Introduction
In many developing urban areas, informal micro and small enterprises (MSEs) are the primary source of income for residents, with their success significantly impacting livelihoods While some entrepreneurs thrive, many petty traders and low-wage workers struggle to make ends meet, indicating that the earnings potential of numerous entrepreneurs remains untapped due to economic constraints like entry barriers and limited access to credit These challenges highlight the need for policy interventions, such as micro-credit programs, to support informal entrepreneurs The combination of entry barriers and capital market imperfections contributes to the variability in success among these entrepreneurs Poverty trap models suggest that returns on capital in this sector are often minimal or non-existent, as increased competition among the poor diminishes potential profits Only those who are financially stable or can secure credit can access higher returns, making the returns to capital in MSEs a crucial indicator of the unrealized potential within informal entrepreneurship.
Despite extensive literature on the informal sector in developing countries, empirical studies on entry barriers and capital returns in micro and small enterprises (MSEs) are surprisingly limited Early insights indicate that the informal sector encompasses a diverse range of activities Research consistently shows high returns on capital, often exceeding 60 percent annually For example, a study of Sri Lankan microenterprises revealed a significant positive correlation between cash transfers and real profits, estimating returns between 55 to 70 percent per year Similarly, research on informal Mexican enterprises indicated high returns at low capital levels, with variable start-up costs by sector In Sub-Saharan Africa, evidence also points to exceptionally high capital returns, such as an average annual marginal rate of return of 113 percent found in rural Kenyan retail firms, highlighting the heterogeneity across different enterprises.
2 See, for example, Banerjee and Newman (1993), Aghion and Bolton (1997) or Lloyd-Ellis and Bernard (2000). whether firms take advantage of quantity discounts from wholesalers Both procedures yield very similar estimates
The reasons behind the notable high returns at low capital levels remain unclear, although some studies indicate that capital market constraints may play a significant role (Banerjee and Duflo, 2004; Schündeln, 2006; de Mel et al., 2008).
High returns in micro and small enterprises (MSEs) highlight their significant potential, as they contribute substantially to urban employment Research by Brilleau, Roubaud, and Torelli (2005) indicates that informal sector employment in urban West Africa consistently exceeds 70 percent, emphasizing the vital role of MSEs in the economy.
This paper estimates capital returns for micro and small enterprises (MSEs) in West Africa and investigates the entry barriers to small-scale economic activities Specifically, it addresses whether informal activities face high start-up costs relative to entrepreneurs' income and wealth, how capital returns fluctuate with the size of the capital stock, and the underlying causes of observed capital return patterns Utilizing a unique cross-sectional micro data set from informal enterprises across seven West African economic capitals, the study closely follows the methodology of McKenzie and Woodruff (2006) to provide insights into these critical questions.
The remainder of the paper is organised as follows Section 2 outlines our analytical framework and formulates the hypotheses that are tested in Section 3 Section 4 concludes.
Analytical framework and hypotheses
This article presents a model where potential entrepreneurs encounter entry barriers and non-convex production technologies, leading to testable assumptions regarding capital market imperfections In the context of developing countries, the literature highlights incomplete capital markets as a significant economic constraint, as noted by Tybout (1983) and Bigsten et al (2003) When credit contracts are difficult to enforce, capital does not flow to its most productive uses, resulting in unequal marginal returns across entrepreneurial activities Consequently, due to varying costs of capital linked to wealth differences and collateral capacity, borrowers may be compelled to invest in diverse technologies, as discussed by Banerjee and Duflo (2005).
The informal sector can be segmented based on varying entry barriers related to skill and capital requirements, as noted by Fields (1990) and Cunningham and Maloney (2001) This concept is incorporated into various economic development models and poverty trap theories, highlighting the significance of wealth distribution, as discussed by Banerjee and Newman (1993) and Galor and Zeira (1993).
In the context of Brilleau, Roubaud, and Torelli (2005), informal sector employment refers to jobs in firms without formal written accounts or tax registration, making such employment nearly universally classified as informal Economic activity segmentation and the coexistence of varying returns stem from the interplay of non-convex production technologies and capital market imperfections When entrepreneurial ventures require startup capital that is inaccessible through capital markets, individuals with limited resources are effectively barred from entering, resulting in a cycle where impoverished individuals remain trapped in low-productivity jobs Consequently, a higher proportion of initially poor individuals leads to an increased prevalence of low-productivity industries, perpetuating a poverty trap within the economy.
These models suggest that at minimal capital levels, returns are nearly negligible, only increasing significantly after surpassing a certain threshold In this simplified scenario, entrepreneurs aim to maximize profit, defined as the difference between output and capital costs, while adhering to borrowing constraints To generate any output using neoclassical technology, an entrepreneur must secure a minimum capital amount; failing to do so results in production costs consuming all resources, leading to zero profit.
Max y rK (1) s.t yf(K) if K K (2) rK y if K K (3)
The entrepreneur will chose his capital stock such that r K f( ) if BK (5)
If his borrowing constraint is binding, i.e BK, then the entrepreneur will be indifferent between different sizes of capital stock, as he earns zero profits anywhere between
0 Returns to an additional unit of capital, i.e ' ( K ), will hence be 0 between
Once an entrepreneur's borrowing capacity exceeds the threshold K, they experience significantly high marginal returns, which diminish to zero upon reaching the optimal capital level K* The relationship between marginal returns to capital and the borrowing constraint B is illustrated in the accompanying graph.
