An Assessment of the Investment Climate in Tanzania

Transcription

1 An Assessment of the Investment Climate in Tanzania May 2009 Revised Draft DO NOT CITE OR QUOTE WITHOUT PERMISSION Volume 1: Overview 1

2 Abbreviations and Acronyms BMK FDI GDP ICA ILO IMF ISO LLC LPI MOID MW NBC NMB PAYE PPP ROSCA SME SMLE SSA TANESCO TFP TIN TSH TPA TRA UNDP UNIDO US VAT Bahati Milk Kiosk Foreign Direct Investment Gross Domestic Product Investment Climate Assessment International Labour Organization, International Monetary Fund International Standards Organization Limited Liability Company Logistics Performance Index Ministry of Infrastructure Development Megawatt (1,000,000 watts) National Bank of Commerce National Microfinance Bank Pay As You Earn Price Purchasing Parity Rotating Savings and Credit Association Small and Medium-Sized Enterprise Small, Medium-Sized and Large Enterprise Sub-Saharan Africa Tanzania Electric Supply Company Limited Total Factor Productivity Taxpayer Identification Number Shilling (Currency) Tanzania Ports Authority Tanzania Revenue Authority United Nations Development Program United Nations Industrial Development Organization United States Value Added Tax 2

3 EXECUTIVE SUMMARY The Investment Climate Assessment (ICA) for Tanzania looks at the obstacles that affect firm competitiveness and incentives to grow and invest. The ICA is based on results from the World Bank Enterprise Survey, a survey of 419 small, medium and large enterprises (SMLEs) and 65 microenterprises conducted in late The survey collected detailed information on firm performance, what managers see as the main obstacles that they face, and objective data on the investment climate. Other sources of information are used to supplement the survey. Macroeconomic background Tanzania s economy has grown quickly in recent years, partly because of the rapid growth of mining. Construction, trade and tourism, and manufacturing have also been growing quickly. Tanzania s export performance has also improved with exports increasing from about 15 percent of Gross Domestic Product (GDP) in 1999 to 22 percent of GDP in 2006, mostly because of a large increase in gold exports. The Investment Climate in Tanzania Tanzania s investment climate often compares favorably with the investment climates of its small, land-locked neighbors such as Burundi, Rwanda, and Uganda. For example, the cost of exporting goods overseas from Tanzania and the regulatory burden of doing so is lower than from the three neighboring countries. In addition, the burden of regulation is lower in Tanzania than in Rwanda or Burundi. And firms in Tanzania faced fewer losses because of power outages than firms in Uganda or Burundi do. Under these circumstances, it is not surprising that Tanzanian firms are, on average, more productive and profitable than firms from its three landlocked neighbors. Tanzania, however, does not face the same geographical and physical constraints that its small, land-locked neighbors face. As a result, Tanzania will not be competing as directly for investment with these countries as it will be with other coastal economies with greater opportunities. Unfortunately, Tanzania s investment climate compares less favorably with the investment climates of other countries in Sub-Saharan Africa with similar favorable locations such as Kenya and South Africa. Moreover, it compares even less favorably with the investment climates of other emerging markets that have successfully diversified into export-oriented manufacturing such as Mauritius, China, Malaysia and Thailand. The 2008 ICA is the second ICA for Tanzania. An earlier assessment, based on a 2003 survey, was written in In some areas, such as tax administration and access to finance, Tanzania s investment climate appears to have improved. In most other areas, however, such as power and transportation, governance, and the burden of regulation, the investment climate does not appear to have improved significantly and has sometimes deteriorated. 3

4 Firm Performance On average total factor productivity (TFP) improved by between about 3 and 6 percent a year for manufacturing firms in Tanzania between 2003 and Despite this, competitiveness remains a problem. Although the median manufacturing firm in Tanzania was more productive than the median firms in small, land-locked Burundi, Rwanda and Uganda, productivity is lower than in many other nearby resource-based coastal economies such as Kenya and South Africa and is far lower than in countries such as China, Mauritius and Thailand that have successfully diversified into export-oriented manufacturing. Other evidence also suggests that firms in Tanzania are not highly competitive. Even compared to other countries in the region, few firms have entered international markets. Fewer manufacturing firms in Tanzania (14 percent) exported than in either Uganda or Rwanda (19 and 28 percent). Moreover, in several of the comparator countries (such as South Africa, Thailand, Mauritius and Malaysia) more than 60 percent of the manufacturing firms in the survey exported. Tanzania s poor performance is even more noticeable given that Tanzania is a coastal economy unlike other countries in the region where few firms export (such as Burundi and Uganda). Although labor productivity increases with firm size in most countries, the difference between large and small enterprises is large in Tanzania. Moreover, large firms also appear to be very profitable. Although they might be more productive than small firms because they are capital intensive or because they hire better educated or more highly skilled workers, it could be because they face little competition an interpretation consistent with their high profitability. That is, the combination of high productivity and high profitability might suggest that large firms have market power. Power As well as being asked about firm performance, firm managers were also asked subjective and objective questions on the investment climate. In particular, they were asked what they see as the biggest problems that they face. By far, the biggest concern was power. Close to nine out of ten SMLEs and microenterprises said that it was a serious problem and close to threequarters said it was the most serious obstacle to doing business in the country. The timing of the survey probably contributed to the extremely high concern. Although the poor performance of the power sector has been a problem for many years, the Enterprise Survey took place during a serious power crisis. Growing demand and a steep drop in hydroelectric generation capacity led to load shedding and almost daily outages for many firms. Consistent with firm perceptions, objective measures of power sector performance also suggest that power is a serious problem in Tanzania. For the month before the survey (i.e., during the crisis), the average manager reported 22 outages almost every working day for many firms. Although this stresses that power outages posed a serious problem for firms during the crisis, outages were common even before the crisis. On average, managers reported about 9.1 outages each month in an average month in Moreover, as discussed in the 2004 Investment Climate Assessment, firms reported as many outages in 2003 as they did in

5 The frequent power outages impose a significant burden on firms in Tanzania, have a large impact on firm competitiveness, and undermine Tanzania s attractiveness as an investment location. The average firm reported that outages resulted in losses equal to almost 10 percent of sales in 2005 especially high since this was before the crisis. Insufficient power supplies negatively affect the competitiveness of firms in Tanzania. Although the losses in Tanzania that result from power outages were similar to losses in other countries in the region such as Burundi, Rwanda and Uganda, they were far higher than in better performing economies in Asia. Median firms in China, Thailand and Malaysia reported losses that were less than 2 percent of sales about a fifth of the losses reported in Tanzania. These relatively larger losses mean that Tanzanian firms are likely to have to price their products higher to recoup these losses and are therefore less competitive against similar products produced by firms in China, Malaysia or Thailand. Access to Finance Although far fewer SMLEs and microenterprises in Tanzania said that access to finance was a serious obstacle than said the same about electricity, access to finance ranked as a greater problem than other areas of the investment climate. About 40 percent of SMLEs and about 50 percent of microenterprises said that access to finance was a serious constraint for their enterprises. Under these circumstances, it is not surprising that firms in Tanzania remain heavily dependent on retained earnings to finance new investment and working capital. Although SMLEs in Tanzania use bank financing to similar degrees as SMLEs in other countries in the region, the difference between Tanzania and comparator countries outside the region is considerably larger. Firms in Tanzania finance about 8 percent of new investment with bank financing and about 85 percent with retained earnings. In comparison, firms in South Africa finance 16 percent of their new investment with bank financing and only 59 percent with retained earnings. Firms in Mauritius and the Asian comparators finance over a third of their new investment with bank financing and less than half with bank financing. Access to finance is a larger constraint for small firms and microenterprises than it is for large enterprises. Microenterprises and small firms were more likely to say that access to finance was a problem than large firms (50 percent and 40 percent compared to 28 percent) and were less likely to have bank financing (17 percent and 24 percent compared to 83 percent). Although there is some market segmentation in most countries, the difference between small and large firms is larger in Tanzania than in most of the comparator countries. Although many firms did not use bank financing, this does not appear to be due only to a lack of demand. Many firms that had not applied for bank financing reported that they did not do so because application procedures were too complex. In fact, over a quarter of firms said that this was the main reason that they had not applied. This proportion is far higher than in most of the comparator countries for example only 9 percent of firms in Kenya and 6 percent of firms in Uganda said the same. 5

