Truth on the Market

Academic commentary on law, business, economics and more

Archive for November, 2009

« Previous Entries
Next Entries »

PeaceHealth and De Facto Exclusive Dealing, Part III

Posted by Thom Lambert on November 13, 2009

Josh’s thoughtful response (Bitchslap? Nah.) to my post criticizing the Ninth Circuit’s recent Masimo decision raises a number of important matters. I started to just submit a comment to Josh’s post, but then I figured a reply was post-worthy. (I don’t want the antitrust nerds who read these technical posts — and here’s to you, dweeby friends! — to miss the discussion.)

Here’s a run-down of the discussion so far:

I criticized the Masimo court for reasoning that a plaintiff complaining of a rival’s bundled discounting may proceed on a de facto exclusive dealing theory when (1) there’s no express covenant of exclusivity (so it is merely the defendant’s low prices that are leading buyers to forego rivals) and (2) the defendant’s bundled discount would pass muster under the Ninth Circuit’s PeaceHealth standard. Under that so-called “discount attribution” standard, a bundled discount is immune from liability if each product in the bundle is priced above cost after the entire amount of the bundled discount is attributed to that product alone.

The basis for the PeaceHealth court’s safe harbor was a desire to immunize bundled discounts that cannot exclude any single-product rival that can produce its own product as efficiently as the defendant. After all, any bundled discount that results in above-cost pricing on each bundled product, even after the entire discount amount is attributed to that product, could be matched by, and thus cannot exclude, any equally efficient single-product rival. Because such discounts do not make it impossible for equally efficient rivals to stay in the market, the PeaceHealth court reasoned, they are not unreasonably exclusionary.

In my initial post, I argued that this same analysis should apply even when the bundled discount is so successful that it induces customers to drop rivals’ brands and to purchase only from the discounter. If a discounter doesn’t procure a contractual covenant of exclusivity and instead just offers a really attractive discount, that discount, which provides immediate consumer benefit, should be immune from liability as long as it doesn’t threaten to exclude an equally efficient rival. The discount can’t pose such a threat unless it’s so extreme that an equally efficient single-product rival (who must match the entire dollar amount of the bundled discount on its single product) could not match it, and that will be the case only if the discount results in a below-cost price on some item in the bundle after the entire amount of the discount is attributed to that item. Thus, I argued, an allegation (or even proof) that a bundled discount resulted in de facto exclusive dealing should not alter the PeaceHealth safe harbor.

Now, there are two ways one might interpret my position. I could be saying simply that the optimal liability rule would not condemn de facto exclusive dealing induced solely by bundled discounts that pass muster under PeaceHealth. I could appropriately stake that claim, even if I acknowledged potential anticompetitive harm from such discounts, if I concluded that the likelihood and magnitude of such harm was low and the likelihood and magnitude of harm from “false positives” (i.e., improperly condemned discounts) was high. Alternatively, I could be making a more aggressive economic claim, asserting that de facto exclusive dealing accomplished via bundled discounts that satisfy PeaceHealth cannot be anticompetitive.

In his thoughtful and nuanced post, Josh suggests he might, given more empirical analysis, agree with the former position (about what liability rule is optimal, given error costs). But he rejects the more aggressive economic claim. That claim, he maintains, is deficient because it fails to account for the fact that bundled discounts resulting in de facto exclusive dealing may “raise rivals’ costs” even if the discounts would pass muster under PeaceHealth.

Here’s the intuition (my understanding of it, at least): A bundled discount that’s big enough to usurp lots of business from the discounter’s rivals, but not big enough to run afoul of PeaceHealth, may still preclude those rivals from attaining minimum efficient scale (i.e., the output level at which all scale economies are exhausted). The discount may thus prevent rivals from achieving equivalent efficiency with the discounter. That would be anticompetitive. Thus, it is not true, as I said, that “[w]hen it comes to bundled discounts, … there can be no anticompetitive harm in the form of predation, unreasonable exclusion, or foreclosure if the competitive product is priced above the defendant’s cost once the entire discount is attributed to that product.” Rather, Josh suggests, a bundled discount that results in substantial foreclosure of the discounter’s rivals may raise those rivals’ costs and thus be anticompetitive, even if the discount would pass muster under PeaceHealth and therefore not amount to predation.

