Policy Research Working Paper 11235 What Do Small and Informal Household Enterprises Want? Nicolas Cerkez Wendy Cunningham Sarika Gupta Felix Lung Social Protection and Labor A verified reproducibility package for this paper is Global Department available at http://reproducibility.worldbank.org, October 2025 click here for direct access. Policy Research Working Paper 11235 Abstract A large share of workers in Sub-Saharan Africa earn a could as wage earners, with wage earners confirming the living through informal, low productivity household observations. However, a combination of family and busi- enterprises. While structural transformation toward ness constraints and shocks may hinder their ambitions, formal wage employment is viewed as the long-term path ability to act on their goals, and realization of those goals. to improving livelihoods, progress has been slow. In the That said, two-thirds of micro-enterprise owners said they meantime, small enterprises will remain a key source of would accept a wage job if it offered wages on par with employment for many years to come, making it important their current earnings. This suggests that households will to better understand how to help such enterprises thrive. continue to prefer firm ownership in the short run until This paper uses original survey data from 1,526 poor indi- structural transformation can improve earning potential viduals across Liberia, Niger, and Senegal to examine the of wage employment in the long term. The results sug- aspirations and constraints of urban household enterprise gest that household enterprise owners require a dual policy owners. The results suggest that most surveyed business approach: one that improves current enterprise conditions owners voluntarily started their businesses, are satisfied while advancing longer-term structural reforms to expand with their jobs, and aspire to and have plans to expand access to quality wage employment. their businesses. Most report that they earn more than they This paper is a product of the Social Protection and Labor Global Department. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at wcunningham@worldbank.org, sringwala@worldbank.org, flung@worldbank.org and nicolas.cerkez@qeh.ox.ac.uk.A verified reproducibility package for this paper is available at http://reproducibility.worldbank.org, click here for direct access. RESEA CY LI R CH PO TRANSPARENT ANALYSIS S W R R E O KI P NG PA The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team What Do Small and Informal Household Enterprises Want?∗ Nicolas Cerkez† Wendy Cunningham Sarika Gupta Felix Lung Authorized for distribution by Camilla Holmemo, Practice Manager, Social Protection and Labor Global Department, World Bank Group Keywords: informal urban labor markets, aspirations, job satisfaction JEL Codes: J46, L26, M20, M50, J28 ∗ Cerkez: nicolas.cerkez@qeh.ox.ac.uk, Oxford University; Cunningham: wcunningham@worldbank.org, World Bank; Gupta: sringwala@worldbank.org, World Bank; Lung: flung@worldbank.org, World Bank. We are grateful to the data collection team, and in particular to Filippo Cuccaro and Ababacar Gueye, for their help in assembling this dataset and thank Francisco Campos, Sara Johanssen, Raphaela Karlen, William Maloney, Federica Ricaldi, and Gabriel Ulyssea for helpful comments. Funding from the World Bank’s Rapid Social Response Adaptive and Dynamic Social Protection (RSR-ADSP) Trust Fund is gratefully acknowledged. † Corresponding author. 1 Introduction A significant share of people in the developing world work in small, informal enterprises with extremely low productivity and thus low earnings (ILO, 2018). Structural transformation toward larger, formal firms, and the wage jobs therein, is widely seen as a way to improve job quality and livelihoods (Bandiera et al., 2022; Gollin and Kaboski, 2023). However, this process takes time. For instance, the ILO estimates that the share of wage employment in modern firms in Sub-Saharan Africa increased by only three percentage points over 25 years (1995-2020).1 Due to the slow pace of structural transformation in many economies, informal micro-enterprises will likely be a primary source of jobs for a long time. Public policies can support the largely informal micro-enterprise sector to be a source of better evalo-S´ jobs and poverty reduction (Banerjee et al., 2015; Hirvonen et al., 2024; Ar´ anchez et al., 2024), but the success of such policies will depend first and foremost on the willingness and capacity of potential and existing micro-entrepreneurs to take part in this kind of work.2 There is thin empirical evidence on the preference for micro-entrepreneurship as a viable source of good jobs, especially in slow growing, low-income economies.3 This is what this paper seeks to explore. By engaging directly with owners of micro and often informal household enterprises in three West African countries, we seek to understand what these micro-entrepreneurs aspire to and what prevents them from achieving it, given current market conditions.4 By incorporating individuals’ perspectives, we aim to inform policies that align with the im- mediate challenges faced by household enterprise owners. We answer five questions: (i) What are individuals’ motivations for becoming a microentrepreneur? (ii) How satisfied are they with their jobs? (iii) What business aspirations do they have, what steps have they taken towards achieving them, and what are their key constraints? (iv) How do their preferences relate to their business 1 World Development Indicators. Modeled ILO Estimates “wage and salaried workers, % of total employment.” 2 The ongoing debate about whether micro-entrepreneurship is a place to wait for a wage job or a utility- maximizing decision motivated us to better understand people’s willingness be micro-entrepreneurs, as well as how willingness may shape other constraints and business outcomes. 3 Fox and Sohnesen (2012, 2013) find that micro-entrepreneurs in the Republic of Congo, Tanzania, and Mozam- bique cite both pull (good job) and push (bad job) factors as motivations for owning a business. Choto et al. (2014) report a similar mix of pull and push factors among South Africa’s self-employed population. Qualitative data from entrepreneurs in Rwanda, South Africa, and Senegal uniformly reported satisfaction with being self-employed (Knox et al., 2019). A similar conclusion can be drawn from Blattman and Dercon (2018), who observed Ugandan employees leaving their jobs to become self-employed. Nigerian youth are as likely to aspire to self-employment as to wage jobs (Babangida, 2023). 4 We use the term “household enterprise” rather than “informal enterprise” in this paper to allow for the fact that some enterprises that may be informally organized—namely those that have a weak division between home and enterprise finances, location, labor, input, and outputs—may be registered with authorities. 2 outcomes? And, as a proxy for market conditions created by structural transformation, we ask (v) Would job aspirations be different if formal firm employment were a more lucrative option? West Africa is a good place to explore these research questions. It is home to some of the poorest countries in the world (Sinha et al., 2024), and structural transformation has been notably slow (World Bank Group, 2018, 2020, 2025). The region has the highest rate of informality globally, with most informal work taking the form of household enterprises (ILO, 2018; Danquah et al., 2021; Cunningham et al., 2024). Many micro-entrepreneurs in this sector face severe disadvantages, including lower profits, fewer assets, fewer growth prospects, and higher risk aversion as compared to modern or formal entrepreneurs (Co et al., 2006; Fox and Sohnesen, 2012; Grimm et al., 2012; Berner et al., 2012; Margolis, 2014; Adoho and Doumbia, 2022). Despite the widespread nature of this work, there is limited evidence on the job preferences of poor household enterprise owners. For instance, Falco et al. (2015) use panel data from Ghana to show that job satisfaction increases when individuals transition from formal wage work to self-employment but decreases when the transition goes the other way.5 While there is more robust evidence on the constraints entrepreneurs face (Bossuroy et al., 2017; Fox and Sohnesen, 2012; Kweka and Fox, 2011; Turkson and Codjoe, 2020), much less is known about how these constraints relate to individual preferences and aspirations. Our survey data come from a sample of 1,526 poor individuals—both enterprise owners and non- owners—across urban Liberia, Niger, and Senegal. Respondents were interviewed via phone surveys. For our main analysis, we pool the samples across countries and present pooled estimates and estimates disaggregated by respondent gender. Country-specific results are provided in appendices. We present five main findings. First, we show that 53 percent of household enterprise owners report voluntarily choosing to start their business, while 26 percent say they defaulted into en- trepreneurship.6 The remaining respondents fall into both categories. Economic considerations are cited as the most prominent reason for starting a business: many see entrepreneurship as a more profitable option than wage labor and a means of income diversification. Independence also matters: respondents frequently cite the satisfaction of being their own boss. Second, we find that household enterprise owners are generally more satisfied with their jobs than wage earners. 80 percent of household enterprise owners report being either satisfied or very 5 Studies on preferences in Senegal (Knox et al., 2019), Nigeria (Diao et al., 2020), and Burkina Faso (Nagler, 2017) reveal that household enterprise ownership is seen as highly desirable in Senegal but is viewed as no more desirable than wage work in Nigeria and Burkina Faso. 6 We classify as “voluntary” those individuals who, for example, started their business to diversify income or because they viewed it as offering higher earnings than other forms of work. In contrast, the “default” category includes those who, for example, report starting a business due to a lack of alternative job opportunities. 3 satisfied with their job, compared to 65 percent of wage earners. Consistent with this, nearly 70 percent of household enterprise owners consider their current job to be their ideal job, while another 18 percent say they would prefer to run a different kind of business. In contrast, only 30 percent of wage earners view their current job as ideal—and a substantial share (40 percent) identify running a business as their ideal occupation. Finally, 80 percent of all respondents—owners and non-owners—state that they would like their children to become household enterprise owners. Third, we find that household enterprise owners are aspirational yet constrained. Over 80 percent of household enterprise owners aspire to grow their business by increasing profits, boosting sales, or reaching new customers. Only about 5 percent express a desire to formalize their enterprise. Respondents cite a wide variety of business and personal constraints, with the most frequent ones being lack of capital, access to finance, and conflicting family responsibilities. Many also report being exposed to shocks such as rising household expenses, price shocks, and natural disasters. Despite their ambitions, most owners take only modest steps toward achieving their stated aspirations by employing low-cost, short-term strategies.7 More than half of respondents report having taken no action to pursue their aspirations. The data suggest that constraints and shocks may indirectly affect entrepreneurial success by shaping aspirations or the ability to act on them. A higher degree of family constraints and business shock exposure are associated with reduced business aspirations. For women, greater exposure to family shocks is negatively associated with having acted to realize business aspirations. Fourth, we ask if aspirations shape business outcomes and find that various measures of en- trepreneurial aspirations generally do not correlate with firm size, measured by number of workers, or firm formality. Owners who voluntarily started their business report higher job satisfaction, consider their current job to be ideal, have higher aspirations, and tend to employ more non-family workers. However, after controlling for owner and business characteristics, these associations largely disappear. Voluntarily starting a business continues to significantly correlate with the number of non-family employees. Likewise, owners who voluntarily start their business, report higher job satisfaction, and express higher aspirations are more likely to have registered their businesses— though these associations, like those with firm size, fade once owner and business characteristics are accounted for. Fifth, we find that economic considerations strongly shape the preference for household en- 7 The short-term, low-cost strategies include lowering sales prices and increasing marketing activities. They avoid costly or longer-term commitments that may be difficult to reverse, such as hiring staff or using new technology. 