69767 THE WORLD BANK GROUP Report No. Côte d’Ivoire Investment Climate Survey Report Africa Finance and Private Sector Group (AFTFP) This Investment Climate Survey Report for Côte d’Ivoire evaluates the country’s business environment by (i) analyzing barriers to private sector investment and growth and how they vary among different types of firms, (ii) benchmarking the Ivorian investment climate and firm performance to that of other countries and (iii) providing recommendations to promote and strengthen the private sector. The ICSR is supported by the statistical analysis of a survey of firms based in Côte d’Ivoire’s major urban centers, as well as data from other sources. Vice President: Obiageli Ezekwesili Country Director Madani Tall Sector Director Marilou Uy Sector Manager Peter Mousley Task Team Leader Alvaro González TABLE OF CONTENTS Chapter 1 – Macroeconomic Context ............................................................................................................................... 5 Introduction ...................................................................................................................................................................... 5 Economic growth and poverty reduction ............................................................................................................. 6 Macroeconomic performance in Côte d’Ivoire and abroad ........................................................................... 6 Inflation .............................................................................................................................................................................. 7 Exchange rate ................................................................................................................................................................... 7 Manufacturing firm performance and exports ................................................................................................... 8 Macroeconomic performance and the investment climate ........................................................................... 9 Political instability....................................................................................................................................................... 10 Summary of findings .................................................................................................................................................. 11 Chapter 2 – Manufacturing ................................................................................................................................................ 12 Data: Enterprise Surveys .......................................................................................................................................... 12 Labor productivity ...................................................................................................................................................... 13 Capital productivity .................................................................................................................................................... 14 Export performance.................................................................................................................................................... 16 Firm-level performance ............................................................................................................................................ 17 Productivity.................................................................................................................................................................... 19 Summary of findings .................................................................................................................................................. 25 Chapter 3 – Access to Finance .......................................................................................................................................... 26 Financial system overview ...................................................................................................................................... 26 Perceptions of access to finance ............................................................................................................................ 31 Structure of firm financing....................................................................................................................................... 33 Applying for credit ...................................................................................................................................................... 35 Access to bank financing ........................................................................................................................................... 37 Characteristics of lines of credit, loans and overdrafts ................................................................................ 39 Access to trade credit ................................................................................................................................................. 41 Summary of findings .................................................................................................................................................. 42 Chapter 4 – Informal Sector .............................................................................................................................................. 43 Characteristics of formal and informal firms ................................................................................................... 43 Productivity comparisons ........................................................................................................................................ 44 Formality vs. informality .......................................................................................................................................... 47 Formality vs. informality: a Cross-Country Comparison............................................................................. 48 Conclusion ..........................................................................................................Error! Bookmark not defined. Chapter 5 – Labor Market .................................................................................................................................................. 50 Characteristics of the formal labor market ....................................................................................................... 50 Education, training, health and wages of formal sector employees ....................................................... 53 Chapter 6 –Investment Climate ....................................................................................................................................... 61 Political instability....................................................................................................................................................... 63 Regulation and corruption ....................................................................................................................................... 64 Electricity ........................................................................................................................................................................ 65 References ................................................................................................................................................................................ 67 TABLE OF FIGURES Figure 1-1: International comparison of Manager perceptions of political instability .................................................... 9 Figure 2-1: International comparison of value added in manufacturing .............................................................................12 Figure 2-2: International comparison of value added per worker, by firm size ...............................................................14 Figure 2-3: International comparison of median returns to capital, by firm size ...........................................................16 Figure 2-4: International comparison of export presence, by firm size ...............................................................................17 Figure 3-1: international comparison of bank deposits over GDP, 1990-2007 ................................................................27 Figure 3-2: International comparison of ratio of private credit by deposit-taking banks to GDP, 1990 - 2007..........................................................................................................................................................................................................27 Figure 3-3: International comparison of ratio of private credit to deposits, 1990 - 2007 ...........................................27 Figure 3-4: International comparison of bank concentration, 1993-2007 .........................................................................28 Figure 3-5: International comparison of perceptions about financing problems ............................................................31 Figure 3-6: International comparison of bank financing as share of total investments................................................35 Figure 3-7: Reasons why firms needing credit do not apply .....................................................................................................36 Figure 3-9: Reasons given for rejecting of application for credit ............................................................................................37 Figure 4-1: Labor productivity in formal and informal firms ...................................................................................................45 Figure 4-2: Kernel density estimate of log value added per worker ......................................................................................45 Figure 4-3: Kernel density estimate of log sales per worker .....................................................................................................45 Figure 4-4: Access to infrastructure services, by legal status ...................................................................................................47 Figure 5-1: International comparison of rigidity of employment ...........................................................................................52 Figure 5-2: International comparison of severance pay ..............................................................................................................52 Figure 5-3: Average education level for typical production worker ......................................................................................54 Figure 5-4: International comparison of training offerings by firms .....................................................................................55 Figure 5-5: International comparison of manufacturing workers’ monthly cash earnings ........................................58 Figure 6-1: Firm perceptions of most severe constraints ...........................................................................................................62 Figure 6-2: International comparison of likelihood of bribe requests ..................................................................................63 Figure 6-3: Political stability percentile rank ...................................................................................................................................64 Figure 6-4: International comparison of losses due to power outages ................................................................................65 Figure 6-5: International comparison of generator usage ..........................................................................................................66 LIST OF TABLES Table 1-1: Macroeconomic indicators ................................................................................................................................................... 7 Table 2-1: Characteristics of sample ....................................................................................................................................................13 Table 2-2: Comparator countries ...........................................................................................................................................................13 Table 2-3: International comparison of median capital intensity, by industry .................................................................15 Table 2-4: International comparison of median capital productivity ...................................................................................15 Table 2-5: International comparison of share of manufactures in merchandise exports ............................................16 Table 2-6: Median productivity, by firm characteristics .............................................................................................................19 Table 2-7: TFP estimation, frontier model. ........................................................................................................................................21 Table 2-8: TFP model with input interactions .................................................................................................................................22 Table 2-9: Evidence of economies of scale in Agro-Processing. ...............................................................................................23 Table 3-1: Doing Business rankings ......................................................................................................................................................30 Table 3-2: Doing Business credit indices ............................................................................................................................................30 Table 3-3: Doing Business indices on property, investors, contracts and business closings, .....................................31 Table 3-4: Perceptions of access to finance as major/severe obstacle .................................................................................32 Table 3-5: Probit estimate of firm perceptions of finance as a constraint ..........................................................................33 Table 3-6: Sources of financing for fixed assets or working capital .......................................................................................33 Table 3-7: Bank financing of fixed assets, by firm type ................................................................................................................34 Table 3-8: Percent of firms applying for credit ................................................................................................................................36 Table 3-9: Characteristics of firms rejected for credit ..................................................................................................................37 Table 3-10: Share of firms with access to lines of credit/loans and overdraft facilities ...............................................38 Table 3-11: Probit estimate of determinants of access to bank financing ...........................................................................39 Table 3-12: Share of credit market, by type of Institution..........................................................................................................39 Table 3-13: Interest rates for lines of credit/loans and overdrafts ........................................................................................40 Table 3-14: Average interest rate, by type of firm .........................................................................................................................40 Table 3-15: Access to trade credit, by type of firm ........................................................................................................................41 Table 4-1: Enterprise Survey criteria for firm formality .............................................................................................................43 Table 4-2: Characteristics of microenterprises ...............................................................................................................................44 Table 4-3: Correlates of firm productivity .........................................................................................................................................46 Table 4-4: Costs of formality and informality ...................................................................................................................................47 Table 4-5: Correlates of formality ..........................................................................................................................................................48 Table 5-1: Trade union presence, by type of firm...........................................................................................................................51 Table 5-2: Average educational achievement in manufacturing firms .................................................................................55 Table 5-3: Formal training in the manufacturing sector .............................................................................................................56 Table 5-4: Determinants of employee training programs ..........................................................................................................56 Table 5-5: International comparison of health-related absenteeism and costs ...............................................................57 Table 5-6: Monthly cash earnings for production workers ........................................................................................................58 Table 5-7: Determinants of wages: Firm-level estimates ............................................................................................................59 Table 6-1: International comparison of political stability ranking. ........................................................................................64 ACKNOWLEDGEMENTS This report was prepared by a team consisting of Alvaro González (AFTFW, task team leader), Riham Shendy (AFTFW, Young Professional), Philippe Alby (Consultant), Nicoletta Berardi (Consultant), Carlos Espina (Consultant), Vijaya Ramachandran (Consultant) and was edited by Robin Kraft (Consultant). Lorenzo Bertolini (AFTFW, Senior Private Sector Development Specialist) was a key member of the team and provided in-country context to the findings of this report. John L. Nasir (Lead Private Sector Development Specialist, MNSFP) and Jean Michel Marchat, Senior Private Sector Development Specialist, MNSFP) served as peer reviewers for this study and provided a number of suggestions and useful comments. Helpful comments were also received from Phillip English, Katrina Sharkey, Joelle Businger, Peter Mousley and participants at the review meeting and several members of the Côte d’Ivoire country team. EXECUTIVE SUMMARY In the 1960s and 70s, Côte d’Ivoire emerged as one of Africa’s few stable and economically successful countries. However, a deep recession has gripped Côte d’Ivoire through most of the 1980s, induced by an overvalued currency, a fall in global commodity prices and the high cost of servicing debts incurred in the 1970s. The country then experienced a collapse of political order starting in 1999 that degenerated into civil war in September 2002. The resulting economic decline has aggravated the already very high unemployment, especially among youth, and increased poverty levels from 38.2 percent of the population in 2002 to 43.2 percent in 2006. Since 2006, there has been modest growth, in part due to strong performance in the cocoa and telecommunications sectors, as well as increased oil production and prices. However, the global economic crisis has affected Côte d’Ivoire’s economic prospects through reduced demand for some of the country’s commodity exports. In addition, the economic forecast for Côte d’Ivoire remains heavily dependent on the domestic political climate. Finally, while membership in the Franc Zone and the recent fall in international oil and food prices have helped to contain inflation, they have not eliminated inflationary pressures altogether. Optimism regarding Côte d’Ivoire’s macroeconomic performance is justifiable, but conditioned on restored political stability. Cocoa exports remain the engine of the economy, and infrastructure and the industrial sector are relatively well developed. Membership in the CFA zone has kept monetary policy stable as well. However, post-conflict economic success will rely heavily on economic diversification that enables the country to better respond to external economic forces. Given an increasingly stable macroeconomic and political environment, the World Bank viewed it necessary to begin to understand developments in the microeconomic environment and for that reason launched this report on the investment climate. The Report points out two important findings; (a) rates of investment are not likely to increase substantially as long as there is any