Streetwise Professor

I have always admired Thomas Sowell greatly.  I read his Knowledge and Decisions during my first year of graduate school, and it had a profound effect on my thinking and way of looking at the world.  I also read most of his work on discrimination, and found his book on Marxism to be a classic in intellectual history: a truly fair and penetrating appraisal of Marx, and Marxists (the two things aren’t the same, which is one of the points of the book).  Conflict of Visions is a classic: there is no better book for understanding the intellectual roots behind the political divisions in the West.  Yes, Sowell has strong views, but he is like John Stuart Mill in his ability to characterize his opponent’s views fairly and objectively.

Sowell has also been the subject of some the the ugliest personal attacks directed against any scholar over the last 40 years.  For the progressive left, to be black and on the right is to be an anathema who must be destroyed.  Yet he has handled the vicious attacks on him with equanimity and humor, and has never been cowed.

It is therefore beyond humbling to have Dr. Sowell come to my defense, and take on the New York Times “hit piece.”

All I can say is that I am honored beyond words.  Thank you, Dr. Sowell.

The NYT hatchet job has elicited quite a bit of discussion on the web, and on Twitter.  Several major bloggers have saddled up and ridden to the defense of Scott Irwin and I.  Pride of place goes to Felix Salmon, who wrote a long and thoughtful post at Reuters. This is particularly gratifying, because Felix and I have had our differences in the past (e.g., over the empty creditor problem) but I have always found him to be extraordinarily fair and open minded.  Felix wrote his piece before I posted mine, meaning that his was based on publicly available information (including his past interactions with me) which confers a considerable degree of objectivity: most importantly, it shows that information that the NYT ignored or distorted was there in plain sight.  Felix is like a judge entering a summary judgment motion against the NYT, with prejudice, making it unnecessary for the defense even to present its case.  I think it’s correct to say Felix’s verdict is pretty brutal, but eminently fair.

Other members of the cavalry include Jeff Carter’s Points and Figures blog (who was the first to arrive on the scene), Peter Klein at Organizations and Markets, Sonic Charm (or is it Crimsonic?) at Rhymes With Cars and Girls (who always cracks me up, whatever he’s calling himself these days), and John Hinderaker at Powerline. Peter and John point out the double standard under which academics who work with private business are often accused of being corrupt, whereas the motives of those who accept money from the government are almost never questioned.   Peter focuses on the Fed, John on climate research.  The common conclusion of all is that the NYT piece was a deeply biased smear written in bad faith with ulterior motives.  Crimsonic calls for a retraction.  Interesting idea.

Felix’s post and mine also unleashed a torrent of tweets.

Thanks to all.  I’m sure this isn’t over yet.  So keep your horses watered and your powder dry.

A few days ago the WSJ reported that US securities exchanges are close to upgrading one of the Securities Information Processors, the part of the RegNMS market structure that links the various exchanges.  The SIP consolidates and disseminates quote data from the various exchanges.  It provides the linkages that are necessary for the links-and-information structure created by RegNMS.

The SIPs have been a source of chronic complaints in the industry.  The current fix was compelled by a breakdown in the SIP over the summer that caused a shutdown in trading of some stocks.  But even when the SIPs work, they disseminate information more slowly than the private connections between exchanges and HFT firms.  This gives an advantage to HFT traders.

The WSJ story indicates that the operating system of the SIP is obsolete.

This issue illustrates an inherent problem with a utility-type model in securities markets. I discussed this problem in my research in the late-90s and early-00s, including in a JLEO article that was published in 2002.  That research was focused on a CLOB model, but the same issue is relevant for a the current SLOB (Simulacrum of a Limit Order Book) model. Specifically, a utility CLOB (or SLOB) has strong public good aspects.  All market participants benefit from better technology and improved performance, but would prefer that others pay for it.  Collective action and public goods problems tend to lead to under provision of quality and technological improvement.

SIPs are industry public goods, and face the same problems.  No wonder they are obsolescent.

There are other problems with the linkages approach that undermine incentives to  invest in SIP quality and technology.  The whole idea behind the linkages approach is to facilitate competition between exchanges.  But exchanges have no incentive to facilitate competition between themselves.  Competition is likely to be weaker with a kludgy linkage system, so why should exchanges push to improve it?  Not to mention that some interests, notably HFT, which would face a diminished competitive position with  more advanced SIP, also have an incentive to impede updates.

This is an inherent challenge to the linkages approach.  Its effectiveness depends on, well, the technology that links markets.  But the incentives to invest in those linkages are suboptimal due to the public good problems, exacerbated by the fact that competing exchanges don’t have an incentive to intensify competition between them.

But as I’ve often written, the alternatives have their own problems.  Most notably, if you don’t mandate linkages between markets, order flow network effects tend to cause the dominance of a single exchange.   So you need linkages and socialization of order flow if you want competition, but there will be underinvestment in the linkages.

US securities markets are committed to the linkages approach.  Public goods problems that provide inefficiently weak incentives to invest in the links necessary to make the approach work mean that it will be beset by chronic difficulties.

Matt Leising had a nice article a few days back about the CFTC’s rule that does not treat US Treasury securities as “qualifying” liquid resources for CCPs.  Instead, under new regulation 33-33 they must obtain “prearranged and highly reliable funding.” Based on Fed rules, this means that a CCP must get  committed line of credit from banks.   This imposes a substantial cost on CCPs, because under new Basel III rules, committed lines impose a large capital charge on the issuing banks.  For purposes of calculating capital, the banks have to assume that the lines are fully drawn.  This capital cost will be passed onto CCPs.

It is ironic that outgoing (resisting the great urge to snark) Chairman Gary Gensler repeatedly argued that one of the main benefits of the Frankendodd clearing mandate is that it would reduce the interconnectedness of the financial markets, especially interconnectedness through derivatives contracts.  Now he has pushed through a regulation that mandates an interconnection among major financial institutions via a derivatives channel: the lines connect derivatives CCPs to major banks.   I have long pointed out that Gensler’s claim that clearing would reduce interconnectedness was grossly exaggerated, and arguably deceptive.  Instead, I pointed out that the mandate would reconfigure-and is reconfiguring-the topology of the network of connections between financial firms.  What the CFTC has done is dictate what that form of interconnection will be.  This particular dictate is extremely problematic.

A CCP needs access to liquidity in the event of a default of a clearing member.  The CCP needs to pay obligations to the winning side of the market, in cash, in a very tight time window.  Failing to make these variation margin payments could impose financial distress on those expecting the cash inflow, and more disturbingly, call into question the solvency of the clearinghouse.  This could spark a run in which parties try to close positions in order to reduce exposure to the CCP.  Given that this is likely to occur in highly unsettled market conditions, such fire sales (and purchases) will inevitably inject substantial additional volatility into price that can exacerbate pressures on the clearing mechanism.

