How Costly Is Corporate Bankruptcy for Top Executives?

To what extent are CEOs filing for bankruptcy tainted by the bankruptcy event? On the one hand, the CEO bears a major responsibility for the firm going broke. After all, the filing might have been avoided if the CEO had managed to reduce firm leverage or otherwise reorganize debt claims in time to stay out of court. On the other hand, CEOs going through bankruptcy likely gain valuable experience from the crisis. The net impact of these two opposing effects on executive reputation is an open empirical question.

In the paper, How Costly is Corporate Bankruptcy for Top Executives?, which was recently made publicly available on SSRN, we provide some first systematic estimates of top executives’ personal costs of corporate bankruptcy. The estimates are based on 324 large public companies filing for Chapter 11 bankruptcy over the past two decades.

The study provides evidence on the following three questions. First, do top executives experience large personal losses (both income and wealth) when filing for bankruptcy? Second, do creditor control rights influence the probability of CEO departure and the income losses? Third, do ex ante predicted personal losses affect CEO’s decision to leave the firm and their compensation contract design?

Why are these questions important? While corporate bankruptcy imposes economic and emotional hardship on executives and other employees in general, the costs incurred by the top executives are of particular interest in corporate finance. High expected personal costs may cause risk-averse executives to hedge against default by reducing corporate leverage and perhaps under-invest in risky corporate projects, thereby resulting in a potentially important form of agency costs associated with debt.

We are particularly interested in how bankruptcy filing affects the future labor income stream of top executives. Lack of personal data on executive compensation has prevented earlier studies from properly estimating this income effect. Our study solves much of this data issue by tracking post-bankruptcy CEO employment using SEC filings, news articles, executive directories, and social networks. Given the new employment category we estimate the associated employment income and change in labor income associated with departing the bankrupt firm.

We find that as many as half of the CEOs are able to maintain full-time executive employment, either at a new company or at the restructured company emerging from Chapter 11. This is a substantially greater likelihood of maintaining full-time executive employment thanwhat is implied by earlier studies, which, for lack of data, have essentially been forced to assume zero post-departure executive income.

For the group of CEOs maintaining full-time executive employment, the median change in total compensation (the sum of all income sources – salary, bonus, equity-based pay, and severance pay if any) is statistically indistinguishable from zero. It appears that these executives’ labor market reputations are not tainted by the bankruptcy filing.

For the remaining CEOs (i.e., those who do not find new executive employment), bankruptcy filing is costly. They experience a median compensation loss until retirement with a present value of about $4 million (in 2009 dollars)—or about five times their pre-bankruptcy compensation level. Results suggest that these CEOs are strongly penalized by the bankruptcy event.

In addition, a typical CEO, whether or not she maintains full-time executive employment, suffers substantial losses from her equity holdings in the bankrupt firm. The loss is about $2 million in stock and option holdings when measured from the last fiscal year prior to bankruptcy through the bankruptcy filing, and about twice that amount when measured from two years prior to filing.

On a more detailed level, our analysis suggests that incumbent CEOs (defined as CEOs in place or internally promoted within two years prior to the Chapter 11 filing) are more likely to be forced to leave (some by senior creditors) and to suffer relatively large income losses.

Also, CEOs with long tenure at the firm suffer greater income losses. One possible interpretation is that longer tenure cause the CEO’s managerial skill set to be specialized to the business of the bankrupt firm and thus difficult to transfer to a new employer – and so the wage loss from having to leave the firm is greater. Our study also shows that CEOs are more likely to experience large post-bankruptcy income losses when there is industry-wide financial distress.

Over the past two decades, creditor influence over and control rights in Chapter 11 proceedings has increased substantially through actions such as post-petition financing, prepackaged filings, and loan-to-own. One measure of creditor influence over the bankruptcy proceedings is the existence of debtor-in-possession financing. Using this measure, greater creditor control is associated with a greater probability that the CEO departs the bankrupt firm. On the other hand, our paper finds that CEOs are less likely to be forced out by creditors representing large suppliers.

Finally, our analysis shows that the cash portion of the total compensation is increasing in the predicted CEO income loss – as if CEO labor contracts provide a partial hedge against risk-averse CEOs’ equity losses from the bankruptcy event. We also find that greater predicted CEO income loss from bankruptcy increases the probability of forced turnover, suggesting that some low quality managers stay with the distressed firm in an attempt to keep extracting labor market rents, until forced out by creditors.

In sum, the evidence shows that the median top executive tends to suffer smaller personal losses during bankruptcy than conventionally believed. Executives who appear to earn labor market rents tend to suffer larger losses than the median. Creditors play an active role in CEO turnover, suggesting a disciplining effect of both creditor control rights and labor market repricing of executives.

The full paper is available for download here.