4 Risk and risk aversion can also create such poverty traps.
This study proposes two key hypotheses for investigation: firstly, the presence of a threshold K in the distribution of initial investments made by micro and small enterprises (MSEs) should be evident; secondly, capital returns are expected to be low at minimal investment levels, increasing but diminishing as capital approaches the threshold K This theoretical perspective challenges the majority of empirical findings discussed earlier The following analysis will determine whether this framework is applicable to the economies under consideration.
Entry costs and capital returns in African MSEs
Data
We test these hypotheses by using data from a set of surveys (1-2-3 surveys or Enquêtes 1-2-
3) in seven economic capitals of the West-African Economic and Monetary Union (WAEMU) in the early 2000s 5 A 1-2-3 survey is a multi-layer survey organised in three phases and specially designed to study the informal sector 6 Phase 1 is a representative labour force survey collecting detailed information on individual socio-demographic characteristics and employment Phase 2 is a survey which interviews a representative sub-sample of informal production units identified in Phase 1 The focus of the second phase is on the characteristics of the entrepreneurs and their production unit, including the characteristics of employed workers It also contains detailed information on input use, investment, sales and profits. Phase 3 is a household expenditure survey interviewing (again) a representative sub-sample of Phase 1 The data of all three phases is organised in a way so that it can be linked For this paper we use data from Phase 2 which hence is a sub-sample of informal entrepreneurs in seven West-African urban centres (Brilleau, Ouedraogo, and Roubaud, 2005).
Basic MSE characteristics
The 1-2-3 surveys define informal enterprises as production units that (a) do not have written formal accounts and/or (b) are not registered with the tax administration Part (b) of this definition varies slightly between countries, as registration may not always refer to registration with tax authorities The so-defined informal sector accounts for the vast majority of employment in the WAEMU cities covered by the surveys, as illustrated in Table
1 The share of informal sector employment exceeds 70 percent in all cities considered – in Cotonou and Lomé even 80 percent Employment in informal firms is typically self- employment, i.e the employed individual is also the MSE owner, but employed and/or
The urban centers of Abidjan, Bamako, Cotonou, Dakar, Niamey, Lomé, and Ouagadougou were the focus of surveys conducted by AFRISTAT and National Statistical Institutes (INS) with DIAL's support This research was part of the Regional Program of Statistical Support for Multilateral Surveillance (PARSTAT) between 2001 and 2003 For further details on the data, refer to Brilleau, Ouedraogo, and Roubaud (2005).
6 See Roubaud (2008) for a detailed assessment of this type of survey instrument. helping family- and non-family workers account for 30 to 40 percent of employment in this sector.
The 1-2-3 surveys do not (explicitly) apply a size criterion, but more than 90 percent of the enterprises employ a maximum of three people including the owner and possibly employed family members As shown in Table 1, around 70 percent of informal enterprises function in
In the 'pure self-employment' mode, enterprises are typically operated solely by the owner, resulting in an average size of just 1.6 individuals, including both family and non-family members This data, derived from a sample of 6,521 informal enterprises across seven countries, will serve as the foundation for future empirical analyses Notably, among these enterprises, 243 micro and small enterprises (MSEs) reported no profits, while 892 MSEs indicated zero capital stock.
Despite their small size, these enterprises have been operating for an average of over seven years, although the median age is notably lower at just five years The owner's experience often falls short of the business's age, largely due to the fact that some micro and small enterprises (MSEs) are passed down within families On average, MSE owners have completed only 3.7 years of schooling, and approximately half of them are women.
Informal enterprises report average monthly profits of approximately 380 International Dollars (Int $), with a median profit of 112 Int $ These profits are calculated by subtracting expenses for hired labor from the value added, which is determined by sales minus input costs for products intended for resale The comprehensive questionnaire includes detailed sections on the sales of both transformed and non-transformed products, as well as services offered Additionally, it addresses input costs, encompassing raw materials, intermediates, resale products, taxes, rents, and utility expenses, all for the previous month It's important to note that interest payments are not deducted from the value added.
The average capital stock among micro and small enterprises (MSEs) is approximately 1,000 Int $, largely influenced by a few MSEs with exceptionally high capital In contrast, the median capital endowment for MSEs is only 75 Int $ Capital stock is assessed based on the replacement value of all business-related assets, including establishments, machinery, furniture, vehicles, and utilities Entrepreneurs report the equipment used in the past year along with its replacement value, though it's unclear if the equipment is solely used for business or other household purposes Additionally, the calculation of capital stocks does not account for inventories or raw material stocks, which will be considered in future analyses of entry barriers and returns to capital.
