Impact of Access to Finance, Corruption and Infrastructure on Employment Growth: Putting Africa in a Global Context

Transcription

1 Impact of Access to Finance, Corruption and Infrastructure on Employment Growth: Putting Africa in a Global Context Mary Hallward-Driemeier and Reyes Aterido * October 2007 Abstract: Comparing employment growth rates across Sub-Saharan Africa with that of the rest of world, the averages are very similar. However, the composition is different. In Africa, employment growth is relatively concentrated in the smallest firms, with medium and large firms growing less rapidly than in other parts of the world. Part of this can be understood by the business environment in which firms operate. On average, firms in Africa face greater challenges in accessing finance, reliable infrastructure services and other public services, all conditions that serve to shift downward the distribution of firms. Unreliable infrastructure not only lowers the growth of large firms, it encourages the growth of micro-firms in Africa. Part of this is due to its effects on the choice of technology and by encouraging greater substitution of capital for labor. Unfortunately, improved access to finance and public services areas associated with positive effects on employment growth of all firms are less beneficial in Africa, particular for micro-firms. Thus, the same level of finance translates into smaller increases in employment growth in Africa compared to other developing regions. That micro firms are expanding employment opportunities can be a good thing, but business environment conditions that do not provide incentives for them to keep growing larger raises the stakes for reforms. * The views expressed here do not necessarily represent the views of either the World Bank or the Inter- American Development Bank or their Boards of Executive Directors.

2 I. Introduction Recent years have seen an encouraging increase in GDP per capita growth in Sub- Saharan Africa as greater macroeconomic stability has been achieved in more countries and additional reforms have been undertaken. Regional growth rates over the last five years have been higher than in Latin America and the OECD countries, with the average growth rate just under 5 percent. This paper looks at the patterns of employment growth comparing the recent experience in Africa with other regions, and comparing the patterns across different sizes of firms. 1 This paper focuses on the role that the business environment plays in explaining the patterns of growth across firm sizes and how this contributes to differences between Africa differs from other regions. It uses data from over 11,500 firms in 32 Sub-Saharan countries included in the World Bank s Enterprise Surveys. To put the experience of Sub-Saharan Africa into a larger global context, results are compared with those of 67 developing countries in other regions. Key areas that are examined include access to finance, infrastructure services, regulations and governance. Particular attention is paid to the impact of these conditions across the spectrum of firm sizes. The data shows that the overall rate of employment growth in Africa is similar to the average in the other regions (Table 1). However, what is different is the types of firms that are growing across the regions. When one includes high-income countries in the comparison, significantly more of this growth is concentrated in the smallest firms in Africa (those with less than 10 employees), while medium and large firms are growing slower than those in other regions. 2 However, when high-income countries are excluded, the pattern by size of firms remains, although no longer at the same level of significance. Whereas young firms are generally relatively more dynamic, we also find this is less true in Africa. Firms in smaller cities and towns are growing more slowly than those in the capital or other larger cities, with the gap even larger in Africa. The paper traces how the business environment in which a firm operates accounts for these growth patterns. It finds that a weak business environment shifts downwards the distribution of firms. It does this in part by lowering the growth of larger firms. But excessive red-tape, corruption and unreliable infrastructure can also have the effect of raising the relative growth of micro firms. Thus, it is possible to have high employment growth even with weak business environment conditions, but the effects are likely to be 1 The paper looks at patterns across Africa in comparison with other developing country regions. Clearly there are differences within Africa; the approach used here could also be applied to an analysis of single countries. 2 One question is whether the higher growth of micro firms is translating into more firms entering the next size category and succeeding in becoming small. Unfortunately this does not appear to be the case. In the other developing regions, 7.8 percent of micro firms grew sufficiently to pass the 10 employee mark, but only 3.8 percent of firms in Africa did so. Other firm characteristics have the expected sign in terms of their correlation with growth. New firms are growing faster, which is not surprising as many are also quite small in size. Foreign owned firms are growing about 4 percent faster and exporters 8.5 percent faster. On the other hand, government owned firms are growing 2.5 percent slower. Compared to garments (the omitted category), textiles, food and agro-industry, and chemicals/plastics are growing faster, while retail employment somewhat slower. 1