A CCP holding Treasuries posted as IM by the defaulting CM can sell them to raise the cash.  Alternatively, it could repo them out.  During most periods of financial turbulence-and financial crisis-which is likely to be either the cause or effect of the default of one or more large CMs, there is a “flight to quality” and Treasury security prices rise and there is a rush to buy them by investors seeking a safe haven.  Moreover, under such circumstances the Fed will perform its lender of last resort function, and readily accept Treasuries as collateral: even if CCPs could not access the Fed directly*, they could access it indirectly.   Thus, in “normal” crises, Treasuries should be highly liquid, and a ready source of cash that can be used to meet variation margin obligations.

Put differently, from a liquidity perspective, Treasuries are a negative beta asset: they become more liquid when overall market liquidity declines-or verges on collapse.  This is a highly desirable attribute.  Another way to characterize it is that from a liquidity perspective, Treasuries have right way risk.

Bank lines are very different.  Banks become stressed during crisis situations, and face a higher risk of being unable to perform on credit lines under these circumstances.  (Indeed, what if the defaulter is one of the suppliers of a committed line?) Banks fighting for survival but which can perform might try to evade this performance during stressed market conditions, which in a tightly coupled system (and clearing is a source of tight coupling) can be extremely disruptive: a few minutes delay in performing could cause a huge problem.  And if the banks do perform, doing so poses the substantial risk of increasing their risk of financial distress.  That is, committed lines are positive beta from a liquidity perspective: that is, they pose wrong way risks.   If drawn upon, these lines can be an interconnection that is a source of contagion from a derivatives default to systemically important banks, precisely at the time that they are least able to withstand the shock.

In the event, a CCP that does collect Treasuries as IM can likely use these right way assets to raise the cash need to meet its obligations, and can avoid drawing down on its committed line.  But that would mean that the committed line is superfluous, and imposes unnecessary costs on the CCP, and hence on the users of the clearing system.

I also conjecture that having met its liquidity requirements with a committed line, pursuant to the CFTC reg, CCPs would  have a weaker incentive to take Treasuries as collateral, and a stronger incentive to permit the posting of lower quality assets (or incentivizing such posting by reducing haircuts assessed to such collateral) for IM. This would mitigate the cost impact to users that results from the CCP having to secure the committed line, and pay for it (the cost being passed onto the users), thereby reducing the loss of trading/clearing volume and the associated revenues.  This would increase the odds that the line will be drawn on (because the lower quality assets pose a substantial risk of becoming illiquid during a crisis situation-they embed wrong way risk too).  I’ll have to think this through more, because the situation is somewhat complex: it depends on the pricing of the line, which will depend on the likelihood it will be drawn against, and the market conditions at the time it is.  This will depend in part on the quality of collateral that the CCP collects.  I’m not sure of what the equilibrium outcome will be, but I suspect that mandating the obtaining of lines will undermine incentives to demand the posting of high quality collateral.  If it does, this is a bad outcome that increases wrong way and systemic risks.  If it doesn’t, then the cost of the lines is superfluous and a burden on clearing and derivatives trading.

There is one scenario in which Treasuries would not be good collateral: if the financial crisis (and default of a CM or CMs) was the result of a fiscal crisis in the US, or a default (real or technical)  of the kind feared during the last (but the last, most likely) debt ceiling standoff.  But that’s an Armageddon scenario in which banks are likely to be highly stressed and unable to perform, or in which they would incur exceptional and arguably existential costs if they did.  Put differently, there’s likely no good source of liquidity in this scenario, and the CFTC rule will hardly make a difference.

In sum, it is highly unlikely that bank lines are a better source of liquidity, especially under crisis situations, than Treasuries.  Indeed, they are plausibly worse, and actually create an interconnection that can transmit a shock to the derivatives market (and the CCP that clears it) to systemically important banks: this is the exact opposite of what clearing was supposed to achieve. The cost of the lines, which is likely to be substantial, particularly given their necessary size, is a deadweight burden on the markets: all pain, no gain.

Other than that, the rule is great.  And a fitting parting shot from Gensler.

* Frankendodd makes it difficult for US CCPs to obtain Fed liquidity support.  This is a serious mistake that could come back to haunt us in some future crisis.  To work effectively, the LOLR must be able to direct liquidity to where it’s needed,  quickly and efficiently.  CCPs could be a major source of liquidity demand in future crises, which makes isolating them from the Fed highly dangerous, and the invitation to an ad hoc response in some future crisis.

The CFTC has released its new position limits proposed rule.  It is an improvement on the earlier version, but still lacks a serious logical and empirical basis, try as the agency might to muster up a justification.

It is an improvement primarily for two reasons.  First, it has more sensible aggregation standards.  Whereas in the 2011 proposed rule, an entity would have to aggregate positions in every entity of which it had a 10 percent or greater stake in order to determine its overall position for the purpose of determining compliance with the limit, now that threshold has been increased to a more reasonable 50 percent.

Second, the new proposal does not include “class limits”.  Under the 2011 rule, swaps and futures were treated as different classes, and long positions in one class were not offset against economically equivalent short positions in the other.  Thus, a firm could have a zero economic exposure because a swap position was offset by an equivalent futures position, but it could be in violation of the position limit.  This made no economic sense, and was likely a surreptitious way of hamstringing swap dealers’ ability to enter into transactions with the massive passives who bother Chilton (and Gensler) so much.  Along with many others, I submitted comments criticizing this approach, and lo and behold, it has disappeared.  Now position limits are based on net exposures, meaning that a swap dealer can hedge a swap entered into with a customer using futures, in any quantity, without falling afoul of the position limit rule.

Unfortunately, not all nonsensical features of the 2011 proposal have vanished along with the class limits.  Most notably, the 2013 proposal not only includes, but extends the scope of, the “conditional limit” on positions held during the last 5 days of the a contract month.  Under this provision, the limit on delivery-settled contracts during the last 5 days is 25 percent of the deliverable supply, but the limit on otherwise identical cash-settled contracts is 5 times larger (i.e., 125 percent of deliverable supply).

As I commented in 2011-based on a paper I published in the Journal of Business in 2000-this makes no economic sense.  A trader that is long a delivery settled contract can manipulate by standing on excessive deliveries, thereby effectively buying up too much of the deliverable supply.  But a trader that is long a cash-settled contract can have the same price impact by buying the same quantity in the cash market.  Under the conditional limit rule, the holder of the cash-settled contract can get 5 times the benefit from this price impact as the holder of a delivery-settled one. The logic of the 25 percent of deliverable supply limit implies that the CFTC believes that taking delivery of something less than 25 percent of deliverable supply permits the long with a futures position of 25 percent of deliverable supply to manipulate profitably.  This means that the holder of a cash-settled position equal to 125 percent of deliverable supply can manipulate far more profitably.