6 Competition with informal firms About 29 percent of SMLEs and about 25 percent of microenterprises, some of whom might also be informal, said that competition from the informal sector was a serious constraint on doing business. Although this suggests far less concern than about power or access to finance, it also suggests that competition from the informal sector is a serious problem for many firms. Microenterprise managers were asked what they saw as the biggest barriers to becoming formal. The biggest barrier was taxes about 37 percent of microenterprise managers said that tax rates were a serious obstacle to becoming formal and about 31 percent said the same about tax administration. They also said that the cost of registering discouraged microenterprises from doing so about a quarter said that the cost of registering was a serious barrier to becoming formal. Consistent with this, objective data also suggests that the cost of registration is high. At the beginning of 2008, the 12 procedures associated with registration took 29 days to finish and the cost was 41.5 percent of per capita gross national income (GNI). Although the cost of registering is only slightly higher than in Kenya (39.7 percent of per capita GNI) and lower than in Uganda (100.7 percent), it is far higher than in the best performing comparator countries such as South Africa, Mauritius and Thailand where it only costs between 5 and 6 percent of per capita GNI to finish all the procedures to start a business. Tax rates SMLE managers were more likely to say that tax rates were a problem than other areas of the investment climate except power and access to finance. Moreover, as noted above, microenterprise managers reported that high tax rates were one of the primary obstacles to becoming formal in Tanzania. Although tax rates are usually among the greatest concerns in Enterprises Surveys, objective data suggests that taxes are high in Tanzania. Although the standard corporate tax rate does not appear to be high, the Value-Added Tax (VAT) rate is higher than in most of the comparator countries. Moreover, after accounting for differences in depreciation rates, investment incentives, loss carry-forwards, and other things, the total tax rate that firms pay (45.1 percent of profits), which does not take the high VAT rate into account, is also high. In summary, although taxes are within the range observed in the comparator countries, they are on the upper end of this range. Summary Although the Tanzania s investment climate compares favorably with the investment climates of its small, land-locked neighbors such as Burundi, Rwanda and Uganda, it compares less favorably with investment climates in other countries in Sub-Saharan Africa and East Asia that are more favorably situated as potential investment locations. The Investment Climate Assessment discusses each area of Tanzania s investment climate and proposes some policy recommendations to improve it. 6

7 The Investment Climate Assessment has two volumes. The first provides a brief overview of the main issues and some policy recommendations. The second provides far greater detail on each area of the investment climate and includes appendices that outline the econometric analysis that underlies some results in the first volume. 7

8 Table of Contents Executive Summary... 3 Overview... 9 I. Macroeconomic Background II. Productivity III. Perceptions about the Investment Climate IV. Electricity V. Access to Finance VI. Informality VII. Tax Rates VIII. Tax Administration IX. Transport X. Education and Training XI. Regulation XII. Conclusions and Policy Recommendations References Endnotes

9 OVERVIEW 1. The goal of the Investment Climate Assessment (ICA) for Tanzania is to evaluate the investment climate in Tanzania in all its operational dimensions and to promote policies to strengthen the private sector and encourage broad-based economic growth. Sustained improvements in living standards depend on broad-based growth, which will only occur if firms improve their productivity by investing in human and physical capital and technology. But firms will only do this when the investment climate is favorable. 2. The main source of information for the ICA is the Enterprise Survey a survey of 419 small, medium and large enterprises (SMLEs) and 65 microenterprises. The survey collects information on firm performance, the labor market, the financial sector, infrastructure, and the regulatory environment. Information from the survey will be supplemented with information from other sources including the Doing Business Report; analytical reports by the World Bank, the International Monetary Fund (IMF), other international organizations and the Government of Tanzania; and academic papers and reports. 3. Because Enterprise Surveys have been conducted in many countries throughout the world, Tanzania s investment climate can be compared with those of other countries. One advantage that the Enterprise Survey (ES) has over other firm surveys is that the World Bank has conducted similar surveys in over 100 countries throughout the World. Because the sampling and survey methodology and the questionnaire are the same, it is possible to benchmark Tanzania s investment climate against those of other countries. Throughout the report, Tanzania s investment climate is compared to the investment climates of three groups of countries: (i) nearby countries in East Africa (Burundi, Rwanda and Uganda); (ii) countries in Sub-Saharan Africa (SSA) that are more attractive for investment outside of the natural resource sector either due to their coastal location or because they have successfully diversified out of primary production into other sectors (Kenya, Mauritius, South Africa and Swaziland); and (iii) several fast growing countries in East Asia that have also successfully diversified out of primary production into export-oriented manufacturing (China, Malaysia, and Thailand). 4. Although the first group of countries provides natural comparators, it is important to note that Tanzania should, in many ways, be a more attractive investment location than these other economies. In particular, all of the neighboring countries except Kenya are small and land-locked attributes that will make them less attractive as investment locations than Tanzania. 1 In this respect, the second and third groups of countries, which are generally more attractive as investment locations might be a better set of comparators. 5. The 2008 ICA is the second Investment Climate Assessment for Tanzania. An earlier assessment, based upon a survey completed in 2003, was completed in One of the goals of the 2008 ICA is to see how much progress has been made since the previous survey (see Box 1 below). 9