remaining uncertainty with respect to the political situation; and (b) the large informal sector, commonly one potential pole of economic dynamism, is in survivalist mode, possibly also in abeyance while the political uncertainty is minimized. This Investment Climate Survey Report (ICSR) indentifies key areas where investment climate reforms may be warranted. This ICSR are largely based the analysis of Enterprise Survey data; which are surveys of mostly manufacturing, formal sector enterprises in the major industrial centers of an economy. Manufacturing is chosen as the lens through which the investment climate is viewed because the complex nature of manufacturing activities can best expose how well a wide set of the economy’s markets function—particularly labor and capital markets. How labor is hired, employed, trained, and at times let go is a function of the labor laws and regulation, the quality of the educational or other training institutions, and the supply (demographics) and demand (employment growth) for labor, among other variables. Similar factors hold for what affects the function of capital (or financial) markets; laws and regulations, the quality of the institutions that apply these rules and the sophistication and efficiency of exchange mechanisms. An assessment of the performance of manufacturing firms, by estimating firm-level productivity, is also an important component of this ICSR. Economists know little about what affects firm-level productivity directly, but they do know that in business environments where firms are allowed some flexibility with the use of labor and capital, these environments are more likely to result in more productive firms. In 1 addition, in economies with investment climates where failure of experimentation is not as costly, firms will be more productive and higher firm-level productivity will result in economies that grow faster. This Investment Climate Survey Report for Côte d’Ivoire is cognizant of the recent economic crisis and political turmoil that has enveloped Côte d’Ivoire. With that in mind, we begin by taking stock of Côte d’Ivoire’s achievements with respect to growth and poverty reduction, comparing its growth to relevant comparator countries. Then we use data from the World Bank’s Enterprise Surveys to examine current perceptions of the economic environment in Cote d’Ivoire as compared to nearby countries, and close with a review of the political instability that continues to plague Côte d’Ivoire. Since this ICSR is largely based on the analysis of the manufacturing sector, it is important to note that Côte d’Ivoire is one of the very few countries in Sub-Saharan Africa with a ratio of manufacturing to GDP higher than 15 percent. In fact, the share of manufacturing in the Ivorian economy is 19 percent, similar to high growth economies such as Morocco and Vietnam. Few countries in Sub-Saharan Africa have managed to enter international markets for manufactured goods, and Côte d’Ivoire’s strong export performance is at least in part due to its large manufacturing sector. While the manufacturing sector is comparatively well-developed, the labor productivity is relatively low in Côte d’Ivoire. The median firm produces approximately US$1677 of value added per worker, which is lower than the corresponding figure for all the comparator countries. Firms in Kenya, Cameroon and South Africa produce between five and six times as much per worker. As expected, labor productivity is higher in larger Ivorian firms. However we note that compared to other low-income countries in Sub-Saharan Africa, Côte d‘Ivoire’s gap in labor productivity between large and small enterprises is by far the largest, at over $38,000. The low capital intensity of manufacturing firms in Côte d’Ivoire is also problematic. Given the strong bias in the Ivorian economy towards agriculture and commodities, firms tend to have less machinery and equipment per worker than firms in the comparator countries. The median Ivorian firm has about US$340 of machinery and equipment per worker, while those in South Africa and Kenya have about US$9000 and US$12,000, respectively, of machinery and equipment per worker. However, the returns to capital are high suggesting that investments in capital are likely to be profitable to most enterprises. Given low labor productivity and low capital intensity, Ivorian firms compare unfavorably with firms in comparator countries with respect to international trade. Less than 10 percent of manufacturing firms in Côte d’Ivoire export goods, as compared to 20 percent of firms in all nearly all comparator countries. Despite the anemic participation of manufacturing firms in export markets, Côte d’Ivoire is a major player in international commodities markets, which suggests that the exports are highly concentrated. Comparing export participation by firm size reveals that nearly 55 percent of large firms are exporters as opposed to only 6 percent of small firms. Large agro- processing firms dominate cocoa and cotton exports. But few other firms are able to export, as compared to their peers in comparator countries. The relatively low capital intensity of manufacturing firms is not a result of poor functioning financial markets because the financial sector of Côte d’Ivoire is the largest and most 2 diversified in the West African Economic and Monetary Union (WAEMU). The banking sector is the largest component of the financial sector in Côte d’Ivoire, and included 19 commercial banks at the end of 2008. The microfinance sector accounts for a small fraction of financial assets, but contributes significantly to access to financial services through its large client base with approximately 958,000 accounts. The insurance sector is comprised of 33 companies, while the retirement system is dominated by two public institutions. There are also three nonbank financial institutions, and one financial institution that enables microfinance institutions to secure credit lines. While financial intermediation is well developed, Ivorian firms rarely rely on financing from the banking system. Internal funds and retained earnings represent by far the main source of financing for firms in Côte d’Ivoire, both for fixed assets and for working capital. However, excessive reliance on internal funds is a sign of potentially inefficient financial intermediation. Funds borrowed from banks on average represent only 3.5 percent of working capital and 1.5 percent of fixed assets. Purchase credits from suppliers and advances from customers are also limited, representing on average only 3.5 percent of working capital and 3.1 percent of fixed assets. This level of supplier and customer credit is much larger than that of Cameroon and Mali, and similar to that of South Africa. The limited role played by the Ivorian banking system may be the result of two phenomena. First, the rejection rate of applications is higher than the rate of accepted applications, suggesting credit rationing by banks. Second, many firms that need financing don’t apply, resulting in self- rationing. Indeed, the large majority of enterprises voluntary exclude themselves from the formal financial market, most often because they believe that application procedures are too complex or that collateral requirements are too high. Microeconomic indications from the financial markets are also consistent with capital waiting in the sidelines, both demand and supply, possibly waiting for clearer signals with respect to political stability. Macro trends hint at the same story. Gross fixed capital formation in the private sector rose by 12.7 percent in 2009, owing to investments in mining and increased imports of capital goods. In the public sector, the slowing of major infrastructure projects resulted in only a slight rise in investment (1 percent in 2009). Overall, the investment rate will not exceed 10.9 percent of GDP, a very low level compared with the other WAEMU countries. It should increase to 12.8 percent of GDP in 2010. The informal sector in Cote d’Ivoire looks quite different than the formal sector. The data suggest that formal firms tend to be larger, with 63 percent of the formal enterprises employing three to five permanent full time workers as compared to only 38 percent of informal firms. The educational attainment of firms’ top manager is also strikingly different for formal firms, where for 50 percent of firms, the top manager has a secondary education or better. The corresponding level of educational attainment in informal firms is 38 percent. Formal firms are also more likely to have been started to take advantage of a business opportunity (77 percent vs. 62 percent) rather than simply as a response to poor employment prospects. Finally, 81 percent of formal businesses are located in permanent or non-moveable premises, whereas only 67 percent of informal firms are located in these kinds of offices. While the data support a positive relationship between productivity and formality, this may be caused by other characteristics correlated with legal status. Econometric analysis suggests that 3 formality is significantly correlated with higher levels of firm productivity, measured as value added per worker. This positive relationship between formality and firm performance is robust to controls for access to services like electricity, telephone, sewage and water. It is also robust to using sales per worker as an alternative measure of productivity (results not reported for space reasons). Finally, we find that manager university education is significantly and positively correlated with productivity. The informal sector in Côte d’Ivoire is largely comprised of “survivalist� firms as opposed to enterprises that could take advantage of the expanded opportunities that could be expected with formalization. Less productive firms tend to be informal, suggesting that owners of such firms could be better off entering the labor market as wage labor. One way to help informal firms in Côte d’Ivoire would be to invest in education or vocational training, as we find that higher levels of productivity are associated with higher educational attainment of top firm management. The investment climate reform agenda is fairly large in Cote d’Ivoire. According to the World Bank’s Doing Business report, Côte d’Ivoire’s ranking improved in 2009, gaining five places over 2008. The country remains in the 20 bottom countries, however, ranking 163rd out of 183 countries. Indicators on the business environment put Côte d’Ivoire amongst the least competitive countries, far behind the African and West African averages. Private sector development is subject to serious constraints: cumbersome administrative procedures; the lack of a framework law on small and medium-sized enterprises (SMEs); the fact that the supply of education and training does not match the needs of business and the resurgence of corruption. About 68 percent of the surveyed firms consider corruption a "major" or "very severe" obstacle. This translates into a non- negligible cost of nearly 12 percent of firms’ total annual sales. Corruption and lack of transparency have been a problem for years in the cocoa and coffee regulatory bodies, government procurement, and the oil/energy sector. However, the severe economic downturn since the conflict erupted in 2002 has led to higher levels of corruption at all levels of public administration. Transparency International’s Corruption Perception Index (CPI), which measures perceptions of corruption by businesspeople and country analysts, showed a decline in the country’s ranking from 75th of 99 countries in 1999 to 150th of 180 countries in 2007. Given this fairly large private sector development agenda, this document serves as a basis to reinitiate the policy dialogue. Many of the findings are indicative, not conclusive, and therefore require further investigation. The scope of the analysis was limited, not comprehensive, with respect to geographic coverage and sectors unaddressed. For this reason, there is need to build on this Report to incorporate an analysis of the Central Northwest and other important economic sectors. Finally, and most importantly, there is need to incorporate the views, experiences of both the public and private sector; to understand the institutional capacity of each and the goals and perspective that will prove essential in any comprehensive, private sector development strategy. 4 Chapter 1 – MACROECONOMIC CONTEXT A favorable macroeconomic environment is essential for private sector-led growth and development. This chapter looks at Côte d’Ivoire’s (i) economic growth and poverty reduction trends and conditions; (ii) macroeconomic performance as compared to other countries; (iii) trends in inflation, interest and exchange rates; and (iv) success in exporting. We also analyze managers’ perceptions of the macroeconomic environment, which are good indicators of expectations and future investment. Finally, we pay special attention to political instability in Côte d’Ivoire since it is a clear obstacle to future growth of investment and the private sector. INTRODUCTION In the 1960s and 1970s, Côte d’Ivoire emerged as one of Africa’s few stable and economically successful countries. However, a deep recession gripped Côte d’Ivoire through most of the 1980s, induced by an overvalued currency, a fall in global commodity prices and the high cost of servicing debts incurred in the 1970s. The country then experienced a collapse of political order starting in 1999 that degenerated into civil war in September 2002. The resulting economic decline has aggravated the already very high unemployment, especially among youth, and increased poverty levels from 38.2 percent of the population in 2002 to 43.2 percent in 2006. Real GDP per capita has declined with an annual rate of 2.1 percent between 2002 and 2006. There has been modest growth since then, thanks to strong performance in the cocoa and telecommunications sectors, as well as increased oil production and prices. However, the global economic crisis has affected Côte d’Ivoire’s economic prospects through reduced demand for some of the country’s commodity exports. In addition, the economic forecast for Côte d’Ivoire remains heavily dependent on the domestic political climate. Finally, while membership in the Franc Zone and the recent fall in international oil and food prices have helped to contain inflation, they have not eliminated inflationary pressures altogether. This Investment Climate Survey Report explores the consequences of the economic crisis and political turmoil that has enveloped Côte d’Ivoire. Most of the analysis in this report is based on firm-level performance in the manufacturing sector, but we begin by looking at the macroeconomic fundamentals that affect manufacturing firms. While microeconomic performance is the basis of our analysis, macroeconomic fundamentals are crucial to establishing the incentives that will stimulate economic growth through improved productivity and increased employment. Broad-based economic growth will only take place when firms improve worker productivity by investing in human capital, physical capital and technological capacity defined as investment in knowledge, equipment and organizational structures. But firms will only invest if the macroeconomic fundamentals are in place. We start by taking stock of Côte d’Ivoire’s achievements with respect to growth and poverty reduction, comparing its growth to relevant comparator countries. Then we use data from the World Bank’s Enterprise Surveys to examine current perceptions of the economic environment, and close with a review of the political instability that continues to plague Côte d’Ivoire. 5 ECONOMIC GROWTH AND POVERTY REDUCTION Economically speaking, the last decade has been one of the worst in Côte d’Ivoire’s history. Its GDP was stagnant until 2003, after which it grew modestly at rates ranging from 1.8 percent in 2004 to 2.2 percent in 2008. This has largely been due to increases in international commodity prices (cocoa and coffee), strong performance in telecommunications, and increased oil production garnering high prices internationally. Côte d’Ivoire’s real GDP per capita in 2003 was about the same as it was in the early 1960s, and its capital stock was only somewhat higher than in the immediate aftermath of independence in 1960. While total factor productivity (TFP) estimates indicate that TFP per capita did grow after independence, it reached a plateau in 1976-1978 and subsequently shrank, as did per-capita output due to a secular deterioration in terms of trade. The outcome has been an increase in the rate of poverty from 32.3 percent in 1993 (when President Houphouët-Boigny died) to an estimated 42-44 percent in 2003 in the immediate aftermath of civil war. Although per capita income (purchasing power parity) of US$1,648, is somewhat high for the region, poverty continues to be a major problem. According to the most recent UN data, 48.8 percent of the population lives on less than US$2 per day, while another 14.8 percent live on less than US$1 per day. Côte d’Ivoire’s Human Development Index was 166th out of 177 countries. MACROECONOMIC PERFORMANCE IN CÔTE D’IVOIRE AND ABROAD Côte d’Ivoire’s poor economic performance in recent years is particularly notable compared to its peers ( 6 Table 1-1). Côte d’Ivoire is the only country of this group to have experienced a recession between 2000 and 2007. Indeed, several other countries performed quite well, and all of the comparator countries had higher figures for gross capital formation as a percentage of GDP. Net foreign direct investment (FDI) as a percentage of GDP is similar to comparators, though it has been volatile. FDI rose after the currency devaluation in 1994 from US$78 million to US$450 million in 1997, then dropped to US$215 million in 2002 according to the UN Conference on Trade and Development (UNCTAD). The bulk of foreign investment came from France, and was directed towards oil exploration, electricity, telecommunications, commodity export processing and transportation. FDI inflows recovered substantially in 2004 and 2005, reflecting increased investment in the oil sector, but fell to US$253m in 2006 with the ongoing political crisis and economic stagnation. 