This rule is passing strange especially considering that the CFTC has filed a manipulation claim against a trading firm-Parnon-that allegedly used cash market purchases to distort futures prices.  The strategy that the CFTC finds illegal in Parnon would benefit both cash-settled or delivery-settled contracts.  Under the conditional limit rule, the agency provides an incentive for the holders of cash-settled contracts to engage in such strategies.

What’s more, whereas the 2011 rule would have potentially constrained such strategies by limiting ownership of physical supplies by the holder of a cash-settled derivatives position to 25 percent of deliverable stocks, the new rule imposes no such constraint.  Instead, it imposes a requirement to report cash market holdings, and hopes that market oversight officials at exchanges will intervene if it looks like a manipulation is in progress.

None of this makes economic sense, and is in fact logically contradictory.  What makes it worse is that whereas in the 2011 proposal the conditional limit applied only to natural gas futures, it now applies to all contracts subject to position limits.

A big chunk of the proposal is a long-winded justification of the rule, in an effort to overcome the legal challenges that saw a federal court reject the 2011 proposal.  It seems to me that the commission is merely restating the arguments that failed in court, so I don’t really see how this helps.

A couple of things struck me as quite interesting.  In both the 2011 and 2013 proposals, the CFTC utilized the Hunt silver case and the Amaranth case as examples of the kind of destabilizing conduct that the rule would prevent. But whereas the 2011 rule argued that large leveraged positions (like those held by the Hunts) posed a threat to market stability, the word “levered” does not appear in the 2013 proposal’s discussion of the kinds of positions that are particularly worrisome.

This is interesting to me, in part, because my criticism of the 2011 rule as over-inclusive focused on the leverage issue.  If leverage is the problem, I argued, why impose the rule on unlevered positions, like those held by ETFs (which are fully collateralized) or real money investors? If leveraged positions were the real problem, the rule was over inclusive because it would apply to unlevered positions too.  Arguably, this argument had some impact, as all reference to the unique dangers posed by leveraged positions disappeared between 2011 and 2013.

But dropping the argument doesn’t eliminate the problem.  The CFTC has not shown how large, unlevered positions with trading decisions made by many individuals (as is the case with ETFs like USO or USNG) pose a threat to market stability in the way the Hunts did, or can cause unwarranted fluctuations in commodity prices.

Relatedly, the 2011 rule argued that position limits would reduce systemic risk.  I commented that it is wildly implausible that even a disorderly liquidation of a large position in a small market could destabilize the financial system.  Would a disorderly liquidation in the milk futures market really pose a threat to world financial stability?  Seriously?

And wouldn’t you know, the 2013 justification does not mention systemic risk.

More generally, the two-two, mind you-cases that the CFTC puts forth to justify its imposition of limits is highly unpersuasive.  I regularly use the Hunt case as an example of how speculation can distort prices.  Or, should I say, the example.  It is pretty much the exception that proves the rule. It occurred over 33 years ago.

Talk about ancient history.  When do we get to the part about the Trojan War?

The commission’s analysis, such as it is, of the Amaranth episode relies heavily on a report from the Senate Subcommittee on Investigations.  This report in no way demonstrates that Amaranth distorted prices, or that the liquidation of its position when it ran into financial trouble caused unwarranted price fluctuations.

A more plausible story is that Amaranth took a huge bet on natural gas spreads based in large part on a view that 2006 would be another big hurricane season.  When September came with no hurricanes, and none in prospect, spreads collapsed and Amaranth lost a lot of money.  The loss does not demonstrate that Amaranth distorted spreads, just that it took a big bet on these spreads, and bet wrong.

It’s not the job of the regulators to prevent market participants from losing big, if those big bets do not distort prices, either when they are initiated, or when they are liquidated.  (In the event, Amaranth’s positions were assumed-quite profitably-by JP Morgan and Citadel, with no notable dislocation in market prices.)

Moreover, even taking the commission’s view, the two episodes do not support the level of limits it proposes to impose.  The 2013 limits (like those proposed in 2013) would limit an individual firm to between 2.5 percent and 10 percent of open interest: the bigger the contract, the smaller the limit as a fraction of open interest.

But the Hunts and Amaranth had around 40-50 percent of the open interest in big contracts-at least 4 times, and arguably close to 20 times the limits the commission has proposed.  So even if you could argue that a 40 percent position could distort prices, or result in unwarranted price fluctuations when the position is liquidated, that doesn’t imply that a 10 percent or 5 percent or 2.5 percent of open interest position would.  There is thus a huge gap between the sizes of the positions that the CFTC claims, based on historical experience, distort markets, and the sizes of the limits it is proposing.

This is related, in a way, to cost-benefit analysis.  The agency is required to compare the costs and benefits of the regulation, but its analysis is perfunctory, at best. In particular, it does not provide a persuasive cost-benefit analysis of any major attribute of the rule, most notably of the size of the limits.  It basically adopts a precautionary principle approach: big positions could cause big problems, so they should be constrained as a precaution.  It asserts-but does not show-that liquidity will not be unduly restricted as a result.  This is insufficient.

Chilton-who has announced his departure-has claimed that the new proposal is immune to legal challenge.  I wouldn’t be so sure.  Even if the court accepts the verbose arguments relating to the commission’s obligation-or not-to make a finding that the limits are necessary, the agency is still vulnerable on the adequacy of its cost-benefit analysis.  Judge Wilkins never ruled on this issue in his 2012 decision, and in my opinion the CFTC is still vulnerable here.

It is not just the cost-benefit trade-off of the rule overall, but the specific choices made.  Most notably, the size of the limits needs to be justified better on a cost-benefit basis.  Similarly, the conditional limit needs to be so justified: and I’ll say that it is impossible-impossible-to do so, if one is constrained to using, you know, rigorous economics.

The rule is unbelievably long.  Five-hundred thirty three double spaced pages.  533.  (395 pages excluding appendices and attachments.) I have read most of it, but have to spend more time digesting the bona fide hedging portion of the rule.  More on that later.

Oh.  There are 846 footnotes: this is where the land mines lurk.  And there was much rejoicing.  By the lawyers, anyways.

I’ve been beavering away at the clearing section of my next book, which will be titled Market Macrostructure.  It’s been something of a struggle, because there are so many aspects to this issue that it is challenging to organize the material in a logical fashion.  I analogize it to trying to write a history of a complicated battle, where many things are happening simultaneously over space, and these things interact.  Inevitably, the narrative must jump around in either space or time or both, and the writer must summarize some material about one action at one time to relate it to the narrative of what is going on at another place at another time.

So it is with writing an analysis of clearing.  There are many moving parts that interrelate and interact, so there is always the challenge of relating detailed analyses of important pieces to one another, and to clearing as a whole, and to the trading process (execution and clearing) as a whole.

One virtue of this struggle, however, is that it can lead  the writer to new insights.  Conceptualization at a fairly highly level facilitates organization, and the process of conceptualization can lead to new ways of understanding.  So it is, I hope anyways, with the clearing analysis.