10 Box 1: The 2003 Investment Climate Assessment The 2008 ICA is the second ICA for Tanzania. An earlier assessment, based upon a survey completed in 2003, was completed in The results of the earlier report include: Firms in Tanzania were not highly competitive. Although labor productivity was higher than in Uganda, it was lower in the more successful manufacturing countries such as Kenya and China. Productivity was a particular problem for small enterprises. Finally, human capital and technology use was relatively low. Workers were less well educated than in Kenya or Uganda and firms were less likely to use computers and than competitors from other low income countries. Problems with competitiveness were also reflected in the poor export performance of manufacturing firms in Tanzania. Firms in Tanzania were about 18 percentage points less likely to export than similar firms in Kenya. The report argued that this partly reflected problems with trade and customs regulations. Burdensome regulation contributed to high levels of informality. The report notes that estimates suggest that up to 58 percent of gross national income is generated by the informal sector higher than in Kenya, Uganda, India or China. In part, this seemed to reflect the heavy burden of regulation and corruption both of which can discourage firms from becoming formal. It took 35 days and cost an amount equal to 304 percent of per capital GNI to register a new business. This also contributed to corruption firms were more likely to report that they needed to pay bribes to get things done than in the comparator countries. There were problems in several additional areas of the investment climate that were obstacles to firm operations and growth. Firms were most likely to say that tax rates, electricity, tax administration, corruption, and the cost of and access to financing were obstacles to their firm s operations and growth. The objective data was generally consistent with the idea that these were problems. Firms were less likely to have loans, reported spending more time with tax officials and had greater losses due to power outages than in the best performing comparator countries. One of the goals to the 2008 investment climate assessment will be to see how much progress has been made in these areas since the previous survey Source: Regional Program on Enterprise Development (2004). I. Macroeconomic Background Tanzania s economy has grown quickly in recent years, partly due to rapid growth in the mining sector. GDP growth averaged 6.5 percent per year between 2000 and 2007, a significant increase from the 1990s, when growth averaged 3.1 percent per year. The fastest growing sector is mining (see Table 1). Between 2001 and 2006, mining grew at an average rate of 15.7 percent per year. Construction, trade and tourism, and manufacturing have also been growing relatively quickly. Agriculture has been growing more modestly with average growth averaging only 5 percent per year since

11 Table 1: Mining and quarrying, along with construction, are the fastest growing sectors Average Agriculture Manufacturing Mining and quarrying Trade & tourism Construction Transport & Communication Finance & Business Services Electricity & Water Public Administration Source: The President s Office, Planning and Privatization (2007). 7. Tanzania s export performance has also improved significantly over the past decade. Exports increased from about 15 percent of GDP in 1999 to 22 percent of GDP in This is encouraging given that Tanzania s location means that it could serve as an attractive destination for investment in export-oriented industries something that might be more difficult for its landlocked neighbors such as Burundi, Rwanda and Uganda. 8. Despite the rapid growth in GDP and exports, Tanzania remains poor. Tanzania is a low income country with GDP per capita of $995 (United States [US] $ in Price Purchasing Parity [PPP] adjusted prices). 3 Recent estimates suggest that 36 percent of the population lives below the national poverty line, with poverty higher in rural areas (39 percent) than in urban areas (29 percent). 4 The World Bank estimated that 58 percent of the population lived on one dollar a day (PPP) in 2000 and 90 percent lived on two dollars a day. Figure 1: Agriculture remains an important sector in Tanzania. Services, 39 Agriculture, 45 Construction, 6 Manufacturing, 6 Mining and Quarrying, 3 Source: National Bureau of Statistics (2007). 9. Agriculture is the largest sector of the economy in Tanzania and the manufacturing sector remains small. Agriculture accounted for 45 percent of GDP in 2006 (see Figure 1). 11

12 US $ (000s) This is not surprising given that about 80 percent of the population lives in rural areas (International Monetary Fund, 2007). Services are also important, accounting for about 39 percent of GDP. In contrast, manufacturing is considerably less important, accounting for only about 6 percent of GDP in Despite its rapid growth and its importance with respect to exports, mining remains small in terms of its contribution to GDP and employment. Mining and quarrying accounts for only about 3 percent of GDP (see Figure 1). Although mining plays a relatively modest role in the economy, this sector has accounted for much of Tanzania s improved export performance. Exports of gold, in particular, have become particularly important (see Figure 2). Between 2000 and 2006, gold exports, including ore, increased from about $114 million in 2000 (17 percent of exports) to about $770 million (46 percent of exports) in Most remaining exports are of agricultural goods and food products. Manufactured goods make up a smaller share of exports. Figure 2: The increase in exports is mostly due to large increase in gold exports. $1,000,000 Exports by sector (US$, 000s) $750,000 $500,000 Fish Coffee Tobacco Gold $250,000 $0 Source: UN Comtrade/World Bank WITS. II. Productivity The historically modest size of the manufacturing sector in Tanzania reflects that manufacturing firms have not been highly competitive. The 2004 ICA showed that although labor productivity was higher than in Uganda, it was lower in the countries that have been more successful in export-oriented manufacturing such as Kenya and China Although this is a concern, firms in Tanzania are more productive on average than firms in Tanzania s land-locked neighbors. As in 2003, the median manufacturing firm in Tanzania was more productive in 2006 than the median firm in Uganda (see Figure 3). 7 The median firm was also more productive than median firms in Burundi and Rwanda two countries that had not been surveyed when the 2004 ICA was written. 12

13 13. Moreover, productivity has improved since Although survey data is not ideal for evaluating changes in firm productivity over time, total factor productivity (TFP) on average appears to have improved by between about 3 and 6 percent per year for manufacturing firms in the Enterprise Survey between 2003 and Despite improvements in productivity, productivity remains far lower than in the countries that have been more successful in diversifying into export-oriented manufacturing. Tanzania s land-locked neighbors, however, might not be the most relevant comparators. With its attractive coastal location that should make exporting overseas easier, Tanzanian firms could be competing with firms in other similar attractive locations. Labor productivity, however, is lower on average in Tanzania than in most of these comparator countries. For example, it is less than half of average labor productivity in Kenya or Thailand and less than one-third of labor productivity in China. Figure 3: Despite improvements in productivity since 2003, firms do not appear competitive relative to firms in the best performing countries. Value-Added per Worker (in US$) Percent of firms that export South Africa Malaysia China Mauritius Thailand Malaysia Mauritius Thailand South Africa China Kenya Rwanda Uganda Burundi Kenya Rwanda Uganda Burundi Tanzania Value-Added per Worker(US$ 2005) Tanzania Percent of firms that export Source: World Bank Enterprise Surveys. Note: See Chapter 2 in Volume 2 for a discussing of methodology. Cross country comparisons are for manufacturing firms only. Value-added per worker is for the median firm in each country. 15. Other evidence also suggests that firms in Tanzania are not highly competitive. Exporting is a good indicator of competitiveness because whereas even inefficient firms can compete in local markets when they are protected from foreign competition by tariff and nontariff barriers and transportation costs, only efficient firms can overcome these barriers and compete on international markets. 9 Even compared to other countries in the region, few Tanzanian firms have entered export markets (see Figure 3). 10 Fewer manufacturing firms in Tanzania exported (14 percent) than in either Uganda or Rwanda (19 and 28 percent respectively). Tanzania s relatively poor performance in this respect is even more remarkable 13