4 trepreneurship. Eighty percent of household enterprise owners and wage employees reported that owning a business was more remunerative than working for someone else. However, when asked whether individuals would be willing to take up wage employment if it offered the same income as their current self-employment—a reality not available at this time, but potentially in the future— around two-thirds of household enterprise owners said they would. We interpret this as an expression of short-term constraints in the face of long-term structural transformation. In other words, once structural transformation boosts the availability of equally paid wage jobs, a significant share of people would be willing to move to those jobs. That said, one-third of respondents are not inter- ested in a wage job that pays the same as they are currently earning, suggesting that they value non-pecuniary aspects of self-employment. This is especially observed among micro-entrepreneurs who are satisfied in their current business. Our results suggest that, given the scarcity of reliable and well-paid wage jobs, micro-entrepreneurs would benefit from public support to reach their business aspirations. Our descriptive findings— particularly regarding the role of constraints and shocks in shaping aspirations and individuals’ ability to act on them—align with a prominent intervention design, namely multi-faceted interven- tions comprising both social assistance and micro entrepreneurship training elements that address multiple constraints and that are scalable in the face of shocks. The respondents’ business aspira- tions are also shaped by risk, suggesting the need to experiment further with mechanisms to reduce the impact of, or compensate for, business, family, and natural shocks (Guven et al., 2021). Our evidence also suggests that micro-entrepreneurs would like policymakers to simultaneously pursue structural transformation to grow higher paying formal sector jobs. The paper contributes to several strands of the literature. First, it expands on a small body of empirical research that uses qualitative data, limited survey questions, and indirect methods to explore the job aspirations of poor, informal entrepreneurs in Sub-Saharan Africa (Kweka and Fox, 2011; Fox and Sohnesen, 2012; Choto et al., 2014; Bossuroy et al., 2017; Knox et al., 2019; Danquah et al., 2021). Our study advances this literature by providing the first systematic, rigorous, large- scale and quantitative evidence on the preferences of household enterprises in the region.8 It builds on the findings in Falco et al. (2015) by exploring aspirations beyond job satisfaction. Second, more broadly, this paper adds to the meager literature from Sub-Saharan Africa on whether owners of largely informal household enterprises are queueing for wage jobs (Harris and Todaro, 1970) or are optimizing income relative to constraints (Maloney, 2004; Ulyssea, 2020). Our 8 See Dalton et al. (2018) for work on Indonesia. 5 findings are aligned more closely with the latter view. However, our results also offer the caveat that household enterprise owners are not averse to wage jobs once those opportunities garner improved earnings potential. The rest of the paper is organized into three sections. Section 2 describes the data and empirical strategy. Section 3 presents the results. Section 4 discusses and concludes. 2 Data and Empirical Strategy Our focus is on urban residents in West Africa, the poorest region in the continent. We draw our sample from Liberia, Niger and Senegal, which collectively characterize heterogeneity in the region. These countries represent low-and lower-middle income economies; relatively large (Niger) and small (Liberia) territorial sizes; coastal and landlocked borders; South, East and West location in the region; and different agro-ecological zones. Roughly 36 (51) [9] percent of the population in Liberia (Niger) [Senegal] live on less than US$2.15 per day. The urbanization rate stands at 53 (17) [49] percent and an estimated 55 (42) [40] percent of the respective populations own an enterprise.9 2.1 Sample and Survey We draw a random sample from social databases in each country that were designed to identify poor households to receive social programs. In Liberia, respondents were sampled from the Liberia Household Social Registry which includes data from 11 poor urban communities in Monrovia. For Senegal, we use the National Single Registry (Registre National Unique) which contains data on poor households across the country. We randomly select respondents in urban Dakar, Diousbel, Saint-Louis, and Ziguinchor. For Niger, we rely on the Covid response survey by the National Mechanism for Prevention and Management of Food Crises which contains data on households from poor neighborhoods in seven of the eight major urban areas. We randomly select respondents in Niamey, Zinder, and Maradi. The sampling was designed to ensure representativeness by area, reflect the gender distribution of the underlying database, and over-represent entrepreneurs. Sampling was stratified by the com- munity (in Liberia), the arrondissement (in Niger), and the department (in Senegal), as well as the gender of the household head, selecting a final sample of 501 households in Liberia, 699 households 9 The data in the paragraph are drawn from the World Development Indicators and Cunningham et al. (2024). 6 in Niger, and 708 households in Senegal. The randomization was set-up such that 80 percent of the respondents in each country were household enterprise owners, and 20 percent did not own a household enterprise. Table A1 provides an overview of the sample by country.10 To ensure mutual clarity about terms, the surveyor shared the following on beginning the inter- view: “A household enterprise is a business run by individuals who are self-employed. An individual is self-employed when he/she has no obligations (contractually or informally) with an employer.” All survey respondents were between 18 and 65 years old at the time of the survey. Consultants and a data collection firm carried out the survey through phone interviews between November 2021 and January 2022 for Liberia, and between January 2022 and March 2022 in Senegal and Niger. 2.2 Sample Household and Business Characteristics Sample household characteristics are summarized in Table 1. For household enterprise owners, the pooled sample across all three countries is 48 percent male, ranging from 31 percent in Senegal to 64 percent in Niger.11 The average age of respondents is about 44 years (ranging from 39 years in Liberia to 50 years in Senegal). One-third of the sample’s business owners completed primary school (14 percent in Senegal to 71 percent in Liberia). Households have an average of 9 members, ranging from 6 members in Liberia to 11 in Senegal. An average of 3 children per household are younger than 14 years old. On average, 94 percent of respondents own a mobile phone or a tablet with little variation across countries.12 Around 6 percent of respondents own a laptop or computer, ranging from 4 percent in Niger and Senegal to 10 percent in Liberia. Columns 5 and 6 display the same averages for men and women, showing few differences except that men tend to be more educated and are more likely to own a laptop or computer. For non-household enterprise owners, sample household characteristics are largely the same. Fifty-seven percent of non-household enterprise owners are employees. In what follows, we focus only on respondents who are employees when discussing non-household enterprise owners. Business characteristics differ across countries (Table 2). The average age of our sample’s house- hold enterprises is ten years, ranging from six years in Liberia to more than 12 years in Senegal. 10 Interviews were conducted with the household head and their spouse in 66 and 25 percent of cases, respectively. Interviews about household enterprise owners were always conducted with the actual household enterprise owner. 11 The outlier in Senegal is explained by the structure of the underlying registry that oversampled women. 12 This is significantly higher than national averages, but to be expected given both the rural-urban divide and the fact that the underlying registries may have targeted mobile phone owners or distributed phones as a part of the registration process. The national averages are 64 percent in Liberia, 52 percent in Niger and 83 percent in Senegal as per 2021/22 FINDEX data. 7 This is relatively old, as compared to data from other studies.13 Most of our sample consists of petty traders (60 percent), indicating that these are very small businesses, which is confirmed by looking at the number of family and non-family employees. The businesses surveyed have on average less than one family or non-family employee. Only 17 percent of the businesses in our sample are formally registered with any authority, ranging from 13 percent in Senegal to 27 percent in Liberia. This is roughly in line with regional estimates that place urban informality in Western Africa at 87 percent (ILO, 2018). Nearly 78 percent of the businesses in the sample report revenue greater than USD 50 per month, which is just below the extreme poverty line for a single-member household.14 An estimated 43 percent of respondents run businesses requiring electricity to produce goods or render services, of whom only 59 percent are connected to the electricity grid. Water is needed in 40 percent of surveyed enterprises and 43 percent said that road conditions are important for the successful operation of the business. Columns 5 and 6 of the table show that businesses run by women tend to be 4-years younger, are more likely to be petty traders, are three times less likely to be registered, and employ nine times fewer non-family employees, as compared to male-owned businesses. In summary, while there are many similarities across respondents from the three countries (re- spondent age, phone ownership, dominant household enterprise sector, high informality, few employ- ees), there is some heterogeneity. The Liberian sample is more educated and has smaller households than in the other two countries. Their enterprises are half as old as in Niger and Senegal, and about twice as likely to be formal. The Senegalese sample is more female than the other countries. 2.3 Empirical Strategy We explore five questions, using simple summary statistics and regressions. The summary statistics provide averages on the variables of interest, while the regressions allow us to identify which, if any, individual or enterprise characteristics are more (or less) correlated with the corresponding dependent variable. We run the following OLS regression yijc = αc + x1 2 ic β1 + xjc β2 + ϵijc (1) 13 For example, Grimm et al. (2012) found that survivalist enterprises across seven urban centers in West Africa had been operating for an average age of eight years. 14 We do not have data on the exact value of revenue, profits, or income. We only know what “revenue band” a business falls into. The revenue bands are: US$0-US$5, US$5-US$25, US$25-US$50, US$50-US$100, US$100-US$150, and greater than US$150. 8 where yijc is an outcome of individual i with enterprise j in city c, αc are city fixed effects, and x1 ic and x2 jc are 1 × k and 1 × p vectors of variables of interest and controls. We report robust standard errors. Unless otherwise noted, the same set of control variables are used across all regressions. Per- sonal characteristics (x1 ic ) include the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent completed primary education. Household enterprise characteristics (x2 jc ) include the age of the household enterprise in years, a dummy indicating if the business owner is a petty trader, a dummy indicating the registration status of the business, a dummy indicating if the business generates more than US$50 in revenue per month, and a dummy indicating Covid exposure. Throughout the paper, we display in the main tables and figures the results for the full sample pooling the three countries, as well as results by gender. We run each regression with and without the control variables. We present disaggregated results by country in the appendix tables and figures. Given the cross-sectional nature of the data, all results are to be interpreted as correlations and not as causal relationships. 2.4 Outcome Variables We are interested in five outcomes: motivation for being an entrepreneur, job satisfaction, aspira- tions, firm outcomes, and long-run aspirations. 2.4.1 Motivation for Being an Entrepreneur We first explore whether household enterprise owners voluntarily opt-in to starting their businesses or instead feel somewhat forced to do so. The survey asked individuals “Why did you start a household enterprise, rather than another type of activities/business?” Respondents were given ten possible answers (and allowed multiple responses), which we use to group owners into three categories: opt-in, forced-in, and both. The opt-in category includes the responses: “to diversify income,” “flexibility in work hours,” “it is a good business opportunity,” “I do not like to rely on others,” “higher income than in other kinds of work,” “being my own boss is more satisfying than working for someone else,” “can work from home,” and “gives more time with children/family.” The “forced-in” category includes the responses: “easier than finding other work” and “no other job was available.” The “both” category includes those individuals who select answers from both 9 the opt-in and the forced-in sets. For conditional estimates, we create a dummy variable that takes a value of 1 if the business owner opted in to start their household enterprise and 0 if they were forced to start their business. Respondents who cite both opt-in and forced-in responses are not included in the regression. 