This post is my first attempt to crystalize those thoughts; indeed, one of the original purposes of the the blog was to provide a place where I could think out loud and commit to pixels some early thoughts to be refined going forward in more formal writing.

The issue that I have been grappling with is why are some trades cleared, and others not?  Why do we see exchanges that trade cleared contracts operate side by side with bilateral markets trading similar contracts (and sometimes nearly identical ones) that aren’t cleared?  What determines the division of trade between these alternatives?

This is an issue that I’ve been looking at since the ’90s, and I identified some factors, but I was never completely satisfied.  But putting together some pieces that I have written about before, I think I’ve come up with a more complete explanation that captures some of the salient aspects of the economics of clearing and counterparty risk generally.

The first piece is that in theory-and indeed, in most theoretical treatments of clearing by academics (including yours truly)-clearing can operate as a classical risk pooling/insurance mechanism, in which market participants pool counterparty risk.  To the extent that this risk is idiosyncratic, such pooling allocates risk more efficiently and makes risk-averse participants better off.

But as I pointed out in my earliest work, and emphasized more in some later papers, like any risk sharing arrangement, mutualization of counterparty risk creates the potential for adverse selection and moral hazard problems.  Moral hazard problems are likely to be particularly acute.  Clearing participants can affect the distribution of the default losses they impose on the mutualization pool by adjusting the riskiness of their trading positions in the cleared derivatives.  They can also do so by adjusting the risks of their balance sheets, through, for instance, adjusting their trades in non-cleared derivatives (or in derivatives cleared at another CCP), changing their leverage, or adjusting the risk of other assets on their balance sheets

The prospect for moral hazard will inevitably lead to limits on the amount of insurance provided through the clearing mechanism.  That is, not all default risk will be mutualized.

Clearinghouses use margins to limit the amount of risk that is mutualized.  Only losses on defaulted positions in excess of margin posted by the defaulter are mutualized.  The higher the margin cover, the lower the level of risk sharing.

In practice, CCPs utilize a “defaulter pays” model in which margin covers losses on defaulted positions with extremely high probability, e.g., 99.7 percent of the time.  In a defaulter pays model, the amount of risk mutualization is very low.  CCPs are not, therefore, primarily an insurance mechanism.  They insure only tail risks (which has important implications for systemic risk and wrong way risk).

Note LCH.Clearnet’s boast that it had collected far more margin than necessary to cover the realized losses on the Lehman’s derivatives positions that it cleared.   The CME also had more than enough Lehman margin to cover losses on its positions (although there were shortfalls on some product segments that were covered by excessive margins on others).

At the Paris conference I attended in September, the head of Eurex Clearing, Thomas Book, was adamant that the goal of his CCP, and of CCPs generally, was to avoid mutualizing risk if at all possible.  His answer surprised Bruno Biais, whose model of clearing in the paper he presented focuses on the role of the CCP as a default risk insurer.

I confess that I have been inadequately appreciative of this point as well.  Understanding its implications has important consequences, as I hope to show in a bit.

To summarize.  CCPs generally operate on a defaulter pays basis: this is also sometimes referred to as a “no credit” system.  That term will help illuminate the differences between cleared and uncleared markets.  The defaulter pays system means that the amount of risk shared through a CCP is extremely limited.  This limitation on risk sharing is best explained as the consequence of moral hazard.

In contrast to a cleared market operating on the no credit model, dealers in bilateral OTC markets historically extended credit to derivatives counterparties.  Put differently, OTC deals often bundled a derivatives trade with credit provision.  That is, dealers often willingly took on exposure to a default loss when entering into a derivatives deal with a customer.  (This is not true for all types of customers.  For instance, hedge funds typically had to post margin.)  Taking credit exposure is equivalent to extending credit to the counterparty.

Now consider whether a firm will prefer to trade a cleared derivative, requiring the posting of a high margin and which thus embeds no credit, or prefers instead to trade an otherwise identical OTC product that does embed credit.  To fix ideas originally, let’s consider a firm that is cash constrained.  Therefore, if it wants to trade the cleared product, it must borrow to fund the initial margin.  The cleared derivative is a no credit transaction, but that doesn’t mean that moving to clearing necessarily reduces the amount of credit the firm obtains: it can borrow the money needed to post margin, and indeed, may have to borrow it.

One source of credit is dealer banks.  So the firm could either enter into an uncleared bilateral trade with a dealer, or borrow money from the dealer bank to fund the IM.  (The argument doesn’t really depend on dealing with the same bank on the derivatives deal and the borrowing: this just facilitates the exposition.)

Here’s were the loser pays aspect of margin comes in.  Let’s say that the firm is selling a derivatives contract with payoff P. If the firm defaults, the OTC counterparty’s exposure is max[P,0].  But in a loser pays model, the margin M is almost always greater than max[P,0].  Thus, the borrowing from bank to fund margin almost always exceeds the default loss that the bank would incur if it entered into a bilateral deal with the firm.

I can show formally that if the firm already has debt outstanding, under standard pro rata/pari passu default loss allocation mechanisms, holding everything equal,  the bank’s default losses if it extends credit to the firm to fund margin almost always exceed, and never are smaller than*, the default losses that it would incur if it had entered an uncleared bilateral trade with the firm.  This, in turn, will make the cleared transaction more expensive for the cash-constrained firm, and it will prefer to trade the OTC product.

There are at least a couple of reasons why the cleared transaction can be more expensive.  One is what is effectively a debt overhang problem.  In order to induce the bank to lend the margin for posting at the CCP, the firm must promise it higher payments in non-default states than it has to promise the bank in these states when it trades OTC instead.  Since the firm’s managers, acting in the interest of equity, only care about payoffs in non-default states, this means that returns to shareholders are lower when it borrows to fund margin than when it deals OTC.  This can be seen another way.  I can also show formally that the payoffs to the firm’s other creditors are almost always higher, and never lower, if it borrows to fund margin than if it trades OTC.  Thus, the value of the the firm’s non-margin-related debt is higher if it trades a cleared product and funds the margin by borrowing, than if the firm uses the OTC product.  Since the value of the firm’s assets doesn’t differ in the cleared vs. uncleared cases, and since value is conserved, this means that equity is less valuable when the firm trades cleared products than bilateral ones.   Some of the benefit of borrowing to fund margin flows to other creditors; this is where the analogy to debt overhang comes in.

This is most easily seen in the following scenario.  The firm can become insolvent when max[P,0]=0, i.e., the bilateral contract is out of the money to the bank, and the bank suffers no loss due to default, and all the losses of insolvency would fall on other creditors.  However, if the bank had lent the firm money to fund margin, it would suffer a loss on the margin loan in this circumstance.  This loss would reduce the loss suffered by the other creditors.

