variable (having demand for credit) is equal to one if the firm has credit or if it applied and was turned down, or if it did not apply and it provided a reason other than ―I do not need credit‖. Alternatively the variable is equal to zero if the firm has no credit and it did not apply for any credit arguing that it does not need it. Having constructed an ―outcome variable‖ (having credit) and a selection variable (having demand for credit), the determinants of access to credit are estimated. using a two-step selection model technique. The variables affecting the demand for credit and the probability of obtaining credit are: being an SME (i.e. having less than 100 employees), being an exporter, being externally audited, being partially or entirely foreign owned, the log of the firm‘s age, and having a limited liability status (incorporated). In order to identify the model, variables that affect the demand for credit but not the probability of obtaining credit are needed. Thus, two dummies are included: one for firms that provide training to their permanent workers and another for firms that consider macroeconomic stability to be a major obstacle to their activities. The latter variable aims to capture the macroeconomic environment that affects a firms‘ perception about its future prospects. The estimation exercise is conducted for the Uruguayan firms alone as well as for the pooled data from Uruguay and its peer countries, where country and industry dummies were also included. A regression was also estimated for the determinants of ―financing depth‖, a categorical variable that contains a count of how many financing products a firm reports. One point is added for each of the following: having access to credit (loan, credit line or overdraft facility), bank financing for working capital, bank financing for investment, and issuance of equity. The results of the regressions for Uruguay alone and the pooled sample are presented in Table A.3..2 in Annex III. 33 Figure 3.4. Reasons for not Applying for Credit Figure 3.5. Reasons for Loan Rejection (percent of all firms not demanding credit) (percent of all credit applicants that are rejected a loan) Not applied Collateral 75.2 Others No need 53.6 Others 26 Profitability 63 0.7 15.1 6.9 Too complex Expect not approved 9.4 2.6 procedures 1.5 15.8 0.3 5.8 0 Uruguay. Correlations between Collateral Value Table 3.1.Debt level too high Credit history Loan size High interest rate and other Loan Characteristics Collateral Incomplete Predicted Collateral value Maturity Profitability profitability Loan size application Note: ―Not applied‖ is estimated as a percent of all firms. Other Collateral value categories are estimated as percent of all firms not demanding credit. 1 Source: World Bank Investment Climate Survey, 2006 Source: World Bank Investment Climate Survey, 2006 Loan size 0.0545 1 Maturity 0.1565* 0.0775 1 6. According to our quantitative Profitability Predicted -0.0253 -0.0953 0.1075 1 analysis, SMEs have less demand profitability 0.1563* -0.0533 -0.0287 0.2677* 1 for credit, lower probability of Note: * significant at 10%; ** significant at 5%; *** significant at 1% obtaining it when needed, while Predicted profitability is estimated as the fitted value of the regression of the (log) audited and older firms have profit/capital ratio on country, industry, size, dummies, plus dummies for ownership status measures such as incorporated and foreign ownership. greater demand and access to Source: Investment Climate Survey credit. SMEs also have less access to a diverse range of financial products. Externally audited firms offer more reliable information, mitigating information asymmetries and facilitating external financing. Older firms maybe also have a more stable growth path than younger and smaller firms reducing financing risks. For Uruguayan firms alone, exporters also seem to have a higher demand for credit, but no greater access, while foreign owned firms have less demand probably because they have alternative sources of funding abroad. For the pooled sample, exporters and foreign owned firms generally have better access to financial markets. The regressions also confirm that Uruguayan firms generally have less demand for credit and access than many other comparator countries after controlling for sector, size, and other firm characteristics (Table A.3.2). Our regression results also indicate that access to finance and investment are positively correlated (Table A.3.3) in Annex III. 7. The low demand for credit in Uruguay may be explained by an increased risk- averse attitude following the crisis, the presence of highly leveraged firms, or entrepreneurs uncertainty about the macroeconomic scenario (Figure 2.2). The family structure of most businesses in Uruguay, does not motivate the leverage of own resources for business expansion and investment opportunities. A number of enterprises that had to restructure their debts after the crisis may be reluctant to seek further credit. High interest rates do not seem to be a reason for lack of credit demand, since lending rates in Uruguay, particularly in dollar denominated loans, are modest in real terms and have remained fairly stable as further discussed in the next section. The investment climate survey also confirms that interest rates are not holding back credit demand. Only 15.8 percent of firms indicated that they did not apply because of high interest rates (Figure 3.4). 34 Figure 3.6 Value of Collateral (as percent of loan value) 8. Lack of collateral constitutes the prime reason for the rejection of credit applications. 240 According to the investment climate survey, for 200.6 enterprises that applied for bank credit and were 159.8 rejected, the main reason for the rejection was 149.9 141.8 140.0 136.3 130.4 125.2 lack of collateral (45 percent) and then poor credit 115.0 111.2 104.3 history (9.1 percent) (Figure 3.5). High levels of 88.5 indebtedness and low profitability were not important, most likely because firms in those conditions do not apply for credit. 0 9. Collateral requirements tend to be l ay ia ia a nd le o nd zi ia a s ic ic tv tin ak hi iu ra an gu la la R ex La C rit B ov en ai uv Ire ru ta M au Th Sl rg os U th higher in Uruguay (Figure 3.6) compared to M A Li C peer countries. An OLS regression was Source: World Bank Investment Climate Survey conducted to estimate the determinants of the ―collateral pledged as percent of the loan value‖ for Uruguay and the pooled sample. The regression results indicate that the collateral value is not significantly correlated to any of the firm characteristics that affected access to credit such as size, age, ownership status, or being externally audited. Collateral value is significantly correlated to loan maturity but not loan value (Table 3.1). The results also confirmed that collateral requirements (as percent of the loan value) are higher in Uruguay than in comparator countries after controlling for firm and loan characteristics and sectors. In the pooled regression, all comparator countries, except Lithuania and Argentina, had coefficients that were negative and significant suggesting that in Uruguay requested collateral as percent of loan value tends to be higher. This pattern can be explained by the banks‘ perception that lending to Uruguayan enterprises is risky and/or the lower level of collateral recovery in Uruguay in cases of creditors‘ default. 3.3 SUPPLY SIDE CONSTRAINTS 10. Although the banking Figure 3.7. Banking Credit Remains below system is liquid, the use of bank Pre-crisis Levels credit by Uruguayan enterprises is very modest. In the wake of the Banking Credit (as % of GDP) devastating financial crisis, the banking system has only cautiously 80 renewed lending to enterprises. The 70 regulatory and institutional 60 50 environment constrains the ability of 40 financial institutions to expand 30 finance in a prudent manner. In 20 addition, capital markets are still 10 Jan-1999 Jul-1999 Jan-2000 Jul-2000 Jan-2001 Jul-2001 Jan-2002 Jul-2002 Jan-2003 Jul-2003 Jan-2004 Jul-2004 Jan-2005 Jul-2005 Jan-2006 Jul-2006 small and concentrated on public debt issues. Source: Central Bank of Uruguay Bank Credit to Enterprises 35 11. The Uruguayan banking system has recovered from the severe crisis of 2002,3 but it is now smaller with reduced lending portfolios. The banking sector is now comprised of large, Figure 3.8. Uruguay. Credit to the Private public banks and a small number of mostly Sector (1985-2005) foreign owned banks. (See Table A.3.1 in Annex III).4 Improvements in the macroeconomic context since 2003 have contributed to the Credit to Private Sector 100 recovery of the banking system. Banks have 80 increased their liquidity (43 percent of total assets at the end of 2006 compared with 14 60 percent at end of 2001), reduced duration 40 mismatches, and are now adequately capitalized. Equity capital accounted for 16.8 percent of risk- 20 weighted assets at end-2006, compared to a 0 1985 1990 1995 2000 2005 reported 9 percent at end-2001. Asset quality has improved with non-performing loans at about Firms Families 3.6 percent of total loans. They are also well Source: Central Bank of Uruguay provisioned. Commercial banks‘ assets are growing, having reached about US$14 billion at end-2006, but are still below the pre-crisis level of US$21 billion. Bank lending to the private sector dipped from about 75 percent of GDP before the crisis, to 22 percent at end-2006, reflecting the decline in deposits, the rise in demand for liquid assets, and the write-off of bad assets (Figure 3.7). The current high liquidity of the banking system could allow extensive credit supply. However, banks are constrained in their ability to lend, among other reasons, because new prudential regulations limit foreign currency lending to non-foreign currency earners. Yet, deposits are still mostly in dollars. 12. Banks direct a significant share of new lending to family versus enterprises’ credits. Higher interest rates for peso denominated consumer loans than for corporate loans are an incentive for banks to concentrate in consumer lending. The average interest rate on peso denominated loans to families has dropped from about 55 percent at end-2004 to about 38 percent at end-2006, but remains much higher than the 10.2 percent average interest rate charged on loans to enterprises. This may explain why consumer credit accounted for 63 percent of total new credit in local currency as of December 2006. The average interest rate charged by banks for dollar denominated loans has remained stable, ranging between 7 and 8 percent, despite the steady decline in Uruguay‘s risk premium (the difference between the average lending rate and the LIBOR rate for 180 days) from about 700 basic points in December 2003 to around 200 basic points since early 2006. In addition, the introduction of innovative lending techniques such as scoring methods has also facilitated the growth in consumer lending and the development of standardized products at a lower cost. By contrast, the risks of SMEs tend to be more opaque and heterogeneous making it more difficult to rely on scoring methods (De la Torre et al 2006). 13. According to the investment climate survey, private banks are a more important source of bank credit to enterprises than state banks, especially for working capital (Figures A.3.3- A.3.4). However, this gaps narrows for fixed assets financing, which requires longer maturities. 3 The withdrawal of nonresident deposits—accounting for over 50 percent of total deposits at end-2001—due to the crisis in Argentina triggered a general run on deposits in Uruguayan banks. Four insolvent banks were closed and some large depositors lost part of their savings. The cost of resolving the banking crisis was estimated at about 20 percent of 2002 GDP. 4 The two large public banks are the Bank of the Oriental Republic of Uruguay (BROU) and the National Mortgage Bank (BHU). The latter is no longer authorized to take deposits excluding peso savings plans. 36 14. To compensate for lack of bank credit, other lending mechanisms are starting to be developed but at a gradual pace. The use of financial instruments such as leasing and factoring is growing. Only financial enterprises regulated by the Central Bank can engage in financial leasing. A total of 17 entities (including 14 banks) had registered leasing transactions by the end of 2006. There are no restrictions on factoring, and any lender can discount receivables. Banks provide this service and they all discount post-dated checks. Cooperatives emerged as sources of credit for small enterprises during the crisis. The cooperative sector has since then contracted5, accounting now for about a tenth of one percent of financial sector assets. Legal and Institutional Constraints to Bank Credit in Uruguay 15. Improvements in the legal and institutional framework could facilitate bank credit in Uruguay. The most concerning gaps have been the tax regime, the legal protection of creditors‘ rights, and inadequate information on enterprises. The financial crisis exposed many of these deficiencies spurring a number of vital reforms, some of which were recently implemented but others are in progress or have yet to be launched. Among recent accomplishments, the enactment of a new tax law in late 2006 is expected to reduce the cost of credit. It eliminates a number of unusual taxes that applied to financial intermediaries in addition to the profit and net worth taxes, such as the asset tax (IMABA) and an additional tax on the credit portfolio (ICOSIFI). However, the value added tax (VAT) applied on interest paid on bank loans remains and has only been reduced from 23 to 22 percent.6 16. There is still scope for improving the legal protection of creditors’ rights and further decreasing the risk of lending to enterprises. The outdated and fragmented insolvency legislation does not provide an efficient means for creditors to obtain the collection of their claims. Insolvency proceedings are not commonly used in Uruguay because the current legislation regulating corporate insolvency neither facilitates the reorganization of viable businesses, nor allows the efficient liquidation of non-viable ones. Moreover, the current system acts discourages the early disclosure of problems by enterprises. Aware of these challenges, the government submitted in mid-2006 to Parliament a draft Bankruptcy Law that seeks to provide a new framework for the reorganization and liquidation of enterprises. The approval of the bill, expected by end-2007, would greatly foster early disclosure of financial problems, the recovery of distressed businesses and the efficient liquidation of non-viable enterprises. 17. Although the legislation governing security interests is adequate, the forced collection mechanism would benefit from higher effectiveness and lower costs. This may partly explain the higher collateral requirements imposed on Uruguayan firms, as was noted earlier.7 BHU is the only institution that has the right to out-of-court executions, which reduces the process to a four- month period until the auction is conducted. The Trust Law approved in 2003 allows for the constitution of guarantee trusts, which circumvent the not very efficient enforcement of pledges. 5 Reasons for the contraction in the cooperative sector include more stringent regulations and the closure in 2006 of the most important institution representing two third of the cooperative system, after experiencing important solvency problems. The institution was purchased by a Venezuelan Bank (BNADES). 6 The tax reform has eliminated the VAT exemption on consumer loans (crédito social), which applied only to BROU and discriminated against private banks. 7 Enforcement proceedings are very lengthy in Uruguay, especially in the phase of the auction of the executed assets. Executory proceeding where defenses are raised can take up to two years, and 3 or 4 more years can elapse from the judgment until effective collection. There is no out-of-court enforcement system applicable to pledge credits to lessen the delay of judicial enforcement procedures. 37 The use of this type of trusts is slowly increasing. Alternative out-of-court mechanisms also deserve to be considered. 18. The banking system cannot fully rely on the financial information of enterprises aggravating information asymmetries between lenders and creditors. Uruguay‘s statutes on accounting and auditing are based on an outdated framework. Changes have been introduced by the government and the professional bodies since 1991, but on a piecemeal approach, creating duplication in certain areas and gaps in others.8 The government has initiated a program to consolidate and modernize the legal framework for accounting and auditing and to strengthen the technical capabilities of the regulatory entities with a view to improving enforcement.9 19. The databases of companies’ financial statements have limited information useful for creditors. The National Registry of Commerce could support the financial system and bank lending by maintaining an updated database of the companies‘ bylaws and other critical information. As noted in Chapter II, the registry has been operating with limited financial and human resources and cannot offer up to date and relevant information. It would benefit from increased capacity to monitor quality of data and sanction non-compliance. The Auditoría Interna de la Nación (AIN), responsible for maintaining bylaws and updated financial statements of large corporations, faces similar challenges. Despite the passage of regulations on the disclosure of financial statements in 2001, compliance remains modest at about 40 percent. AIN has the legal authority to enforce compliance but needs greater capacity to do so effectively.10 20. Similarly, lack of interconnection of property and commercial registries diminishes their effectiveness. In addition to the National Registry of Commerce, the Dirección General de Registros (DGR) operates other key registries such as the property registry. The system currently allows queries through the internet on the national registries of personal acts and commerce, and on the registry of property of Montevideo. The DGR is in the process of implementing a Single Registry System which will interconnect and allow for remote queries on all the registries it operates. The modernization of the AIN and the DGR will be very helpful to streamline the cost of starting a company and enhance the transparency of financial information on enterprises and on property pledged as collateral. 21. Databases on debtors offer a great untapped potential. The Central Bank operates a public credit registry that receives and disseminates information from regulated entities, but it is only used for supervisory purposes and focuses on negative information. There are also two privately run debtor databases, which are fragmented, shared only among participating creditors, and constrained to negative information. As a result, banks have scarce information to assess risk and apply adequate risk methodologies. Moreover, banks tend to use the same credit rating process for all companies without separate processes for large, medium and smaller companies, implying excessive administrative costs. This situation is starting to change slowly, as some private banks are trying to capture part of the SME segment and introducing new technologies such as ―credit scoring‖. Banks‘ 8 Some of the International accounting norms (IAS) started to be implemented since 1991. Some of International Financial Reporting Standards (IFRS) were adopted in 2004, but the international standards have been extensively modified and the latest standards need to be adopted. 9 The programs seeks to enhance the technical capability of the regulatory entities to keep abreast with the most current versions of the IFRS, strengthen audit practices through enforcement actions and an enhanced professional licensing process, and adopt legal definitions for the audit of financial statements of listed as well as non listed enterprises (other than banks) that comply with international standards. 10 The legal framework establishes that corporations have to present evidence of filing statements with IAN when submitting their tax statements. 38 management of credit risk is likely to benefit from improvements in the information collected and shared by the public and private credit registries. Changes to the legal framework already allow for sharing of positive and negative information by public and private registries. However, the issuance of necessary regulations and the development of a protocol for exchanging information between the Central Bank and the private registries is pending.11 These improvements would help the regulated financial debtors to expand their credit services to small firms that are currently only served by non- regulated credit institutions. 22. Higher penetration of electronic payment instruments would increase the information available to differentiate good borrowers. Cash and checks are the main means of retail payments in Uruguay. The modest use of electronic payment instruments such as direct debit means that credit reporting systems do not have easy access to data on certain type of payments (such as utilities), which are useful indicators to establish the payment pattern of individuals. Capital Markets at a Nascent Stage 23. Uruguay’s capital markets play a modest role in private investment financing. Capital markets are small and dominated by sovereign debt issues. The 2002 crisis shattered confidence in the market at a time when it was barely taking off. Outstanding equity and corporate debt in December 2006 was US$191 million, substantially below 1999 levels. The stock markets‘ capitalization in Uruguay, about 2.5 percent of GDP, is low even for emerging markets standards, and turnover accounts for a small share of market capitalization (Figure 3.9). 24. Capitalized pension funds have grown rapidly, but their Figure 3.9. Uruguay’s Capital Markets in International Perspective (2005) portfolios are overly concentrated on public debt. As of end-2006, individual capitalized accounts had reached US$2.3 billion, about 14 percent of GDP, from less than US$0.5 billion in 1998. Most of these funds are invested in government or Central Bank debt (82 percent). Banks‘ certificates of deposit account for only 9 percent, and investments in trusts, bonds and equities for a mere 4 percent. Some pension funds have reached the maximum investment allowed in government issues and are rapidly accumulating cash given the Source: World Development Indicators Database limited domestic supply of instruments and tight restrictions on investments abroad. Insurance companies are the second most important institutional investors with investments amounting to 3 percent of GDP but face similar investment constraints as pension funds. 11The legal framework for credit reporting system has substantially improved with the enactment of two specific laws. The first one, approved in 2004, recognizes the right to know, update and correct information in private and public credit databases, and the second one passed in January 2006 clarifies that bank secrecy in Uruguay does not apply to bank assets. 39 25. Laws on investments funds, factoring, and trust funds, which are the basis for asset securitization, are in place. Nevertheless, an increased use of these instruments would be beneficial. Public offering of corporate bonds almost disappeared after problems with the largest issuer in 1998 and the 2002 financial crisis. Table 3.2: Outstanding Securities Investment funds also had a promising start but nearly December 2006 (US$ million) faded in the wake of the financial crisis, and no Total Issues Public Sector 11,608 securitizations were made under the Securitization and Total Issues Private Sector 445 Factoring Law. The approval of the Trust Law (Ley de Negotiable debentures 208 Fideicomisos) in 2003 generated high expectations based Equity 191 on the extraordinary development of structured Financial trusts 23 finance12 in other Latin American countries, where asset backed securities worth over US$12 billion were Investment certificates 21 Source: Banco Central de Uruguay (BCU) (2006) issued in local markets in 2005, particularly in Mexico and Brazil. However, only a few highly structured securitization deals have been completed in Uruguay in recent years. 