2.4.2 Job Satisfaction We use responses about subjective work-related well-being. As argued in Falco et al. (2015), subjec- tive measures of job satisfaction are viable proxies for job quality and well-being. These measures are particularly useful when considering the utility of different employment sectors since the fac- tors that affect job satisfaction, and the weights that each respondent assigns to those factors, are mostly unmeasurable and may vary across individuals with different preferences. Subjective indi- cators allow the individual to identify the factors and weights to arrive at a latent job satisfaction measure. We rely on two variables to assess subjective well-being. First, we use the responses to a 5-point Likert scale for the question: “All things considered, how satisfied are you with your current job?” For the regressions, we construct a dummy variable where “satisfied” or “very satisfied” take a value of 1 while the other responses take a value of 0. Second, we asked respondents “If you could have any job you wanted, what job would you really like to have?” providing them with six possible answers: “my current job,” “running a different business,” “work as a professional (doctor etc.),” “working abroad,” “work as salary employee in private sector,” or “work as salary employee in public sector (wage job).”15 We create a dummy variable that takes a value of 1 if the respondent’s current job is their ideal job to use in the regression analysis. We run two sets of regressions. First, we estimate differences in job satisfaction between house- hold enterprise owners and employees. Second, we estimate the gap between household enterprise owners and employees in their response to whether the current job is their ideal job. 2.4.3 Aspirations and Constraints To measure aspirations of household enterprise owners we again use perception questions, allowing the respondent to include measurable and unmeasurable factors in the response, and assign indi- vidual weights to each. Specifically, we asked individuals “What are your plans for the household 15 For respondents who do not own a household enterprise, the difference between the answers “current job” and “work as salary employee in private sector” are that the latter measures the desire for another private sector job. 10 enterprise next year?” offering ten dimensions of aspirations as response options: (i) no change, (ii) close the business, (iii) hire more workers, (iv) sell to new clients, (v) increase profits/sales, (vi) change location, (vii) register the firm, (viii) improve the quality of the goods/services, (ix) change activity in the same sector, and (x) change sector. Individuals stated whether each of the ten possible answers represents one of their aspirations. We use these categories when presenting descriptives. Second, we zoom in on the most prevalent business aspirations (options iii, iv, v, and viii) and create a continuous variable that counts the number of business aspirations a business owner holds, ranging from 0 to 4.16 This is a different way of asking individuals about aspirations than previous approaches in the literature, capturing a dichotomous measure of several potential aspirations rather than the standard approach of creating a continuous measure of aspirations on one dimension.17 To capture information on why owners do not take steps to realize their aspirations, we consider the main challenges faced by business owners, as well as the shocks they have experienced in the past 6 months. Enterprise owners were asked “What are the main challenges in running the household enterprise?” Responses range over financial (i—ii), inputs and competition (iii—ix) infrastructure (x—xiii), supply side (xiv—xvi), crime and corruption (xvii—xviii), and family (xix—xx) issues: (i) shortage of money, (ii) access to finance or loans, (iii) poor quality of inputs, (iv) input prices, (v) access to equipment, (vi) loss of perishable goods, (vii) high rent, (viii) low sales or competition, (ix) non-paying customers, (x) electricity access, (xi) road conditions or transport, (xii) internet connectivity, (xiii) water access, (xiv) limited own skills, (xv) limited worker skills, (xvi) low avail- ability of workers, (xvii) crime, (xviii) harassment by authorities, (xix) family responsibilities, and (xx) illness or poor health. For the regressions, we construct a variable called “business constraints” that counts how many of the first 18 constraints listed above the respondent identifies, and another variable called “family constraints” that captures whether the respondent mentions constraint (xix), (xx), or both. 16 This variable is a continuous measure of business aspirations. Instead of measuring the degree or intensity of an aspiration of an individual (as the measures in the literature do with, for example, profits), it counts the number of dimensions of aspirations an individual holds. The two are not the same. 17 The more conventional approach usually focuses on only one dimension (namely profits) and asks business owners to state or predict by how much they want to increase profits within, for example, the next year (e.g., Janzen et al., 2017; Dalton et al., 2018). This yields a continuous measure of aspirations. The advantages of our approach are obvious: we can capture ten dimensions of aspirations instead of just one. We believed this to be important when writing the survey as aspirations contain more layers than just profits. The drawback, however, is just as obvious: we are left with a dichotomous measure of aspirations, while the more conventional approach followed in the literature often relies on continuous measures. These continuous measures allow researchers to measure things like the aspiration gap (e.g., Ray, 2006; Genicot and Ray, 2017), something we cannot do. 11 We carry out a similar process to identify the shocks faced by the respondent in the 6 months prior to the survey, asking owners “Has your household been negatively affected by the following problem over the past 6 months?” offering 15 responses that can be grouped into three categories. Four family shocks are queried: death/disability of a working household member, illness of an earning household member, job loss of a household income contributor, and increase in household expenses. Three natural shocks are: floods, droughts, and Covid disruption. The remaining eight shocks relate to the business: business failure, theft of money/property, lack of workers for household enterprise, increase in input prices, fall in the selling price of products or services, shortage of inputs, damage of a productive asset, and damage to the operating space. We create three categories of shocks (family, nature, business) for use in the regressions. We evaluate business owners’ aspirations in multiple steps. First, we regress the number of business expansion aspirations on our satisfaction measurements, demographics, and firm char- acteristics. Second, we consider what steps, if any, business owners take to achieve their stated aspirations. In the absence of panel data (Dalton et al., 2018), we ask business owners to tell us what steps they have taken in the past 12 months to achieve each of the ten aspirations we ask about. Finally, we assess the role of constraints and shocks in hindering proactivity in realizing aspirations. 2.4.4 Firm Outcomes We ask if the motivation for starting the business, satisfaction, and aspirations actually matter for business outcomes. We estimate the correlation between those variables and two proxies for firm outcomes: First, the size of the firm as measured by the number of paid employees, and second, whether the firm is registered with authorities, as a proxy for firm formality.18 2.4.5 Long-run Aspirations We end by asking respondents “If offered a wage job that pays the same as you take home now from the household enterprise, would you take it?” For our analysis, we create a variable that equals one if the respondent answers “yes.” For the regression, we limit the sample to household enterprise owners and regress the willingness to accept a wage job on respondents’ subjective well-being and ideal job. 18 Due to anomalies in the profits variable, we cannot use it as an outcome variable. See footnote 14. 12 3 Results The data confirm that urban poor household enterprise owners largely opted to start their busi- nesses, are predominately satisfied with their job, and view it is as their “ideal job.” They aspire to expand their enterprises, yet business, family and nature-related factors constrain them in act- ing on these aspirations. Their job preferences are shaped by market opportunities, as two-thirds would be open to taking a wage job if it paid as well as owning a micro-enterprise. However, the fact that one-third would not make the switch suggests that factors other than earnings affect micro-entrepreneurs’ preferences. 3.1 Poor Urban Africans Largely Opt into Household Enterprise Own- ership Household enterprise owners overwhelmingly opted to start their business. Table 3 summarizes the motivational responses. The top panel displays the three types of business owners in our data, i.e., individuals who opted in to business ownership, those who were forced in, or both. The other two panels display the ten reasons individuals were presented with: the first eight being opt-in reasons (middle panel) and the last two being forced-in reasons (bottom panel). Overall, 53 percent of our respondents only listed job-related factors that were better addressed through opening a business rather than wage employment. Another 26 percent were enterprise owners only due to forced-in factors, and about 21 percent fall into the “both” category. Women are more likely than men to have started their household enterprise as a last resort, though these women are a minority of all self-employed women. When considering positive factors that led to the opening of the business, economic consider- ations dominate, with individuals viewing opening a business as a good opportunity (18 percent) that is more profitable than wage labor (24 percent) and allows for income diversification (27 per- cent). Furthermore, individuals like independence as they indicate that it is more satisfying to be one’s own boss (19 percent) and that they do not like to rely on others (17 percent). Among the two “negative” factors individuals were presented with, 32 percent state that it is easier to open a business than to find wage labor while 19 percent mention that no other work was available. This is a striking finding since it suggests that among poor household enterprise owners, only a minority are there because they are unable to find a wage job. 13 The high level of opt-in household enterprise ownership is consistent across countries (Table A2). Niger has the largest share of those who opted to start businesses, where only 17 percent of the sample did not give an opt-in reason for firm start-up. Senegal had the most reluctant household enterprise owners, with one-third of the sample only giving forced-in reasons for start-up, mostly due to greater ease in starting a business than in finding another type of work. The motivation for starting a household enterprise differs little by personal or business charac- teristics. Table 4 shows that men are 9 percentage points more likely to have opted in to start their business than women. Individuals with at least a primary education are more likely to opt into business ownership than those with less education. The significance of this effect disappears when splitting the sample by gender in Columns 2 and 3, though this may be due to a lack of power as the magnitude of the coefficients is comparable across columns.19 3.2 Household Enterprise Owners Are Satisfied in their Jobs Household enterprise owners are satisfied with their jobs both in absolute and in relative terms in comparison with those who are not owners. Figure 1 Panel (a) shows that 80 percent of household enterprise owners are “satisfied” or “very satisfied” with their jobs as compared to 65 percent of employees (Panel (c)). In line with this, Figure 1 Panel (b) shows that almost 70 percent of enterprise owners view their current job as their ideal job, with another 18 percent saying that they would like to run a different business. In contrast, only 30 percent of wage earners consider their current job to be ideal (Panel (d)), and in fact a larger percentage (40 percent) say running a business is their ideal job. To approach the question of preferences from a different perspective, we asked respondents whether they would like their children to become household enterprise owners. Regardless of gender and whether respondents themselves were business owners, more than 80 percent of respondents stated that they would like to see their children become household enterprise owners (Figure 3). We cannot conclude that household enterprise ownership is the optimal job in an unconstrained context, but it does seem better than the alternative, even for those (wage workers) who work in an alternative sector (see section 3.5). The job satisfaction trends are similar across genders (Figure 2, Panels (a)—(d)). Male and female household enterprise owners have similar views on their ideal job (Panels (e) and (f)), while female employees are nearly 25 percentage points more likely than male employees to prefer running 19 The results by country are presented in Table A3. 14 a business (Panels (g) and (h)). Household enterprise owners across countries are consistent in terms of their satisfaction with their jobs and their idea of an ideal job (Figure A1), but employee responses differ by country (Figure A2). Job satisfaction is particularly high among Nigerien employees (80 percent) as compared to the other two countries, though not as high as the satisfaction of Nigerien household enterprise owners (90 percent). Similarly, a higher proportion (45 percent) of Nigerien employees say that they are working in their ideal job, as compared to Senegalese or Liberians employee, though, again, much less than Nigerien household enterprise owners (nearly 80 percent). Our regressions confirm the summary statistics. Household enterprise owners are 20 percentage points more likely than employees to say that they are satisfied with their jobs (Table 5). Similarly, household enterprise owners are 40 percentage points more likely than employees to view their current employment as ideal (Table 6).20 These results are robust when controlling for respondent demographics. The gap is higher for women than men. When disaggregating by country, the results hold, though the satisfaction estimates are not statistically significant for Niger (Table A4). This may reflect the very high level of satisfaction among all Nigerien workers, as reflected in the unconditional correlations (see Figure A2). The gap in the ideal job is particularly high in Senegal, as compared to the other countries (Table A5). Economic and non-economic factors underpin the preference for owning a household enterprise. More than 80 percent of household enterprise owners say that they can make more money by running a business than by being an employee. Among the non-economic factors in the survey, more than 80 percent of respondents appreciate the time flexibility and working conditions that come from running a household enterprise (Panels (a), (c), (e), and (g) of Figure A3). The picture remains unchanged for employees (Panels (b), (d), (f), and (h) of Figure A3). In sum, not only do they overwhelmingly value the non-monetary benefits of being self-employed, but they do not have to sacrifice income to achieve them. 20 We also find that respondents who opted into starting their business are more likely to state that their current job is their ideal job. These results are available upon request. 