So what to make of all this? Read the rest of this entry »

Posted in antitrust, economics, law and economics, markets, regulation | 5 Comments »

Hedge Funds & The SEC

Posted by J.W. Verret on November 11, 2009

Thanks so much to my new colleagues at Truth on the Market for inviting me to join an impressive group of scholars on a blog I have followed for many years now. Chairman Frank of House Financial Services has an interesting thought on how to prevent the next financial crisis.  Let’s require hedge funds, along with other private pools of capital, to register with the SEC as investment advisers.  But wait, wasn’t Bernie Madoff a registered investment adviser?  Yes he was.  And did the SEC’s random compliance inspections detect his fraudulent scheme?  No, unfortunately they didn’t.  Ribstein offers more detail on the OIG report on the Madoff scandal here.  But surely, the SEC’s Division of Enforcement responded to multiple tips about Madoff?  Well, actually, they investigated Madoff multiple times.

Here are a couple of useful clips from the SEC Inspector General’s Report on Madoff: “The OIG investigation found the SEC conducted two investigations and three examinations related to Madoff’s investment advisory business based upon the detailed and credible complaints that raised the possibility that Madoff was misrepresenting his trading and could have been operating a Ponzi scheme. Yet, at no time did the SEC ever verify Madoff’s trading through an independent third-party, and in fact, never actually conducted a Ponzi scheme examination or investigation of Madoff. The first examination and first Enforcement investigation were conducted in 1992 after the SEC received information that led it to suspect that a Madoff associate had been conducting a Ponzi scheme…. In the examination of Madoff, the SEC did seek records from the Depository Trust Company (DTC) (an independent third-party), but sought copies of such records from Madoff himself. Had they sought records from DTC, there is an excellent chance that they would have uncovered Madoffs Ponzi scheme in 1992.”

Interesting.  We imagine the SEC is familiar with the most basic principle of auditing, that obtaining third party verification for randomly selected transactions from sources unlikely to have participated in a possible fraud is the best way to verify reported results.  The SEC should be familiar with the principle, as it has always played an important role in oversight of auditing principles used by accounting firms.   This principle is important for auditing, but it becomes even more important for an investigation triggered by allegations of fraud.

In fact, the SEC’s compliance staff is so backed up that many hedge funds who voluntarily registered with the SEC (despite the SEC’s mandatory registration requirement being overturned by the DC Court of Appeals in 2006) have complained that despite their high compliance costs for registration they haven’t been given compliance reviews for three years.  As such, they can’t obtain the benefit of telling potential clients they’ve been inspected despite having invested in registration and compliance.

I have an alternative plan for hedge fund oversight, not that the Hill is listening, in Dr. Jones and the Raiders of Lost Capital: Hedge Fund Regulation Part II, an SRO Proposal.  My suggestion is to take the same approach that we do for broker-dealer regulation, and create a Self-Regulatory Organization for hedge funds.  The SEC could facilitate this by, for instance, responding to a mandatory requirement for hedge fund registration by offering exemptive relief for funds registered with the Hedge Fund SRO.  To keep the SRO from becoming a vehicle for industry capture, the SEC would play a role in:

i) vetting candidates to the Board to make sure a diversity of fund size and strategy is represented;

ii) approving amendments to the charter of the SRO; and

iii) overturning SRO rules only if they failed a high standard of deference.

The Madoff scandal also puts the Financial Industry National Regulatory Authority (FINRA) in a bad light for failing to respond to Madoff tips as well.  And there is something to be said for the idea that doing nothing at all could be the best option.  But compared to the Frank proposal, we could expect a Hedge Fund SRO to internalize more of the costs of compliance and result in a more efficient outcome than direct regulation by the SEC.  I also developed this idea before the PCAOB constitutionality challenge lit up, so perhaps my idea isn’t quite legal.  But that doesn’t mean it might not be a good idea.

Posted in markets | Comments Off

Gelbach, Helland and Klick on Single Firm, Single Event Studies

Posted by Josh Wright on November 11, 2009

Larry Ribstein points to the new paper from Gelbach, Helland and Klick on Valid Inference in Single Firm, Single Event Studies.  This is an important paper with implications for finance, securities litigation and antitrust where event studies are frequently used as economic expert evidence.  Ribstein gives a good, non-technical explanation of its contribution:

Essentially what’s happening is that single-firm event studies are determining the existence of abnormal returns against an assumption that the firm’s returns are “normally” distributed under a bell-shaped curve. “Abnormal” refers to returns located around the “bell’s” right and left sides. The problem is that returns are often not normally distributed, and you can’t determine if the observed returns are abnormal if you don’t know the shape of the curve. The paper proposes “a very simple but statistically sound alternative,” the “SQ” test, which does not present the problem of assuming a normal distribution.

The slightly more technical version is as follows.  The standard approach to the event study is comparing t-ratios to critical values drawn from the normal distribution.  So if we are interested in abnormal returns, and those returns come from a normal distribution, the standard approach will perform pretty well.  But the evidence is that abnormal returns are non-normal, and GHK present a lot of evidence on that score.  One might also justify the standard approach in a context with a lot of events. But in the law and economics literature, event studies with small numbers of firms and a small number of events (often 1) are common — and of policy importance.  The primary contribution of GHK is to offer an alternative approach with better statistical properties for these types of studies.

