Regulating Banker Pay

The WSJ reports that the Fed plans extensive regulation of bank pay policies to prevent the taking of excessive risk:

Under the proposal, the Fed could reject any compensation policies it believes encourage bank employees -- from chief executives, to traders, to loan officers -- to take too much risk. Bureaucrats wouldn't set the pay of individuals, but would review and, if necessary, amend each bank's salary and bonus policies to make sure they don't create harmful incentives. ... Its strategy appears to go further than what some in the industry were expecting, given that it would apply to many employees, not just top earners. It would go beyond a more generic list of "best practices" that many thought the central bank would craft.

The proposal will likely push banks to use "clawbacks" -- provisions to reclaim the pay of staffers who take risks that hurt their firms -- in certain pay packages, among other tools, to punish employees for taking excessive risks with their firms' money. The central bank could also demand that more pay be offered through restricted stock or other forms of long-term compensation designed not to reward short-term performance.

The proposal raises four concerns. First, risk and return are inextricably linked. If you want a higher rate of return, you have to take greater risks. Financial institutions owned by shareholders thus must take into account the basic fact that shareholders have a strong incentive to favor risky projects. Because creditors have a prior claim on the firm’s assets and earnings, they get paid first; shareholders get the residual—whatever is left over. Shareholders thus prefer projects offering potentially high rate of returns, so there will be something left over after the creditors get paid.

If compensation policy encourages banks to reduce the risks they take, that policy will inevitably result in a lower rate of return for shareholders. This will significantly distort the flow of capital. Shareholders will divest bank stocks and invest in industries offering a higher rate of return. This will raise the cost of capital for banks and perhaps have the unintended consequence of making banks weaker rather than safer.

Second, whether or not that specific unintended consequence comes to fruition, it's critical to remember that government regulation of compensation has been characterized by unintended consequences. The Journal explains:

 Previous efforts to regulate pay have often failed or even backfired, sending pay higher. A 1980s law restricting "golden parachutes" helped spread what had been a rare practice. A 1993 law limiting annual salaries led to bigger stock-option grants, pensions and deferred compensation. Former Securities and Exchange Commission Chairman Christopher Cox once said the law belongs in the "Museum of Unintended Consequences."

For example:

Jonathan Macey, a professor at Yale Law School, says the Federal Reserve plan "responds to an acute problem" of taxpayers losing when a bank blows up, but ceding all the gains to traders and executives when the bank makes a profit.

The proposal could have unintended consequences, however, he said. If the Fed takes the 25 biggest banks and works to make sure they are consistent in the way they pay people, it "may make things worse because it contributes to lemming-like, herd behavior," he said. "Bubbles are caused by that behavior. What you want is diversity and heterogeneity between firms."

A third problem is that the proposal may not only distort the capital market, it may also distort the job market. If the best and brightest perceive finance as offering lower income potential, they'll be incentivized to look elsewhere. Again, the proposal thus could have the unintended consequence of making harder to strengthen the banking sector.

Finally, how do you tell whether a policy will encourage "excessive risk"? How do you decide what level of risk is appropriate versus excessive? Once you make that determination, how do you figure out whether a proposed pay policy will encourage the latter rather than the former? Why do we think government bureaucrats are going to be any good at answering those questions?

My guess would be that government bureaucrats will be over conservative. They won't want the Washington Post and NY Times writing articles about how they were asleep at pay switch. If so, all the problems outlined above will be exacerbated.

I'm not saying that risk management failures didn't contribute to the financial crisis. They did. I'm not saying that pay policies didn't contribute to those failures. They did. 

I'm just saying, the Fed damn well better be sure it knows what it's doing here ... and I'm dubious that it does.

Comments

[PatHMV]

Certainly government regulation will only complicate problems, as you demonstrate.

But it seems that the recent financial crisis revealed some systemic defects in compensation systems for financial agencies. Individual officers and employees were able to make significant salaries and bonuses without undertaking any risk comparable to the long-term risk to which they exposed their principles. Smaller banks originated bad loans and profited when they shortly thereafter sold to larger banks or Fannie or Freddie. The traders for those larger banks were paid based on the immediate short-term profits from those purchases, so they had no particular incentive to pay close attention to the long-term risks associated with the mortgages being purchased.

I can't help but wonder if some personal liability being attached to the highest ranking officers and directors of financial institutions, in the event that the institution becomes so insolvent that it must be bailed out, might not help restore a bit of sanity. That would undoubtedly make investments a bit more conservative, but I'm not certain that would be a bad thing, given our recent experiences.

[Brian]

Prof. Bainbridge is too intelligent to include a claim from the authority of the abysmal SEC chairman Christopher Cox. Tut tut.