15 3.3 Household Enterprise Owners Aspire to Grow Their Profits, but Not to Formalize If household enterprise owners do not aspire to close their firms and transition to perceived less profitable wage jobs, what do they aspire to? Most household enterprise are not stagnant holding patterns for salaried labor. Owners aspire to expand their businesses. Panel (a) of Figure 4 displays the share of individuals who indicated they had any of the ten aspirations in the survey. Three stand out: increasing profits/sales, selling to new clients, and improving the quality of goods/services, with roughly 60 percent, 40 percent, and 20 percent of the sample, respectively, naming each. The other seven aspirations are of much lower importance, including the aspiration of formalizing a firm through registration (about 5 percent) and the aspiration to hire more workers (only about 10 percent). When grouping the ten aspirations into conceptually similar categories, aspirations related to expanding the business clearly dominate, cited by over 80 percent of respondents (Panel (b)). The other groups, not surprisingly given the results from Panel (a), do not matter much. Men and women express similar aspirations for their businesses (Figure A4). The country disaggregation finds that Niger and Senegal follow the overall trend, while Liberian entrepreneurs are particularly focused on increasing profits/sales and improving the quality of their products (Figure A4). The conditional estimates show that entrepreneurs who are happier with their business have more aspirations to expand it (Table 7). Individuals who say that they are willing to accept an outside equivalent salaried job have fewer such aspirations (Table 7). This effect remains highly significant even after controlling for personal and business characteristics. Personal characteristics are not significant predictors of an individual’s aspirations. Business characteristics, however, are. Being a petty trader, having businesses with revenues above US$50 per month, and having more non-family employees are associated with having more growth aspirations. The broad take-aways hold when looking at trends within gender. Columns 4—9 of Table 7 show one main difference from the aggregate trends. The positive correlation between number of non-family employees and business aspirations is driven by men. This is due to the fact that women barely employ non-family employees (see Table 1). The results are similar when disaggregating by country. Table A6 reruns the regressions for each country. While the results here are less consistent, the willingness to accept an outside offer still exhibits a clear and strong negative correlation with business aspirations in Niger and Senegal. 16 Personal characteristics still do not seem to matter in any of the three countries, and revenue and being a petty trader still has a positive effect in Niger and Senegal. Although household enterprise owners in our sample aspire to expand their businesses, they only take modest actions to do so. Panels (a), (b), and (c) of Figure 5 display the actions business owners have taken to achieve their stated aspirations for the three most frequently mentioned aspirations. The striking pattern across all three panels is that more than half of surveyed individuals report to have done nothing to achieve their aspirations. The actions that they state as having taken are easier to carry out and are less permanent, such as marketing activities or making special deals, as opposed to hiring new workers or using new technologies. This may reflect constraints to taking more costly and permanent actions. Household enterprise owners report financial, personal, and market-related challenges that hin- der the realization of aspirations. Panel (a) of Figure A5 displays constraints that business owners currently face. Twelve constraints included in the survey affect at least 10 percent of respondents. The most cited difficulties relate to financing: 80 percent identify a “shortage of money” and 50 percent have trouble with “access to finance and loans.”21 However, the third most frequently cited constraint is personal, with 41 percent of respondents identifying “family responsibilities” as a con- straint to the business. Own illness is cited by 25 percent of the sample. A range of barriers to inputs affect 15-30 percent of respondents, including prices, quality, and access. Market-related challenges round out the challenges. Regression analysis suggests that owners with family constraints seem to have fewer aspirations to expand their business (Table A7) and, if they are women, are less likely to take action to realize their aspirations (Table A8). Business constraints do not affect expansion aspirations nor the propensity to act on them, with one exception: women who face more business constraints are less likely to attempt to expand (Tables A7 and A8). Business and natural shocks directly affect the formulation and pursuit of aspirations and may indirectly affect them by shaping constraints. The top five shocks faced by owners were: increases in household expenses (49 percent), being affected by Covid (49 percent), increase in input prices (34 percent), illness of a household member (26 percent), and flood exposure (20 percent) (Figure A5 Panel (b)). Regression analysis shows a positive correlation between the number of family, business or nature shocks and the number of business constraints. Business and family shocks are positively 21 A follow-up question finds that the financing is primarily needed for business purposes—to purchase raw mate- rials, improve the business premises, and acquire/maintain equipment—though more than 10 percent identify food consumption. 17 correlated with non-business constraints, pointing to a blurring of household and enterprise risk management.22 Owners who report having experienced more business shocks in the last 6 months have fewer aspirations to expand their business (Table A7). This correlation is particularly driven by women. Those who experienced natural shocks, in contrast, had a slight tendency to have greater aspirations for firm expansion (Table A7). No category of shocks induced owners to undertake actions to expand their business (Table A8), with women who experienced business shocks being particularly averse to expansion. 3.4 More Satisfied Entrepreneurs Have Larger, but Not Necessarily Formal, Businesses This section examines whether previously analyzed factors—such as whether owners started their businesses voluntarily, their job satisfaction, perception of their current job as ideal, and their aspirations—influence business outcomes, measured by the number of employees and whether the firm is formally registered. Table 8 shows that opting in to start an enterprise and identifying household enterprise ownership as the ideal job are positively correlated with a larger business size. Columns 1 and 6 in Table 8 show that a business owner who opted to start the business has 0.37 more non-family employees than those who were forced to start the business. Similarly, entrepreneurs who report that they are in their ideal job employ more paid workers (Columns 3 and 6). Business and personal characteristics play strong roles in determining business size (Table 8, Columns 5 and 6). Businesses tend to be larger when they are run by men, when the owner is younger, when the firm is older, and when it is formally registered. Being a petty trader is significantly and negatively correlated with business size. The results on the motivation for starting a business, job satisfaction, ideal job, and aspirations from Table 8 are largely driven by businesses owned by men (Table 9). Women do not show strong correlations (Table 10). Furthermore, the coefficient estimates in the sample containing only women is an order of magnitude smaller than in the sample containing only men.23 Firm owner’s job satisfaction, identification as ideal job, opting into starting the business, or the number of aspirations could not be identified as positively correlated with firm formality once control 22 Regression estimates are available upon request. 23 The aggregate results are largely driven by Liberia and Niger (Tables A9, A10, and A11). In Senegal, nothing but the motivation for starting the household enterprise was identified as statistically significant. 18 variables were introduced (Table 11, Column 6). In line with literature findings, the data show that the registration status of the firm depends significantly on personal and business characteristics. If the business owner is male, older, and more educated, the firm is more likely to be registered, while petty traders are less likely to be registered (Table 11, Column 7). These results are driven by male household enterprise owners (Tables 12 and 13).24 3.5 In the Long Run, Wage Work Can Be an Attractive Option Despite the very positive views of owning a household enterprise, wage employment remains a valued option. When asked whether household enterprise owners would accept a wage job that paid as much as they earned in self-employment, around two-thirds of household enterprise owners state that they would accept the offer (Table 14). The response is similar across countries (Table A15). This trend is partly driven by those who are unhappily running a household enterprise. Table 14 reports that household enterprise owners who are happy with their jobs are 13 percentage points less likely to accept that outside offer, while those who see their current job as ideal are 27 percentage points less likely to accept an equivalently paid employee position. The results are similar for both men and women. The benefits of job security may play into the decision, as 60 percent of respondents felt that working as an employee offers more job stability (Panel (a) of Figure A6). Non-household enterprise owners also view working as an employee as providing more job stability (Panel (b) of Figure A6). Based on these data, our interpretation is that self-employment pays more and offers more flexibility but has a high level of uncertainty, making wage employment an attractive alternative if available at equal pay, which is often not the case. 4 Conclusion and Policy This paper illustrates that micro-entrepreneurship is a viable option for poor West Africans in constrained labor markets but may play a lesser—though still important—role once structural transformation improves access to and value of wage employment. Our results suggest that most surveyed business owners voluntarily started their business, are satisfied with their jobs, and aspire to and have plans to expand their businesses. Most report that they earn more than they could 24 The aggregate results are driven by Liberia and Niger (Tables A12, A13, and A14). 19 as wage earners, with wage earners confirming the observations. However, a combination of family and business constraints and shocks may hinder their ambitions, ability to act on their goals, and realization of those goals. That said, two-thirds of micro-enterprise owners said they would accept a wage job if it offered wages on par with their current earnings. This may indicate that today’s micro-entrepreneurs are not averse to wage work in principle. But they would rather be household enterprise owners than work as an employee at the current market wage. The observation that one-third of household enterprise owners would not take an equally paid wage job aligns with the broader literature that argues that unobservable factors, such as entrepreneurial spirit and other non-pecuniary benefits of being self-employed, are sufficiently valued so micro-entrepreneurship may persist even as earnings across self-employment and wage employment equalize for poor workers.25 The results suggest that household enterprise owners support a dual policy approach, valuing policies to enhance their enterprises, as well as structural reforms that lead to more remunerative wage employment opportunities in the long run. Our descriptive findings point to two types of interventions. First, multi-faceted packages of support have repeatedly shown to be a cost-effective strategy to help poor entrepreneurs operating in constrained economies to overcome their many challenges. Programs that provide a package of basic entrepreneurship and life skills, financial literacy, personal initiative coaching, and seed capital, often referred to as “economic inclusion” or “graduation” programs, have been implemented in more than 50 countries and evaluated through dozens of randomized controlled trials. They are often coupled with social assistance interventions such as cash transfers, often scalable in the face of shocks. While there is continuing debate about the most effective program design and transfer values (Banerjee et al., 2024), they regularly have high rates of return as measured by income, business revenue, savings, and accumulation of productive assets; create sustainable jobs that persist long after beneficiaries graduate; and are relatively low-cost (Banerjee et al., 2015; Haushofer and Shapiro, 2016; Bandiera et al., 2017; Blattman et al., 2020; Egger et al., 2022; Balboni et al., 2022). Second, household enterprise owners would benefit from policies enabling better protection from shocks. A range of personal, business, and natural disaster shocks limits business aspirations and impedes the success of many small urban businesses. There is growing evidence on risk management tools that are suitable for household enterprise owners who tend to have low earnings, uncertain 25 See Maloney (1999, 2003, 2004); Perry et al. (2007) for Latin America and Danquah et al. (2019) for Africa. 