Most domestic firms in developing countries are, and will remain, small, employing only a handful of workers. Can policy-makers identify potential high-growth entrepreneurs with the ability to grow from small to large? And, if so, is direct policy action, such as targeted financial support for these high-growth entrepreneurs, effective? A much discussed World Bank study by McKenzie in 20151 answers both questions affirmatively. But do the findings hold up to scrutiny?
A business plan competition in Nigeria, which provided cash grants for entrepreneurs to start new firms and expand existing firms, was used to investigate these questions. The World Bank study used a randomized controlled trial, allocating an average of US$50,000 (out of a total pool of US$36 million) to each winning entrepreneur. The grants were presented in four tranches, each one contingent on the entrepreneur achieving specified outcomes. Tracking applicants using surveys over three years, the study found large impacts on the rate of start-up firms, survival of existing firms, and employment, sales and profit figures. Firms receiving the cash injection were, for example, more likely to employ 10 or more workers than the control group. The World Bank study concluded that these effects were caused by the capital injection, rather than alternative explanations (such as participation in business training programmes or being a winner of the business plan competition). The widely quoted finding of the study is that the programme enhanced employment creation in a more cost-effective way than large-scale policy efforts, such as those in the United States.
While the use of the randomized controlled trial is an important methodological strength of the World Bank study, other aspects urge a much more cautious view of the extent to which this study can inform evidence-based policy. Three issues stand out: self-reporting effects, spillover effects and the sustainability of results.
The study relies on self-reported results. When respondents are aware that they are part of a study, there is a tendency to overestimate or over-report their results. This is known as the ‘Hawthorne effect’. It is likely that the Hawthorne effect played a role in the World Bank study: the US$50,000 prize was paid in four tranches contingent on achieving the pre-specified outcomes, providing an incentive to over-report results. Indeed, the pattern of the findings indicates positive effects of the cash injections for all types of programme outcomes, regional samples and outcomes, which is highly unlikely in view of the heterogeneity of the firms in the business plan competition.
The study ignores negative spillover effects. A second concern is the spillover effects of firms receiving a cash injection on other firms in the market. The capital injection induces competition and increases labour mobility: firms receiving the cash injection can steal markets, crowd out other firms and even poach labour (hiring the most productive workers from firms not receiving the grant). Therefore, the capital injection creates a negative impact on other competing firms.
Results are unlikely to be sustained. The World Bank study reports that the results of the direct policy action on the treated group are robust and consistent, but it is doubtful that the results can be sustained. The effects of the business plan competition will wane over time. One clear concern is corruption: the 2015 Corruption Perception Index ranks Nigeria 136 out of 167 countries. The study suggests that “… winning the competition could give the firm some protection against government officials asking for bribes or otherwise inhibiting firm productivity, since now the firm is seen as a favoured firm which should not be touched”.2 The presence of corruption implies short-lived impacts on firms receiving the cash injection and Nigeria would be better advised to take direct policy action to tackle corruption instead of providing large capital injections.
ADD YOUR COMMENT
Youth Business International
World Bank Group