The SQ test that they propose works is related to a Chow test.  Recall that a Chow test is aimed at detecting a structural break in the relationship between Y and X between two sets of observations given the normality of regression residuals.  A Chow test focuses on testing whether the slopes are the same across both periods of observations.  The logic of the GHK test is to apply a sort of structural break test which aims at testing whether the abnormal returns before and after the break come from the same distribution.   GHK come up with the SQ test which provides a test statistic and uses critical values which are estimated from the empirical distribution of predicted residuals from earlier in the sample.

GHK have a great example in the text of the paper discussing how the SQ test would apply in practice:

To illustrate using the example on which we focus below, suppose that a firm discloses that its past quarterly earnings were substantially below the level claimed in an earlier earnings statement. A class of plaintiffs file an action under SEC rule 10b-5, citing standard fraud-on-the-market arguments. In light of Dura’s requirement that plaintiffs establish loss causation pursuant to the firm’s corrective disclosure, an expert witness wants to test the null hypothesis that the corrective disclosure had no effect on a firm’s stock price; the alternative hypothesis is that the event has reduced the value of the firm’s stock.

To use our method, the witness would obtain data on the security’s daily return and the market return for both the event date and a set of, say, n = 99 pre-event observations. She would then use OLS estimation to estimate the firm’s beta and the coefficient on an event dummy, with the latter coefficient being the estimated event effect. All of these steps are taken in both the standard approach and in ours. To implement our test, the analyst would then calculate the fitted residuals from the estimated model, sort them, and find the 5th most negative value among the
non-event dates. She would reject the null hypothesis if the coefficient on the event dummy were less than or equal to this value. Remarkably, the Type I error rate of this test converges to 0.05, regardless of the shape of the true distribution of abnormal returns.

The intuition for this result is simple. In a sample of 99 randomly drawn variables, the fifth most negative element is the sample 0.05-quantile. It has long been known that sample quantiles are consistent estimators of population quantiles, so that the sample 0.05-quantile of a large collection of abnormal returns is an excellent estimate of the 0.05-quantile of the true underlying probability distribution for the abnormal returns. As we discuss below, this quantile is the key estimand in assessing whether a single event on a known date significantly reduced a firm’s value.  While the details below involve some technical points, this simple example illustrates how easy our sample quantile (SQ) test is to use in practice.

What interested me in this paper the most was thinking about its application to antitrust event studies.  There is now a substantial literature on event studies in antitrust — which frequently focus either on antitrust litigation or mergers.  For example, McAfee and Williams (1988) suggest that the standard event study methodology cannot detect anticompetitive mergers using data from a horizontal merger “known” to be anticompetitive.   Bittlingmayer and Hazlett (2000) use the event study methodology to test the reaction of financial markets to the antitrust litigation against Microsoft.  Eckbo (1983), Garbade, Silber and White (1982), Werden and Williams (1989) and Eckbo and Wier (1985) are all examples in this literature.  There is of course now also the debate on the value and appropriate use of merger retrospectives to evaluate merger policy (see, e.g. Carlton 2008).  Weinberg offers an excellent and up to date literature review.  There are also examples of litigated merger decisions that rely on event study evidence.

All of this raises important issues about anticompetitive policy generally, but also expert evidence.  Ribstein raises the possibility of judges using their own experts to overcome the bias of hired experts tied to their own standard theories.  Oddly enough, the rate of judges appointing their own experts in antitrust cases is infinitesimal despite the growth in econometric and economic sophisticated of evidence over the last couple of decades so I’m a bit skeptical of  this solution in that context.   But maybe I shouldn’t be.  In the meantime, I think the GHK paper raises all sorts of interesting issues for event study analysis in antitrust and more generally.

Enjoy the paper.

Posted in antitrust, economics, law and economics | Comments Off

The UPS v. FedEx "bitchfight"

Posted by Geoffrey Manne on November 10, 2009

Hilarious video from reason.tv.

(HT: Luke Froeb)

Posted in business, markets, musings, politics, regulation | 6 Comments »

Should PeaceHealth Apply to De Facto Exclusive Dealing Claims?

Posted by Josh Wright on November 10, 2009

Thom answers this question in the affirmative in his excellent post about the Ninth Circuit’s analysis in Masimo and is disappointed that the Ninth Circuit rejected the discount attribution standard as the sole test for Section 2 in favor of a separate inquiry as to whether the bundled discount arrangement resulted in a substantial foreclosure of distribution and competitive harm.  Thom describes this reasoning as “sorely disappointing.”  I’m tentatively not convinced things are as bad as Thom sees them and want to explain why.  Maybe Thom can persuade me that I ought to be more upset about Masimo than I am.