12Structured finance involves the pooling of assets and the subsequent sale to institutional investors of claims on the cash flows backed by these pools. An important characteristic is the delinking of the collateral assets pool from the credit risk of the originator, through the use of a special purpose vehicle, the trust fund. 40 Legal and Institutional Constraints to the Development of Capital Markets 26. Overall, there are challenges in the regulatory and institutional Table 3.3. Protecting Investors in 2006 framework that have not favored Table 3.3. Protecting Investors in 2006 Investor capital markets development. As Disclosur Director Shareholder Protection Investor Country e Index Disclosur Liability Index Director Suits Index Shareholder Index Protection noted earlier, Uruguay lies behind Argentina Country 6 e Index 2 Liability Index Suits6Index 4.7 Index comparable high middle income Argentina Brazil 6 5 2 7 6 4 4.7 5.3 countries in terms of creditors' and Brazil Chile 5 8 7 6 4 5 5.3 6.3 shareholders‘ rights (Table 3.3). The Chile Costa Rica 8 2 6 5 5 2 6.3 3 modest level of disclosure stands out. Costa IrelandRica 2 10 5 6 2 9 3 8.3 The capital markets‘ legal and regulatory Ireland Latvia 10 5 6 4 9 8 8.3 5.7 Latvia Lithuania 5 6 4 4 8 6 5.7 5.3 framework does not provide sound Lithuania Mauritius 6 6 4 8 6 9 5.3 7.7 guidelines on the disclosure of Mauritius 6 8 9 7.7 Mexico 8 5 5 6 information by the issuers and reporting Mexico Slovakia 8 2 5 4 5 7 6 4.3 standards by the intermediaries. The Slovakia Thailand 2 10 4 2 7 6 4.3 6 legal treatment of economic groups is Thailand Uruguay 10 3 2 4 6 8 6 5 weak hindering control of insider Uruguay 3 4 8 Note: Higher values indicate greater investor protection. 5 trading. Source: Doing Business Database (2007), www.doingbusiness.org Note: Higher values indicate greater investor protection. Source: Doing Business Database (2007), www.doingbusiness.org 27. Corporate governance rules could be strengthened to increase the credibility of issuers. Most often, boards in Uruguay play a secondary role in the governance of listed companies. Majority shareholders generally dominate the appointments of the boards, which in turn tend to be controlled by management. The number of independent directors could be increased. Public disclosure of non- financial information such as ownership structure and related party transactions for listed companies offers room for improvement.13 Internal auditors (sindico) could contribute more to good governance but their practices would need to improve in line with international standards. Uruguay could benefit from stronger minority shareholders‘ rights, enhanced disclosure of related party transactions, and more independent and active boards. Enforcement powers and technical capacity of the Securities Market Regulator, a department of the Central Bank could be strengthened to encourage the development of efficient and transparent markets and the restoration of public trust in them. The current law does not explicitly allow the regulator to impose fines on issuers, and its authority and jurisdiction overlap with those of the self-regulatory stock exchanges.14 Insufficient financing has constrained its technical capacity in the past. 28. To spur capital market development, the government has formulated a strategy based on four complementary pillars. The first pillar consists of a new capital markets law that will strengthen minority shareholders‘ rights and expand the authorities of the Securities Market Regulator to regulate and supervise the market, particularly in the areas of corporate governance for listed companies, licensing standards for intermediaries, and disciplinary powers. The draft law is under advanced preparation. There are also plans to amend the Company Law to strengthen corporate governance practices. The second pillar will increase funding and technical capacity of the Securities Market Regulator and move towards a risk-based supervision framework. The third pillar will upgrade certification standards and support professional training of key market intermediaries such as traders, dealers, issuers, and professional bodies. The fourth pillar will seek to enhance the market infrastructure, especially clearance and settlement arrangements. 13 Uruguay. Corporate Governance Country Assessment ROSC. January 2006. 14 Ibid. 41 29. The recently approved tax law has reduced the tax burden but has not eliminated the tax bias towards bank versus debt financing. Under the previous tax law, interests paid on bank loans to enterprises could be fully deducted for the calculation of the borrower‘s corporate income tax, but there were ceilings on the deduction of interest paid by the issuer on corporate debt. The new tax law removes this ceiling if the debt is issued through public subscription and the instruments are listed in the stock exchange. Few enterprises, however, can meet listing conditions. Harmonizing the tax treatment for bank debt and corporate debt financing would be beneficial. 30. Differences in tax treatment will affect the development of structured finance. Companies can deduct interest paid on debts contracted with financial trusts. However, if financial trusts issue certificates of participation instead of debt certificates, the interest paid is treated as a dividend and cannot be deducted from the corporate income tax. This different tax treatment does not follow a robust justification since the economic results of both types of certificates are the same. A VAT is imposed on debt contracted with financial trusts. In addition, the tax treatment of financial trusts is subject to high discretionality allowing the government to exonerate them of taxes on a case by case basis.15 Such discretionality generates market asymmetries and delays the formation of new financial trusts. 31. Structured finance in Uruguay has also been limited by regulatory gaps. More than 80 trust companies have been registered since the enactment of the Trust Law, but only a few highly structured deals requiring special laws and tax concessions have taken place.16 There are no cases of simple securitizations of commercial assets or credit cards, or short term securitizations, common in Argentina and other countries such as Chile and Mexico. Structuring a financial trust can take up to two and a half years in Uruguay due to the lack of standardized contracts and a lengthy approval process. Financial trusts are subject to complex regulations which are costly to meet. The market regulator has not speedily approved such transactions due to resource constraints and concerns that these trusts could be used to shield assets from creditors. Lowering and simplifying the tax burden, standardizing contracts, and strengthening the capacity of the regulator to analyze the special features of these contracts would facilitate a wider use of financial trusts. 3.4 CONCLUSIONS AND POLICY IMPLICATIONS 32. The Uruguayan financial system has not recovered to its pre-crisis levels, but the government is taking important measures to restore confidence and promote market development. As a result, the capacity to regulate and supervise the banking system is much stronger today, and banks are adequately capitalized. A number of additional initiatives are underway or planned to enhance the enabling environment for financial intermediation, especially to strengthen creditors‘ and investors‘ rights and improve market transparency and information available on enterprises. Most noteworthy among these are the draft Bankruptcy Law, currently under discussion in Parliament, the capital markets law under preparation, and the plan to enhance market transparency. An active engagement 15 A law approved in May 2007, for example, exonerates financial trusts from the corporate income tax and net worth tax if they are structured under the Inter American Development Bank (IADB) Global Multisectoral Loan with resources from that loan. 16 The first securitization in Uruguay under the Trust Law was that of UTE (the electricity company) for US$25 million in December 2004, guaranteed with the stream of future payments of UTE‘s clients. Two other securitizations have been recently structured with the intervention of the Municipality of Montevideo (IMM). One deal is to issue debt for up to US$8.4 million to finance the expansion of a highway by a concessionaire, guaranteed by future incomes by the IMM. The other deal involves issuing debt for US$22.5 million to finance urban transporters in Montevideo, through the securitization of contributions to a fund of 5 percent of the sale of transport tickets. 42 and dialogue with market agents would facilitate the design and implementation of these reforms and development of additional initiatives to further financial sector growth. 33. The approval of the Bankruptcy Law will be critical to allow for the orderly exit of non- viable firms and the financial restructuring of those with a potential. These legal reforms would enhance creditors‘ rights. In parallel, an accelerated procedure for execution of collateral such as out-of-court settlement could be considered. If this is not deemed feasible, current procedures could be modified to make them more abbreviated, focusing on ways that involve the least possible judicial intervention.17 Also, the courts could benefit from more resources that would allow them to hire qualified staff and modernize equipment and thus speed up proceedings. 34. The enactment of the draft capital markets law is recommended. It would lay the foundations to strengthen the enforcement capacity of the Securities Market Regulator, enhance disclosure and practices of issuers and market intermediaries, and discourage conflicts of interest such as insider trading. To be effective, legal reforms have to be complemented with programs to strengthen the capacity of the Securities Markets Regulator. 35. On-going initiatives to enhance the quality of information are also vital to market development. Effective financial intermediation relies on tools that help mitigate information asymmetries. The on-going program to reform the regulatory framework for accounting and auditing constitutes a critical step in this direction. The modernization and automation of the National Registry of Commerce, the Property Registry, and the AIN could also be accelerated. Notwithstanding that some progress has been achieved, the practices and quality of information of these institutions could be improved in order to meet international standards and allow financial intermediaries to easily access critical financial information on enterprises and on property pledged as collateral. 36. In addition to on-going reforms, additional measures can be implemented to foster access to finance, especially to SMEs. In Uruguay, the number of listed firms is likely to remain small, and as a result, many of the proposed reforms on accounting and auditing standards and corporate governance would apply to a limited market segment. In this vein, special simplified accounting rules and a corporate code of conduct could be developed for non-listed enterprises and SMEs. The development of peer groups would offer a possibility to certify compliance with these codes of conduct. The expectation is that banks would place less emphasis on real guarantees for loans to enterprises that comply with these codes of conduct and accounting standards. To create incentives for banks to take into account their clients‘ compliance with these standards, the banking regulator could possibly reduce provision requirements on loans to enterprises that meet minimum standards. 37. The effectiveness of the existing credit reporting system could be easily improved. The Central Bank could issue regulations on the sharing of positive and negative information by public and private registries. Also, a protocol for exchanging data between the Central Bank credit registry and the private credit reporting industry could be developed. These improvements would help expand credit services by the regulated financial institutions to small firms and the self-employed, some of whom are currently limited to non-regulated credit institutions. 17The latter, for example, could be limited to the formal analysis of the guarantee or the issuance of an order to dispose the debtor of the guarantee. 43 38. The securitization of SME-related assets merits serious consideration.18 Banks and non- bank financial institutions such as leasing and factoring firms that provide financing to SMEs can finance themselves through the securitization of their SME portfolios of loans, leases and receivables. Securitization can expand the portfolios of financial institutions without having to increase their capital requirements. Securitization of SME-related assets requires a vast pool of cash- flow generating assets, which is not yet available in Uruguay. Multi-originator securitization of SME- related assets could be a way to reach the critical mass needed to cover the cost of the transaction and attract institutional investors. The Government can foster securitization by streamlining regulations and the approval of structured financial products and offering guidelines to standardize the basic characteristics of the SME-related assets. The Government could also consider the provision of partial-credit guarantees to reduce the financial cost of these transactions and provide comfort to institutional investors, until they become more familiar with this type of transactions. Box 3.2 below presents three international examples of structured financed for SMEs. The Corporación Nacional para el Desarrollo (CND) could play a role in the development of similar products in Uruguay as further discussed below. Is there a Role for Pro-active Market Development Policies? 39. Given the depth of the financial crisis, rebuilding the confidence of financial agents could take time. The reform initiatives discussed above are critical to help create the regulatory and institutional environment necessary for private financial agents to renew active lending and financial markets to expand. Building an institutional environment akin to OECD standards could take some time. While full market confidence is renewed and the institutional environment is built, there is scope for some transitory government interventions to improve access to finance and promote the development of capital markets.19 Demand for finance is also likely to increase over the next few years if the economy recovery is sustained. These government interventions would need to be selective and take place in collaboration with the private sector to promote financial development. In sum, pro-market activism would need to foster the development of the market, rather than seek to substitute it.20 These interventions could seek inter alia to facilitate financing to SMEs through partial guarantees or structured financed products. 40. In this line, the Government plans to give the CND a more prominent role in the development of financial markets. Until recently, the CND had basically operated as a venture capital fund and credit provider for companies of its own group, and as fund administrator. Many of these forays resulted in heavy losses for the CND and tarnished its image as a financial agent. Under new leadership, the institution has prepared a strategic plan for 2006-2010 that emphasizes its role as development agency through four main lines of action: (i) a second-tier banking activity through regulated and non-regulated financial institutions; (ii) the development of new instruments to facilitate access to credit (e.g., a guarantee fund); (iii) improving the quality of information on enterprises; and (iv) supporting institutional coordination for enterprise development projects. It would retain a residual venture capital activity. The CND is making efforts to expand its financing sources through local and multilateral financing sources. 18 For a detailed exposition of the use of securitization to enhance SME financing see Cheikhrouhou et al. (forthcoming). 19 A. de la Torre, J. C. Gozzi, and S. L. Schmukler. ―Innovative Experience in Access to Finance: Market Friendly Roles for the Visible Hand?‖. April 19, 20006. 20 This approach differs from interventionist views held in the 1950s and 1960s throughout many emerging economies that called for an intense public involvement in the allocation of financial resources through direct interventions rather than interventions focused on working with and developing the market. The general experience with such interventions in emerging markets was not very successful (De la Torre et al. (2006). 44 Box 3.2. International Examples of Structured Financed for SMEs Case 1. Structured Finance for SMEs Receivables: Design of a Special Purpose Vehicle in Peru Large corporations that sell to SMEs find it difficult to discount their receivables through a ‗Factor‖ because of the uncertainty regarding the risks associated with these enterprises. The corporations, however, tend to have good knowledge of their SME clients and could potentially generate a portfolio of SME related receivables attractive to investors. Securitization of these receivables could expand the provision of credit to SMEs by corporations at lower rates and longer terms. In Peru, the World Bank is helping design a structured finance program to securitize SME receivables by medium or large firms. The mechanism will operate as follows:  The receivables portfolios of SME clients are analyzed using a financial platform that acts as a back office to manage the risks associated with the SMEs. This platform: (i) allows the connection between corporations and financiers; (ii) uses a web-based tool to evaluate and administer the credit risk of the SME clients and the accounts receivables; and (iii) incorporates the functions of credit bureau, credit rating and credit scoring.  A Special Purpose Vehicle (SPV) is established in a bank to securitize the portfolios of SME receivables grouped or segmented by their risk characteristics under the operating platform. The SPV mobilizes funds from banks and institutional investors issuing bonds and/or participations. The novel aspect of this structured finance is the use of the financial platform to analyze and administer the risks associated with the SMEs. Case 2. Partial Guarantees on Securitization: The case of FTPYME in Spain The Fondo de Titulizacion de Activos Pyme (FTPYME) was created in 1999 to support the securitization of SME loans from commercial lenders. By securitizing loans, lenders obtain liquidity and capital relief that can be used to originate more SME loans. The program provides a partial credit guarantee from the Spanish Treasury to qualifying transactions, to reduce the cost of the transactions and add comfort to the institutional investors.  The program requires that at least 80 percent of the securitized portfolio be loans to SMEs. Originating banks are required to use at least 80 percent of the proceeds to provide additional loans to SMEs.  The program provides a partial guarantee to AA-rated tranches within a given transaction. It guarantees pools of assets from multi-originator transactions including different types of assets such as loans, leasing receivables, and commercial mortgages. The market for the securitization of SME-related assets in Spain reached E8.8 billion in 2004, and FTPYME securities are widely accepted by local and international investors. Case 3. Structured Finance for Working Capital Financing: The Case of FIRA in Mexico FIRA, Agricultural Related Trust Funds, created by the Mexican Central Bank, has developed a structured finance program, involving Ocean Garden (a large shrimp distributor), shrimp producers and feed suppliers, and private banks. Ocean Garden signs supply agreements with individual producers and advances working capital finance. Credit rights are then transferred to a trust fund and sold to banks. The transaction helps to deal with information asymmetry problems. Ocean Garden provides know-how in screening and monitoring producers. The pooling of debt obligations allows banks to diversify their risks and avoid exposure to a specific producer. Banks do not face Ocean Garden‘s credit risk. SPV bankruptcy is remote. To align incentives, all industry participants provide liquid guarantees to cover initial credit losses. Producer and feed suppliers provide guarantees for specific loans covering initial credit losses up to a certain level. Ocean Garden provides a general guarantee covering initial credit losses up to a certain level. Once these guarantees are exhausted, FIRA provides a guarantee that partially covers second losses. Source: H. Cheikhrouhou et. al. (forthcoming) and others. 45 41. The CND is already working on the implementation of a National Guarantee System (GNS). Interested financial institutions will join the GNS and use it when the firm can not provide the requested guarantees. The GNS will cover up to 50 percent of the credit. Credit guarantees could expand lending to SMEs by reducing the risk exposure (or perception of) risk exposure by banks. International experiences with credit guarantees, however, have been mixed depending on incentives granted. CND could benefit from the experiences of FOGAPE (Fondo de Garantía para Pequeños Empresarios) in Chile (Box 3.3) and FNG (Fondo Nacional de Garantías) in Colombia, two of the most successful guarantee funds in the region. 42. Governance changes would help the CND become a more effective instrument for financial market development. Notwithstanding the very positive changes brought about by the new management, the CND‘s public image has been affected by its historical performance. Historically, it was sometimes used as an instrument to rescue distressed companies or support crisis resolution in the banking sector. Strengthening its governance and accountability would mitigate the risks of possible political interference in the future, enhance its credibility vis-à-vis the market, and make recent institutional changes more sustainable. International experience indicates that well governed development banks has rested on the following elements: a focused mandate; board independence through the introduction of independent and professional board members; clear rules for appointment and dismissal of board members21; high accountability of the board; strong public disclosure and annual submission of audited financial statements to Parliament; and the creation of risk management and other internal committees (Scott forthcoming). Box 3.3. Partial Credit Guarantees: The case of FOGAPE in Chile FOGAPE is a state-owned fund, established in 1980 to provide partial guarantees to credits issued by commercial lenders to small, medium and micro enterprises. The rules for FOGAPE‘s guarantees include:  Eligible firms must have sales within established limits and meet creditworthiness standards. Loans to be guaranteed should not exceed 3,000 UF (about US$4,800) and mature within 10 years.  Guarantees are allocated through a closed auction held several times a year. Each bidder submits a bid with the amount of wanted rights and the maximum desired coverage rate over the total amount lent. Guarantees are given to financial institutions that request the lowest coverage rates, until the total amount auctioned has been assigned. There is a limit to the guarantee (80 percent or 50 percent) and to the amount guaranteed for each lender. FOGAPE charges a fee and commissions between 1 and 2 percent depending on the lender‘s riskiness.  FOGAPE does not evaluate the creditworthiness of the borrower or its performance. Currently there are 16 participating financial institutions. The default rate is similar to the default rate for the whole banking system. FOGAPE seems to be financially self-sustainable. Source: www.fogape.cl 21 Board member should only be removed for just cause to safeguard independence. 