7 TABLE 1-1: MACROECONOMIC INDICATORS (averages, 2000-7) GDP growth Population growth Inflation, GDP deflator Gross capital formation Net FDI (% (annual %) (annual %) (annual %) (% of GDP) GDP) Burkina Faso 5.2 3.1 2.0 16.8 1.6 Cameroon 3.7 2.2 2.3 18.3 1.7 Côte d'Ivoire -0.2 2.2 2.8 10.1 2.0 Ghana 5.2 2.3 21.2 27.0 2.8 Kenya 4.0 2.6 4.9 17.5 0.6 Madagascar 3.6 2.8 10.8 20.5 3.3 Mali 5.4 3.0 4.1 23.5 3.9 Senegal 4.2 2.6 2.6 23.3 1.1 South Africa 4.2 1.4 7.4 17.7 1.6 Togo 2.0 2.7 0.7 18.7 3.3 Source: World Bank. Note: Data are averages for available data between 2000 and 2007. INFLATION As stated earlier, membership in the Franc Zone has helped to contain inflationary pressures in line with the priority of the Banque Centrale des Etats de l’Afrique de l’Ouest (BCEAO) to maintain the CFA franc’s peg to the euro. Inflation from 2000 to 2007 was at the low end of the scale of comparator countries ( 8 Table 1-1). Inflation fell rapidly after devaluation in 1994 to 2.5 percent in 1996, and averaged 2.9 percent through the end of the decade. In 2001, inflation accelerated to 4.3 percent due to rising oil prices, higher electricity tariffs and a new value-added tax. Although average inflation fell for much of 2002, prices rose sharply given civil war-related scarcities, and annual inflation fluctuated between 1.4 percent and 4 percent over the next several years. With the global rise in commodity prices in the second half of the decade, inflation increased rapidly. Even as prices dropped with the onset of the economic crisis in 2008, the annual rate of inflation that year was still 6.3 percent. The recent rise in commodity prices in international markets is expected to generate inflationary pressures, with year-on-year inflation reaching 5.3 percent by the end of 2010. The link to the euro limits flexibility and autonomy vis-à-vis monetary policy, leaving fiscal policy as the main instrument for managing inflation and the economy as a whole. This has been a source of tension with the IMF over the years, as economic and political crises have caused rapid accumulation of both domestic and external arrears in payments. EXCHANGE RATE While gains in competitiveness were largely preserved in the aftermath of the 1994 devaluation, a real appreciation of the CFA franc started in late 2000. After devaluation, the real effective exchange rate stabilized about 30 percent lower than the pre-1994 level. But from October 2000 to late 2003 there was a real appreciation of 20 percent. Part of this reflected the appreciation of the euro vis-à-vis the dollar. But at the sectoral level, competitiveness does appear to have worsened due to a decline in productivity and increased costs of production, transportation and trade for industries in war-affected areas. Some regional trade and transshipment has been diverted to neighboring countries such as Ghana. This is likely due to poor logistical capacity in Côte d’Ivoire due to the war. Recent appreciation in the real exchange rate and the consequent decrease in competitiveness of agricultural exports has generated much debate within Côte d’Ivoire over a possible second depreciation. But there is significant political opposition to devaluation since it would lead to a short-term fall in living standards and could affect the region’s reputation for macroeconomic stability. The CFA franc is pegged to the euro at a fixed rate of CFAfr 655.96:€1. The US dollar appreciated sharply against the euro in the second half of 2008 and is expected to maintain some of that appreciation from US$1.47:€1 to an expected average of US$1.39:€1 in 2010. One likely forecast is that the CFA franc will depreciate from its 2008 level of CFAfr 448:US$1 to an average of 474:US$1 in 2010, weakening arguments for a more flexible exchange rate regime. MANUFACTURING FIRM PERFORMANCE AND EXPORTS An increase in exports is critical to accelerating growth and reducing poverty. But improvements on the export front do not yet look promising. The Ivorian economy changed very little in the last two decades and agriculture still dominates the economy, accounting for over 26 9 percent of GDP and providing employment for 49 percent of the labor force. Côte d’Ivoire is the world’s largest producer of cocoa, accounting for approximately 40 percent of the global supply. Côte d’Ivoire’s economic growth tends to reflect fluctuations in revenue from this all-important crop, which has withstood the strain of political conflict and provided badly needed revenue for the government. However, recent analysis of the cocoa sector suggests that revenue from exports has helped finance armed forces on both sides of the civil war. Services are mostly led by trade and transportation activities, and accounted for 49.8 percent of GDP in 2005. However, the protracted conflict and political instability has diverted trade, with landlocked neighbors turning to ports in Ghana, Togo and Senegal. Ports in Tema (Ghana) and Lomé (Togo) in particular have seen traffic increase at the expense of Abidjan over the last four years. The industrial sector, including construction, accounted for 22.3 percent of GDP in 2005. Prior to the current crisis, manufacturing activities had expanded rapidly, led by agro-processing as well as a diverse range of consumer goods produced for domestic and export markets. Manufacturing growth turned negative in 2000, however, dragged down by construction-related industries. A nascent recovery in early 2002 was undermined by the outbreak of civil war in September of that year. The economy has not recovered from the resulting sharp contraction because of political uncertainty and other underlying weaknesses. Export volume fell 8.7 percent between 1994 and 1999 and by 2.3 percent from 2000 to 2006. This suggests that Côte d’Ivoire may find it difficult to expand its export basket beyond cocoa, coffee and oil – the effects of a recent civil war and ongoing political crisis notwithstanding. As we shall see in Chapter 2, only 6.5 percent of firms responding to the latest Enterprise Survey depend directly or indirectly on exports for 20 percent or more of total sales. The overall real appreciation of the CFA franc in the past eight years has reduced the competitiveness of the country’s agricultural and industrial exports, while the manufacturing sector is suffering from increased competition from cheap goods from Asia. It appears that political instability and the resulting economic stagnation will limit investments in most sectors for the medium to long term. MACROECONOMIC PERFORMANCE AND THE INVESTMENT CLIMATE It is important to understand how the decisions of managers are affected by these macroeconomic circumstances. Enterprise Surveys collect information on the investment climate, including such topics as corruption, competition from the informal sector, political instability, and worker education and skills. While subjective data like these have limitations, managers’ views about the macroeconomic environment and other things are nonetheless useful because they likely affect the level and type of investments, and may explain patterns we see in other data on Côte d’Ivoire. We find that political uncertainty is embedded in managers’ perceptions of the macroeconomic circumstances of Côte d’Ivoire. Comparing these perceptions to those in comparator countries highlights the importance of restoring managers’ confidence (Figure 1-1). More than 87 percent of firm managers highlight political instability as a major or very severe obstacle to firms operations and development, far more than in relatively stable countries like Senegal, South Africa, Morocco or 10 Mali. This is high even compared to countries that have experienced political turmoil in recent years (Togo, Madagascar, Burkina Faso or Kenya). FIGURE 1-1: INTERNATIONAL COMPARISON OF MANAGER PERCEPTIONS OF POLITICAL INSTABILITY (percent of manufacturing firms indicating political instability is a major or severe problem) Political instability in Côte d’Ivoire is ranked as the greatest constraint to doing business by firm managers, just ahead of access to finance. The third and fourth constraints for Ivorian firms, namely corruption and crime, theft and disorder, may well be a result of the underlying political instability. POLITICAL INSTABILITY We now describe briefly the nature of political instability in Côte d’Ivoire that has gripped the country for more than 15 years. Côte d’Ivoire became independent in August 1960, with Félix Houphouët-Boigny taking office as president. President Houphouët-Boigny came to dominate the country’s political life, and in the 1960s and 1970s presided over Côte d’Ivoire’s emergence as one of Africa’s few stable and economically successful countries. This was achieved by cautious redistribution of cocoa export revenues to manage political conflict through clientelism and paternalism, the migration of labor from the north and neighboring countries to agriculturally productive areas in the south, and strong bilateral support from France. In the 1980s, commodity prices fell and Côte d’Ivoire began to face serious economic and social problems. As President Houphouët-Boigny’s grip on power weakened, popular dissent increased and by the beginning of the 1990’s demonstrations and strikes had become commonplace. With this decline in commodity prices came increasing fiscal constraints and mounting debts. The death of President Houphouet-Boigny in 1993 was the tipping point after which a series of periodic constitutional and economic crises wreaked havoc on the economy. As the economy contracted in the 1990s, Ivoirité, or “Ivorianness�, emerged as central to a wide range of socioeconomic issues around civil and economic rights, land ownership and national governance, and access to power. 11 After the death of President Houphouet-Boigny, a power struggle began between his appointed successor Henri Konan Bédié and his prime minister Alassane Ouattara. President Bédié began consolidating his own power, moving his loyalists into key positions in the administration and sidelining those sympathetic to Mr. Ouattara. As a consequence, President Bédié was re-elected with 95 percent of the vote in the October 1995 presidential election. Support for the Bédié government started to flag in 1998 as the economy showed signs of faltering, but Mr. Ouattara continued to be sidelined over questions about his origins and “Ivoirité�. On December 24, 1999, President Bédié was overthrown in a bloodless coup led by a group of young army mutineers headed by General Robert Gueï, who himself was ousted in October 2000. Despite substantial misgivings, France and other influential partners eventually endorsed the election of Laurant Gbagbo in the absence of a better alternative. A coup attempt in September 2002 and the ensuing descent into civil war took the country entirely by surprise. Several days of violence in Abidjan led to the assassination of General Gueï and Emile Boga Doudou, the interior minister, as well as attempts on the life of Mr. Ouattara and other leading political figures. A military front hardened along an east-west line, splitting the country almost exactly in two. Fearing a prolonged civil war that could engulf other countries in the region, the French government stepped up its military presence in Côte d’Ivoire, agreeing to police a ceasefire line. However, several setbacks to the peace process took place between the end of 2002 and 2004. One major blow came in November 2004 when four days of anti-French violence rocked Abidjan. With international concern growing, mediation efforts were consolidated. The main outcome was a decision that each party be allowed to present the candidate of its choice in the election, thereby allowing Mr. Ouattara to run. This election was supposed to pave the way for increased political stability, but it has been delayed several times due to political disagreements, and administrative constraints concerning disarmament and national identification. These constraints will continue to affect the process, and voter registration is unlikely to be completed until mid-2010. As a result, the election will not take place until late 2010 at the earliest. Although all major political parties have accepted the delays, they are becoming increasingly frustrated with the situation. If the poll is postponed 2010, the risk of violence between supporters of the different parties will increase. SUMMARY OF FINDINGS Optimism regarding Côte d’Ivoire’s macroeconomic performance is justifiable, but is very much conditioned on restored political stability. Cocoa exports remain the engine of the economy, and infrastructure and the industrial sector are relatively well developed. Membership in the CFA zone has kept monetary policy stable as well. However, post-conflict economic success will rely heavily on economic diversification that enables the country to better respond to external economic forces. 12 Chapter 2 – MANUFACTURING Firm performance and how it may be affected by the investment climate is central to the inquiry of this chapter. Manufacturing firm performance is mostly evaluated by looking at labor productivity and total factor productivity. The analysis finds that firms in Côte d’Ivoire are less productive than firms in most comparator countries. We also find large disparities between firm performance and these performance disparities are correlated with particular firm characteristics, mostly firm size. Because firm characteristics are important in explaining firm performance, we explore firm size and other characteristics such as a firm’s use of technology, a firm’s export and/or import orientation, a firm’s use of skilled labor and management’s educational achievement, and a firm’s use of financial instruments to finance their operations.. INTRODUCTION Côte d’Ivoire is one of the very few countries in Sub-Saharan Africa with a ratio of manufacturing to GDP higher than 15 percent (Figure 2-1). In fact, the share of manufacturing in the Ivorian economy is 19 percent, similar to high growth economies such as Morocco and Vietnam FIGURE 2-1: INTERNATIONAL COMPARISON OF VALUE ADDED IN MANUFACTURING (19 and 21 percent, respectively). Few countries in Sub-Saharan Africa have managed to enter Mauritius international markets for manufactured goods, and South Africa Cote d'Ivoire Côte d’Ivoire’s strong export performance is at least Lesotho Cameroon in part due to its large manufacturing sector. Mozambique Seychelles Namibia In this section we explain the observed Senegal performance of the Ivorian manufacturing sector as Burkina Faso Malawi measured by the productivity of manufacturing Madagascar Kenya enterprises. We begin by comparing Ivorian Zambia Uganda manufacturing firms to those in other countries in Eritrea Ghana Sub-Saharan Africa, looking at labor and capital Guinea-Bissau productivity across countries then examining total Tanzania Congo, Dem. Rep. factor productivity among Ivorian manufacturing Sudan Rwanda firms. Chad Congo, Rep. Cape Verde Ethiopia Angola DATA: ENTERPRISE SURVEYS Guinea Botswana Mali We use various measures of manufacturing firm Nigeria performance from World Bank Enterprise Surveys1 0 10 20 30 in order to calculate capital and labor productivity, Value added in manufacturing (% of GDP) as well as to determine export performance and identify firm characteristics most associated with Source: World Bank (2008) higher or lower levels of performance. The 1See enterprisesurveys.org. Interested readers should look for the “Full Survey Data� option and fill out a confidentiality agreement to get access to all data sets and questionnaires. 13 Enterprise Surveys also provide useful information on firm perceptions of the investment climate. The data used in this analysis are comprised of responses from a sample of 526 firms, of which 36 percent are in manufacturing (Table 2-1). Most firms (92 percent) are based in the Abidjan metropolitan area, with fewer than 8 percent based in Yamoussoukro and San Pedro. There are 131 microenterprises with fewer than five employees, but because of differences in sampling strategies we focus on small, medium and large enterprises in this section. TABLE 2-1: CHARACTERISTICS OF SAMPLE Total Number of firms 526 City: Establishment size (employees): Abidjan 92% Micro (0-4) 25% San Pedro 7% Small (5-19) 48% Youssoukro <1% Medium (20-99) 19% Large (100-499) 8% Ownership: Sectors: Corporate 15% Manufacturing 36% Individual proprietorship 72% Trade (Wholesale and Retail) 20% Other 13% Other 44% Source: Enterprise Surveys, World Bank Because similar surveys have been conducted in close to 100 countries throughout the world, including approximately 40 countries in Sub-Saharan Africa, we can benchmark Côte d’Ivoire’s investment climate against those of other low-income countries in the region (Table 2-2). TABLE 2-2: COMPARATOR COUNTRIES Burkina Faso Cameroon Ghana Kenya Togo Madagascar Mali South Africa Senegal LABOR PRODUCTIVITY Labor productivity is relatively low in Côte d’Ivoire. The median firm produces approximately US$1677 of value added per worker, which is lower than the corresponding figure for all the comparator countries (Figure 2-2). Firms in Kenya, Cameroon and South Africa produce between five and six times as much per worker. As expected, labor productivity is higher in larger Ivorian firms. However we note that compared to other low-income countries in Sub-Saharan Africa, Côte d‘Ivoire’s gap in labor productivity between large and small enterprises is by far the largest, at over $38,000. 14 In this sample of Ivorian manufacturing firms, the median large enterprise is more productive than the median large enterprise in the comparator countries. The median small enterprise, on the other hand, is less productive. While it is common to have large differences between the productivity of small and large firms, the gap in Côte d‘Ivoire is particularly large. FIGURE 2-2: INTERNATIONAL COMPARISON OF VALUE ADDED PER WORKER, BY FIRM SIZE South Africa Kenya Togo Cameroon Senegal Median Ghana Small Madagascar Large Mali Burkina Faso Côte d'Ivoire 0 10,000 20,000 30,000 40,000 50,000 Source: Enterprise Surveys, World Bank Notes: Comparisons only include manufacturing enterprises. Data were collected between 2002 and 2008 depending on the survey period for each country. Value added is calculated by subtracting intermediate inputs and energy costs from sales. Employees include both permanent and temporary workers. Data collected prior to 2005 are converted to 2005 figures using GDP deflators. CAPITAL PRODUCTIVITY The low capital intensity of manufacturing firms in Côte d’Ivoire is also problematic. Given the strong bias in the Ivorian economy towards agriculture and commodities, firms tend to have less machinery and equipment per worker than firms in the comparator countries (Table 2-3). The median Ivorian firm has about US$340 of machinery and equipment per worker, while those in South Africa and Kenya have about US$9000 and US$12,000, respectively, of machinery and equipment per worker. Capital intensity is relatively low in Côte d’Ivoire, but returns are high (Table 2-4). Return to capital, as defined by the ratio of valued added to the value of capital stock, is 3.3--the highest among its peers,. However, because cross-industry comparisons may be distorted by cross-sector variation, the more relevant statistic on returns to capital looks at firms in the same industry group, such as garments or agro-processing. In the case of agro-processing, Ivorian firms’ returns to capital are far higher, at 8.6, whereas the next-highest returns are approximately 2.8 in Ghana and South Africa. The high returns to capital enjoyed by agro-processing firms may reflect the degree of specialization in agricultural commodities like cocoa and cotton, as well as increasing returns to 15 scale of their production technology. This is consistent with the high labor productivity of large firms highlighted in Figure 2-2. Returns to capital are high for both small and large firms and in fact exceed the returns for firms in all comparator countries . Returns to capital for large firms are three times the returns for firms in South Africa, the second-most productive comparator country. Note that this extremely high productivity of capital, coupled with high labor productivity, is again consistent with increasing returns to scale in Côte d’Ivoire’s economy (see Figure 2-3). TABLE 2-3: INTERNATIONAL COMPARISON OF MEDIAN CAPITAL INTENSITY, BY INDUSTRY All Agro-processing Garments Burkina Faso 1,194 1,964 - Cameroon 2,084 6,579 893 Côte d'Ivoire 343 417 250 Ghana 407 455 327 Kenya 12,187 11,976 10,874 Madagascar 1,819 2,046 626 Mali 574 861 201 Senegal 1,739 2,838 1,004 South Africa 8,861 7,642 4,726 Source: Enterprise Surveys, World Bank and author calculations. Note: Calculations based on WDI nominal exchange rate. TABLE 2-4: INTERNATIONAL COMPARISON OF MEDIAN CAPITAL PRODUCTIVITY All Agro-Processing Burkina Faso 1.5 0.7 Cameroon 2.9 0.8 Côte d'Ivoire 3.3 8.6 Ghana 2.4 2.8 Kenya 0.9 0.8 Madagascar 1.0 0.6 Mali 3.4 1.6 Senegal 1.9 1.8 South Africa 2.0 2.7 Source: Enterprise Surveys, World Bank and author calculations. 16 FIGURE 2-3: INTERNATIONAL COMPARISON OF MEDIAN RETURNS TO CAPITAL, BY FIRM SIZE Côte d'Ivoire Mali Cameroon Ghana Burkina Faso South Africa Small firms Senegal Kenya Large firms Madagascar 0 5 10 Returns to capital Source: Enterprise Surveys, World Bank EXPORT PERFORMANCE Côte d‘Ivoire’s dominant role as a cocoa exporter is TABLE 2-5: INTERNATIONAL reflected in the macroeconomic data on exports. In South COMPARISON OF SHARE OF MANUFACTURES IN Africa and Togo, manufactured goods exports account for a MERCHANDISE EXPORTS greater proportion of exports than in many other African countries (Table 2-5). However, Côte d‘Ivoire’s productivity is % substantially higher than most of its neighbors, reflecting its Cameroon 3 position as the world’s leading exporter of cocoa. The cross- Côte d'Ivoire 18 country comparison provided below presents a picture of Ghana 11 Côte d‘Ivoire’s place among comparator peers. Kenya 37 As expected, given low labor productivity and low capital Madagascar 57 intensity, Ivorian firms compare unfavorably with firms in Mali 3 comparator countries with respect to international trade Senegal 36 (Figure 2-4). Less than 10 percent of manufacturing firms in South Africa 51 Côte d’Ivoire export goods, as compared to 20 percent of firms Togo 62 in all nearly all comparator countries. Despite the anemic Source: World Bank (2008) participation of firms in export markets, Côte d’Ivoire is a major player in international commodities markets, which suggests that the exports are highly concentrated. Comparing export participation by firm size reveals that nearly 55 percent of large firms are exporters as opposed to only 6 percent of small firms. This is consistent with the other empirical findings that larger firms tend to be more efficient and as a result are better suited to exporting. Large agro-processing firms dominate cocoa and cotton exports. But few other firms are able to export, as compared to their peers in comparator countries. 17 FIGURE 2-4: INTERNATIONAL COMPARISON OF EXPORT PRESENCE, BY FIRM SIZE Kenya Ghana Madagascar South Africa Togo Senegal All firms Cameroon Small firms Burkina Faso Large firms Mali Côte d'Ivoire 0% 25% 50% 75% 100% Percent of firms export Source: Enterprise Surveys, World Bank. FIRM-LEVEL PERFORMANCE We now look at the performance of domestic firms, comparing manufacturing firms on several dimensions including size, location, ownership structure and export status. We find significant differences in the performance of small and large domestic firms. We begin with measures of firm performance in 18 Table 2-6 grouped by firm characteristics. There are three major patterns in the data that are worth highlighting. Large firms demonstrate generally better performance, while small firms have higher unit labor costs. The difference between exporting and non-exporting firms is striking as well. Exporting firms are likely able to pay higher wages because they can compensate for higher unit labor costs with relatively efficient production. Exporters must also survive in highly competitive world markets. Foreign-owned firms follow a similar pattern, in large measure because most exporters are foreign-owned. 