Rohini Pande and coauthors in VoxDev: How does microenterprise growth impact child outcomes?

By Patrick Agte, Arielle Bernhardt, Erica Field, Rohini Pande, and Natalia Rigol

This article first appeared in VoxDev on October 12, 2023

Many poverty reduction programmes emphasise small enterprise development as a means of generating self-sustaining income growth for the poor. We know less about how microenterprise growth impacts child outcomes, especially human capital investment. Do business growth opportunities for poor households improve their children's educational attainment, thereby disrupting the intergenerational transmission of poverty? In general, we would anticipate that income gains lead parents to increase investments in children’s education. Yet among entrepreneur households, parents must trade-off investments in their children’s human capital with business investment opportunities. Income gains from enterprise growth may also raise demand for child labour, limiting investment in schooling.

In our research (Agte et al. 2023), we evaluate how entrepreneur parents resolve these trade-offs and investigate if the decision to invest in children’s schooling versus a household enterprise varies with the parents’ own level of education. To investigate this question, we consider the long-term impacts of a liquidity shock intervention for poor entrepreneurs, which resulted in short-run microenterprise growth. Our study setting is urban India, where educational attainment has risen dramatically in recent decades. While primary school completion was only 78% among urban adults of the same generation as our study sample parents, it is now nearly universal (National Family Health Survey 2021). Similarly, college completion has increased from 22% to 44% over the past 19 years. At the same time, though, India has one of the lowest rates of relative intergenerational educational mobility in the world (Asher et al. 2022). Will income gains for entrepreneur parents result in more education for all, or only for children of better educated parents?

We revisit microfinance borrowers in the city of Kolkata over a decade after they participated in a field experiment in which they were randomly assigned to either a traditional microfinance contract (control group) or one with a flexible repayment schedule that encouraged business investment (treatment group). For borrowers assigned to the standard contract, repayment began two weeks after loan disbursement, while for borrowers assigned to the flexible contract there was a grace period of eight weeks before repayment began.  Relative to the traditional contract, the flexible repayment contract generated rapid business growth: three years after the intervention, the treatment group had 41% higher business profits and 19% higher household income than the control group (Field et al. 2013). 

The modal household in our study had two children, at least one of whom was of school-going age (7-17) at baseline. To evaluate the impact of experimentally-generated business growth on these children’s outcomes, we conducted an in-depth household survey with study participants 11 years post-intervention. The survey collected educational and socio-economic outcomes for all children of study participants, including those who had left the household. 

Gains in enterprise profits generated by the flexible microcredit contract leads treatment households to significantly increase investments in their children’s education. School-aged children in treatment households are, for instance, more than twice as likely to attend private secondary school and benefited from 21% higher spending on after-school tutoring. Gains in higher education are substantial: children in treatment households are 10 percentage points more likely to attend college, a 37% increase in attendance rate when compared to control group children of the same age. Overall, the increase in education spending accounts for roughly 10% of the treatment-induced increase in household income.

Since our goal is to understand how treatment impacted relative intergenerational mobility, our analysis considers both average effects and differences between children of more- and less-educated parents. This heterogeneity analysis reveals striking differences in investment response to treatment by parental education. Among households in which both parents are literate, treatment increases secondary school completion by 12 percentage points and college attendance by 15 percentage points. Children with at least one illiterate parent, on the other hand, are 14 percentage points less likely than their control counterparts to complete secondary schooling and experience no change in college attendance.

Consistent with an investment trade-off, long-run household business outcomes exhibit the opposite pattern with respect to parents’ literacy. In the short run, both types of households – those with literate parents and those with at least one illiterate parent – report income and profit gains as a result of the flexible credit contract treatment. But, 11 years post-intervention, these economic gains only persist among illiterate treated households, who report a 45% increase in profits and a tripling of enterprise capital compared to control peers. Conversely, among literate parent households, income and business profits converge between treatment and control groups by the time of our long-term follow-up survey. 

In line with the decrease in schooling we observe for children of illiterate parents, we also find that treatment leads to an increase in child labour within these households. Only 2% of children in the control group report having worked in self-employment when under the age of 18, but among children of illiterate parents, treatment leads to a 6 percentage point increase in this activity. This result is consistent with there being complementarities between capital and labour in household enterprises: since illiterate parent households tend to put their short-run income gains towards capital investments in their business, they have a higher need for workers. 