20 income, and a lack of access to public risk management systems (e.g., pensions, health insurance). For example, cash transfers that are triggered by a shock—such as a pandemic (Gentilini, 2022) or a natural disaster (Bowen et al., 2022)—can offset the cost of shocks. Universal health insurance may enable greater entrepreneurship among the poor. Public finances can be used to encourage savings behaviors, such as matching grants for micro-enterprise owners who voluntarily contribute to publicly managed savings funds. Finally, while the paper does answer some questions, it highlights further questions for future research. In particular, there is a need for panel data that tracks individuals and their aspirations over time. Such data would allow researchers to examine whether and how people act on their goals, what prevents them from doing so, and how different constraints or interventions shape both aspirations and outcomes. 5 Bibliography Adoho, F. M. and Doumbia, D., 2022. ‘Informal sector heterogeneity and income inequality: Evi- dence from the Democratic Republic of Congo’, Journal of Economic Development 47(4), 55–77. evalo-S´ Ar´ anchez, I., Heisey, J., Chaudhary, S., Clay, T. and Strokova, V., 2024. ‘The state of economic inclusion report 2024’, World Bank Publications-Books . Babangida, J. S., 2023. Breaking barriers to inclusion: Assessing job security, school-to-work transi- tion and discouragement effects of youth in Nigeria, Tech. rep., STEG (Structural Transformation and Economic Growth), CEPR. Balboni, C., Bandiera, O., Burgess, R., Ghatak, M. and Heil, A., 2022. ‘Why do people stay poor?’, The Quarterly Journal of Economics 137(2), 785–844. Bandiera, O., Burgess, R., Das, N., Gulesci, S., Rasul, I. and Sulaiman, M., 2017. ‘Labor markets and poverty in village economies’, The Quarterly Journal of Economics 132(2), 811–870. Bandiera, O., Elsayed, A., Heil, A. and Smurra, A., 2022. ‘Presidential address 2022: Economic development and the organisation of labour: Evidence from the jobs of the world project’, Journal of the European Economic Association 20(6), 2226–2270. e, W., Shapiro, J., Thuysbaert, Banerjee, A., Duflo, E., Goldberg, N., Karlan, D., Osei, R., Parient´ B. and Udry, C., 2015. ‘A multifaceted program causes lasting progress for the very poor: Evidence from six countries’, Science 348(6236), 1260799. Banerjee, A., Hanna, R., Olken, B. A. and Lisker, D. S., 2024. ‘Social protection in the developing world’, Journal of Economic Literature 62(4), 1349–1421. 21 Berner, E., Gomez, G. and Knorringa, P., 2012. “Helping a large number of people become a little less poor’: The logic of survival entrepreneurs’, The European Journal of Development Research 24, 382–396. Blattman, C. and Dercon, S., 2018. ‘The impacts of industrial and entrepreneurial work on in- come and health: Experimental evidence from Ethiopia’, American Economic Journal: Applied Economics 10(3), 1–38. Blattman, C., Fiala, N. and Martinez, S., 2020. ‘The long-term impacts of grants on poverty: Nine- year evidence from Uganda’s youth opportunities program’, American Economic Review: Insights 2(3), 287–304. Bossuroy, T., Koussoube, E. and Weeden, E., 2017. ‘Design of productive inclusion interventions: Identifying constraints and opportunities for household agriculture and enterprises’, World Bank Mimeo . Bowen, T., del Ninno, C., Andrews, C., Coll-Black, S., Gentilini, U., Johnson, K., Kawasoe, Y., Kryeziu, A., Maher, B. and Williams, A., 2022. ‘Adaptive social protection: Building resilience to shocks’, World Bank: Washington, DC . Choto, P., Tengeh, R. K. and Iwu, C. G., 2014. ‘Daring to survive or to grow? the growth aspirations and challenges of survivalist entrepreneurs in South Africa’, Environmental Economics (5, Iss. 4), 93–101. Co, M. J., Gronewald, J., Mitchell, B., Nayager, T., van Zyl, J. and Visser, K., 2006. Entrepreneur- ship: Fresh Perspectives, Pearson Education, Cape Town. Cunningham, W., Newhouse, D., Ricaldi, F., Seuyong, F. T., Viollaz, M. and Edochie, I., 2024. ‘Urban informality in sub-Saharan Africa: Profiling workers and firms in an urban context’, Policy Research Working Paper #10703. Washington DC: World Bank . uschenp¨ Dalton, P. S., R¨ ohler, J. and Zia, B., 2018. ‘Determinants and dynamics of business aspira- tions: Evidence from small-scale entrepreneurs in an emerging market’. Danquah, M., Schotte, S. and Sen, K., 2019. Informal work in sub-Saharan Africa: Dead end or steppingstone?, number 2019/107, WIDER working paper. Danquah, M., Schotte, S. and Sen, K., 2021. ‘Informal work in sub-Saharan Africa: Dead end or stepping-stone?’, IZA Journal of Development and Migration 12(1). Diao, X., Kweka, J., McMillan, M. and Qureshi, Z., 2020. ‘Economic transformation in Africa from the bottom up: New evidence from Tanzania’, The World Bank Economic Review 34(Supple- ment 1), S58–S62. Egger, D., Haushofer, J., Miguel, E., Niehaus, P. and Walker, M., 2022. ‘General equilibrium effects of cash transfers: Experimental evidence from Kenya’, Econometrica 90(6), 2603–2643. 22 Falco, P., Maloney, W. F., Rijkers, B. and Sarrias, M., 2015. ‘Heterogeneity in subjective well- being: An application to occupational allocation in Africa’, Journal of Economic Behavior & Organization 111, 137–153. Fox, L. and Sohnesen, T. P., 2012. ‘Household enterprises in sub-Saharan Africa: Why they matter for growth, jobs, and livelihoods’, World Bank Policy Research Working Paper (6184). Fox, L. and Sohnesen, T. P., 2013. ‘Household enterprises in Mozambique: Key to poverty reduction but not on the development agenda?’, World Bank Policy Research Working Paper (6570). Genicot, G. and Ray, D., 2017. ‘Aspirations and inequality’, Econometrica 85(2), 489–519. Gentilini, U., 2022. ‘Cash transfers in pandemic times: Evidence, practices, and implications from the largest scale up in history’, World Bank: Washington, DC . Gollin, D. and Kaboski, J. P., 2023. ‘New views of structural transformation: Insights from recent literature’. Grimm, M., Knorringa, P. and Lay, J., 2012. ‘Constrained gazelles: High potentials in West Africa’s informal economy’, World Development 40(7), 1352–1368. Guven, M., Jain, H. and Joubert, C., 2021. ‘Social protection for the informal economy: Operational lessons for developing countries in Africa and beyond’, World Bank: Washington, DC . Harris, J. R. and Todaro, M. P., 1970. ‘Migration, unemployment and development: A two-sector analysis’, The American Economic Review 60(1), 126–142. Haushofer, J. and Shapiro, J., 2016. ‘The short-term impact of unconditional cash transfers to the poor: experimental evidence from Kenya’, The Quarterly Journal of Economics 131(4), 1973– 2042. Hirvonen, K., Leight, J. and Zafar, S., 2024. The effectiveness of cash and cash plus interventions on livelihoods outcomes: Evidence from a systematic review and meta-analysis, IFPRI Discussion Papers 2262, International Food Policy Research Institute (IFPRI). ILO, 2018. ‘Women and men in the informal economy: A statistical picture’, International Labour Office, Geneva . Janzen, S. A., Magnan, N., Sharma, S. and Thompson, W. M., 2017. ‘Aspirations failure and formation in rural Nepal’, Journal of Economic Behavior & Organization 139, 1–25. Knox, A. J., Bressers, H., Mohlakoana, N. and De Groot, J., 2019. ‘Aspirations to grow: when micro-and informal enterprises in the street food sector speak for themselves’, Journal of Global Entrepreneurship Research 9, 1–24. Kweka, J. and Fox, L., 2011. ‘The household enterprise sector in Tanzania: Why it matters and who cares’, World Bank Policy Research Working Paper (5882). 23 Maloney, W. F., 1999. ‘Does informality imply segmentation in urban labor markets? Evidence from sectoral transitions in Mexico’, The World Bank Economic Review 13(2), 275–302. Maloney, W. F., 2003. Informal self-employment: poverty trap or decent alternative?, in ‘Pathways out of poverty: Private firms and economic mobility in developing countries’, Springer, pp. 65–82. Maloney, W. F., 2004. ‘Informality revisited’, World Development 32(7), 1159–1178. Margolis, D. N., 2014. ‘By choice and by necessity: Entrepreneurship and self-employment in the developing world’, The European Journal of Development Research 26, 419–436. Nagler, P., 2017. ‘A profile of non-farm household enterprises in sub-Saharan Africa’, Maastricht Economic and Social Research institute on Innovation and Technology (UNU-MERIT) working Paper 48. Perry, G., Maloney, W., Arias, O., Fajnzylber, P., Mason, A. and Saavedra, J., 2007. Informality: Exit and exclusion, World Bank: Washington, DC. Ray, D., 2006. ‘Aspirations, poverty, and economic change’, Understanding Poverty 1, 409–421. Sinha, N., Narayan, A. et al., 2024. Leveling the playing field: Addressing structural inequalities to accelerate poverty reduction in Africa, World Bank: Washington, DC. Turkson, D. and Codjoe, E. A., 2020. ‘The household enterprise sector in Ghana: Overview and challenges’, International Journal of Business and Management 15(8). Ulyssea, G., 2020. ‘Informality: Causes and consequences for development’, Annual Review of Economics 12(1), 525–546. World Bank Group, 2018. ‘Country Partnership Framework for the Republic of Niger for the Period FY18-FY22’, World Bank: Washington, DC . World Bank Group, 2020. ‘Country Partnership Framework for the Republic of Senegal for the Period FY20-FY24’, World Bank: Washigton, DC . World Bank Group, 2025. ‘Country Partnership Framework for the Republic of Liberia for the Period FY25-30’, World Bank: Washington, DC . 24 6 Tables and Figures Table 1: Sample Characteristics HHE Owners Non-HHE Owners (1) (2) (3) (4) (5) (6) (7) All Liberia Niger Senegal Men Women All Male 0.477 0.491 0.636 0.311 0.584 (0.500) (0.501) (0.482) (0.463) (0.494) Age 43.777 38.898 41.323 49.608 43.107 44.365 43.188 (11.628) (10.920) (11.227) (9.957) (11.417) (11.784) (12.691) Primary education 0.330 0.708 0.250 0.144 0.395 0.272 0.387 (0.470) (0.455) (0.433) (0.352) (0.489) (0.445) (0.488) Household size 9.134 5.957 9.580 10.930 9.105 9.169 7.937 (4.955) (2.825) (4.783) (5.234) (5.037) (4.880) (4.495) Number of children below 14 3.473 2.385 4.266 3.452 3.831 3.153 2.939 (2.463) (1.637) (2.597) (2.518) (2.658) (2.225) (2.213) Owns a mobile phone or tablet 0.938 0.930 0.955 0.928 0.970 0.910 (0.241) (0.255) (0.207) (0.259) (0.171) (0.287) Owns a laptop or computer 0.058 0.105 0.041 0.040 0.096 0.023 (0.233) (0.307) (0.199) (0.197) (0.295) (0.149) Employed 0.573 (0.495) Observations 1526 400 557 569 727 798 382 Notes: The table displays mean sample characteristics and standard deviations (in parentheses) for a variety of characteristics of household enterprise (HHE) and non-household enterprise (non-HHE) owners. These include the gender and age of the respondent in years, a dummy indicating whether the respondent completed primary education, the household size, the number of children below 14 living in the household of the respondent, a dummy indicating whether the respondent owns a mobile phone or a tablet, a dummy indicating laptop or computer ownership, as well as a dummy variable indicating whether the respondent is employed. Columns 1—6 display the results for HHE owners, while Column 7 focuses on non-HHE owners. Column 1 displays the characteristics across the full HHE owner sample, Columns 2—4 split that sample by country, and Columns 5 and 6 split that sample by gender. 25 Table 2: Household Enterprise Characteristics (1) (2) (3) (4) (5) (6) All Liberia Niger Senegal Men Women Age of household enterprise 10.475 6.384 11.242 12.585 12.770 8.387 (9.533) (6.366) (9.479) (10.517) (10.468) (8.048) Petty trader 0.601 0.640 0.532 0.640 0.419 0.766 (0.490) (0.481) (0.499) (0.481) (0.494) (0.424) Business registered 0.174 0.273 0.151 0.127 0.270 0.086 (0.379) (0.446) (0.358) (0.333) (0.444) (0.281) Revenue above 50USD per month 0.778 0.865 0.622 0.843 0.816 0.747 (0.416) (0.342) (0.486) (0.364) (0.388) (0.435) Number of family employees 0.680 0.885 0.691 0.524 0.674 0.685 (1.071) (1.104) (1.191) (0.884) (1.149) (0.997) Number of non-family employees 0.482 0.415 0.654 0.362 0.902 0.100 (1.382) (1.192) (1.577) (1.285) (1.856) (0.459) Business requires electricity 0.433 0.527 0.402 0.397 0.462 0.407 (0.496) (0.500) (0.491) (0.490) (0.499) (0.492) If electricity required, business connected to grid 0.589 0.550 0.643 0.571 0.592 0.585 (0.492) (0.499) (0.480) (0.496) (0.492) (0.494) Business requires water 0.389 0.307 0.506 0.332 0.393 0.386 (0.488) (0.462) (0.500) (0.471) (0.489) (0.487) Road conditions are important for business 0.428 0.560 0.339 0.422 0.487 0.373 (0.495) (0.497) (0.474) (0.494) (0.500) (0.484) Observations 1526 400 557 569 727 798 Notes: The table displays mean sample characteristics and standard deviations (in parentheses) for a variety of household enterprise characteristics. These include the age of the household enterprise in years, a dummy indicating if the business owner is a petty trader, a dummy indicating the registration status of the business, a dummy indicating if the business generates more than 50USD in revenue per month, the number of family and non-family employees, a dummy indicating whether the business requires electricity to produce goods or render services, a dummy indicating that the household enterprise is connected to the electricity grid (conditional on requiring electricity to run the business), a dummy indicating that the business needs water for its activity aside from typical consumption and hygiene, and a dummy indicating whether road conditions are an important factor for the successful operation of the business. Column 1 displays the characteristics across the full sample, Columns 2—4 split the sample by country, and Columns 5 and 6 split the sample by gender. 