The Limits to Compensation in the Financial Sector

In recent years bonuses tied to performance have become commonplace in banks and other financial institutions; indeed they now constitute a major part of employee compensation. The practice was originally justified by academic work on principal-agent contracts, which argued that performance bonuses would better align the interests of managers and shareholders. In this article we argue that such schemes are not well-suited to aligning these interests in the financial sector. There are two reasons for this failure. First, new financial products make it easy to create the appearance of superior performance over long periods of time even though the outsize returns are merely being driven by hidden tail risk. We show that it is virtually impossible to create performance contracts that get around this problem. Second, the complexity of new products and the size of modern financial institutions make it extremely difficult (and costly) to monitor risky activities directly. As in the first case, compensation schemes, including deferred compensation, are inefficient substitutes because it is easy to escape detection for long periods of time. This opens the door for outright fraud. We argue that a greater emphasis on ethical values, e.g., a duty of care to customers and shareholders, is more likely to produce effective reforms.