Let me start with two preliminary points.

First, I agree that bundled discounts are generally pro-competitive for all of the reasons Thom states as well as some others.  While there is some empirical evidence that bundled discounting appears in highly competitive markets where anticompetitive theories do not apply, suggesting pro-competitive efficiencies, but little empirical verification of a high likelihood of competitive harms.

Second, despite our agreement about the generally efficiency of bundled discounting, Thom’s claim that a bundled discount distribution arrangement cannot result in anticompetitive effect is overstated as a matter of economic theory.  My basic point is that it is possible, as a matter of economic theory, for distribution arrangements involving bundled discounts that satisfy the PeaceHealth safe harbor to result in anticompetitive effects.  Despite this economic point, I’m not sure that Thom and I disagree on the ultimate appropriate legal treatment of bundled discounting.  I’ll get back to that.

Now, to defend my claim.

Let’s start with Thom’s position that, contra the Ninth Circuit, a bundled discount scheme that satisfies PeaceHealth’s discount attribution test (i.e. prices are still above cost after the discount is fully attributed to the competitive product in the bundle) should be immune from Section 2 liability even if the arrangement results in the “foreclosure” of a sufficient share of distribution to deprive rivals of the opportunity to have access to a critical input (such as shelf space) required to achieve minimum efficient scale.

What is the anticompetitive story in these “bundled discount as de facto exclusive dealing” set of cases?  Put simply, the anticompetitive theories are based on the notion that the monopolist’s distribution arrangement will deprive the rival of the opportunity to reach minimum efficient scale through the foreclosure of access to some critical input do not depend on offering distributors a price that fails the discount attribution standard.  A broad set of “exclusionary distribution” cases allege that various forms of marketing arrangements between manufacturers and retailers result in a situation where the monopolist is purchasing exclusion + distribution rather than just distribution.

The economic literature giving rise to these anticompetitive theories of exclusive dealing as “raising rival’s costs” is about the conditions under which manufacturers will be able to purchase exclusion from downstream firms and the price that they will have to pay to do so.  Manufacturers make payments to distributors for access to shelf space in a lot of ways: lump sum payments such as slotting fees, rebates, loyalty discounts, bundled discounts, RPM, cooperative marketing dollars, trade promotions, and more.  But the key question should not turn on the form of those payments.  It should turn on whether the contracts satisfy the conditions necessary for anticompetitive harm: are rivals foreclosed from a sufficient share of distribution that they cannot achieve minimum efficient scale?

This begs the question: is a price that fails the discount attribution test a necessary condition for the above set of theories to operate?  I don’t think so as a matter of theory.  One can think of the raising rivals’ costs theories of distribution as the manufacturer paying a set of distributors to join the manufacturer’s cartel.  What payment would be sufficient to sustain that agreement without defection (distributors would all have the standard incentive to cheat)?  The answer to that question depends on a lot of things: upstream and downstream entry conditions, switching costs, number of distributors, the existence and magnitude of economies of scale or scope, etc.  But I don’t think that there is any reason to believe that economic theory provides a linkage between passing the discount attribution test and failure to satisfy the necessary conditions for standard raising rivals’ cost-based exclusion theories.  Thus, in theory one suspects that there are distribution arrangements that could logically survive PeaceHealth but also potentially create anticompetitive effects because they satisfy the conditions of the exclusion theories.  Let’s call that set of agreements X.

The existence of X doesn’t necessary mean that I disagree with Thom about the appropriate legal rule.  If X is very small such that it would be more costly to identify these agreements and prosecute them, one could justify Thom’s rule on those grounds.  If enforcement actions against X would lead to substantially greater error costs than Thom’s rule, one could also justify his position on those grounds.  The existing empirical evidence, to my knowledge, is insufficient to make such fine grained determinations.  However, the same evidence also tells us that manufacturer arrangements to pay for distribution and promotion are incredibly common, provide benefits to consumers, and occur in competitive markets.  Indeed, I’ve written a great deal about the set of conditions under which the normal competitive process generates payments for distribution. As such, I agree with Thom that it is incredibly important to establish workable and broad safe harbors in this area that minimize error costs. What I reject is the strong economic claim that appears in Thom’s post:

When it comes to bundled discounts, which generally reflect (or promote) cost-savings and which provide an immediate benefit to consumers, there can be no anticompetitive harm in the form of predation, unreasonable exclusion, or foreclosure if the competitive product is priced above the defendant’s cost once the entire discount is attributed to that product.