19 TABLE 2-6: MEDIAN PRODUCTIVITY, BY FIRM CHARACTERISTICS Value Labor Unit Capital Per Added per Cost per Labor Worker Capital Worker Worker Costs (sales value) Productivity $ $ % $ % All 1678 645 37% 343 335% Sector Garments 1553 501 40% 250 420% Food and Beverage 5626 1182 28% 417 861% Chemicals 18278 2028 18% 464 376% Plastics 42139 3864 15% 5173 444% Other 1673 835 28% 1043 115% Size Small (5-19 employees) 1824 626 23% 184 453% Medium (20-99 employees) 1605 1081 53% 626 154% Large (more than 100 employees) 38709 2630 9% 5173 945% Manager Education University or higher 13042 3338 28% 1794 195% Vocational 1723 913 37% 348 400% Secondary or less 1648 584 40% 343 287% Exports Exporters 13042 3709 28% 1788 263% Non-exporter 1673 609 40% 343 376% Foreign Ownership Foreign Owned 9619 1335 28% 469 444% Domestic 1673 613 40% 343 335% Internet Use Internet User 6082 1739 28% 626 154% Non-user 1553 584 40% 250 420% Bank Credit Bank Credit 13042 3709 28% 5173 1527% No Bank Credit 1673 609 40% 343 295% Source: Enterprise Surveys, World Bank Note: Firms where data available. Capital-related data only for manufacturing firms. Median weights used. PRODUCTIVITY Estimation of productivity. We now turn to estimating total factor productivity. The econometric models control for capital and labor2 and use dummies to control for sector and firm size.3 By including these dummy variables, we allow average productivity to vary across sectors and by firm size. Because firms in different sectors use different production technologies, the coefficients for capital and labor will differ across sectors. For this reason, we interact sector dummies with capital 2 Results are similar using log sales as the dependent variable, and including intermediate inputs in addition to labor and capital. 3 We ignored location controls given the structure of the data described at the beginning of the chapter. The vast majority of the firms sampled are in Abidjan. 20 and labor to control for sector-specific production technologies (Table 2-7).4 A joint significance test on the interaction terms rejects the null hypothesis that the coefficients are equal across sectors. This suggests that it is inappropriate to pool enterprises from different sectors into a single model without controlling for sector differences in factor intensity. Earlier discussion of capital intensity and capital and labor productivity pointed to their relationship with firm size. So we hypothesize that firms of different sizes use different production technologies and interact size-specific dummies with the labor and capital variables to test this hypothesis. As with sector-specific technologies, the statistical test for these interaction terms rejects the hypothesis that technologies used do not vary across firm size. This also suggests that it is inappropriate to pool estimates across firms of different sizes. Size and sector dummies. The magnitude and statistical significance of the coefficients for the size dummies support the findings of the previous section that small establishments on average perform substantially less well than larger enterprises. Because we hold all other factors constant, we can establish that size is not a proxy for another variable, such as sector, and that small firms in Côte d’Ivoire indeed perform less well. 4The coefficients on labor and capital are allowed to vary across sectors and the model is estimated after adding the sector dummies and interaction terms to the regression. These coefficients are not reported in the table due to space constraints. 21 TABLE 2-7: TFP ESTIMATION, FRONTIER MODEL. Dependent variable: log of Value Added Dependent Variable: Log of Value Added (1) (2) (3) (4) (5) (6) (7) (1) (2) Production function: Production Function Labor (log) 1.3535 1.316 (0.2156)*** (0.2122) Capital (log) 0.31*** 0.33*** 0.24*** 0.23*** 0.25*** 0.27*** Capital (log) 0.13*** 0.2843 0.236 (0.0918)*** (0.0942 (0.08) (0.08) (0.09) (0.08) (0.08) (0.08) (0.04) Establishment characteristics: Incorporated (dummy) 0.440 Labor (log) 1.27*** 1.26*** 1.19*** 1.27*** 1.34*** 1.30*** 1.46*** (0.1806 (0.34) (0.33) (0.35) (0.33) (0.35) (0.35) (0.36) Establishment < 5 years (dummy) -0.331 (0.1614 Establishment Characteristics Female-owned establishment (dummy) 0.117 (0.158 Industrial zone (dummy) -0.246 (0.178 Incorporated (dummy) -0.60 Globalization: -0.53* Exporter (dummy) (0.52) (0.32) Importer of inputs (dummy) Establishment < five years old (dummy) -0.39 0.56** Foreign-owned establishment (dummy) (0.37) (0.27) Technology: Globalization Establishment uses foreign-licensed technology (dummy) Exporter (dummy) 0.50 Establishment uses IT (dummy)0.53** (0.49) Human Capital: (0.27) Importer of Inputs (dummy) 0.03 Skilled laborforce (dummy) 0.02 (0.58) (0.30) Manager has at least a tertiary education (dummy) Foreign-owned establishment (dummy) 0.71* Finance: 0.42 Establishment has bank account (dummy) (0.42) (0.26) Overdraft facility (dummy) Technology Control variables: Establishment uses foreign-licensed technology (dummy) 0.95* Location 0.31 yes yes Sector yes yes (0.55) Size (0.33) yes yes Establishment uses IT (dummy) 0.93** Constant 0.87** 10.6795 11.631 (1.7309)*** (1.7756) (0.39) Observations (0.43) 292 292 Standard errors in parentheses Human Capital * significant at 10%; ** significant at 5%; *** significant at 1% Skilled labor force (dummy) 0.51 0.25 (0.50) (0.56) Manager has at least tertiary education (dummy) 1.22*** 0.44 (0.43) (0.49) Finance Overdraft facility (dummy) 0.16 0.02 (0.43) (0.44) Establishment has bank account (dummy) 0.64 0.50 (0.41) (0.36) Control Variables Size yes yes yes yes yes yes yes Sector yes yes yes yes yes yes yes Constant 9.08*** 9.08*** 10.63*** 9.88*** 9.50*** 9.22*** 11.11*** (1.79) (1.76) (2.06) (1.76) (1.68) (1.77) (1.46) N 82 82 82 82 76 82 76 note: *** p<0.01, ** p<0.05, * p<0.1 Standard errors in parenthesis. 22 TABLE 2-8: TFP MODEL WITH INPUT INTERACTIONS Dependent variable: log of Value Added Dependent Variable: Log of Value Added (1) (2) (3) (4) (5) (6) (7) (1) (2) Production function: Production Function Labor (log) 1.3535 1.316 (0.2156)*** (0.2122) Labor (log) 2.25*** 2.27 1.84* 1.93** 2.17 2.14* Capital (log) 0.61 0.2843 0.236 (0.0918)*** (0.0942 (0.84) (1.41) (0.98) (0.82) (2.08) (1.21) (1.17) Establishment characteristics: Incorporated (dummy) 0.440 Capital (log) 0.35 0.34 0.37 0.30 0.00 0.17 0.25 (0.1806 (0.27) (0.32) (0.33) (0.28) (0.35) (0.32) (0.39) Establishment < 5 years (dummy) -0.331 (0.1614 Establishment Characteristics Female-owned establishment (dummy) 0.117 (0.158 Industrial zone (dummy) -0.246 (0.178 Incorporated (dummy) 0.03 Globalization: 0.40 Exporter (dummy) (1.19) (1.01) Importer of inputs (dummy) Establishment < 5 years old (dummy) 0.00 0.09 Foreign-owned establishment (dummy) (0.28) (0.68) Technology: Globalization Establishment uses foreign-licensed technology (dummy) Exporter (dummy) -0.03 Establishment uses IT (dummy) -0.78 (1.13) Human Capital: (0.82) Importer of Inputs (dummy) 1.18 Skilled laborforce (dummy) 0.23 (0.93) (0.50) Manager has at least a tertiary education (dummy) Foreign-owned establishment (dummy) 0.49 Finance: 1.15 Establishment has bank account (dummy) (1.01) (0.71) Overdraft facility (dummy) Technology Control variables: Establishment uses foreign-licensed technology (dummy) 1.70*** Location -0.13 yes yes Sector yes yes (0.57) Size (0.68) yes yes Establishment uses IT (dummy) 0.27 Constant 0.21 10.6795 11.631 (1.7309)*** (1.7756) (0.30) Observations (0.63) 292 292 Standard errors in parentheses Human Capital * significant at 10%; ** significant at 5%; *** significant at 1% Skilled labor force (dummy) -1.44*** -0.50 (0.14) (0.43) Manager has at least tertiary education (dummy) 1.72*** 1.78** (0.59) (0.74) Finance Overdraft facility (dummy) 0.32 -0.30 (0.65) (0.71) Establishment has bank account (dummy) 0.96** 0.89*** (0.44) (0.29) Control Variables Size yes yes yes yes yes yes yes Sector yes yes yes yes yes yes yes Interactions with Inputs yes yes yes yes yes yes yes Constant 4.97 4.98 6.12 6.47 11.25* 7.85** 11.31* (3.84) (3.89) (4.03) (4.19) (5.86) (3.84) (6.16) N 82 82 82 82 76 82 76 note: *** p<0.01, ** p<0.05, * p<0.1 Standard errors in parenthesis. Source: Enterprise Surveys and author calculations. Table 2-7 and Table 2-8 display the results for several specifications to estimate TFP. To save space, the tables do not include results from all the regressions that were run. We begin by presenting in column 1 of Table 2-7, the results for a basic production function, where the only variables are capital, labor, control variables such as sector (plus interaction terms with labor and capital), location and firm size (plus interaction terms with labor and capital). We then add variables to elucidate the relationship between TFP and establishment characteristics (column 2), the effects of globalization (column 3), the use of technology (column 4), management education and skill and sophistication of the labor force (column 5) and access to finance (column 6). The final specification in column 7 includes all variables. We estimate a Cobb-Douglas production function using a standard stochastic frontier approach, in order to distinguish between technical efficiency (i.e. how well the individual firm performs relative to the best firms in the sample) and random noise due to shocks or measurement error. 23 Economies of scale. The analyses of performance through measures of capital and labor productivity suggest substantial differences between firms of different sizes: Small firms perform considerably worse than large firms. The results of the multivariate econometric model confirm the univariate results – coefficients for firm size variables indicate that large firms have significantly greater TFP than small firms. We test for a direct relationship between firm size and performance by examining economies of scale across manufacturing sectors. Economies of scale would result in large enterprises being consistently more productive than small enterprises. To test for this phenomenon, we add up the coefficients for labor and capital from the TFP estimation. A sum greater than one suggests that total production would more than double with a doubling of the number of workers, the amount of capital and the amount of intermediate inputs. For the agro-processing sector, at conventional significance levels we reject the null hypothesis that the coefficients sum to one, providing evidence of economies of scale (Table 2-9). TABLE 2-9: EVIDENCE OF ECONOMIES OF SCALE IN AGRO-PROCESSING. Test of constant returns to scales Sum of chi-squared p-value coefficients Agro-processing 2.36 4.66 0.0309 Overall 2.6 4.65 0.0311 Source: Enterprise Surveys and author calculations. Establishment characteristics. We also examine correlations of various establishment characteristics, such as legal status and firm age, to TFP, controlling as before for firm size and sector in order to determine the direct correlation between these characteristics and TFP. In developing countries, legal systems tend to be weak, and may offer little credible protection against unreasonable liability. Incorporation is one mechanism of limiting such risk, and could have some impact on firm performance. In Côte d’Ivoire we do not find a relationship between productivity and organizational structure. When we account for input interactions, the results also indicate that establishments five years old or younger tend to be more productive than other establishments, and that these young, foreign-owned firms tend to grow faster than more mature firms. One explanation is that firms that profitably entered the market after the turmoil of 2002 must be very productive to survive. Export orientation. According to the economics literature, links to the global economy confer performance advantages to firms through a number of different channels, including exploitation of economies of scale and the increasing degree of competition they face. Recent research emphasizes the relationship between economic openness – a broader concept encompassing imports and international investment in addition to exports – and firm performance and economic growth.5 5 See Dollar and Kraay (2001) for an extensive survey of the related literature supporting the argument for globalization. 24 To examine the relationship between such linkages to the world economy and manufacturing performance in Côte d’Ivoire, we include three dummy variables:  whether an establishment directly or indirectly exports more than ten percent of its output;  whether the establishment imports any of its inputs;  whether ten percent or more of the establishment is foreign-owned. We recall that exporting and non-exporting firms displayed major differences in performance on several different measures. Surprisingly, we find that export orientation is not correlated to performance for manufacturing firms in Côte d’Ivoire. Other measures of export orientation are also statistically insignificant. The coefficient for the foreign ownership dummy, for example, is not significant. This is surprising because other studies have found that foreign-owned firms perform better on average, usually because they have access to better technology and better access to financial markets, and can more easily exploit the comparative advantages of producing across several countries. These results may be due to the correlation between size and export orientation; the size variable may well be picking up the efficiency gains. Technology. Column 4 displays the econometric results with variables representing different measures of technology use included in the model. First, the coefficient of a dummy variable for use of foreign-licensed technology is positive and statistically significant, suggesting that firms using these technologies are more productive than those that do not. Second, the coefficient for the variable indicating whether a firm uses e-mail to communicate with clients and suppliers was found to be positively but insignificantly related to firm performance. Human capital. The results in Column 5 include two variables on human capital. First, the coefficient for a dummy variable indicating whether the firm’s top manager graduated from college suggests that managerial educational achievement and productivity are positively and significantly related. Second, a continuous variable indicating the ratio of skilled to unskilled workers employed by the firm indicates that worker education is unrelated to firm performance. Access to finance. The positive relationship between firm performance and use of financial instruments is well established. Recent research6 using detailed firm-level data identifies several ways in which access to finance can improve performance. For example, access to formal financial services helps firms exploit investment opportunities, finance production and process innovation, and choose more efficient organizational structures and asset portfolios. For these reasons, we include a dummy variable indicating whether a firm has a bank account, and another indicating whether it has access to an overdraft facility. However, the coefficients for both variables are insignificant. 6Beck, T., A. Demirgüç-Kunt (2006). Small and Medium-Size Enterprises: Access to Finance as a Growth Constraint, Journal of Banking and Finance, forthcoming. 25 SUMMARY OF FINDINGS Ivorian firms are less productive than those in the comparator countries mainly because they are less capital and skill intensive, and are relatively small aside from a few large productive firms. Labor productivity appears to be particularly high among large agro-processing firms, as we would expect from such a big player in the world cocoa market. Small firms in Côte d’Ivoire are less productive and less capital intensive than small firms in other low income countries in Sub-Saharan Africa, and are primarily oriented toward the domestic market. Firms that do export have above- average productivity rates. Large disparities in manufacturing firm performance are particularly striking, whether we consider differences between small and large firm performance due to differences in capital intensity and production technologies, or between agro-processing firms or firms in the garment sector. Surprisingly, access to finance turned out not to be a major issue in the estimation exercise. While it is difficult to link poor firm performance to poor access to finance, this financial dimension of the investment climate is worth exploring further because financial access is often strongly associated with the performance of top-performing firms. 26 Chapter 3 – ACCESS TO FINANCE Factor markets are critical to the performance of firms, and anomalies in these markets may impose severe constraints related to labor regulation, worker education, and to access and cost of finance. In this chapter we will examine access to finance, leaving labor markets to Chapter 5. Firms in Côte d’Ivoire perceive limited access to finance and high borrowing costs as the main constraints on their growth. In this chapter we will analyze firms’ external sources of financing, focusing in particular on the banking system. It is important to note that although the availability and cost of capital depends on both supply and demand, we will limit our analysis of supply to an overview of the banking system7, and focus on various aspects of financial demand and firm perceptions of access to finance. FINANCIAL SYSTEM OVERVIEW The financial sector of Côte d’Ivoire is the largest and most diversified in the West African Economic and Monetary Union8 (WAEMU). The banking sector is the largest component of the financial sector in Côte d’Ivoire, and included 19 commercial banks at the end of 2008. The microfinance sector accounts for a small fraction of financial assets, but contributes significantly to access to financial services through its large client base with approximately 958,000 accounts. The insurance sector is comprised of 33 companies, while the retirement system is dominated by two public institutions. There are also three nonbank financial institutions, and one financial institution that helps microfinance institutions secure credit lines. Figure 3-1, Figure 3-2, and Figure 3-3 compare the health of Côte d’Ivoire’s financial sector to that of its peers on three indicators: the ratio of bank deposits to GDP, private bank credit to GDP, and private credit to deposits, respectively. With a few exceptions – most notably South Africa – we see that financial systems deteriorated in many countries in the 1990s. But while financial sector performance has generally improved in Côte d’Ivoire, it has lagged behind regional averages in recent years as a result of economic and political crises. Bank deposits represent a small share of GDP, the result of a gradual decline from 20 percent of GDP in the 1990s. Deposits staged a mild recovery from 13 percent of GDP in 2001 but reached only 16 percent by 2007. Other countries in the region followed similar trajectories, but several experienced rapid growth in deposits, most notably Senegal, Kenya and Ghana (Figure 3-1). The ratio of M2 to GDP was about 27 percent in 2007, relatively low compared to South Africa’s 63 percent or an average of 36 percent in Sub-Saharan Africa. Credit to the private sector as a share of GDP is lower in Côte d’Ivoire (16 percent) than the average in Sub-Saharan Africa (43 percent) and 10 times lower than in South Africa (Figure 3-2). 7The analysis of the finance system is largely based on a recent report that has been jointly produced by the World Bank and the IMF (World Bank and International Monetary Fund 2009). 