Streetwise Professor

January 28, 2011

No Cigar, John

Filed under: Commodities,Derivatives,Economics,Energy,Exchanges,Financial crisis,Politics — The Professor @ 2:31 pm

Writing in Commodities Now, Reuters reporter John Kemp takes issue with my criticism of the CFTC proposal on position limits. I appreciate the attention, but I’m underwhelmed by his arguments.

There are a lot of details, but the gravamen of his complaint is that my view on the purpose of position limits is far too narrow:

Pirrong defines the rationale for intervention too narrowly. Large leveraged positions run by the Hunt Brothers did indeed threaten to bring down their own brokers and several other financial institutions. But systemic risk is not the only reason for regulators to intervene to prevent market manipulation.

Kemp directs his beef at the wrong guy.  His real problem is with the CFTC because these were the only–I repeat only–reasons the CFTC set out it its NOPR.  My critique of the proposal is that it was a vastly overinclusive way to achieve the objectives the CFTC set out.  Again, the objectives the CFTC set out.  I was responding to what was in the proposal, but I did mention quite pointedly that the CFTC was utterly silent on whether speculation was culpable in the mid-2000s price spikes, or those going on now.  How is it possible to respond to arguments and assertions that are not made?

Since it was the CFTC that was setting out the rationale for its proposed intervention, if that rationale is too narrow, Kemp should direct his ire where it belongs–the CFTC–not where it doesn’t–yours truly.

Kemp also critiques my views on manipulation and the proper way to deter it.  Oh no!  Please don’t throw me in the manipulation briar patch, Brer Fox!

Actually, born and bred in the briar patch, John.  We are agreed that past efforts to deter manipulation through ex post enforcement have been troubled, at best.  In this, Kemp is no doubt late to the game, because I’ve been making that argument since the mid-1990s.

That said, I’ve noted that manipulation is not inherently “unprosecutable” (a word Kemp lifts from Jerry Markham).   The problems with detecting, deterring, and punishing manipulation can be corrected by a proper understanding of the economics, and a judicious use of, yes, econometrics–something that apparently sends Kemp into apoplexy.

Moreover, I have made very concrete recommendations as to how manipulation law can be reformed to address the problems it has experienced.  (Not that you would know that from reading Kemp’s description of my views and arguments.)  The problem is that instead of taking a new, economics-based approach to the problem, Congress keeps returning to the vapory SEC 10(b)-5 “manipulative contrivance or device” standard, like a dog returning to its own vomit.  Doing the same thing over and over again and expecting different results is insane.  I’ve recommended specific and practical ways to stop the insanity.

As an aside, Kemp laments that manipulators have only been subject to civil, rather than criminal, sanctions.  There are a couple of issues here.  The first is that criminal sanction–imprisonment–is socially inefficient when defendants are not judgment proof.  The second is that Kemp’s lament actually supports the point I made in the previous paragraph.  Judge Miller in Houston found the existing anti-manipulation statutes unconstitutionally vague, and threw out criminal indictments against BP propane traders.  The suggestions as to how to correct the law would go a long way to addressing that vagueness problem that impedes criminal prosecution.

Speaking of econometrics, here Kemp is extremely misleading, and puts words into my mouth that I did not utter, and which do I endorse.  He says that I propose that the CFTC not “set limits or commence enforcement proceedings unless econometric tests showed an unequivocal effect” [emphasis added].

If there is anything unequivocal, it is that his statement is unequivocally false, for I never said any such thing.  Indeed, insofar as manipulation is concerned, he turns reality on its head.  For what I have shown in numerous articles, a book, and in testimony, is that econometric evidence and economic analysis generally is the best way of proving manipulation, rather than being an obstacle that makes prosecution of manipulators virtually impossible.  It is a sword that can be wielded against manipulators, not  a shied that defends them: although it is effective as a shield protecting the innocent.  Rigorous analysis of data is the best way to cut through trader BS, and prosecutor ignorance.

And that’s the point: we want methods that avoid false positives and false negatives.  My work shows that econometrics, when combined with rigorous economic reasoning and other economic tests that are not econometric per se, can do just that.

Before characterizing my views, John, you might want to be more familiar with what they actually are.

Even with respect to the issue of whether speculation has distorted prices in ways other than the Hunt-type scenario, I have been quite open about the basic limits of econometric analysis.  Here’s what I wrote in a piece in Regulation last summer:

One of the most important functions of markets is to discover prices; to provide a mechanism whereby information about supply and demand dispersed among millions can be aggregated.  Thus, no one has the information to know what the price “should” be; if someone did, we wouldn’t need markets.  But this always makes it very difficult to rebut assertions that a particular market price is distorted by speculation, or to prove that they were.