If the plaintiff is making a predation claim involving bundled discounts, I think the PeaceHealth standard is workable and useful and we should keep it.  A potential case might even be made, as discussed, to justify PeaceHealth as the universal standard for bundled discount claims even when they alleged exclusionary deprivation of scale because we think X is sufficiently small or unimportant or especially susceptible to Type I error.  But I don’t read Thom as making that case.  Perhaps he is and I hope he’ll clarify.

To repeat: I just don’t think that there is any reason to believe that exclusion in the sense defined here is not theoretically possible as a matter of economics because we observe a price that passes PeaceHealth.  As such, I don’t want to throw out foreclosure analysis as an important and relevant part of the antitrust inquiry.  Let me end with a few words in defense of foreclosure analysis which I think gets a bad rap nowadays.

There are costs to keeping the foreclosure analysis, and having two standards for two different allegations of anticompetitive harm.  Beyond that, of course, foreclosure analysis is full of its own complications, e.g. foreclosure of what? does duration of contract matter? what about staggered expiration dates?  But despite its complications and the potential for abuse, the foreclosure analysis asks the right question in deprivation of scale questions and the one that we know is explicitly linked to an important necessary condition of a very large set of the theories of harm alleged in monopolization cases.  Getting a legal standard reasonably tied to the necessary conditions for anticompetitive harm, as Thom knows from his important work in the RPM area, is not always an easy thing to do in antitrust.

By the way, I think that my objection here survives Thom’s “Hydra critique” that the mode of antitrust analysis should be a function of economic substance rather than form.  I agree that the critical question is whether the conduct is likely to impair the competitive process to the detriment of consumers.   The point here is that the the deprivation of scale claims are or at least can be, as a matter of economic substance, different than pure price predation claims.

The critical economic point is that the set of distribution arrangements must, as the literature says, raise a rival’s cost of operating or impair his ability to exist.  Those arrangements that do not should not trigger antitrust violations.  And of course, those that do not satisfied a necessary but not sufficient condition for competitive harm.  The key point is that in cases involving allegations of deprivation of scale, the economic consensus is that those claims require allegations of exclusion require foreclosure sufficient to deprive rivals the opportunity to compete for minimum efficient scale.  If we are ready to accept that this is the state of economic consensus, then we ought to explicitly include this showing in the part of the plaintiff’s burden.  The antitrust law currently attempts to get at this inquiry through foreclosure analysis, requiring something around 40 percent foreclosure share in de facto exclusionary cases.  That seems sensible to me.

Antitrust can handle different standards.  If the plaintiff is alleging deprivation of scale, lets make substantial foreclosure a necessary (but not sufficient) condition.  If the plaintiff is alleging a price predation argument that does not depend on deprivation of scale, PeaceHealth is a safe harbor.  Would that be so bad?  And one more question for discussion purposes, if Thom is right about PeaceHealth in the context of bundled discounts, doesn’t this also apply to any payment distribution?  For example, I think the logic clearly applies that single product loyalty discounts ought to be analyzed the same way, i.e. we should use discount attribution to apply the discount on so-called non-contestable units to the contestable ones and apply the same filter.  But if that’s true, exclusive dealing with discounts is a loyalty discount where the threshold volume is set to 100% of the distributor purchases.  If that’s right, Thom are you arguing that we should get rid of all exclusive dealing law whenever there is a discount scheme?

Posted in antitrust, economics | 1 Comment »

Hayek and the Fall of the Wall

Posted by Thom Lambert on November 10, 2009

As we reflect back on the remarkable events of twenty years ago, and as we witness more and more centralized economic planning in our own society, we should pause to remember what the fall of the wall revealed.

Consider Alan Greenspan’s account (The Age of Turbulence, pp. 131-32):

Controlled experiments almost never happen in economics. But you could not have created a better one than East and West Germany, even if you’d done it in a lab. Both countries started with the same culture, the same language, the same history, and the same value systems. Then for forty years they competed on opposite sides of a line, with very little commerce between them. The major difference subject to test was their political and economic systems: market capitalism versus central planning.

Many thought it was a close race. West Germany, of course, was the scene of the postwar economic miracle, rising from war’s ashes to become Europe’s most prosperous democracy. East Germany, meanwhile, became the powerhouse of the Eastern bloc; it was not only the Soviet Union’s biggest trading partner but also a country whose standard of living was seen to be only modestly short of West Germany’s. . . .

The fall of the wall exposed a degree of economic decay so devastating that it astonished even the skeptics. The East German workforce, it turned out, had little more than one-third of the productivity of its Western counterpart, nothing like 75 percent to 85 percent. The same applied to the population’s standard of living. . . . The extent of the devastation behind the iron curtain had been a well kept secret, but now the secret was out.