8 In French: Union économique et monétaire ouest-africaine (UEMOA) 27 In 1990, the ratio of bank credit to bank deposits was high in Francophone countries of West Africa and peaked at 1.85 in Côte d’Ivoire (Figure 3-3). It then decreased dramatically in several countries, including Senegal and Cameroon, but the deterioration was extremely severe in Côte d’Ivoire, cutting the ratio in half in less than two decades. FIGURE 3-1: INTERNATIONAL COMPARISON OF BANK DEPOSITS OVER GDP, 1990-2007 0.7 Burkina Faso 0.6 Cameroon 0.5 Côte d'Ivoire 0.4 Ghana Kenya 0.3 Madagascar 0.2 Mali Senegal 0.1 South Africa 0 Togo 1990 1993 1996 1999 2002 2005 FIGURE 3-2: INTERNATIONAL COMPARISON OF FIGURE 3-3: INTERNATIONAL COMPARISON OF RATIO OF PRIVATE CREDIT BY DEPOSIT-TAKING RATIO OF PRIVATE CREDIT TO DEPOSITS, 1990 - BANKS TO GDP, 1990 - 2007 2007 2 Burkina Faso 0.8 1.8 Burkina Faso Cameroon 0.7 1.6 Cameroon Côte d'Ivoire 0.6 1.4 Ghana Côte d'Ivoire 0.5 1.2 Kenya Ghana 1 Madagascar 0.4 Kenya 0.8 Mali 0.3 Madagascar 0.6 Senegal 0.2 Mali 0.4 South Africa Senegal 0.1 0.2 Togo South Africa 0 0 Togo 1990 1993 1996 1999 2002 2005 1990 1993 1996 1999 2002 2005 Source: Enterprise Survey, World Bank The start of the civil war in 2002 heavily affected the Ivorian financial system. Banking activities ceased entirely in the center, north and west of the country until the Ouagadougou peace agreement was signed in 2007, and disbursements of technical and financial support have been disrupted. The microfinance sector, for example, lost support from long-time development partners and was unable to cover losses. Delayed donor disbursements for crisis recovery and budget support, along with difficulties mobilizing funds in regional markets, led to government arrears to private companies. Arrears amounted to about 4 percent of GDP in 2008, and undermine government credibility while representing a source of vulnerability for the financial system as a whole. Performance of the banking sector is mixed. Eight banks and one non-bank financial institution faced difficulties in 2008 due to undercapitalization, governance problems, or the risky economic and political environment. Bank solvency as measured by the risk-weighted capital adequacy ratio 28 (CAR) has declined markedly from 12.2 percent in 2002 to 9.25 percent at the end of 2008. Banks affiliated with multinational corporations have shown greater resilience to shocks, while smaller banks and local banks are severely undercapitalized. The share of capital and the number of financial institutions under state control has increased in recent years, from 21.5 percent to 28.6 percent in the period from 2006 to 2008 as the government assumed control of three insolvent banks. These three institutions were intended to intervene in specific sectors (housing, agriculture, and small and medium enterprises) underserved by other banks, but in fact operate as conventional commercial banks. Assuming control of these banks has increased the government’s exposure to risk. The National Investment Bank (BNI in French), the most important development finance fund, is also the recipient of funds from the state. BNI has engaged in conventional commercial bank activities similar to those undertaken by the private banks. However, its developmental impact would likely be higher if it operated on a for-profit basis, eliminating its dependence on government subsidies. Asset quality has improved in recent years, with non-performing loans (NPLs) peaking at 26 percent in 2004 and returning 19 percent – roughly pre-crisis levels – by 2008. Average return on assets (ROA) and return on equity (ROE) both rose from 2005 to 2008, from 0.3 to 1.6 percent and from 2.3 to 16 percent, respectively. After a long decline in concentration of the banking sector from 1993 to 2005, the trend started to reverse in 2005, and by 2008 three banks controlled 94 percent of assets (Figure 3-4). A recent increase in minimum capital requirements for non-bank financial institutions contributed to this oligopolization of banking, pushing the sector to a degree of concentration last seen in 1994. Similar trends occurred in Côte d’Ivoire’s neighbors, though typically not with the same intensity. Ghana was the only country that experienced a continuous decline in the concentration of the sector. In Côte d’Ivoire, lower minimum capital requirements for certain types of financial institutions that do not take deposits from the public might result in greater competitiveness in the marketplace with a higher rate of entry of new financial institutions. FIGURE 3-4: INTERNATIONAL COMPARISON OF BANK CONCENTRATION, 1993-2007 1 Burkina Faso Cameroon 0.9 Côte d'Ivoire 0.8 Ghana Kenya 0.7 Madagascar 0.6 Mali Senegal 0.5 South Africa Togo 0.4 1993 1996 1999 2002 2005 Source: Enterprise Survey, World Bank Bank lending is also relatively concentrated in loans to a few large customers, implying that SMEs represent a small share of bank credit. According to some reports, the 10 largest borrowers 29 account for about 18 percent of the sector’s entire loan portfolio; 50 borrowers account for 52 percent of loans. Loans not reported to the credit rating agency because they fall below the reporting threshold (CFAF 10 million) accounted for just 25 percent of total outstanding credit in 2008. For the sake of comparison, we note that these relatively small loans in 2007 accounted for 6 percent in Senegal, 29.4 percent in Togo, and 29.8 percent in Burkina Faso. Microfinance. The microfinance sector in Côte d’Ivoire attracts more customers than banks do. Nevertheless, it has the lowest penetration rate (25 percent) in the WAEMU, after Niger, according to the calculations of the Central Bank of West African States. Rural penetration is low, and large sections of the country are not served at all as a result of the socio-political crisis. The National Union of COOPECs of Côte d'Ivoire (UNACOOPEC-CI) accounted for 85 percent of the activity in the sector at the end of 2007, and financial difficulties in member institutions may cause a sector-wide crisis. The financial position of many microfinance institutions (MFIs) is extremely worrisome, and the major networks are virtually bankrupt. The sector as a whole posts negative equity, and a quarter of customer deposits have been wiped out by losses. The weight of microfinance assets in the financial sector is low, at 4 percent, but the social impact of the sector is significant owing to the large number of clients. The National Microfinance Commission (CNM in French), an MFI supervisory agency unique to the subregion, and other supervisory authorities have been unable to manage the growth and current financial deterioration of the microfinance sector. For example, the CNM has been unable to control the creation of new MFIs, which often start operating before they are licensed. Overall, rehabilitation of the sector is a prerequisite for the integration of microfinance into the traditional financial sector, including the restoration of close ties between banks and MFIs. Insurance. Côte d’Ivoire’s insurance sector is small, but leads among countries in the Inter- African Conference on Insurance Markets. In 2008, the market was comprised of 11 specialized life insurance companies and 22 property and casualty insurance companies, as well as one reinsurance company. The insurance sector could potentially play an important role in Côte d’Ivoire’s development by intermediating risk and providing access to long-term financial resources, but remains fairly under-developed. Retirement. The retirement system is dominated by two public institutions: the National Social Security Fund (CNPS in French) manages the retirement and workplace accident funds for private sector wage earners; the General Retirement Fund for Government Employees (CGRAE in French) is reserved for civil servants. Life insurance companies manage limited complementary retirement savings plans. Actuarial studies indicate that pension funds managed by the CNPS and the CGRAE are not financially viable in the long term without significant reforms. Doing Business and banking. Doing business indicators developed by the World Bank underscore the necessity of reform. The indicators rank countries along 10 dimensions, including ease of getting credit and dimensions that indirectly influence access to finance, such as ease of registering property, protecting investors, enforcing contracts or closing a business. These direct and indirect indicators of access to finance suggest that businesses face a difficult financial environment in Côte d’Ivoire. Overall, Côte d’Ivoire ranks very poorly (168th) on the Doing Business indicators, below a number of other countries in the region (Table 3-1). Breaking down the “Getting Credit� indicator 30 into its constituent parts, we see in Table 3-2 that Côte d’Ivoire typically lags far behind the regional average. However, on the indicators listed in Table 3-3 that are directly related to access to finance, Côte d'Ivoire appears in a more favorable – but still mixed – position. This overview of Côte d’Ivoire’s financial system and comparison with other countries in Sub- Saharan Africa suggest serious challenges going forward. The fairly strong financial system at the beginning of the 1990s has gradually weakened, and the civil war has hindered its recovery. Socio- political instability and weaknesses in the business environment have undermined the quality of bank portfolios and constituted major impediments to the financing of small and medium-sized enterprises (SMEs). TABLE 3-1: DOING BUSINESS RANKINGS Ease of Doing Registering Getting Protecting Enforcing Closing a Business Property Credit Investors Contracts Business Côte d'Ivoire 168 145 150 155 127 71 Burkina Faso 147 114 150 147 110 112 Cameroon 171 143 135 119 174 98 Ghana 92 33 113 41 47 106 Kenya 95 125 4 93 126 79 Madagascar 135 152 167 57 155 183 Mali 156 99 150 147 135 117 Senegal 157 166 150 164 151 80 South Africa 34 90 2 10 85 76 Togo 165 155 150 147 154 97 Source: Doing Business 2010 TABLE 3-2: DOING BUSINESS CREDIT INDICES Strength of legal Depth of credit Public credit registry Private credit rights index information index coverage (% of bureau coverage (0-10) (0-6) adults) (% of adults) Côte d'Ivoire 3.0 1.0 2.7 0.0 Burkina Faso 3.0 1.0 1.9 0.0 Cameroon 3.0 2.0 1.8 0.0 Ghana 7.0 0.0 0.0 0.0 Kenya 10.0 4.0 0.0 2.3 Madagascar 2.0 1.0 0.1 0.0 Mali 3.0 1.0 4.0 0.0 Senegal 3.0 1.0 4.4 0.0 South Africa 9.0 6.0 0.0 54.7 Togo 3.0 1.0 2.7 0.0 Sub-Saharan Africa 4.6 1.5 2.4 4.5 Source: Doing Business 2010 31 TABLE 3-3: DOING BUSINESS INDICES ON PROPERTY, INVESTORS, CONTRACTS AND BUSINESS CLOSINGS, Registering Protecting Enforcing Closing a Business Property Investors Contracts Time Strength of Time Time Recovery (days) protection (0-10) (days) (years) rate (cents/$) Côte d'Ivoire 62.0 3.3 770.0 2.2 34.0 Burkina Faso 59.0 3.7 446.0 4.0 21.7 Cameroon 93.0 4.3 800.0 3.2 25.5 Ghana 34.0 6.0 487.0 1.9 24.0 Kenya 64.0 5.0 465.0 4.5 31.6 Madagascar 74.0 5.7 871.0 N/A 0.0 Mali 29.0 3.7 626.0 3.6 20.9 Senegal 124.0 3.0 780.0 3.0 31.6 South Africa 24.0 8.0 600.0 2.0 32.2 Togo 295.0 3.7 588.0 3.0 26.6 Sub-Saharan Africa 80.7 4.4 643.9 3.4 17.0 Source: Doing Business 2010 PERCEPTIONS OF ACCESS TO FINANCE A majority of firms in Côte d’Ivoire perceive access to finance as the main obstacle to development, even more so that political instability, which is listed as the main constraint for a quarter of firms. Although the financing obstacle is widely cited by firms in several neighbors, the situation in Côte d’Ivoire is particularly severe, with 70 percent of firms listing it as their main constraint (Figure 3-5). FIGURE 3-5: INTERNATIONAL COMPARISON OF PERCEPTIONS ABOUT FINANCING PROBLEMS 80% 60% perception of financing 40% problems as most serious 20% obstacle to development 0% perception of financing problems as "major" or "very severe" obstacle to development Source: Enterprise Surveys, World Bank Note: Firms in the manufacturing sector. 32 Many firms share the view that access to finance is a TABLE 3-4: PERCEPTIONS OF major or very severe obstacle. Table 3-4 shows that different ACCESS TO FINANCE AS MAJOR/SEVERE OBSTACLE types of firms have somewhat different perceptions. The most affected enterprises are small, domestically oriented9, Firm characteristics % of firms belong to the manufacturing sector, have fewer than 20 Manufacturing 85 employees10 or operate in the region of San Pédro. Exporting Services 51 and foreign firms and subsidiaries of larger companies are Residual 61 less likely to cite access to finance as a major or very severe Micro 71 constrain, as they are more likely to have access to finance in other countries. Foreign 50 We estimate a Probit model to single out the main Ivorian 71 determinants of a firm’s perception that access to finance is a major or very severe obstacle to their growth. The Large 37 dependent variable is whether a firm states that financial Medium 61 access and costs are "major" or "very severe" obstacles. The Small 69 results are reported in Table 3-5. Consistent with the descriptive statistics above, firm size (number of employees) and the exporter dummy are negatively and significantly Abidjan 67 correlated with this view. In other words, large and San Pédro 84 exporting firms tend to feel less constrained by finance than Yamoussoukro 45 enterprises that have fewer employees and are oriented towards domestic markets. Firms in the manufacturing Subsidiary 50 sector appear more sensitive to financial constraints than Non-subsidiary 69 enterprises operating in other sectors. The firm owner’s nationality and the firm’s age and status as a subsidiary do Source: Enterprise Survey, World Bank not significantly affect the probability that a firm perceives finance as a major or severe constraint. In conclusion, available data concerning firm perceptions of the financial sector in Côte d’Ivoire suggest that 1) firms lack confidence in the financial system, and 2) perceptions vary to some extent with size, sector and commercial orientation. 9 Firms are classified as "domestic" or "Ivorian" if less than 10 percent is owned by foreigners, and are otherwise "foreign". 10 A firm employing less than 20 workers is defined "small", between 20 and 99 "medium", and more than 99 "large" 33 TABLE 3-5: PROBIT ESTIMATE OF FIRM PERCEPTIONS OF FINANCE AS A CONSTRAINT Perception of access to external financing as "major" or Dependent Variable: "very severe" constraint Constant 0.618 (5.97)*** Sector: Manufacturing versus residual 0.316 (2.17)** Services versus residual -0.053 (-0.34) Microenterprises versus residual -0.060 (-0.36) Size: Medium versus small -0.280 (-2.01)** Large versus small -0.662 (-2.64)*** Group -0.284 (-1.52) Region: San Pedro versus Abidjan 0.279 (1.26) Yamoussoukro versus Abidjan -0.427 (-0.72) Foreign ownership 0.167 (-0.25) Exporter -0.673 (-2.37)** Age of firm 0.006 (0.07) STRUCTURE OF FIRM FINANCING Firms rarely rely on financing from the TABLE 3-6: SOURCES OF FINANCING FOR banking system. Internal funds and retained FIXED ASSETS OR WORKING CAPITAL earnings represent by far the main source of % of % of financing for firms in Côte d’Ivoire, both for fixed fixed working assets and for working capital (Table 3-6). assets capital However, excessive reliance on internal funds is a Internal funds/Retained 93.1 89.2 sign of potentially inefficient financial earnings intermediation. Funds borrowed from banks on Trade credits 3.1 3.5 average represent only 3.5 percent of working Banks 1.5 3.5 capital and 1.5 percent of fixed assets. Purchase credits from suppliers and advances from Non-bank financial 0.3 0.3 customers are also limited, representing on institutions average only 3.5 percent of working capital and Source: Enterprise Survey, World Bank 3.1 percent of fixed assets. This level of supplier 34 and customer credit is much larger than that of Cameroon and Mali, and similar to that of South Africa. The average share of fixed assets financed with internal funds or retained earnings ranges between 86 percent for enterprises operating in the manufacturing and service sector to 95 percent for microenterprises. Internal funds are most frequently used by young and small enterprises (defined either in terms of sales volume or number of employees) and by firms that are not subsidiaries of another firm. Small firms’ heavier reliance on financing working capital internally is widespread in the region – occurring in Mali, Burkina Faso, Ghana, Senegal and Cameroon – but the case of Côte d’Ivoire is particularly extreme. The use of funds borrowed from banks to finance the purchase of fixed assets varies considerably. Firms in the service sector finance on average 12 percent of fixed assets, while the corresponding figure in manufacturing is only 2 percent and barely above 0 percent for microenterprises. The proportion of fixed assets financed by banks is more significant for large firms and for those operating in the region of San Pédro (Table 3-7). When it comes to making investments, firms still eschew the banking system for financing to a striking degree compared to firms in neighboring countries (Figure 3-6). In sum, Ivorian firms rely almost exclusively on internal funds for financing, receiving only a minuscule share from banks, whereas in other countries in Sub-Saharan Africa firms turn to banks much more often for financing. In the Ivorian case, firms clearly act on their negative perceptions of banks, behavior that immediately raises questions about the reasons for those perceptions. We will investigate this in the next section. TABLE 3-7: BANK FINANCING OF FIXED ASSETS, BY FIRM TYPE Firm % of fixed characteristics assets Manufacturing 2 Abidjan 1 Services 12 San Pédro 4 Residual 1 Yamoussoukro 0 Micro 0 Large 8 Foreign 1 Medium 1 Ivorian 2 Small 1 Subsidiary 2 Non-subsidiary 1 Source: Enterprise Surveys, World Bank Note: Firms in the manufacturing sector. 35 FIGURE 3-6: INTERNATIONAL COMPARISON OF BANK FINANCING AS SHARE OF TOTAL INVESTMENTS South Africa Burkina Faso Kenya Togo Cameroon Senegal Ghana Madagascar Côte d'Ivoire Mali 0 10 20 30 Source: Enterprise Surveys, World Bank. APPLYING FOR CREDIT Data from the Enterprise Survey for Côte d’Ivoire indicate that applications for lines of credit or loans are rare. Overall, only 9 percent of firms applied to a financial institution for one or more lines of credit or loans in 2007. The demand for bank financing is very heterogeneous across industries. In particular, most applications as a share of all firms in an industry are filed by firms in the food industry (45 percent), plastics and rubber (33 percent), IT (30 percent), and transportation (21 percent). We also note that applications for credit varied substantially depending on a number of characteristics summarized in Table 3-8. Older firms are more likely to apply, as are firms that have an internationally recognized quality certification or multiple owners, or if the current owner started the business or owns at least part of the land occupied by the firm. Data from the Enterprise Survey for Côte d’Ivoire indicate that of firms not applying for credit, only 13 percent didn’t actually need it. The remaining 87 percent cite other reasons summarized in Figure 3-7. Half of the firms that do not apply for credit cite the complexity of application procedures as motivating their decision. Small and medium-size firms are particularly affected (44 and 42 percent, respectively), while large firms are less affected (8 percent). This complexity also affects young firms and firms owned by one person, as well as Ivorian firms (46 percent versus 16 percent of foreign firms). 36 TABLE 3-8: PERCENT OF FIRMS FIGURE 3-7: REASONS WHY FIRMS NEEDING CREDIT APPLYING FOR CREDIT DO NOT APPLY Firm characteristics % of firms Manufacturing 7 Services 16 Other 15 Residual 5 Expect rejection 2 Micro 10 Loan is of insufficient size or 9 maturity Foreign 8 Unfavorable interest rates 15 Ivorian 10 Collateral requirements Large 28 17 too high Medium 6 Small 9 Abidjan 10 San Pédro 8 Application too complex 50 Yamoussoukro 0 Subsidiary 6 Non-subsidiary 9 Percent of firms Source: Enterprise Surveys, World Bank. While it is true that firms in Côte d’Ivoire rarely apply for loans or lines of credit, those that do apply are more likely to be rejected (42 percent) than accepted (39 percent). The remaining 19 percent did not know the status of their application. As shown in Table 3-9, the rejection rate is higher for microenterprises and medium-size firms than for large ones, which is reflected in the low application rates seen before. Finally, it is extremely surprising that firms with externally audited financial statements are often rejected. This counterintuitive finding suggests severe anomalies in the functioning of financial markets in Côte d’Ivoire or in the reliability of external auditing. Overall, more than two-thirds of applications are rejected because firms cannot provide guarantees or are insufficiently profitable (Figure 3-8). Insufficient profitability is the most common reason for rejecting applications of firms operating in the manufacturing and service sector, while the lack of guarantees is often the alleged reason for rejecting microenterprise applications. Medium-size and older firms are often rejected because of high levels of indebtedness, while young firms are not profitable enough. Financing requests of exporting firms are frequently refused due to problematic credit histories and domestically oriented firms and non-subsidiary firms are deemed insufficiently profitable. 