3 seen in the composition of who is growing. In the case of Africa, average growth has been strong, but it has been strongest among the smaller firms. In analyzing the evidence, the paper examines both differences across regions and within a country across different sizes of firms. It finds significant differences in the impact of the business environment stemming from two sources. First, objective conditions can vary across countries or even within country for different sizes of firms. Second, the same business environment conditions may lead to different employment responses by different groups of firms. Objective conditions are often more challenging for firms in Africa. Power outages are more common, firms receive less formal external financing and can face greater delays in their interactions with officials. The paper also find that micro and small firms face significantly greater interruptions in infrastructure services, have less access to formal finance and pay more in bribes as percentage of sales-- than do larger firms. On the other hand, larger firms spend significantly more time dealing with officials and red tape. Thus, the differences in the underlying objective conditions faced by firms, even within the same country, are substantial. While these weaknesses in the business environment have been found to retard growth (Dollar et al., Aterido et al.), employment growth is comparable or even higher in Africa. This is explained differences in response to the same condition by different firms. On the one hand, responses to business environment conditions are often more muted in Africa. And, second, the pattern of discouraging the growth of large firms while micro firms expand is more pronounced in Africa. The analysis finds that for the same frequency of outages, firms in Africa cut their employment by less. While delays and inefficient services dampen growth, more interactions with government officials are generally associated with higher employment growth outside of Africa. This is consistent with these interactions leading to access to needed services and that these services are beneficial to firms. 3 However, there is not this beneficial association in Africa where there is little correlation between interactions with officials and employment growth. On the other hand, corruption is associated with lower growth in other regions, but not so in Africa. Improved access to finance is one area that leads to higher growth across the size spectrum. However, the extent of the benefits for the same level of finance is lower in Africa. Second, the paper also finds evidence that the same conditions have differential or non-linear effects on employment growth by size. Thus, the same extent of external finance is more beneficial to smaller firms. Less reliable infrastructure (particularly losses from electricity outages) is a significant deterrent to employment growth. However, there is also evidence that its effects are more detrimental for larger firms and that it can encourage growth of smaller, less capital intensive firms, particularly in Africa. Inefficient government services slow the growth of larger firms while encouraging firms 3 There is a limit to this finding. Adding a quadratic term shows that additional interactions can have a negative marginal effect, and even a net negative impact at levels actually experienced by about 20 percent of firms in the sample. 2

4 to remain small. Access to finance and the quality of government services thus emerge as the two issues that help account for the relatively faster growth of micro-firms in Africa. That firms in Africa respond to challenges in the business environment by reducing employment by less helps to keep employment growth relatively high in the region. However, the substitution away from capital and the growth of micro firms over larger firms implies that productivity will be lower. Raising the growth of microenterprises is a desirable outcome-- but it would be more encouraging if business environment reforms could help achieve greater growth rates among SMEs and large firms too. This would also strengthen the incentives of firms not merely to expand in the informal sector but to join the formal sector proper and to increase the productive use of their resources. II. Literature Review There are two strands of literature that this work draws on. The first is the growing importance given to measures of institutions or the business environment in which firms operate and how they can impact employment growth across countries. The second is work done at the firm-level within Africa. a) Business environment and employment growth across countries There is interest in looking at which dimensions of the business environment matter for growth and improved performance. Publications such as the Global Competitiveness Report and Doing Business are widely quoted and their rankings quoted in the media and by those advocating reforms that would improve their country s rankings. The macroeconomic literature stresses the importance of institutions in explaining longer run growth patterns. Several papers that examine a particular dimension of the business environment (e.g. finance or corruption) have found significant results. However, looking only at a single dimension carries the risk of omitted variable bias. And these aggregate indicators do not allow one to explore the rich diversity of experiences within a single country. The Enterprise Surveys include firm-level measures of a wide array of business environment indicators. While subjective rankings of constraints are collected, most indicators focus on the costs (monetary or time) associated with completing various transactions or interactions with government officials. This work uses this data from 100 countries to put Africa s employment growth experience in a larger context. This work builds on earlier studies of the effects of regulations on employment growth (Botero et al, 2004; Heckman and Pagés, 2004, Micco and Pagés (2006)), addressing three potential shortcomings in the papers discussed above. First, the data contains actual measures of enforcement of regulations rather than de jure formal requirements that may have little relation to what is experienced on the ground. Second, there are significant variations in the business environment not only across countries but also within countries, e.g. across different sizes of firms. Third, with a few exceptions, studies tend to focus on one given factor or policy, such as firing and hiring regulations (Botero et al. 2004), regulations on new businesses (Klapper et al. 2006), or restricted access to finance (Levine 2005). They are, therefore, potentially liable to omitted 3

5 variable bias, as other non-considered aspects of the business environment can also affect firms decisions. The focus here is on regressions that combine multiple dimensions of the business environment in a single regression. This not only addresses potential bias in the estimates, it allows one to test directly which dimensions have the biggest impact on firm performance. b) Business environment and firm performance in Africa The collection of new firm-level datasets in Africa 4 have enabled new research into the determinants of firm performance within Africa. Eifert, Gelb and Ramachandran (2007) show how much higher indirect costs of production are for African firms compared to firms in Asia. Higher costs for transportation, electricity, water, security, marketing and accounting services can more than offset any advantage from greater productivity on the factory floor or from lower wage costs. A number of papers have looked at the effects of different individual dimensions of the business environment. Thus, Bigsten et al. find evidence of credit constraints among manufacturing firms in 6 countries in Africa. Collier and Gunning argue that poor infrastructure is significant in promoting the proliferation and growth of small firms as local markets remain small. However, Clarke finds only weak evidence that transportation constraints lower exports. Reinikka and Svensson 2002 find that weak public infrastructure lowers private investment using firm data from Uganda, thus lowering capital intensity. Fisman and Svensson 2007 use panel data from Uganda to show that corruption has significant effects in lowering sales growth, effects three times larger than the effect of taxes (Fisman and Svensson 2007). Beyond the direct benefits to individuals of expanding employment opportunities, there are additional benefits to having firms grow. One is the effect on productivity. With greater economies of scale, firms can raise their productivity and use their inputs more efficiently. They can be more likely to adopt more advance technologies, reinforcing greater productivity (Collier, 2000; Bigsten andd Soderbom 2006). And size has been found to be a significant predictor of the probability of exporting, controlling for sector and capital intensity (Rankin, Soderbom and Teal 2006). Given the smaller market size in Africa, this offers a significant way to expand market opportunities. It has also been found to have learning-by-exporting benefits as firms gain greater knowledge about additional production, managerial and distribution techniques (Bigsten et al. 2004; van Biesebroeck 2005). These studies indicate that business environment conditions do affect firm decisions. This paper is the first to examine a range of them together and to link them to employment growth in Africa. It is also the first to undertake a systematic comparison between Africa and the broader set of developing countries. It also addresses endogeneity concerns associated with subjective rankings of constraints by using 4 A significant source of such data are the Regional Program on Enterprise Development (RPED) surveys carried out by the World Bank since the early 1990s. They were a precursor to the current Enterprise Surveys that have expanded the objective indicators collected as well as being fielded in countries around the world. 4