46 CHAPTER 4. LABOR MARKETS, SKILL DEVELOPMENT, AND PRODUCTIVITY IN URUGUAY Abstract: Labor productivity as well as labor market dynamics and efficiency are crucial elements for improving the Uruguayan Investment Climate. Unfortunately, cyclical factors (particularly the 1999-2002 economic crisis) together with structural components have been responsible for the slower pace of Uruguay’s labor productivity and labor markets improvements. Recent events suggest that labor relations can be improved to the benefit of the labor market and employment generation. This chapter complements several data sources and the findings from the focus groups with workers and entrepreneurs to shed more light on these issues. Several key findings emerge. First, both workers and entrepreneurs value strong labor relations and future collaboration can build on traditional tripartite collaboration and the wage councils. Second, strong labor relations are important to modify rigidities in labor regulations, in particular strict working hours and severance payments, while maintaining adequate social protection. Fine-tuning wage regulations would also be desirable. Finally, policies and programs on training can be streamlined focusing on developing transferable skills and promoting more active investment by firms, especially SMEs. 4.1. INTRODUCTION 1. Labor productivity as well as labor market dynamics and efficiency are essential ingredients of the Investment Climate in Uruguay as reported in Chapter 2. During the last decade, Uruguayan labor markets have been characterized by high levels of unemployment, informality, and volatile earnings. These characteristics have been linked to cyclical factors including the latest economic crisis, but also, structural factors such as regulations and workers‘ skills. This chapter focuses on those structural factors and explores the role of labor market relations and regulations, as well as skills development, on employment generation, informality, firm productivity, and earnings‘ growth in Uruguay. Figure 4.1: Unemployment and Growth in Uruguay 2. Various data and methods 20 are combined to provide a view of 15 the labor market from both the demand (firms) and supply side 10 (workers). Data sources include the Percentage 5 World Bank ICA Survey (2006), 0 World Bank Doing Business -5 Indicators (2006), and the Expanded Household Survey, which contains a -10 skills and training module, conducted -15 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 by the Instituto Nacional de Estadística (INE) (2006). Moreover, special Year Unemployment rate Grow th in GDP per capita (annual %) focus groups with workers and entrepreneurs were conducted to Source: Socio-economic database for Latin America and the Caribbean (SEDLAC) and Word Development Indicators understand their views on critical topics such as job creation, labor 46 regulations, labor relations, and informality.1 The remainder of the chapter is organized in four sections. Section 4.2 reviews recent trends in the labor market identifying the need for greater employment generation and reverse the growing informality trend. Section 4.3 analyzes the impact of labor regulations and relations on the labor market, especially on informality and productivity. Section 4.4 analyzes the impact of skills development on workers and firms, the determinants of firms‘ investments in training, and public policies to foster workers‘ training. The last section offers some concluding remarks and policy implications. 4.2. RECENT LABOR MARKET TRENDS 3. During the last decade, the Uruguayan labor market has been characterized by high unemployment and informality levels, and volatile earnings. In general, it has followed the economic cycle with systematic deteriorations during recessions but has seen more modest improvements during recovery periods. A number of structural factors, which are explored later in this chapter, help to explain these downward rigidities. Changes would be desirable to help revert stagnant labor market conditions to the benefit of workers and competitiveness of firms. 4. Unemployment had fallen to 11 percent in mid-2006. Although this constitutes a substantial reduction from its peak of 17 Figure 4.2: Informality and Growth in Uruguay 2 percent in 2003, it does not yet seem enough to reach a desired equilibrium in the labor 50.0 market from efficiency and a welfare 40.0 standpoint (Figure 4.1). Unemployment is 30.0 Percentage particularly harsh among adult women, those 20.0 younger than 25, and middle-age laid-off 10.0 0.0 workers due to the industrial restructuring. -10.0 5. Despite being a major concern for -20.0 1989 1992 1995 1996 1997 1998 2000 2001 2002 2003 2004 2005 the Uruguayan government, informality Year has registered no reductions during the Informality rate Grow th in GDP per capita (annual %) last 15 years (Figure 4.2).3 Similarly to Sources: Authors‘ estimates based on Uruguay‘s Encuesta Nacional unemployment dynamics, increases in de Hogares (ECH) and ENHA, and World Development Indicators. informality during recessions have been larger than reductions during recovery. Average growth elasticity of informality during the 1989- 2005 period was -0.42 percent, reaching -2 percent during the height of the economic crisis. 1 Six focus groups with workers and entrepreneurs were conducted. Only unionized workers participated, which could have possibly introduced a certain bias in the results. Unionized workers tend to participate in larger firms. According to the Investment Climate survey, workers are unionized in 37.5 percent of large firms compared to 10.2 percent in medium firms. See Box A.4.1 in Annex IV for additional information on the focus groups and methodology followed. 2 Women were also more disadvantaged in terms of unemployment duration. On average, unemployment lasted 3.4 months among women and 2.9 months among men (ENHA, 2006). See Table A.4.1 in Annex 4 for additional information on unemployment duration. 3 There are several operational definitions of informality, given that, by nature, informal firms and workers tend to be under-registered. According to the International Labor Organization (ILO), informal workers include all salaried workers in micro-enterprises (less than 10 workers for Uruguay), non-professional self-employed and zero-income workers. Another operational definition of informality focuses on social protection of workers. This dimension of informality has also registered an increasing trend in Uruguay (see Rofman and Lucchetti, 2006). 47 Informality levels in Uruguay are lower than in Argentina, Mexico and Brazil, although higher than in Chile and Costa Rica and other high middle-income countries in Eastern Europe such as 6. Latvia. Informality growth in Figure 4.3: Distribution of Informal Workers across Uruguay between 1992 and 2004 was Economic Sectors the second highest in Latin America (World Bank 2007c) and remains a concern to workers, the government and the private sector as noted in Agric./cattle/forest./hunt., 14.1% Chapter 2. The resilience of Domestic services, 19.1% Fishing, 0.2% Mining, 0.2% informality points to rigidities in the Other comm., soc., activ./services, 4.8% Manufacture, 11.2% labor market, which could be partly caused by labor regulations. Indeed, Social and health services, 2.5% Electricity, gas, water, 0.1% rigidities in labor regulations are T eaching, 1.2% Pub. adm and defense, soc. Construction, 8.8% commonly cited in the literature as sec. plan., 0.1% Commerce, vehicle/home factors creating incentives to enter the Real state activities, 5.9% appliance repairs, 25.4% Financial services, 0.3% informal sector. The possible causes T ransport, storage, communication, 4.3% of informality are further explored in Hotel and restaurants, 1.9% the following section. Informality in Uruguay is concentrated among men, Source: Own estimates based on microdata from ENHA, 2006. the less educated and those not benefiting from training, those earning less than half the minimum wage, and rural workers (Figure A.4.1 in Annex 4). Also, commerce and domestic service sectors cluster more informal workers (Figures 4.3). 7. Real wages in Uruguay have shown extreme volatility, following the economic cycle, and a general stagnation over the longer trend. Overall, earning levels in 2006 are similar to those observed during the early 1990s (Figure 4.4). The lowest hourly wages were Figure 4.5: Real Hourly Wages in the Figure 4.4: Real Hourly Wages Formal and Informal Sector Real Hourly Wages (1997 prices) 100 90 33,0 80 31,0 70 29,0 60 Percentage 27,0 50 UY Pesos 40 25,0 30 23,0 20 21,0 10 19,0 0 17,0 1989 1992 1995 1996 1997 1998 2000 2001 2002 2003 2004 2005 15,0 Formal Entrepreneurs Formal Large firms Formal Public sector 1992 1995 1996 1997 1998 2000 2001 2002 2003 2004 2005 2006 Formal SE professionals Informal salaried Small firms Informal SE Unskilled Source: SEDLAC based on microdata from the ENH , ENHA, and Source: SEDLAC based on microdata from the ENH National Statistics Institute Uruguay and, ENHA registered by informal self-employed (SE) and informal non-professional workers confirming the correlation between informality and low productivity in Uruguay. Formal salaried workers in the private sector have exhibited a reduction in real hourly wages, which could be linked to productivity stagnation or productivity losses. The following sections explore structural factors that could be affecting the performance of the labor market. 48 4.3. LABOR MARKET REGULATIONS AND RELATIONS 8. How do labor regulations and labor relations impact Uruguay’s labor market? According to the Uruguay Investment Climate Survey, labor regulations constitute the fifth most important obstacle to entrepreneurs (Chapter 2). Nearly one third of entrepreneurs indicated that labor regulations affected their hiring and firing decisions. Given the pattern of high unemployment and informality, this section analyzes how wage and non-wage regulations may be contributing to these rigidities. The views of workers‘ and entrepreneurs‘ on the status of labor relations are also presented. The findings indicate that both workers and entrepreneurs value constructive relations. Wage Regulations 9. Minimum wages are an important policy Table 4.1: Wage Council Coverage According to instrument to protect vulnerable workers and Occupational Category, 2006 establish a floor for the labor markets. Yes No Nonetheless, minimum wages can cause Private wage-earners unemployment or informality among those Formal 32.6% 67.3% workers that it seeks to protect (Maloney et al. Informal 3.8% 96.2% Public wage-earners 2001 and World Bank 2007). The minimum wage Central administration 27.1% 72.8% and sector based wage council negotiations barely Article 220 40.9% 59.0% impact unemployment but seem to contribute to Public firms 50.1% 49.8% informality. In recent decades, the minimum wage Regional government 40.3% 59.7% has been relatively low, and prior studies that Others 25.5% 74.5% analyzed the impact of the minimum wage found Non-dependent 1.6% 98.3% that that it plays a minor role in the determinants Note: The information is derived from the question: “Did wage increase according to what the Salary Council established? of unemployment, including unemployment Source: Instituto Nacional de Estadística (INE) de Uruguay among the youth (Pages, 2005, Maloney and (2006) Nunez, 2001, and Amarante, Figure 4.6: Minimum Wage Impact on Informality: 2003). The Tripartite Wage Marginal Effects (Probit Results) Councils, in operation since 0.60 0.52 1985, determine the central 0.50 minimum wage in Uruguay. In 0.40 addition, differentiated wage 0.30 scales are also determined per 0.20 0.22 sector, creating in practice 0.07 0.10 sector-based differentiated wage 0.10 0.03 0.040.04 0.03 0.03 regulations. Table 4.1 displays 0.00 -0.02 the percentage of workers that experienced wage increases in -0.04 -0.06 -0.10 -0.09 -0.11 -0.13 2006 due to agreements of the -0.13 -0.20 -0.17 -0.14 -0.30 wage councils. Using a Manufacture Construction Commerce Hotel/Rest Transport Dom. Services modified probit model, this ICA total less than 1/2 MW 1/2 - 2 MW expands on previous studies by Note: Numbers are marginal effects of interactions between MW distribution dummies (less than half MW, between half and twice MW, more than twice MW) and economic analyzing the impact of the sector dummies. sector-based wage council Source: Authors‘ estimates based on ENHA, 2006, Weighted sample. negotiations on informality (Box 49 4.1). Various models were tested indicating robust outcomes to different specifications. The most salient results are presented in this section, and a more complete set of results can be found in Table A.4.2 (Annex 4). Box 4.1: Methodology for Estimating the Impact of the Minimum Wage on Informality: Modified Probit Model. The conceptual model utilized for the informality probit model estimation uses a modified version of the Maloney et al. (2004) methodology (utilized for analyzing unemployment). This model captures the impact of the minimum wage incidence and other labor regulations on the probability of being informal. That is, in addition to the traditional demographic, employment characteristics, and geographic variables, the model includes earnings-to- minimum wage ratio distribution dummies, capturing individuals earning less than half the minimum wage (MW), between half and twice the MW, and above twice the MW. The economic sector dummies and the interaction effects between sector dummies and earnings-to-minimum wage ratio dummies are added to capture sector- specific minimum wage and other labor regulations effects on informality. The mathematical model is as follows: Inf , prob( z  1) i  �1 X i  � 2 Ei  � 3 R(wi , mw)  � 4 s R(wi , mw) * ES is  � 5 Ai  � i  The dependent variable ( Inf , prob( z  1) i ) is a dichotomous variable equal to 1 if the individual works in the informal sector and 0 if the worker participates in the formal sector.  X i includes demographic variables such as age (particularly interesting to focus on younger groups), sex, race, education (both formal and job training), and job tenure.  Ei contains employment characteristics such as dummies for economic sector (commerce, manufacturing, domestic services, and construction), and underemployment. The omitted category for economic sector is agriculture.  R(wi , mw) is a vector of dummies capturing various levels of the ratio individual wage/earnings to the minimum wage (MW) level (per hour). The null hypothesis is that workers earning the MW per hour or less are more likely to be informal. Three dummies are utilized to capture this distribution: below half the MW, between half and twice the MW, and above twice the MW.  R(wi , mw) * ES is captures the interaction effects between minimum wage levels and various economic sectors. In Uruguay, Tripartite Wage Councils set wage increases and other wage policies (including sector minimum wages) by sector. The sector dummy and the sector coefficient capture the impact of sector specific wage regulations.  Ai is a vector with geographic characteristics such as urban/rural residence, Montevideo vs. other areas.  �i is the error term.  The subscript i represents the individuals and s the economic sector. Source: Neumark et al. (2004). 10. The probit estimates show that wage increases (due to minimum wage increases or negotiations in the wage councils) may have different results by sector and for different groups of workers. Figure 4.6 shows the different marginal effects of increasing the minimum wage by one dollar over the probability of becoming informal by economic sectors. The probability of informality among workers in the construction sector, especially those earning between half a minimum wage and twice the minimum wage could increase. In manufacture, the average impact of an increase in the minimum wage is a reduction in informality. Only those workers earning a wage close to the minimum wage (between half and twice the minimum wage) 50 would experience an increase in the probability of becoming informal. In the case of hotels and restaurants an increase in the minimum wage would reduce the overall probability of being informal, including low wage workers (earning closer to the minimum wage and less than half a minimum wage). In the case of domestic services, an increase in the minimum wage would not increase informality in the most vulnerable groups but could increase overall levels of informality in the sector. 11. These findings lead to important policy implications, particularly regarding optimal minimum wages by economic sector. Increasing the minimum wage could have undesired effects on informality. This evidence points to the need for fine-tuning wage regulations such as the minimum wage and providing information on the potential impact of wage setting on informality to wage councils. It is important to adapt regulations to the needs of specific sectors such as agriculture, construction, commerce, and transport. The role of wage councils could be enhanced to include the goal of informality reduction within their sectors. Additional analyses could help further understand the differentiated impact per sector. 12. In addition, our analysis shows that acquiring skills--through formal schooling, job-related training, and work experience--reduces a worker’s probability of being informal. Moreover, skills‘ development through training is particularly important, and transferable skills, i.e., skills that can be utilized in several occupations, increase the likelihood of accessing a formal job.4 Table A.4.2 (Annex 4) shows overall results from the informality Probit estimation. Further explorations on this issue are presented in Section 4.4. Non-wage Labor Regulations 13. The investment climate survey indicates that severance payments followed by restrictions on working hours and temporary employment constitute the two most important obstacles to hiring and firing (Table 4.2). Some variations are observed across economic sectors and firm size. As many as twenty percent of surveyed entrepreneurs considered tight regulations on working hours and temporary workers to be the most important hiring and firing obstacle. In Uruguay, the work week is strictly regulated to 40 hours for most activities, with limited flexibility to up to 48 hours (8 hours per day and 6 days a week) for blue- collar and up to 44 hours for administrative workers. Additional hours must be paid at double the original rate. Part-time work and other weekly hour regimes are regulated exceptions, mainly applying to sectors with strong seasonal activity. 4 See Annex A.4.2 for a detailed description of the definition of transferable and non-transferable skill training. 51 Table 4.2: What Constitutes the Most Important Obstacle to Hiring and Firing? Restrictions on hours Retirement Costs of health Severance worked and temporary Others Total benefits insurance pay workers Firm Size Small and Medium 19.0% 9.6% 19.3% 33.1% 19.0% 100% Large 5.1% 9.4% 25.5% 25.5% 34.5% 100% Economic Sector Food 0.0% 10.5% 48.8% 22.2% 18.5% 100% Garments 9.5% 40.5% 35.3% 9.5% 5.2% 100% Textiles 0.0% 10.2% 0.0% 29.9% 59.9% 100% Chemicals and pharmaceuticals 21.4% 3.2% 11.0% 41.1% 23.3% 100% Other manufactures 4.9% 0.0% 9.1% 72.7% 13.3% 100% Retail 29.0% 17.7% 27.2% 17.2% 9.0% 100% Other services 29.0% 0.0% 11.8% 44.1% 15.1% 100% Construction and Transport 0.0% 0.0% 39.7% 0.0% 60.3% 100% Total 16.9% 9.6% 20.2% 31.9% 21.4% 100% Source: World Bank Investment Climate Survey (2006) 14. Internationally, Uruguay ranks high on working hours’ rigidity (Figure 4.7). Previous studies have also documented that rigid regulations on the work-week create considerable labor market inflexibilities (Heckman and Pages, 2000 and Packard 2005). Specifically, they have a negative impact on women and students Figure 4.7: Rigidity of Working Hours Index that are searching for part-time Selected Countries, 2006 employment. An extensive body of 90 literature on gender equality in the 80 workplace in Latin America confirms that flexible working 60 60 60 60 60 hours is a desired quality for female workers, especially during child 40 40 40 40 rearing years. Women would even choose to work in the informal 20 20 20 sector, giving up formal job benefits, in order to attain the flexible schedule that is necessary 0 to balance work and family needs.5 l nd ia ay ia a nd Th le a a nd zi a s ic ni si tv tin ak hi iu ra gu la Allowing greater flexibility in the la la R ay ua La C rit B ov Po en ai Ire ru ta al th au Sl rg os U M Li M A work week in Uruguay would be C Note: The rigidity of hours index has 5 components: (i) whether night work is welfare enhancing, especially for unrestricted; (ii) whether weekend work is unrestricted; (iii) whether the workweek women, and ease their participation can consist of 5.5 days; (iv) whether the workweek can extend to 50 hours or more (including overtime) for 2 months a year; and (v) whether paid annual vacation is in the formal labor market. 21 working days or fewer. Source: World Bank, Doing Business Report 2006 15. Non-wage labor costs (especially severance and retirement benefits) are one of the highest in the region and 5 For empirical evidence in Latin America and the Caribbean and policy options, see Cunningham (2005), World Bank (2007d), and Inter-American Development Bank (2005). For evidence in Uruguay see Espino and Amarante (2007). 52 have long been considered as factors contributing to the sluggish formal job creation (World Bank 2007d). About 32 percent of surveyed entrepreneurs indicated that severance payments have become the most important obstacle to hiring and firing. This obstacle is of particular concern to small and medium enterprises. Severance pay is very costly and impairs the labor market restructuring. From a social protection point of view, it also constitutes an inefficient form of unemployment insurance since it leaves younger workers less protected against unemployment. At the same time, the unemployment insurance program presents a number of gaps, including low coverage (only 6 percent of the unemployed benefit from it), lack of clarity on extension of benefit rules, use of insurance for lay-off cases, and poor administrative controls (Pages 2005)6. Uruguay would benefit from a more flexible severance payments regime combined with improvements to the unemployment insurance scheme. This policy option would reduce non-wage costs, while enhancing labor market efficiency and social protection instruments. Colombia‘s labor reform of 2002 illustrates the potential gains in labor market efficiency while simultaneously improving social protection (World Bank 2005). This reform created an unemployment insurance scheme, funded by resources from the parafiscales that had collected severance payments. This initial reform did not completely eliminate all severance payments or linked individual contributions to unemployment insurance, but the World Bank (2005) found that it had some gains in formal employment and improved protection of unemployed workers. 16. Retirement benefits constitute the third most important obstacle to hiring and firing. Retirement benefits are part of Uruguay‘s comprehensive social protection policies and are highly valued by society. Reforms to address these costs are likely to be politically challenging. Notwithstanding this, recent efforts to eliminate special exemption regimes for some economic sectors have been a positive step and are helping to create a more level-playing field and to diminish existing distortions. Labor Relations 17. Harmonious labor relations are essential to build worker and employer partnerships for economic growth. Strong labor relations are also the basis of partnerships for knowledge creation, innovation, and skills development. Recent events in Uruguay suggest that labor relations can be enhanced.7 To understand labor relations in Uruguay and gather first-hand information, special focus groups with unionized workers and entrepreneurs were conducted in April 2007. Focus groups discussions revealed encouraging findings regarding the high value that partnership promotion and tripartite dialogue have to unionized workers and entrepreneurs (Box 4.2). This shared appreciation together with the long tradition of tripartite collaboration in Uruguay sets a good basis for strengthening labor relations. However, they have different views on the role that takeovers have played and the role of subcontracting. Other areas where there is a shared understanding include the importance of skills development and the current disconnect between skills‘ availability and market requirements. Both workers and entrepreneurs are supportive of wage councils, but the latter expressed a need for wage councils to take into account firms‘ competitiveness in the market. Workers and entrepreneurs shared a common 6 Insurance is limited to 6 months but can be extended up to a year. There are, however, no clear rules defining benefit extension. 