14 given than unlike other countries in the region where few firms export (e.g., Burundi and Uganda), Tanzania is not land-locked. 16. Tanzania compares even less favorably with respect to exporting when compared with the other comparator countries. In several of the comparator countries (e.g., South Africa, Thailand, Mauritius and Malaysia) more than 60 percent of the manufacturing firms exported some part of their production. Like Tanzania, most of the other comparators are better situated for exporting than Tanzania s small, land-locked neighbors are. Labor Costs Given that firms in Tanzania do not appear to be highly competitive, it is important to see whether this is due to high labor costs. Although the median firm in Tanzania spends more per worker on wages, salaries and other benefits than in several other countries in the region, wages are lower than in most of the more successful manufacturing economies in Sub-Saharan Africa and Asia (see Figure 4). For example, the median firm in China spends about $1,250 per worker on wages, salaries, and benefits and the median firm in Thailand spends about $1,750 per worker 50 percent higher than and twice as high as in Tanzania respectively. In this respect, the cost of labor does not appear to be a major drag on firm competitiveness. Figure 4: Although labor productivity is higher than in many nearby countries, costs are lower than in economies that have been more successful in manufacturing and so remain competitive. Per worker labor cost (2005 US$) Malaysia Thailand China India South Africa Mauritius Swaziland Kenya Unit labor costs Malaysia Thailand India China South Africa Mauritius Swaziland Kenya Rwanda Uganda Burundi Rwanda Uganda Burundi Tanzania $0 $5,000 $10,000 $15,000 Labor Costs per Worker (US$ 2005) Tanzania -20% 0% 20% 40% 60% Unit Labor Cost (as % of value-added) Source: World Bank Enterprise Surveys. Note: All values are for the median firm in each country. Cross-country comparisons are for manufacturing firms only. 18. Because simple wage comparisons do not take differences in productivity into account, it is better to look at an alternate measure of labor costs to get a better idea about how they affect competitiveness. Unit labor costs (labor costs as a percent of value-added) are a measure of labor costs that make it easier to assess the net impact of labor costs on competitiveness by 14

15 taking differences in productivity into account when assessing labor costs. Unit labor costs are higher when higher labor costs are not fully reflected in higher productivity. When unit labor costs are higher (i.e., when labor costs are higher compared to productivity), all else equal, firms will find it more difficult to compete on international markets than when they are lower. Although unit labor costs are not the only factor that affect competitiveness for example, they do not take the cost of capital or capital intensity into account they are a better measure of competitiveness than labor costs alone. 19. Tanzania compares relatively favorably with other low-income countries in Sub- Saharan Africa with respect to unit labor costs. Although wages are slightly higher in dollar terms than in Rwanda, Burundi, and Uganda, labor productivity is also higher. Because the difference is greater for labor productivity, unit labor costs are actually lower than in any of these other countries. 20. Unit labor costs also compare favorably with some of the successful manufacturing countries in Sub-Saharan Africa and Asia. Although unit labor costs are considerably higher than in China or India and slightly higher than in Kenya and Thailand, they are about the same as in Swaziland, are slightly lower than in Malaysia and are far lower than in South Africa and Mauritius. Compared with these economies, the low level of productivity in Tanzania is set off with relatively low wages. This suggests that productivity would have to improve if Tanzanian firms were to remain competitive while paying higher wages. 21. Although the minimum wage does not appear to be set so high that it would have a large impact competitiveness, increases might mean that it has a greater impact in the future. About 10 percent of the workers that were interviewed for the Enterprise Surveys reported earning less than the mandated minimum wage that was in place when the survey was run (the old minimum wage). Given that wages are generally higher in larger firms, it is not surprising that most of these workers worked in firms with less than 50 workers (see Figure 5). About 13 percent of workers in small firms reported wages below the old minimum wage compared with 7 percent of workers in mid-sized firms and no workers in large firms. A new, higher, minimum wages was set in January If the new minimum wage is enforced, more workers (about 20 percent of workers interviewed in the worker module of the Enterprise Survey) would have been below the threshold. Moreover, this is true for workers across the spectrum of firm sizes. Indeed, many of the workers in larger firms reported wages below the new minimum wage. 22. Another issue is that it is likely that enforcement of those minimum wages would be highly difficult and most likely somewhat ad hoc. Further, the fact that exemptions have been granted to larger firms and firms that export appears to be counter to what the data show, as well as introducing a significant source of non-transparency in the system. 15

16 % of workers below minimum wage Figure 5: Many workers are paid less than the new minimum wage. 25 workers below old minimum wage workers below new minimum wage Total Source: World Bank Enterprise Survey. Note: Firms with more than 300 workers are excluded because few firms are in this size category. Differences in productivity across different firms within Tanzania 23. Although labor productivity increase with firm size in most countries, the difference between large and small manufacturing enterprises is particularly great in Tanzania. As in most countries, productivity in Tanzania increases with firm size (see Figure 6). The median small enterprise produces about $2,100 of output per worker compared to about $5,000 for the median medium-sized enterprise and about $14,200 per worker for the median large enterprise. This is a far larger difference than in most countries. Although small enterprises in Tanzania are less productive than small firms in Kenya, large firms are more productive than similar firms in Kenya. 24. Although the high productivity of large firms could be encouraging, it might be because that they face little competition. Firms in Tanzania especially large firms are relatively profitable. Return on sales (profits over sales) is 21 percent for the median SMLE in Tanzania (see Table 2) higher than for most other countries in Africa. Moreover, as with productivity, profitability appears to be increase significantly by size and is particularly high for large manufacturing firms returns on sales are about 29 percent for the median large firm. 16

17 Value-added per worker (2005 US$) Figure 6: Although small firms in Tanzania are relatively unproductive, large firms appear to be relatively productive. $20,000 $15,000 Small Medium Large $10,000 $5,000 $0 Kenya Tanzania Uganda Source: World Bank Enterprise Surveys. Note: See Chapter 3 in Volume 2 for methodology. Cross-country comparisons are for manufacturing firms only. All values are for the median firm in that category in each country. 25. This suggests that high productivity observed among large firms might reflect market power rather than true productivity. Because of the way that total factor productivity is calculated, firms that have market power appear to be more productive than they really are. 12 It seems that this might be the case in Tanzania. First, as noted above, large firms are also very profitable, possibly suggesting that they have market power. Second, very few large firms from Tanzania export and even fewer say that they see international markets as their main market. Only about 5 percent of large firms in Tanzania said that international markets were their most important market compared to close to 20 percent of large firms in Kenya and Uganda. Given that Tanzania has a relatively favorable location for exporting (i.e., it is not landlocked and most industry is located in Dar es Salaam), this provides further evidence that market power might be important. That is, if large firms in Tanzania really were productive, we would expect that they would be able to compete on international markets. Finally, when asked about the competition that they face, large Tanzanian firms competing in domestic markets were more likely to say that they had no competitors than similar firms in Kenya or Uganda (about 17 percent in Tanzania but only 4 percent in Uganda and 5 percent in Kenya). Overall this suggests that although large firms in Tanzania might be more productive than small firms because they are capital intensive, because they hire better educated or more highly skilled workers, or because they use more advanced technology, these do not appear to be the only reasons large firms might only appear to be productive because they have market power. 17