6 objective measures and by using location-sector-size averages instead of firms own responses. It also considers whether selection is an issue driving the results. III. Data To better address questions about business environment conditions and their impact on the performance of a wide variety of firms, the World Bank launched its program of Enterprise Surveys in To date, it has interviewed over 70,000 entrepreneurs and senior managers in 99 developing countries and 8 high income countries. Of particular interest here are the 11,600 interviews in 32 countries in Sub- Saharan Africa. The Enterprise Surveys have four distinguishing features that make it particularly useful in this study (see appendix for more details). First, it can benchmark not only subjective rankings of investment constraints to business performance (e.g. the extent to which electricity is rated as a problem), but also objective measures of these constraints (e.g. the frequency and duration of outages, production lost from outages, and the use and cost of generators). Second, it covers a wide range of issues: from access to financial and infrastructure services, crime, corruption and government regulations, allowing a ranking of these issues. Third, the data can also go beyond benchmarking to test directly the impact of these objective conditions on the actual performance of the firm, how the actual business environment conditions affect the productivity and employment growth of respondents. Fourth, large, randomly selected samples of firms allow for results to be compared across types of firms, with particular attention paid to firm size. For many of the countries in the region, this is the only source of detailed information on firm performance and disaggregated objective indictors of a wide variety of business environment indicators. This paper looks at Africa in comparison to the rest of the world (ROW). The ROW, in fact, is largely the rest of the developing world. There are 5 west European countries included, but 67 are other developing countries. Combining the other regions does hide differences between East Asia and Latin America, or South Asia and Eastern Europe. However, it also made it harder to see the broader story of how trends in Africa stand out. Overall, the general pattern of which dimensions of the business environment matter are consistent across all the regions, so that combining the other regions in a comparison with Africa seemed justified. IV. Business environment: Comparing constraints between Africa and Rest of Developing World To examine the possible role of the business environment in explaining these patterns, we begin with what entrepreneurs themselves identify as leading obstacles. Respondents are asked to rank 17 potential constraints and the degree to which they are obstacles to the operation and growth of their business. 5 Figure 1 shows the share of 5 Given concerns about differences in respondents willingness to complain, the ratings are converted to a relative score. Subtracting the mean complaint for each respondent acts like an individual fixed effect (see Hallward-Driemeier and Aterido, 2007). The relative rankings of issues should help motivate which issues are worth exploring in more detail. Infrastructure, finance, regulation and corruption are all listed as top 5

7 firms that report a given obstacle as being an above-average constraint. Electricity and access to finance stand out both for the share of firms that report these issues as problematic in Africa and because of the gap with other regions. Electricity is the top constraint in the Africa region, but is much lower down the list in all other regions except South Asia. On the other hand, a number of potential issues are reported as relatively less concerning for African entrepreneurs. Of particular interest are issues of interactions with the government (e.g. dealing with licenses) and corruption that are reported as relatively less constraining than in other regions. 6 One of the benefits of the Enterprise Survey data is that these subjective responses can then be correlated with more objective measures of how these areas of the business environment impact firms. As shown in Figure 2, there is a reason why infrastructure is the area with the biggest differences in severity of constraint. Firms in Africa experience greater delays in getting access to infrastructure service, experience more frequent interruptions in service and incur greater losses from these interruptions than do firms in other regions. The time to get a phone line is 57 percent longer, the frequency of power outages is more than twice as high in Africa, and interruptions of water is 80 percent higher. The percent of production lost due to power outages is 4 percentage points higher in Africa. Clearly this is an area where African firms face greater challenges than there counterparts in most other regions, affecting their choices of technology, capital intensity and growth potential. Access to finance is another area where African entrepreneurs face greater challenges. This is both true for simpler financial services such as an overdraft, where on average 12.3 percent fewer firms have access to an overdraft in Africa, as well as longer term bank loans. The share of working capital financed by banks is 3.5 percentage points lower in Africa, which is about a fifth lower given the reliance on retained earnings. Even trade credit extended between firms is lower. These findings corroborate emphasis place on electricity and finance in the subjective rankings. On the other hand, corruption was seen as relatively less constraining. The frequency of bribes is reported as comparable across the regions, although the frequency with which gifts are expected during inspections is somewhat lower. However, the cost of the bribes paid to get things done as a share of sales is 0.89 percentage points higher in Africa. Yet, this is not reported as constraining. As discussed below, the impact of these bribes on employment growth is actually more muted in Africa. So, indeed, the evidence seems consistent with the perception that corruption is less of an obstacle, even if it is as prevalent. Figure 2 also shows the extent to which objective conditions not only vary across regions, but also across firm sizes. Within Africa, micro and small firms spend less time with officials, while larger firms, exporters and foreign firms spend a great deal more constraints in at least one region, but there are also particularly striking here for the differences in their relative importance across regions. 6 A complete list of constraints is available on request and is discussed in the longer Working Paper version of this paper. One interesting point to note too is that while foreign owned firms report being relatively less constrained in the other regions, this is not true for Africa. Foreign firms are particularly concerned with corruption and crime, the two areas that had received lower rating overall in Africa. 6