37 TABLE 3-9: CHARACTERISTICS OF FIRMS REJECTED FOR CREDIT Firm characteristics % of firms Firm characteristics % of firms Manufacturing 36 New business 44 Services 27 Existing business 22 Residual 27 Micro 56 Payments for security 69 No payments 27 Large 30 Medium 47 International certification 6 Small 43 No international certification 21 Subsidiary 31 Audited financial statements 78 Non-subsidiary 43 Not audited 28 Source: Enterprise Survey, World Bank. FIGURE 3-8: REASONS GIVEN FOR REJECTING OF APPLICATION FOR CREDIT Already highly Other indebted 36 35 2 21 3 2 Unacceptable Insufficient Poor credit Unknown collateral profitability history Source: Enterprise Survey, World Bank. Percent of firms In sum, the banking system responds to the financing needs of Ivorian firms only to a very limited extent. Firms rarely rely on the banking system because they feel uneasy with the application process, seldom apply and are often rejected – often due to lack of sufficient guarantees or insufficient profitability. This may explain why access to finance is seen as the main obstacle to the development of the private sector in Côte d’Ivoire. ACCESS TO BANK FINANCING Few firms access bank financing in Côte d’Ivoire. Only 16.5 percent of enterprises declare having loans, lines of credit, or overdraft facilities, while 59 percent have a checking or savings account. Access to a line of credit or a loan from a bank is limited to 1 and 5 percent, respectively. These figures are much lower than the average of Sub-Saharan African countries. However, Table 3-10 shows that access to finance varies widely across firms on a number of dimensions. Enterprises operating in the service sector, for example, are twice as likely as manufacturing firms and eight times more likely than microenterprises to obtain bank financing. Large enterprises are more than four times as likely as small firms to have access to financing, bested only by exporters, which are eight times more likely than small firms to have access. 38 Table 3-10 also shows that about 13 percent of enterprises in Côte d’Ivoire benefit from overdraft facilities. Such facilities are a potentially important source of financing for firms that would otherwise find it difficult to access lines of credit for loans. Overdraft facilities are also more flexible with repayment than traditional loans and transaction costs typically are not too onerous. Unfortunately, overdraft facilities appear to be granted to the same types of firms that can access loans and lines of credit, and thus do not actually provide a true alternative to traditional financing for most firms. TABLE 3-10: SHARE OF FIRMS WITH ACCESS TO LINES OF CREDIT/LOANS AND OVERDRAFT FACILITIES Lines of Overdraft Lines of Overdraft credit/loans (% of Firm credit/loans (% of Firm characteristics (% of firms) firms) characteristics (% of firms) firms) Manufacturing 8 16 Abidjan 7 13 Services 16 31 San Pédro 1 16 Residual 11 11 Yamoussoukro 0 12 Micro 2 10 Subsidiary 8 21 Large 27 52 Non-subsidiary 6 13 Medium 6 11 Small 6 13 Exporter 43 66 Non-exporter 6 13 Source: Enterprise Survey, World Bank Firms’ vastly differentiated access to bank financing is the result of a complex interaction of various factors, but effective policies depend on understanding their relative importance. Here, we use a probit model to estimate the probability that a firm has access to bank financing, as a function of a number of firm characteristics. The dependent variable is equal to one if the enterprise has a loan, a line of credit or an overdraft facility, and zero otherwise. The estimated coefficients are presented in Table 3-11. These results confirm that firm size is positively and significantly correlated with access to finance, as is a firm’s subsidiary status and having audited financial statements. 39 TABLE 3-11: PROBIT ESTIMATE OF DETERMINANTS OF ACCESS TO BANK FINANCING Dependent variable: Access to bank financing Constant -0.978 (-8.35)*** Sector: Manufacturing versus residual -0.099 (-0.65) Service versus residual 0.224 (1.31) Size: Medium versus small 0.306 (1.98)** Large versus small 1.087 (3.74)*** Group 0.647 (2.84)*** Region: San Pedro versus Abidjan -0.048 (-0.18) Foreign ownership -0.135 (-0.68) Exporter 0.251 (0.75) Age of firm 0.008 (1.29) Certification by external auditor 0.467 (2.63)*** Land ownership -0.165 (-1.13) International quality certification -0.132 (-0.40) Observations 490 Note: Standard errors corrected for heteroskedasticity à la White; Z-statistics in parentheses; Statistical significance: * at 10% level, ** at 5% level, *** at 1% level. Source: Enterprise Survey, World Bank. CHARACTERISTICS OF LINES OF CREDIT, LOANS AND OVERDRAFTS Lines of credit and loans are often granted by TABLE 3-12: SHARE OF CREDIT MARKET, BY private banks, while less than one third of the time by TYPE OF INSTITUTION non-bank financial institutions. Table 3-12 illustrates Share of credit (%) the dominance of commercial banks in the supply of Private commercial banks 67.7 credit over state-owned banks and non-bank financial Non-bank financial institutions 30.9 institutions. Comparing the institutions providing State-owned banks 0.9 credits in Côte d’Ivoire and in a neighboring country like Ghana suggests that the extremely limited role Other 0.5 played by state-owned banks is a peculiarity of Côte Source: Enterprise Survey, World Bank d’Ivoire, while the leading role played by private banks is common. Nominal interest rates on lines of credits and loans are about 15 percent on average (Error! eference source not found.), slightly above the average in Sub-Saharan Africa. However, this is less than interest rates in other countries like Ghana, and as shown previously in Figure 3-7, firms in Côte d’Ivoire do not perceive high interest rates to be the main reason not to apply for credit. The average interest rate paid varies widely with firm characteristics (Table 3-14). Firms in the service 40 sector and firms that export or are subsidiaries tend to enjoy lower interest rates. Small enterprises, often with little collateral and limited bargaining power, on average pay almost double the rate enjoyed by large firms. One third of enterprises state that they do not know the maturity of their loan, while the duration is not specified for about 14 percent. For firms that do know the maturity of their loan, most are either two months (38 percent) or 12 months (42 percent). This is strikingly short when compared to averages of six and 11 years in Mali and Senegal, respectively. TABLE 3-13: INTEREST RATES FOR LINES OF CREDIT/LOANS AND OVERDRAFTS Credit Interest Rate lines/loans Overdrafts 1 to 5 4% 24% 6 to 10 13% 20% 11 to 15 55% 35% 16 to 20 7% 18% 21 to 25 21% 1% Greater than 25 - 2% Source: Enterprise Survey, World Bank TABLE 3-14: AVERAGE INTEREST RATE, BY TYPE OF FIRM Average interest rate (%) Average interest rate (%) Manufacturing 11 Exporter 11 Services 9 Non-exporter 16 Residual 17 Micro Land owning 9 Non-land owning 19 Large 10 International quality Medium 12 certification 10 Small 17 Without quality certification 16 Group 13 Certification by external auditor 11 Not group 16 No external auditor 17 Source: Enterprise Survey, World Bank Most credit requires collateral, since the legal and judicial environment limits access to credit and offers few guarantees for banks willing to finance SMEs. However, collateral is required less systematically than in other countries in the region, like Burkina Faso or Mali. At the sectoral level, there are stark disparities: some collateral is always asked of micro-enterprises, while three quarters of manufacturing and only one quarter of service firms are required to provide collateral. The value of collateral is often 100 percent of the value of the credit, and 73 percent on average. But this is rather favorable compared to much larger collateral requirements in other Sub-Saharan countries like Togo or Cameroon. Commercial banks’ practice of accepting only land or property as collateral is not conducive to financing SMEs, however, which generally do not have these types of assets. Accounts receivables and inventory are used by 78 percent of enterprises as collateral. Land 41 and buildings owned by the establishment (26 percent) and machinery and equipment (22 percent) are also used. Personal assets of the owner are used to guarantee 8 percent of loans. The nominal interest rate of overdraft facilities in Côte d’Ivoire is about 12 percent. However, interest rates vary considerably (Error! Reference source not found.), and depend on firm haracteristics. Firms operating in the service sector pay lower interest rates than manufacturing firms and microenterprises. Establishments that at least partly own the land they occupy enjoy relatively low interest rates as well. Interestingly, there is no clear relationship between the size of a firm (either in terms of employees or sales) and interest rate it pays for overdrafts. ACCESS TO TRADE CREDIT Purchases on credit from suppliers and advances TABLE 3-15: ACCESS TO TRADE CREDIT, BY from customers may be an alternative source of TYPE OF FIRM financing for enterprises that do not have access to % of firms bank credit or overdraft facilities. Although trade Manufacturing 50 credits are an oft-neglected component of firm Services 67 financing, they may prove very useful in a context of rationed bank credit. Moreover, trade credits are Residual 42 sometimes used to signal firm quality as a borrower Micro 57 and may improve a firm’s ability to subsequently obtain bank credits. Large 79 Medium 48 In Côte d’Ivoire, 54 percent of firms receive some Small 54 trade credits, making them the most widespread source of external financing for firms. In particular, 27 percent of enterprises purchase on credit from Group 63 suppliers and 42 percent benefit from advances from Not group 53 customers. However, the absolute value of trade credits remains limited, representing between 3 and Exporter 43 4 percent of firm financing on average. As already Non exporter 54 noted, the share of trade credits in sources of financing varies greatly across countries in Sub- Saharan Africa, from 10 percent in Cameroon to less Land ownership 61 than 1 percent in Mali. The contribution in Côte No land ownership 51 d’Ivoire is similar to that in South Africa and above the average in Sub-Saharan Africa. International quality certification 89 Trade credits are more often available than other Without quality certification 48 forms of credit to firms that otherwise have very limited access to bank financing, such as Certification by external auditor 73 microenterprises. However, as with other forms of No external auditor 51 credit, access is not evenly distributed across different types of firms (Table 3-15). Source: Enterprise Survey, World Bank. 42 SUMMARY OF FINDINGS Bank credit is a source of financing for only a limited number of firms in Côte d’Ivoire, while most firms rely on internal funds or retained earnings. The deficient role played by the Ivorian banking system may be interpreted as the result of two phenomena. First, the rejection rate of applications is higher than the rate of accepted applications, suggesting credit rationing by banks. Second, many firms that need financing don’t apply, resulting in self-rationing. Indeed, the large majority of enterprises voluntary exclude themselves from the formal financial market, most often because they believe that application procedures are too complex or that collateral requirements are too high. Only 6 percent of firms have a loan or a line of credit at their disposal and the latter are often relatively expensive and short-term. Moreover, only 13 percent of enterprises have overdraft facilities. Access to bank financing greatly varies with firm characteristics. Loans, lines of credit and overdraft facilities tend to be granted to enterprises with similar characteristics and they are very hard to access for other types of firms. Thus, firms are unable to substitute one form of external financing for another. Bank credits are most often provided to large firms, subsidiaries of larger firms, service-sector firms and export-oriented firms. Other enterprises have very limited access to finance and trade credits are often their sole form of external financing. The use of trade credits is widespread, but it accounts for only a limited share of total financing. 43 Chapter 4 – INFORMAL SECTOR In this section we examine microenterprises and informality in Côte d’Ivoire, focusing on firm characteristics, performance and potential. In particular, we consider the relationship between legal status and productivity in Côte d’Ivoire as compared to other countries. To investigate these issues we analyze the Microenterprise Survey for Côte d’Ivoire, which covers 90 firms registered with the Chambre des Metiers et de l’Artisanat, the relevant Chamber for firms with less than 10 employees. We limit our sample to firms with five or fewer employees – a commonly applied threshold for defining microenterprises – which leaves a sample of 80 firms. We must then categorize firms as formal or informal. Informality is typically defined according to a number of criteria, including the number of employees; registration status, whether for a business name, tax purposes, an operating license or association with a chamber; non-compliance of employment contracts with standard labor, tax, social security or benefits legislation; and the location of the business, whether fixed or mobile, or run out of a home. The Enterprise Survey data contain information on proxies for legal status centering on firm registration. We categorize firms based on registration for taxes, classifying 45 percent of our sample as formal. More than 90 percent of these same firms also satisfy the other formality criteria described in Table 4-1. TABLE 4-1: ENTERPRISE SURVEY CRITERIA FOR FIRM FORMALITY % of Criterion sample Registered for taxes or obtained a tax identification number 45 Registered business name with appropriate government institution 47 Registered with any government agency charged with commercial registration 49 Obtained operating, trade or other business license 77 CHARACTERISTICS OF FORMAL AND INFORMAL FIRMS In Table 4-2 we outline the distinguishing characteristics of both formal and informal microenterprises.11 The data suggest that formal firms tend to be larger, with 63 percent of the formal enterprises employing three to five permanent full time workers as compared to only 38 percent of informal firms. The educational attainment of firms’ top manager is also strikingly different for formal firms, where for 50 percent of firms, the top manager has a secondary education or better. The corresponding level of educational attainment in informal firms is 38 percent. Formal firms are also more likely to have been started to take advantage of a business opportunity (77 percent vs. 62 percent) rather than simply as a response to poor employment prospects. Finally, 81 percent of formal businesses are located in permanent or non-moveable 11 We note again that these descriptive statistics do not represent all informal microenterprises in Côte d’Ivoire, but rather only those registered with the Chambre des Metiers et de l’Artisanat. 44 premises, whereas only 67 percent of informal firms are. Legal status does not vary significantly with business sector, management experience or female participation. TABLE 4-2: CHARACTERISTICS OF MICROENTERPRISES Informal (%) Formal Both (%) (%) Size of Firm (permanent full-time employees, including managers): 1 employee 22 5 15 2 employees 40 32 37 3-5 employees 38 63 49 Full time paid female employees No female employees 66 70 68 1 female employee 18 14 16 2 female employees 11 11 11 3-5 female employees 5 3 4 Sector Manufacturing 45 43 44 Services 42 38 40 Other (IT, hotels, construction, transportation) 13 19 16 Educational attainment of top Manager: Less than secondary 68 50 61 Secondary and vocational 23 28 25 University or higher 9 22 14 Ownership and management of firm Ownership is all or majority female 18 16 17 Motivation to start the business Take advantage of a business opportunity 62 77 69 Jobs and opportunities were absent or unsatisfactory 38 23 31 Years of experience of top manager 10 11 10 Business premises Permanent or non moveable 67 81 73 PRODUCTIVITY COMPARISONS In this section we examine the differences in productivity levels between formal and informal micro enterprises. We employ two proxies for productivity. The first and more accurate measure is valued added per worker, with value added computed as total annual sales in 2007 less total annual cost of raw materials, electricity and immediate goods used in production. Labor costs and costs that do not go into the production of the good (e.g. communications) are excluded. In order to avoid reducing the number of observations, and to account for concerns over the reliability of cost data, we adopt sales per employee as a second measure of labor productivity. Figure 4-1 demonstrates that mean and median productivity in the sample using value added per employee is considerably higher for formal microenterprises than for their informal counterparts. 45 FIGURE 4-1: LABOR PRODUCTIVITY IN FORMAL AND INFORMAL FIRMS Mean value added/employee Median value added/employee Informal Mean sales/employee Median sales/employee Formal 0 2000 4000 6000 Labor productivity (US$) In Figure 4-2 and Figure 4-3 we investigate the dispersion of productivity around the median firm. The kernel density estimate of labor productivity in formal firms is to the right of that of informal firms, confirming that formal microenterprises are more productive. FIGURE 4-2: KERNEL DENSITY ESTIMATE OF LOG VALUE ADDED PER WORKER Micro informal Micro formal FIGURE 4-3: KERNEL DENSITY ESTIMATE OF LOG SALES PER WORKER Micro informal Micro formal 46 While these data support a positive relationship between productivity and formality, we recognize that this may be caused by other characteristics correlated with legal status. To address this concern, in Table 4-3 we report the results of regressing productivity on variables likely to be associated with firm performance including managerial education level, sector, gender composition, startup motivation, and type of firm premises. This confirms that formality is significantly correlated with higher levels of firm productivity measured as value added per worker. This positive relationship between formality and firm performance is robust to controls for access to services like electricity, telephone, sewage and water. It is also robust to using sales per worker as an alternative measure of productivity (results not reported for space reasons). Finally, we find that manager university education is significantly and positively correlated with productivity. TABLE 4-3: CORRELATES OF FIRM PRODUCTIVITY Dependent Variable: Log Productivity (Value added in thousands divided by number of workers) (1) (2) (3) (4) (5) (6) Firm manager education: Secondary or vocational 0.152 0.332 0.579 0.306 0.257 0.330 (0.44) (0.87) (1.48) (0.88) (0.74) (0.95) University or higher 0.816 0.876 1.694 0.949 0.953 0.560 (1.69)* (2.14)** (2.82)*** (1.91)* (2.00)* (1.35) Formal 0.909 0.816 0.915 0.916 0.641 0.596 (2.84)*** (2.73)*** (2.65)** (2.83)*** (2.13)** (1.91)* Manufacturing sector 0.129 (0.44) Number of employees -0.089 (0.64) Majority female ownership 0.049 (0.10) Business started because opportunity 0.393 (1.29) Business location is permanent/fixed 0.615 0.780 0.680 0.603 0.599 (1.55) (1.68)* (1.51) (1.41) (1.47) Connection to Services: Electricity -0.357 (1.03) Water -0.113 (0.36) Sewage 0.507 (1.66) Telephone 0.615 (2.12)** Observations 63 58 49 60 60 57 R-squared 0.20 0.25 0.35 0.23 0.26 0.24 Robust t-statistics in parentheses, * significant at 10%; ** significant at 5%; *** significant at 1%. Notes: Reference groups for education, sector and reason for startup are the following: Education: firms where the top manager has less than secondary education. Sector: firms in services, IT, hotels, wholesale, and construction, among others. Reason for startup: “jobs and opportunities were absent or non satisfactory.� 47 FORMALITY VS. INFORMALITY Figure 4-4 displays the difference between formal and informal microenterprises’ access to basic infrastructure services. While formal firms are only slightly more likely to have access to electricity, they are twice as likely to have access to water, sewage and telephone service. There is less difference with regard to access to finance. Checking accounts are popular with both types of firms, and both have difficulty obtaining overdraft facilities and lines of credit. That said, formal microenterprises are more likely to have checking accounts and have better access to overdraft facilities. FIGURE 4-4: ACCESS TO INFRASTRUCTURE SERVICES, BY LEGAL STATUS Water Telephone Sewage Overdraft Electricity Formal Credit Informal Checking account 0% 25% 50% 75% 100% Share of firms with access While the choice of formality may be attributed to benefits associated with access to infrastructure and financial services, the choice of informality may be triggered by the the regulatory burden of tax compliance or the low cost of informality due to the weak rule of law. Table 4-4 illustrates the high cost of formality in Côte d’Ivoire. The financial and administrative burdens of tax compliance are considered greater constraints to doing business by informal firms than by formal ones. This table also illustrates the potential costs of informality. Visits from tax inspectors are one indicator of enforcement of tax laws. Thirty-one and 62 percent of informal and formal firms, respectively, reported such a visit, but only 20 percent of formal firms reported an expectation of bribes (as compared to 30 percent of informal firms). TABLE 4-4: COSTS OF FORMALITY AND INFORMALITY Informal Formal Both (% of firms) (% of firms) (% of firms) Financial burden of tax compliance is a major or severe obstacle to doing business 86 72 79 Administrative burden of tax compliance is a major or severe obstacle 73 56 65 Visited by tax inspector in the last year 31 62 45 Gift or informal payment was expected or requested 30 20 24 Source: Enterprise Survey, World Bank Table 4-5 reports a more thorough investigation of the correlates of firms’ decisions to formalize. This confirms that access to services and having a bank account are positively and significantly associated with formality. 