7 Unfortunately, we do not have any information about sales of or damage to capital goods
An initial assessment of MSE heterogeneity reveals distinct characteristics across capital quintiles The first quintile, primarily consisting of self-employed individuals engaged in trading and services, operates with minimal capital and generates average profits of around 200 Int $, nearly double the median In the second quintile, entrepreneurs resemble those in the first but are less educated and earn approximately 30 Int $ less monthly The third quintile shows a significant profit increase of over 70 percent compared to the second, with average capital stock quadrupling, yet still remains low at about 80 Int $ Owners in this group have slightly more education than those in lower quintiles, though female ownership is less common, and firms are larger Transitioning from the third to the fourth quintile mirrors the previous shift, with capital stock again quadrupling, increased owner education, larger firm sizes, and a higher proportion of male owners, while monthly profits rise by an average of 70 Int $.
The disparities in capital and profits between the fourth and fifth quintiles are significant, with the average capital stock of micro and small enterprises (MSEs) in the fifth quintile reaching nearly 5,000 Int $ Additionally, these enterprises enjoy considerably higher monthly profits compared to others Entrepreneurs in this quintile are often more educated than the average, and over 50% of them employ at least one additional worker.
The descriptive statistics reveal significant heterogeneity within the informal sector, particularly among micro and small enterprises (MSEs) in the bottom 40 percent of capital distribution, which share common characteristics This variability, including profit differences, appears linked to capital stock and the specific sector of activity According to Table 3, the leading sector is 'petty trading' at 27.1 percent, followed by 'other manufacturing and food' at 16.1 percent, and 'other services' at 11.8 percent, while the transport sector has the smallest share at 4.6 percent due to high start-up costs The distribution of industries aligns with household demand patterns, as small services and food-related businesses dominate budgets, whereas transport and repair services occupy a lesser share Notably, the industry distribution is fairly consistent across the seven countries studied, with exceptions in Ouagadougou and Niamey, where 'other manufacturing and food' is more prevalent and 'hotels and restaurants' are less common Interestingly, the industry composition does not significantly correlate with GDP per capita, as wealthier cities like Abidjan and Dakar exhibit similar distributions to those of Niamey and Lomé.
Entry barriers
This analysis investigates the presence of entry barriers in informal activities, focusing on micro and small enterprises (MSEs) We anticipate that these businesses will participate in both low-capital subsistence activities and more capital-intensive ventures that necessitate significant initial investments.
Certain industries, such as trade, typically require a higher level of initial investment in equipment compared to others, like transport To understand these differences, we first analyze the distribution of initial equipment investments In addition, we examine other startup costs, including expenses related to additional inputs and inventory.
Before analyzing the entry barriers quantitatively, we present evidence from surveys of entrepreneurs regarding the challenges they encounter Table 4 highlights the percentage of entrepreneurs facing specific issues, focusing on micro and small enterprises (MSEs) in the clothing and apparel sector, categorized by age to assess variations in challenges at startup Two primary issues emerge: a lack of demand, characterized by insufficient clients and excessive competition, and difficulties in accessing capital, including credit, location, and necessary equipment While only 25% of MSEs report challenges in obtaining raw materials, this issue is more pronounced among younger firms Demand-related issues are significant across all age groups, with half of the firms indicating inadequate access to credit The perception of credit problems may be influenced by past experiences of credit applications, particularly among older firms, while younger firms frequently cite credit constraints related to locality and equipment as substantial entry barriers Other constraints, such as the lack of qualified personnel and governance issues, appear less critical, and there is no indication that these problems intensify when firms first commence operations.
The descriptive statistics presented are influenced by the presence of constrained firms that do not establish a shop, leading to potential bias in the analysis This limitation is acknowledged and remains unresolvable given the current dataset.
Our cross-sectional dataset enables the identification of investment paths for each enterprise asset, as we have access to purchase dates and establishment dates To estimate initial equipment investment, we utilize the accumulated investment made during the first year of operation Due to anticipated measurement errors in the investment history of micro and small enterprises (MSEs), our analysis focuses exclusively on enterprises established four years prior to the survey, resulting in a sub-sample of 3,144 informal enterprises.
The analysis begins with an overview of initial investments across various industries, as illustrated in Table 5 This table presents the replacement value of business assets accumulated during the first year of operation, categorized by specific quantiles of the initial investment distribution These statistics are derived from pooled data encompassing all seven countries included in the dataset.
Initial investment levels in equipment are generally low across various industries, with 31% of enterprises reporting no initial investment at all This trend is particularly evident in less capital-intensive informal sectors like repair services and hospitality For instance, the median petty trader invests less than 10 Int $ in their first year Although 29% of transport sector enterprises report zero initial investment, those that do invest allocate significantly more—approximately five times the amount seen in repair services and clothing sectors The top 25% of transport sector investors spend over 3400 Int $ within the first two years There is notable variability in initial investment across industries, as illustrated by the distribution of log initial investments, which suggests potential entry barriers, particularly in the service sector at around 50 Int $ The manufacturing sector exhibits several discontinuities, while the transport sector displays a distinct spike in investment above 1000 Int $ Overall, this analysis indicates that significant entry barriers related to equipment investment are not strongly supported.
9 It turns out that the distributions of start-up costs across industries in the different countries are fairly similar to those reported in Table 6.