26 Table 3: Motivation for Starting Household Enterprise (1) (2) (3) All Men Women Opted in to start business 0.527 0.608 0.453 (0.499) (0.489) (0.498) Forced to start business 0.264 0.204 0.319 (0.441) (0.403) (0.466) Both 0.209 0.188 0.228 (0.407) (0.391) (0.420) To diversify income 0.270 0.319 0.227 (0.444) (0.466) (0.419) Flexibility in work hours 0.030 0.030 0.030 (0.172) (0.172) (0.171) It is a good business opportunity 0.177 0.242 0.118 (0.382) (0.429) (0.323) I do not like to rely on others 0.168 0.155 0.180 (0.374) (0.362) (0.385) Higher income than in other kinds of work 0.242 0.245 0.239 (0.428) (0.430) (0.427) Being my own boss is more satisfying than working for someone else 0.189 0.266 0.118 (0.391) (0.442) (0.323) Can work from home 0.064 0.024 0.101 (0.245) (0.152) (0.301) Gives more time with children/family 0.060 0.018 0.098 (0.238) (0.133) (0.298) Easier than finding other work 0.316 0.274 0.354 (0.465) (0.446) (0.478) No other work was available 0.191 0.145 0.232 (0.393) (0.353) (0.422) Observations 1517 722 794 Notes: The table describes the motivation of household enterprise owners for starting their business. Column 1 displays the overall means/shares and standard deviations (in parentheses), while Columns 2—3 split the sample by gender. When asked about their motivation for starting the business, owners were presented with ten possible answers, all of which are summarized in the bottom two panels. We take their answers and create three “types” of respondents, shown in the top panel. Individuals who answered “to diversify income”, “flexibility in work hours”, “it is a good business opportunity”, “I do not like to rely on others”, “higher income than in other kinds of work”, “being my own boss is more satisfying than working for someone else”, “can work from home”, and “gives more time with children/family” are classified as having “opted in” to business ownership. On the other hand, individuals who answered “easier than finding other work” and “no other job was available” are classified as having been forced to start business. In between, the third type (“both”) is an individual who picked answers from both categories. In Senegal, nine individuals do not fit in any of the three categories. The shares displayed here are thus conditional on fitting into one of these three categories. 27 Table 4: Predictors of Motivation for Starting Household Enterprise Owner opted in to start business All Men Women (1) (2) (3) Personal Characteristics Male 0.093∗∗∗ (0.031) Primary education 0.066∗∗ 0.047 0.066 (0.033) (0.045) (0.049) Business Characteristics Age when HHE owner started HHE -0.000 -0.001 0.001 (0.001) (0.002) (0.002) Petty trader -0.034 -0.025 -0.023 (0.030) (0.038) (0.047) Mean of outcome 0.666 0.749 0.587 City Fixed Effects Yes Yes Yes Robust Standard Errors Yes Yes Yes Observations 1128 542 586 Notes: The table displays OLS regressions of a dummy variable that equals to one if the household enterprise owners opted to start their business, where “opt-in” is defined as in Table 3. The variable equals to zero if the business owner had no other choice but to open the household enterprise, where the definition again follows the one adopted in Table 3. All regressions include city fixed effects and report robust standard errors. Personal characteristics include the gender and a dummy indicating whether the respondent completed primary education. Household enterprise characteristics include the age of the household enterprise owner when they started the household enterprise and a dummy indicating if the business owner is a petty trader. When calculating the age of the household enterprise owner when they started the household enterprise we assume that individuals below 16 years of age inherited the business and exclude them from the analysis. Column 1 (2) {3} run regressions using the full (male only) {female only} sample. 28 Table 5: Job Satisfaction of Household Enterprise Owners and Non-Household Enterprise Owners Respondent is satisfied with job All Men Women (1) (2) (3) (4) (5) (6) Household Enterprise Owner? Yes 0.205∗∗∗ 0.200∗∗∗ 0.178∗∗∗ 0.178∗∗∗ 0.255∗∗∗ 0.251∗∗∗ (0.032) (0.032) (0.037) (0.037) (0.065) (0.065) Personal Characteristics Male -0.018 (0.019) Age 0.001 0.000 0.001 (0.001) (0.001) (0.001) Primary education 0.007 0.009 0.015 (0.022) (0.030) (0.033) Mean of outcome 0.834 0.829 0.837 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 1745 1744 890 890 854 854 Notes: The table displays OLS regressions of a dummy variable that equals to one if the respondent is satisfied with their current job. The right-hand side variables include a dummy variable indicating whether the respondent is a household enterprise owner, the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent completed primary education. All regressions include city fixed effects and report robust standard errors. Columns 1—3 (4—6) {7—9} run regressions using the full (male only) {female only} sample. 29 Table 6: Ideal Job of Household Enterprise Owners and Non-Household Enterprise Owners Respondent sees current job as ideal job All Men Women (1) (2) (3) (4) (5) (6) Household Enterprise Owner? Yes 0.400∗∗∗ 0.400∗∗∗ 0.366∗∗∗ 0.368∗∗∗ 0.503∗∗∗ 0.480∗∗∗ (0.033) (0.034) (0.041) (0.041) (0.058) (0.059) Personal Characteristics Male 0.015 (0.023) Age 0.005∗∗∗ 0.004∗∗∗ 0.005∗∗∗ (0.001) (0.001) (0.001) Primary education -0.004 0.004 -0.006 (0.027) (0.038) (0.039) Mean of outcome 0.695 0.712 0.679 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 1738 1737 886 886 851 851 Notes: The table displays OLS regressions of a dummy variable that equals to one if the respondent indicates that their current job is their ideal job. The right-hand side variables include a dummy variable indicating whether the respondent is a household enterprise owner, the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent completed primary education. All regressions include city fixed effects and report robust standard errors. Columns 1—3 (4—6) {7—9} run regressions using the full (male only) {female only} sample. 30 Table 7: Predictors of Business Aspirations Number of “expand the business” aspirations (0-4) All Men Women (1) (2) (3) (4) (5) (6) (7) (8) (9) Job Satisfaction Satisfied with job 0.371∗∗∗ 0.362∗∗∗ 0.391∗∗∗ 0.373∗∗∗ 0.364∗∗∗ 0.344∗∗∗ (0.061) (0.072) (0.088) (0.108) (0.086) (0.096) Outside Option Accepts equiv. salaried outside offer -0.223∗∗∗ -0.153∗∗∗ -0.283∗∗∗ -0.190∗∗ -0.158∗∗ -0.105 (0.048) (0.057) (0.068) (0.083) (0.067) (0.079) Personal Characteristics Male 0.023 0.044 (0.062) (0.060) Age -0.001 -0.002 -0.004 -0.006 0.001 0.000 (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) Primary education 0.126∗ 0.105 0.068 0.039 0.166∗ 0.152∗ (0.068) (0.067) (0.108) (0.108) (0.087) (0.086) Business Characteristics Age of household enterprise -0.007∗∗ -0.007∗∗ -0.011∗∗ -0.010∗∗ -0.002 -0.003 (0.004) (0.003) (0.005) (0.005) (0.005) (0.005) Petty trader 0.156∗∗∗ 0.167∗∗∗ 0.159∗ 0.160∗ 0.143 0.159∗ (0.059) (0.059) (0.084) (0.083) (0.087) (0.088) Revenue above 50USD per month 0.328∗∗∗ 0.290∗∗∗ 0.403∗∗∗ 0.361∗∗∗ 0.286∗∗∗ 0.256∗∗∗ (0.064) (0.063) (0.098) (0.096) (0.086) (0.086) Number of non-family employees 0.071∗∗∗ 0.063∗∗∗ 0.074∗∗∗ 0.064∗∗∗ 0.103 0.085 (0.024) (0.023) (0.024) (0.024) (0.106) (0.105) Affected by Covid 0.080 0.072 0.115 0.104 0.054 0.043 (0.056) (0.055) (0.083) (0.081) (0.075) (0.075) Mean of outcome 1.359 1.421 1.305 City Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Yes Yes Yes Observations 1525 1148 1148 727 510 510 797 638 638 Notes: The table displays OLS regressions of the number of “expand the business” aspirations respon- dents hold, as defined in Figure 4. The right-hand side variables include a dummy indicating whether the respondent is satisfied with their job, another dummy variable indicating if the respondent would accept an equivalently salaried wage job, and personal and business controls. Personal characteristics include the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent com- pleted primary education. Household enterprise characteristics include the age of the household enterprise in years, a dummy indicating if the business owner is a petty trader, a dummy indicating the registration status of the business, a dummy indicating if the business generates more than 50USD in revenue per month, the number of non-family employees, and a dummy indicating Covid exposure. All regressions include city fixed effects and report robust standard errors. Columns 1—3 (4—6) {7—9} run regressions using the full (male only) {female only} sample. 31 Table 8: Predictors of Non-Family Employees, Full Sample Number of non-family employees (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.373∗∗∗ 0.232∗∗∗ (0.080) (0.077) Job Satisfaction Satisfied with job 0.179∗∗ 0.117 (0.074) (0.090) Ideal Job Current job is ideal job 0.219∗∗∗ 0.169∗ (0.070) (0.090) Aspirations No of business aspirations 0.150∗∗∗ 0.063 (0.042) (0.049) Personal Characteristics Male 0.434∗∗∗ 0.498∗∗∗ (0.074) (0.090) Age -0.006∗ -0.008∗ (0.003) (0.004) Primary education 0.217∗∗ 0.165 (0.090) (0.111) Business Characteristics Age of household enterprise 0.012∗∗ 0.010∗ (0.005) (0.006) Petty trader -0.474∗∗∗ -0.506∗∗∗ (0.079) (0.095) Business registered 0.483∗∗∗ 0.517∗∗∗ (0.145) (0.173) Revenue above 50USD per month 0.159∗∗∗ 0.124 (0.061) (0.077) Affected by Covid -0.007 -0.007 (0.074) (0.089) Mean of outcome 0.482 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 1200 1526 1522 1525 1148 901 Notes: The table displays OLS regressions of the number of non-family employees on the respondents motivation for starting the household enterprise (see Table 3), the respondent’s job satisfaction and ideal job (see Figure 1), the number of “expand the business” aspirations (see Figure 4), as well as a range of personal and business controls. Personal characteristics include the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent completed primary education. Household enterprise characteristics include the age of the household enterprise in years, a dummy indicating if the business owner is a petty trader, a dummy indicating the registration status of the business, a dummy indicating if the business generates more than 50USD in revenue per month, the number of non-family employees, and a dummy indicating Covid exposure. All regressions include city fixed effects and report robust standard errors. 32 Table 9: Predictors of Non-Family Employees, Men Only Sample Number of non-family employees (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.487∗∗∗ 0.323∗ (0.150) (0.166) Job Satisfaction Satisfied with job 0.341∗∗ 0.131 (0.140) (0.178) Ideal Job Current job is ideal job 0.419∗∗∗ 0.238 (0.137) (0.181) Aspirations No of business aspirations 0.278∗∗∗ 0.149 (0.077) (0.099) Personal Characteristics Age -0.010 -0.010 (0.008) (0.009) Primary education 0.373∗ 0.315 (0.202) (0.249) Business Characteristics Age of household enterprise 0.017∗ 0.012 (0.009) (0.010) Petty trader -0.727∗∗∗ -0.739∗∗∗ (0.142) (0.163) Business registered 0.670∗∗∗ 0.719∗∗∗ (0.214) (0.246) Revenue above 50USD per month 0.305∗∗ 0.164 (0.140) (0.167) Affected by Covid -0.050 -0.015 (0.157) (0.181) Mean of outcome 0.902 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 586 727 726 727 510 412 Notes: The table reproduces Table 8, but runs the regressions for men only. More details on the content of the table can be found in Table 8. 33 Table 10: Predictors of Non-Family Employees, Women Only Sample Number of non-family employees (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.013 0.041 (0.044) (0.034) Job Satisfaction Satisfied with job 0.062∗∗∗ 0.042 (0.024) (0.032) Ideal Job Current job is ideal job 0.007 0.054 (0.040) (0.042) Aspirations No of business aspirations 0.003 -0.001 (0.019) (0.026) Personal Characteristics Age -0.003∗ -0.005∗∗ (0.002) (0.002) Primary education 0.047 0.012 (0.043) (0.049) Business Characteristics Age of household enterprise 0.004 0.002 (0.002) (0.003) Petty trader -0.146∗∗ -0.146∗∗ (0.059) (0.072) Business registered 0.119 0.154 (0.086) (0.110) Revenue above 50USD per month 0.079∗ 0.092∗∗ (0.040) (0.047) Affected by Covid 0.019 0.004 (0.033) (0.039) Mean of outcome 0.100 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 613 798 795 797 638 489 Notes: The table reproduces Table 8, but runs the regressions for women only. More details on the content of the table can be found in Table 8. 34 Table 11: Predictors of Registration Status of Household Enterprise, Full Sample Household enterprise is registered (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.059∗∗ 0.003 (0.023) (0.025) Job Satisfaction Satisfied with job 0.044∗ 0.027 (0.026) (0.034) Ideal Job Current job is ideal job 0.016 -0.002 (0.022) (0.028) Aspirations No of business aspirations 0.027∗∗ 0.000 (0.012) (0.014) Personal Characteristics Male 0.100∗∗∗ 0.110∗∗∗ (0.025) (0.028) Age 0.003∗∗∗ 0.004∗∗∗ (0.001) (0.001) Primary education 0.096∗∗∗ 0.115∗∗∗ (0.030) (0.034) Business Characteristics Age of household enterprise 0.001 0.000 (0.001) (0.002) Petty trader -0.086∗∗∗ -0.083∗∗∗ (0.025) (0.029) Number of non-family employees 0.041∗∗∗ 0.039∗∗∗ (0.011) (0.012) Revenue above 50USD per month 0.052∗∗ 0.047∗ (0.023) (0.027) Affected by Covid 0.049∗∗ 0.048∗∗ (0.021) (0.024) Mean of outcome 0.174 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 1200 1526 1522 1525 1148 901 Notes: The table copies Table 8, with the only difference that the outcome variable here is a dummy indicating if the business is formally registered. For more details, see Table 8. 