Taken together, our results show illiterate parents are more likely to reinvest income gains from business expansion in their enterprise, whereas literate parents are more likely to invest in their children's education. There are two central explanations why investment patterns differ so substantially by parental education, despite comparable short-run income gains: differences in expected returns to child schooling and differences in credit constraints. We find little evidence of credit constraint differences among clients in our sample, the majority of whom are second-time borrowers with similar repayment behavior and comparable short-run returns to capital. We therefore conclude that differences in expected returns to education between more- and less-educated households are likely the primary driver of divergent household investment responses to microenterprise growth.

There are multiple reasons why expected returns to children’s schooling might differ by parental education: first, it may be that less educated parents are less able to assist their children with schoolwork (Todd and Wolpin 2007, Banerji et al. 2017) or spend less time on child care (Guryan et al. 2008). Less-educated parents may struggle to guide their children through the educational system due to limited exposure to successful pupils in their social circles (Sequeira et al. 2016). Alternatively, it may be that the key difference lies in perception of returns to children’s schooling: multiple empirical studies document that less-educated households underestimate returns to education.  Regardless of whether the difference is real or mis-perceived, the net result is that less educated parents have lower expected returns than their more educated counterparts, which should give rise to lower educational investments.

Our findings demonstrate how a positive shock to household liquidity has long-term consequences for the next generation by raising human capital investment in children. From a measurement standpoint, our findings emphasise the importance of long-term follow-up surveys as well as evaluating intervention impacts using a broad definition of the household. To estimate intergenerational treatment effects, we needed data on all children who had ever been born, not just those who were living at home at the time of our follow-up survey. From a policy standpoint, our findings emphasise the need to look at relative treatment effects among more and less vulnerable populations, and not exclusively at average treatment effects (Deaton and Cartwright 2018). Average educational gains in our sample were accompanied by losses in relative intergenerational educational mobility, highlighting the trade-offs inherent in encouraging microenterprise growth as an anti-poverty strategy. 

Agte, A, A Bernhardt, E Field, R Pande, and N Rigol (2023), "Investing in the Next Generation: The Long-Run Impacts of a Liquidity Shock," NBER WP No. 29816.

Almas, I, A W Cappelen, K G Salvanes, E Ø Sørensen, and B Tungodden (2016), "What Explains the Gender Gap in College Track Dropout? Experimental and Administrative Evidence," American Economic Review, 106(5): 296–302.

Asher, S, P Novosad, and C Rafkin (2022), "Intergenerational Mobility in India: New Methods and Estimates Across Time, Space, and Communities," Conditionally accepted, American Economic Journal: Applied Economics

Attanasio, O P and K M Kaufmann (2014), "Education choices and returns to schooling: Mothers’ and youths’ subjective expectations and their role by gender," Journal of Development Economics, 109: 203–216.

Avitabile, C and R de Hoyos (2018), "The heterogeneous effect of information on student performance: Evidence from a randomized control trial in Mexico," Journal of Development Economics, 135: 318–348.

Banerji, R, J Berry, and M Shotland (2017), "The impact of maternal literacy and participation programs: evidence from a randomized evaluation in India," American Economic Journal: Applied Economics, 9(4): 303–37.

Boneva, T, M Golin, and C Rauh (2021), "Can perceived returns explain enrollment gaps in postgraduate education?" Labour Economics, 101998.

Brown, P H (2006), "Parental education and investment in children’s human capital in rural China," Economic Development and Cultural Change, 54(4): 759–789.

Chakravarty, S and A Agarwal (2021), "Perceived returns to education and its impact on schooling decisions," Working Paper.

Deaton, A and N Cartwright (2018), "Understanding and misunderstanding randomized controlled trials," Social Science & Medicine, 210: 2–2.

Delavande, A and B Zafar (2019), "University choice: The role of expected earnings, nonpecuniary outcomes, and financial constraints," Journal of Political Economy, 127(5): 2343–2393.

Field, E, R Pande, J Papp, and N Rigol (2013), "Does the Classic Microfinance Model Discourage Entrepreneurship Among the Poor? Experimental Evidence from India," American Economic Review, 103(6): 2196–2226.

Jensen, R (2010), "The (perceived) returns to education and the demand for schooling," Quarterly Journal of Economics, 125(2): 515–548.

Guryan, J, E Hurst, and M Kearney (2008), "Parental education and parental time with children," Journal of Economic Perspectives, 22(3): 23–46.

Nguyen, T (2008), "Information, role models and perceived returns to education: Experimental evidence from Madagascar," Unpublished manuscript.

Sequeira, S, J Spinnewijn, and G Xu (2016), "Rewarding schooling success and perceived returns to education: Evidence from India," Journal of Economic Behavior & Organization, 131: 373–392.

Todd, P E and K I Wolpin (2007), "The production of cognitive achievement in children: Home, school, and racial test score gaps," Journal of Human Capital, 1(1): 91–136.