18 26. Table 2: Profitability (return on sales), by firm size. All Small Large Mauritania 3% 3% 0% Gambia 9% 3% --- Ghana 9% 9% 8% Guinea-Bissau 10% 9% --- Rwanda 12% 8% 16% Uganda 13% 13% 21% Congo, DR 15% 17% 15% Botswana 17% 13% 36% Guinea-Conakry 18% 18% 20% Angola 19% 19% 9% Burundi 20% 20% -1% Tanzania 21% 18% 29% Kenya 24% 17% 23% Namibia 28% 28% 20% Swaziland 31% 38% 35% Source: World Bank Enterprise Survey. Note: Data is from between Return on sales is sales less costs (materials, wages, other miscellaneous costs) divided by sales. This is used rather than return on assets due to difficulty of measuring assets accurately. Comparisons are for manufacturing firms only. All values are for the median firm in each country. Observations presenting medians for groups with less than five observations are dropped. 27. Other factors also affect productivity in Tanzania. As well as the stong assocation between productivity and size, other associations can also be observed between enterprise characteristics and firm productivity. 13 The most robust associations are: (i) firms that use technology more intensively are more productive than other firms. In particular, firms that are International Standards Organization (ISO) certified and that have their own website are much more efficient than other firms (51 percent and 43 percent more productive respectively) and (ii) enterprises that provide their own transportation are about 32 percent more efficient than less vertically integrated firms. As discussed in the main volume, it is difficult to draw strong conclusions from this analysis due to the possiblity of reverse causation. III. Perceptions about the Investment Climate In addition to asking managers to provide information on firm productivity, the Enterprise Survey asks managers about the investment climate. The survey asks two kinds of questions: (i) subjective questions about what they see as the major obstacles that their firm faces; and (ii) objective questions such as production lost due to power outages, whether the firm has a loan or overdraft facility, and amount of time managers spend dealing with government regulations. 29. Manager s perceptions about the investment climate provide a very useful starting point for an analysis of investment climate constraints. Although perception-based data has drawbacks (see Chapter 3 in Volume 2 for discussion), enterprise managers have a better idea about the obstacles they face than anyone else. Therefore as a starting point for the analysis, this section looks at what enterprise managers say are the biggest obstacles that they face

19 % of firms saying issue is serious problem Results from 2006 Enterprise Survey 30. Problems with power were by far the greatest concern for both SMLE and microenterprise managers. Close to nine out of ten SMLE and microenterprise managers said that power was a serious problem for their firm (see Figure 7). This was significantly more than the number that rated any of the other constraints as a serious problem. As discussed below, the concern about the power sector partly reflects the crippling power crisis that was occurring at the time of the survey. 31. Access to finance and competition with informal firms were also ranked as serious problems by both SMLE and microenterprise managers. Only about four out of ten SMLE managers and about five out of ten microenterprise managers said that access to finance the next greatest constraint was a serious problem. Although competition with informal firms was a lesser concern than electricity and access to finance for both SMLE and microenterprise managers, about one quarter of managers of both types of firm said it was a serious problem. Figure 7: SMLEs and microenterprises have similar views on the investment climate in Tanzania with electricity and access to finance rating far above other constraints. 100% 80% SMLEs Microenterprises 60% 40% 20% 0% Source: World Bank Enterprise Survey. 32. Although in many countries SMLE and microenterprise managers have very different concerns about the investment climate, they have similar concerns in Tanzania. 16 With a few exceptions such as tax rates microenterprise and SMLE managers had very similar concerns. As well as having similar views about electricity, access to finance and competition with the informal sector, SMLE and microenterprise managers also seemed to broadly agree on the areas of the investment climate that were less serious problems. Relatively few (one-fifth or less) of enterprise managers of either type of enterprise rated telecommunications, political instability, worker skills and education, crime, transportation, the courts, and most areas of regulation (e.g., labor, trade, or business registration) as serious problems. The most notable difference between the views of microenterprise and SMLE manager is with respect to their 19

20 % of firms saying issue is serious problem views about tax rates. Whereas about two out of five SMLE managers said that tax rates were a serious problem, only one out of five microenterprise managers said the same. Comparisons with the 2003 Enterprise Survey 33. These concerns were very similar to the concerns of SMLEs interviewed for the earlier 2004 Investment Climate Assessment. It is interesting to compare results from the 2006 Enterprise Survey with results from the earlier 2003 Enterprise Survey. 17 In the 2003 survey, firms greatest concerns were tax rates, electricity, cost of finance, tax administration, corruption, access to finance and macroeconomic instability. These were similar to the concerns expressed in In particular, electricity, tax rates, and corruption remained among the top concerns in Figure 8: Other than electricity, fewer enterprises said that most other areas of the investment climate were serious problems in 2006 than in % 80% 60% % 20% 0% Source: World Bank Enterprise Surveys. 34. There were, however, some noticeable differences: Firms were far more likely to indicate that electricity was a serious problem in 2006 than they were in About 70 percent of manufacturing firms said electricity was a serious problem in 2003 (the second biggest constraint) compared to about 90 percent in This emphasizes that although problems in the power sector are not new, the crisis seriously amplified concerns about the sector in the 2006 survey. Except for transportation, firms were less likely to indicate that all other areas of the investment climate were serious problems in 2006 than they were in The most likely reason for this is that the power crisis overshadowed concerns about problems in other parts of the investment climate to such a degree that it changed managers responses on other questions. That is, it is possible that managers used the power crisis as a reference point, thinking, for example, corruption is a less serious problem than 20

21 power. Since we said power was a very severe problem, we should say that corruption is only a major or moderate constraint. Although fewer managers indicated that most areas of the investment climate were serious problems in 2006 than in 2003, differences were larger for some obstacles than for others. In particular, far fewer firms said tax administration was a serious problem in 2006 than in 2003 (about 35 percent fewer) and it fell from the 3 rd to the 6 th greatest constraint of the 15 constraints common to the two surveys. Concern about tax rates and corruption also appears to have declined somewhat, although both remained among the top concerns in Box 2: Reforms since the 2003 Investment Climate Assessment The 2003 ICA included suggestions for reforms in several areas of the investment climate. Recommendations included reducing the burden of taxation; improving the performance of the power sector; increasing access to credit, especially for micro and small enterprises; improving tax administration; reducing corruption; maintaining macroeconomic stability; increasing ties to the international economy; and improving enterprise productivity. This box summarizes some of these recommendations. Reducing the Burden of Taxation on Formal Enterprises. In the 2003 Enterprise Survey, firm managers were more likely to say that tax rates were a serious problem than any other area of the investment climate. The 2003 ICA suggested that improvements in tax collection, which could reduce the relative burden that taxes impose on formal enterprises, would be useful and that they would reduce economic distortions. The report also recommended taking steps to reduce corruption in tax administration. As discussed in Volume 2, the Government of Tanzania, with support from the World Bank, has taken steps to improve tax administration are reduce corruption in the tax authority. Although concern about tax rates remains relatively high, enterprises appear less concerned about tax administration in the 2006 survey than they did in the 2003 survey. These reforms have broadened the tax net. The tax yield more than doubled from SHS 1.4 trillion in FY 2003/04 to SHS 3.4 trillion in FY 2007/08 and the ration of taxes to GDP ratio has increased to 14.9 %. The number of registered taxpayers increased from 190,000 in July 2003 to 381,000 in March Steps have also been taken to improve governance in revenue administration, and a TRA anti-corruption strategy has been designed as part of the National Anti-Corruption Strategy and Action. Improving the Performance of the Power Sector. The poor performance of the power sector was identified as a serious problem in Tanzania in the 2003 ICA. Electricity was second only to tax rates in the percentage of manufacturing enterprises that rated it as a major or very severe problem. Reforms in the power sector have been modest. For example, the ICA recommended that the government should ensure that TANESCO can finance its operations in the medium term without government assistance. However, the Government only agreed to a recovery plan only in 2007 and in 2008 further progress in implementation of the TANESCO financial recovery plan were still needed by Government, (e.g., the conversion of the IPTL plant from HFO to gas firing; and tariff adjustments that may be required to strengthen TANESCO s revenue base by October 2008). The 2003 ICA also recommended a series of actions including: (i) unbundling generation, transmission, and distribution; (ii) restructuring power sector debt; and (iii) establishing an independent regulatory body and revising supporting legislation to establish a successful track record with respect to regulatory decision making. These actions have not been taken and some, such as unbundling, have been deemed infeasible for Tanzania. Enhancing the Effectiveness of Financial Services and Access to Credit. The ICA 2003 suggested several reforms to strengthen the financial system in ways that improve access for micro and small enterprises. These included reforms related to the privatization and regulation of 21