35 Table 12: Predictors of Registration Status of Household Enterprise, Men Only Sample Household enterprise is registered (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.056 0.030 (0.041) (0.049) Job Satisfaction Satisfied with job 0.113∗∗∗ 0.052 (0.042) (0.061) Ideal Job Current job is ideal job 0.049 0.019 (0.037) (0.053) Aspirations No of business aspirations 0.029 -0.021 (0.019) (0.024) Personal Characteristics Age 0.007∗∗∗ 0.007∗∗∗ (0.002) (0.002) Primary education 0.130∗∗ 0.143∗∗ (0.052) (0.060) Business Characteristics Age of household enterprise 0.000 -0.000 (0.002) (0.003) Petty trader -0.130∗∗∗ -0.111∗∗ (0.040) (0.044) Number of non-family employees 0.040∗∗∗ 0.039∗∗∗ (0.011) (0.012) Revenue above 50USD per month 0.094∗∗ 0.117∗∗ (0.046) (0.050) Affected by Covid 0.068∗ 0.068 (0.039) (0.044) Mean of outcome 0.270 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 586 727 726 727 510 412 Notes: The table reproduces Table 11, but runs the regressions for men only. More details on the content of the table can be found in Table 11. 36 Table 13: Predictors of Registration Status of Household Enterprise, Women Only Sample Household enterprise is registered (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.023 -0.008 (0.024) (0.027) Job Satisfaction Satisfied with job -0.009 0.002 (0.031) (0.036) Ideal Job Current job is ideal job -0.008 -0.017 (0.023) (0.028) Aspirations No of business aspirations 0.013 0.018 (0.013) (0.017) Personal Characteristics Age 0.000 0.001 (0.001) (0.001) Primary education 0.055 0.073∗ (0.035) (0.040) Business Characteristics Age of household enterprise 0.002 0.002 (0.002) (0.002) Petty trader -0.062∗ -0.071∗ (0.033) (0.037) Number of non-family employees 0.057 0.066 (0.041) (0.045) Revenue above 50USD per month 0.018 -0.013 (0.024) (0.031) Affected by Covid 0.027 0.029 (0.023) (0.026) Mean of outcome 0.086 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 613 798 795 797 638 489 Notes: The table reproduces Table 11, but runs the regressions for women only. More details on the content of the table can be found in Table 11. 37 Table 14: Predictors of Willingness to Accept Equivalently Salaried Waged Job Owner would accept equivalently salaried wage job All Men Women (1) (2) (3) (4) (5) (6) (7) (8) (9) Job Satisfaction Satisfied with job -0.136∗∗∗ -0.073∗∗ -0.124∗∗∗ -0.097∗ -0.142∗∗∗ -0.059 (0.031) (0.035) (0.044) (0.052) (0.043) (0.048) Ideal Job Current job is ideal job -0.266∗∗∗ -0.218∗∗∗ -0.267∗∗∗ -0.217∗∗∗ -0.264∗∗∗ -0.214∗∗∗ (0.023) (0.028) (0.035) (0.044) (0.032) (0.037) Personal Characteristics Male 0.010 (0.032) Age -0.003∗∗ -0.002 -0.003∗ (0.001) (0.002) (0.002) Primary education -0.073∗∗ -0.050 -0.098∗∗ (0.037) (0.056) (0.049) Business Characteristics Age of household enterprise -0.001 0.000 -0.004 (0.002) (0.003) (0.003) Petty trader 0.059∗ 0.016 0.107∗∗ (0.030) (0.044) (0.044) Revenue above 50USD per month -0.029 0.025 -0.070 (0.033) (0.054) (0.043) Affected by Covid -0.006 -0.077∗ 0.052 (0.028) (0.042) (0.037) Mean of outcome 0.653 0.640 0.666 City Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Yes Yes Yes Observations 1525 1521 1145 727 726 509 797 794 636 Notes: The table displays OLS regressions of a dummy variable that equals to one if the household enterprise owner would accept an equivalently salaried waged job, and zero otherwise. The right-hand side variables include a dummy variable indicating whether the household enterprise owner is satisfied or very satisfied with their job, a dummy variable indicating whether the household enterprise owner believes their current job to be their ideal job, as well as a range of personal and business characteristics. Personal characteristics include the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent completed primary education. Household enterprise characteristics include the age of the household enterprise in years, a dummy indicating if the business owner is a petty trader, a dummy indicating the registration status of the business, a dummy indicating if the business generates more than 50USD in revenue per month, the number of non-family employees, and a dummy indicating Covid exposure. All regressions include city fixed effects and report robust standard errors. Columns 1—3 (4—6) {7—9} run regressions using the full (male only) {female only} sample. 38 Figure 1: Job Satisfaction and Ideal Job (a) Job Satisfaction, HHE Owners (b) Ideal Job, HHE Owners (c) Job Satisfaction, Non-HHE Owners (d) Ideal Job, Non-HHE Owners Notes: The figure displays respondents’ job satisfaction (Panels (a) and (c)) and ideal job (Panels (b) and (d)) for household enterprise owners (Panels (a) and (b)) and non-household enterprise owners (Panels (c) and (d)). 39 Figure 2: Job Satisfaction and Ideal Job by Gender (a) Job Satisfaction, Male HHE Owners (b) Job Satisfaction, Female HHE Owners (c) Job Satisfaction, Male Non-HHE Owners (d) Job Satisfaction, Female Non-HHE Owners (e) Ideal Job, Male HHE Owners (f) Ideal Job, Female HHE Owners (g) Ideal Job, Male Non-HHE Owners (h) Ideal Job, Female Non-HHE Owners Notes: The figure displays respondents’ job satisfaction (Panels (a), (b), (c), and (d)) and ideal job (Panels (e), (f), (g), and (h)) for household enterprise owners (Panels (a), (b), (e), and (f)) and non-household enterprise owners (Panels (c), (d), (g), and (h)). 40 Figure 3: Do you want your children to become HHE owners? (a) HHE Owners (b) Non-HHE Owners (c) Male HHE Owners (d) Male Non-HHE Owners (e) Female HHE Owners (f) Female Non-HHE Owners Notes: The figure displays respondents’ answer to a question whether they want their children to become household enterprise owners. Panels (a), (c), and (e) displays answers for household enterprise owners and Panels (b), (d), and (f) show answers for non-household enterprise owners. Panels (a) and (b) ((c) and 41 } sample. (d)) {(e) and (f)} use the full (male only) {female only Figure 4: Aspirations (a) Aspirations (b) Grouped Aspirations (c) Distribution of Business Aspirations Notes: The figure displays various aspirations of respondents. First, Panel (a) shows the raw share of respondents whose aspirations is any of the ten possible answers. Second, Panel (b) groups the various aspirations from Panel (a) into five groups. Specifically, “expand the business” contains “hire new workers”, “sell to clients”, “increase profits/sales”, and “improve the quality of goods/services”. “Change activity or sector” contains “change activity (within sector)” and “change sector”. “Register or change location” contains “register the firm” and “change location”. Finally, “No change” contains “no change” and “close the business” contains “close the business”. Lastly, Panel (c) displays the distribution of the “expand the business” aspirations. 42 Figure 5: Actions Taken to Reach Aspirations (a) Sell to New Clients (b) Increase Profits (c) Improve the Quality of Goods and Services Notes: Panels (a), (b), and (c) of the figure display what actions individuals have taken to reach the three most states aspirations: selling to new clients, increasing profits, and improving the quality of goods and services. 43 A Appendix Tables and Figures Table A1: Sample Description Country # Cities N # Women # HHE Owners N/# Cities # Women/# Cities # HHE Owners/# Cities Liberia 1 501 242 400 501 242 400 Niger 3 699 241 557 233 80.333 185.667 Senegal 4 708 474 569 177 118.500 142.250 Notes: The table describes the three samples used in the paper. A1 Table A2: Motivation for Starting Household Enterprise by Country Liberia (1) (2) (3) All Men Women Opted in to start business 0.425 0.474 0.379 (0.495) (0.501) (0.486) Forced to start business 0.273 0.214 0.325 (0.446) (0.411) (0.470) Both 0.302 0.311 0.296 (0.460) (0.464) (0.457) Observations 400 196 203 Niger (1) (2) (3) All Men Women Opted in to start business 0.688 0.692 0.680 (0.464) (0.462) (0.468) Forced to start business 0.169 0.164 0.177 (0.375) (0.371) (0.383) Both 0.144 0.144 0.143 (0.351) (0.352) (0.351) Observations 557 354 203 Senegal (1) (2) (3) All Men Women Opted in to start business 0.439 0.587 0.374 (0.497) (0.494) (0.484) Forced to start business 0.354 0.273 0.389 (0.479) (0.447) (0.488) Both 0.207 0.140 0.237 (0.406) (0.348) (0.426) Observations 560 172 388 Notes: The table reproduces the top panel in Table 3, but splits the shares by the three countries in our sample: Liberia, Niger, and Senegal. More details on the content of the table can be found in Table 3. A2 Table A3: Predictors of Motivation for Starting Household Enterprise by Country Owner opted in to start business Liberia Niger Senegal (1) (2) (3) Personal Characteristics Male 0.178∗∗∗ 0.008 0.127∗∗ (0.064) (0.041) (0.063) Primary education -0.002 0.083∗∗ 0.091 (0.072) (0.042) (0.066) Business Characteristics Age when HHE owner started HHE 0.005∗ -0.003 -0.002 (0.003) (0.002) (0.002) Petty trader 0.004 -0.006 -0.078 (0.061) (0.040) (0.061) Mean of outcome 0.609 0.803 0.554 City Fixed Effects Yes Yes Yes Robust Standard Errors Yes Yes Yes Observations 272 429 427 Notes: The table reproduces Table 4, but runs the regressions separately for each country in our sample. More details on the content of the table can be found in Table 4. A3 Table A4: Job Satisfaction of HHE Owners and Non-HHE Owners by Country Respondent is satisfied with job Liberia Niger Senegal (1) (2) (3) (4) (5) (6) Household Enterprise Owner? Yes 0.292∗∗∗ 0.287∗∗∗ 0.080∗∗ 0.084∗∗ 0.293∗∗∗ 0.300∗∗∗ (0.076) (0.077) (0.040) (0.040) (0.056) (0.055) Personal Characteristics Male 0.010 (0.049) Age 0.002 0.001 -0.001 (0.002) (0.001) (0.001) Primary education -0.078 0.038 0.051 (0.054) (0.026) (0.035) Mean of outcome 0.633 0.917 0.893 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 444 443 649 649 652 652 Notes: The table reproduces Table 5, but runs the regressions separately for the three countries in our sample. More details on the content of the table can be found in Table 5. A4 Table A5: Ideal Job of HHE Owners and Non-HHE Owners by Country Respondent sees current job as ideal job Liberia Niger Senegal (1) (2) (3) (4) (5) (6) Household Enterprise Owner? Yes 0.267∗∗∗ 0.253∗∗∗ 0.345∗∗∗ 0.353∗∗∗ 0.529∗∗∗ 0.529∗∗∗ (0.074) (0.074) (0.055) (0.055) (0.047) (0.048) Personal Characteristics Male 0.003 (0.051) Age 0.007∗∗∗ 0.005∗∗∗ 0.002 (0.002) (0.001) (0.002) Primary education -0.101∗ 0.035 0.060 (0.056) (0.041) (0.045) Mean of outcome 0.535 0.788 0.715 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 437 436 649 649 652 652 Notes: The table reproduces Table 6, but runs the regressions separately for the three countries in our sample. More details on the content of the table can be found in Table 6. A5 Table A6: Predictors of Business Aspirations by Country Number of “expand the business” aspirations (0-4) Liberia Niger Senegal (1) (2) (3) (4) (5) (6) (7) (8) (9) Job Satisfaction Satisfied with job 0.116 0.152 0.488∗∗∗ 0.422∗∗∗ 0.700∗∗∗ 0.660∗∗∗ (0.080) (0.095) (0.119) (0.149) (0.128) (0.148) Outside Option Accepts equiv. salaried outside offer -0.085 -0.117 -0.325∗∗∗ -0.118 -0.190∗∗ -0.177∗ (0.080) (0.101) (0.075) (0.092) (0.090) (0.102) Personal Characteristics Male 0.020 0.031 0.036 0.038 0.081 0.110 (0.098) (0.098) (0.096) (0.094) (0.130) (0.124) Age 0.003 0.003 -0.002 -0.004 -0.003 -0.004 (0.004) (0.005) (0.004) (0.004) (0.005) (0.005) Primary education 0.091 0.085 0.095 0.066 0.175 0.122 (0.103) (0.105) (0.111) (0.109) (0.147) (0.149) Business Characteristics Age of household enterprise -0.011 -0.012 -0.010 -0.007 -0.005 -0.006 (0.007) (0.007) (0.006) (0.006) (0.005) (0.005) Petty trader 0.076 0.097 0.054 0.077 0.370∗∗∗ 0.346∗∗∗ (0.109) (0.110) (0.090) (0.091) (0.116) (0.113) Revenue above 50USD per month -0.009 -0.027 0.357∗∗∗ 0.339∗∗∗ 0.480∗∗∗ 0.381∗∗∗ (0.111) (0.116) (0.086) (0.086) (0.130) (0.128) Number of non-family employees 0.093∗∗ 0.089∗∗ 0.082∗∗ 0.074∗∗ 0.055 0.043 (0.045) (0.045) (0.036) (0.036) (0.040) (0.039) Affected by Covid 0.276∗∗ 0.273∗∗ 0.070 0.073 -0.040 -0.068 (0.107) (0.106) (0.085) (0.084) (0.098) (0.096) Mean of outcome 0.945 1.552 1.462 City Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Yes Yes Yes Observations 400 310 310 556 367 367 569 471 471 Notes: The table reproduces Table 7, but runs the regressions separately for the three countries in our sample. More details on the content of the table can be found in Table 7. A6 Table A7: Business Aspirations, Constraints, and Shocks Number of “expand the business” aspirations (0-4) All Men Women (1) (2) (3) (4) (5) (6) (7) (8) Constraints Number of business constraints (0-18) 0.006 0.002 0.004 0.000 (0.008) (0.010) (0.013) (0.015) Number of family constraints (0-2) -0.125∗∗∗ -0.130∗∗∗ -0.165∗∗∗ -0.103∗ (0.036) (0.042) (0.064) (0.059) Shocks Number of shocks to business (0-8) -0.067∗∗∗ -0.071∗∗∗ -0.046 -0.095∗∗∗ (0.018) (0.021) (0.031) (0.030) Number of shocks to family (0-4) -0.003 -0.007 -0.023 0.003 (0.029) (0.033) (0.048) (0.045) Exposed to a disaster 0.113∗∗ 0.028 -0.065 0.085 (0.057) (0.066) (0.110) (0.087) Personal Characteristics Male 0.000 0.021 (0.062) (0.062) Age -0.001 -0.001 -0.005 -0.005 0.002 0.000 (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) Primary education 0.122∗ 0.136∗∗ 0.071 0.075 0.158∗ 0.172∗∗ (0.068) (0.067) (0.110) (0.107) (0.087) (0.087) Business Characteristics Age of household enterprise -0.007∗∗ -0.007∗ -0.010∗∗ -0.010∗∗ -0.003 -0.002 (0.004) (0.003) (0.005) (0.005) (0.005) (0.005) Petty trader 0.148∗∗ 0.145∗∗ 0.154∗ 0.146∗ 0.136 0.124 (0.059) (0.059) (0.083) (0.084) (0.087) (0.087) Revenue above 50USD per month 0.341∗∗∗ 0.344∗∗∗ 0.415∗∗∗ 0.431∗∗∗ 0.298∗∗∗ 0.297∗∗∗ (0.065) (0.064) (0.098) (0.098) (0.086) (0.086) Number of non-family employees 0.068∗∗∗ 0.073∗∗∗ 0.071∗∗∗ 0.075∗∗∗ 0.092 0.103 (0.024) (0.023) (0.025) (0.024) (0.104) (0.102) Affected by Covid 0.092 0.111∗∗ 0.127 0.148∗ 0.065 0.082 (0.056) (0.056) (0.083) (0.083) (0.076) (0.076) Mean of outcome 1.359 1.421 1.305 City Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Yes