Streetwise Professor

In the aftermath of the financial crisis, CFTC head Gary Gensler believes that OTC derivatives markets need regulation in the worst way, and that he’s just the man to do it.  Gensler testified in front of Congress, and presented an outline of his ideas for regulation of the OTC markets.  It is now clear how Gensler was confirmed in his post, despite his involvement (salutary, in my view) in the (wrongly, in my view) widely excoriated Commodity Futures Modernization Act:  he agreed to push all the derivatives policy nostrums emanating from Capitol Hill.

There is enough here for several posts, and I’ll probably do just that, time and travel schedule permitting.  The Cliffs Note version is: I disagree with just about everything, and think that most of the arguments raised in the letter are wrong, when they’re not retarded.  (Can you say that anymore?)  Actually, that’s not quite right.  There are NO arguments in the letter to speak of.  Just a series of unfounded assertions on which Gensler erects a set of far-reaching policy proposals.  The Monty Python “Argument Sketch” goes to Washington.

Today I’ll focus on the issue that constitutes the most important part of Gensler’s letter: clearing.  It presents a cartoonish version of what clearing does, and grotesquely misstates the economic implications of clearing.  Gensler presents clearing as a panacea, a silver bullet, that will banish systemic risk.  Based on this distorted depiction of the costs (none) and benefits (incalculable) of clearing, Gensler recommends that virtually everything be cleared, and that non-cleared product be subjected to intense scrutiny.

Gee.  If it’s such a great idea, why didn’t anybody think of it before?  That’s the first clue–and it’s a big one–as to why Gensler’s advocacy (which, admittedly, just echoes Geithner’s and Congress’s) is almost certainly based on fantasy rather than fact.  

Gensler’s most grotesque distortion is to assert that bilateral markets are too interconnected, and that cleared markets would be less so:

One of the lessons that emerged from this recent crisis was that institutions were not  just “too big to fail,” but rather too interconnected as well.   By mandating the use of  central clearinghouses, institutions would become much less interconnected, mitigating  risk and increasing transparency.   Throughout this entire financial crisis, trades that  were carried out through regulated exchanges and clearinghouses continued to be  cleared and settled.  

Where do I start? First, clearinghouses CREATE interconnections, albeit different ones than in the OTC market; clearinghouses formalize a specific and extensive form of interconnection between systemically important market participants, and their customers. They are a means of SHARING default risks, and determining the allocation of losses from default. Sharing means a connection.  By definition.  The members of the clearinghouse share the risks of default. The clearinghouse members and their customers are therefore interconnected, and the failure of one has financial implications for the financial health of others. Moreover, the central clearing structure can actually lead to greater concentration of risk than occurs in the bilateral market, and result in less capital being available to absorb default losses.  That is, the particular interconnections in a cleared market can be more systemically threatening than those in a bilateral one.  It is false, misleading, and borders on the fraudulent for Gensler to claim that clearinghouses make markets “much less interconnected.” It is false, false, FALSE.  

Clearing changes the nature of the interconnection.  It is incumbent on advocates of the imposition of clearing to demonstrate that this change is an improvement.  Gensler doesn’t even begin to try to do so.  He just misleadingly asserts the issue out of existence.  Out of such sloppiness, bad policy is made.

The last sentence of the quoted paragraph is also misleading. First, other than the AIG trades (which I’ll get to in a minute), products OUTSIDE the regulated exchanges and clearinghouses also continued to be traded and settled. Lehman’s collapse did NOT lead to a cascade of failures, for instance. Second, the distribution of products traded on exchanges and in the OTC markets is not determined exogenously, so the comparison is not a fair one. Where things trade and the institutions of trading are endogenously determined by the actions of market participants. In my academic work, I argue that some products are suited for clearing, some not, and that trading patterns reasonably reflect those differences. Therefore, a comparison of the performance of cleared and non-cleared markets does hold everything equal. The stuff that is cleared is basically the easier stuff–hence you would expect fewer problems there.  

I’ve heard this argument about how well futures and options clearinghouses worked during the crisis several times now.  Every time I’ve heard it, those advancing it have failed to consider the possibility that this comparison is superficial and incomplete, and it ignores the fact that despite all of the hyperventilating, bilateral markets continued to work too.  

And for those with short memories, I suggest a look at the near death experience of the CBT and CME clearinghouses on 20 October, 1987, and the problems with COMEX clearing on Silver Tuesday in March 1981 (when the Hunts imploded) and during the mid-80s, to get a better understanding of the potential frailties of cleared markets, and how they can actually create systemic risks.  

I would also submit that many of Gensler’s proposals actually would lead to increased interconnectedness. For instance:

Central counterparties should also be required to have fair and open access criteria that allow any firm that meets objective, prudent standards to participate regardless of whether it is dealer or a trading firm. Additionally, central clearinghouses should implement rules that allow indirect participation in central clearing.

Extending the membership of clearinghouses by government fiat means that more firms are members of the clearinghouse–and hence interconnected to the failure of any other member. So much for reducing interconnection.