Entrepreneurs face additional costs when starting a business, including recurring expenses for raw materials and inventory, which must be financed before sales occur While our dataset does not provide a precise calculation of start-up costs, we present monthly recurring expenses alongside initial investments and median profits as reference points These monthly expenses serve as an imperfect proxy for start-up costs, particularly in industries like wholesale and retail, where last month's inventory purchases may reflect these costs Conversely, petty trading can often begin with a fraction of reported monthly inventory purchases Additionally, inputs may only be purchased after receiving an order, suggesting these costs may be more variable than fixed Table 7 further illustrates the relative size of different components of start-up costs by reporting median monthly micro and small enterprise (MSE) profits.
While labor expenses may appear minimal, non-labor expenses can significantly exceed initial investments, particularly in manufacturing and construction where raw materials are crucial In trading activities, inventory costs dominate non-labor expenses Equipment investment plays a major role in start-up costs for manufacturing, except in food processing, and the disparity between mean and median non-labor expenses is less pronounced at lower capital levels Additionally, in trading, the costs associated with building inventories can be as critical as the initial investment in equipment.
A comparison of start-up costs and median monthly profits reveals that equipment and non-labor expenses can be significant in certain industries, while remaining minimal in others For example, the transport and clothing sectors require nearly three months' earnings to offset median initial investments When considering both equipment capital and ongoing monthly costs, the total start-up expenses amount to less than 30 Int $ combined.
Approximately 12 percent of Micro and Small Enterprises (MSEs) engage in informal activities, which typically feature low entry barriers Nonetheless, most informal activities do involve some fixed costs for entry It's important to consider that the costs mentioned should be viewed as an estimated upper limit.
Returns to capital
This study focuses on estimating returns to capital across varying levels of capital stock Due to the cross-sectional nature of the data, we can only analyze total capital stock rather than returns on initial or additional investments Our empirical model indicates that the profits (π ihj) of micro and small enterprises (MSE) in household h, located in country j, depend not only on capital (K ihj) but also on a set of exogenous variables (X ihj) and two unobserved factors: one at the household level ( hj), such as household wealth, and another at the individual level ( ihj), which we associate with entrepreneurial ability These unobserved factors influence profits directly and also impact the size of the capital stock.
( ihj hj ihj ihj hj ihj ihj f K X
In log-linearised form and with u ihj , a random error, the equation can be expressed as ihj ihj hj ihj ihj K ihj K X u
The study examines the observable exogenous characteristics of micro and small enterprise (MSE) owners, focusing on their years of schooling, experience, and gender It incorporates total labor input measured in hours, which encompasses both household and hired labor Additionally, the analysis includes industry and country dummies, along with interaction terms for industry and country, to account for variations in returns to capital, labor, and schooling across different contexts.
The cross-sectional estimation of equation (7) faces several potential biases that may affect the accuracy of the parameter K One significant issue is the omission of variables correlated with both capital stock and profit, such as the unobserved abilities of entrepreneurs, which can lead to an upward bias in K Additionally, reverse causality poses a challenge, as higher profits can facilitate faster capital accumulation, further contributing to this upward bias Lastly, classical measurement errors in both profits and capital stocks can result in a downward bias of K We will outline our approach to addressing these biases in the following sections.
To assess heterogeneity in returns based on capital stock, we divided the sample into three categories: low (below 150 Int $), medium (between 150 Int $ and 1000 Int $), and high capital stock (above 1000 Int $) This classification was informed by the distribution of initial investments and non-parametric analyses of capital profitability, which indicated that capital profitability is significantly high at low capital levels but declines rapidly as capital increases Notably, at around 150 Int $, the decrease in profitability becomes less pronounced The chosen thresholds ensure that the sub-samples remain adequately sized, resulting in approximately 50% of enterprises categorized as low-capital, 30% as medium-capital, and 20% as high-capital MSEs.
In this study, we employ a double-log specification to analyze various samples, regressing log profits on log capital and log labor using OLS This approach assumes a constant capital elasticity of profits, with marginal returns of capital influenced by capital profitability (π/K) Specifically, marginal returns are derived from the product of βK and (π/K), and since the estimated elasticity reflects an average effect, we calculate average marginal returns based on the mean of (π/K) Additionally, we assess returns at varying levels of capital stock, which demonstrate distinct capital profitability The first set of results, shown in Table 7, involves interacting log capital with country dummies, using Dakar (Senegal) as the reference category The second set of regressions in Table 8 incorporates interactions of capital with industry dummies, with manufacturing serving as the reference category.
The analysis in Table 7 reveals that the full sample estimations account for a significant portion of profit variation, with R-squared values consistently between 0.3 and 0.4 The profit elasticity of capital varies by country, ranging from 0.18 to 0.25, indicating marginal returns to capital (MRK) of 3 to 13 percent per month at country-specific mean capital-profitabilities Notably, the results highlight exceptionally high marginal returns at low capital levels, particularly in most countries where they exceed 70 percent monthly, driven by high profit-capital ratios However, only Niamey shows a statistically significant difference from the base coefficient for Dakar, suggesting that parameter equality across countries for low capital stocks is largely upheld.
In our analysis, we conducted a regression of monthly profits using a second-degree polynomial for both capital and labor, without applying logarithmic transformations The results obtained from this approach are comparable to those derived from the double-log specification, and further details can be requested from the authors.
