Streetwise Professor

It seems that some in Congress are slowly making it down the clearing learning curve:

Manufacturers such as  Caterpillar Inc. and  Apple Inc. wouldn’t have to post extra cash as collateral for hedging transactions under legislation proposed to tighten oversight of the $592 trillion derivatives market.

The draft bill released by House Financial Services Committee Chairman  Barney Frank on Oct. 2, drawn from Obama administration proposals, would require the most common and actively traded over-the-counter derivatives contracts to be bought and sold on exchanges or processed through a regulated trading platform. It would impose new rules and collateral requirements, except on so-called end users such as Caterpillar.

. . . .

The draft would expand the administration’s exemption of end-users from collateral and clearing requirements. The administration proposal, released Aug. 11, would give a dispensation only to companies that use so-called hedge accounting. Frank’s proposal would exempt all end-user transactions and would let companies post non-cash items as collateral to satisfy margin requirements, Coleman said.

In brief, the new proposal would not require clearing for any deal in which one of the parties was not a swap dealer.  This is an improvement.  I’ve been arguing here on SWP, in my academic research, and my recent paper for Brookings that the cash flow and liquidity demands that clearing and daily mark-to-market impose are very onerous for hedgers and other market participants.  It seems that corporate end users were able to get that message through to Frank.  Good for them.  It also suggests that there may be a glint of understanding–albeit only a glint–that OTC market arrangements have some economic benefits, and are not instruments of the devil.

But it appears that dealers will be required to clear inter-dealer transactions.  This raises many questions, which if addressed forthrightly, would undermine the case for mandate.

Dealers certainly have had the ability to form inter-dealer clearinghouses in the past.  Why haven’t they?  The common story is that they like non-cleared deals because they profit at the expense of their customers from the lack of fungibility and liquidity.  But that argument holds no water when it comes to purely inter-dealer trades.

More fundamentally, a major fear relating to OTC markets is that the interconnections among dealers are systemically risky.  That is, because of all the inter-dealer trading, the default of one dealer affects the financial condition of others.

But, how does inter-dealer clearing change that?  Default losses are shared among dealers now because they trade with one another.  Default losses will be shared among the same dealers via a clearinghouse.

Moreover, the sharing mechanisms are different in a crucial way–that does not favor clearing.  In a bilateral market, if dealer X defaults, those who trade with X may suffer losses.  In a cleared market, if dealer X defaults, those who originally traded with it do not bear all the loss: it is shared among all the other members of the clearinghouse.  This reduces the incentive of market participants to monitor X’s creditworthiness, because they are not on the hook directly for the entire amount of counterparty exposure they create by trading with X.  The clearinghouse monitors, but as I’ve argued repeatedly (a) the clearinghouse is unlikely to have as good information as the other dealers, and (b) its incentives are weaker because it is a delegated monitor subject to low power incentives.

There’s another landmine hiding here that hasn’t drawn attention.  Note that the members of a clearinghouse recognize that they have delegated their risk exposure to somebody else: the clearinghouse and other clearinghouse members.  They recognize further that the clearinghouse mechanism thereby creates a moral hazard.  What is the conventional response to such a problem?  To limit recourse.  That is, clearinghouse members will limit the risk exposure that they will take on from the clearinghouse.  This is true of every major clearinghouse: the losses that members have to absorb are limited.

Put it all together, and you realize that (a) moral hazard and monitoring issues can lead to poorer control of counterparty risk, and (b) the limitations on member contributions means that a clearinghouse could well fail because of a loss that exceeds the amount of capital its members are obligated to commit.

Thus, the mandate for inter-dealer clearing doesn’t make any sense either.

In fact, one of the main virtues of clearing is that it increases the fraction of their contractually promised payments that non-members receive.  Note that the Board of Trade Clearing Corporation always used to say that no CUSTOMER had ever suffered from a default loss, not that no member had ever suffered from such a loss.  Inter-dealer clearing slices and dices risks differently; it does NOT eliminate them.  Indeed, it can (due to the moral hazard and adverse selection problems) lead to an accumulation of greater risks, and a less efficient allocation of those risks, than is the case in a bilateral market.

In other clearing news, CFTC Chairman Gensler made some interesting remarks in an interview with Futures Industry Magazine (registration required).  Interesting, but misguided.

First, he comes down to requiring clearinghouses take deals from any execution platform: he says clearinghouses should “not be so inextricably linked to an exchange.”  As I’ve argued for several years on SWP and in my academic writings, basic transactions costs considerations likely make vertical integration of execution and clearing economically efficient.  The argument that this arrangement impedes competition (which Gensler makes) fails on traditional Chicago grounds.  It is certain that there is neither persuasive theory nor convincing evidence that would support as radical a step as Gensler advocates.

Second, Gensler advocates that clearinghouses should be open access in the sense that “membership is open to the non-dealer community” and that clearinghouses should have “open governance” in which non-dealers would have a voice.

Now, I’ve shown in some academic writing that closed membership can be exploited to permit the clearinghouse to in effect cartelize a market, so I am somewhat sympathetic to the idea of open access.  However, I’ve also shown in my academic work that membership heterogeneity can be very, very costly.  It increases the costs of governance because it is necessary to accommodate more diverse preferences and interests.  Moreover, heterogeneity means that it is possible to redistribute wealth among the different members through pricing policies, and in the case of a clearinghouse, margin and capital requirements.  To mitigate rent seeking, cooperative organizations like a clearinghouse with disparate members will (a) implement very low-powered incentives, and (b) erect intricate, formal, and cumbersome governance mechanisms (e.g., committee-dominated decision making).  These arrangements are costly.  Thus, requiring open access and open governance would have unintended consequences: clearing organizations would be sluggish organizations with weakly-incentivized managements.

As always, there are trade-offs.  The question is: who is best situated to evaluate these trade-offs?  I think that it is far more efficient for government to let market participants decide on their own arrangements, and limit its interventions to situations in which these cooperative arrangements stifle competition.  Government is ill-suited to evaluate the costs and benefits of alternative ways of sharing risk–including counterparty risk–and of organizing and managing specialized organizations like clearinghouses.  Barney Frank and Gary Gensler should leave these tasks to people who have the expertise and internalize the costs and benefits, and limit the government to policing anti-competitive behavior by collective organizations like clearinghouses.  And even there, a considerable amount of circumspection is in order, for as Ronald Coase pointed out long ago, not every non-standard form of organization or arrangement is necessarily anti-competitive in nature.  It may be designed instead to economize on transactions costs.

At least Barney Frank has made a step towards understanding the intricacies of something as complex as clearing.  Would it be too much to hope that he keeps pulling on the string until the entire case for clearing mandates unravels?  Probably.