I would also note that Gensler completely overlooks that mandating open access to clearing will increase the heterogeneity of the membership of these organizations, which will have important implications for their governance. In my work on exchanges I note how heterogeneity greatly complicates governance, tends to impede decision making, leads to divisive political infighting, and on and on. (The paper I am presenting in Florence focuses on the effects of heterogeneity of the membership of financial cooperatives like exchanges and clearinghouses affects their governance.)  

Now, I have also shown in my academic work that membership limitations can also be used to turn the clearinghouse into a cartel. But nonetheless, it is very dangerous to blithely dictate the membership policies of such systemically important institutions without even the slightest consideration of how these diktats will influence governance–and how governance might affect policies relating to margining, capital, risk management, and systemic risk. Governance matters, and the heterogeneity of membership that Gensler’s recommendations would create would inevitably affect the politics of exchange governance, and the efficiency of clearinghouse operations.  Not that Gensler–or anybody else flogging this policy–even acknowledges the governance implications.

(I would further note that the possibility that the members of a clearinghouse can effectively use this institution to operate a cartel undermines arguments that anti-competitive motivations underly dealers’ refusals heretofore to clear many products. This provides further support for my argument that there are other costs associated with clearing, namely costs relating to asymmetric information, that outweigh the private benefits that clearing can provide the members of the clearinghouse.)

Gensler brings up AIG to justify his arguments. This, in my view, is the modern financial equivalent of arguendo ad Hitlerum: invoking AIG to justify any policy without a full and fair discussion of just how that proposed policy would have affected AIG’s behavior or the market, is irresponsible and usually reflects a lack of thought. It, like arguments ad Hitlerum, is an appeal to emotion rather than fact and analysis.

Sadly, Genlser’s invocation fits that characterization:

By novating contracts to a central clearinghouse coupled with effective risk management practices,  the failure of a single trader, like AIG, would no longer jeopardize all of the counterparties to its trades.

Wrong again, Mr. Gensler. This seems to suggest that a clearinghouse makes default risk Disappear!, like some amazing magician. No. It does not. It redistributes the risk. As I showed in It’s a Wonderful Life: AIG Edition, (a) it is highly unlikely the products AIG traded would have been cleared in any event, and (b) if they had, it is unlikely that clearing would have materially mitigated the implications of an AIG collapse, and indeed could have made it worse.  This is related to the interconnection point.  Once AIG blew up, the members of the clearinghouse would have been on the hook just as its bank counterparties were.  And guess what?  The members of the clearinghouse would likely have been the same bank counterparties.  

There is no acknowledgement in Gensler’s prepared testimony of even the possibility that clearing may be inappropriate for some products, and that requiring clearing of these products may result in greater costs than benefits.  Indeed, the fact that clearing has not succeeded for numerous products, and has not even been attempted for some, raises quite clearly (no pun intended) the possibility that the costs indeed exceed the benefits.

Think of it this way.  By recommending a mandate, Gensler (and others like Geithner who have taken similar positions) is asserting implicitly the existence of a market failure.  The key word is “asserting.”  He is not specifically identifying what that market failure is, and how his proposal corrects it.  Why, if clearing is so socially beneficial, has it not been adopted?  What market failures, what transactions costs, are preventing the adoption of this socially optimal arrangement?  If it is not possible to present a coherent, plausible, and evidence-backed case for the existence of such a market failure, the presumption should be that the market knows better than Mr. Gensler, and that market participants have adopted one set of institutional arrangements for some products and another set for other ones because costs and benefits differ across products and users.  Mr. Gensler apparently believes one size fits all.  The market has been acting as if it believes something very different.  Why is the market wrong?  It may be–but it is incumbent on Gensler and others to make the case.

This implicit assertion of market failure pervades the Gensler letter.  His recommendations to dictate how instruments are traded, with a decided preference for exchange trading, represent another example.  We see invocations of transparency and other Motherhood-and-Apple-Pie virtues to justify dictating market structure, with no serious discussion as to why market participants have apparently chosen suboptimal arrangements.  I will hopefully return to this part of the proposal soon.    

The Gensler testimony also suggests–though it is somewhat ambiguous on this score–that the government will establish margin requirements on all derivatives trades.  One could also read the letter to mean that it will require clearinghouses to set prudent margin requirements.  Either alternative is problematic to the extent that the government and the clearinghouse–but especially the former–are unlikely to have better information for setting these requirements for some products than some private market participants.  This is problematic–as is the possibility that a government regulator or a clearinghouse may be subject to political pressures that lead it to set margin requirements based on rent seeking considerations, rather than prudential ones.

I could go on, but it’s late.  Suffice it to say that Gensler is predicating a massive change in the regulation of financial markets based on a set of implicit assertions.  He is implicitly asserting the existence of a plethora of market failures.  Not one of which does he demonstrate by thorough argument or evidence, beyond the ritual incantation of the Ghost of AIG.  He does not even identify what the market failures are.  This is the stuff that public policy disasters are made of.   If there are market failures that need fixing, it should be straightforward to identify what they are, and to show how the proposed policy will improve the allocation of resources by reducing the severity of these market failures.  

But Gensler doesn’t even try to do that.  That shouldn’t be sufficient, but given that Gensler is primarily serving as Charlie McCarthy to Congress’s Edgar Bergen, sadly it will be more than enough.