OTC is cheaper than cleared products in other models of capital structure.  For instance, moral hazard (or adverse selection) mean that the firm will be credit constrained: the amount it can borrow is limited by its collateral, and/or the amount of cash flows that it can credibly pledge to lenders.  The same formal analysis implies that more cash flows in non-default states must go to supporting a margin loan in a defaulter pays clearing model than in a bilateral transaction.  This leaves less cash flows to support other borrowing, so by borrowing to fund margin loans the firm must borrow less to support other investments (which, in this sort of model, it has insufficient equity to fund itself).  Thus, borrowing to fund margins on cleared transactions crowds out borrowing to fund positive NPV investments.

This analysis implies that this cash constrained firm will choose to trade OTC rather than cleared products with defaulter pays margins funded with loans.  The bank is indifferent, because it will price the product or the loan to cover its costs, but the firm is always better off  with the bilateral trade because it has to pay the bank less if it trades OTC than if it borrows to fund margin.

Moreover, the analysis implies that if the firm is forced to clear, it will either scale back its derivatives trading (because the cleared transaction is more expensive), and/or reduce its investments in positive NPV projects. Cutting back derivatives trading is costly if this trading reduces the deadweight costs of debt, for instance.  Indeed, I can show that the cost of margin is especially high when the derivatives trade is a “right way” risk, as would occur when the firm is hedging.

Clearing mandates are therefore expensive for such firms, and there is a legitimate reason to exempt such firms from clearing requirements.  (Note that even if these firms are exempted, other rules affecting dealer banks, e.g., punitive capital charges on OTC derivatives trades, can induce an inefficient use of cleared transactions.)

This analysis explains the preference of many firms, especially corporate end users, for uncleared OTC trades that embed credit, as opposed to cleared transactions that must be funded by increased borrowing.  This, in turn, can explain the growth of OTC markets relative to exchange traded markets from the 1980s onwards.

It is useful to step back a bit here, and understand what is really going on.  In essence, in this model clearing is expensive because it causes one agency problem to exacerbate another.  CCPs adopt defaulter pays because of an agency problem: the moral hazard associated with risk sharing.  Debt is expensive or constrained for firms because of agency problems (e.g., debt overhang problems, or constraints on borrowing due to moral hazard).  Effectively, the firm must obtain more credit to support a cleared position than an uncleared one, and  the cost of debt arising from agency problems makes this higher level of credit more expensive.

This analysis raises the question of why firms would ever choose to clear.  There are a couple of answers to that.

First, there may be other (private) benefits.  For instance, netting economies may be greater with clearing.  Of course, the efficiency effects of netting are equivocal (because netting primarily has the effect of redistributing losses among creditors), but as a positive matter is is pretty evident that netting offers private benefits, and thus the mulitilateral netting that can occur in clearing, but not to the same degree in bilateral trades, could induce some traders to prefer clearing.

Second, the analysis started from the assumption that the firm at issue is cash constrained and hence has to borrow to fund margin on the cleared trade.  Some firms are not.  One example would be an ETF like USO or USNG, which collect the entire notional value of derivatives in cash from their investors.  These firms do not require any credit to meet margins.  Large real money funds, like a Pimco, that collect cash from investors and use derivatives to gain exposure to price risks would be another.

Even many hedgers may not suffer from cash constraints that limit their ability to trade cleared contracts.  Consider commodity trading firms.  They typically use derivatives to hedge inventories of commodities.  Banks, in turn, are willing to lend against these inventories as collateral.  Thus, the commodity trader can fund margin using borrowings secured by commodity inventories, and the lender does not share in default losses pro rata with other creditors.  This type of borrowing is fundamentally different than the borrowing considered above, in which the firm borrows against its balance sheet and default losses are shared with other creditors.

Thus, the simple models would predict that whereas cleared derivatives used to hedge liquid inventories are as cheap or cheaper than uncleared derivatives, it is much more expensive to use cleared derivatives to hedge cash flows on illiquid assets, or to hedge broad balance sheet risks.  This is largely consistent with my understanding of the pattern of usage of cleared and uncleared derivatives.

This model, which combines a model of the cost of risk sharing at a CCP with a model of the capital structure of firms, has both positive and policy implications.  In particular, it can explain the adoption of defaulter pays by CCPs.  It can also explain the disparities between OTC and cleared markets when CCPs utilize defaulter pays.  Moreover, it demonstrates that clearing mandates can be inefficient if they are applied too broadly.  One source of this inefficiency is that the mandate leads to a perverse interaction between agency problems.

Of course, a rationale for clearing mandates is that clearing reduces systemic risk.  Anyone who has read my work will know that I am dubious of that rationale on many grounds, but it is worthwhile to consider the implications of the foregoing analysis for systemic risk.

The model is too sparse to make very strong conclusions, but one consideration does stand out.   If firms borrow from OTC dealer banks to fund margins, holding the rest of the firm’s liabilities and assets constant, these banks suffer larger losses when the firm goes bankrupt if they lend to them to fund margins on  derivatives trades than if they enter into identical uncleared OTC derivatives trades.  As noted before, there is a distributive effect: other creditors suffer smaller losses.  The systemic implications of this redistribution depend on the relative systemic importance of the banks and the other creditors.  Dealer banks are definitely systemically important, but other creditors may be too.  Therefore, it is not evident how this cuts, but if one believes that large financial institutions that serve as OTC dealers are especially crucial for systemic stability, moving to cleared trades would tend to increase systemic risk because clearing actually increases their credit exposures to customers.

But again, caution is warranted here.  The redistribution result holds capital structure and derivatives trades constant, but of course these will be different if firms have to clear than if they don’t.  These changes are difficult to predict, and the systemic riskiness of different configurations is even more difficult to compare given how little we really know about the sources of systemic risk.   But the fact that clearing can lead to adverse interactions between agency problems should raise concerns about the systemic implications of forcing clearing.

*To be more precise.  When P>M, the CCP and the lender suffer default losses, and the sum of these default losses is the same as the bilateral counterparty would incur on an uncleared trade.  Relatedly, other creditors of the bankrupt firm suffer the same default losses in the cleared and uncleared cases when this condition holds.  They suffer smaller losses whenever this condition does not hold and the firm goes bankrupt.

In September, the CFTC’s new SEF rule threw swap market participants into a panic when they realized that footnote 88 buried within the rule would require instruments that were thought to have been outside the scope of the rule to be traded on the new SEFs:

Further muddying the waters is a footnote that was written into the margins of the final rules that could end up having a major impact. Footnote 88 requires certain entities that offer trade execution of swaps on a multiple-to-multiple basis to register as SEFs, even if they only execute or trade swaps that are not subject to the trade execution mandates.

This could lead to a significantly expanded SEF universe, which is already expected to be crowded. Trading participants would have to code and test to these various SEFs, some of which will have hastily assembled their platforms in order to be compliant. The on-boarding process will be rushed for participants in order to generate needed liquidity at the Oct. 2 starting point.

