August 24, 2010

Are Watchdogs Corrupt, or Just Biased?

Regulatory agencies favor firms that have good reputations.

In early 2010, the National Highway Transportation Safety Administration (NHTSA) issued a wide-scale recall of some models of Toyota cars after a flurry of reports of spontaneous acceleration and disabled brakes. The extent of the recalls — a dozen models and millions of cars — was as startling as the fact that Toyota, long regarded as a standard-bearer for automotive safety, was slow to alert regulators when the company discovered that serious problems might exist with some braking systems.

Regulators have not been immune to the fallout, as the public wondered how such egregious safety flaws could have gotten past the NHTSA, echoing other recent incidents that have called into question the efficacy of other gatekeeper agencies. Why did the SEC fail to act when financial services firms took sky-high risks with investors’ money? How could the Minerals Management Service fail to require petroleum giant BP to secure permits designed to protect sea life in the Gulf of Mexico prior to the Deepwater Horizon disaster?

The conventional interpretation of regulatory failures is that watchdog agencies are too easily influenced by the companies they are charged with policing and are held hostage, or captured, by the firms’ enormous resources and a seemingly unlimited ability (using lobbyists and lawyers) to wield influence. Safety and consumer advocates worry that as corporate power grows, regulators are finding it increasingly difficult to withstand the pressures and temptations.

But Professor Jerry Kim hypothesized that regulatory agencies favor certain firms not as a result of lobbying or corruption but instead because some firms have reputations for producing safe or effective products: the better a firm’s reputation, the better its regulatory outcome. “Using reputation as a proxy for safety may be a way regulators manage uncertainty,” Kim says, “and protect the agency from fallout — especially in an age where political scrutiny and rapid scientific development and innovation have upped the ante and the complexity of interactions between regulators and firms.”

Kim tested his hypothesis using data from more than 800 drugs that were submitted to the FDA for review by dozens of pharmaceutical firms between 1990 and 2004. He assigned each firm a measure of knowledge status — its intellectual expertise and experience in its drug portfolio — as a proxy for reputation. Kim then looked at the length of time between a drug’s submission and when it was approved to market (the very last stage of FDA approval, after which a pharmaceutical firm can begin to market the drug).

Controlling for all other factors, Kim found that the FDA approved drugs far more quickly for the top 15 percent of high-reputation firms — an average of 218 days sooner — than it approved drugs from other firms. Political contributions and a firm’s experience in dealing with the regulator, on the other hand, did not lead to a meaningful reduction in approval time.

The lag in approval time translates into a significant economic impact: by some estimates, a one-month delay in approval time results in about $40 million of lost revenue. Firms with drugs that don’t move through the approval process quickly miss out on significant revenue.

Furthermore, the effect of a standout reputation in a single major drug category was strong enough to spill over into other drug categories. A firm known for its cancer drugs facing approval for a different class of drug, such as a new nasal spray for asthmatics, could benefit from its reputation for cancer drugs even if it had no track record with asthma drugs.

This halo effect signaled to Kim that political motivations might explain the reliance on reputation. Consider the FDA’s typical response after a drug recall: it slows approval for all drugs, suggesting that the agency is trying to protect and repair its own reputation to retain functional autonomy and avoid scrutiny. But high-status firms suffered the smallest penalty, as, in a time of crisis, the FDA looked toward those firms it trusted most.

“It’s ironic that the firms that suffer the most due to drug recalls, the lower-status firms, are not the firms that are usually responsible for recalls, the ones with stellar reputations,” Kim notes. He suggests that lower-status firms (which are typically younger and smaller) may be able to offset the bias towards higher-status firms (which are typically older and larger) by carving out expertise niches that the more favored firms can’t or won’t enter, which can earn the notice of regulators.

Kim found no evidence that firms with high status actually produced better quality products overall, which suggests that firms that benefit from regulatory bias should exercise caution. “These firms shouldn’t assume they can skate by without further consequences down the road, which may be what happened in the case of Toyota,” he says.

Kim’s research also offers some insight into the role of gatekeepers more broadly. Critics are gatekeepers of a sort, weeding out boilerplate movies, books, or music from masterpieces. Similarly, journalists vet information about affairs of state and business so the public can form opinions and respond with dollars or votes.

“However trusted these gatekeepers may be, they aren’t prone to corruption so much as they are just human, and prone to bias,” Kim says.

Jerry Kim is assistant professor of management at Columbia Business School.