Hayek, it seems, was right. May we never forget.

Posted in economics, markets | 1 Comment »

Oracle is nonplussed; the DOJ is . . . plussed?

Posted by Geoffrey Manne on November 9, 2009

The European Commission has issued a Statement of Objections in response to Oracle’s proposed acquisition of Sun.  The deal had already cleared the DOJ’s review.  Oracle is none too happy about the development, issuing a strongly-worded statement.  Here’s a taste:

The database market is intensely competitive with at least eight strong players, including IBM, Microsoft, Sybase and three distinct open source vendors. Oracle and MySQL are very different database products. There is no basis in European law for objecting to a merger of two among eight firms selling differentiated products. Mergers like this occur regularly and have not been prohibited by United States or European regulators in decades.

The DOJ also issued a statement.  Here’s a taste:

After conducting a careful investigation of the proposed transaction between Oracle and Sun, the Department’s Antitrust Division concluded that the merger is unlikely to be anticompetitive. This conclusion was based on the particular facts of the transaction and the Division’s prior investigations in the relevant industries. The investigation included gathering statements from a variety of industry participants and a review of the parties’ internal business documents. At this point in its process, it appears that the EC holds a different view. We remain hopeful that the parties and the EC will reach a speedy resolution that benefits consumers in the Commission’s jurisdiction.

I love that last sentence–totally non-committal.

OK–I don’t want to make too much of this.  The DOJ is not necessarily happy about the Commission’s intervention here, and is probably just being polite (although maybe a little ire is called for).  And, actually, the DOJ press release goes on to make a nice, succinct argument for why the EU’s action is probably misguided.  Notes the DOJ:

The Division concluded, based on the specific facts at issue in the transaction, that consumer harm is unlikely because customers would continue to have choices from a variety of well established and widely accepted database products. The Department also concluded that there is a large community of developers and users of Sun’s open source database with significant expertise in maintaining and improving the software, and who could support a derivative version of it.

The fact that both Richard Stallman and Microsoft are lobbying against the acquisition (at least without divestiture of MySQL) is a sure sign that it is procompetitive.  Of course, this is also the same Commission that handed over Microsoft’s intellectual property to the likes of Oracle and MySQL in a claimed effort to bolster competition in server software.  Maybe it’s just engaging in a little more fine-tuning of its industrial policy now.

Posted in antitrust, markets, technology | 2 Comments »

Some Links

Posted by Josh Wright on November 9, 2009

  • Larry Ribstein on exempting small firms from SOX
  • Bernie Sanders’ “Too Big to Fail, Too Big to Exist” Bill (but see here)
  • More Professor Birdthistle on Jones v. Harris
  • Michael Ward on the economics of H1N1 (here, here and here)
  • Lots of blogging on the meat market — but I’ve seen nobody discuss what I thought was the most surprising event at the conference, i.e. the disappearance of Starbucks from the hotel

Posted in blogging, music | Comments Off

The CFPA's Effect on Consumer Credit and A Wager Proposal for Professor Levitin

Posted by David Evans on November 4, 2009

Professor Adam Levitin is not impressed by our prediction of the effect on consumer credit of the CFPA.  Readers might recall that, using estimates from the literature on the effect of regulatory shocks on interest rates and of the long-term debt elasticity, we offered a (in our words) “rough calculation” of the “lower bound” of the effect of the CFPA Act on consumer credit at 2.1%.  Professor Levitin says that we just “make up the numbers” and that they do not pass the “straight-faced test.”  In his paper (and second blog post) Professor Levitin offers more of the same formula: a combination of assertions unsupported by evidence, ad hominem attacks, and insistence to his prior assumption that the CFPA will reduce the cost of credit without imposing serious regulatory costs (again, without substantiation).  He writes that his real problem with our analysis is that “The key point here, however, is the impact of the legislation is speculative and certainly not susceptible to precise statistical predictions.”

We continue to be puzzled as to how a carefully qualified estimate of a lower bound could be confused with a precise statistical prediction.  Levitin’s response, unfortunately, does not address our lengthy explanation of how the CFPA will increase lender costs and continues to assert without evidence that failures of consumer protection caused the financial crisis.