“For some of our members who have decided that they’re going to connect to each one of the platforms that’s available, they’re going to need this unique coding and connectivity to these new platforms,” Nevins says. “That’s a lot of work in a very short period of time.”Those who don’t want to connect to each venue will have to decide quickly which SEFs they want to work with, and they still need to hurry the coding and connection of pipes to the SEFs.

But two can play the footnote game.  Today Bloomberg reports that the banks have found a footnote in the rule that allows them to shelter some trades from the SEF execution mandate.  Footnote 513.  (The number itself should give an indication of how bloated and complex the rule is.)

Wall Street, trying to preserve profits from swap trading in the face of tougher scrutiny from Washington, has found a new way to keep some of its overseas deals private. It’s called Footnote 513.

Banking lawyers have seized on the wording of the footnote, contained in an 84-page policy statement issued in July by the main U.S. regulator of derivatives. The largest banks told swap brokers in late September that the language means certain swaps still don’t fall under the agency’s new trading rules, according to three people briefed on the discussions.

London-based ICAP Plc (IAP), one of the largest swap brokers, told the banks it didn’t agree with their interpretation, said the people, who spoke on condition of anonymity because the discussions weren’t public. Other brokers accepted the banks’ position and have been trading billions of dollars in contracts outside the new regulatory system, the people said.

The deals in question include swaps for foreign clients that are arranged by U.S.-based traders or brokers and then booked through a U.S. bank’s overseas affiliate. Bank lawyers say that if the trade goes through an affiliate it can stay private and doesn’t need to be handled on one of the public electronic trading platforms approved by the Commodity Futures Trading Commission, two people said.

Other than lawyers, who thinks this Duel of Footnotes is anyway to run a railroad?  (I know, I know: most lawyers think this is stupid too.  It’s just that they’ll be about the only ones that profit from it.)

And remember it’s not just footnotes.  It’s commas too.  As I wrote about a year ago, the decision overturning the CFTC’s position limit rule hinged in part on the placement of commas.

The whole structure is still under construction, and it is already requiring a plethora of ad hoc fixes, in the form of exemptions and no action letters, which Commissioner Scott O’Malia trenchantly criticized.

And it’s not just in the US.  The rules and paperwork are metastasizing in Europe too.  As one commentator puts it, “Regulation descends into a Kafkaesque bureaucracy.”  Yeah.  This will work.  Especially when their Kafkaesque bureaucracy clashes with our Kafkaesque bureaucracy.  Kafka squared. Woot!

And it’s not just derivatives.  It’s hedge funds.  And I could go on.

The hedge fund article also points out something I’ve been hammering on for years now: perversely, since compliance costs have a huge fixed cost element, the regulatory onslaught creates scale economies that favor concentration and consolidation, and which can potentially reduce competition.  You are seeing it in hedge funds.  You are seeing it in banking.  You are seeing it in FCMs.

And I do mean perverse, because among the ostensible purposes of these regulations were mitigating the too big to fail problem, and promoting competition.  The dramatic increase in regulatory overhead that favors the big over the small is completely at odds with these purposes.

Moreover, don’t forget another point that I’ve made.  These rules are systemic in nature, in the sense that they affect myriad financial market participants, and in pretty much the same way.  Meaning that if one of the rules is fatally flawed, it could lead to a serious systemic problem.

Which raises the question: what footnote-or what comma-in what rule will create a future crisis?  Perhaps that’s a little hyperbolic, but not outrageously so.

SEC Chair Mary Jo White has made something of a stir with her recent speech on equity market structure.  One interesting thing was her attempt to downplay regulation as a cause:

Although some have argued otherwise, these developments are not attributable solely to regulatory choices. Competition plays a powerful role. Well before Regulation NMS, market participants were trading in dark pools and trading with highly automated strategies. Many jurisdictions around the world with different regulatory structures than ours are dealing with analogous issues related to off-exchange venues and automated trading.

This is a slippery and somewhat disingenuous argument. It starts out with an Obama-esque straw man: I don’t know of any credible analyst of these issues who claims the problems in market structure are “attributable solely” to regulatory choices.  Having knocked down that straw man, she focuses on somewhat off-topic technological issues and thus totally sidesteps saying what role regulation, and notably RegNMS have played.

I generally agree with her emphasis on empirical evidence as the basis for policy choices.  But the empirical research has to consider all potential issues-including potential regulatory culpability.

The part of White’s speech that attracted the most comment relates to exchange competition and the self-regulatory model:

Exchange Competition and Self-Regulatory Model

Another set of assumptions about our current market structure is related to the nature of exchange competition and the nature of the self-regulatory model itself.

Equity exchanges today operate fully electronic, high-speed trading systems using a business model that mostly was developed by electronic communications networks, or ECNs, prior to Regulation NMS. Indeed, in many ways, today’s exchanges are yesterday’s ECNs.

Exchanges differ from ECNs, however, in significant respects. Exchanges, for example, continue to exercise self-regulatory functions, even as they operate as for-profit entities.

This model for exchanges has encountered challenges. As I noted earlier, for example, the “lit” exchanges no longer attract even one-half of long-term investor orders.

From time to time, equity exchanges have adopted trading models that use different fee structures or attempt to focus on different priorities, such as order size or retail investor participation. These models have been met with mixed success, which raises the question as to whether exchanges have a real opportunity to develop different trading models that preserve pricing transparency and are more attractive to investors.

As is true for all important aspects of our current market structure, the current nature of exchange competition and the self-regulatory model should be fully evaluated in light of the evolving market structure and trading practices. This evaluation should include whether the current exchange regulatory structure continues to meet the needs of investors and public companies. Does it provide sufficient flexibility for exchanges to implement transparent trading models that can effectively compete for investor orders? Does the current approach to self-regulation limit or support exchange trading models?

This evaluation should also assess how trading venues can better balance their commercial incentives and regulatory responsibilities. For example, is there an appropriate balance for exchanges in key areas, such as the maintenance of critical market infrastructure? And are off-exchange venues subject to appropriate regulatory requirements for the types of business they today conduct?

One thing that jumps out at me is the suggestion that there is a tension between the for-profit model and self-regulatory responsibilities, with the corollary suggestion (rather obliquely made) that for-profit exchanges should be stripped of some regulatory responsibilities.

Let me break it to everyone: there is a tension between the not-for-profit model and self-regulatory responsibilities as well.  Not-for-profit exchanges were not charitable organizations.  They were established and run by very profit driven intermediaries (brokers, liquidity suppliers) who adopted the non-profit form to prevent redistribution of wealth among heterogeneous members.  This form was economically rational in floor trading days, when intermediaries and exchanges had highly specific capital that needed protection against expropriation. A big technological shock-the move to electronic trading-meant that the non-profit form was not longer needed to protect these specific investments, or to prevent exchanges from using pricing of their services to redistribute wealth among members.

As I wrote about extensively during the mid-90s, traditional exchanges did a good job at some self-regulatory functions, and a bad job at others.  Manipulation was one that exchanges policed quite badly, for instance: contract enforcement was one they did quite well.