8 time. On the other hand, larger firms in Africa spend less time in inspections. The evidence on access to finance shows that trends that favor larger firms are even more prevalent in Africa. The pattern on corruption is that larger firms are relatively less likely to pay bribes and payments too are higher the smaller the firm. The infrastructure variables do not show a different pattern by size in Africa compared with other regions. This evidence on the objective conditions demonstrates that there are large and significant differences, not only in conditions facing African firms compared to those in other developing countries, but conditions also vary within countries. Smaller firms do face greater costs associated with corruption and have less access to finance. Larger firms tend to spend more time with officials and have more inspections, although less so in Africa. The next section then looks to see how these differences then translate into employment growth in different regions and across the size distribution of firms. VI. The Business Environment and Employment Growth Table 2 then relays these objective measures of the business environment to firms growth performance. The first three columns control for country characteristics (GDP per capita, GDP growth, inflation, exports) and thus include both within country and between country variation. A dummy is included for Africa. It remains insignificant (col 1); controlling for country characteristics and measures of its business environment Africa s growth is not different from the other regions. 7 The Africa dummy is also interacted with the four main areas of interest of the business environment to test if the impact of the business environment differs across regions (each is discussed separately below). Incorporating multiple dimensions of the broader business environment simultaneously deals seriously with concerns of omitted variable bias of papers that only include a single dimension, e.g. labor regulations or finance. 8 The second set of columns in Table 2 then looks in more detail at how the business environment impacts firms of different types (columns 4-7) 9. They report the effects of the within-country variation, with the results for Africa run in parallel for those of the other regions. The business environment variables are then interacted by size dummies to test for non-linear responses. 7 Using an alternative measure for infrastructure (frequency of outages rather than cost of outages) does yield a positive dummy for Africa (col 2); given the frequency of outages, Africa grows unexpectedly fast. However, the dummy then disappears again when one controls for the use of generators (col. 3), a private solution to disruptions on the public grid. Indeed, given distributions are more frequent in Africa, it is perhaps not surprising that the use of generators (both in terms of numbers of firms that use them and the share of electricity they can represent) is higher in Africa. 8 One concern is that endogeneity is operating at a different level, that firms might be selecting locations based on the quality of their business environment. A correlation between a better environment and better performance could simply reflect that better firms are more likely to locate there. To address this concern, the results are repeated, excluding all firms that are most likely to be footloose, or likely to choose a location other than the one where the entrepreneur is from, i.e. foreign owned firms and large domestically owned firms. The results are extremely robust, casting doubt that selection is much of the story. 9 To better capture potentially omitted variables capturing economy conditions, the regressions include country dummies. The sample for Africa is then run in parallel to the one for the other regions combined. 7

9 Finance: Of all the areas in the business environment, improved access to finance has clear benefits to firms of all sizes (Table 2). A 10 percent increase in the share of investments financed through bank loans (equivalent to doubling the average share) is associated with a 3 percent increase in employment growth. This result is robust to alternative measures of finance, including formal bank financing of investment to trade credit among firms (Table 3, Col 1-2). There is some indication the benefits of external finance are not as extensive in Africa, but the effect is not significant at the 10 percent level. When comparing the impact across different firms within regions, the impact of improved finance is still positive and significant in Africa, but the coefficient is one third of that of the other regions. The benefits of finance are most strongly felt among the smallest firms. These are the firms that have the lowest levels of access and have the largest response to a given increase in finance. For large firms in Africa, the overall result is often not even significant. This somewhat more muted response to finance in Africa could well be due to interactions with other dimensions of the business environment. Improving access to finance alone does not address other important constraints that firms could be facing. Infrastructure: Losses stemming from power outages lower overall employment growth in all regions (Table 2, Col 1). While these losses are greatest in Africa, for the same loss, firms in Africa change employment levels by less. Looking simply at the frequency of outages provides insights into this finding. Controlling for the frequency of outages results in a positive coefficient on the Africa dummy variable; given the frequency of outages, Africa has stronger than predicted job growth (Table 2, Col 2, 6-7). Some of this can be explained by the need for firms to rely on their own generator (Table 2, Col 3). Given the more frequent interruptions, firms in Africa are more than 20 percent more likely to have a generator. Controlling for whether a firm has a generator removes the significance of the Africa dummy. Another explanation is also clear when looking across firms of different size. Looking within countries, while the overall impact of outages is negative on medium and larger firms, for the micro firms is actually positive (Table 4, Col 1). The effect of outages thus appears to be to tilt growth towards the micro firms. Part of this pattern stems from micro-firms being less vulnerable to costs associated with outages as they have made different choices in terms of technology. To test this, we look at firms capital intensity and changes in capital intensity (Table 4). These results are only available for a subset of countries as the information was not included in all the surveys. The results do indicate that in areas with more frequent outages, firms are likely to have significantly lower capital intensity, particularly at the micro level. Looking at changes over time, firms in Africa are more likely to be lowering their capital intensity when outages are more frequent. This provides evidence that the unreliable infrastructure is a large part of the story of growth opportunities being greater for micro-firms, in part as they substitute away from capital to labor. Regulatory environment: There are three dimensions of the regulatory environment that are examined (Table 2 and Table 3, Col 5-8). Most often studies equate 8

