In the world of academia, there are debunkings that take issue with the assumptions made by other scholars, or the choice of data, or the interpretations made therefrom, that are worded so carefully and politely or are so buried in jargon that, unless one is a specialist in the field, one has a hard time evaluating them. And then there are debunkings that sweep down like cohorts of Assyrians, gleaming in purple and gold, leaving not a wall standing or a blade of glass growing in their wake.
So it is with Georgetown scholar Adam Levitin's full-scale frontal assault on a recent study, funded by the American Banker's Association, attacking the proposed Consumer Financial Protection Agency.
Authored by David S. Evans of University College London and the University of Chicago Law School and Joshua D. Wright from the George Mason University School of Law, "The Effect of the Consumer Financial Protection Agency Act of 2009 on Consumer Credit" claims that the CFPA would have substantial negative effects on consumers and the larger economy. Specifically, say the authors, it would "increase the interest rates consumers pay by at least 160 basis points, reduce consumer borrowing by at least 2.1 percent, and reduce the net new jobs created in the economy by 4.3 percent."
(One hundred and sixty basis points is equal to a 1.6 percent rise in interest rates.)
Here's how Adam Levitin summarizes how Evans and Wright concocted their numbers in a blog post at Credit Slips.
How, you might ask, did anyone possibly arrive at such precise predictions based on legislation that does not create any substantive [new] regulation of the credit industry, but would merely transfer largely existing powers to a new agency?
The short answer: just make up the numbers. I kid you not. Evans and Wright selectively chose a study on the impact of a different regulation (interstate banking restrictions) on credit cost. They briefly argue it is analogous to the CFPA Act, which they claim will have double the impact. (Why double? Why not?) Then they take that number and multiply it by an elasticity metric for the demand impact. And for the coup-de-grace, they take a misleading number on net job creation and conjecture with no basis that it would be reduced by 5 percent. These numbers are presented as "plausible, yet conservative" assumptions.
If some of that is Greek to you, well, yes, it is to me also. But for the full annihilation, one must refer to Levitin's new paper, "A Critique of Evans and Wright's Study of the Consumer Financial Protection Agency Act."
Let me inject my own disclaimer. I tend to suspicious of, prima facie, industry-funded studies that purport to tell me why new regulation would be bad for their business, consumers, children and other living things. And in this case, in which the banking industry is pushing back against rules designed to help prevent a repeat of the catastrophe that it was a major player in precipitating, the predisposition to be critical is overwhelming. I've also been admiring Levitin's consumer-friendly and credit industry-critical work for several years now. That said, this particular debunking seems to stand above the normal crowd, both in its thoroughness and in its ferocity.
As the basis for their claim that the CFPA will make credit more expensive, Evans and Wright cite one study that shows that state regulation of interstate banking resulted in an 80 basis point rise in credit costs. And then, argues Levitin, they make a questionable logical jump:
Levitin:
Regulation A resulted in the cost of credit rising by X. Therefore, regulation B will result in the cost of credit rising by 2X. For Evans and Wright's syllogism to hold, Regulation B must equal 2 x Regulation A. The two regulations in point are so different as to make comparison a stretch, much less to allow for a precise mathematical expression ..."
How are they different?
It is not surprising that jurisdictions that limited the number of bank competitors and prevented the formation of economics of scale would have higher credit prices than jurisdictions that did not. But it hardly follows that the CFPA would have a similar effect. The legislation is simply not analogous. The CFPA Act does not limit the number of firms that can compete in the market. It does not prevent economies of scale. It does not create any new substantive regulation. Instead, the CFPA Act grants the CFPA rule-making powers that already exist for various federal agencies. The Federal Reserve Board, OCC, OTS, NCUA, FDIC, and FTC already have the power to prohibit unfair and deceptive actives and practices. The CFPA does not create a private right of action. It preserves existing rights of action for federal and state government, but there is no reason to expect a barrage of litigation from governmental units. Most legal interpretative issues get resolved out of court or in a single litigation. Similarly, the risk of inconsistent state regulation is vastly overstated. As long as there is an active federal regulator, like a CFPA, states are unlikely to engage in more vigorous regulation, and even if they did, states tend to enact similar regulations, and industry tends to respond by conforming with the strictest level of regulation. The key point here, however, is the impact of the legislation is completely speculative and certainly not susceptible to precise statistical predictions.
There's plenty more where that came from, if you are looking for details. As for ferocity? Evans and Wright qualify their assertions by noting repeatedly that their assumptions are "plausible" (and at one point, they even deploy the almost impossible to disprove "It is not implausible ..." formulation). Levitin unloads.
To be sure, Evans and Wright take care to repeatedly caveat their conclusions as "plausible." But scholarship does not operates on the level of mere plausibility. Thus, it's plausible that Jimmy Hoffa is still alive. It's plausible that Hitler escaped his bunker and took up residence in South America. And it's plausible that I have a bridge to sell you in Brooklyn. But plausible is not the same as likely or even reasonably inferred. To write a study on "The Effect of the Consumer Financial Protection Agency Act of 2009 on Consumer Credit," and have the lead finding simply be something that is "plausible" is too coy. This is not a luxury in which academics can engage.
Which leads to some of the stronger language I have seen in the economic to and fro over the effects of regulation:
Evans and Wright's lengthy piece makes numerous tendentious claims and questionable assumptions. But at its heart are a trio of claims so outlandish, so unsupported by evidence, and so coyly phrased that they cannot pass for good faith scholarship...
There is room for robust academic debate about consumer credit regulatory policy. There are plenty of issues to legitimately question about the creation of a CFPA, including whether it would result in decreased credit availability and/or higher costs of credit. But there is a world of difference between the honest, good faith exploration of such claims and making arguments so fallacious that they do not pass the straight-face test of scholarship. Mere plausibility does not cut it; "non-implausibility" even less so. Evans and Wright are serious scholars from whose work I have learned a great deal in the past. Sadly, their study of the CFPA is not worthy of their names.
The white glove has been thrown. Choose your weapon, sir!
Shares