The analysis excludes enterprises that report zero capital or profits, resulting in a total of 5,403 observations from an initial 6,584 Potential biases from this exclusion will be discussed later Additionally, influential outliers identified through the DFITS-statistic are removed from both the overall sample and sub-samples, following the cutoff-value recommendations by Belsley, Kuh, and Welsch (1980).
The procedure significantly reduces the sample size, with losses ranging from 5 to 10 percent, which may be attributed to measurement and reporting errors as well as the high heterogeneity present in informal micro and small enterprises (MSEs).
In this section, we will explore several key findings that could not be included in a table due to space constraints For those interested, all tables are available upon request from the authors.
13 All the results are robust to slight variations in the thresholds.
At higher capital levels, marginal returns to capital decrease significantly, even though the log capital coefficient increases with capital stocks Profit-capital ratios are notably lower at elevated capital levels In the medium capital range of 150 to 1000 Int $, there are substantial differences in log capital coefficients across countries For Abijan, Bamako, and Niamey, the interaction terms are strongly negative, indicating that the correlation between log profits and log capital is negligible in this capital range, with implied marginal returns to capital (MRKs) potentially being negative Conversely, other countries exhibit MRKs around 13 to 14 percent, except for Senegal, which shows a remarkable monthly return of 38 percent Above 1000 Int $, there is less variability, with only Cotonou and Niamey displaying significantly lower capital coefficients than Dakar, contributing to their reduced MRKs In contrast, Lomé's low returns are attributed to a diminished profit-capital ratio, while other countries maintain monthly capital returns between 7 and 14 percent.
Identifying patterns in the results is challenging, potentially due to variations in the level of development and the differing industry compositions of informal micro and small enterprises (MSEs) across countries To address this, we conducted regressions that focus on industry-capital interactions rather than country-capital interactions, as our dataset lacks the necessary size for a comprehensive analysis We have categorized industries into four main groups: (1) Manufacturing, (2) Construction, Hotels, Transport, (3) Trade, and (4) Repair and Other Services.
The findings indicate that the observed cross-country heterogeneity can be partially attributed to variations in industry compositions Notably, there is significantly less heterogeneity in the capital coefficient within industries compared to across countries At medium and high capital levels, none of the industry-capital interaction terms are significant, revealing marginal returns of 4 percent for the repair and other services sector, and 6 to 7 percent for all other sectors.
The latest analysis reveals increased variability in returns at lower capital levels, with marginal returns (MRKs) ranging from 47% for repair services to 268% for sectors like construction, hotels, and transport Despite these fluctuations, the key takeaway remains unchanged: there are consistently high marginal returns to capital when capital levels are low.
An analysis of variation within the trade sector, the only industry with adequate observations, reveals that marginal returns to capital at low levels of capital stock exceed 65%, with the exceptions of Lomé and Bamako, where they are 35% and 33%, respectively.
15 Within these aggregate sectors, capital coefficients were found to be homogeneous.
Returns to capital with a household fixed-effect
To address the limitations of using imperfect proxies for ability, incorporating household fixed effects into regression analyses offers a viable alternative Many households operate multiple enterprises, enabling the examination of profit and capital variations among firms within the same household This approach effectively eliminates omitted household-level variables from the estimation, potentially reducing ability bias It is reasonable to assume that entrepreneurs within the same household share greater similarities in ability compared to those outside the household, enhancing the reliability of the results.
Analyzing intra-household differences allows us to test the assumption that factor returns are equalized across various activities within the household A rational household should achieve this equilibrium; otherwise, there would be opportunities for Pareto-improving reallocations of factors If we observe discrepancies in marginal returns to capital, it indicates potential inefficiencies in capital allocation These inefficiencies may stem from non-cooperative behavior within the household or from non-linearities in capital stocks that hinder return equalization Furthermore, differing risks associated with activities may lead households to maintain portfolios with varying risk profiles, necessitating the equalization of risk-adjusted returns While risk is likely a primary reason for differences in returns within households, it is also possible that distinct activities, managed by different individuals, face varying constraints such as unequal access to capital Therefore, the fixed-effects results presented should be interpreted not only as a robustness check but also as an initial exploration of the underlying causes of the observed return patterns to capital.
Before analyzing the results, it's important to highlight that fixed-effect estimation may introduce selection bias due to its limitation to households with multiple enterprises The findings from the fixed effects estimates are detailed in Table 9, which includes data from 946 households that collectively own 2,079 enterprises.
In this analysis, the no-log specification indicates zero coefficients for capital, while the log specification enables the testing of marginal returns through capital profitabilities, utilizing at least two Mean Squared Errors (MSEs) MSEs reporting zero profits or zero capital are excluded, along with influential outliers, resulting in a significant reduction of the sample size The initial estimates are derived from 600 households operating 1,301 firms Alongside the previously mentioned double-log specification, a model without logarithmic transformation is also estimated, where the coefficients in the no-log specification can be directly interpreted as the marginal return to capital.
