Brown, Others Stoke Fears of GAO Bias in 'Too Big to Fail' Report

WASHINGTON — Members of the Senate Banking Committee and others are raising concerns about the credibility of a pending watchdog report analyzing whether some banks are still "too big to fail," suggesting some experts it relies on may be too closely tied to Wall Street.

The Government Accountability Office is due this spring to release the second of two reports examining the issue, with this one focused on providing an estimate of the implicit subsidies the biggest banks maintain from a market belief the government will rescue them.

While the report is highly anticipated, some are questioning its value ahead of the release, citing the influence some experts tied to the banking industry could have in determining how that analysis is conducted. It was an issue raised by Simon Johnson, a professor at the Massachusetts Institute of Technology, and echoed by Sen. Sherrod Brown, D-Ohio, during a Senate Banking Committee hearing on Wednesday.

"Professor Johnson notes that the GAO weights all so-called expert opinion equally, regardless of whether the work is being produced in the public interest or at Wall Street's behest," Brown, who chaired the hearing, told a GAO official. "You know the influence that Wall Street has, their economic power has been enhanced, their political power has been enhanced… It's pretty clear that people want to make sure who these people are that are talking to you and who's paying whom, and that you work even harder at doing that."

Appearing on the panel alongside the GAO representative, Johnson said the report's conclusions will be suspect if the agency does not pay adequate attention to potential conflicts of interest.

"If the GAO cannot sort out sensible analysis from sophisticated lobbying, then its important follow-up report on the current value of implicit subsidies to large banks is unlikely to have much value," Johnson said. "The negative reputational effect on the credibility of the GAO, Congress, and the executive branch would be considerable and most unfortunate."

But Lawrence Evans, director of financial markets and community investment at the GAO, pushed back on the critique, disagreeing that the claim is warranted.

"I don't think, first of all, that's a fair criticism based on this report, because where we draw from some of the experts is just in identifying some of the challenges to implementing some of the provisions of Dodd-Frank," he said. "We are balanced, fact-based in our approach. We reach out to all stakeholders. We talk to all affected parties. We do speak with banks about how they fund themselves. We speak with investor groups."

He added that the GAO is clear in laying out the scope and methodology of its reports, noting that as economists turn to the second report, they are being "quite rigorous" in their evaluation of previous approaches to the issue.

"So we are aware of lobbying power. We are aware of conflicts of interest," Evans said. "In fact, when we reached out to academics to look at our model and our model specifications, one of the things that we explored was conflicts of interest. We wanted to make sure that these individuals were free from those types of influences."

Johnson shot back that the GAO also reached out to him for expert advice as it conducted its work, noting that nobody asked him about his conflicts of interest.

"They didn't ask me if I had a conflict of interest. They didn't ask me who pays for various parts of my activity. I mean, I would be happy to tell them. In fact, I put a lot of disclosure on my website. I hope that that was an exception, an aberration," he said. "I know quite a few people who work for the industry without necessarily fully volunteering in all instances the full extent of the income and other benefits they derive from" that work.

Evans countered that Johnson and others like him have large bodies of work that can be reviewed, arguing as well that interviews with various experts don't influence the agency's final opinions.

"We reach out to a wide range of stakeholders. We're interested in their opinions. Those opinions don't influence what we say or what we think at the end of the day," he said. "And again, the academic studies have clear methodologies. So you can tell when they're using untoward assumptions or undertaking methodologies that raise questions about the validity of the estimates."

The issue of researcher independence has been gaining increasing attention on Capitol Hill, after Sen. Elizabeth Warren, D-Mass., urged big banks last month to voluntarily disclose their contributions to think tanks. During the hearing, Brown cited a recent report by the New York Times detailing the undisclosed ties some academics have maintained with Wall Street, even as they vigorously support industry-friendly policies in letters and papers.

Overall, Evans said the GAO report would rely on clear data.

"At the end of the day this is about investor perceptions — investor expectations about how regulators will behave in an instance of distress. So in our work, we will be looking at how any subsidy we estimate might change over time," he said. "It is an empirical question to ask whether investors' expectations have been changed as a result of Dodd-Frank. We will also be talking to a lot of investors and other open market participants."

Lawmakers also spent some time discussing the GAO's first report, released in November, that detailed the amount of support the government provided banks during the financial crisis.

Many bankers and industry representatives maintain that the Dodd-Frank Act has effectively eliminated the problem of "too big to fail."

But Warren, among others, suggested more should be done even as regulators continue to finalize the financial reform law. (Brown and Sen. David Vitter, R-La., have introduced a bill that would significant raise capital requirements on the largest banks.)

"Even if the regulatory agencies issue all of the Dodd-Frank rules that remain and then rigorously enforce all of those rules, do you believe that would solve the 'too big to fail' problem?" Warren asked the witnesses except for Evans.

The panelists all said "no."

"I'm going to answer your question by saying no — and hell no," said Harvey Rosenblum, a professor at Southern Methodist University. "We cannot have this Dodd-Frank which is 850 pages roughly, probably will be 30,000 pages of regulations to implement it. It's not workable."