10 regulation with red tape and highlight its negative effects. Such an interpretation can be confirmed here, using as a measure the time it takes to get goods through customs (Table 3, Col 5-6). Longer delays and bureaucratic inefficiencies are costly and lower growth. However, this is not the only dimension that matters. A second measure looks at the consistency with which regulations are enforced (Table 3, Col 7-8). In the rest of the developing world, greater consistency has a significant positive impact on growth. And it is an effect that holds across all sizes of firms. Unfortunately, this is not true in Africa. In Africa, the overall effect is not significant. A beneficial impact is only significant for the largest firms. This is an area where improvements could have a large impact on improving smaller firms to grow. A third measure, the overall time managers spend with officials, appears to provide a middle ground between the other two measures. On the one hand, interactions with officials are associated with access to some public services. And this has a significant positive effect on employment growth, an effect that is particularly concentrated among micro firms (the effect is half that of micro firms for medium and large firms). This is good news for encouraging the smallest firms to grow and underscores the potential benefits of becoming more formal. However, there are two caveats to this result. First, there is a limit to the benefits. After a point, continuing to raise the time managers spend with officials loses its positive impact. At about 15 percent of management time, the marginal impact of additional interactions with the government is negative. This is the case for over a fifth of the firms in the sample. By 25 percent of time, the overall impact is actually detrimental. Second, the effect in Africa is again different from the other developing regions. The benefit to micro firms is present. However, it is offset for small, medium and large firms, such that there is no significant impact of management s time with officials on their growth. One interpretation is that the benefits of gaining access to government services do not exceed the costs of complying with regulations and paying taxes. Thus, greater time with officials can shift growth towards the smallest firms that are better able to stay under the radar screen and avoid as many of the regulatory requirements. The lack of benefits even for larger firms does bring into question the quality and efficiency of the services provided. Together, these three results highlight the importance of improving the efficiency and quality of public services to encourage private sector growth. Interactions with the government can clearly have benefits, but currently the evidence shows this is less the case in Africa. This dulls the incentives to become formal and results in lower growth rates for small, medium and large firms. Governance: The findings on corruption within Africa reinforce the need to address regulatory consistency and the importance of improving the incentives of firms, particularly small ones, of wanting to interact with the government. Within countries, the detrimental impact of corruption is significant in the other developing regions overall but not so in Africa. While there are no particular differences 9

11 in impact across sizes of firms in the other regions, in Africa, there is a negative impact, but it is concentrated among small firms (Table 2, Col 4-5). As these are ones that are often just over the threshold in terms of higher regulatory standards, this finding indicates a discouraging incentive for firms to expand. 10 Overall: African firms seem less sensitive to conditions in the business environment, for the same business environment obstacles they respond by lowering employment by less. They reduce employment by less in the face of unreliable infrastructure and corruption; they also raise employment by less from interactions with officials and access to finance. However, underlying this is a compositional shift downwards in the distribution of firm sizes. In particular, it is micro firms that grow relatively more when outages are more frequent and when the net benefits to greater regulation are low (or even negative). On the other hand, if micro firms grow relatively more in the challenging business environment, it is small firms that stand to gain the most from improvements. 11 This is particularly true when combining the different objective conditions faced by firms of different sizes with the differences in marginal impact of the conditions on growth. While high employment growth is desirable, improving these key elements in the business environment will improve the incentives of all firms to expand and to use resources more productively. VIII. Conclusion The paper has documented how the business environment in which firms operate varies significantly across types of firms. Firms in Africa do face greater obstacles in terms of finance, infrastructure, public services and governance. And within Africa, the constraints are often greatest for the smallest firms. What is striking, however, is that weak business environment conditions do not necessarily translate into lower employment growth. Rather, the more challenging business environment conditions are associated with shifting down the size distribution, lowering the relative growth of larger firms, or in some cases expanding micro-firms. In particular, unreliable electricity and less efficient government services are associated with relatively higher growth for microfirms and lower growth for larger firms. This is consistent with firms choosing less capital-intensive production technologies and incentives to stay below the radar screen of officials. The evidence in this paper points to four areas where reforms could improve the incentives of all firms to grow, and, in particular, for micro firms to join the ranks of more formal small firms: First, improving access to finance would help raise firm growth. It would particularly benefit the smaller firms that have less access and for whom a given increase in finances is associated with a greater increase in employment. 10 For bribes, an alternative measure of the frequency of bribes is found to be less significant than the actual amount of money paid. This is certainly plausible as presumably it is the actual monetary costs that matter, and whether it is paid in one or in five payments is secondary. 11 In fact, in other work (Aterido et al. 2007), the merging of micro and small firms into a single category or the elimination of the micro firms from the analysis greatly mutes the message that it is the smaller firms that would benefit from a stronger investment climate. 10