7 Most firm takeovers took place at the end of 2005 and beginning of 2006. 53 concern that the economic recovery is occurring without generating new employment. Boxes 4.3 and 4.4 gather the views of workers and entrepreneurs, respectively, as articulated in the focus groups. Box 4.2: Views of Workers and Entrepreneurs on Labor Relations: Focus Groups Summary Workers Firms The Government is fully including workers and Unfair and asymmetrical treatment from the government. Forced addressing their concerns. The Government‘s inclusion takeovers by workers (―ocupaciones‖) are becoming the norm has generated a positive impact on union‘s bargaining instead of a last recourse. power and membership. Firms have been growing without expanding Strong disincentives to expand employment due to potential employment. labor litigation if future lay offs are needed. Firms abuse subcontracting and extend it to workers Subcontracting is a valid modality to expand employment and who are clearly long term staff. increase competitiveness. Wage councils do a good job at determining and Wage councils work well but should take into account regulating some wages by sector. productivity and competitiveness. Strongly favored promoting firm/worker partnerships and tripartite dialogue. Source: Focus Groups conducted by the Foro Consultivo Económico Social, 2007. Box 4.3: Workers’ Views on Labor Relations and Other Key Labor Issues Participating workers considered that current tensions in labor relations are part of a transition underway due to the implementation of wage councils and a better balance in employee-employer relations. Unionization rates and workers‘ participation is recovering slightly as a result of the wage councils´ framework. Workers acknowledged that wage councils‘ agreements are hard to enforce on small firms which show resistance. Workers agree with current proposals to amend the subcontracting regime (Ley de Terciarizaciones) and think that it may increase formalization of firms. They expressed, however, concerns regarding potential loopholes of the new subcontracting regime and difficulties in regulating the new service provider firms. Overall, workers agreed with entrepreneurs on the need to increase the importance and quality of secondary and tertiary education as well as training programs, given labor market demands. In addition, they provided specific suggestions for improving/reforming labor regulations and institutions. Regarding training, participating workers proposed to limit training programs to those with completed primary education; set tripartite agreements with JUNAE Labor Exchange for companies to receive training benefits; provide incentives to companies hiring workers that participate in JUNAE‘s Labor Exchange program; and grant ISO certification to JUNAE (Junta Nacional del Empleo – National Employment Committee) and the Instituto Nacional del Empleo (National Employment Institute). On informality, workers recommended to write binding contracts for companies that fail to comply with wage council agreements and sanitary rules and forbid them to function as government suppliers, and to provide certificates to secure firms´ compliance with norms of public institutions such as Dirección General Impositiva or Tax Directorate and the Banco de Previsión Social or Social Security Institute. Finally, they suggested to expand the authority of the Dirección Nacional del Trabajo or National Work Directorate from mediating in labor conflicts to having punitive competencies; provide resources to workers to mobilize union recruitment across the country; grant tripartite inspection power to unions to investigate the execution of labor norms and other formal complains; and update the currently operational 1940 Collective Negotiation Law. Source: Focus Groups conducted by the Foro Consultivo Económico Social, 2007. 54 Box 4.4: Entrepreneurs’ Views on Labor Relations and other Key Labor Issues “ The government is supporting the union movement but not employment� Overall, participating entrepreneurs perceive that labor relations are asymmetric (with the government prioritizing workers’ demands over firms). Participants indicated that labor market regulations are too strict (preventing firms from creating new employment and over-utilizing the subcontracting regime) and that wage councils largely benefit large over SMEs. In addition, the mismatch between workers‘ skills and skills needed by firms is large. Entrepreneurs sensed considerable asymmetries in labor relations, with the government penalizing firms over workers rather than promoting a win-win framework. They argued that a major outcome of the current asymmetry in the law is the unions‘ use of forced take-overs (Ley de Ocupaciones) to control firms. Participants expressed that although this law was conceived as a last resource for unionized workers, it has become a regular tool, even creating disincentives to foreign investment. A participant presented this particular example: “(…) some entrepreneurs have not invested in Uruguay because of the take-over legislation (Ley de Ocupaciones). There was a famous case of a large Uruguayan printing business that tried to attract foreign investment unsuccessfully. Foreign investors stressed that workers “take-overs� and conflictive labor relations were a major problem (…) Ultimately, the Uruguayan firm had to close.� Regarding labor regulations, entrepreneurs noted that labor markets need to be more dynamic, with higher turnover, so that firms can hire according to their labor demands. They stated that, contrary to developed countries, in Uruguay, “if you hire a worker there is no way to fire him later.‖ As a consequence, they mentioned, ―firms are trying to grow without generating employment‖, and are investing in labor saving technology. They argued that “here the last option is to hire an additional worker‖. Also, their utilization of the subcontracting special regime (terciarizaciones) is considered an alternative tool to satisfy firms‘ seasonal labor demands. Unfortunately, they noted, this regime has been over-utilized, even by the government, and it is likely to be eliminated. Entrepreneurs proposed to “(…) simplify regulations but improve enforcement, especially for firms in the informal sector.� On the wage councils, participants mentioned that they were a good institution for wage regulation, but nowadays these councils are not taking into account important competitiveness factors for different economic sectors. Some participants believe that the wage councils ―benefit large firms vis-a-vis small and medium enterprises�, since the former can sustain higher wages that affect the competitiveness of smaller firms and their capacity to generate additional formal jobs. Regarding skills, entrepreneurs consider that there is a mismatch between skills created by the school system and public training programs, and their needs. Quoting directly from one participant: “(...) in Uruguay businesses and the education system are, in fact, divorced, except for the case of some public programs (…)� In addition to the quality and relevance of tertiary education, participants said that access to this educational level needs to be expanded. Also, entrepreneurs highlighted the importance of flexibility and transferability of skills: “(…) education programs cannot foresee future labor demand needs …that is why training should focus on developing transferable skills.� Source: Focus Groups conducted by the Foro Consultivo Económico Social, 2007 18. Recent tensions have marked a departure from Uruguay‘s long tradition of tripartite collaboration, but the value that workers and employers grant to good labor relations provides a basis for returning to a healthy cooperation. In particular, the Uruguayan Labor Working Group (or Subgrupo de Trabajo 10), established in the mid-1990s to support the MERCOSUR negotiation 55 process, offers a useful example of constructive and valued cooperation among workers, employers, and the government. It focuses on employment, job quality, migration, and skills development issues that are directly related to the MERCOSUR implementation. As part of its functions, the group provides labor market information on member countries and has commissioned a comparative study of labor legislation in the MERCOSUR member countries.8 The experience of Ireland is also relevant to Uruguay, providing a good example of how to attain higher labor market flexibility while maintaining its traditional and solid institutional framework (Box 4.5). Box 4.5: Ireland’s Successful Experience of Good Labor Relations Over the past decade, Ireland has experienced a rapid growth and an overall social recovery. During 1987, Ireland was undergoing a major economic crisis with high government debt, non-existent investment, and high unemployment (despite the large emigration flow). At that time, the Irish Programme for National Recovery (PNR) was negotiated, as the first of five partnership agreements. By 2000, social partnership negotiations shifted emphasis with the Programme for Prosperity and Fairness (PPF). Labor and skill deficits replaced the previous high unemployment problem (O‘Connell, 1999; Baccaro and Simoni, 2004). In addition, large foreign investment flows established a more productive and dynamic manufacturing sector. Although some debate exists on the actual effect of partnership negotiations on Ireland‘s economic recovery, studies indicate that social partnership negotiations have positive effects on the wage formation process (Baccaro and Simoni, 2004). Several explanations were presented regarding this relationship. Baccaro and Simoni argue that social partnership negotiations altered the process of wage setting, potentially increasing the competitiveness of the more dynamic sectors, and hence, the whole Irish economy. Although public sector wage increases created the bases for private sector wages, they primarily affected the traditional and less dynamic manufacturing sector, commonly established based on productivity gains. Wages in the modern, more dynamic sector remain stable, despite the major increases in productivity. Also, Teague (1995) attributes the positive outcomes observed in Ireland to its social partnership system, which incorporates civil associations, in addition to the traditional union, employers, and government groups. Source: Driffill (2006), OECD (2004), Baccaro et al. (2004), O‘Connell,(1999), and Teague (1995). 4.4. DEVELOPING THE SKILLS FOR THE FUTURE 19. Skills development in Uruguay is vital to a dynamic labor market and productivity together with flexible labor regulations and strong labor regulations. Skills are a key complement to technological advance, and technological developments are increasingly biased towards skilled workers (De Ferranti et al. 2002). The Uruguay IC survey and subsequent focus groups interviews conducted with workers and entrepreneurs confirm the importance of skill development. Estimates of productivity growth using the IC survey data indicate that increasing investments in training relative to the 75th percentile of a group of comparator countries (Figure 2.23) could increase firm productivity by about 16 percent. About 45 percent of surveyed entrepreneurs cited that an inadequately workforce is a moderate obstacle to optimal operations and growth. The focus groups also highlighted the importance that workers and entrepreneurs attribute to skills formation and concerns, especially by entrepreneurs, that there is a certain disconnect between requirements and availability (Boxes 4.3-4.4). 20. Skills can be acquired through the formal schooling system and through subsequent training in the workplace. Uruguay has made steady progress but continues to 8See www.cinterfor.org.uy. 56 face challenges on both fronts. The country has a long history of good educational attainment and compares favorably to Latin America, but it is not keeping up 21. with changes in the OECD and East Asian countries. The stock of human capital expanded continuously from 1960 to 2000. Uruguay was a pioneer in Latin America mandating 9 years of primary Table 4.3 Years of Schooling education in the 1970s (World Bank 2005a). The pace of Uruguay and Selected Countries expansion, however, has been slower than rapidly growing East Asian economies and most OECD countries. While Uruguay 1960 2000 raised average years of schooling from 5.0 to 7.3 from 1960 to Argentina 4.99 8.49 2000, Finland, starting from a similar level in 1960, was able to Brazil 2.83 4.56 9 double school attainment by 2000 (Table 4.3) . Korea, which Chile 4.99 7.89 lagged Uruguay in 1960, has more than tripled schooling Finland 5.4 10.1 Korea 3.2 10.5 attainment and surpassed Uruguay during the same period. The Mexico 2.41 6.73 growing gap could jeopardize Uruguay‘s international Uruguay 5.03 7.25 competitiveness. The results of the International Student Spain 3.64 7.25 Evaluation Program (2003) on basic competencies of reading, Source: De Ferranti et al. (2003) math and science also showed that Uruguay performed better than most of its neighbors but was below the OECD (Table 4.4). 22. Public spending on education (3.6 percent of GDP) is smaller than the average of Latin America (4.5 percent) and of high income countries (5.3 percent). Issues of access and equity remain. Only 30 percent of students in the lowest income group complete secondary education compared to 80 percent in the highest income group (World Bank 2005a). Access is even more skewed for tertiary education. World Bank (2005b) conducted an assessment of the educational system and explored strategic policy options for addressing current gaps. These options considered inter alia improvements in the Table 4.4 Student Performance. PISA Ratings. Uruguay and Selected Countries overall administration of public education; amendments to the curricula Reading Mathematical Scientific Average to make it more relevant to the literacy literacy literacy performance knowledge economy; enhancing teacher Finland 544 543 548 545 incentives by increasing the autonomy South Korea 525 557 550 543 and responsibility of school directors, OECD average 500 494 500 498 improving teacher evaluations, and Spain 493 476 491 487 linking remuneration to performance; Uruguay 422 434 438 431 and increasing the incentives of schools. Mexico 422 387 422 410 Chile 410 384 415 403 Reforms were also suggested to address Argentina 418 388 396 401 equity gaps and cost-efficiency issues, Brazil 396 334 375 368 especially in tertiary education.10 Source: World Bank (2005b) and OECD (2003). Economic Surveys Chile. Volume 2003/17. Paris 23. Training in the workplace is a critical complement to skill development through the formal schooling system. Uruguayan firms, however, under- invest in training. Complementing prior analysis on the formal schooling system, this section explores the benefits of investing in training to workers and firms; the profile of firms investing 9 Average school attainment refers to the number of years of education in the population over 25. 10 See World Bank (2005b) for a further discussion of current challenges and strategic policy options. 57 in training; and public programs to foster training. Lastly, it offers policy recommendations to enhance training related policies. The analysis uses data from both the IC survey and the expanded household survey with a training module. Training Enhances Firms’ Productivity and Workers’ Earnings 24. Training enhances the productivity of Uruguayan firms and workers. Chapter 2 simulations found that productivity of Table 4.5 : Impact of Training on Labor Productivity at the Firm Uruguayan firms could be increased Level Using Propensity Score Matching Method through enhanced training (Figure 2.23). (Uruguayan pesos per year) To test the robustness of these findings, Number of firms Average Standard an analysis on the impact of training at the Treated Control treatment error t-statistic firm level through a non-parametric effect Value added statistical method was conducted. per worker 126 136 16377*** 3676 4.5 Propensity score matching compares labor Last fiscal year productivity of firms in the treatment sales per 126 179 49999*** 15538 3.2 group (firms providing training) and worker control group (firms not providing Average wage 126 148 6002*** 1177 5.1 training) and finds a positive effect on * significant at 10% ** significant at 5% *** significant at 1% productivity among treatment firms (Table Source: Authors’ estimates using Investment Climate Survey Data, 2006. Weighted sample 4.5). Matched firms in both groups are statistically identical in a set of characteristics (training determinants). As a consequence, differences in observed productivity between these two groups can be attributed to actual training provision. Only firms in the manufacturing sector are included in these estimates. Box 4.6 discussed the methodology in detail. 11 25. Uruguayan workers also benefit from training. Investments in training have a significant positive impact on workers‘ employability and earnings. We evaluate the effects of training— total training as well as transferable and non-transferable training--on employability and earnings using the expanded household survey of (of 2006). The survey contains a skills and training module that inquires on training during the last 5 years. Box A.4.2 (Annex 4) summarizes the classifications of training courses on transferable and non-transferable skills and presents the main characteristics of workers benefiting from training. Workers participating in transferable skills courses tend to be more educated and have longer job tenure. We use Probit models to analyze training effects on workers‘ employability and a Mincer-type model to analyze the impact on earnings. Both estimations show robust results to different model specifications. Detailed model specifications and regression results are presented in Tables A.4.5 and A.4.6 (Annex 4). The main findings are summarized below. 11 Using a non-parametric method is a good approach to complement estimates of Chapter 2 and explore the robustness of the impact of training provision. Empirical evidence suggests that avoiding functional form restrictions can significantly reduce bias (Dehejia and Whaba, 1998; Smith and Todd, 2000; Smith, 2000). However, non-parametric approaches have their own set of econometric complications, including the need for quite large sample sizes, not offering advances over traditional regression analysis in the treatment of unobserved heterogeneity, and not explicitly dealing with issues of simultaneity. 58 Box 4.6: Impact of Training Provision on Labor Productivity at the Firm Level: Propensity Score Matching Methodology: The compares labor productivity of firms in a ―treatment‖ group (firms that provided training) to a ―control‖ (firms that did not provide training) group. Matched firms in both groups are statistically identical in a set of characteristics (training determinants) and differences in observed productivity between these two groups can be attributed to actual training provision. The methodological steps are presented below. Step 1: Estimate a Probit equation of the determinants of training investment. Table A.4.2 (Annex IV) contains the results of several specifications of a Probit model of the determinants of training investment in Uruguay using ICA survey data (2006). All these specifications were tried but only a few allowed for propensity score matching (Step 2). For matching purposes, the most relevant determinants of training provision are: number of years in operation, manager‘s experience, firm size, foreign ownership, access to credit, technology use (proxied through web use), percentage of full time workers, percentage of female workers, percentage of unionized workers, and controls for employee‘s education level. Step 2: Use propensity score matching to pair firms that provided training to those who—although did not provide training—have similar estimated a-priori probability of investing in training. The methodology requires that the range of a priori probabilities be similar for treatment and control groups, i.e., that there is an area of common support. Step 3: Compare the means of matched ―treatment‖ and ―control‖ groups for the variables of interest. In this context, the variables of interest are proxies of labor productivity. Source: Smith (2000) Figure 4.10: Impact of Training Courses on Figure 4.11: Impact of Training Courses on Employability Earnings Dependent Variable equals 1 if currently employed and Dependent Variable: Logarithm of wage/earnings 0 if currently unemployed 0.30 0.05 0.04 0.27 0.04 0.25 0.23 0.04 0.03 0.20 Marginal Effect 0.03 Marginal Effect 0.17 0.03 0.15 0.02 0.02 0.01 0.10 0.01 0.05 0.01 0.00 0.00 Any tra ining No n-tra ns fe ra ble Tra ns fe ra ble Any t raining course Non-t ransferable T ransferable t raining c o urs e - M o de l I tra ining c o urs e - tra ining c o urs e - - Model I t raining course- course- Model II M o de l II M o de l II Model II Notes: Marginal effects are significant at 1% Notes: Marginal effects are significant at 1% Models I and II specifications are presented in Table A.4.5 (Annex IV) Models I and II specifications are presented in Table A.4.7 Source: Authors‘ estimates based on ENHA, 2006, Weighted (Annex IV) sample. Source: Authors‘ estimates based on ENHA, 2006, Weighted sample 26. Training improves earnings’ conditions and the impact is greater for transferable skills (Figure 4.11). Those receiving transferable skills increased their earnings by 27 percent compared to a 17 percent increment, among those receiving non-transferable training. It is 59 possible that transferable skills grant them greater mobility, and consequently they are more capable of finding better jobs or negotiating promotions. The impact on employability is more modest (Figure 4.10) and is greater for those benefiting from non-transferable skills. It is possible that individuals participating in specific (non-transferable) training courses are doing so because they already have a concrete position to enter, and thus they exhibit greater employability. Firms Under-invest in Training 27. Despite the potential large Figure 4.12: Percentage of Firms Providing returns, few Uruguay firms provide Training to Full Time Employees formal training compared to other 80% 74% 76% upper middle income countries (Figure 70% 67% 72% 4.12). Moreover, these investments are 60% mainly concentrated among large firms. 52% Only 8 percent of small firms‘ employees 50% 46% report having received firm training, 40% compared to 28 percent and 36 percent 30% 25% 25% among medium and large firms‘ 20% employees (Household survey 10% estimations). That firms underinvest in 0% training is consistent with the literature. Argentina Brazil Chile M exico Costa Rica Uruguay Ireland (2005) Thailand Employee turnover may prevent firms Source: World Bank, Investment Climate Survey, 2006 from retaining productivity gains derived from training investments. Lack of information and finance constraints, which tend to be more pronounced for small firms, may also lead to underinvestments. Acemoglou (1998), Barron et al. (1999), and Holtmann et. al Idson (1999) present a literature review on the matter. The IC survey, however, indicates that the problem is more acute in Uruguay. 28. The determinants of investments in training by firms were explored using a probability model with IC survey data (Table 4.6). Probit specifications are tested including workforce education characteristics, firms‘ investment on technology, access to foreign capital and credit markets, firm size, economic sector, and sales, and manager‘s experience. Workforce composition variables such as share of unionized and female workers are also included. 