In these papers, I identified a couple of factors that affected the effectiveness of exchange self-regulatory efforts.  One was whether the exchanges internalized the benefits.  In the case of manipulation, for instance, the effects of corners and squeezes on the derived demand for the services of exchange members did not reflect the effects of corners and squeezes on market participants generally.  The price and quantity of exchange services depends on the impact of manipulation on marginal customers, but inframarginal customers bore the brunt of manipulations.  Moreover, some effects were public bads, such as the effect of manipulation on the informativeness of prices.

Another factor is the distributive effects across exchange members.  Self-regulations that led to overall efficiency gains but hurt some members were often not adopted due to such conflicts: the battles over grain warehouses in Chicago in the 1860s and 1870s is an example.

Similar problems afflict for profit exchanges, though internalization problems are probably the most acute now: the movement to the for-profit form has reduced the conflicts within exchanges.

Back around 2000, when the move to for-profit exchanges was gaining momentum, I wrote a paper on the implications of this for self-regulation.  I put it aside.  I should probably dust it off and revise, but the basic conclusion holds true, I think: for-profit exchanges have good incentives to regulate some things, bad incentives to regulate others.  The task is to identify these various strengths and weaknesses, and allocate regulatory responsibilities accordingly, always keeping in mind that the alternative-public regulation-has its weaknesses and strengths too.

The one major issue that has brought this to the fore is the breakdowns in the SIPs that connect exchanges and are necessary for the information-and-linkages approach adopted by the SEC in RegNMS (and Congress under the original NMS mandate) to work.  This is predictable, based on the analysis presented above.  SIPs have some attributes of public goods, and no exchange internalizes the benefits and costs of SIP performance.  Not surprising, then, that this has proved to be a weak link in the current market structure.  But since SIPs have natural monopoly characteristics, it’s not an easy task to determine how they should be owned, priced, and governed.  But the question hasn’t been framed this way, which means that it’s unlikely to be answered properly.  (Recall that the old Intermarket Trading System was also a chronic problem in the pre-RegNMS world: it was sort of a joke, actually.)  SIPs are analogous to transmission in electricity, and transmission has always proved the most difficult part of the system to regulate and price.  (Yes, there are important technological differences, but the similarities are relevant.)

Some things in White’s speech appear contradictory.  For instance, she questions whether a one-size-fits-all market structure is appropriate, then denigrates exchange efforts to experiment with different models: “These models have been met with mixed success, which raises the question as to whether exchanges have a real opportunity to develop different trading models that preserve pricing transparency and are more attractive to investors.”  It seems to me that exchanges have had plenty of opportunities, and that the mixed success suggests that the potential for differentiation (including moving away from one-size-fits-all) is rather limited.

There is one form of differentiation that White seems to be quite suspicious of: trading on dark venues:

A steadily increasing percentage of trading occurs in “dark” venues, which now appear to execute more than half of the orders of long-term investors.

Combine this with her desideratum for trading: models “that preserve pricing transparency and are more attractive to investors.”  Well, here we have a great example of a non-one-size-fits-all model that is obviously more attractive to some investors, but it bothers White (and regulators generally).  One of the things that makes dark venues popular to long-term investors is precisely the lack of price transparency.   Ironically, the SEC has always claimed the markets should favor long-term investors, and indeed White starts her speech by arguing that the markets cannot serve their capital formation function without long-term investors.  Thus, the “and” in that phrase is problematic.

One interpretation is that White doesn’t want one-size-fits-all, but that the varieties that have evolved-dark venues that accommodate long-term investors and lit venues that accommodate some long-term investors and most others-aren’t to her liking.  This is not a convincing position.  It begs the question of why the long-term investors White favors prefer a venue she disfavors.  It also doesn’t come to grips with the literature which shows that off-exchange trading venues reduce trading costs for uninformed, liquidity traders-which can include long term investors who initiate and liquidate positions for non-information driven reasons.  It also is in tension with her statement that for-profit exchanges with a strong profit motive are trying to innovate: if they are losing business to dark venues as a result of, say, predatory conduct carried in lit markets, they have an incentive to devise rules and pricing structures that curb the predatory conduct and thereby attract the return of those who have fled to dark venues.

It also bears emphasis, for this point is often forgotten, that dark venues have always been with us, and that large, long-term investors have been the users of these venues.  Back in the day, this was the function of block markets.  Dark venues have largely replaced the block markets, serve the same function, and service the same investor base.  That’s true even though much of the trading on dark venues is not in blocks (although some dark markets are essentially for block trading).  As Duane Seppi demonstrated years ago, block trading and the protocols surrounding it were ways of screening out informed traders.  Dark venues have other means of performing the same function, without imposing the constraint that trades occur in blocks.

Summing up, it’s good to see these market structure issues get such serious attention.  It’s less encouraging to see that regulators are focusing on secondary or tertiary issues (e.g., whether self-regulation is incompatible with for-profit exchanges) and have certain preconceptions that makes them unduly suspicious of departures from one-size-fits-all trading that accommodate the needs of an important category of market users-indeed, the category that the SEC has long said it desires to protect. (As an aside, which I may expand upon later, the SEC’s preferences, and its suspicion of dark markets, is likely driven by political economy considerations.  Namely, exchanges exert more political influence, and have influenced the SEC to advance their interests.)

In my opinion, the SEC’s immediate priority should be fixing the SIP (if it deems that links-and-information is sacrosanct).  This fix should look at ownership, organization, and governance, based on an understanding of the public good attributes of the SIP.  (Longer term, the RegNMS links-and-information approach needs a new look.)  In addition, the empirical approach that White rightly advocates should look at dark venues through objective eyes, free from prejudices which are all too apparent (as the very term “dark” demonstrates).  It’s especially important that the reasons for the commercial success of dark markets be understood and explained before it is lamented.  A case can be made that the non-one-size-fits-all model that works best could consist of dark and light venues operating side-by-side, serving different investor clienteles.   It’s certainly a model that has existed over time, and in a wide variety of different countries.  Maybe there’s something to it.  Perhaps we should understand what that might be, before monkeying with market structure any more.

The BIS has been one of the major advocates for mandated clearing.  They have produced an analysis claiming that mandated clearing will increase GDP growth by .1 percent or more per annum.  I criticized this calculation claiming that it was predicated on a fallacy: namely, that multilateral netting and collateralization, result in a reduction in the costs of OTC derivatives borne by banks, and thereby reduce the risk that they will become dangerously leveraged.  In fact, these measures redistribute losses, and will not affect the overall leverage of a financial institution in the event of an adverse shock to its balance sheet.

