IT HAS become commonplace to suggest that failure is good for entrepreneurs. In this view, failure that comes early in a founder's career can teach them important lessons about doing business and harden them up for the next start-up attempt.

In the US, a tolerance for entrepreneurs who have gone bust is held up as a vital ingredient in the country's entrepreneurial culture, while Europe's stigma of failed businesspeople is seen as a brake on economic growth.

To address this, legislators in many European countries have sought to make it easier for former business owners to start again.

Everyone seems to accept that entrepreneurs do learn and that there is an economic penalty for preventing this from happening.

But do entrepreneurs actually learn from their mistakes?

The empirical evidence in support of this learning process is flimsy. If learning were pervasive, we would expect statistical evidence that people with a history in business were more likely to start a successful business than novices.

In the UK, the evidence is that novices are neither more nor less likely to have a business that grows or survives than experienced founders. In Germany, where more extensive statistical work has been undertaken, those whose business had failed had worse-performing businesses if they restarted than did novices.

One reason for this is the role of chance in determining whether a business prospers. In spite of volumes of books identifying simple recipes for success, the reality is that starting a business is risky.

The outcome depends heavily on luck - whether parking is suddenly banned outside your shop, or whether you or a member of your family become ill. This unpredictability means it is difficult for entrepreneurs to learn. The best analogy is a lottery: it is not possible to learn to win a lottery.

The only way for an individual to increase their likelihood of a win is for them to buy more tickets. Likewise, the circumstances in which a new business is created and the challenges it faces are almost certainly different from those faced by the owner in their second, third and fourth business.

So why do some people start many businesses? Part of the answer is optimism. The evidence is clear that business owners are more optimistic about their future earning prospects than employees. This optimism is consistently misplaced, but it suggests as a key character trait that such individuals always expect to win the lottery. This helps to explain why the US seems to spawn so many individuals who have done well after successive business failures. To revisit the lottery analogy, the "price" of the lottery tickets in the US is low; people can restart a business with far fewer penalties than in Europe.

Since the costs of failure are lower, more businesses are founded and there are more eventual winners. This does not mean founders have "learnt" to succeed: they are simply more optimistic.

Does it matter whether entrepreneurs learn? If individuals want to start businesses and believe they have learnt from past errors, why not let them get on with it?

Predictably, it is not as simple as that: policymakers have used a belief in the power of entrepreneurial self-improvement to enact legislation that is at best misguided and at worst harmful.

Some commentators believe that the UK's Enterprise Act 2002, which enabled individuals to extract themselves more easily from personal bankruptcy to start a new enterprise, is at the heart of much of the consumer uncertainty associated with the credit crunch.

In the US, research shows that those states with generous bankruptcy laws are the ones where it is most difficult to obtain bank finance to start a new business.

The assumption that entrepreneurs use the lessons of their own experience to improve their chances of creating a series of profitable businesses is not borne out by the evidence. Success in business remains, as in life, a lottery.

David Storey is associate dean at Warwick Business School