29. Firms with a higher share of less than complete high school tend to invest less on training programs. This supports the idea that firms require workers that can be trained in the use of new technologies. These results are consistent with prior studies that also pointed to the relevance of secondary education for firms to invest in training.12 The results also show that firms using new technologies in the production process are more likely to invest in training, confirming the complementarity between new technologies and training. Access to credit markets and foreign capital show positive effects on firms‘ training provision. Size is also correlated to training provision with small and firms underinvesting relative to large firms, which suggests that public policies may need to consider targeting the former. Public policies in supporting of training are discussed below. 12 World Bank (2003) –Learning and Skills development report 60 30. According to the results in Table Table 4.6: Determinants of Training Provision by Firms 4.6, underinvestment in training is particularly severe in firms with a large Variables Model percent of female workers. This result Firm manager's experience -0.011 [11.33]*** may be due to occupational segregation by Workforce education gender. Differentials in occupational Primary complete -0.249 [2.80]*** composition become occupational High school incomplete -0.194 [2.02]** segregation by gender when individuals High school complete -0.082 [0.91] who are identical in all productivity-related Superior incomplete 0.022 [0.11] characteristics choose a job or can only New technology in production process 0.133 [6.26]*** find a job in certain occupations, based Firm years in business 0.003 [6.79]*** solely on gender roles. High levels of Foreign capital 0.293 [5.14]*** occupational segregation by sex may have Percentage of unionized workforce 0.002 [4.78]*** important effects on male and female Percentage of female in the labor force -0.339 [7.05]*** Large firm 0.15 [5.12]*** welfare and labor market efficiency. In Garments 0.327 [6.75]*** particular, there is evidence that it Textiles 0.267 [7.19]*** contributes to under-investment in female Chemicals & Pharmaceuticals -0.064 [2.24]** education and training in technical Other Manufactures 0.02 [0.52] occupations and/or an under-investment Access to credit 0.073 [3.26]*** in male education and training in service Log last fiscal year sales per worker 0.004 [0.47] occupations (World Bank, 2007d and IDB, Observations 2299 2004). Pseudo R2 0.2217 Log Likelihood -1030 Training and Skill Development Naïve 0.263 Policies and Programs Predicted P. 0.213 Note: Absolute value of z statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 31. There is a strong interest by the Source: Authors’ estimates based on ENHA, 2006. Weighted sample. government to foster workers’ training, but the supporting institutional framework is fragmented and there are too many dispersed programs. There is a wide number of public programs and institutions supporting training in Uruguay, but there is scope for improving coordination of these policies and decreasing existing overlaps. The three main institutions are DINAE, JUNAE, and the National Labor Skills System or Sistema Nacional de Competencias Laborales (SNC). Their policies and programs support the general labor force as well as more vulnerable groups such as women, the unemployed, the youth, people with disabilities, and rural workers. The DINAE, created in 1992 as a Directorate within the Ministry of Labor and Social Security, is the main public unit responsible for formulating policies on worker training and skill development and also designs and oversees training programs. Broadly speaking, DINAE‘s programs seek to enhance the quality of training and improve the competitiveness of individuals at greater unemployment risk by increasing their qualifications and offering re-training. It also sponsors a system of job intermediation. Box 4.7 summarizes its main programs. 32. The JUNAE is a tripartite body presided by DINAE’s Director, a unions’ representative, and an employers’ representative. It plays an advisory role to the DINAE but also has specific executive functions such as designing and overseeing the implementation of 61 training and retraining programs13; managing the retraining fund (Fondo de Reconversión Laboral)14; and mediating between DINAE and local governments on training policies and programs. JUNAE has been meeting these original objectives and making progress in the design of programs for unemployed workers.15 Some programs, however, show duplication with DINAE‘s initiatives and would benefit from more systematic impact evaluations. Box 4.8 summarizes a pilot program targeted to micro and small enterprises. A rigorous evaluation of this pilot to determine its impact would be crucial before launching it at a larger scale. Box 4.7: Main Skill Development Programs by DINAE Centralized Programs  Uruguay Activo – System of Job Intermediation. The program has a web-based data processing system with information from job seekers and employers. The system provides a search engine and matches workers profile to posted job positions. Firms can also register to carry out their own searches for employees and pre-select candidates.  PROCAL (Programa de Capacitación Laboral para Trabajadores en Seguro de Paro). It focuses on laid off workers with unemployment insurance  Training Programs: -Training Program for Employed Workers (Programa de Formación para Trabajadores en Actividad). It supports on the job training so that workers can keep their jobs and is co-financed by the re-training labor fund (by JUNAE see below). It targets firms with 20 or more employees. Eligible firms must send a proposal and an agreement signed by the firm and union or organization representing workers in the firm. -Program of Productive Training (PROCAPRO): It supports training to groups of workers, civil society groups, and cooperatives with a productive project. The project‘s feasibility is evaluated by DINAE before enrolling the group in this program. -Special Training Programs: The program sponsors reliability and safety at the workplace. Decentralized Programs  Training Program for Rural Workers: This program supports rural firms and self-employed rural workers. It promotes the use of appropriate technologies that favor the productive re-conversion of small producers.  Training Program for Individuals with Disability (PROCLADIS)  Emprende Uruguay: The program promotes job creation through entrepreneurship. It is financed by the re- training labor fund and managed by the Catholic University of Uruguay (UCUDAL).  PROIMUJER (Promoción de la Igualdad de Oportunidades para la Mujer en el Empleo y la Formación Profesional). It seeks to improve the employability and labor insertion of women with limited resources. The target group is women from urban or rural areas who have lost their jobs or are looking for their first jobs and are facing obstacles to enter the labor market.  PROJOVEN: It supports training and labor insertion of youth (18 to 24 years) from low income households. Eligibility criteria include: having difficulties to find a job, incomplete high school education, and limited experience. The program favors young people that are heads of households or have children. Note: Decentralized programs are implemented in coordination with other entities. Source: DINAE. See http://www.mtss.gub.uy/dinae.htm 33. General training and lifelong learning policies are designed by the National Skills System or Sistema Nacional de Competencias Laborales (SNC). The SNC has a 13 Programs can be implemented by public or private providers. 14 The fund is financed with contributions by the government, employers and employees. 15 Source: Centro Interamericano de Investigación y Documentación sobre Formación Profesional. See www.cinterfor.org.uy. 62 consultative body with representatives from DINAE, unions, employers, chambers, the formal schooling and training systems, and experts funded by International Labor Organization (ILO). Its consultative body has undertaken numerous reports on skill development systems in other countries and their applicability to Uruguay. It also coordinates programs by unions, universities, and firms, such as internship programs. Chambers of commerce and industry also coordinate specific skill development programs with the SNC. Box 4.8: JUNAE: Small and Medium Enterprises Pilot Program (FOMYPES) FOMYPES is a pilot program of the JUNAE offering financial assistance for training to micro and small firms (less than 20 workers). The program finances up to 80 percent of the cost of the training. Eligible micro and small firms must be located in the departments of Canelones, Colonia, Maldonado, Montevideo, Paysandú, San José y Tacuarembó. Firms have to be formal (registered and active contributors to the tax and social security systems). The program seeks to improve small firms‘ management and productivity efficiency; support employment creation; encourage firms‘ and workers‘ formalization; and foster synergies between firms and their workers Source: JUNAE, available at http://www.junae.gub.uy/sin_flash/principal.htm Policy Priorities on Training 34. A skills’ development policy framework needs to address market failures in training provision, which will include both programs to reduce unemployment and to enhance the productivity of employed workers and firms. Based on the above findings, three policy lessons emerge in meeting those objectives. First, there is a degree of duplicity among public or quasi-public bodies involved in the design and promotion of training programs, and there are too many dispersed training initiatives. Strong coordination together with greater clarity of functions and allocation of responsibilities will lead to a more integrated and efficient use of public resources. While some evaluations of training programs have been undertaken in the past, they have not been conducted in a systematic way. Regular monitoring and independent evaluations would shed light on the impact and effectiveness of these initiatives, providing constructive feedback on policy design. 35. Second, the above findings indicate that when supporting vulnerable workers greater emphasis is needed on training programs fostering transferable skills. Some program redesign with the participation of entrepreneurs and unions would be desirable. Based on the ENHA survey (2006), public programs finance larger proportion of non-transferable courses (36 percent) versus transferable courses (28 percent), whereas firms place greater emphasis on transferable training (31 percent) versus non-transferable (56 percent). The earlier findings also indicate that training on transferable skills had a higher impact on workers‘ earnings relative to training on non-transferable skills. 36. Third, providing incentives to increase firm training is important to enhance firm productivity and increase workers’ earnings. Moreover, SMEs need to be the primary target of these incentives and proactive measures are needed to reach them. Programs will also be more effective if they are offered as part of a more comprehensive package of technical assistance support including incentives to embrace new technologies. Mexico‘s Program of Comprehensive Quality and Modernization or Programa de Calidad Integral y Modernización (CIMO) is a good example of collaboration between the government and SMEs associations to actively seek out SMEs and provide a comprehensive support package (Box 4.9). See also Tan (2002) for 63 an assessment of the Malaysia training fund which was very successful in increasing training by its participating firms and their productivity. Recently, a few training initiatives have been implemented in Uruguay with the collaboration of unions and the Association of Small and Medium Enterprises. More systematic efforts in this direction would be desirable. Box 4.9: Mexico: Program of Comprehensive Quality and Modernization (CIMO) The CIMO Program has offered subsidized training and technical assistance to SMEs since 1987. The Ministry of Labor and Social Protection is responsible for implementing the program through a decentralized network of promotional training units (local associations and chambers of commerce). The CIMO‘s main objectives are to improve workers‘ productivity and quality; to promote quality control systems, human resource management, and labor relations in enterprises; to encourage industrial clusters and inter-firm linkages; and to match the training supply with the enterprises‘ skill needs within each region. Several innovative characteristics have been incorporated in this program. First, in addition to training, the CIMO program provides technical assistance to participating firms. Training alone would not have a significant effect on productivity. Second, the CIMO is proactive, seeking and engaging SMEs, through its decentralized network system. Third, financial subsidies, rather than direct provision of training, are granted to SMEs. Technical support is provided for evaluating the firm‘s actual needs (production, skills, or other constraints). Training and technical assistance is offered on the basis of this initial assessment, prioritizing the use of services from local providers. Source: Tan et al (2007). Evaluating Mexico‘s Small and Medium Enterprise Programs. The World Bank. 4.5. CONCLUSION 37. The earlier findings provide some interesting policy implications to enhance Uruguay’s investment climate and encourage greater employment generation in the formal sector. In their own words, both workers and entrepreneurs expressed a value for strong labor relations and favored promoting partnerships and tripartite dialogue to set the basis for growth and employment generation. They both expressed concerns that firms are growing without generating sufficient employment and there are important gaps in workers‘ skills. Indeed, the econometric analysis indicates that training has a payoff to both firms and workers. Investments in transferable skills seem to have a higher impact on workers‘ earnings. The focus groups, however, suggested some areas where there was not yet a shared understanding between entrepreneurs and employees pointing to the need for further dialogue, for example on labor regulations. Both entrepreneurs and workers indicated a role for wage councils as a wage setting framework at the sectoral level, but there was not yet a full understanding on the framework. 38. Future tripartite collaboration can draw on Uruguay’s history of dialogue among the government, employees, and employers, for example, on Mercosur negotiations. International experiences also provide useful lessons. The case of Ireland, a small open economy, is relevant to Uruguay because it achieved labor market flexibility while maintaining its traditional and solid institutional framework. In the early 1970s, a national consensus that involved both unions and employers and cut across political lines was reached becoming an important pillar of its new competitiveness strategy. Ireland achieved greater flexibility and productivity emphasizing shared benefits for both employers and employees. 39. Strong labor relations are vital to achieve more flexible labor regulations, while maintaining adequate social protection. The econometric analysis suggests that, in some 64 sectors, wage councils regulations have increased the likelihood of informality. Providing information to wage councils on the impact of regulations on informality and adapting regulations to the needs of specific sectors is vital. 40. Introducing a more flexible regime for working hours would be welfare enhancing, especially for women and the youth, two groups that are prone to informality and unemployment. Uruguay‘s legislation imposes tight restrictions on working hours by international standards. Evidence from other countries in Latin America corroborates the importance of flexible working hours to ease the participation of women in the labor force. According to the investment climate survey, entrepreneurs highlighted severance payments as the most important obstacle to hiring and firing. One option for consideration would be to enhance the unemployment insurance scheme, which is a superior social protection instrument, and make the severance payment scheme more flexible to reduce its costs on employment generation. This matter merits further investigation. 41. Despite a high payoff, the quantitative analysis indicates that Uruguayan firms do not invest sufficiently in training. The deficiency is most pronounced for SMEs. Public policies on training should tackle market failures and would benefit from addressing the retraining needs of the unemployed as well as from enhancing the productivity of employed workers. Distinct but complementary policies are necessary. Based on the analysis, three policy lessons emerge in meeting those objectives. First, it would be desirable to improve coordination across public and quasi-public bodies involved in the sector and provide greater clarity on responsibilities to avoid possible duplication. The design of training policies and programs could also benefit from regular monitoring and independent evaluations. Second, training programs fostering transferable skills deserve more emphasis. Third, public incentives to foster training by firms need to be primarily targeted at SMEs, use proactive policies to reach them, and be coordinated with other initiatives to enhance their productivity. 65 CHAPTER 5. CONCLUSION 1. Uruguay stands out in selected investment climate areas, in particular public integrity. However, its overall performance in international perspective is not as favorable. Improving the investment climate will be crucial to sustain the rapid growth observed in recent years. Access to finance by enterprises is modest. A series of important reforms to develop the enabling regulatory environment for finance are underway. In parallel, there is a transitory role for the CND to engage in pro-active market development policies within a sounder corporate governance framework. 2. The analysis indicates that some labor regulations are conducive to informality, even though they were intended to protect vulnerable workers. Labor regulations can be made more flexible and provide adequate social protection. Investing in training would enhance firms‘ and workers‘ productivity. A wide range of policies and programs to support training are already in place. These could benefit from streamlining, greater focus on transferable skills, and further promotion of more active investment by firms, especially SMEs. Our analysis also points to the complementarity between training and technology. The investment of Uruguayan firms on innovation and new technologies is generally modest. Innovation has already become a policy priority, and the challenge is to maintain continuity due to lags in innovation policy returns. 3. Red tape is prevalent and the quality of public institutions could be increased in Uruguay. A number of institutional reforms are planned or under implementation. The reforms of the tax and customs administrations are generally progressing well. For the most part, other regulatory and institutional reforms that would ease doing business or would seek to promote investment are advancing at a slower pace and could receive a higher priority. A culture of high quality public service with streamlined regulations could be promoted more widely. Infrastructure is not an important investment climate obstacle at this stage, but it would be desirable to prepare medium-term development strategies for key sectors in advance of growing demand. The country is already facing challenges in meeting demand in the energy sector and is lagging in new telecommunications services such as broadband penetration. Fostering a more competitive environment and strengthening regulatory bodies would help to attract higher investment to the infrastructure sector and increase efficiency. In areas where Uruguay has encouraged competition, services have rapidly expanded and efficiency has risen. 4. Addressing investment climate challenges requires social partnerships involving the government, employers, trade unions, and civil associations. The progress attained on the macroeconomic front in recent years reflects an underlying consensus that sound macroeconomic management is equally vital to growth and poverty reduction. A similar consensus across social partners would be desirable to tackle labor relations and other investment climate challenges and to maintain the rapid growth of recent years. The findings of the focus groups with workers and entrepreneurs are promising, indicating that they both value good labor relations and share a common concern that the economic recovery is not generating sufficient employment in the formal sector. 5. Although social institutions are unique to each country, international experiences can be illustrative. Ireland experienced a major economic crisis in 1987, which led to the Irish Program for National Recovery. Since then, Ireland has seen rapid growth and improved standards of living. Social partnerships agreements have also been important to the development of national competitiveness agendas in northern European countries such as Finland. Until the early 1990s, 66 Finland also had several state monopolies and was heavily regulated, but a deep economic crisis in the early 1990s motivated social partnerships in support of a broader national competitiveness strategy. Investing in workers‘ skills is the trade unions‘ foremost priority as the best instrument to foster employment and rising earnings. Like Finland and Ireland, Uruguay is a small country with social capital, stable democratic institutions, and an educated population, which provide the basis for similar social partnerships. 67 ANNEX I: SAMPLING METHODOLOGY The Uruguay Investment Climate Survey sample included 621 firms in Montevideo and the department of Canelones. With the exception of a small change in firm size definitions, the survey followed the sampling methodology for the global roll-out of Investment Climate Surveys. The sample consists of a range of small, medium, and large firms in the food, clothing, chemical, and retail sectors, as well as a residual category for other types of firms (Table A.1.1). The survey is based on the standard Investment Climate Survey questionnaire to ensure international comparability. The data has been collected by face-to-face interviews of company management, performed by trained enumerators during the period March to August 2006. Data collection was overseen by the Enterprise Analysis Unit of the World Bank‘s Investment Climate Department. Table A.1.1: Investment Climate Survey Sample Breakdown Food Clothing Chemical Retail Residual Total Montevideo Small (1-19) 34 59 53 42 30 218 Medium (19<100) 33 41 57 25 37 193 Large (100=<) 16 9 7 14 36 82 Subtotal 83 109 117 81 103 493 Canelones Small (1-19) 20 6 3 27 20 76 Medium (19<100) 12 2 2 14 9 39 Large (100=<) 3 0 0 2 4 9 Subtotal 35 8 5 43 33 124 Total 118 117 122 124 136 617 Note: The total sample size was 621. However, 4 firms are not categorized because they did not answer the industry or city questions; hence, the above total is only 617 firms. Sampling structure. The sample was selected using stratified random sampling based on three levels of stratification: industry, establishment size, and region. Stratified random sampling was preferred over simple random sampling for several reasons:  To obtain unbiased estimates for different subdivisions of the population with some known level of precision.  To obtain unbiased estimates for the whole population. The whole population, or universe of the study, is the non-agricultural economy. It comprises: all manufacturing sectors (group D), construction (group F), services (groups G and H), and transport, storage, and communications (group I). Groups are defined following ISIC revision 3.1. Note that this definition excludes the following sectors: financial intermediation (group J), real estate and renting activities (group K, excluding sub-sector 72, IT, which was added to the population under study), and all public or utilities-sectors.  