Policymaking always utilises a broad brush with which to redraw the lives of individuals. However, though broad, with the right evidence this brush can be narrowed by taking account of the heterogeneity of human behaviour. Just consider the many and varied schemes designed to support entrepreneurs. Putting aside the debate over whether or not this is the best use of tax revenues, nobody could deny that if we are to spend money on promoting entrepreneurs we should do so in most efficient way.

In “Entrepreneurship Training, Risk Aversion and Other Personality Traits: Evidence from a Random Experiment”, Robert W. Fairlie and William Holleran from the University of California draw on data from Growing America through Entrepreneurship (Project GATE), the largest randomised control experiment on providing entrepreneurship training ever conducted in the United States. Fairlie and Holleran find that:

[I]ndividuals who are more risk tolerant benefit more from entrepreneurship training than individuals who are less risk tolerant. The estimated interaction effects are large: averaging our estimates across the three waves implies that individuals who have a one standard deviation higher level of risk tolerance experience a 2.9 percentage point larger increase in business ownership and a 3.7 percentage point larger increase in the likelihood of starting a business from receiving the treatment than individuals with the lower level of risk tolerance.

This is a useful insight and suggests that we should consider identifying specific groups that may benefit more or less from government programmes to help people start a business. There can be no sure-fire way for spotting the next Zuckerberg, but we can increase the odds. Interestingly, Fairlie and Holleran also find “no evidence that individuals who are more innovative benefit more from entrepreneurship training than individuals who are less innovative.”

As the paper states: “some of the most disadvantaged groups such as at-risk youth and individuals with a criminal background have high levels of risk tolerance, and thus might benefit more for entrepreneurship training than more traditional job training programs.” There might be something in this: John Timpson has found ex-offenders fit in well with his unique entrepreneurial, bottom-up model for running his high street retailer.

As things stand in the UK, we have a remarkably limited understanding whether the schemes used to support entrepreneurship are doing any good. According to Gov.uk, business owners have 278 schemes to choose from. With proper analysis it might turn out that this is the correct number and they are being targeted at exactly the right group in the most efficient way. But I doubt it.

Philip Salter is director of The Entrepreneurs Network.