22 microfinance institutions and efforts to improve the institutional environment, such as increasing access to credit information and improving contract enforcement and the efficiency of the judicial system. The government completed the major reform of privatizing the state-owned National Microfinance Bank (NMB) as recommended. Other reforms such as the privatization of the Tanzania Investment Bank (TIB), and Tanzania Postal Bank (TPB) were not completed along with some other suggested reforms, i.e. the regulation of smaller microfinance institutions (MFIs) adopting a light handed and delegated approach to small providers of basic financial services as recommended is still not operational. More recently a Leasing law was passed but regulations to implement the law have still to be developed to become operational. Some of the institutional reforms have also not been completed. The establishment of a credit information bureau is still in process and is still not operational, initial improvements in the commercial justice have not been followed up with reforms in civil procedures and other courts leading little or no change of the three to four year period. Reducing corruption. The government started working to repeal the Anti-Corruption Act of 1971 and replace it with a new Anti-Corruption Law in The Anti-Corruption Act was repealed in 2007 and a new law The Prevention and Combating of Corruption Act 2007 enacted. The revision will widen the definition of corruption, increase protection for whistleblowers, and place the onus on those being investigated to prove that assets were not amassed through corrupt or unethical behavior. The government has also made some progress in improving the institutional framework related to corruption (e.g. the National Anti-Corruption Strategy and Action Plan, NASCAP ( ) is being implemented by Government expanding anti-corruption initiates further to local council level and establishing closer collaboration with civil society. In addition, the 2003 ICA recommended changing the incentives that firms face to offer bribes. Recommendations included streamlining regulation to reduce the burden on enterprises and reducing the discretion that officials have in interpreting regulations. Such measures were particularly recommended for agencies, such as the Tanzania Revenue Authority, where corruption was identified as a particular problem. While the Tanzania Revenue Authority has implemented a series of activities since 2003 (see TRA box), not much has happened in other related agencies such as the Land or business registry. Improving Ties to the International Economy. To improve international competitiveness, the ICA 2003 recommended a review of existing practices in customs administration and trade regulation. It recommended that these reforms should focus on shortening processing times and reducing the burden created by excessive regulation. The Government largely implemented successful reforms at customs by introducing risk based assessment and larger penalties for willful corruption of freight forwarders and clearing agents. In addition, the customs administration is also implementing a fully integrated computerized documentation system to allow the filling of documents before the freight arrival. However, these measures will not improve processing time due to the limitations at the Port of Dar Es Salaam. Increasing Labor and Total Factor Productivity. Enterprises in Tanzania, in 2003, were less likely to have formal training programs than similar enterprises in Kenya or China, but those that did were more productive. The 2003 ICA recommended encouraging enterprises to invest in worker training to improve firm-level productivity. The government should facilitate training and apprenticeship programs that are designed and implemented by the private sector. The government was also encouraged to consider exploring ways to improve access to business education. The government recently introduced a matching grant program which is also intended to address worker education and training, along with a program to upgrade the curriculum and faculty at business. However, because of their recent introductions, it is difficult to properly evaluate the success of these programs. Crime and transportation declined less than other constraints, although neither ranked among the top constraints in either survey. Both moved from among the lowest 22

23 concerns in the 2003 survey to somewhere near the middle in the 2006 survey. This suggests that these areas might require greater attention to avoid further deterioration in perceptions. Comparisons between the 2003 and 2006 surveys are very difficult for access to finance. In 2006, more firms said access to finance was a serious constraint than any area of the investment climate except power. In 2003, it ranked below tax rates, tax administration, and corruption as well as power. It is important to note, however, that the wording of the question changed between the two surveys. In 2003, it was described as access to finance (collateral) and there was a separate question for cost of finance (interest rates). In 2006, it was described as access to finance (availability and cost). Given that cost of finance was the second largest constraint in 2003, the discrepancy might be due to change in wording rather than a change in availability of financing. The changed question suggests that it will be particularly useful to also look at objective data on access to finance to see whether access improved between the two surveys. IV. Electricity Firm managers were far more concerned about the performance of the power sector than they were about any other area of the investment climate. Close to nine out of ten SMLEs and microenterprise managers said that it was a serious problem and close to threequarters said it was the most serious obstacle to doing business in the country. 36. Problems in the power sector are not unique to Tanzania. Although concern is particularly high in Tanzania, more than seven in ten firms said that power was a serious obstacle in about half of the low-income countries in Sub-Saharan Africa where surveys were conducted in 2006 and 2007 (see Figure 9). 37. The timing of the survey probably contributed to the extremely high level of concern. Although, as discussed below, the poor performance of the power sector has been a problem for many years, it is important to note that the Enterprise Survey took place during a serious power crisis. Growing demand and a steep drop in hydroelectric generation capacity led to load shedding and almost daily outages for many firms at the time of the survey (see Box 3). Moreover, it was not clear how long this crisis would continue, adding to uncertainty about the cost that the crisis would impose on firms. 23

24 Box 3: The 2006 Power Crisis. At the time of the survey, the Tanzania power system had about 1,192 Megawatts (MW) of installed generating capacity (permanent plants). A significant part of this capacity is hydroelectric capacity, which accounted for about 562 MW of total installed capacity (47 percent of the total). The remainder (630MW) was thermal capacity. A serious drought that lasted for several years resulted in a significant and continuous drop in reservoir water levels. This resulted in available hydropower capacity dropping from 562 to 300MW. This shortage meant that Tanzania Electric Supply Company Limited (TANESCO) started to ratio power in February This in turn resulted in serious load shedding and almost daily outages in some parts of the country. The problems can be seen in the Enterprise Survey data. Whereas the average firm in the Enterprise Survey reported about 9 outages per month in 2005, the average firm reported about 22 per month in It was estimated that the 2006 energy crisis might have reduced GDP growth by as much as 2 percentage points. Significant rainfalls during the short rains season in late 2006 and early 2007 restored full hydroelectric capacity, with the company lifting power rationing on December 28 th, This allowed TANESCO to avoid further generation shortfalls. TANESCO also contracted about 140MW of leased capacity for two years and 40 MW of leased capacity for one-year. About 150MW was scheduled to come on line by the end of Although generation capacity is now sufficient for the near term, the crisis in 2006 greatly affected the private sector and highlights the impacts of power shortages and uncertainty on businesses. Moreover, it was not a new problem Tanzania has faced three energy crises over the last decade. This emphasizes the importance of improving sector performance in the medium-term. Source: World Bank (2007b). 38. Although both firm perceptions and some objective indicators are likely to be negatively affected by the crisis, it is important to note that problems in the power sector predate the 2006 crisis. As noted above, close to 70 percent of firms in the 2003 Enterprise Survey said that power was a serious problem for them at that time. Moreover, as discussed below, the results from the 2006 survey suggest that problems in the sector were evident before the crisis started. 39. Consistent with firm perceptions, objective measures of power sector performance also suggest that power was a serious problem in Tanzania at the time of the survey. In addition to asking firm managers about whether power was a serious problem, the Enterprise Survey also asked firms about the frequency of outages and the losses that this caused for the firms. For the month prior to the survey (i.e., during the crisis), managers reported an average of 22 outages per month (see Figure 9). Moreover, the median manager reported 30 outages per month. This was higher than in most other countries in SSA, many of whom were having their own crises at the time. 24