Moreover, in that article, I use some econometrics, but only very little.  I instead utilize some non-econometric approaches to test the implications of the speculative distortion story.  Thus, to attribute to me a slavish and myopic devotion to econometrics is just wrong.  I have a very catholic view of what kind of evidence is persuasive.

But I am a firm believer that data and evidence should be decisive.  Kemp argues that “policymakers must act on the balance of probabilities.”  Fine.  But just where, pray tell, are those probabilities to come from?  Kemp bashes econometrics, but advances no alternative.  Zero.  Zip.  Nada .

Sorry, but that just doesn’t cut it.  Econometrics is one tool that can be used to inform judgments about probabilities.  There are other tools–qualitative comparisons, and rigorous analysis of the economic logic of purported causal connections between speculation and price distortions, for instance–that can also inform those judgments.  The best approach integrates statistics, econometrics, qualitative comparisons and rigorous economic reasoning.  That’s the kind of approach I favor, and the kind of approach I’ve used–though you would never guess that in a million years if your only impression of my work is Kemp’s description thereof.

In the end, Kemp takes refuge in the position that as a matter of law, Congress has required the imposition of position limits, and that’s the end of that.  He worries that the CFTC could be sued if it failed to adopt such limits.

But even that legalistic defense is incomplete.  Congress has mandated the imposition of limits to achieve a specific purpose: the prevention and diminution of “unwarranted fluctuations” in prices.  Limits that are unnecessary to achieve that goal–because such unwarranted fluctuations are extremely unlikely to occur–or limits that are so overinclusive that they impede the achievement of other Congressionally mandated objectives, shouldn’t hide behind the fact that courts may find the exercise of such powers Constitutional.  “We do it because we can” is a recipe for regulatory mischief that undermines the very goals that Kemp claims to endorse.

My criticism of the CFTC NOPR is that it (a) identified a way in which large positions could cause unwarranted fluctuations, and (b) imposed limits that were completely unnecessary to address the very problem the CFTC identified.  Maybe Kemp has different theories regarding (a).  If so, he should spell them out, and specify his proposed remedies so that they can be evaluated and critiqued in order to ensure that they are indeed problems and the remedies are an efficient and effective way to address them.

In the end, I guess I do agree with Kemp on one thing: it comes down to an issue of the burden of proof.  I believe in markets more than regulators.  He apparently takes the opposite view.  I am more than willing to debate this anytime, anywhere.

But contrary to what Kemp insinuates, I do not believe that markets are unerring, that market power is not a problem, and that certain laws and regulations can improve the performance of markets.  I am not some sort of Dr. No, insisting upon unreasonable level of empirical support to reject the hypothesis that the Market Is Always Right.  (Indeed, a former colleague, upon reading my work on manipulation, referred to me as a “Chicago apostate.”)  I have a presumption in favor of markets, but it is a rebuttable presumption.  And with respect to manipulation in particular, I have both rebutted the widespread view that manipulation is not a problem, and recommended specific, targeted legal responses to mitigate that problem.

When it comes to speculation and position limits, however, those who have questioned markets have come no where near to rebutting the presumption in favor of markets.  I note–indeed emphasize–again that the CFTC does not even try beyond the specific areas identified in the NOPR: areas which Kemp believes are too narrow.  But Kemp hasn’t provided a credible alternative either.

Given the havoc that bad regulations routinely inflict on markets, a credible story and at least some evidence should be required to support a particular policy proposal.  A fair acknowledgement of the potential for unintended consequences is also appropriate.   I note too that Kemp does not even attempt to critique the unintended consequences of the CFTC proposal that I lay out in my comments.

When it comes to the specifics, Kemp ironically endorses my view–although he does not acknowledge this.  He says:

Reversing the burden of proof, it is worth asking opponents why anyone would want to run positions above the very high levels set out in the CFTC’s proposals. Are there legitimate reasons to want to run very large net positions, including in the spot month? Or is the only reason to exercise market power?

Uhm, that’s very close to what I recommend.  Give the CFTC the authority to question the legitimacy of very large positions, and intervene accordingly if there is no legitimate reason.  But that’s a far, far, different thing from a blanket prohibition of positions in excess of a particular size.

In sum, I stand by my specific criticisms of the CFTC position limit proposal.  Moreover, I emphasize that Kemp has presented a distorted view of my arguments, beliefs, and work.  His arms must be tired from giving a straw man such a thrashing.

But I’m still standing.

Print Friendly, PDF & Email

No Comments »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress