Becoming a successful entrepreneur requires the right combination of factors, impeccable timing, and often sheer luck. In her latest research, Wharton management professor Valentina Assenova examines what it takes to help entrepreneurs who need more luck than most — those who live in places where inequality is institutionalized and the obstacles to success are even higher to hurdle.

Inspired by a trip to South Africa, Assenova tracked the improvement rates for small businesses in Soweto whose owners received help through an incubator. She found that entrepreneurs who were paired with mentors learned the kind of expertise that enabled them to grow their businesses significantly.

Do founders learn during early-stage venture incubation? If so, what role does learning play in subsequent performance? To answer these questions, I use data from 2,153 applicants to an incubator in Johannesburg, South Africa, and study the relationship between changes in founders’ learning during incubation and their ventures’ post-incubation revenues and profits.

Further, I examine how founders’ random assignment to coaches in the incubator affected learning and compare changes in revenues, costs, employment, and rates of business registration within incubated and non-incubated ventures using a differences-in-differences design. I find that learning during incubation among founders was associated with profitability and revenue growth post-incubation. Every one standard-deviation increase in learning, as measured by changes in entrepreneurs’ mid-point and final exam scores, was associated with a 109% increase in monthly profits and a 136% increase in monthly revenues during the 11 months after venture incubation.

Further, coach centrality moderated the observed benefits of incubation, with a one unit increase in a coach’s centrality being associated with up to 25% increase in founders’ learning during incubation. These findings suggest that early-stage venture incubation promotes learning and that learning contributes to greater scaling and profitability among incubated ventures.

Existing research at the nexus of institutional theory and entrepreneurship suggests that lowering institutional barriers to forming, growing, and exiting new firms through legal and regulatory reforms can affect the types of startups that entrepreneurs found in a region. These institutional changes could affect entrepreneurs’ willingness to partner with organizational sponsors to develop high-growth startups and potentially enhance these sponsors’ ability to select high-growth startups to fund and develop.

This study evaluates these ideas by developing and testing three hypotheses, that institutional reforms (a) improve entrepreneurs’ perceived benefits of the resources that organizational sponsors provide, (b) increase the number, quality, and diversity of applicants for these resources within sponsors’ local ecosystems, and (c) enhance sponsors’ capacity to select high-growth startups into their cohorts.

To evaluate these hypotheses, I analyze data from 13,770 applicants to venture accelerators over multiple application cycles between 2016-2018 in 170 countries, using a difference-in-differences design that exploits institutional reforms that aimed to reduce the time and procedures to start new firms, obtain credit, and resolve bankruptcy for entrepreneurs.

The findings have important implications for how institutional reforms affect early-stage entrepreneurship and the capacity of organizational sponsors to select high-growth early-stage startups to fund and develop within their local ecosystems.