25 Figure 9: Power is a serious problem in many countries in Sub-Saharan Africa Guinea-Bissau Guinea-Conakry % reporting power is serious problem Zambia Mozambique Kenya Mauritania Nigeria Rwanda Mali Senegal Congo Burundi Gambia Uganda Ghana Tanzania 0% 25% 50% 75%100% % of firms Median days with outages in an ave. month Mozambique Zambia Mali Mauritania Kenya Guinea-Bissau Tanzania Ghana Rwanda Burundi Uganda Senegal Congo, DR Gambia Tanzania 2006 Nigeria Guinea-Conakry No of days of outages Ave days with outages in an ave. month Mozambique Zambia Mali Mauritania Kenya Guinea-Bissau Tanzania 2005 Ghana Rwanda Burundi Uganda Senegal Congo Gambia Tanzania 2006 Nigeria Guinea-Conakry No of days of outages Source: World Bank Enterprise Surveys. Note: Comparisons based upon data from the surveys in Africa in 2006/07 are for all firms (not just manufacturing firms). Outliers (firms more than 3 standard deviations from the mean) are dropped for days of outages when calculating means. Data are for year prior to survey (2005 for Tanzania). 40. Although this emphasizes that outages became very common during the crisis, they were common even before the crisis. Firms were also asked about the losses and outages during 2005 before the full impact of the crisis had been felt. Although recall data must always be treated cautiously, and it is possible that firm managers might have overestimated losses and outages in 2005 due to the crisis coloring their answers, this gives some indication of the impact of the crisis. 19 Managers reported an average of about 9 outages per month in an average month in 2005 with about 6 outages per month. Although lower than in the very worst performing countries, this is still higher than in most other countries in Sub-Saharan Africa (see Figure 9). Moreover, as discussed below, most countries in SSA perform very poorly in this respect relative to countries outside of the region. This suggests that problems in the power sector have had a negative impact on firm competitiveness in Tanzania even before the crisis. 41. The frequent power outages impose a significant burden on firms in Tanzania, negatively affect firm competitiveness, and undermine Tanzania s attractiveness as an investment location. Frequent and long power outages result in high indirect costs for firms (e.g., due to running a generator) and can contribute to lost production. For SMLEs reporting outages in an average month in 2005 and most firms did report outages even before the crisis manufacturing firm managers reported losses that were equal to about 9 percent of sales on average (see Figure 10), with the median firm reporting losses of 5 percent of sales. 20 This also has a large negative effect on firm competitiveness. Although losses in Tanzania were similar or slightly lower than losses in other countries in the region, they were far higher than in the successful manufacturing economies in Africa and Asia. Average losses were less than 2 percent of sales in China, Thailand and Malaysia about one-fifth of the level of losses reported in 25

26 Tanzania Tanzania 2003 Kenya Rwanda Burundi Uganda South Africa China Thailand Malaysia Mauritius Losses (% of sales) Tanzania and the median firm reported losses of 0 percent in most of these countries (i.e., less than half reported any losses). This gives firms in these countries a large cost advantage over firms from Tanzania and will make it difficult for Tanzanian firms to compete with firms from Asia and even from regional competitors such as Kenya, which has a slightly better situation in this respect on international markets. Figure 10: SMLEs in Tanzania reported very high losses due to power outages Losses dues to unreliable power as percent of sales Mean Median Source: World Bank Enterprise Surveys. Note: Outliers more than three standard deviations from the mean are excluded. Cross-country comparisons are for manufacturing firms only. 42. Firms also reported high losses due to power outages in the 2003 Enterprise Survey. Although losses were high in Tanzania in 2005, they were also quite high in the 2003 Enterprise Survey. 21 In both cases the average firm reported losses equal to between about 7 and 9 percent of sales (see Table 3). Although as discussed in detail in Appendix 1.2 in Volume 2, differences in sampling might make it difficult to compare results from the two surveys, in this case the difference does not appear to be due to this. The similar levels of losses are visible looking only at firms that were interviewed in both 2003 and 2006 (i.e., the panel firms). Table 3: Losses due to outages were higher in 2003 than in % of sales lost due to outage (average) All manufacturing SMLEs Panel Firms % of sales lost due to outage (medians) All manufacturing SMLEs Panel Firms Source: World Bank Enterprise Surveys. Note: Averages for Panel firms are averages for firms that were in both the 2003 and 2007 surveys. Because the 2003 survey only covered manufacturing comparisons are only for manufacturing. 26

27 43. Losses during the crisis in 2006 are likely to have been even higher than the losses that SMLEs reported in It is important to note that these losses were for the fiscal year before the crisis hit Tanzania that is, the question was asked about fiscal year 2005 not fiscal year Since the crisis did not hit Tanzania until early 2006 and reported outages were far more frequent in 2006 than in 2005 it is likely that losses during the crisis were even higher than this. V. Access to Finance Although far fewer SMLEs and microenterprises in Tanzania said that access to finance was a serious obstacle than said the same about electricity, access to finance ranked as a greater problem than any other area of the investment climate. About 40 percent of SMLEs and about 50 percent of microenterprises said that access to finance was a serious constraint for their enterprises. Financial Sector Development 45. Tanzania s banking sector has been significantly restructured over the past twenty years. At the beginning of the 1990s, Tanzania s banking sector was highly concentrated and was dominated by state-owned banks. Since the early 1990s, the Government of Tanzania has implemented a series of reforms that have liberalized banking and reduced the dominant role of the state Despite these reforms, the financial sector remains small and underdeveloped by international standards. Credit to the private sector was equal to about 15 percent of GDP by mid-june This is slightly larger than in Uganda and about the same size as in Rwanda but smaller than in the other comparator countries (see Figure 11). Credit to the private sector was equal to about 35 percent of GDP in Burundi, Kenya and Swaziland and between 80 and 120 percent of GDP in Mauritius, Thailand, South Africa, Malaysia and China. Other measures of financial sector development such as bank credit and money and quasi-money tell a similar story. The stock market is also relatively small and illiquid. 27