12 However, improved access to finance alone is not sufficient, particularly where other constraints are significant. Second, addressing weaknesses in infrastructure, particularly reliable electricity and transportation would reduce the distortions affecting the choices of technology and capital intensity that should lead to higher productivity and the more efficient use of resources. Third, improving government services would significantly lower incentives to remain very small so as to avoid costly burdens of compliance and increase the benefits of public services (property rights, access to credit etc.). In Africa, where there is less evidence of the beneficial effects of interactions with government official, work to reduce red-tape and to improve the quality of government services would do a great deal to help firms grow and to encourage formality. Fourth, corruption is a corollary to government efficiency and is associated with the extent of discretion exercised by officials. Lowering discretion has a large payoff, but currently in Africa it is only shared by the largest firms. Again, this is a disincentive for the small firms to grow. While a weak business environment is associated with lower growth, this effect is more muted in Africa. Underlying this are incentives that encourage the expansion of micro firms. While greater employment is surely desirable, reforms that would improve the incentives and opportunities for firms to grow into SMEs and larger firms would have even greater payoffs for employment and the productive use of resources. 11

13 References Azam, Jean-Paul, Augustin Fosu, and Njuguna S. Ndung u Explaining Slow Growth in Africa. African Development Review 14 (2): Aterido, Reyes, Mary Hallward-Driemeier and Carmen Pages Investment Climate and Employment Growth: The Impact of Access to Finance, Corruption and Regulations Across Firms. World Bank Working Paper. Beck, Thorsten, Aslı Demirgüç-Kunt, and Vojislav Maksimovic Financial and Legal Constraints to Firm Growth: Does Firm Size Matter? Journal of Finance 60 (1): Biggs, Tyler, Vijaya Ramachandran, and Manju Shah (1999). The Determinants of Enterprise Growth in Sub-Saharan Africa: Evidence from the Regional Program for Enterprise Development, RPED Discussion Paper 135. Washington D.C: The World Bank. Bigsten, Arne, Paul Collier, Stefan Dercon, Marcel Fafchamps, Bernard Gauthier, Jan Willem Gunning, Abena Oduro, Remco Oostendorp, Cathy Pattillo, Måns Söderbom, and Albert Zeufack Credit Constraints in Manufacturing Enterprises in Africa. Journal of African Economies 12 (1): Bigsten, Arne and Måns Söderbom What Have We Learned From a Decade of Manufacturing Enterprise Surveys in Africa? World Bank Research Observer. Oxford University Press, vol. 21(2), pages Collier, Paul, and Jan Willem Gunning. 1999a. Explaining African Economic Performance. Journal of Economic Literature 37 (1): b. Why Has Africa Grown Slowly? Journal of Economic Perspectives 12 (3): Demirgüç-Kunt, Aslı, and Vojislav Maksimovic Institutions, Financial Markets, and Firm Debt. Journal of Financial Economics 54 (3): Dollar, David, Mary Hallward-Driemeier and Taye Mengistae Investment Climate and Firm Performance in Developing Countries. Economic Dvelopment and Cultural Change. vol. 54, issue 1, pages 1-31 Eifert, Benn, Alan Gelb, and Vijaya Ramachandran Business Environment and Comparative Advantage in Africa: Evidence from the Investment Climate Data. RPED Report 126. Washington, DC: World Bank. Fafchamps, Marcel and Måns Söderbom (2004). Climate for Job Creation. mimeo. Part of Employment Issues Regional Stocktaking Review, Africa Region of the World Bank. Washington, D.C: The World Bank. 12

14 Fisman, Raymond. and Jakob Svensson Are corruption and taxation really harmful to growth? Firm level evidence. Journal of Development Economics, 2006 Hallward-Driemeier, Mary and Reyes Aterido Comparing Apples to Apples: How to Make (More) Sense of Subjective Rankings of Constraints to Business. World Bank Working Paper. Rienikka, Ritva and Jakob Svensson, 2002, Coping with Poor Public Capital. Journal of Development Economics, 69(1), Söderbom, Måns and Francis Teal (2004), Size and Efficiency in African Manufacturing Firms: Evidence from Firm-Level Panel Data, Journal of Development Economics 73, pp Söderbom, Måns, Francis Teal and Alan Harding (2006) The Determinants of Survival among African Manufacturing Firms Economic Development and Cultural Change, volume 54 (2006), pages Tybout, J. (2000), Manufacturing Firms in Developing Countries: How well do they do, and why? Journal of Economic Literature 38(1): Van Biesebroeck, Johan, Firm size matters: growth and productivity growth in African manufacturing. Economic Development and Cultural Change 53 (3), (April). World Bank World Development Report: A Better Investment Climate for Everyone. Washington, DC: World Bank Doing Business in 2006: Creating Jobs.Washington D.C: The World Bank. 13

