Asset Transfers for (Pre-) Entrepreneurs: Evidence from Chile

The original theory of microcredit was that it offered the opportunity for poor households to create profitable microenterprises. But there were always households left behind—those that were too poor to create a microenterprise or plausibly repay even the very small loans on offer.

One attempt to address these households, usually called the “ultra poor,” was to create an asset transfer and training program that would allow them to “graduate” into standard microcredit. BRAC’s Targeting the Ultra Poor program is perhaps the best known of these. Evaluations of TUP-style programs have been mixed – with some showing no effect and others strong effects. It seems that a major factor is local labor markets—when ultra poor households have good wage labor alternatives, asset transfers do not help much. When local labor markets are thin or non-existent, asset transfers can make a big difference.

A new study from Claudia Martínez A., Esteban Puentes, and Jaime Ruiz-Tagle from the University of Chile investigates an asset transfer program that targets a slightly different population, although one with its own struggles in accessing formal financial products, and offers new insight into asset transfer programs and labor markets.

Researchers evaluated the state-run micro-entrepreneurship support program (MESP) in Chile, where the loan coverage rate for micro-entrepreneurship is around 45%. MESP seeks to serve the potentially large demographic of entrepreneurs that may be excluded from business financing. While the study participants did not fit the criteria of the global definition of ultra-poor, they still live on around $3 a day - many were unemployed but some were already microentrepreneurs. The program provides an initial in-kind transfer of around 600USD as well as several months of business training and mentoring visits. Like the TUP programs, the treatment provides a supportive influx of assets and more holistic model than a simple loan. Using an RCT, researchers found that MESP “did significantly increase individuals’ employment and income, by 18% and 32% respectively after one year and significantly improves the business practices of its beneficiaries.”

An interesting finding from this study is that there were larger impacts on self-employment and income on those who were unemployed compared with those who were either wage workers or already entrepreneurs at baseline—a finding that aligns with the evidence from ultra poor programs. Also, the effect on wage earners was only positive for low-income individuals. The study authors explain that the capital transfers and training may help the unemployed with the “fixed cost” of being an entrepreneur, while those who entered the study with their own business in place have already paid this cost. Additionally, there are lower opportunity costs in devoting time and resource to the program for someone who is unemployed.

Some entrepreneurs received a second round of asset transfers four months into the program. While this group did show greater improvements in terms of employment and income than the control, there was no statistical significance between this “MESP+” group and those who completed the standard program. That result supports the theory that one reason that microenterprises don’t grow is that the marginal returns to capital decrease rapidly (albeit from a very high level)—for more on this, check out Chapter 9 in Banerjee and Duflo’s Poor Economics. For governments or NGOs with limited budgets, this apparent flattened production function suggests it is more beneficial to provide small transfers to larger numbers of recipients than larger transfers to fewer recipients.

Martínez et al. also note that the cost of the program is recovered within 24-27 months of increased labor income. However, this assumes that income increases are sustained during this time as the length of evaluation for this study was only one year. It would be interesting to investigate the long-term impact of the MESP program as well as questions around the growth of these enterprises. Did they stay micro or scale up? Were entrepreneurs able to hire more employees, thus creating additional employment growth? While this research shows promising and new information about asset transfer programs, there are still remaining questions around how to sustain and multiply employee opportunities through entrepreneurship development.

Reviews of similar programs that provide capital have found little support for long-term growth or scaling for SMEs. David McKenzie and Christopher Woodruff provide recommendations on improving these types of program evaluations to help guide policymakers since currently there is little data currently to show which elements of business training and financing best assist SMEs.