To make sure that the final total sample includes establishments from all different sectors and that it is not concentrated in one or two industries.  To exploit the benefits of stratified sampling where population estimates, in most cases, will be more precise than using a simple random sampling method (i.e., lower standard errors, other things being equal.) 68 Sampling implementation and questionnaires. Given the stratified design, a sample frame containing a complete and updated list of establishments was required. For Uruguay, the National Statistics Institute‘s (INE) Permanent Register of Economic Activities for 2004 was used. A secondary source for company statistics was the sample frame of Economic Entities from the 1997 National Economic Census. To correct for the presence of ineligible units—which comprised about 10 percent of the sample frame—adjustments were made to the computation of weights (see below). The survey was implemented following a two-stage procedure. In the first stage, a screener questionnaire was applied over the phone to determine eligibility and to make appointments; in the second stage, a face-to-face interview took place with the Manager/Owner/Director of each establishment. Three different versions of the questionnaire were used. The Core Questionnaire included all common questions asked to all establishments. The Manufacturing Questionnaire added some specific questions relevant to the sector, and the Services Questionnaire added questions relevant to either retail or IT. Weights. Since the sampling design was stratified and employed differential sampling, individual observations were weighted to make inferences about the population. Under stratified random sampling, unweighted estimates are biased unless sample sizes are proportional to the size of each stratum. With stratification, the probability of selection of each unit is, in general, not the same. Consequently, individual observations must be weighted by the inverse of their probability of selection.1 Special care was given to the correct computation of the weights. Totals within each region/industry/size stratum were adjusted to account for the presence of ineligible units (non- existing units, public establishments, and non-business units). The information required for the adjustment was collected in the first stage of the implementation (the screening process). Using this information, each stratum cell of the universe was scaled down by the observed proportion of ineligible units within the cell. Once an accurate estimate of the universe cell (projections) was available, weights were computed using the number of completed interviews. Non-response. Survey non-response must be differentiated from item non-response. The former refers to refusals to participate in the survey altogether whereas the latter refers to the refusals to answer some specific questions. Survey non-response was addressed by maximizing efforts to contact establishments that were first selected in the sample and by trying to keep a tight control over the process of substitutions. Item non-response was addressed by two strategies: a) for sensitive questions that could generate negative reactions from the respondent, such as corruption or tax evasion, enumerators were instructed to collect the refusal to respond as a different option from ―do not know‖; and b) for information that establishments may consider too private to share, such as financial information, sample sizes were inflated by 25 percent to account for a margin of non-response. 1 This is equivalent to the weighted average of the estimates for each stratum, with weights equal to the population shares of each stratum. 69 ANNEX 2A: PRODUCTIVITY ANALYSIS This section examines in detail the effects of the investment climate on productivity in Uruguay. A subset of questions that capture the quality of the investment climate is selected from the Enterprise Survey according to two criteria: first, questions should (as much as possible) reflect objective measures, rather than perceptions. Subjective (perception) indicators (e.g., whether crime is a major obstacle to business) are difficult to compare across regions or countries since it is possible that individuals‘ standards differ due to cultural, geographic, or other reasons. On the other hand, objective measures (e.g., losses from theft) offer more precise assessments of investment climate quality that one can easily relate to performance indicators. Second, the interpretation of these questions with respect to their effect on productivity should be unambiguous. For example, the use of courts to solve payment disputes could reflect a well- functioning legal system, but it could also be the case that, because the legal system works well, firms have sufficient alternative means to settle their disputes outside courts, and thus avoid costly and lengthy legal procedures. In the latter case, we should observe a low proportion of disputes settled in court, but it would be wrong to conclude that the legal system does not function well. Further, variables are selected that have emerged in previous analyses of the Uruguayan economy as relevant investment climate aspects, as well as those where Uruguay seems to fare poorly when compared to its peer group countries.2 This selection still leaves a large number of potential variables to include in the analysis. In order to isolate a small group of variables that actually have an effect on productivity, the concept of Escribano and Guasch (2005) is applied moving from a general specification with many variables to a more parsimonious one by narrowing the selected variables to keep those that affect productivity significantly. However, the methodology applied in this study departs from Escribano and Guasch (2005) in several respects. The first estimations are conducted with fewer variables because of limits in the Uruguayan sample size and because priors are used about the relevant investment climate variables for Uruguay. In addition, we conduct the regression analysis for Uruguay data alone as well as for Uruguay data pooled with data from its peer countries.3 The latter may yield more efficient estimates because if investment climate conditions affect all the firms in a country in a similar fashion, then it is likely that individual analysis will not be able to capture the relevance of particular investment climate factors. Pooling the data also allows simulation exercises to estimate the productivity gains that Uruguayan firms could attain if investment climate conditions were to improve to a certain benchmark within the pool of countries. Two specifications are estimated. To investigate the impact of investment climate attributes on the labor productivity, the following is estimated (1) 2 For example, as noted in previous Chapters, the majority of firms see competition from the informal sector and access to finance as major obstacles to their operations. 3 The list of peer countries includes Argentina, Brazil, Chile, Costa Rica, Ireland, Latvia, Lithuania, Mauritius, Mexico, and Slovakia. Thailand is excluded from the analysis in this Chapter due to lack of governance questions. 70 where Y is output measured as firm sales or value added and L is labor; IC refers to investment climate variables; C to other firm control variables that are not directly related to the investment climate but that can affect enterprise performance, and v is a random error term. The following are included as control variables: sector, region, and country effects. These are variables that can capture unobservable firm characteristics not directly related to the investment climate but that can affect firm performance. Also, the estimations control for the (log) age of the firm, a size dummy, a dummy for (partly or fully) foreign-owned firms, and a dummy for exporting firms. The sample is limited to the manufacturing sector, so as not to impose an identical functional form for the production function for manufacturing, services, and construction. To analyze the impact of the investment climate on wages, a variation of equation (1) is estimated where the dependent variable is substituted by the average wage per worker paid by the firm. 4 (2) The dependent variables are in turn, (log) sales per worker, (log) value added per worker, or labor productivity (equation 1); and (log) average monthly wage (equation 2). We are interested in both how the investment climate affects firms‘ productivity and the impact on workers‘ wages. The selected explanatory variables cover four investment climate areas: ‗Governance and Regulations,‘ ‗Access to Finance,‘ ‗Technology Use,‘ and ‗Human Capital Accumulation.‘ Table A.2.1 presents the sample size and mean of each variable for the total sample, and the mean value for each peer group country. As mentioned previously, these variables were selected after performing several regression exercises using a large number of indicators. Other indicators used included inter alia: senior management's time for regulations, number of tax inspections, power outages‘ losses, wait time for telephone connection, transport losses dummy, time for clearing customs (imports), use of trade credit for investment (and working capital) dummy, use of financial institutions for investment (working capital) dummy, quality certification dummy, and share of skilled production workers. The fact that this second set of variables was not significant does not imply that the IC conditions that they are supposed to capture are not relevant. It is possible that the corresponding IC conditions are not well reflected, partly because the enterprise survey questions related to governance and regulation are more extensive and detailed relative to questions in the skills, technology, and finance areas. In the area of ‗Governance and Regulations‘, two variables are selected. The first one, crimeloss, captures the country‘s rule of law by measuring the costs of security and crime as a percentage of sales. It adds security expenses and sales lost due to theft, robbery, or arson. The average cost of crime is 1.48% of sales for the entire sample with 1.15% for Uruguay. These averages can be interpreted as an approximation of the impact of crime at the firm level, but they do not reflect the incidence of crime for the society as a whole. The second variable, reported, measures regulatory compliance and equals the percentage of sales a ‗typical firm‘ reports to the tax authorities. For the entire sample, on average, firms report 77.6% of total sales for tax purposes. Firms in Uruguay report on average 85.7%. Chile stands at the top of the sample with a 97.2% reporting rate. In ‗Access to Finance‘, the variable not credit constrained is constructed. This is a dummy variable that equals one if the firm has a loan, a credit line, or an overdraft facility, or if has not requested any loan because it does not need credit. Hence it is a measure of whether firms‘ credit needs are being 4Averagemonthly wage is calculated as monthly labor costs (including compensations and benefits) divided by total employees. 71 met in the financial system. In the full sample, 79.8% of firms claim to be not credit constrained. Uruguay is slightly below the average. The country with the highest percentage of non-constrained firms is Ireland with 97.4%. Less than half of all firms in Uruguay have a loan or credit line, and less than 65% have any form of credit (overdraft, loan, or credit line), which implies that a relatively large share of firms claim that they do not need credit. Perhaps firms have become more risk averse in the wake of the economic crisis or firms have low growth prospects since they continue to perceive macroeconomic instability as one of the most important investment climate constraints (Figure 2.2). The selected ‗Technology Use‘ variable, website, is equal to one if the firm uses its own website to communicate with clients and suppliers. For the full sample, 49% of firms have their own website, but there is a large variance across countries. In Slovakia, 94.7% of firms report having a website, while only 31% of Mexican and Uruguayan firms report having one. Finally, for ‗Human Capital Accumulation‘, we select training, which is equal to one if the firm has established training programs for its permanent workers. On average 43.5% of firms in the entire sample provide training for their permanent workers versus only 24.5% of firms in Uruguay. At the other end, 92% of Slovakian firms offer training. In the estimations, we need to address the potential problems caused by a low response rate, which reduces the sample size considerably. For instance, Table A.2.1 shows that out of a total of 396 firms in manufacturing in Uruguay, only 225 firms have observations for reported. This represents a non-response rate of 43%. To avoid losing as much as 50% of the sample in some cases, we substitute missing observations in each country with the mean for the corresponding region, industry, and size category. We establish a threshold of 10 observations as a minimum for the mean to be representative of the population. If the region-industry-size category has less than 10 firms (observations) in it, we use the region-industry mean, provided that the minimum number of observations is reached. If not, we use the industry, or region mean, depending on the number of observations in each. If none of these categories contains at least 10 observations, we do not substitute the missing value. The right-hand-side panel of Table A.2.1 provides summary statistics for the sample of IC measures after substitution. The number of observations increases dramatically after substitution, especially in Uruguay and Ireland for most variables, and in several countries for the variable training. Broadly, the average values are very similar to the averages before substitution, except for training, for which it decreases by about 5 percentage points, mostly due to the increase in the number of observations for the Uruguayan and Mexican samples, which have low averages for training. In Uruguay, however, the means before and after substitution are almost identical. Table A.2.2 reports summary statistics for the other control variables. These variables generally have a much higher response rate, so we do not substitute their missing values. The exception is capital stock, which has a non-response rate of 62.3%. However, we do not substitute capital stock numbers because the distribution of capital stock per worker in the Uruguayan sample has an extremely large variance, even within narrowly defined categories. It should also be noted that the response rate for materials is lower than the average; hence the number of observations for value added is considerably lower than for sales. Table A.2.3 provides pair-wise correlations between the dependent variables, the investment climate indicators, and the control variables. Most correlation coefficients are significant at the 10% level, but the correlation is generally low, well below 0.5, especially for the investment climate variables.5 5 The highest correlations between IC variables is 0.25 between using a website and providing training (in the Uruguay sample). In addition we computed the correlation between the percentage of sales reported for taxes and 72 We then proceed to run regressions for equations (1)-(2) above both for the Uruguayan sample alone and for the pooled sample including all 11 countries. Thailand is excluded from the regressions as it does not have any data on regulatory compliance. We use ordinary least squares (OLS) to estimate the effect of each investment climate attribute. While we cannot determine the direction of causality from this exercise, we can already observe which investment climate attributes are associated with better firm performance. Reverse causality issues are further discussed below. Table A.2.4 reports regression results for the OLS estimations. All reported standard errors are robust to the presence of heteroskedasticity. For labor productivity (column 3), we find that crimeloss, reported (regulatory compliance), and website have the expected sign and are significant at least at the 10% level for the Uruguay sample. The variable nonconstrained has a positive but not significant coefficient. Interestingly, for the IC variables most coefficients (except for training) are not statistically different between the regressions with only Uruguayan firms and the regressions with the pooled data. This confirms to a large extent our assumption that, after controlling for the industry, size composition of the sample and other unobservable aspects, the effects of the investment climate on an ‗average‘ firm‘s productivity is the same across the countries in the sample. Finally, for average wage we find that crimeloss and training affect wages significantly both in the Uruguay sample and in the pooled sample (where only website is not statistically significant). However, here we find that only two variables are within the same confidence interval for the Uruguay and the pooled regressions, which suggests that we cannot assume that the effects on wages are similar in Uruguay and the rest of the sample; specifically, effects in Uruguay appear to be larger in magnitude. Potential Gains from Improvements in Investment Climate Conditions We proceed to calculate the potential gains in productivity for Uruguayan firms were the IC conditions to improve up to a certain (admittedly arbitrary) benchmark level.6 We perform this simulation exercise for labor productivity, using the results from the value added regression reported in column (4) of Table A.2.4. For each one of our investment climate variables, we choose three benchmark levels: (i) the highest 75th percentile of the entire sample (25th percentile in the case of costs of crime) within the same industry and size; (ii) the average in Chile; and (iii) the average in Ireland.7 We then take the difference between the benchmark value and the country‘s industry-size average for each variable. Specifically, we calculate:   potential productivity gainU ,i ,s  IC B,i ,s  IC U ,i ,s * coefIC (4)   where IC B,i ,s  IC U ,i ,s is the difference between the ‗best practice‘ (meaning, the best 75th percentile within the same industry and size, the average for Chile, or the average for Ireland) and the average by industry and size in Uruguay. We multiply this difference by the regression the percentage of workers reported for taxes, as an alternative measure of regulatory compliance. This correlation is 0.47. 6 This exercise is similar to Loayza, Fajnzylber and Calderón (2005), and Loayza, Oviedo and Servén (2004). 7 Due to differences in the industries covered in each country‟s survey, in (ii) and (iii) we only use the overall country average as a benchmark, instead of taking averages by industry and size. 73 coefficients from the OLS estimations reported in column (4) of Table A.2.4 for the pooled sample. We use the pooled OLS regressions as they provide coefficients that we can attribute to all countries, and even if we cannot infer causality from them, we can establish the increases in productivity associated with a better investment climate. Finally, we average out all industry-size gains to obtain one number by country. Figure A.2.1 reports the results for Uruguay and its peer countries. The first striking result from these simulations is that Uruguay is systematically among the three countries that would gain the most in all performance indicators from improving their investment climate, together with Mexico and Brazil. Not surprisingly, Chile and Ireland are among the countries with the smallest potential gains, when comparing each country to the top 75th percentile. For example, if Uruguay reached the top 75th percentile of all six indicators its firms would enjoy a 22 percent higher productivity. The largest gains would be achieved by improving in the area of technology use where reaching the 75th percentile is associated with about 6.5% higher productivity. Labor productivity is also higher with increased training (4.2%) and regulatory compliance (5.8%). Our results indicate that small firms have considerably larger potential gains than medium and large firms. In addition to comparing the performance of Uruguayan enterprises to a hypothetical best practice, we compare them to Chile and Ireland which depict the best investment climate in the sample. The aggregate potential gains for Uruguay are high when compared to Chile (21%) and even higher with respect to Ireland (30%). (See Figure A.2.1.) Are Exporters Different? One important question for Uruguay is whether firms that enter the global markets perform better than domestically-oriented firms. A priori, it is reasonable to expect that firms that reach to larger markets are able to exploit economies of scale and scope that allow them to raise their productivity. At the same time, global markets are much more competitive and only the best performers survive. Indeed, the literature usually finds significant exporter premia for several measures of performance in most countries. However, there are different views as to what makes exporters more productive. On the one hand, the ‗learning by exporting‘ hypothesis states that exporters start out with the same productivity as non-exporters, and that exposure to foreign (more competitive) markets puts pressure on firms to become more productive or be selected out. An alternative view is that exporters are more productive before they begin to export and that being more productive is what encourages them to enter the export market. In this study, we perform two simple exercises, the first aimed at testing whether there is an exporter premium in Uruguay, and the second to examine whether investment climate conditions affect exporters differently than non-exporters.8 Unfortunately, due to the lack of time-series data and the limited size of the sample, we could not test the ‗self-selection‘ versus ‗learning by exporting‘ hypotheses.9 First, we find that exporters in Uruguay perform better than domestically-oriented firms. Table A.2.5 reports exporter premium regressions, which resemble those from Table A.2.4 but without the investment climate variables. We find that exporters depict 47% higher sales per 8 Here we include all sectors: manufacturing, services, and construction. 9 Casacuberta and Gandelman (2007) perform a detailed study of exporter premia in Uruguay using firm census data. They find that allowing factor elasticities to vary between exporters and non-exporters provides a much more accurate measure of productivity for each group, and they find a significant output gap between both groups. In addition, they find that differences in factor use explain most of the output gap differences, while differences in TFP explain roughly one-third of the difference but only for firms that export less than 50 percent of their sales. 74 worker and 45% higher average wage than non-exporters. This is true after controlling for several firm characteristics as in columns (2) and (4). Second, we examine exporter premium by firm size. We divide the sample into small (less than 20 employees), medium (20-99 employees), and large (100 or more employees), and we estimate regressions similar to Table A.2.5 (second column of each performance measure). Table A.2.6 reports the results from these regressions. A caveat, however, is necessary. Dividing the sample by size implies that for each regression the number of observations falls substantially, which could possibly make our results less robust in this exercise. Interestingly, we find that the exporter premium is more significant for small firms than for medium and large firms. One possible explanation for this finding is that productivity differences between exporters arise from economies of scale and scope, and these are mainly related to size. For small firms, only those that are highly productive relative to their peers are able to compete in export markets. Finally, in Table A.2.7, we report OLS regression results similar to those of Tables A.2.4 but restricting the sample to exporting firms only. The purpose of this exercise is to examine whether exporters suffer from investment climate conditions more or less than non-exporters. We find that all coefficients for the exporter-only regressions are not statistically different to the regressions with all firms (except for audited in the labor productivity regression). However, there is a smaller number of significant coefficients which is probably due to the much smaller sample (140-167 observations). 