The fixed-effects estimates align closely with those obtained without fixed effects, indicating that capital returns are consistent, particularly at low capital levels where marginal returns reach approximately 90 percent These returns may be influenced by the high risks associated with activities in this capital range, as non-linearities in capital stocks are unlikely to account for intra-household differences In the medium range, capital shows no significant impact, though its magnitude remains similar to previous estimates, with monthly marginal returns around 9 percent At higher capital levels, intra-household variations may arise from activity-specific capital constraints and potential non-linearities, particularly in machinery investment The slightly elevated returns in the fixed-effect model could reflect a selection bias towards more talented households with at least two MSEs, a phenomenon likely more pronounced at higher capital levels Additionally, the capacity of diversified households to assume greater risks may contribute to achieving these higher returns.
The fixed-effects estimates indicate significant returns at low capital levels, although the reduced sample size warrants caution in interpreting these findings Nevertheless, these results suggest that risk is a key factor in understanding the high returns associated with lower capital investments.
The sub-samples categorized by capital size consist solely of households where all enterprises possess a capital stock that aligns with the specified criteria, such as a capital stock of less than 150 Int $.
21 This also holds when we estimate the earlier specification without fixed effects on the much smaller samples.
When interpreting fixed effects estimates, it is crucial to recognize that the two primary variables, profits and capital stock, may be subject to measurement error This issue is exacerbated when relying solely on within-household variation Consequently, such measurement inaccuracies could lead to a bias in the returns to capital, skewing the results toward zero, which contrasts with the potential for ability bias.
Capital stocks and profits among various micro and small enterprises (MSEs) within the same household exhibit sufficient variability to facilitate the estimation of a fixed-effects model However, characteristics such as education and experience show limited variation within the household Furthermore, it is noteworthy that some MSEs are managed by the same individual.
Some more thoughts on the causes
This paper does not delve deeply into the causes of capital return patterns, leaving that for future research However, it provides preliminary evidence on potential channels explaining these patterns, particularly focusing on the risks associated with varying levels of capital If the observed high returns are indeed linked to capital levels, then risks should be greater at lower capital levels, as indicated by fixed-effects estimates Additionally, we attempt to proxy capital constraints and analyze these proxies across different capital levels, anticipating that micro and small enterprises (MSEs) with lower capital will face more constraints than those with higher capital.
Our survey presents several opportunities to create risk proxies, despite the challenges of doing so with cross-sectional datasets We begin by developing traditional proxies for risk, specifically focusing on the variation in profits or sales This variation is measured at the country-sector level, ensuring that industries are detailed while maintaining a minimum of 30 observations in each country-sector cell This approach provides valuable insights into risk assessment.
123 country-sector cells, for which we compute the coefficients of variation in profits and sales Second, we use business risk perceptions of the entrepreneur Specifically, we set a
Approximately 60 percent of micro and small enterprises (MSEs) perceive the lack of clients or excessive competition as significant business risks, leading to a 'risk-of-closure dummy' value of 1 Table 10 presents the sample means of these risk proxies across various capital stock levels, highlighting the challenges faced by entrepreneurs in this sector.
The descriptive statistics indicate that risk may partially account for the observed return patterns, as evidenced by the coefficient of variation in profits and sales being lowest for higher capital levels Notably, the coefficient of variation for profits, a more reliable risk indicator, is higher at low capital levels compared to other groups This suggests that high capital MSEs, which yield lower returns, are more common in sectors with stable profit variations However, the differences in these indicators are not statistically significant, with standard errors ranging from 0.5 to 0.9 Furthermore, the analysis reveals that the greatest threats to business survival occur at medium capital levels, contradicting the notion that marginal returns correlate with higher risks.
A detailed analysis of risk indicators, such as by country or capital profitability, yields inconsistent results, suggesting that the evidence only weakly supports risk as a primary factor influencing capital returns This lack of definitive findings may stem from inadequate proxies for risk and risk aversion, as well as a simplistic empirical approach Additionally, the relationship between risk and returns may be influenced by capital market constraints, which our analysis does not account for.
23 The corresponding question in the survey reads ‘which are major threats to the existence of the MSE’.
In our analysis, we explore the potential impact of capital constraints on the observed patterns, as detailed in Table 11, which presents three proxies for capital constraints categorized by capital stock range Notably, for lower levels of capital, where we observe significantly high marginal returns, it is anticipated that micro and small enterprises (MSEs) face substantial capital limitations.
Table 11 indicates that micro and small enterprises (MSEs) with low capital stock face greater capital constraints, with 88% relying solely on personal savings for financing, in contrast to 81% and 77% of those with medium and high capital stock, respectively Additionally, 14% of low capital MSEs report liquidity constraints, compared to only 10% in higher capital groups Analyzing MSEs by household wealth reveals that 32% of high capital MSEs are situated in the wealthiest households, although there are instances of affluent households operating low capital MSEs and impoverished households managing high capital MSEs.
While the findings align with expectations, they do not sufficiently demonstrate the significance of capital constraints, as the descriptive statistics appear unclear The notable presence of low-capital micro and small enterprises (MSEs) among affluent households suggests that additional factors may influence capital accumulation Many households tend to engage in extensive growth by investing in multiple small firms rather than establishing a single large one, indicating a preference for risk aversion rather than capital constraints Furthermore, as highlighted by McKenzie and Woodruff (2006), MSEs theoretically have the capacity to reinvest their substantial returns for capital accumulation, implying that capital constraints may only partially account for high returns.