15 Figure 1: Subjective ranking of constraints: the share of firms that report issue as an above-average obstacle to the growth of their business. Firms in Africa see access to finance and electricity as relatively more constraining Access to finance Electricity Other region Africa Other region Africa Percent of firms Percent of firms micro 1-10 small medium large micro 1-10 small medium large 200+ while firms in other regions see interactions with the government and corruption as greater constraints. Licenses Corruption Other region Africa Other region Africa Percent of firms Percent of firms micro 1-10 small medium large micro 1-10 small medium large

16 Figure 2: Differences in objective conditions are often more challenging in Africa and for smaller firms. Share of investments financed with bank loans Days without power Other developing regions Africa Other developing regions Africa Percent 15 Percent micro 1-10 small medium large micro 1-10 small medium large 200+ Management time with officials Frequency of bribes to 'get things done' Other developing regions Africa Other developing regions Africa Percent 8 6 Percent micro 1-10 small medium large micro 1-10 small medium large

17 TABLE 1: PATTERNS OF GROWTH Including high income countries Excluding high income countries (1) (2) (3) (4) d_afr * (0.019) (0.027) (0.020) (0.028) (small)*d_afr (0.023) (0.024) (medium)*d_afr * (0.030) (0.030) (large)*d_afr * (0.030) (0.030) small *** *** *** *** (0.013) (0.015) (0.014) (0.016) medium *** *** *** *** (0.017) (0.019) (0.017) (0.019) large *** *** *** *** (0.021) (0.023) (0.021) (0.023) mature *** *** *** *** (0.009) (0.009) (0.010) (0.010) older *** *** *** *** (0.010) (0.010) (0.010) (0.010) non-capital, small city *** *** *** *** (0.011) (0.011) (0.011) (0.011) export 0.085*** 0.084*** 0.086*** 0.086*** (0.008) (0.008) (0.008) (0.009) foreign ownership 0.044*** 0.044*** 0.045*** 0.046*** (0.007) (0.007) (0.007) (0.007) government ownership ** ** * * (0.012) (0.012) (0.013) (0.012) constant (0.174) (0.174) (0.339) (0.340) SECTOR DUMMY YES YES YES YES COUNTRY CONTROLS YES YES YES YES Observations R-squared Robust standard errors in parentheses * significant at 10%; ** significant at 5%; *** significant at 1% Country controls: inflation, lagged GDP growth, GDP per capita, GDP per capita squared, and exports as share of GDP. 16

18 TABLE 2: IMPACT OF BUSINESS ENVIRONMENT ON EMPLOYMENT GROWTH IN AFRICA (1) (2) (3) (4) (5) (6) (7) All countries (excluding high income) AFR WORLD AFR WORLD d_afr== *** sh-invest-fin 0.004*** 0.007*** 0.004*** 0.008*** (0.027) (0.026) (0.035) (0.001) (0.001) (0.001) (0.001) sh-invest-fin 0.003*** 0.003*** 0.002*** (small)*sh-invest-fin ** ** (0.000) (0.000) (0.001) (0.001) (0.001) (0.001) (0.001) (d_afr==1)*sh-invest-fi (medium)*sh-invest-f * ** * *** (0.001) (0.001) (0.001) (0.002) (0.001) (0.002) (0.001) bribe * (large)*sh-invest-fin ** ** ** *** (0.002) (0.002) (0.032) (0.002) (0.001) (0.002) (0.001) (d_afr==1)*bribe bribe * ** (0.004) (0.003) (0.047) (0.004) (0.004) (0.004) (0.004) loss-power *** (small)*bribe ** * (0.002) (0.006) (0.004) (0.005) (0.004) (d_afr==1)*loss-power 0.005** (medium)*bribe * (0.002) (0.008) (0.006) (0.008) (0.006) lgdysnopower_csci *** (large)*bribe ** (0.004) (0.007) (0.015) (0.007) (0.016) (0.006) (d_afr==1)*lgdysnopower_csci ** loss-power *** ** (0.009) (0.013) (0.003) (0.003) generator_csci 0.005*** (small)*loss-power (0.000) (0.003) (0.003) (d_afr==1)*generator_csci *** (medium)*loss-powe *** (0.001) (0.004) (0.005) mng-time 0.007*** 0.007*** 0.008*** (large)*loss-power *** (0.001) (0.001) (0.001) (0.004) (0.008) (d_afr==1)*mng-time *** *** *** dys-no-power 0.048** (0.002) (0.002) (0.002) (0.023) (0.010) small *** *** *** (small)*dys-no-power (0.010) (0.009) (0.013) (0.016) (0.009) medium *** *** *** (medium)*dys-no-power (0.015) (0.013) (0.020) (0.022) (0.011) large *** *** *** (large)*dys-no-power (0.016) (0.014) (0.025) (0.024) (0.011) mature *** *** *** mng-time 0.010*** 0.013*** 0.009** 0.012*** (0.008) (0.007) (0.009) (0.004) (0.002) (0.004) (0.002) older *** *** *** (small)*mng-time ** *** * *** (0.009) (0.008) (0.010) (0.004) (0.002) (0.004) (0.002) Export 0.083*** 0.086*** 0.077*** (medium)*mng-time ** *** *** (0.007) (0.006) (0.008) (0.004) (0.002) (0.004) (0.002) foreign 0.036*** 0.042*** 0.021** (large)*mng-time *** *** (0.008) (0.007) (0.009) (0.004) (0.002) (0.005) (0.002) government 0.031*** ** small * *** *** (0.018) (0.009) (0.015) (0.036) (0.024) (0.041) (0.024) SECTOR CONTROLS YES YES YES medium ** *** *** COUNTRY CONTROLS YES YES YES (0.053) (0.034) (0.062) (0.032) Constant 0.628*** 0.317** 0.927*** large *** *** ** *** (0.169) (0.144) (0.257) (0.068) (0.038) (0.079) (0.033) Observations FIRM CONTROLS YES YES YES YES R-squared SURVEY DUMMIE YES YES YES YES Robust standard errors in parentheses Observations * significant at 10%; ** significant at 5%; *** significant at 1% R-squared