Stephen Cecchetti, the head of the BIS’s Monetary and Economics Department has responded to this sort of criticism.  Here’s his argument regarding multilateral netting:

Before turning to the costs, I think it is worth responding to criticism of this approach. First, some critics have argued that, by focusing on derivatives exposures, the Group has ignored the impact of multilateral netting on other unsecured creditors. The main claim is that multilateral netting dilutes other unsecured creditors.

This is correct. Multilateral netting does dilute non-derivatives-related claims to some extent. However, this is neither new, nor is it unique to derivatives. In fact, repos, covered bonds and any other secured loans result in dilution and subordination. For repos, that counterparties can close out a position and seize collateral in default has led to comment and worry for some time. Given that this is all well known, I would think that it is already reflected in the pricing of the instruments involved. Presumably no one will be terribly surprised by this when it is applied to derivatives, and so the impact will be muted. It is a stretch to see how this redistribution of a part of the risk associated with OTC derivatives transactions increases systemic risk.

This argument totally fails to meet the criticism.

First, there has been some repricing, but it is incomplete.  Repricing only takes place to the extent that adoption of mandated clearing occurs, or is at least anticipated, and does not occur for derivatives contracts and other liabilities until they mature.  Since many derivatives and other liabilities have maturities extending well beyond the clearing implementation dates, these have yet to be repriced.

Yes, the most run-prone liabilities, such as short term debt, have been repriced, but that’s not all that important.  Even if banks adjust their capital structures to reflect the repricing, they will still have large quantities of such liabilities which are now more junior and which will pay off less in states of the world when a bank is insolvent.  The higher yield received in normal times compensates the creditors for the lower returns that they get in bad states of the world, and most notably in crisis times.  And it is exactly during these crisis times that these liabilities are a problem.  Due to repricing, there may be a lower quantity of such short term debt, but this debt will be more vulnerable to runs as a result of the subordination inherent in multilateral netting.  Excuse me if I don’t consider that a clear win for reduced systemic risk, and fear that it in fact represents an increased systemic risk.  That is no stretch at all.  None.  Cecchetti’s belief that it is a stretch reflects a cramped and incomplete analysis of the implications of subordination.

In terms of the BIS’s claim that will boost economic growth, Cecchetti’s argument does not rebut in any way what I have been saying.  The BIS argument is based on a belief that netting makes the pie bigger.  My argument is that it does not make the pie bigger, but just resizes the pieces, making some smaller and others bigger.  Nothing in what Cecchetti writes demonstrates the opposite, and in fact, his acknowledgement that netting dilutes other creditors is an admission that the effects of netting are redististributive.

Once that is recognized, the entire premise behind the BIS macroeconomic analysis of the OTC derivative reforms collapses, and the conclusions collapse right along with it.

Cicchetti mischaracterizes the critique when he insinuates that it means that netting increases systemic risk.  I’ve said that’s one possible outcome, but mainly have used the redistribution point to refute the claim (made by the BIS and others) that netting reduces systemic risk. The channel by which the BIS claims it will is predicated on the belief that netting reduces default losses rather than reallocates them.  If they only reallocate them-which Cicchetti effectively acknowledges-this channel is closed, and the asserted benefit does not exist.  It’s very simple.

Therefore, Cicchetti may “remain convinced that the Group’s analysis accurately captures the benefits of the  proposed reforms,” but his conviction is based on faith rather than economic reasoning: his attempted defense notwithstanding, the Group’s analysis is contrary to the economics.

Here’s what Cicchetti says about collateral:

A second concern that has been raised is that the regulatory insistence on increased collateralisation will simply redistribute counterparty credit risk, not reduce it. To see the point, take the simple example of an interest rate swap. The primary purpose of the swap is to transfer interest rate risk. But the mechanics of the swap mean that there is always a risk that the parties involved will not pay. This is a credit risk. In the case where the swap is completely uncollateralised, it is clear that the instrument combines these two risks: interest rate (or market) risk, and credit risk.

Now think of what happens if there is collateralisation. At first it appears that the credit risk disappears, especially if there is both initial margin to cover unexpected market movements and variation margin to cover realised ones. But the collateral has to come from somewhere. Getting hold of it by borrowing, for example, will once again create credit risk.

The point is that, by collateralising the transaction, the market and credit risk are unbundled. I would argue fairly strenuously that unbundling is the right thing to do. Unbundling forces both the buyer and the seller to manage both the interest rate and the counterparty credit risks embedded in a swap contract. In the past, some parties seem to have simply ignored the credit component. Unbundling sheds light on the pricing of the two components of the contract. A more transparent market structure with more competitive pricing will almost surely result in better decisions and hence better risk management, risk allocation and ultimately lower systemic risk. The AIG example is a cautionary tale that leads us in this direction.

I agree completely that clearing unbundles price and credit risks.  This is particularly true under a defaulter pays model, in which the CCP members bear very little default risk.   In fact, this is the focus of a chapter I’m writing for my next book.

But Cicchetti’s assertion that unbundling is a good thing begs a huge question: why were risks bundled in the first place?  By way of explanation, sort of, Cicchetti asserts, without a shred of evidence, that market participants often ignored credit risk bundled in derivatives trades.  Even if that’s true, why would they necessarily take it into consideration merely because it’s unbundled?  I think the most that can be said is that ex post it appears that market participants underestimated credit risk prior to the crisis.   And if they did this with derivatives, they did it with unbundled credits too: look at the massive repricing of corporate paper and the virtual disappearance of unsecured interbank lending starting in August 2007.

But more substantively, there can be good reasons for bundling market risk and credit risk.  I explore these reasons in detail in my Rocket Science paper, which has been around for years.  In particular, there can be informational synergies.  These are quite ubiquitous in banking, and explain a variety of phenomena, such as compensating balances which require firms that borrow from a bank to hold some portion of the loan in deposits at the same bank: this is a form of bundling.  Moreover, bundling can be a way of controlling a form of moral hazard, namely asset substitution/diversion.

At the very least, bundling is so ubiquitous in banking (and finance generally, e.g., trade credit) that it is extremely plausible that there are good economic reasons for it.  The reasons for this practice should be understood before implementing massive policy changes that forcibly eliminate it for massive quantities of contracts.  It is rather frightening, actually, that Cicchetti/the BIS are so cavalier about this issue, and are so confident that they know better than market participants.

And again, even if credit risk is priced more accurately in an unbundled world (which Cicchetti asserts rather than demonstrates), bank capital structures in an unbundled world can be fragile and a source of systemic risk.  For instance, collateral transformation trades used to acquire collateral to post as CCP margin are arguably very fragile and systemically risky even if they are priced correctly.

What’s needed is a comparative analysis of the fragility/systemic riskiness of the bundled and unbundled structures, and this BIS/Cicchetti do not provide.

Cicchetti’s speech suggests that BIS has heard the criticisms of clearing mandates, but doesn’t really understand them, or is so invested in clearing mandates that it refuses to take them seriously.  Regardless of the reasons, one thing is clear.  The BIS has taken a big swing at a rebuttal, and missed badly.