75 Table A.2.1: Summary Statistics of Regression Variables No substitutions With substitutions crimeloss reported noconstr website training crimeloss reported noconstr website training Uruguay Mean 1.151 85.665 0.816 0.312 0.245 1.151 86.294 0.816 0.312 0.238 Observations 395 225 394 396 360 396 396 396 396 396 Argentina Mean 1.875 82.805 0.833 0.741 0.521 1.873 82.65 0.833 0.743 0.506 Observations 743 662 742 739 650 746 746 746 746 746 Brazil Mean 2.219 67.351 0.85 0.731 0.671 2.221 67.42 0.85 0.731 0.671 Observations 1626 1513 1634 1641 1638 1641 1641 1641 1641 1641 Chile Mean 0.858 97.241 0.946 0.738 0.707 0.854 97.203 0.946 0.737 0.706 Observations 661 650 684 637 685 688 688 688 688 688 Mexico Mean 1.255 72.514 0.748 0.313 0.246 1.258 72.403 0.749 0.314 0.248 Observations 1156 1023 1119 1141 1118 1161 1161 1161 1161 1160 Costa Rica Mean 1.755 71.631 0.848 0.769 0.464 1.726 71.464 0.848 0.768 0.464 Observations 289 287 342 143 343 343 343 343 343 343 Mauritius Mean 1.399 87.887 0.933 0.447 0.626 1.379 88.019 0.933 0.449 0.627 Observations 149 155 179 179 179 184 184 184 184 184 Ireland Mean 0.504 96.573 0.974 0.902 0.781 0.433 96.929 0.974 0.89 0.78 Observations 167 171 265 173 301 304 305 310 305 311 Latvia Mean 1.385 93.3 0.853 0.735 0.656 1.385 93.294 0.853 0.735 0.655 Observations 34 30 34 34 32 34 34 34 34 34 Lithuania Mean 0.356 91.028 0.867 0.889 0.45 0.356 90.889 0.867 0.889 0.443 Observations 45 36 45 45 40 45 44 45 45 45 Slovakia Mean 1.867 98 0.946 0.947 0.919 1.852 97.582 0.944 0.947 0.919 Observations 36 31 37 38 37 38 38 38 38 3 Total Mean 1.48 77.597 0.798 0.49 0.435 1.478 77.576 0.798 0.493 0.379 Observations 7244 5334 7404 7135 7352 7549 6164 7555 7550 7555 76 Table A.2.2. Summary Statistics of Firms’ Characteristics used as Control Variables SME age foreign exporter audit Uruguay Mean 0.673 2.991 0.082 0.394 0.212 Observations 395 394 396 395 391 Argentina Mean 0.43 3.117 0.121 0.524 0.64 Observations 743 746 746 746 741 Brazil Mean 0.182 2.855 0.054 0.308 0.191 Observations 1635 1641 1641 1635 1635 Chile Mean 0.266 3.263 0.157 0.43 0.601 Observations 688 688 688 688 670 Mexico Mean 0.607 2.464 0.032 0.066 0.191 Observations 1161 1092 1161 1161 1126 Costa Rica Mean 0.63 2.819 0.09 0.389 0.56 Observations 343 341 343 342 343 Mauritius Mean 0.205 2.981 0.103 0.68 0.869 Observations 166 181 184 181 183 Ireland Mean 0.543 3.042 0.1 0.57 0.942 Observations 311 172 311 307 172 Latvia Mean 0.441 2.605 0.353 0.588 0.636 Observations 34 34 34 34 33 Lithuania Mean 0.4 2.388 0.356 0.644 0.556 Observations 45 45 45 45 45 Slovakia Mean 0.316 2.742 0.211 0.737 0.784 Observations 38 38 38 38 37 Total Mean 0.512 2.786 0.077 0.285 0.391 Observations 7519 7340 7556 7538 7337 77 Table A.2.3: Partial Correlations between Performance and Investment Climate variables (Uruguay and Full Sample, Manufacturing) Losses from sec. Log value added Access to credit Foreign-owned External audit Log sales per compliance Regulation per worker Log wage Exporter Web use Training & crime Log age worker SME Uruguay Log sales per worker 1 Log value added per worker 0.9073* 1 Log wage 0.6557* 0.6899* 1 Losses from sec. & crime -0.2646* -0.2276* -0.2178 1 Regulation compliance 0.2305* 0.2517* 0.1987* 0.0196 1 Access to credit 0.2644* 0.2910* 0.2649* -0.0868* 0.0961* 1 External audit 0.3038* 0.3317* 0.2997* -0.0672 0.1325* 0.1491* 1 Web use 0.2655* 0.3320* 0.3178* -0.0437 0.1320* 0.2168* 0.1865* 1 Training 0.2596* 0.3153* 0.3426* -0.0043 0.1332* 0.0793 0.2562* 0.2451* 1 SME -0.2449* -0.2387* -0.2473 -0.0002 -0.1317* -0.1435* -0.2168* -0.2835* -0.2432* 1 Log age 0.2356* 0.3216* 0.2989* -0.1069* 0.0762 0.2375* 0.0966* 0.2198* 0.1056* -0.1604* 1 Foreign-owned 0.2669* 0.2527* 0.2662* -0.0465 0.0574 0.1232* 0.3141* 0.0749 0.1626* -0.1534* 0.0564 1 Exporter 0.2643* 0.2713* 0.3253* 0.0453 0.1712* 0.0825 0.2357* 0.3074* 0.2804* -0.4038* 0.0766 0.0811 1 Full sample Log sales per worker 1 Log value added per worker 0.9319* 1 Log wage 0.7043* 0.7362* 1 Losses from sec. & crime -0.1626* -0.1242* -0.0834* 1 Regulation compliance 0.2898* 0.2785* 0.2470* -0.0497* 1 Access to credit 0.1364* 0.1135* 0.0860* -0.0632* 0.0827* 1 External audit 0.2202* 0.1591* 0.0699* -0.0768* 0.2349* 0.1622* 1 Web use 0.2191* 0.2115* 0.2180* 0.0305* 0.0918* 0.1037* 0.0481* 1 Training 0.1599* 0.1318* 0.0692* 0.0189 0.1066* 0.1343* 0.2614* 0.2382* 1 SME -0.0594* -0.0144 0.0824* -0.0037 -0.0517* -0.1486* -0.2417* -0.1677* -0.3326* 1 Log age 0.2057* 0.1984* 0.2174* -0.0246* 0.1259* 0.1160* 0.0864* 0.1607* 0.0811* -0.1446* 1 Foreign-owned 0.2077* 0.1734* 0.0950* -0.0272* 0.1402* 0.0664* 0.2405* 0.0750* 0.1749* -0.1935* -0.0072 1 Exporter 0.2493* 0.2210* 0.1320* -0.0474* 0.1736* 0.1433* 0.3268* 0.2081* 0.2987* -0.3329* 0.1317* 0.2982* 1 * significant at 10% level 78 Table A.2.4: Productivity Regressions Sales per worker Value added per worker Wage Uruguay Full Sample Uruguay Full Sample Uruguay Full Sample Costs of security and theft -5.204*** -4.473*** -3.231*** -3.381*** -2.724*** -1.906*** [0.823] [0.705] [0.820] [0.706] [0.540] [0.590] Tax compliance 0.737** 0.466*** 0.689* 0.408*** 0.114 0.196*** [0.326] [0.070] [0.407] [0.071] [0.257] [0.059] Access to credit 0.186 0.162*** 0.114 0.169*** 0.139 0.132*** [0.166] [0.044] [0.208] [0.048] [0.123] [0.035] Use of website 0.086 0.204*** 0.247* 0.205*** 0.12 0.056 [0.114] [0.043] [0.134] [0.045] [0.100] [0.038] Training 0.124 0.134*** 0.076 0.146*** 0.249** 0.100*** [0.138] [0.038] [0.170] [0.040] [0.119] [0.034] Log of firm's age 0.089 0.128*** 0.202*** 0.146*** 0.123** 0.152*** [0.064] [0.024] [0.077] [0.025] [0.052] [0.020] Audited 0.288* 0.286*** 0.385* 0.246*** 0.224** 0.102*** [0.164] [0.043] [0.196] [0.046] [0.107] [0.037] Foreign owned 0.348* 0.499*** 0.314 0.401*** 0.279** 0.353*** [0.180] [0.068] [0.206] [0.072] [0.123] [0.059] Exporter 0.367*** 0.317*** 0.383** 0.391*** 0.264** 0.221*** [0.124] [0.043] [0.151] [0.047] [0.108] [0.039] Log capital per worker 20-99 employees 0.067 0.021 -0.028 -0.05 0.01 -0.058 [0.124] [0.044] [0.150] [0.047] [0.106] [0.036] 100+ employees 0.247 -0.007 -0.019 -0.025 0.169 -0.116** [0.203] [0.060] [0.300] [0.064] [0.141] [0.052] Garments -0.845*** -0.595*** -0.689*** -0.452*** -0.242* -0.168*** [0.215] [0.064] [0.243] [0.068] [0.145] [0.055] Textile -0.441 -0.105 -0.301 -0.014 0.007 0.166*** [0.271] [0.071] [0.243] [0.072] [0.173] [0.058] Metals & machinery 0.792*** -0.065 0.039 1.686*** 0.258*** [0.193] [0.069] [0.072] [0.147] [0.062] Chemicals 0.280* 0.330*** 0.376* 0.412*** 0.669*** 0.395*** [0.151] [0.080] [0.192] [0.086] [0.141] [0.070] Electronics -0.085 0.098 0.262*** [0.109] [0.105] [0.087] Non metallic -0.113 -0.229*** -0.299 -0.245*** 0.11 -0.094 [0.160] [0.086] [0.208] [0.092] [0.190] [0.074] Other manufacturing -0.103 -0.343*** -0.451 -0.265*** 0.305* 0.037 [0.182] [0.063] [0.356] [0.070] [0.162] [0.055] Argentina -0.111 -0.012 0.196 [0.129] [0.145] [0.132] Brazil -0.436*** -0.587*** -0.474*** [0.098] [0.105] [0.097] Chile 0.351** 0.132 0.181 [0.142] [0.148] [0.126] Mexico -0.256** 0.05 -0.266** [0.116] [0.128] [0.128] Costa Rica -0.472*** -0.544*** -0.371*** [0.105] [0.112] [0.085] Mauritius -0.972*** -0.294 -0.532*** [0.140] [0.180] [0.126] Ireland 0.955*** 1.233*** 1.477*** [0.197] [0.293] [0.105] Latvia -0.045 -0.106 0.03 [0.159] [0.201] [0.165] Lithuania 0.107 0.019 -0.1 [0.158] [0.250] [0.187] Slovakia -0.169 -0.154 0.188 [0.243] [0.200] [0.126] Constant 8.868*** 8.958*** 8.029*** 8.321*** 4.836*** 4.988*** [0.319] [0.118] [0.407] [0.128] [0.254] [0.102] Observations 322 4814 215 4308 273 4669 R-squared 0.379 0.309 0.421 0.284 0.461 0.224 Robust standard errors in brackets * significant at 10%; ** significant at 5%; *** significant at 1%. Within country region dummies included 79 Table A.2.5: Exporter Premium Dependent variable: Sales per worker Wages (1) (2) (3) (4) Exporter 0.492*** 0.471*** 0.448*** 0.450*** [0.122] [0.115] [0.103] [0.095] Log of firm's age 0.215*** 0.180*** [0.060] [0.046] Foreign owned 0.712*** 0.621*** [0.134] [0.099] SME (<100) -0.205* -0.049 -0.216** -0.077 [0.105] [0.109] [0.088] [0.087] Services 0.658*** 0.708*** 0.056 0.092 [0.117] [0.115] [0.098] [0.096] Construction & Transport 0.371* 0.304 0.645*** 0.588*** [0.206] [0.199] [0.152] [0.136] Constant 10.028*** 9.205*** 5.843*** 5.147*** [0.100] [0.231] [0.085] [0.162] Observations 506 505 456 456 R-squared 0.08 0.15 0.13 0.22 Robust standard errors in brackets * significant at 10%; ** significant at 5%; *** significant at 1% Regional dummies included 80 Table A.2. 6: Exporter Premium by Size Number of employees Sales per worker < 20 [20-99] 100+ Log of firm's age 0.033 0.322*** 0.314** [0.088] [0.089] [0.146] Foreign owned 0.454 0.645*** 0.995*** [0.306] [0.204] [0.207] Exporter 0.769*** 0.161 0.115 [0.210] [0.170] [0.219] Services 1.058*** 0.386** -0.045 [0.162] [0.188] [0.267] Construction & Transport 0.373 0.407 -0.577* [0.371] [0.313] [0.294] Constant 9.520*** 9.094*** 9.286*** [0.282] [0.326] [0.529] Observations 215 210 80 R-squared 0.2 0.13 0.31 Wage Log of firm's age -0.008 0.275*** 0.333*** [0.074] [0.067] [0.096] Foreign owned 0.411** 0.487*** 0.815*** [0.196] [0.143] [0.163] Exporter 0.523*** 0.374*** 0.152 [0.192] [0.140] [0.214] Services 0.164 -0.084 -0.102 [0.147] [0.158] [0.212] Construction & Transport 0.537*** 0.652*** 0.131 [0.193] [0.149] [0.345] Constant 5.556*** 4.910*** 4.937*** [0.224] [0.224] [0.389] Observations 203 178 75 R-squared 0.08 0.25 0.41 Robust standard errors in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 81 Table A.2.7: Investment Climate Effects on Exporters’ Productivity Dependent variable: Sales per worker Wages (1) (2) - - 2.880** Costs of security and theft 5.792*** * [-8.216 - [-3.741 - -3.369] -2.020] Tax compliance 1.071** 0.583 [0.015 - [-0.267 - 2.127] 1.434] Access to credit 0.548* 0.24 [-0.038 - [-0.198 - 1.134] 0.678] Audited 0.096 0.14 [-0.310 - [-0.122 - 0.502] 0.401] Use of website 0.380** 0.185 [0.025 - [-0.104 - 0.736] 0.473] 0.588** Training 0.392** * [0.025 - [0.294 - 0.760] 0.883] 0.240** Log of firm's age 0.124 * [-0.055 - [0.072 - 0.303] 0.408] 0.454** Foreign owned 0.648*** * [0.201 - [0.141 - 1.095] 0.767] SME (<100) 0.207 0.044 Of them on a timely manner, and lack a Vietnamese say the timing of [-0.184 - [-0.330 - the ISA to sell U.S. and how would I have enough money on the 0.598] 0.418] Services 0.147 -0.339 [-0.455 - [-0.838 - 0.749] 0.160] 0.666** Construction & Transport -0.27 * [-0.721 - [0.267 - 0.182] 1.066] 4.296** Constant 8.156*** * [6.984 - [3.496 - 9.328] 5.097] Observations 167 140 R-squared 0.346 0.46 Robust 95% confidence intervals in brackets * significant at 10%; ** significant at 5%; *** significant at 1% Region dummies included 82 Figure A.2.1: Differences in Value Added per Worker Associated with a Better Investment Climate A. To the 75th Percentile of the Sample Mexico 32% Brazil 22.4% Uruguay 22% Costa Rica 20.6% Mauritius 17.9% Argentina 17.4% Latvia 10.8% Lithuania 10.2% Slovakia 7% Chile 6.9% Ireland 3.2% percentage gain Crime & security costs Regulatory compliance Access to credit Technology use Human capital accumulation B. To the Average of Chile Mexico 27.7% Uruguay 20.8% Brazil 19.1% Costa Rica 18.7% Argentina 14.5% Mauritius 13.1% Lithuania 7.5% Latvia 5.8% Slovakia 3.4% Ireland 0.3% percentage gain Crime & security costs Regulatory compliance Access to credit Technology use Human capital accumulation C. To the Average of Ireland Mexico 34.8% Uruguay 29.7% Brazil 24.9% Costa Rica 23.8% Mauritius 19% Argentina 17.7% Latvia 11.5% Lithuania 9.2% Chile 6.2% Slovakia 5.1% percentage gain Crime & security costs Regulatory compliance Access to credit Technology use Human capital accumulation 83 ANNEX 2B: DETERMINANTS OF BRIBE PAYMENTS IN URUGUAY Table A2B.1: Determinants of Bribe Payments in Uruguay Regression Results Dependent variable: dummy=1 if firm reports payments when dealing with the government. Uruguay Full Sample (continued) Full Sample Requested public services 0.083** 0.067*** Argentina 0.416*** [2.00] [4.71] [5.46] Trusts courts' enforcement power 0.005 -0.061*** Brazil 0.509*** [0.14] [4.20] [7.07] Access to finance -0.012 -0.039** Chile -0.033 [0.22] [2.06] [-0.51] Regulatory compliance -0.198** -0.270*** Mexico -0.064 [2.12] [9.32] [-0.95] Log of firm's age -0.023 -0.014 Costa Rica 0.077 [1.17] [1.48] [1.27] Exporter 0.077* -0.017 Mauritius -0.049 [1.66] [0.99] [-0.56] Foreign ownership -0.008 -0.036 South Africa -0.162*** [0.17] [1.47] [-2.75] Medium (20-99) 0.016 -0.003 Ireland 0.024 [0.33] [0.20] [0.25] Large (100+) -0.076 -0.050** Latvia 0.23** [1.61] [2.41] [1.99] Lithuania 0.359*** [4.5] Slovakia 0.505*** [6.52] Observations 186 4903 Pseudo R2 0.11 0.35 Notes: the dependent variable is a dummy variable that takes the value 1 if the firm agrees that informal payments are necessary to obtain government contracts, or if the firm has had to make informal payments to obtain any of the following: electricity connection, telephone connection, construction permit, import license, operating license, or if it had to make an informal payment whenever inspected by tax officials. The variable is equal to zero if the firm answers "No" to all these questions. Robust z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. Industry effects are included in both regressions. 84 ANNEX 2C: SELECTED INTERNATIONAL INDICATORS Table A2C.1:Payroll Taxes for Pensions and All Social Insurance, Selected OECD and Latin American Countries Source: Palacios and Pallares-Miralles (2000), as cited in World Bank (2005a). 85 Table A2C.2 International Comparison for VAT rates Standard rate Other positive rates (percent) (percent) Argentina 21 10.5; 27 Australia 10 Austria 20 10; 16 Belgium 21 6; 12 Brazil 1/ 20.5 22 Bulgaria 20 Canada 7 Chile 19 Colombia 16 7; 10; 20; 35 Costa Rica 13 5 Czech Republic 19 5 Denmark 25 Estonia 18 5 Finland 22 8; 17 France 19.6 2.1; 5.5 Germany 16 7 Greece 18 4; 8 Hungary 25 5; 15 Iceland 24.5 14 Ireland 21 4.3; 13.5 Israel 17 9 Italy 20 4;10 Korea 10 Latvia 18 5 Lithuania 18 5; 9 Mexico 15 10 Netherlands 19 6 New Zealand 12.5 Norway 24 12 Peru 19 Poland 22 3; 7 Portugal 19 5; 12 South Africa 14 Spain 16 4; 7 Sweden 25 6; 12 Switzerland 7.6 2.4; 3.6 Turkey 2/ 18 1.; 8 United Kingdom 17.5 5 Uruguay3/ 22 (basic) 10 (minimum) Sources: International Bureau of Fiscal Documentation (IBFD, 2004); and Corporate Taxes 2003-2004, Worldwide Summaries in Bird and Gendron (Forthcoming) (PricewaterhouseCoopers). 1/ Effective rates of 7.5 percent and 13.6 percent apply for interstate transaction between registered taxpayers. 2/ Rates of 26 percent and 40 percent rates apply to luxury goods. 3/ Effective July 2007 86 ANNEX 3: SELECTED INFORMATION ON URUGUAY’S FINANCIAL SECTOR AND INTERNATIONAL COMPARISONS Figure A.3.1. Bank Financing of Working Capital Figure A.3.2. Bank Financing of Fixed Assets (as percent of total) (as percent of total) 45.7 58.2 31.7 27.3 26.6 17.3 16.3 25.2 22.1 12.5 10.5 10.5 14.8 12.9 7.8 7.8 6.9 11.8 10.1 9.7 9.1 8.9 2.0 3.7 3.7 2.2 l ay ia a nd le a ru a o ov d os az i Li ina Ire a s ic ki ni n ic tv hi ic iu gu La l la y la Sl tvia ca nd ile So ta R th a Ire a o at nd r i ua ex a a fr La s t C az rit B ua ki ni ic en ai in Co fric iu A Ri Ch la la th M at au Br ua Th ex ug nt rit rg U h ai Sl ov a A M ut ge M au A Th Ur st C h Li Ar M ut So Source: World Bank Investment Climate Surveys Source: World Bank Investment Climate Surveys Table A.3.1. Financial Structure of the Financial System December 2001 December 2006 Type of Institution Number of Assets in US % of Total Number of Assets in % of Total Institutions millions Assets Institutions US Assets Comercial banks, of which 22 20608 88.2 16 14168 71.6 Public sector-owned banks 2 7456 31.9 3 8307 42.0 BROU 1 4788 20.5 1 5739 29.0 BHU 1 2668 11.4 1 1414 7.1 Nuevo Banco Comercial 1/ 1 1154 5.8 Domestic private banks 3 3172 13.6 0 0 0.0 Foreign banks 2/ 17 9980 42.7 13 5861 29.6 Nonbank intermediaries 25 1141 4.9 13 1917 9.7 Cooperatives 6 478 2.0 2 26 0.1 Finance Houses 7 540 2.3 6 142 0.7 External Financial Institutions 12 123 0.5 5 1732 8.8 Management Companies for 1 12 0.0 4 17 0.1 Savings Society Funds Currency Exchange Houses (as 77 115 0.6 of 31-12-05) Credit Providers 9 218 1.1 Pension Funds 4 1045 4.5 4 2635 13.3 Insurance companies 17 572 2.4 16 733 3.7 Total financial system 68 23366 100 135 19786 100 1/ NBC is a private bank. Public sector owns 40% of total shares. 2/ Includes both foreign branches and subsidiaries. Source: Banco Central del Uruguay 87 Table A.3.2. Estimating the Determinants of Demand and Access to Credit. Regression Results Full Sample Uruguay Demand for Access to Demand for Access to Credit Credit Depth Credit Credit Depth SME (<100) -0.264*** -0.247*** -0.328*** -0.429*** -0.348* -0.351*** [6.06] [5.44] [12.73] [2.78] [1.84] [4.57] Exporter 0.159*** 0.236*** 0.199*** 0.607*** 0.263 0.285*** [3.63] [5.19] [8.18] [3.08] [1.21] [3.26] Audited 0.051 0.171*** 0.158*** 0.319* 0.384* 0.338*** [1.13] [3.71] [6.09] [1.72] [1.81] [3.92] Foreign owned -0.234*** 0.173** -0.174*** -0.334 0.21 -0.118 [4.10] [2.51] [5.30] [1.55] [0.69] [1.17] Log of firm's age 0.049** 0.155*** 0.028** 0.277*** 0.317*** 0.117*** [2.01] [5.93] [2.08] [4.08] [3.26] [3.42] Incorporated -0.078 0.093 0.032 -0.113 0.559** 0.196 [1.42] [1.60] [0.98] [0.49] [2.31] [1.61] Training 0.272*** 0.16 [6.71] [0.73] Macroeconomic instability perception 0.135*** 0.132 [3.42] [0.81] Argentina 0.231*** -0.099 0.046 [2.65] [1.18] [1.03] Brazil 0.128 0.051 0.627*** [1.49] [0.62] [13.44] Chile 0.677*** 0.431*** 0.649*** [6.18] [4.26] [13.02] Mexico -1.011*** -0.184** -0.537*** [13.27] [1.99] [13.00] Costa Rica 0.286** 0.134 0.383*** [2.35] [1.21] [5.86] Mauritius 0.519** 0.22 0.621*** [2.58] [1.31] [7.09] South Africa -0.291*** 0.252** 0.211*** [2.92] [2.20] [3.48] Ireland 0.233* 1.049*** 0.271*** [1.76] [4.82] [3.78] Latvia -0.507*** -0.258** 0.425*** [4.24] [1.96] [4.46] Lithuania -0.502*** -0.328** -0.132 [4.14] [2.53] [1.53] Slovakia -0.755*** 0.009 0.036 [6.31] [0.07] [0.41] Thailand -0.05 0.481*** 1.379*** [0.54] [4.95] [26.75] Constant 0.910*** 0.429*** 0.988*** 0.249 -0.494 0.480*** [6.85] [3.24] [13.84] [0.64] [0.97] [2.62] Robust z statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 88 Table A.3.3. Estimating the Determinants of Investment. Regression Results Determinants of investment a Dependent variable: log of investment (in USD) Uruguay Full sample Access to credit 1.847* 0.389** [1.005] [0.177] Audited 0.017 0.550*** [0.453] [0.142] Foreign owned 0.346 1.056*** [0.586] [0.153] Exporter 0.747 0.381*** [0.611] [0.123] Incorporated -0.186 -0.245* [0.474] [0.149] Log of firm's age -0.579** 0.017 [0.271] [0.078] 20-99 employees 1.548*** 0.962*** [0.442] [0.144] 100+ employees 2.830*** 2.351*** [0.578] [0.172] Argentina 1.054*** [0.293] Brazil 5.563*** [0.494] Chile 4.871*** [0.579] Mexico -0.276 [0.436] Costa Rica 5.283*** [0.578] Mauritius 6.555*** [0.558] Ireland 1.007*** [0.363] Latvia 0.565 [0.361] Lithuania 0.798*** [0.299] Slovakia 0.606* [0.313] Constant 7.441*** 8.058*** [0.726] [0.375] Observations 330 4379 R-squared 0.163 0.462 Note: 1/ The dependent variable is the sum of all expenditures in machinery, equipment, vehicles, land, and buildings, plus the expenditures in R&D in the past fiscal year. Within-country region dummies and sectoral dummies included in both regressions. Robust standard errors in brackets. * significant at 10%; ** significant at 5%; *** significant at 1% 89 ANNEX 4: INDICATORS AND ECONOMETRIC ANALYSIS ON LABOR MARKETS AND SKILLS Table A.4.1: Unemployment Duration (Average number of months) Worker Unemployment Worker Unemployment characteristics duration characteristics Duration Total 3.16 Age Women 3.35 14- 18 3.02 Men 2.91 19- 23 2.91 24-28 2.83 Educational level 29-33 3.07 Primary incomplete 3.03 34-38 3.09 Primary complete 3.18 39-43 3.23 High school incomplete 3.24 44-48 3.90 High school complete 2.97 49-53 3.99 College incomplete 3.13 54-58 4.38 College complete 2.90 59-65 2.95 Source : Own estimations based on microdata from ENHA, 2006 Figure A.4.1: Share of Infomality by Worker’s Characteristics 90% 82% 80% Proportion of informal workers 70% 65% 59% 58% 60% 60% 50% 47% 46% 45% 40% 40% 40% 33% 33% 31% 30% 22% 20% 10% 5% 0% Female Rural Training course No training course Urban Male Half the legal MW or less Secondary complete Secondary incomplete Between half and twice Twice the MW or higher Primary complete Primary incomplete Tertiary complete Tertiary incomplete Source: Own estimates based on microdata from ENHA, 2006 90 Box A.4.1: Methodology of Focus Groups with Workers and Entrepreneurs. Focus groups are a qualitative and exploratory technique for information gathering, which consists of generating free and guided discussions about a specific subject among a previously selected group of people. Focus groups generate first hand information. The added value of a focus group comes from the group interaction. In this dynamic setting, a person‘s response can become a stimulus to another, generating an exchange of responses with richer results than if individuals from the group made independent contributions. For our study, the focus groups were conducted by the Foro Consultivo Económico Social del MERCOSUR and comprised three groups with unionized workers and three groups with entrepreneurs from different economic sectors. No focus group was organized with not unionized workers since these workers interests have been and are still are represented by the Central Unica de Trabajadores. Focus groups with workers included participants from industry, commerce, service, and the energy sectors. Focus groups with entrepreneurs included participants from the following sectors: textile (2), wood (2), forestry (1), education (2), subcontracting (2), industry (3), fishing (1), rural association (1), agro-industry (3), and chambers of industry (3). Focus group structure  Introduction: Project presentation, definition of work session objectives, and methodology description.  Debate: Using the findings of the Uruguay Investment Climate Survey as a starting point, people participated in the group by sharing knowledge and experiences. The group analyzed different opinions offered by individuals through a participative debate.  Main discussion elements: Besides an open discussion on the key investment climate obstacles in Uruguay, the main elements of discussion were labor relations and skills development. In addition, the following issues were raised to clarify the responses of the investment climate survey: competition with the informal economy, crime and violence, and taxes. 