Conclusions
Our analysis of capital entry barriers and returns on investment in informal activities across seven urban centers in West Africa reveals significant challenges for micro and small enterprises (MSEs), particularly when considering operating costs While most informal sectors demonstrate considerable entry barriers, we identified a sub-sector with minimal fixed costs Notably, a small percentage of informal entrepreneurs make substantial initial investments, especially in the transport industry These insights, combined with our descriptive analysis of MSE characteristics, highlight the considerable diversity within informal activities.
24 For each item of capital stock, the entrepreneur is asked for the source of funding From this information, we construct the dummy for ’No access to external capital’.
25 The ‘liquidity constraints’ dummy is set to 1 if entrepreneurs perceive the lack of liquidity as a major threat to survival of their enterprise.
The analysis reveals heterogeneous returns to capital, with exceptionally high marginal returns of at least 70 percent at low capital stock levels However, these returns decline rapidly as investment increases For capital stocks exceeding 150 Int $, monthly marginal returns drop to four to seven percent using a simple OLS method, and approximately nine percent with a household fixed-effects estimator Consequently, the annualized returns at higher capital levels range from 50 to 70 percent, significantly surpassing the interest rates typically charged by micro-credit providers (15 to 25 percent) and falling within the rates of informal money lenders (60 percent and above).
Our research aligns with previous studies on small-scale activities, challenging the notion that the informal sector is solely comprised of low-capital ventures with minimal returns Instead, many micro and small enterprises (MSEs) with low capital stocks demonstrate the potential for high returns, suggesting they can be instrumental in alleviating poverty Although our static analysis does not address the dynamic aspects of this issue, we offer insights into the factors that impede these entrepreneurs from achieving their growth potential.
This article examines capital constraints and risk as factors contributing to high returns despite low capital levels in micro and small enterprises (MSEs) Although MSEs with limited capital are often significantly constrained, their access to capital does not sufficiently account for the substantial variations in returns across different capital stock levels Our innovative approach focuses on the role of risk, interpreting the observed high marginal returns at lower capital levels through a household fixed-effects profit function estimation, which highlights the differing risks associated with informal activities conducted by households.
This study suggests that risk plays a significant role in explaining the high returns on capital in small-scale economic activities However, the findings should be interpreted cautiously, as they are not fully supported by other risk indicators This research serves as an initial step toward understanding the challenges and opportunities faced by informal entrepreneurs in Sub-Saharan Africa A deeper investigation into the reasons behind the variability in returns is essential, especially since informal activities are expected to continue being the primary income source for the region's poor in the coming decades.
Aghion, P and P Bolton, “A theory of trickle-down growth and development,” Review of
Banerjee, A and E Duflo, “Do Firms Want to Borrow More? Testing Credit Constraints Using a Directed Lending Program,” CEPR Discussion Papers 4681, 2004.
Banerjee, A and E Duflo, “Growth Theory through the Lens of Development Economics,” in: Aghion, P and S Durlauf, ed., Handbook of Economic Growth, 473-552, Elsevier, 2005.
Banerjee, A and A Newman, “Occupational choice and the process of Development,”
Belsley, D A., Kuh, E and R.E Welsch, Regression Diagnostics, Wiley, New York, 1980
Bigsten, A., Collier, P., Dercon, S., Fafchamps, M., Gauthier, B., Gunning, J., Oduro, A., Oostendorp, R., Patillo, C., Soderbom, M., Teal, F., and A Zeufack, “Credit Constraints in Manufacturing Enterprises in Africa,” Journal of African Economies, 12, 104-125, 2003.
Brilleau A., Roubaud, F., and C Torelli, “L’emploi, le chômage et les conditions d’activité, Enquêtes 1-2-3, phase ," Stateco, 99, 54-84, 2005.
Brilleau A., Ouedraogo, E., and F Roubaud, “L’Enquête 1-2-3 dans les Principales Agglomérations de l’UEMOA: la Consolidation d’une Méthode,“ Stateco, 99, 15-19,
Cunningham, W., and W.F Maloney, “Heterogeneity in the Mexican Micro-enterprise Sector:
An Application of Factor and Cluster Analysis,“ Economic Development and Cultural Change, 5, 131-156, 2001.
De Mel, S., McKenzie, D., and C Woodruff, “Returns to Capital in Microenterprises: Evidence from a Field Experiment, “ The Quarterly Journal of Economics, 123, 1329-1372, 2008.
Fields, G “Labour Market Modelling and the Urban Informal Sector: Theory and Evidence,” in: Turnham, D., Salomé, B., and A Schwartz, ed., The Informal Sector Revisited, OECD Development Centre Seminar Series, OECD, Paris, 49-69, 1990.
Galor, O and J Zeira, “Income Distribution and Macroeconomics,” Review of Economic
Hart, K., “Informal Income Opportunities and Urban Employment in Ghana,” The Journal of
Henley, A., Arabsheibani, R., and F Carneiro, “On Defining and Measurement the Informal Sector,” World Bank Policy Research Working Paper 3866, World Bank, Washington D.C., 2006.