19 TABLE 3: IMPACT OF BUSINESS ENVIRONMENT ON EMPLOYMENT GROWTH - USING ALTERNATIVE MEASURES AFR WORLD AFR WORLD AFR WORLD AFR WORLD (1) (2) (3) (4) (5) (6) (7) (8) FINANCE CORRUPTION REGULATION REGULATION IC-variable Share sales on credit Freqency of bribes Days import customs Consistency of enforcement finance 0.005*** 0.006*** 0.004*** 0.009*** 0.004*** 0.008*** 0.004*** 0.008*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (small)*finance *** ** *** ** (0.001) (0.001) (0.002) (0.001) (0.002) (0.001) (0.002) (0.001) (medium)*finance *** *** ** ** ** *** (0.001) (0.001) (0.002) (0.001) (0.002) (0.001) (0.002) (0.001) (large)*finance ** ** ** *** *** *** *** *** (0.002) (0.001) (0.002) (0.001) (0.002) (0.001) (0.002) (0.001) corruption * ** (0.004) (0.004) (0.059) (0.058) (0.004) (0.004) (0.004) (0.004) (small)*corruption * * ** * (0.006) (0.005) (0.061) (0.050) (0.006) (0.004) (0.005) (0.004) (medium)*corruption * *** * (0.008) (0.007) (0.075) (0.060) (0.008) (0.006) (0.008) (0.006) (large)*corruption ** (0.017) (0.008) (0.093) (0.063) (0.016) (0.007) (0.016) (0.006) loss-power ** ** *** *** *** *** *** *** (0.003) (0.004) (0.003) (0.004) (0.003) (0.003) (0.003) (0.004) (small)*loss-power (0.003) (0.004) (0.003) (0.003) (0.003) (0.003) (0.002) (0.004) (medium)*loss-power ** ** *** * (0.004) (0.006) (0.004) (0.006) (0.004) (0.005) (0.004) (0.006) (large)*loss-power ** *** *** * (0.004) (0.010) (0.004) (0.009) (0.004) (0.007) (0.004) (0.009) regulations 0.008** 0.013*** 0.009** 0.012*** ** *** (0.004) (0.002) (0.003) (0.002) (0.022) (0.015) (0.024) (0.018) (small)*regulations ** *** ** *** *** (0.004) (0.002) (0.004) (0.002) (0.018) (0.014) (0.026) (0.018) (medium)*regulations * * ** ** (0.004) (0.003) (0.004) (0.002) (0.023) (0.019) (0.032) (0.022) (large)*regulations ** * ** (0.005) (0.003) (0.004) (0.002) (0.041) (0.021) (0.041) (0.020) small *** * *** ** *** ** *** (0.040) (0.029) (0.045) (0.031) (0.048) (0.029) (0.087) (0.063) medium *** ** ** *** *** ** (0.062) (0.046) (0.058) (0.040) (0.068) (0.042) (0.115) (0.078) large ** *** *** *** *** *** *** *** (0.099) (0.048) (0.071) (0.044) (0.094) (0.046) (0.154) (0.072) FIRM CONTROLS YES YES YES YES YES YES YES YES SURVEY DUMMIES YES YES YES YES YES YES YES YES Constant *** 0.260** 0.249*** 0.699*** 0.537*** 0.320** 0.457*** (0.117) (0.075) (0.116) (0.078) (0.112) (0.075) (0.144) (0.073) Observations R-squared Robust standard errors in parentheses * significant at 10%; ** significant at 5%; *** significant at 1% Default variables for each category: finance=sh-invest-fin; corruption=%bribe; regulations=%mngt-time with officials One alternative variable is substituted in each column, with the variable indicated at the top of the column. 18