28 47. Under these circumstances, it is not surprising that firms in Tanzania remain heavily dependent upon retained earnings to finance both new investment (see Figure 11) and working capital. Although SMLEs in Tanzania use bank financing to similar degrees as SMLEs in other countries in the region in terms of financing new investment and working capital, the difference between Tanzania and comparator countries outside of the region is considerably larger. Firms in Tanzania finance about 8 percent of new investment with bank financing and about 85 percent with retained earnings. In comparison, firms in South Africa finance 16 percent of their new investment with bank financing and only 59 percent with retained earnings. Firms in Mauritius and the Asian comparators finance over one-third of their new investment with bank financing and less than half with bank financing. Figure 11: The financial sector is not highly developed in Tanzania. Credit to private sector New Investment China Malaysia South Africa Thailand Mauritius Swaziland Kenya Burundi Rwanda Uganda South Africa Mauritius Malaysia Thailand Kenya Uganda Swaziland Rwanda Burundi Tanzania % of GDP Tanzania 0% 25% 50% 75% 100% % of new investment financed in different ways Retained Earnings Banks Trade Finance Other Source: World Bank (2008c); World Bank Enterprise Surveys. Note: Data for private credit for Uganda are for 2006 and data for Rwanda are for All other data on private credit are for Access to long-term financing appears to be particularly difficult. Long-term loans of over 5 years made up between about 4 and 7 percent of lending in Tanzania between June 2006 and June This suggests that long-term financing is relatively rare. Evidence from the Enterprise Survey is consistent with this. Only 10 percent of firms with loans reported that their most recent loan was for more than 50 months. Similarly, firms appear to have far better access to short-term credit for working capital than long-term financing for new investment. Whereas SMLEs report that they finance more of their working capital with bank financing than SMLEs in Uganda and about the same amount as in Kenya or Swaziland, they finance less new investment in this way than SMLEs in any of the comparator countries. 49. There is, however, some evidence that access to finance has improved since As noted earlier, changes in the survey questions mean that it is difficult to compare perceptions about access to finance between the 2003 and 2006 surveys. Some objective evidence, however, 28

29 suggests that access might have improved since Panel firms firms interviewed in both 2003 and 2006 were more likely to have loans and overdrafts in 2006 than in Other evidence from comparing results from the two surveys which is difficult due to changes in sampling also suggests that access to loans appears to have improved or stayed the same since It is harder, however, to reach strong conclusions with respect to overdrafts. Access to financing for firms of different sizes 50. As in most countries, larger firms have better access to credit than smaller enterprises do. About 83 percent of large enterprises had bank financing loans, lines of credit, or overdrafts compared with 24 percent of small and medium-sized enterprises (SMEs) and 17 percent of microenterprises (see Figure 12). Larger firms were also less likely to say that access to finance was a problem than small firms. Whereas about 28 percent of managers of large enterprises said that access to finance was a problem, about 40 percent of managers of small enterprises said the same. 24 At least in part, this probably reflects that larger, more formal and more profitable firms are easier to lend to in a sustainable way. 25 Figure 12: Large enterprises in Tanzania have good access to credit. Microenterprises SMEs Large Enterprises Swaziland South Africa Rwanda Mauritius Thailand South China Swaziland Thailand Mauritius South China Swaziland Kenya Burundi Uganda Kenya Rwanda Burundi Uganda Burundi Kenya Rwanda Uganda Tanzania Tanzania Tanzania 0% 25% 50% 75% 100% % of firms with bank credit 0% 25% 50% 75% 100% % of firms with bank credit 0% 25% 50% 75% 100% % of firms with bank credit Source: World Bank Enterprise Surveys. Note: Credit means firm has either a loan or overdraft. Cross-country comparisons for SMLEs are only for manufacturing SMLEs. Microenterprise data is only available for countries in SSA. Because of small sample sizes for microenterprise samples, all enterprises are included for microenterprises. 51. Although it is common for large enterprises to have better access than small enterprises do, the difference appears large in Tanzania (see Figure 12). In most countries, there is some degree of market segmentation with respect to access to finance. The gap, however, appears to be particularly large in Tanzania. Whereas manufacturing SMEs in Tanzania are less likely to have bank financing than similar firms in almost all of the comparator countries, large 29

30 manufacturing enterprises in Tanzania are more likely to do so than similar firms in many of these countries (see Figure 12). For example, whereas fewer SMEs have loans or overdrafts than in Swaziland, China or South Africa, more large enterprises do. In this respect, the financial sector appears to serve large firms as well as financial sectors in most other countries serve large firms. The same is not true for SMEs. 52. Microenterprises, in contrast, tend to have better access than similar firms in the comparator countries in Sub-Saharan Africa. 26 Although SMEs were less likely to have bank credit than in the other comparator countries except Uganda, microenterprises were more likely to have bank credit than in some of the comparators. Only about 10 percent of microenterprises in Uganda and 12 percent of microenterprises in Burundi had either loans or overdrafts compared with 17 percent of microenterprises in Tanzania. In this respect the gap between microenterprises and SMLEs is more modest in Tanzania than in most other countries. Reasons for not having loans 53. Although many SMLEs and even more microenterprises in Tanzania do not have any bank financing, it is important to assess why this is the case. Have many firms without financing had loan applications rejected? Or do they not want to take on debt? Or are the terms of the loans that are available unattractive? The Enterprise Survey asks firms some additional questions to address some of these issues. 54. Despite the small number of firms with bank financing, this does not appear to be because many firms have loan applications rejected. In addition to asking firms whether they have bank financing, firms were also asked whether they have applied for a loan recently and if not, why not. Few firms reported that they had applied for a loan in the previous fiscal year only about 19 percent of SMLEs and 21 percent of microenterprises. Given the low number of applications, it is not surprising that few firms had had applications rejected less than 7 percent of SMLEs and 12 percent of microenterprises had had an application rejected. This suggests that self-selection is taking place. 55. But this self-selection does not appear to be because firms do not want loans. Given the high level of concern about access to finance, it is not surprising that the low level of bank financing does not appear to be primarily due to firms not wanting loans. For firms that did not apply for a loan, only about one quarter said that that they did not need a loan (see Figure 13). Although this shows that some firms do not want or need loans, fewer firms said that this was the case in Tanzania than in the comparator countries with comparable data 38 percent of firms in Kenya and 70 percent of firms in Swaziland that did not apply for a loan said they did not need one. Firms in Uganda, Burundi and Rwanda that had not applied were also more likely to say that they did not need one (37 percent, 36 percent and 44 percent respectively). 56. Another common reason that firms gave for not applying for loans was that application procedures were too complex. Over one-quarter of firms said that this was the main reason why they did not apply. This was far higher than in most of the comparator countries for example, only 9 percent of firms in Kenya and 6 percent of firms in Uganda said the same. SMLEs that already had a loan were only slightly less likely to say that this was the case than SMLEs without loans (20 percent compared with 26 percent). Also consistent with this, about 30

31 one-quarter of firms that had applications rejected reported that it was because their application was incomplete potentially suggesting complex application procedures. Figure 13: Many firms without loans reported that they did not want a loan. Size or Maturity are insufficient 7% Collateral requirements too high 14% Other 10% No need for loan 24% Interest rates are not favorable 20% Application procedures too complex 25% Source: World Bank Enterprise Survey. 57. Another common response was that interest rates were not favorable about one-fifth of firms said that the main reason they had not applied was that interest rates were too high. Although this is quite high, it was lower than in either Uganda (36 percent) or Kenya (27 percent), where high interest rates were the most common response for not applying other than that the firm did not need a loan. In this respect, interest rates appear less binding in Tanzania than in these two countries. Other responses (e.g., about collateral requirement of insufficient maturity) were less common. VI. Informality Although competition with informal firms was a lesser concern than electricity or access to finance, it remains a serious concern for many firms (see Figure 14). About 29 percent of SMLEs and about 25 percent of microenterprises, some of whom might also be informal, said that competition from the informal sector was a serious constraint on doing business. 31