Yes Observations 1525 1525 1148 1148 510 510 638 638 Notes: The table displays OLS regressions of the number of “expand the business” aspirations respondents hold, as defined in Figure 4. The right-hand side are variables counting the number of constraints or shocks the household was exposed to, personal controls, and business controls. Personal characteristics include the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent completed primary education. Household enterprise characteristics include the age of the household enterprise in years, a dummy indicating if the business owner is a petty trader, a dummy indicating the registration status of the business, a dummy indicating if the business generates more than 50USD in revenue per month, the number of non-family employees, and a dummy indicating Covid exposure. All regressions include city fixed effects and report robust standard errors. Columns 1—4 (5—6) {7—8} run regressions using the full (male only) {female only} sample. A7 Table A8: Actions Taken to Reach Aspirations, Constraints, and Shocks Has done nothing to reach aspirations All Men Women (1) (2) (3) (4) (5) (6) (7) (8) Constraints Number of business constraints (0-18) -0.005 0.007 -0.004 0.023∗∗∗ (0.005) (0.006) (0.009) (0.009) Number of family constraints (0-2) 0.076∗∗∗ 0.059∗∗ 0.033 0.066∗ (0.022) (0.027) (0.043) (0.035) Shocks Number of shocks to business (0-8) 0.002 0.009 -0.022 0.043∗∗ (0.011) (0.013) (0.019) (0.017) Number of shocks to family (0-4) -0.011 0.007 0.034 -0.014 (0.017) (0.020) (0.030) (0.026) Exposed to a disaster 0.029 0.067 0.084 0.060 (0.033) (0.041) (0.066) (0.054) Personal Characteristics Male 0.002 -0.003 (0.038) (0.038) Age -0.002 -0.002 -0.002 -0.003 -0.002 -0.002 (0.002) (0.002) (0.003) (0.003) (0.002) (0.002) Primary education -0.046 -0.043 -0.049 -0.043 -0.034 -0.047 (0.040) (0.041) (0.063) (0.062) (0.052) (0.053) Business Characteristics Age of household enterprise -0.001 -0.001 -0.001 -0.001 -0.000 -0.001 (0.002) (0.002) (0.003) (0.003) (0.003) (0.003) Petty trader 0.061∗ 0.057 0.010 0.021 0.116∗∗ 0.101∗ (0.037) (0.037) (0.053) (0.053) (0.054) (0.054) Revenue above 50USD per month -0.092∗∗ -0.090∗∗ -0.041 -0.053 -0.115∗∗ -0.109∗∗ (0.041) (0.042) (0.069) (0.069) (0.051) (0.053) Number of non-family employees -0.012 -0.011 -0.008 -0.008 -0.140∗∗∗ -0.143∗∗ (0.014) (0.014) (0.014) (0.014) (0.052) (0.055) Affected by Covid -0.122∗∗∗ -0.127∗∗∗ -0.145∗∗∗ -0.150∗∗∗ -0.089∗∗ -0.097∗∗ (0.032) (0.033) (0.051) (0.051) (0.042) (0.044) Mean of outcome 0.637 0.590 0.682 City Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Yes Yes Observations 1217 1217 893 893 407 407 486 486 Notes: The table displays OLS regressions of a dummy variable indicating if the respondent has done “nothing” to reach their stated aspirations on variables counting the number of constraints or shocks the household was exposed to, personal controls, and business controls. Personal characteristics include the gender and the age in years of the respondent, as well as a dummy indicating whether the respondent com- pleted primary education. Household enterprise characteristics include the age of the household enterprise in years, a dummy indicating if the business owner is a petty trader, a dummy indicating the registration status of the business, a dummy indicating if the business generates more than 50USD in revenue per month, the number of non-family employees, and a dummy indicating Covid exposure. All regressions include city fixed effects and report robust standard errors. Columns 1—4 (5—6) {7—8} run regressions using the full (male only) {female only} sample. A8 Table A9: Predictors of Non-Family Employees for Liberia Number of non-family employees (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.219 0.134 (0.145) (0.149) Job Satisfaction Satisfied with job 0.172 0.146 (0.113) (0.157) Ideal Job Current job is ideal job 0.255∗∗ 0.229 (0.117) (0.144) Aspirations No of business aspirations 0.218∗∗ 0.075 (0.090) (0.119) Personal Characteristics Male 0.382∗∗∗ 0.379∗∗∗ (0.093) (0.124) Age 0.003 -0.001 (0.006) (0.007) Primary education 0.274∗∗∗ 0.258∗∗ (0.095) (0.121) Business Characteristics Age of household enterprise 0.013 0.005 (0.010) (0.009) Petty trader -0.564∗∗∗ -0.510∗∗∗ (0.144) (0.179) Business registered 0.681∗∗∗ 0.828∗∗∗ (0.190) (0.263) Revenue above 50USD per month 0.257∗∗ 0.316∗∗ (0.105) (0.132) Affected by Covid -0.063 0.015 (0.158) (0.200) Mean of outcome 0.415 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 279 400 396 400 310 217 Notes: The table reproduces Table 8, but runs the regressions for Liberia only. More details on the content of the table can be found in Table 8. A9 Table A10: Predictors of Non-Family Employees for Niger Number of non-family employees (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.276∗ 0.049 (0.151) (0.150) Job Satisfaction Satisfied with job 0.355∗∗∗ 0.093 (0.118) (0.143) Ideal Job Current job is ideal job 0.284∗∗ 0.203 (0.140) (0.162) Aspirations No of business aspirations 0.271∗∗∗ 0.130∗ (0.088) (0.078) Personal Characteristics Male 0.159 0.190 (0.115) (0.122) Age -0.010∗∗ -0.012∗∗ (0.004) (0.005) Primary education 0.359∗∗ 0.284 (0.176) (0.195) Business Characteristics Age of household enterprise 0.018∗∗ 0.016∗ (0.009) (0.009) Petty trader -0.283∗ -0.227 (0.155) (0.162) Business registered 0.620∗ 0.649∗ (0.316) (0.330) Revenue above 50USD per month 0.278∗∗ 0.204∗ (0.110) (0.121) Affected by Covid -0.089 -0.098 (0.136) (0.145) Mean of outcome 0.654 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 477 557 557 556 367 317 Notes: The table reproduces Table 8, but runs the regressions for Niger only. More details on the content of the table can be found in Table 8. A10 Table A11: Predictors of Non-Family Employees for Senegal Number of non-family employees (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.533∗∗∗ 0.304∗∗∗ (0.124) (0.108) Job Satisfaction Satisfied with job 0.054 0.158 (0.146) (0.147) Ideal Job Current job is ideal job 0.133 0.088 (0.111) (0.170) Aspirations No of business aspirations 0.043 0.036 (0.051) (0.076) Personal Characteristics Male 0.726∗∗∗ 0.813∗∗∗ (0.164) (0.189) Age -0.011∗ -0.012 (0.006) (0.008) Primary education -0.001 -0.107 (0.187) (0.229) Business Characteristics Age of household enterprise 0.010 0.008 (0.007) (0.008) Petty trader -0.513∗∗∗ -0.708∗∗∗ (0.112) (0.151) Business registered 0.163 0.170 (0.255) (0.294) Revenue above 50USD per month -0.005 -0.056 (0.084) (0.109) Affected by Covid 0.101 0.069 (0.110) (0.137) Mean of outcome 0.362 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 444 569 569 569 471 367 Notes: The table reproduces Table 8, but runs the regressions for Senegal only. More details on the content of the table can be found in Table 8. A11 Table A12: Predictors of Registration Status of Household Enterprise for Liberia Household enterprise is registered (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.131∗∗ 0.030 (0.052) (0.055) Job Satisfaction Satisfied with job 0.011 -0.016 (0.046) (0.058) Ideal Job Current job is ideal job 0.017 0.037 (0.045) (0.057) Aspirations No of business aspirations 0.091∗∗∗ 0.042 (0.031) (0.040) Personal Characteristics Male 0.183∗∗∗ 0.185∗∗∗ (0.050) (0.064) Age 0.006∗∗ 0.006∗∗ (0.002) (0.003) Primary education 0.040 0.075 (0.054) (0.066) Business Characteristics Age of household enterprise 0.003 0.004 (0.004) (0.005) Petty trader -0.055 -0.046 (0.051) (0.058) Number of non-family employees 0.087∗∗∗ 0.084∗∗∗ (0.017) (0.018) Revenue above 50USD per month 0.083 0.032 (0.057) (0.078) Affected by Covid 0.096∗ 0.075 (0.052) (0.062) Mean of outcome 0.273 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 279 400 396 400 310 217 Notes: The table reproduces Table 11, but runs the regressions for Liberia only. More details on the content of the table can be found in Table 11. A12 Table A13: Predictors of Registration Status of Household Enterprise for Niger Household enterprise is registered (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.059 0.025 (0.038) (0.049) Job Satisfaction Satisfied with job 0.093∗∗ 0.058 (0.039) (0.061) Ideal Job Current job is ideal job 0.026 -0.061 (0.036) (0.051) Aspirations No of business aspirations 0.030 -0.024 (0.021) (0.027) Personal Characteristics Male 0.108∗∗∗ 0.120∗∗∗ (0.032) (0.035) Age 0.002 0.004∗ (0.002) (0.002) Primary education 0.069 0.084∗ (0.044) (0.049) Business Characteristics Age of household enterprise 0.001 -0.000 (0.003) (0.003) Petty trader -0.093∗∗ -0.093∗∗ (0.038) (0.041) Number of non-family employees 0.044∗∗ 0.047∗∗ (0.018) (0.018) Revenue above 50USD per month 0.048 0.075∗∗ (0.033) (0.037) Affected by Covid 0.040 0.044 (0.035) (0.038) Mean of outcome 0.151 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 477 557 557 556 367 317 Notes: The table reproduces Table 11, but runs the regressions for Niger only. More details on the content of the table can be found in Table 11. A13 Table A14: Predictors of Registration Status of Household Enterprise for Senegal Household enterprise is registered (1) (2) (3) (4) (5) (6) Motivation for Starting Business Owner opted in to start business 0.015 -0.023 (0.032) (0.034) Job Satisfaction Satisfied with job 0.063∗ 0.071 (0.036) (0.044) Ideal Job Current job is ideal job 0.008 0.006 (0.031) (0.041) Aspirations No of business aspirations 0.001 -0.004 (0.014) (0.017) Personal Characteristics Male 0.013 0.029 (0.049) (0.058) Age 0.001 0.002 (0.001) (0.002) Primary education 0.158∗∗∗ 0.162∗∗ (0.059) (0.069) Business Characteristics Age of household enterprise 0.001 0.001 (0.002) (0.002) Petty trader -0.120∗∗∗ -0.125∗∗ (0.045) (0.052) Number of non-family employees 0.011 0.011 (0.017) (0.018) Revenue above 50USD per month 0.025 0.006 (0.037) (0.047) Affected by Covid 0.018 0.023 (0.029) (0.034) Mean of outcome 0.127 City Fixed Effects Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Observations 444 569 569 569 471 367 Notes: The table reproduces Table 11, but runs the regressions for Senegal only. More details on the content of the table can be found in Table 11. A14 Table A15: Predictors of Willingness to Accept Equivalently Salaried Waged Job by Country Owner would accept equivalently salaried wage job Liberia Niger Senegal (1) (2) (3) (4) (5) (6) (7) (8) (9) Job Satisfaction Satisfied with job -0.018 -0.024 -0.356∗∗∗ -0.189∗∗∗ -0.167∗∗∗ -0.108∗ (0.050) (0.057) (0.031) (0.047) (0.054) (0.060) Ideal Job Current job is ideal job -0.237∗∗∗ -0.210∗∗∗ -0.334∗∗∗ -0.214∗∗∗ -0.236∗∗∗ -0.211∗∗∗ (0.047) (0.055) (0.035) (0.044) (0.039) (0.046) Personal Characteristics Male 0.018 (0.058) Age 0.000 -0.005∗∗ -0.002 (0.003) (0.002) (0.002) Primary education -0.074 -0.075 -0.043 (0.067) (0.058) (0.069) Business Characteristics Age of household enterprise -0.002 -0.002 -0.001 (0.005) (0.003) (0.002) Petty trader 0.150∗∗ -0.031 0.067 (0.059) (0.048) (0.048) Revenue above 50USD per month 0.044 -0.067 -0.022 (0.083) (0.047) (0.057) Affected by Covid -0.012 -0.030 0.029 (0.060) (0.046) (0.043) Mean of outcome 0.635 0.653 0.666 City Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Robust Standard Errors Yes Yes Yes Yes Yes Yes Yes Yes Yes Observations 400 396 307 556 556 367 569 569 471 Notes: The table reproduces Table 14, but runs the regressions separately for the three countries in our sample. More details on the content of the table can be found in Table 14. A15 Figure A1: Job Satisfaction and Ideal Job by Country, Household Enterprise Owners (a) Job Satisfaction, Liberia (b) Job Satisfaction, Niger (c) Job Satisfaction, Senegal (d) Ideal Job, Liberia (e) Ideal Job, Niger (f) Ideal Job, Senegal Notes: The figure reproduces Panels (a) and (b) of Figure 1, but splits the panels by the countries in our sample. A16 Figure A2: Job Satisfaction and Ideal Job by Country, Non-Household Enterprise Owners (a) Job Satisfaction, Liberia (b) Job Satisfaction, Niger (c) Job Satisfaction, Senegal (d) Ideal Job, Liberia (e) Ideal Job, Niger (f) Ideal Job, Senegal Notes: The figure reproduces Panels (c) and (d) of Figure 1, but splits the panels by the countries in our sample. A17 Figure A3: How often is it that... (a) running a business gives you more time flexibility (b) running a business gives you more time flexibility than working for someone else? (HHE) than working for someone else? (Non-HHE) (c) you can make more money running a business than (d) you can make more money running a business working for someone else? (HHE) than working for someone else? (Non-HHE) (e) the working conditions running a business are bet- (f) the working conditions running a business are bet- ter than working for someone else? (HHE) ter than working for someone else? (Non-HHE) (g) running a business gives you time to address fam- (h) running a business gives you time to address fam- ily responsibilities? (HHE) ily responsibilities? (Non-HHE) Notes: The figure displays the raw data to four variables, displayed in the for rows, related to individuals’ job satisfaction for household enterprise owners (Panels (a), (c), (e), and (g)) and non-household enterprise owners (Panels (b), (d), (f), and (h)). A18 Figure A4: Aspirations by Gender and Country (a) Aspirations, Men (b) Aspirations, Women (c) Aspirations, Liberia (d) Aspirations, Niger (e) Aspirations, Senegal Notes: The figure reproduces Figure 4, but splits the panels by gender and the countries in our sample. A19 Figure A5: Constraints and Shocks Faced by Household Enterprise (a) Constraints Faced by Business Owner (b) Shocks Faced by Business Owner Notes: The figure displays the reasons business owners stress as the main challenges facing their business (Panel (a)) as well as the types of shocks business owners were exposed to in the past 12 months. A20 Figure A6: How often is it that working for someone else offers more stability? (a) Household Enterprise Owner (b) Non-Household Enterprise Owner Notes: The figure displays the raw data to a variable related to individuals’ outside option. A21