Kremer, M., Lee, J.N., and J.M Robinson, J.M, “The Return to Capital for Small Retailers in Kenya: Evidence from Inventories,” Mimeo, Harvard University, 2010.
Lloyd-Ellis, H and D Bernhardt, “Enterprise, Inequality and Economic Development,” Review of Economic Studies, 67, 147-168, 2000.
Malony, W., “Informality Revisited,” World Development, 32, 1159-1176, 2004.
McKenzie, D.J and C Woodruff, “Do Entry Costs Provide an Empirical Basis for Poverty Traps? Evidence from Mexican Microenterprises,” Economic Development and Cultural
Moser, C., “Informal Sector or Petty Commodity Production: Dualism or Dependence in Urban Development?,” World Development, 6, 1041-1064, 1978.
Peatti, L., “An Idea in Good Currency and How It Grew: The Informal Sector,” World
Rakowsky, C., “Convergence and Divergence in the Informal Sector Debate: A Focus on Latin America 1984-92,” World Development, 22, 501-516, 1994.
Roubaud F., “Mixed Surveys Using the Modular Approach,” in: ILO Manual on Surveys of
Informal Employment and Informal Sector, ILO, Geneva, 2008.
Schündeln, M “Returns to Capital and Financial Constraints,” Mimeo, Harvard University, 2004.
Schündeln, M., “Modeling Firm Dynamics to Identify the Cost of Financing Constraints in Ghanaian Manufacturing,” Mimeo, Harvard University, 2006.
Tybout, J., “Credit Rationing and Investment Behavior in a Developing Country,” Review of
Udry, C and S Anagol, “The Return to Capital in Ghana,” American Economic Review, 96,388-393, 2006.
Figure 1: Borrowing constraints and marginal returns to capital
Figure 2: Histograms of initial investment (values in current Int $) r f’(K)
Source: Authors’ computation based on 1-2-3 surveys (Phase 2, 2001/02, AFRISTAT, DIAL, INS).Note: The histograms exclude zero investment.
Table 1: Employment by sector in seven West-African urban centres (in percent)
Principal employment Cotonou Ouaga Abidjan Bamako Niamey Dakar Lomé Total
Private informal firm 80.3 73.4 74.7 77.5 71.1 76.4 81.0 76.2 of which
Source: Brilleau, Roubaud and Torelli (2005), and authors’ computations based on 1-2-3 surveys (Phase 2, 2001/02, AFRISTAT, DIAL, INS).
Table 2: Basic descriptive statistics of informal MSEs, by quintiles of capital stock (values in 2001 international Dollar)
Share of pure self- employment 0.69 0.9 0.9 0.7 0.6 0.4
Monthly profit (in 2001 international Dollar) 380.3 112 206.7 179.9 323 412 783.3
Capital stock (in 2001 international Dollar) 997.2 76.8 2.1 23.4 83.6 351.8 4554.4
Notes: Quintiles of capital (min and max capital in Int $ in parentheses).
Source: Authors’ computation based on 1-2-3 surveys (Phase 2, 2001/02, AFRISTAT, DIAL, INS).
Notes: 2001 international dollars are on the basis of the Purchasing Power Parity converters for GDP from the World Development Indicators (World Bank, 2010).
Table 3: Industry composition of informal MSEs by country (number of observations and shares in percent)
Industry/City Cotonou Ouaga Abidjan Bamako Niamey Dakar Lomé Total
Source: Authors’ computation based on 1-2-3 survey (Phase 2, 2001/02, AFRISTAT, DIAL, INS).
Note: Shares (in percent) in italics.
Table 4: Perceived problems faced by MSEs in the clothing and apparel sector by enterprise age
Problem less than 1 year 2-3 years 4-8 years more than 8
Too many regulations and taxes 0.10 0.07 0.09 0.11 0.11
Source: Authors’ computation based on 1-2-3 survey (Phase 2, 2001/02, AFRISTAT, DIAL, INS).
Table 5: Entry barriers to informal enterprises (values in current Int $)
Share 0 init inv Mean Mean (>0) p10 p25 p50 p75 p99
Source: Authors’ computation based on 1-2-3 survey (Phase 2, 2001/02, AFRISTAT, DIAL, INS).
Table 6: Initial investment and other start-up costs relative to income levels (values in current Int $)
Non-labor expenses Raw material Inventory
Mean p50 Mean p50 Mean p50 Mean p50 Mean p50 p50
Source: Authors’ computation based on 1-2-3 surveys (Phase 2, 2001/02, AFRISTAT, DIAL, INS).
Note: Non-labour expenses include raw materials, inventories, and all other recurrent expenses (for example fuel).
Table 7: Returns to capital – results from OLS including capital-country interactions
Additional controls Log labour and Log labour-country interactions, Owner's education and owner's education-country interactions, Owner's experience, owner female, industry dummies, country dummies, country-industry interactions
Implied MRK (at average P and
Source: Authors’ computation based on 1-2-3 surveys (Phase 2, 2001/02, AFRISTAT, DIAL, INS).
p