48 TABLE 4-5: CORRELATES OF FORMALITY Table 5: Correlates of Formality Probit Dependent Variable: Dummy Variable for being Formal (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Firm Manager Education: Secondary or vocational 0.337 0.318 0.166 0.225 0.220 0.301 0.288 0.393 0.196 0.244 (0.95) (0.87) (0.43) (0.58) (0.56) (0.83) (0.80) (0.61) (0.52) (0.65) Attended univerisity or higher 0.650 0.533 1.101 0.839 0.600 0.347 0.482 0.457 0.454 0.275 (1.50) (1.12) (2.22)** (1.91)* (1.44) (0.81) (1.02) (0.86) (1.01) (0.53) Manufacturing -0.126 (0.41) Size of firm (employment) 0.255 0.238 0.128 0.083 0.165 0.191 -0.114 0.207 0.160 (1.93)* (1.63) (0.98) (0.62) (1.22) (1.42) (0.42) (1.56) (1.18) Majority female ownership -0.092 (0.21) Start business because opportunity 0.257 (0.77) Premise location is permanent/fixed 0.333 0.541 0.034 0.059 0.339 0.277 0.409 0.336 0.417 (0.93) (1.38) (0.08) (0.14) (0.87) (0.77) (0.64) (0.92) (1.12) Connection to Services: Electricity -0.248 (0.67) Water 1.010 (2.90)*** Sewage 1.153 (3.33)*** Telephone 0.844 (2.57)** Tax Inspector Visit 0.441 (1.32) Bribe Payment -0.416 (0.73) Bank Account 0.587 (1.83)* Overdraft 1.140 (1.61) Observations 76 71 60 72 72 69 73 31 73 72 Robust t-statistics in parentheses, * significant at 10%; ** significant at 5%; *** significant at 1%. Notes: Reference groups for education, sector and reason for startup are the following: Education: firms where the top manager has less than secondary education. Sector: firms in services, IT, hotels, wholesale, and construction, among others. Reason for startup: “jobs and opportunities were absent or non satisfactory.� FORMALITY VS. INFORMALITY: A CROSS-COUNTRY COMPARISON The choice of comparator countries used in this section is motivated by Gelb et al. (2009). Using data from microenterprise surveys in Southern Africa (South Africa, Namibia, and Botswana) and Eastern Africa (Kenya, Uganda, and Tanzania) Gelb et al. investigate the drivers of the choice to formalize and examine how informal firms should be viewed, whether as sources of future growth and employment or as “survivalist� employment alternatives. Similar to Figure 4-2 and Figure 4-3, they plot kernel density estimates of labor productivity across formal and informal microenterprises in the countries they study. They find that labor productivity is lower in informal firms on average in every country they study, and that the kernel density function of labor productivity in informal firms overlaps that of formal firms. However, the degree of overlap varied significantly across countries, and was much smaller in Southern Africa than in Eastern Africa. 49 Gelb et al. hypothesize that the smaller extent of overlap in labor productivity across the formal- informal enterprises in Southern Africa is due to better enforcement of tax laws and regulations, as well as better access to services for formal firms. Informal firms are more likely to be “survivalist�. That is, owners of informal firms may be better off entering the labor market as wage labor, and would benefit from vocational training. In Eastern Africa, where tax enforcement is weak and informality has little impact on access to services, there are fewer incentives to formalization. Furthermore, informal and formal firms are just as likely to increase productivity and lower costs as they grow. Policies to increase productivity and lower the cost of formality may be helpful in this context. Comparing our findings for Côte d’Ivoire with those in Gelb et al., Côte d’Ivoire seems to fit the Southern Africa pattern better. The kernel density estimates in Figure 4-2 and Figure 4-3 for labor productivity show less overlap between formal and informal firms. Furthermore, the benefits of formality, as indicated by access to basic services, show that formal firms enjoy greater access than their informal counterparts. With respect to the costs associated with informality, informal microenterprises in Côte d’Ivoire face bribe requests from tax inspectors slightly more often than formal firms, but as in other countries they face fewer visits. Overall, visits in Côte d’Ivoire are as likely as in Southern Africa (62 percent of formal firms) but less likely than in Eastern Africa where visits are reported by 77, 86 and 83 percent of firms in Kenya, Tanzania and Uganda, respectively. SUMMARY OF FINDINGS In this chapter we have examined the characteristics and performance patterns of informal and formal microenterprises in Côte d’Ivoire. Formal microenterprises tend to have higher labor productivity and better access to infrastructure and financial services than informal firms. Informal firms are slightly more vulnerable to bribe-taking officials, though less so than peers in other countries. Taken together, this indicates some degree of regulatory enforcement, and thus reduces the disincentive to formalization. The conclusion that emerges is in line with Gelb et al. for the Southern Africa sub-sample. That is, the informal sector in Côte d’Ivoire is largely comprised of “survivalist� firms as opposed to enterprises that could take advantage of the expanded opportunities that could be expected with formalization. Less productive firms tend to be informal, suggesting that owners of such firms could be better off entering the labor market as wage labor. One way to help informal firms in Côte d’Ivoire would be to invest in education or vocational training, as we find that higher levels of productivity are associated with higher educational attainment of top firm management (Table 4-3). In light of the low educational status of most informal entrepreneurs suggested by the data, it is difficult to envision a route to higher- productivity employment that does not include building human capital. 50 Chapter 5 – LABOR MARKET A well-functioning labor market is key to good economic growth and development, sustainable improvements in living standards and firm performance. But firm managers in Côte d’Ivoire are concerned with constraints on growth related to labor regulations and an inadequately educated workforce. In this chapter we discuss the relative importance of institutional rigidities in the labor market, explore key worker characteristics and examine wage-setting behavior in Côte d’Ivoire. CHARACTERISTICS OF THE FORMAL LABOR MARKET The labor force of Côte d’Ivoire was comprised of about 7 million people in 2008. The agricultural sector remains the main supplier of jobs, accounting for an estimated 68 percent of the workforce in 2007. The formal or “modern� labor market – wage employment in the public or private sector – in 2005 represented only 5 percent of the labor market. But these jobs generate relatively high incomes, with typical production workers in manufacturing earning roughly Côte d’Ivoire’s annual per capita GDP. The region of Abidjan is home to 25 percent of the population as well as the majority of opportunities in the modern sector. Political turmoil, civil war and the partition of country that followed have considerably encouraged the informal sector, and disrupted formal means of exchange. Armed conflict, coupled with the need for continued exchange between the two partitioned zones, strengthened opportunities for lucrative informal activities. But besides the effect of these political forces on the Ivorian formal sector, institutional aspects of the labor market – including labor contracts, trade unions, regulations and job-seeking practices – can affect the functioning of the formal market. Labor contracts. As in other comparable countries,12 the firm survey data suggest that more than 80 percent of the full-time workers of formal firms have permanent worker status.13 This large proportion of permanent workers can hamper firm performance given the lack of flexibility to react to changing economic conditions that this implies. Trade unions. Since independence, trade unions have been engaged in politics and currently have an important role in Côte d’Ivoire’s political landscape.14 Félix Houphouët-Boigny, the first president of the republic, was originally a trade union leader and had a major role in Côte d’Ivoire’s independence process. More recently, Guillaume Soro, a former student union leader (FESCI), took part in the military coup of 1999 and led the rebel Patriotic Movement of Côte d'Ivoire (MPCI) in a September 2002 rebellion against the government. Following a peace deal signed in March 2007 he 12In Kenya and Ghana, permanent full-time employees account for 80 and 87.2 percent, respectively, of the full-time manufacturing workforce. In manufacturing, retail and other sectors, the share of permanent, full-time workers among all full-time 13 workers is 81.5, 83.3 and 86.1 percent, respectively. 14The main trade unions are the Union Générale des Travailleurs de Côte d’Ivoire (UGTCI), Confédération des Syndicats Libres de Côte d’Ivoire (Dignité), and the Fédération des Syndicats Autonomes de Côte d’Ivoire. The UGTCI was the sole authorized union from independence until the 1990s, when political reforms permitted the formation of the two other unions. 51 was appointed prime minister. The relationship between trade unions, employers and political authorities has been extremely strained since September 2002.15 In 2009, some 15 percent of surveyed manufacturing firms indicated that some of their employees belonged to a union. However, this proportion drops to only 4 percent in retail sector and other sectors of the economy (Table 5-1). This is well below Kenya’s 51 percent or in Ghana’s 52 percent. In nearly all sectors, the presence of trade unions increases with firm size. Only 13.3 percent of firms with between five and 20 employees indicate that some belong to a trade union, whereas the proportion in large firms (greater than 100 employees) is 55.5 percent. In every sector, trade unions are more common in foreign-owned firms and exporting firms. Taking only those manufacturing firms reporting the presence of at least one trade union, membership is quite high, about 55 percent of workers. This proportion reaches nearly 94 percent in large manufacturing firms. Among manufacturing and retail firms, membership is higher in foreign-owned firms. TABLE 5-1: TRADE UNION PRESENCE, BY TYPE OF FIRM Share of firms whose employees belong to a union Manufacturing Sector Retail Sector Residual By Size Small 13% 2% 2% Medium 9% 29% 5% Large 56% 77% 78% By Ownership Private Foreign Firms 23% 6% 11% Private Domestic Firms 14% 3% 3% By Exporting Status Exporting Firms 69% 50% 29% Non-Exporting Firms 9% 4% 4% Share of employees belonging to a union Manufacturing Sector Retail Sector Residual By Size Small 43% 43% 43% Medium 70% 73% 86% Large 94% 58% 72% By Ownership Private Foreign Firms 67% 63% 63% Private Domestic Firms 52% 38% 63% By Exporting Status Exporting Firms 84% 50% 61% Non-Exporting Firms 32% 56% 64% Source: Enterprise Survey, World Bank Note: Weighted data. Regulation. The major source of labor regulation in Côte d’Ivoire is the New Labor Code of 1995, part of the modernization and recovery program for the Ivorian economy. This labor code liberalized recruitment of workers by ending the monopoly of the public placement office (Office de la Main d’oeuvre). It also simplified procedures for laying off workers for economic reasons and for negotiated breach of contract. At the same time, trade union rights were strengthened by 15According to the trade unions, only a small fraction of the workforce in Côte d’Ivoire is unionized, as persistent civil war hampers efforts to exert worker rights, particularly the right to organize and engage in collective bargaining. Collective bargaining agreements were signed in most of large firms and in the public sector, but application throughout the country has been stymied by prolonged instability. 52 establishing worker representatives in firms and enforcing trade union pluralism. All in all, the labor code strives to restore entrepreneurs’ decision-making capacity and the autonomy of social actors. As shown in Figure 5-1 and Figure 5-2, indicators of labor market regulation are generally favorable in Côte d’Ivoire. The rigidity of employment index is quite low in Côte d’Ivoire, and is less than the average in Sub-Saharan Africa. According to these measures, procedures to employ workers are not especially burdensome, and there are few limitations on hiring, working hours or firing procedures. The rigidity of employment is comparable to that of Mali and South Africa, and only restrictions on working hours are stronger than in other countries of the sub-region. Severance pay, while less than the regional average is higher than in most of the comparator countries – Ghana’s severance pay is particularly generous and pulls up the average. FIGURE 5-1: INTERNATIONAL COMPARISON OF FIGURE 5-2: INTERNATIONAL COMPARISON OF RIGIDITY OF EMPLOYMENT SEVERANCE PAY Senegal Ghana Madagascar Mean Togo Côte d'Ivoire Cameroon Kenya Mean Senegal South Africa Togo Côte d'Ivoire Burkina Faso Mali Cameroon Ghana Mali Burkina Faso Madagascar Kenya South Africa 0 20 40 60 80 0 50 100 150 200 Rigidity of employment index Severance pay (weeks of salary) Source: Doing Business Note: In Figure 5-1Figure 5-2, higher values indicate more rigid employment regulation, with a value of 100 indicating maximum rigidity. Consistent with the Doing Business indicators, the Enterprise Survey data indicate that labor market regulations are relatively less cumbersome than other aspects of the business environment. Fewer than 8 percent of managers consider labor regulations a “major� or “severe� constraint on firm growth, and not one says it is the most significant obstacle. However, these perceptions vary across sectors. Labor regulations are problematic for 15 percent of managers in the manufacturing sector, 2 percent in services, 10 percent in microenterprises, and 4 percent in other sectors. However, firm managers’ relative lack of concern should be considered together with other information from the survey. Based on perceptions data, labor regulations seem fairly unimportant compared to the situation with access to finance or political instability. 53 Labor market information. A well-functioning labor market depends in part on the circulation of information about job opportunities. In Côte d’Ivoire, finding jobs through public announcements or advertisements is very uncommon. Networks of family and friends are more effective, and 56 percent of manufacturing firms use such networks. This proportion is even larger when we consider other sectors of the economy. Public and private placement offices play a very minor role. While foreign-owned firms are more likely to use formal recruitment methods, they also rely heavily on informal methods. These practices may be understood as consequences of informational deficiencies in the labor market. In an economy characterized by an abundant labor supply and difficulties evaluating the real qualifications of job seekers, firms rely on friends, current employees or family ties to collect useful information before hiring. However, informal networks may provide imperfect information as well. EDUCATION, TRAINING, HEALTH AND WAGES OF FORMAL SECTOR EMPLOYEES In following section we analyze the main characteristics of formal sector employees in Côte d’Ivoire. Data are derived from interviews with managers and supervisors and detailed accounting data of the surveyed firms. These characteristics are important explanatory elements to workers’ productivity and firm performance. EMPLOYEE EDUCATION AND TRAINING. The education sector has suffered from neglect in Côte d’Ivoire. The net enrollment rate in primary school dropped from 58.3 percent in 1997 to 56 percent in 2005, while the drop in secondary school enrollment was extremely pronounced over the same period, from 34.1 percent to 20 percent. Class size and the quality of teaching have worsened, resulting in decreasing school achievement rates. Planned improvements, consisting mainly of building new schools, were frozen because of the armed conflict. This conflict has worsened the overall education situation particularly in the north of the country where many schools were closed for months, and only opened again in part thanks to volunteer teachers. Since the beginning of the conflict, public expenditures on education have dropped to make room for military and humanitarian expenditures. University education has also suffered from the political crises and armed conflict. The student union (FESCI), which controls the majority of universities in Abidjan, has been associated with a climate of terror on campuses since 1999. Indeed, this student union has played an important role in the political events of the country and was associated with a number of student murders, as well as demonstrations and riots. Figure 5-3 compares educational attainment in Côte d’Ivoire to that of other countries in Sub- Saharan Africa for which firm survey data are available.16 Forty-one percent of the manufacturing workforce has between seven and 12 years of education, while a relatively large proportion has no education, comparable to Senegal. 16 Firm managers gave the average education level reached by a typical production worker. 54 FIGURE 5-3: AVERAGE EDUCATION LEVEL FOR TYPICAL PRODUCTION WORKER 0 - 3 years 4 - 6 years 7 - 12 years 13 + years South Africa Senegal Mali Madagascar Kenya Ghana Côte d'Ivoire Cameroon Burkina Faso 0% 25% 50% 75% 100% Percent of workers Source: Enterprise Surveys, World Bank. Note: Exclusively for manufacturing firms. Weighted data. Table 5-2 shows that education levels vary across firm size, ownership and exporting status. Large exporting firms have a highly educated workforce, while domestic firms’ share of workers with more than 10 years of education (28 percent) is actually higher than that of foreign-owned firms (13 percent). Unsurprisingly, microenterprises have a less educated workforce than manufacturing firms (11 percent vs. 23 percent of workers have 10 or more years of education). This in line with manager perceptions of education as constraint for firms. Indeed, in the manufacturing sector 43 percent of managers indicate that it is a major or severe obstacle to the current operations of their establishment. In a situation where more than a fourth of the workforce has received no education and where university education is under-developed, professional training provided by firms to their employees is key to improving worker skills. Batra and Stone (2004) describe evidence which indicates that some kinds of external training may in fact help boost firm productivity. But fewer than 20 percent of manufacturing firms in Côte d’Ivoire provide formal training to their employees (Figure 5-4). This figure is lower than the rate in Burkina Faso or Mali, and half that of South Africa. The proportion of firms that provide formal training programs varies significantly from one economic sector to another. Fewer than 7 percent of firms operating in the garment industry have formal training programs, compared to 100 percent of firms in the electronics sector. When firms offer formal training, the proportion of employees benefiting from these programs in the manufacturing sector varies (Table 5-3). Foreign-owned firms tend to offer more formal training program than their domestic counterparts. However, foreign-owned firms tend to favor more training sessions for production workers whereas domestic firms place more emphasis on training non-production workers. These figures provide a summary of worker training schemes in the manufacturing sector, but they do not reveal why some firms offer training and others don’t. We report the results of multivariate analysis in Table 5-4. This indicates that training is correlated with higher educational attainment and as well as with export status. 