Working Paper Series no. 644: Entrepreneurship and Information on Past Failures: A Natural Experiment

Non-technical summary

In this paper, we analyze the causal impact of public information regarding past corporate defaults on entrepreneurs' access to credit and ability to create a new firm. We find that the flagging of past failures causally impacts 1) the restart probability of failed entrepreneurs, 2) the new corporate loans contracted by flagged entrepreneurs, and 3) the interest rate at which they borrow.

To establish the link between information on past failure and entrepreneurship, we exploit a natural experiment from France. Before 2013, Banque de France produced a highly visible record of past corporate bankruptcy for firm managers. Such individual flagging was lasting for exactly three years and was made available to banks via Banque de France's FIBEN database, a widely used scoring system of firms and managers. A 2013 reform suppressed the publication of this flag in the FIBEN data for individuals involved in one bankruptcy only. As a result, more than 140,000 individuals were simultaneously deflagged and informed about it.

Our empirical strategy consists in comparing the dynamics of the entrepreneurs whose flag lasted three-years (i.e., those whose flag removal arrived before the policy change) with the one of the entrepreneurs whose flag lasted less because of the policy change. Once the business cycle and other aggregate characteristics are controlled for, the only difference between those two groups of managers is the duration of the flag. Thus, by looking at what happens when the flag is removed, we can derive what is the effect of the information on past failures on entrepreneurs’ ability to start a business and access to credit.

The following figure plots the dynamics of the probability of starting a business for failed entrepreneurs around the moment they are assigned the flag and the moment their flag is removed. There are three groups of flagged managers: 1) three-year flagged managers, 2) policy-affected managers whose flags last one year, and 3) policy-affected managers whose flags last two years. The first vertical line indicates the assignment of the flag, while the other vertical lines indicate flag removal, which changes depending on the group of managers considered. Comparing the pattern of restart of the policy-affected groups to the one of the pre-reform group (solid line) as a function of the time since flag start, it appears very clearly that the early removal of the flag causes a rise in restarts. In turn, this suggests that as soon as failure information is not immediately visible to bankers, failed entrepreneurs restart more easily.

We also analyze the impact of information on past failures through the prism of entrepreneurs managing multiple companies: After one of the companies fails, the other companies suffer from the entrepreneur's stigma on the credit market. We show that after removal of the failure flag, these companies borrow significantly more. This effect is particularly strong in industries where individual talent matters most.

A key feature of the 2013 reform was that only the failure flag was erased while the information on entrepreneurs’ past failure remained accessible, though at a cost. Thus, an insight of our analysis is that details in the framing of public information strongly matter in shaping the equilibrium. We show that a small friction in accessing information can change drastically how much this information weights in economic agents’ decision. When the flag was highly salient, bankers used to heavily penalize individuals that had experienced failure in the last three years. Once this information is made less visible, albeit accessible at a small cost, we find that it does not weight as much in conditioning access to credit. This suggests that policy makers ought to be careful when making public information highly visible and need to investigate any potential for stigmatization of certain categories of individuals, and its consequence for market efficiency.