More importantly, we’re puzzled by the apparent denial by supporters of the CFPA of the need to provide any evidence that its benefits exceed its costs.  In his blog post, Professor Levitin notes that he “didn’t set out to prove a positive case in the critique and don’t need to do so to make [his] central point” in critiquing our analysis.  That’s fine.  But the central point of the discussion ought to be the costs and benefits of the proposed regulation.  The burden lies with the proponents of this broad sweeping change in the consumer protection regulatory landscape to present some evidence in favor of their proposals.  Professor Levitin further argues that implying a concession from the absence of empirical support in favor of the CFPA as “ridiculous in this context.”  Astute readers may perceive a developing theme.  Professor Levitin and supporters of the CFPA Act have expressed indifference at best to the notion of providing empirical evidence concerning the potential benefits of the CFPA.  Instead, their general strategy has been to nakedly assert, as Professor Levitin does in his analysis, that failures of consumer protection led to the financial crisis and then proceed to offer regulatory proposals under the assumption that they will have benefits and their implementation will occur without cost.  For example, Professor Levitin predicts that issues involving inconsistent regulations between the states will be resolved out of court, that states will likely adopt consistent regulations, and that industry will (apparently costlessly) “conform with the strictest level of regulation.”  As we discuss in the paper, there is simply no evidence to suggest that failures of consumer protection caused the financial crisis.  None.  To reiterate, once again, there is absolutely no evidence whatsoever that  failures in consumer protection precipitated the current financial downturn.

So to the supporters of the CFPA Act, and particularly Professor Levitin, who has criticized our analysis for providing at least some evidence of the legislation’s costs, where is the evidence – any evidence – that the CFPA Act’s benefits will outweigh its costs?  Do you still support the “plain vanilla” provision despite the fact that both a government agency deigning to design financial products and imposing their offering by banks will necessarily involve significant costs?  What are the benefits to outweigh those costs?  What is the proof of those benefits?  Or, even without “plain vanilla,” where is the evidence to suggest any benefits that outweigh the obvious increase in lending costs (and in turn, interest rates) associated with the CFPA’s new regulatory scheme that we lay out in the paper?  We’d like to see two things: a coherent economic explanation of these benefits and some demonstration of empirical evidence suggesting they actually will materialize.  Can the CFPA’s supporters really think that the CFPA and its new regulatory landscape will not increase costs to lenders and reduce consumer credit?  We concede now, as we did in the original paper, that it is quite possible that some of the new proposals could produce benefits.  But as we said then, and repeat now, that discussion should be based on a careful analysis of the costs and benefits of the various proposals.

On to the wager proposal foreshadowed in the title to the post.

The CFPA Act’s supporters have fought vigorously for this piece of legislation.  Professor Levitin appears quite confident that our analysis represents a “scare statistic” meant to avoid rigorous cost-benefit analysis and to ignore precision.   Of course, we find this line of attack ironic in light of the complete absence of empirical evidence in favor of the CFPA Act mustered up by its supporters.  More generally, we’d like to offer Professor Levitin the opportunity to prove that he means what he says about our overestimate of the lower bound of the impact of the CFPA Act on consumer credit and about the beneficial effects of the CFPA Act more generally.  We are economists.  And so we also believe in the power of revealed preferences.  We stand by our estimate of the lower bound at 2.1 percent.  If Professor Levitin is correct that is a ‘scare statistic’ that we’ve inflated from the true number, we would like to provide an opportunity for Professor Levitin to profit from our misguided approach and to test whether he really believes that the effect on consumer credit will be smaller than that.

We propose the following wager to Professor Levitin:

If the effect on consumer credit is less than 2.1 percent, you win and we lose

If and when the CFPA Act is passed, there will be ample data to test the impact of the CFPA on consumer credit directly.  We’re happy to negotiate what methods should be used to calculate the number to both of our satisfaction.  We’re also happy to let you name the stakes.  But let’s make it interesting.  If it’s good enough for Mankiw and Krugman, it’s good enough for us.  What do you say?

Here’s the abstract:

The Consumer Financial Protection Agency Act (“CFPA Act”), introduced by the U.S. Department of the Treasury in June 2009, proposes sweeping regulation of consumer lending and borrowing. As we showed in “The Effect of the CFPA on Consumer Credit” (hereinafter “Evans and Wright (2009)”):

The CFPA Act creates massive litigation exposure for lenders facing (a) potential lawsuits from state and municipal governments for violating more stringent financial protection regulations that those entities can adopt pursuant to the CFPA Act; and (b) litigation under the CFPA Act’s new and undefined standards for engaging in unfair, deceptive, abusive, or unreasonable practices.

The new Agency would impose significant costs on lenders who would be required to: (a) offer to consumers on a preferred basis plain-vanilla products designed by the Agency either before offering their own products or at the same time; (b) seek prior regulatory approval for new lending products which could be defined as minor variations on existing products; (c) face the risk of having lending products banned altogether; and (d) have to comply with various other rules and regulations.

