Is Virtue What We Buy or What We Sell?

If we judge politicians by what they buy, then Eliot Spitzer has clearly violated the public’s trust: he purchased the services of a high-priced prostitute, and may well end his political career as a result. But what if we judge politicians by what they sell? On this score, Spitzer may be one of the few politicians who has not prostituted himself to special interests.

An insightful new contribution to the emerging “forensic economics” literature by my frequent co-author Eric Zitzewitz makes the case that Spitzer is extraordinary because of the high price he placed on his integrity. Let me explain.

Spitzer made his name politically as New York’s attorney general, where he launched a number of high-profile investigations into the securities industry. The trouble with evaluating an attorney general’s performance is that it is very difficult for outsiders to make any inferences about the most important power used by the office — prosecutorial discretion. It may be that an attorney general settles for small fines against a wrongdoer because he is in their pocket; but equally, it may be that the evidence was weak, or the conduct at issue wasn’t that egregious.

Zitzewitz’s analysis focuses on analyzing the set of cases involving mutual fund market timing and late trading. Using available data, he develops some useful proxies of what he calls “the dilution of long-term fund shareholders from arbitrage trading.” Eric is too polite: this is a measure of how much money unsavory fund managers allowed to be diverted from our mutual fund savings. This all occurred because select friends of these managers were given the right to buy into the fund at this morning’s low price, rather than this afternoon’s higher price (even after the markets had risen).

In these cases, the evidence was uniformly strong, and by measuring this “dilution,” Zitzewitz can measure how egregious the misconduct by these firms was. All told, he compiles estimates of the harm — and public records on the restitution ordered — in 20 SEC-negotiated settlements. Of these, 16 involved the New York attorney general, and the resulting restitution typically amounted to about 80 percent of the harm — pretty close to full restitution. But in the 4 settlements in which Spitzer’s office was not involved, the SEC was willing to settle on a figure closer to 7 percent of the total harm.

What causes this different treatment? It can’t be the expertise and manpower of the New York Investor Protection Bureau, which has only about 30 employees, relative to the 1,200 folks working in the SEC’s Enforcement Division. And it isn’t that Spitzer was only interested in the big fish or big headlines: the finding of differential treatment is robust when controlling for the fund’s ability to pay or the timing of the settlement. Zitzewitz also obtains similar results when he uses two useful instrumental variables for whether the New York attorney general will be involved: whether a firm is headquartered in New York, and the percentage of its investors who are residents.

So what explains this disparity? Zitzewitz’s analysis suggests that the key is the SEC’s laxity, rather than Spitzer’s excess zeal. Often we economists worry about “regulatory capture,” and I’m concerned about the SEC’s “revolving door” personnel policy, in which young securities lawyers spend a few years at the SEC before moving on to greener pastures (e.g., working for financial firms against the SEC). Given these incentives, it seems pretty likely that these young lawyers don’t really want to upset potential future employers.

It seems to me that the truly important violations of the public trust are when the power we give our government officials is sold, rather than what government officials choose to buy. Yet our political scandals are too often dominated by private mistakes, rather than public misdeeds. This is why I’m more worried about what the SEC is selling than what Eliot Spitzer has been buying.