55 TABLE 5-2: AVERAGE EDUCATIONAL ACHIEVEMENT IN MANUFACTURING FIRMS 0-3 years of 4-6 years of 7-9 years of 10-12 years of 13 + years of education education education education education Manufacturing Sector 27.7% 19.2% 22.9% 18.4% 4.8% By Sizeclass Small 30.6% 19.8% 16.7% 19.8% 5.5% Medium 24.8% 20.1% 37.9% 14.8% 2.6% Large 6.3% 12,23% 55.4% 21.1% 6.3% By Ownership Private Foreign Firms 22.4% 33.8% 30.0% 7.2% 6.1% Private Domestic Firms 28.0% 16.2% 22.1% 22.8% 5.2% By Exporting Status Exporting Firms 0.0% 8.1% 38/2% 17.7% 39.1% Non-Exporting Firms 30.6% 20.4% 21.3% 18.5% 1.3% Micro Entreprises 28.8% 28.0% 29.9% 9.3% 1.8% By Ownership Private Foreign Firms 28.8% 33.3% 33.3% 6.5% 0.0% Private Domestic Firms 29.0% 26.4% 29.5% 9.9% 2.1% By Exporting Status Exporting Firms 0.0% 50.0% 0.0% 50.0% 0.0% Non-Exporting Firms 29.7% 27.8% 29.8% 8.6% 1.8% Source: Enterprise Survey, World Bank. Weighted data. Note: Percentages refer to the share of firms of a particular type whose workers on average have a given level of educational achievement. FIGURE 5-4: INTERNATIONAL COMPARISON OF TRAINING OFFERINGS BY FIRMS Kenya South Africa Ghana Togo Burkina Faso Madagascar Cameroon Mali Senegal Côte d'Ivoire 0% 25% 50% 75% 100% Percent of firms offering training 56 TABLE 5-3: FORMAL TRAINING IN THE MANUFACTURING SECTOR Formal training offered by firms Private Foreign Firm Private Domestic Firm Total Firms offering formal training (%) 27.6% 18.3% 19.1% % of the workforce receiving training 38.1% 61.2% 43.0% % of the production workers receiving training 68.3% 48.6% 48.2% % of the non production workers receiving training 35.4% 81.0% 58.2% Source: Enterprise Survey, World Bank Note: Weighted data. TABLE 5-4: DETERMINANTS OF EMPLOYEE TRAINING PROGRAMS Dependent Variable Formal Training Formal Training Formal Training Intercept -6.624 -6.685 -6.415 (8.53)*** (8.01)*** (7.66)*** Firm Size Medium 0.217 0.205 0.178 (0.66) (0.62) (0.52) Large -0.090 -0.135 -0.129 (0.19) (0.27) (0.24) Human capital Manager has at least a tertiary education (dummy) 0.675 0.690 0.454 (1.79)* (1.82)* (1.10) Skilled labor force (dummy) 1.116 1.120 1.019 (3.02)*** (2.98)*** (2.67)*** Workforce education present a serious / severe 0.120 0.118 -0.024 obstacle to the current operations (dummy) (0.40) (0.39) (0.08) Weights -0.009 -0.009 -0.010 (1.18) (1.12) (1.10) Other Characteristics Foreign-owned establishment (dummy) -0.394 (1.08) Age of firm 0.019 0.006 (0.44) (0.13) Age of firm (square) -0.000 -0.000 (0.44) (0.12) Exporter (dummy) 0.990 (2.17)** Control for firm industry yes yes yes Control for firm location yes yes yes Number of observations 131 131 131 LR Chi2(21) 34.61 34.81 40.50 Prob>Chi2 0.0070 0.0147 0.0065 Pseudo R2 0.2276 0.2289 0.2663 Absolute value of z statistics in parentheses * significant at 10%; ** significant at 5%; *** significant at 1% 57 HEALTH STATUS The health status of the workers has consequences on their productivity. The situation in Côte d’Ivoire appears to be more worrying than in other African countries where data are available (Table 5-5). Indeed, almost 40 percent of firms face high absenteeism due to sickness. Concerning HIV/AIDS, firms in Côte d’Ivoire seem to be less affected than in Burkina Faso. However, more than 30 percent of the surveyed firms undertake diffusion of HIV prevention messages. Similarly, Ivoirian firms are fairly generous with regard to HIV/AIDS programs and activities. TABLE 5-5: INTERNATIONAL COMPARISON OF HEALTH-RELATED ABSENTEEISM AND COSTS Côte Burkina Ghana Kenya Madagascar Mali Senegal Cameroon Togo d'Ivoire Faso Percent of firms affected by absenteeism due to: Sickness 37 25 22 24 37 26 23 30 17 Care for sick family members or friends 14 20 15 16 34 18 18 20 14 HIV/AIDS 2 6 1 8 0 1 0 8 6 Care for family members or friends with HIV/AIDS 1 6 0 4 1 1 2 6 4 Percent of firms providing: HIV prevention messages 31 34 25 61 19 15 7 43 33 Free condoms 14 14 5 25 5 5 4 25 19 Anonymous HIV testing 10 8 4 7 2 6 2 26 20 Percent of firms imposing pre-employment health check: 16 31 17 - 72 20 30 46 40 Cost of all AIDS/HIV programs and activities $62 $162 $20 - $17 $14 $1 -- $292 Source: Enterprise Surveys 2003-2009, World Bank Note: Exclusively for manufacturing firms. Weighted data. EMPLOYEE PAY It is useful to examine the pattern of earnings of production workers. These workers earnings are the most homogenous of wage categories and thus fairly comparable across firms and countries. In addition, the wage level of production workers serves as a barometer for foreign firms when they decide whether or not to invest in a country. Data on remuneration presented here were provided by managers from the financial statements of their firm. Using these data we study the level and structure of production workers’ remuneration including benefits when applicable. Then, we analyze mechanisms used to determine wages at the firm level. Table 5-6 suggests that production workers’ average remuneration varies widely according to firm size, commercial status (exporting or non-exporting), sector and ownership. Descriptive analysis shows that large, exporting and foreign-owned firms pay their production workers more than other firms. 58 TABLE 5-6: MONTHLY CASH EARNINGS FOR PRODUCTION WORKERS Average monthly cash earnings for production workers (US$) Manufacturing sector 127 By Size Small 106 Medium 163 Large 318 By Ownership Private Foreign Firms 227 Private Domestic Firms 108 By Exporting Status Exporting Firms 295 Non-Exporting Firms 109 By sub-sectors Machinery and equipment 362 Electronics 305 Chemicals 301 Textiles 251 Plastics & rubber 198 Food 177 Basic metals 150 Non metallic mineral products 101 Fabricate metal products 99 Garments 76 Source: Enterprise Survey in Côte d'Ivoire 2009, World Bank Note: Exclusively for manufacturing firms. Weighted data. The data indicate that the average production worker’s remuneration in Côte d’Ivoire is moderate by regional standards. It lies between Burkina Faso and Mali, and well below South Africa but above Ghana. Note that average Ivoirian production worker pay is comparable to most other West African countries (Figure 5-5). FIGURE 5-5: INTERNATIONAL COMPARISON OF MANUFACTURING WORKERS’ MONTHLY CASH EARNINGS Madagascar South Africa Cameroon Kenya Senegal Burkina Faso Côte d'Ivoire Mali Ghana 0 250 500 750 Monthly cash earnings (US$) Source: Enterprise Surveys 2003-2009, World Bank. Note: Exclusively for manufacturing firms. Weighted data. 59 WAGE DETERMINATION Large earnings differentials across sectors in Côte d’Ivoire imply that the labor market is not competitive or is not fully integrated. Several characteristics of the labor market may contribute to this phenomenon, including a minimum wage17 and the balance of power between trade unions and firms. We investigate the determinants of wages by estimating wage equations, with results reported in Table 5-7: Determinants of wages: Firm-level estimat. TABLE 5-7: DETERMINANTS OF WAGES: FIRM-LEVEL ESTIMATES Dependant Variable : Log of average monthly (1) (2) (3) compensation per production worker Intercept 11.047 11.052 11.070 (19.79)*** (19.14)*** (19.18)*** Workforce Characteristics Manager has at least a tertiary education (dummy) 0.563 0.525 0.501 (2.71)*** (2.44)** (2.31)** Skilled labor force (dummy) 0.206 0.185 0.121 (1.02) (0.90) (0.57) Firms Characteristics Trade union presence (dummy) 0.200 0.153 0.132 (0.94) (0.68) (0.59) Medium 0.061 0.047 0.033 (0.36) (0.28) (0.20) Large 0.475 0.338 0.343 (1.90)* (1.27) (1.29) Exporter (dummy) 0.112 0.044 (0.45) (0.17) Foreign-owned establishment (dummy) 0.259 0.287 (1.41) (1.55) Age of firm 0.000 -0.001 (0.01) (0.07) Age of firm (square) 0.000 0.000 (0.24) (0.30) Offer formal training (dummy) 0.247 (1.40) Weights -0.011 -0.010 -0.010 (3.37)*** (3.12)*** (2.98)*** Economic Sectors yes yes yes Firms Localization yes yes yes Number of observations 138 138 137 R-squared 0.43 0.44 0.45 Absolute value of t statistics in parentheses * significant at 10%; ** significant at 5%; *** significant at 1% In the first specification (column 1) we regress the logarithm of the average monthly compensation per production worker on workforce characteristics, firm size and trade union presence. Our results correspond to the findings of Teal et al. (2006): Large firms pay their production workers significantly. However, the presence of trade unions does not affect average monthly wages, while the educational attainment of the firm’s manager is positively correlated with wages. Surprisingly, a skilled labor force – defined as an average of 12 or more years of education – does not result in higher wages. In the second specification (column 2) we introduce more detailed firm characteristics such as export status, nationality of owners and age of firms. None of these new variables is significant, but 17The minimum wage in Côte d’Ivoire was set after the FCFA devaluation in 1994 through tripartite negotiations between workers unions, employers and the government. In the industrial sectors the minimum monthly wage is 36,607 FCFA (74.3 $US). 60 their inclusion entirely dilutes the effect of firm size from the first specification. In our final specification (column 3) we add a variable reflecting the availability of training programs. This variable is not significant, and firm size remains insignificant. We note that the educational attainment of firm management appears to be the main determinant of production worker wages. There is a correlation between the percent of skilled workers as a proportion of a firm’s labor force and the educational attainment of the firm’s management. It for this likely explains why this coefficient was so large in the specification of the wage equation. . 61 Chapter 6 –INVESTMENT CLIMATE This chapter discusses areas of the investment climate indicated by managers to be of concern to them that are not fully covered in other chapters; namely political instability, corruption and regulation, and electricity. These are some of the top constraints rated and ranked as serious obstacles by enterprise managers but not directly covered in other chapters. According to the data collected from enterprises in Cote d’Ivoire, the investment climate reform agenda is large. INTRODUCTION In this chapter we will discuss firm perceptions of the overall investment climate, using subjective Enterprise Survey data to distinguish between laws on the books and reality on the ground. The Enterprise Survey asks firm managers to rate concerns about 15 areas of the investment climate according the degree to which each area is an obstacle to operations. Figure 6-1 reports the share of different types of firms that rate each area as serious obstacles. Concerns about access to finance and political instability stand out above the others. This comes as no surprise, given the political turmoil in the wake of the 1999 coup and the subsequent economic decline. The importance of these factors is robust to the size and capital intensity of the firms surveyed. Fifty percent of managers also said that electricity and crime are a serious concern. But 71 percent and 89 percent of managers considered access to finance and political instability, respectively, to be serious concerns. Access to credit. The banking sector has been relatively resilient for the duration of the crisis, but long-standing weaknesses have been exacerbated. Although the share of nonperforming loans fell from a peak of 26 percent of total loans in 2004 to 20 percent in 2006, the sector continues to face structural weaknesses as well as problems resulting from the ongoing crisis. In particular, compliance with prudential norms has deteriorated, and the sector has suffered from the increase in government arrears and the closure of bank branches in the north, center and west of the country. It is therefore not surprising that about 87 percent of new investments in manufacturing are financed through retained earnings. Chapter three goes into greater detail on access to finance. Corruption. About 68 percent of the surveyed firms consider corruption a "major" or "very severe" obstacle. This translates into a non-negligible cost of nearly 12 percent of firms’ total annual sales. Corruption and lack of transparency have been a problem for years in the cocoa and coffee regulatory bodies, government procurement, and the oil/energy sector. However, the severe economic downturn since the conflict erupted in 2002 has led to higher levels of corruption at all levels of public administration. Transparency International’s Corruption Perception Index (CPI), which measures perceptions of corruption by businesspeople and country analysts, showed a decline in the country’s ranking from 75th of 99 countries in 1999 to 150th of 180 countries in 2007. We do note however that the probability of being asked for a bribe is relatively low, as measured by the graft index (Figure 6-2). 62 FIGURE 6-1: FIRM PERCEPTIONS OF MOST SEVERE CONSTRAINTS access to finance political instability practices of competitors in the informa crime, theft and disorder corruption electricity access to land transport customs and trade regulations inadequately educated w orkforce 0 10 20 30 40 50 60 SMLE Microenterprises Crime, theft and disorder. More than 58 percent of firms consider crime, theft and disorder a "major" or "very severe" obstacle. This is clearly linked to the civil war. Light weapons were abundant before the civil war and are even more so today. Armed gangs in the west have abused civilians under the cover of the civil war, and there have been many reports of atrocities such as rape and mass murder. An estimated 53 percent of surveyed firms declare that their costs for security have increased in the last 10 years. More than one fourth experience losses as a result of theft, robbery, vandalism or arson. Access to land. There is pressure on land as clashes over land ownership have become more frequent over the years in the cocoa region, notably where cocoa workers of foreign origin have become landowners. Landholding laws have had an important role in fueling the civil war and ongoing political turmoil. Transportation. Although Côte d’Ivoire has the most developed road network in West Africa, transportation is considered by more than 38 percent of firms a "major" or "very severe" obstacle. Underinvestment since the 1980s has left many roads in poor condition, and very little road maintenance has taken place since 1999. Furthermore, the large number of checkpoints and the system of formal and informal fees that the national army, the New Forces and local militias have imposed have hampered transportation throughout the country. Tax rates and administration. Overall tax rates seem reasonable, according to the 2008 Doing Business indicators. The overall tax rate on profit is 45.5 percent in Côte d’Ivoire, compared to an average of 71.2 percent in Sub-Saharan Africa and 47.8 percent in OECD countries. However, the tax administration needs improvements. Despite the relatively low tax rates, more than 60 percent of firms consider tax rates and tax administration a "major" or "very severe" obstacle. But on average only about 47 percent of manufacturing firms’ annual sales are declared to tax authorities. 63 POLITICAL INSTABILITY FIGURE 6-2: INTERNATIONAL COMPARISON OF LIKELIHOOD OF BRIBE REQUESTS The political crisis since the first coup d’état in 1999 has seriously undermined government Namibia efforts to launch an effective economic recovery South Africa and development plan. Since May 2007, the Cape Verde government has developed an early program of Botswana reform measures aimed at restoring Zambia macroeconomic stability and advancing Rwanda structural reforms in key sectors like energy, Swaziland cocoa and finance. At the same time, and awaiting Malawi the PRSP II, an Interim Strategy Note was Senegal prepared in coordination with the World Bank. Uganda The Bank’s Interim Strategy in support of Côte Angola d’Ivoire’s emergency recovery has three main Burkina Faso objectives : (a) support stabilization and assist Tanzania Guinea Bissau the government in addressing key factors Cote d'Ivoire contributing to the conflict; (b) assist war- Niger affected populations by way of community Burundi rehabilitation and support to the provision of Gambia basic social services; (c) assist economic recovery Ghana and reform by focusing on economic governance Mali reforms, institutional building, fostering demand Nigeria for governance and accountability and Kenya supporting sustained economic growth. Mauritania Guinea Figure 6-3 presents the latest data on Côte Cameroon d’Ivoire from the Worldwide Governance Congo, Dem. … Indicators (WGI) (Kaufman, Kraay and Mastruzzi 2009).. The indicators measure six dimensions of 0% 50% 100% governance: Voice and Accountability, Political Stability and Absence of Violence/Terrorism, Source: González et al (2007) and Enterprise Surveys, Government Effectiveness, Regulatory Quality, World Bank Rule of Law, and Control of Corruption. They cover 212 countries for 1996, 1998, 2000, and annually for 2002 onward. These indicators are based on several individual variables measuring perceptions of governance, drawn from 35 separate data sources constructed by 33 different organizations around the world. The WGI project then assigns these individual measures to categories capturing the six dimensions of governance. The measure for political stability has been in near-constant decline since 1996, which is consistent with Enterprise Survey data. Table 6-1 reports rankings for Côte d’Ivoire and its comparators over time. 64 FIGURE 6-3: POLITICAL STABILITY PERCENTILE RANK Political Stability 80 70 60 50 40 30 20 10 0 1996 1998 2000 2002 2004 2006 2008 Source: Kaufmann et al. 2009 Note: the error bands have been omitted. TABLE 6-1: INTERNATIONAL COMPARISON OF POLITICAL STABILITY RANKING. Political Stability Percentile Rank (0-100) 1996 2002 2005 2008 Burkina Faso 49 32 43 39 Cameroon 10 24 32 28 Côte d’Ivoire 40 6 1 5 Ghana 39 43 56 47 Kenya 20 16 14 12 Madagascar 44 38 45 30 Mali 57 52 49 36 Senegal 28 35 42 37 South Africa 15 33 47 42 Togo 26 46 7 40 Source: Kaufmann et al. 2009 REGULATION AND CORRUPTION According to many firm managers, the regulations the Enterprise Survey asks about are not serious obstacles to firm operations. Only one in five managers said that business licensing and registration was a serious problem, while one in four said that customs and trade regulation was a serious problem, and only about one in twenty said labor regulation was a serious problem. None of these specific areas of regulation ranked among the managers’ top concerns. Although this might suggest regulation is not a serious problem, it is important to note that the survey’s narrow measures may not be representative of the overall burden of regulation. Other evidence suggests that regulation might be a broader concern. In particular, there was serious concern about corruption. More than two-thirds of managers said that corruption was a serious problem and ranked among their top concerns. Corruption should be seen as a symptom of other problems in the investment climate. Data from the latest round of the Enterprise Survey suggests that corruption in the form of bribes is not a major in Côte d’Ivoire, compared to other countries in Sub-Saharan Africa (Figure 6- 65 2). However, there is room for improvement bribes are more frequent than in some neighboring countries when it comes to requests for certain infrastructure services or permits. ELECTRICITY As in many other countries in Sub-Saharan Africa, the supply of electricity is limited and unpredictable in Côte d’Ivoire. Despite the fact that Côte d’Ivoire is a net exporter of electricity, 61 percent of firms consider electricity to be a major or severe obstacle. Of 1,210 MW of installed capacity of the Compagnie Ivoirienne d’Electricité (CIE), only 740 MW was available. Anticipated peak demand was 730 MW and left little spare capacity, which prompted the CIE to warn customers about possible outages. Firms reporting outages in 2007 experienced losses equal to nearly 8 percent of sales (Figure 6-4). Firms generally cope with outages by adopting low-tech or labor-intensive production processes, or by generating their own power. Although generator use can reduce losses due to outages by allowing firms to continue production, it is usually far more costly than power from the grid. In practice, fewer than 4 percent of Ivorian firms reported owning generators in 2007, the lowest of comparator countries (Figure 6-5). FIGURE 6-4: INTERNATIONAL COMPARISON OF LOSSES DUE TO POWER OUTAGES Togo Madagascar Cote d'Ivoire Ghana Kenya Senegal Cameroon Burkina Faso Mali South Africa 0% 5% 10% 15% Source: Enterprise Survey, World Bank 66 FIGURE 6-5: INTERNATIONAL COMPARISON OF GENERATOR USAGE Angola Guinea Bissau Gambia Guinea Rwanda Kenya Tanzania Malawi Congo, Dem. Rep. Burundi Swaziland Cape Verde Uganda Mauritania Niger Senegal Madagascar Botswana Namibia Ghana Mozambique South Africa Cameroon Mali Burkina Faso Cote d'Ivoire 0% 25% 50% 75% 100% Percent of firms owning or sharing a generator Source: Enterprise Survey, World Bank 67 REFERENCES Batra, G. and A. Stone. 2004. “Investment Climate, Capabilities and Firm Performance: Evidence from the World Business Environment Survey�. World Bank. David Dollar and Aart Kraay, “Growth is Good for the Poor,� Journal of Economic Growth, Volume 7, No 3, September 2002, pp 195-225. Gelb A, T. Mengistae, V. Ramachandran and M. K. Shah (2009). “To Formalize or Not to Formalize? Comparisons of microenterprise data from Southern and East Africa.� Center for Global Development. Working Paper 175. Kaufmann D., A. Kraay, and M. Mastruzzi 2009: Governance Matters VIII: Aggregate and Individual Governance Indicators, 1996-2008. World Bank Policy Research Working Paper No. 4978. McKenzie, D. and Yaye Seynabou Sakho. 2010. “Does It Pay Firms to Register for Taxes? The Impact of Formality on Firm Profitability�. Journal of Development Economics. 91(1). World Bank. Doing Business. Survey data and reports (DoingBusiness.org). World Bank. Enterprise Surveys. Survey data and reports (EnterpriseSurveys.org). 68