This note responds to a recent paper by Professor Adam Levitin offered in response to Evans and Wright (2009). As a prefatory matter, his paper is filled with various ad hominem attacks which we will ignore. Instead, we focus on the substance of the issues in contention. Professor Levitin’s basic substantive objection is that he disagrees with our estimates that the Treasury Department’s bill would increase interest rates by at least 160 basis points and reduce net job creation by 4.3 percent under plausible assumptions. Professor Levitin’s criticisms are misguided and we stand by those numbers as lower bounds on the effect of the Treasury’s CFPA Act on the economy. We also note that Professor Levitin has disputed virtually none of our findings that the CFPA Act would impose high costs on lenders and ultimately result in denying borrowers choice.

We think it is impossible to read the CFPA Act without concluding that lenders will face higher costs as a result of, among other things, dealing with the new Agency, being forced to offer products designed by a governmental body rather than themselves, coordinating the sale and distribution of financial products across regulatory regimes varying across the fifty states, and facing the increased possibility of fines and litigation under a novel and ambiguous “abusive” practices standard. While we believe there is a debate to be had on the costs and benefits of the CFPA Act, it is difficult to fathom a claim that this particular Act will not impose significant costs on lenders and that those costs will not be passed on to borrowers. Sound public policy should be based on a careful analysis of the costs and benefits of the various proposals. We do not believe Professor Levitin has made a constructive contribution to that deliberation but encourage him and others to do so as Congress considers the CFPA Act of 2009.

We encourage interested readers to take a look at our papers for themselves:

David S. Evans and Joshua D. Wright, The Effect of the Consumer Financial Protection Agency Act of 2009 on Consumer Credit

David S. Evans and Joshua D. Wright, A Response to Professor Levitin on the Effect of the Consumer Financial Protection Agency Act of 2009 on Consumer Credit

Posted in business, credit cards, economics, financial regulation, scholarship | 1 Comment »

As New York goes, so goes the FTC?

Posted by Geoffrey Manne on November 4, 2009

The New York Times is reporting that New York’s attorney general, Andrew Cuomo, has filed an antitrust suit against Intel.  According to the report,

The New York move increases the chances that the F.T.C. will take action against Intel, according to a person who was familiar with the state’s investigation but was not authorized to discuss it. Mr. Cuomo’s staff, this person said, regularly communicates and cooperates with the commission’s staff.

“These are separate investigations, but it would be very surprising for New York State to go off on its own without being fairly confident the F.T.C. would pursue Intel as well,” the person said.

The case is modeled on the European Commission’s case against Intel, itself modeled on the ongoing private suit brought by rival, AMD, in Delaware.

Among FTC watchers it is an open secret that the Commission is likely to bring a case against Intel soon, probably before the end of the year.  This just corroborates that intuition.

TOTM readers know that we have had some harsh things to say about the case against Intel.  See, for example, here, here and here.

Thom in particular has had some choice words for New York’s investigation:

So why is Attorney General Cuomo pursuing this European-style, competitor-focused antitrust inquiry? Perhaps he is, as he claims, just looking out for the interests of New Yorkers. But not in their status as computer buyers (where they’d be helped by aggressive price competition). Instead, he’s looking out for their prospects of employment and for his state’s coffers. It seems that Intel’s chief competitor, Advanced Micro Devices (AMD), has promised to build a $3 billion plant in upstate New York. Thus, it makes sense that Mr. Cuomo would jump on the bandwagon led by Senator Charles Schumer and Representative Kirsten Gillibrand (D-NY), both of whom have pushed for a federal inquiry into Intel’s loyalty discounts.

So what we have here is really a protectionist move that may benefit New York’s workers (who may land sweet AMD jobs) and politicians (who’ll have more tax revenue to spend), but at the expense of computer consumers generally. What’s more, other potential discounters will know they might face inquiries from aggressive states like New York, and they may therefore forego consumer-friendly discount arrangements.

It’s bad enough that American businesses have to worry about competitor-focused European regulators. We’re in real trouble if rogue states start acting like antitrust bullies.

The case is almost certainly a bad one.  As I said in the New York Times when the EU decision was announced, “Europe’s case is really predicated on the idea that there will be future harm to A.M.D. . . . It is really hard to find evidence of that now.”  The case is the poster child for intervention based on a decision uninformed by error cost analysis.  Intel’s is an innovative industry, and Intel’s contract practices are themselves innovative means to facilitate its extremely-high-fixed-cost business model.  Absent actual evidence of harm to consumers (and in the face of evidence of great benefit to consumers), the case is built on tenuous theoretical underpinnings with a strong risk and high cost of error.  As Josh and I discuss in our recent paper, this sort of enforcement decision is a recipe for disaster.

Posted in antitrust, federal trade commission, law and economics, technology | 2 Comments »

« Previous Entries
Next Entries »
 
Follow

Get every new post delivered to your Inbox.

Join 1,035 other followers