91 Table A.4.2: Determinants of Informality in Uruguay. Probit Results Dependent Variable: =1 if informal worker, =0 if formal worker Model [I] [II] [III] [IV] Men (=1 if men) -0.113 [-98.00] *** -0.110 [-95.1] *** -0.074 [-62.4] *** -0.074 [-61.7] *** Age (in years) -0.013 [-52.0] *** -0.013 [-49.5] *** -0.005 [-18.3] *** -0.005 [-18.1] *** Age2 0.000 [64.6] *** 0.000 [62.5] *** 0.000 [34.3] *** 0.000 [34.3] *** Years of education -0.018 [-134.2] *** -0.017 [-130.4 *** -0.014 [-100.3] *** -0.014 [-100.2] *** Race (=1 if White) -0.019 [-11.5] *** -0.018 [-10.5] *** -0.008 [-4.8] *** -0.009 [-5.0] *** Took non-transferable training course (=1 if took training) -0.029 [-11.0] *** -0.026 [-10.1] *** -0.004 [-1.5] *** -0.005 [-2.0] *** Took transferable training course (=1 if took training) -0.142 [-64.1] *** -0.137 [-61.7] *** -0.117 [-51.3] *** -0.118 [-51.5] *** Job Tenure (in years) -0.004 [-59.7] *** -0.003 [-54.1] *** -0.002 [-37.7] *** -0.003 [-39.3] *** Underemployed (=1 if underemployed) 0.174 [141.0] *** 0.173 [140.1] *** 0.156 [123.8] *** 0.155 [122.5] *** Urban areas (=1 if urban areas) -0.118 [-65.0] *** -0.116 [-64.2] *** -0.117 [-62.8] *** -0.116 [-62.3] *** Montevideo -0.068 [-61.2] *** -0.065 [-58.7] *** -0.046 [-40.7] *** -0.047 [-41.1] *** Fishing a 0.044 [3.8] *** 0.039 [3.4] *** -0.016 [-1.4] -0.017 [-1.5] Mining -0.151 [-14.4] *** -0.151 [-14.4] *** -0.133 [-11.9] *** -0.134 [-12.1] *** Manufacture -0.160 [-82.0] *** -0.162 [-82.8] *** -0.154 [-76.6] *** -0.173 [-74.9] *** Electricity, gas, water -0.372 [-73.5] *** -0.371 [-73.5] *** -0.369 [-68.3] *** -0.370 [-69.1] *** Construction 0.082 [31.9] *** 0.079 [30.8] *** 0.088 [33.3] *** 0.030 [9.9] *** Commerce, vehicle/home appliance repairs 0.052 [25.2] *** 0.050 [24.3] *** 0.051 [24.5] *** 0.040 [16.6] *** Hotel and restaurants -0.179 [-58.7] *** -0.180 [-59.0] *** -0.161 [-50.9] *** -0.144 [-36.9] *** Transport, storage, communication -0.114 [-46.2] *** -0.112 [-45.4] *** -0.083 [-32.2] *** -0.106 [-37.4] *** Financial services -0.314 [-82.5] *** -0.309 [-79.9] *** -0.291 [-71.4] *** -0.294 [-72.5] *** Real state activities 0.011 [4.1] *** 0.017 [6.3] *** 0.043 [15.5] *** 0.038 [13.7] *** Pub. adm and defense, soc. sec. plan. -0.450 [-139.0] *** -0.450 [-139.2] *** -0.448 [-131.4] *** -0.448 [-132.1] *** Teaching -0.350 [-157.7] *** -0.349 [-156.9] *** -0.330 [-137.3] *** -0.332 [-138.8] *** Social and health services -0.320 [-151.4] *** -0.318 [-149.9] *** -0.300 [-132.6] *** -0.302 [-133.6] *** Other comm., soc., activ./services -0.073 [-27.6] *** -0.069 [-26.0] *** -0.045 [16.4] *** -0.046 [-17.0] *** Domestic services 0.475 [157.0] *** 0.476 [157.0] *** 0.511 [169.6] *** 0.521 [150.9] *** Earnings to minimum wage ratio -0.003 [-47.3] *** Earns less than half minimum b 0.423 [172.0] *** 0.429 [108.1] *** Earns half to twice minimum 0.237 [191.0] *** 0.197 [95.1] *** Earns less than half minimum * manufacture -0.018 [-2.5] *** Earns less than half minimum * construction -0.093 [-10.8] *** Earns less than half minimum * commerce 0.044 [6.8] *** Earns less than half minimum * hotel/restaurant -0.135 [-10.8] *** Earns less than half minimum * transport 0.029 [1.8] *** Earns less than half minimum * domestic service -0.128 [-5.3] *** Earns half to twice minimum * manufacture 0.071 [20.7] *** Earns half to twice minimum * construction 0.219 [43.8] *** Earns half to twice minimum * commerce 0.030 [10.0] *** Earns half to twice minimum * hotel/restaurant -0.044 [-6.6] *** Earns half to twice minimum * transport 0.103 [18.4] *** Earns half to twice minimum * domestic service -0.055 [-9.6] *** Number of observations 1210315 1210315 1210315 1210315 Pseudo R2 0.281 0.282 0.316 0.318 Log likelihood -594328 -593103 -565136 -563491 Naïve probability 0.427 0.427 0.427 0.427 Estimated probability 0.376 0.376 0.378 0.379 Notes: Marginal effects and absolute value of z statistics (in brackets) are reported * significant at 10%; ** significant at 5%; *** significant at 1% a Omitted category Agriculture b Omitted category more than twice MW Population: currently employed individuals 15-65 yrs Source: Own estimations based on microdata from ENHA, 2006, Weighted Sample 92 Table A.4.3: Determinants of Training Provision by Firms Model Variables (I) (II) (III) (IV) (V) (VI) Firm manager's experience -0.023 -0.011 -0.012 -0.018 -0.027 [9.03]*** [11.33]*** [13.46]*** [14.21]*** [9.93]*** Workeforce education Primary complete 0.065 -0.249 [0.97] [2.80]*** High school incomplete 0.095 -0.194 0.123 [1.45] [2.02]** [2.31]** High school complete 0.193 -0.082 -0.076 [2.45]** [0.91] [0.92] Superior incomplete 0.351 0.022 [1.62] [0.11] New technology in production process 0.014 0.218 0.133 0.114 0.063 0.173 [0.90] [3.95]*** [6.26]*** [5.47]*** [2.41]** [3.06]*** Firm years in business 0.003 -0.002 0.003 0.003 0.002 -0.003 [6.86]*** [1.51] [6.79]*** [6.11]*** [4.44]*** [2.61]*** Foreing capital 0.241 0.221 0.293 0.427 0.271 0.057 [5.59]*** [1.56] [5.14]*** [8.09]*** [3.98]*** [0.38] Percentage of non-production employess 0.141 0.368 0.741 [2.70]*** [4.88]*** [4.68]*** Percentage of unionized workforce 0.003 -0.003 0.002 0.002 0.002 -0.003 [8.19]*** [3.66]*** [4.78]*** [5.07]*** [3.56]*** [3.55]*** Percentage of female in the labor force -0.149 0.923 -0.339 -0.226 -0.464 0.642 [4.29]*** [6.73]*** [7.05]*** [5.14]*** [8.05]*** [4.53]*** Large firm 0.132 -0.006 0.15 0.194 0.12 0.128 [5.26]*** [0.10] [5.12]*** [6.42]*** [3.16]*** [1.90]* Garments 0.198 -0.357 0.327 0.317 0.387 -0.287 [5.65]*** [6.86]*** [6.75]*** [6.71]*** [6.30]*** [6.13]*** Textiles 0.11 0.267 0.34 0.435 [3.89]*** [7.19]*** [8.96]*** [7.66]*** Chemicals & Pharmaceuticals -0.033 -0.098 -0.064 -0.014 -0.018 0.108 [1.46] [1.54] [2.24]** [0.51] [0.51] [1.64] Other Manufactures -0.121 -0.155 0.02 0.075 0.133 0.03 [4.73]*** [1.85]* [0.52] [1.79]* [2.36]** [0.31] Log total expenditure in R & D 0.13 0.117 [7.12]*** [6.57]*** The firm have web site 0.166 [3.17]*** Access to credit 0.073 0.1 0.065 [3.26]*** [4.66]*** [2.37]** Log last 3 fiscal year sales per worker 0.016 [1.28] Log last fiscal year sales per worker 0.004 0.017 [0.47] [1.82]* Observations 3195 644 2299 2449 1699 629 Pseudo R2 0.1121 0.3497 0.2217 0.2398 0.2758 0.4208 Log Likelihood -1535 -271 -1030 -1086 -738 -232 Naïve 0.232 0.349 0.263 0.270 0.287 0.334 Predicted P. 0.213 0.303 0.213 0.224 0.233 0.243 Absolute value of z statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 93 Box A.4.2: Classification Training Courses Overall training courses could be grouped in computer application usage; administration and secretarial; technical in agriculture; technical in manufacturing/construction (wood work, textiles, pottery, electricity); technical in health/nurse/psychology; food/restaurant industry; tourism and languages; arts and education; and cosmetology. Training courses were classified as transferable and non-transferable, based on the type of skills the course could provide to the participant. In general, transferable training courses are those that provide skills that could be utilized in various occupations, hence not specific to one type of job or activity. Transferable training courses include classes such as accounting, computer usage, or marketing. Non-transferable training courses are, on the contrary, those that provide very specific skills for particular jobs/activities, such as bar tender to work in the hotel/restaurant industry, or artificial insemination for animals. A third group, among those who reported having taken a training course, was excluded from this classification. This group of individuals reported having taken training courses for personal reasons or did not derive any benefits from it. In many cases these individuals reported having taken courses such as cooking classes, shiatsu, or yoga, which are clearly non-job related. Distribution of Demographic and Experience Characteristics Across Training Course Types. Non-transferable Transferable No training Total training courses training courses course Demographic characteristics Male population 46.3% 51.5% 55.6% 54.9% Average age 37.46 36.55 39.04 38.82 Average years of education 12.18 12.57 9.74 10.02 Race (white) 88.5% 91.6% 88.7% 88.8% Tenure and Technology usage Job tenure 9.69 10.42 8.89 9.02 Experience (first job 22 years or younger) 82.1% 85.0% 85.8% 85.6% Use of technology 67.1% 90.6% 38.4% 42.8% Source: Authors’ estimates based on ENHA, 2006, Weighted sample. 94 Figure A.4.2: Distribution of Employment Figure A.4.3: Distribution of Earnings to Characteristics across Training Courses Minimum Wage ratio across Training Courses 10 0 % 93% 91% 89% 100% 90% 90% 80% 80% 70 % 70% 69% 60% 60% 46% 83% 86% 50 % 50% 40% 28% 26% 25% 40% 30% 19% 20% 20% 30% 10 % 20% 25% 0% 10% 15% 10% No n-Trans ferab le Trans ferab le Training No Training Co urs e 0% 3% 3% 6% Training Co urs es Co urs es Non-T ransferable T ransferable T raining No T raining Course T raining Courses Courses Em plo ye d Info rm a l Wo rke r Unde re m plo ye d Wo rke r Ea rns le s s tha n ha lf o f the M W Ea rns be twe e n ha lf a nd twic e o f the M W Ea rns m o re tha n twic e o f the M W Source: Own estimations based on microdata from ENHA, Source: Authors‘ estimates based on ENHA, 2006, 2006, Weighted Sample weighted sample Table A.4.4: Characteristics of Employed and Unemployed Distribution across Employment Status Distribution across Individual Characteristics Employed Unemployed Employed Unemployed Total 89.0% 11.0% 100% 100% Gender Women 86.0% 14.0% 43.6% 57.6% Men 91.5% 8.5% 56.5% 42.4% Age 14- 18 64.1% 35.9% 3.2% 14.4% 19- 23 74.0% 26.0% 8.7% 24.7% 24-28 85.9% 14.1% 11.1% 14.7% 29-33 91.6% 8.4% 12.5% 9.4% 34-38 91.8% 8.2% 11.8% 8.5% 39-43 93.2% 6.8% 12.3% 7.3% 44-48 93.9% 6.1% 12.4% 6.5% 49-53 93.9% 6.1% 11.3% 6.0% 54-58 94.5% 5.6% 8.9% 4.2% 59-65 93.8% 6.2% 8.0% 4.3% Educational level Primary incomplete 91.0% 9.0% 8.7% 7.0% Primary complete 88.5% 11.5% 21.0% 22.2% High school incomplete 86.5% 13.5% 32.6% 41.3% High school complete 90.8% 9.3% 9.1% 7.5% Superior incomplete 90.2% 9.8% 21.7% 19.2% Superior complete 95.2% 4.8% 6.9% 2.8% Source: Own estimations based on microdata from ENHA, 2006 95 Table A.4.5: Marginal Effects of Employability Probit Models Dependent Variable: =1 if currently employed, =0 if currently unemployed Model [I] [II] [III] Men (=1 if men) 0.050 *** 0.051 *** 0.051 *** [101.9] [102.5] [102.8] Age (in years) 0.017 *** 0.017 *** 0.017 *** [157.3] [156.7] [156.8] Age2 0.000 *** 0.000 *** 0.000 *** [-118.9] [-118.3] [-118.3] Years of education 0.004 *** 0.004 *** 0.004 *** [57.5] [55.1] [55.0] Race (=1 white, 0=others) 0.012 *** 0.013 *** 0.013 *** [16.7] [16.9] [17.1] Took training course (=1 if took training) 0.026 *** [34.1] Took non-transferable training course (=1 if took training) 0.041 *** [37.7] Took transferable training course (=1 if took training) 0.014 *** [14.1] Job experience (=1 if first job at 22 or younger) 0.122 *** 0.121 *** 0.122 *** [154.6] [154.6] [155.0] Use of technology (=1 if used PC past 6 months) 0.024 *** 0.021 *** 0.022 *** [44.6] [37.9] [39.4] Urban areas (=1 if urban areas) -0.039 *** -0.039 *** -0.039 *** [-55.9] [-56.0] [-56.0] Montevideo 0.001 0.000 0.000 [1.3] [1.0] [0.9] Number of observations 1394888 1394888 1394888 Pseudo R2 0.127 0.128 0.128 Log likelihood -415367 -414766 -414530 Naïve probability 0.893 0.893 0.893 Estimated probability 0.916 0.917 0.917 Notes: Marginal effects and absolute value of z statistics (in brackets) are reported * significant at 10%; ** significant at 5%; *** significant at 1% a Omitted category agriculture b Omitted category „more than twice the MW‟ Population: currently employed individuals 15-65 yrs Source: Author‟s estimates based on ENHA, 2006, Weighted sample. 96 Table A.4.6: Mincer Wage Equations Dependent Variable: Logarithm of wage/earnings Model [I] [II] Men (=1 if men) 0.058 *** 0.056 *** [4.6] [4.5] Age (in years) 0.040 *** 0.039 *** [12.0] [12.0] Age2 0.000 *** 0.000 *** [-9.8] [-9.7] Years of education 0.041 *** 0.041 *** [23.1] [23.2] Race (=1 white, 0=others) 0.116 *** 0.115 *** [6.3] [6.3] Took training course (=1 if took training) 0.228 *** [11.2] Took non-transferable training course (=1 if took training) 0.169 *** [5.6] Took transferable training course (=1 if took training) 0.270 *** [10.7] Informality (=1 if worker informal) -0.370 *** -0.369 *** [-28.3] [-28.2] Job Tenure (in years) 0.016 *** 0.016 *** [18.8] [18.7] Urban areas (=1 if urban areas) -0.019 *** -0.019 *** [-1.4] [-1.4] Montevideo 0.223 *** 0.222 *** [16.9] [16.9] Constant 2.206 *** 2.211 *** [29.7] [29.9] Number of observations 28262 28262 Wald X2 4997 5029 Log likelihood -1791891 -1791595 Rho -0.327 -0.331 Sigma 0.735 0.736 Lambda -0.240 -0.244 Notes: Marginal effects and absolute value of z statistics (in brackets) are reported * significant at 10%; ** significant at 5%; *** significant at 1% a Omitted category Agriculture b Omitted category more than twice MW Population: currently employed individuals 15-65 yrs Source: Authors estimates based on ENHA, 2006, Weighted sample 97 References Acemoglu D. and J. Pischke (1998). ―Why Do Firms Train? Theory and Evidence‖ The Quarterly Journal of Economics, 113 (1): 79-119. Altonji J. and J. Spletzer (1991). ―Worker Characteristics, Job Characteristics, and the Receipt of On-the-Job Training.‖ Industrial and Labor Relations Review, 45 (1): 58-79. Amarante V. and A. Espino (2007). ―Informalidad y protección social en Uruguay. La situación de las mujeres.‖ Banco Mundial y el Instituto Nacional de la Mujer, Ministerio de Desarrollo Social de Uruguay. Aw, B.Y. and G. Batra (1998) ―Technological Capability and Firm Efficiency in Taiwan.‖ World Bank Economic Review (12): 59- 79. Baccaro, L. and M. Simoni (2004). “The Irish Social Partnership and the “Celtic Tiger� Phenomenon.� Decent Work Research Program. Discussion Paper 154/2004. International Labor Organization, Geneva. Barron J., M. Berger and D. Black (1999). ―Do Workers Pay for On-The-Job Training?‖ Journal of Human Resources, 34 (2): 235-252. Bernard, A. B. and J. B. Jensen (1999). ―Exporting and Productivity.‖ National Bureau of Economic Research Working Paper 7135. Bishop J. (1998). ―Occupation-Specific versus General Education and Training.‖ Annals of the American Academy of Political and Social Science, 559, The Changing Educational Quality of the Workforce: 24-38. Black S. and L. Lynch (1996). ―Human Capital Investments and Productivity.� American Economic Review, Vol. 86 (May): 263-268. Bosworth and Collins (2003). The Empirics of Growth: An Update. Washington, DC: Brookings Institute. Calderón, C. and L. Servén (2004). The Effects of Infrastructure Development on Growth and Income Distribution. Washington, DC: World Bank. Casacuberta, C. and N. Gandelman (2007). “Explaining Output Differentials by Export Orientation: Evidence from the Manufacturing Sector in Uruguay.� Background paper for the LAC Regional Study on Micro-determinants of Growth. Washington, D.C: World Bank. Cheikhrouhou, H., B. Gwinner, J. Pollner, E. Salinas, S. Sirtaine, D. Vittas. Forthcoming. Structured Finance in Latin America: Channeling Old-Age Savings to Underserved Sectors, Washington, D.C: World Bank. De Ferranti, D., G. E. Perry, I. S. Gill, J. L. Guasch, W. F. Maloney, C. Sanchez-Paramo, and N. Schady (2003). Closing the Gap in Education and Technology. Washington, DC: World Bank. De Ferranti, D., G.E. Perry, D. Lederman, W. Maloney (2002). From Natural Resources to the Knowledge Economy. Washington, DC: World Bank. De la Torre, A., J. C. Gozzi, and S. L. Schmukler (2006a). ―Innovative Experience in Access to Finance: Market Friendly Roles for the Visible Hand?� JEL, April 2006. 98 De la Torre, A., J. C. Gozzi, and S. Schmuckler (2006b). ―Financial Development in Latin America: Big Emerging Issues, Limited Policy Answers.‖ Mimeo. The World Bank. Washington, D.C. Dearden L., H. Reed and J. Van Reenen (2000). ―Who Gains when Workers Train? Training and Corporate Productivity in a Panel of British Industries.‖ Centre for Economic Policy Research. Discussion Paper 2486. Deutsch, R., A. Morrison, C. Piras, and H. Ñopo (2004). "Working within Confines: Occupational Segregation by Gender in Costa Rica, Ecuador, and Uruguay.‖ In Women at Work: Challenges for Latin America. Editor Claudia Piras. Washington, DC: Inter-American Development Bank. DICyT (2006). La Innovación en la Industria Uruguaya (2001-2003): II Encuesta de Actividades de Innovación en la Industria. Montevideo. Dollar, D. and E.N. Wolff (1998). ―Convergence of Industry Labor Productivity among Advanced Economies, 1963-1982.‖ The Review of Economics and Statistics. Vol 70:4, 549-558. Drewry (2006). Drewry Annual Container Market Review and Forecast 2006/2007. Escribano, A. and J. L. Guasch (2005). ―Assessing the Impact of the Investment Climate on Productivity using Firm-Level Data: Methodology and the Cases of Guatemala, Honduras, and Nicaragua.‖ Policy Research Working Paper 3621. World Bank: Washington DC. Gasparini, L. and L. Tornarolli (2007). ―Labor Informality in Latin America and the Caribbean: Patterns and Trends from Household Survey Microdata.‖ Centro de Estudios Distributivos, Laborales y Sociales, Maestría en Economía Universidad Nacional de La Plata. Documento de Trabajo Nro. 0. Hausmann, R., U. Panizza and R. Rigobon (2004a) ―The Long-Run Volatility Puzzle of the Real Exchange Rate.‖ National Bureau of Economic Research Working Paper 10751. Hausmann, R., U. Panizza and R. Rigobon (2004b) ―Financial Imperfections and the Long-Run Volatility of the Real Exchange Rate.‖ Mimeo. Harvard University. Hausmann, R., A. Rodriquez-Clare, and D. Rodrik (2005). ―Towards a Strategy for Economic Growth in Uruguay.‖ Economic and Social Studies Series, RE1-05-003. Inter-American Development Bank: Washington DC. Heckman J. and C. Pages (2003). "Law and Employment: Lessons from Latin America and the Caribbean.‖ National Bureau of Economic Research Working Paper 10129. Holtmann A. and T. Idson (1991). ―Employer Size and On-the-Job Training Decisions.‖ Southern Economic Journal, 58 (2): 339-355. Hujer R., K. Maurer and M. Wellner (1997). ―Estimating the Effect of Training on Unemployment Duration in West Germany: A Discrete Hazard-Rate Model with Instrumental Variables.‖ Department of Economics Johann Wolfgang Goethe-University Working Paper. International Monetary Fund and the World Bank (2006). Uruguay: Financial Sector Assessment. Washington, D.C. Katz E. and A. Ziderman (1990). ―Investment in General Training: The Role of Information and Labour Mobility.‖ Economic Journal, 100 (403): 1147-1158. Kaufmann, D., A. Kraay, and M. Mastruzzi (2006). ―Governance Matters V: Governance Indicators for 1966- 2005.‖ Washington, DC: World Bank. 99 Laborda, L. (2007) ―Análisis de la Eficiencia Técnica en Uruguay: Una Aproximación no Paramétrica a Nivel de Empresa.‖ Mimeo. Washington, D.C: World Bank. Loayza, N., P. Fajnzylber, and C. Calderón (2005). Economic Growth in Latin America: Stylized Facts, Explanations, and Forecasts. Washington DC: World Bank. Loayza, N., Oviedo, A. M. and L. Servén (2004). “Regulation and Macroeconomic Performance.� Policy Research Working Paper No. 3469. Washington DC: World Bank. Loewenstein M. and J. Spletzer (1999). ―General and Specific Training: Evidence and Implications.‖ Journal of Human Resources, 34 (4): 710-733. Maloney, W. F., J. Nuñez, W. Cunningham, N.F.C. Montenegro, E. Murrugarra, M. Santamaria, and C. Sepulveda (2004). ―Minimum Wages in Latin America.‖ Mimeo. Washington DC: World Bank. Martinez-Vasquez J. (2006). ―An Assessment of Pakistan‘s Tax System.‖ Mimeo. Georgia Technology University Monteserín G., D. Chiappori, (2004). ―Financing Decisions: Study for Firms Established in Uruguay.� XIX Jornadas Anuales de Economía – Banco Central de Uruguay. Mroz T. (1987). ―The Sensitivity of an Empirical Model of Married Women's Hours of Work to Economic and Statistical Assumptions.‖ Econometrica, 55 (4): 765-799. Neumark, D., M. Schweitzer; and W. Wascher (2004). “Minimum Wage Effects throughout the Wage Distribution.� The Journal of Human Resources, Vol. 39, No. 2. Spring, pp. 425-450. O‟Connell, P. (1999). “Economic Growth and the Labor Market in Ireland.� Employment and Training Papers No. 44. International Labor Organization, Geneva. Packard, T. (2001). ―Is There a Positive Incentive Effect from Privatizing Social Security? Evidence from Pension Reforms in Latin America.‖ Policy Research Working Paper. Washington, DC: World Bank. Packard, T. (2004) ―Obstacles and Opportunities for Economic Growth in Uruguay's Social Sectors.‖ Background Paper for Uruguay Sources of Growth, Regional Office for Latin America and the Caribbean. Washington, DC: World Bank. Pagés C. (2005). ―Diagnostico del Mercado de Trabajo en Uruguay.‖ Serie de Estudios Económicos y Sociales, RE1-05-013. Washington, DC: Inter-American Development Bank. Palacios, R. and M. Pallarés-Miralles (2000). "International Patterns of Pension Provision." Social Protection Discussion Paper No. 0009, Washington, DC: World Bank. Perry, G. E., W. F. Maloney, O. S. Arias, P. Fajnzylber, A. D. Mason, J. Saavedra-Chanduvi (2007). Informality: Exit and Exclusion. Washington, DC: World Bank. Piras, Claudia (2004). Women at Work: Challenges for Latin America. Washington, DC: Inter-American Development Bank. Prescott, E.C. (1998). ―Needed: A Theory of Total Factor Productivity.‖ International Economic Review, 39, 525- 552. 100 Rada, K. and U. Blotte (2007). ―Improving Business Registration Procedures at the Sub-National Level: The Case of Lima, Peru.‖ Lessons in Advisory Services, International Finance Corporation. Washington, DC: World Bank. Red de Indicadores de Ciencia y Tecnología (RICYT). http://www.ricyt.edu.ar. Ricardo P. (2007). ―Impacto de la Reforma Tributaria sobre los Mercados de Capitales y las Finanzas de las Empresas en Uruguay.‖ Mimeo. Montevideo. Rodrik, D. (2004). ―Industrial Policy for the Twenty-First Century.‖ Cambridge, MA: Harvard University. Rofman R. and L. Lucchetti (2006) ―Pension Systems in Latin America: Concepts and Measurements of Coverage.‖ Working paper, World Bank, Washington DC. Ruane, F. and J. Sutherland (2005). ―Export Performance and Destination Characteristics of Irish Manufacturing Industry.‖ Review of World Economics, Vol. 141 (3). Scott, D. H. (Forthcoming). ―Strengthening the Governance and Performance of State-Owned Financial Institutions.‖ Policy Note. Washington DC: World Bank. Tan H. (2000). ― Technology and Skills Demands: Panel Evidence from Malaysian Manufacturing.‖ Working Paper, The World Bank Institute, Washington DC: World Bank. Tan H. and G. Lopez-Acevedo (2003). ―Mexico In-firm Training for the Knowledge Economy.‖ Policy Research Working Paper 2957, Washington, DC: World Bank. Teague, P. (1995). Pay Determination in the Republic of Ireland: Towards Social Corporatism?, British Journal of Industrial Relations, Vol.33, No.2, June. Trajtenberg, M. (2005). ―Innovation Policy for Development: an Overview.‖ Paper presented at the LAEBA 2005 Meeting in Buenos Aires. Universidad Politécnica de Valencia (2003). ―Evaluación de los Principales Puertos de América del Sur.‖ Corporación Andina de Fomento. Veum J. (1995). ―Sources of Training and their Impact on Wages.‖ Industrial and Labor Relations Review, 48 (4): 812-826. Veum J. (1999). ―Training, Wages, and the Human Capital Model.‖ Southern Economic Journal, 65 (3): 526-538. Winkler H. (2005) ―Monitoring the Socio-Economic Conditions in Uruguay‖ Working Paper, CEDLAS, Universidad Nacional de La Plata. World Bank (Various years). Doing Business. http://www.doingbusiness.org/ World Bank (2007a). Economic Growth in Latin American and the Caribbean: A Microeconomic Perspective. Latin America and the Caribbean Region. Washington, D.C. World Bank. World Bank (2007b). ―Project Appraisal Document for Promoting Innovation to Enhance Competitiveness.‖ World Bank Report No. 36600-UY. Washington, D.C. World Bank (2007c). Development Policy Review on Infrastructure. Washington, D.C. 101 World Bank (2007d). ―Job Creation in Latin America and the Caribbean: Recent Trends and the Policy Challenges.‖ Mimeo. Washington, DC. World Bank (2006a). ―Doing Business 2007: How to Reform.‖ Washington, DC. World Bank (2006b). ―Mexico‘s Competitiveness: Reaching Its Potential.‖ World Bank Report No. 35388-MX. Washington, DC. World Bank (2005a). ―Uruguay - Social Program Development Policy Loan Project.‖ World Bank Report No. 32289-UY. Washington, DC. World Bank (2005b). ―Uruguay Sources of Growth: Policies for the Development of Human Capital, Integration, Competition and Innovation.‖ World Bank Report No. 31737-UY. Washington, DC. World Bank (2005c). “Colombia Labor Market Adjustment, Reform, and Productivity: What are the Factors that Matter?� World Bank Report No 32068-CO. Washington, DC. World Bank (2004a). World Development Report 2005: A Better Investment Climate for Everyone. New York: Oxford University Press. World Bank (2004b). Uruguay Poverty Update 2003. Washington, DC: World Bank. World Bank (2001). Latin America: From Natural Resources to the Knowledge Economy. Office of the Chief Economist, Latin American Region. Washington, DC: World Bank. World Bank and World Health Organization (2004). World Report on Road Traffic Injury Prevention. World Economic Forum (2005). The Global Competitiveness Report 2005–2006. Geneva: World Economic Forum. Zweimüller J. and Winter-Ebmer (2000). ―Firm-specific Training: Consequences for Job Mobility.‖ Institute for the Study of Labor (IZA), Discussion Paper 138. 102