It is ironic that this story about UBS came out on the day I arrived in Switzerland.
UBS has received subpoenas from the SEC, the US Commodity Futures Trading Commission and the US Department of Justice in connection with investigations regarding submissions to the British Bankers’ Association, which sets LIBOR rates. UBS understands that the investigations focus on whether there were improper attempts by UBS, either acting on its own or together with others, to manipulate LIBOR rates at certain times. In addition, UBS has received an order to provide information to the Japan Financial Supervisory Agency concerning similar matters. UBS is conducting an internal review and is cooperating with the investigations.
I wrote several of posts about Lie-bor back before the financial crisis climaxed. The story seemed to fade from view very quickly, despite the fact that the effects of a distortion in the LIBOR index due to misreporting could lead to the transfer of Carl Sagan-esque billions and billions of dollars between swap counterparties and floating rate borrowers and lenders. Moreover, since unlike in the natural gas misreporting episode where some firms had an incentive to report prices that were too high (and did) and other firms had an incentive to report prices that were too low (and did), here all the potential misreporters had the same incentive: to underreport the rates at which they borrowed. This would tend to cause the effects of misreporting to be greater, as misreports would have no tendency to cancel out.
I surmised at the time that despite the fact that there was some pretty convincing stories to support the claims, government regulators did not seem to pursue the allegations aggressively because (a) that was the least of the problems in the banking sector and regulators had other fish to fry, and (b) because of those other problems, regulators were beyond loath to compound them and pour additional burdens on banks that were already reeling and threatened to bring down the financial system. Regulators were looking to put out fires, not start them. LIBOR was something that regulators, like Scarlet O’Hara, could worry about . . . tomorrow.
Those days of forebearance, it appears, are gone. If you are a there’s-got-to-be-a-pony-in-there-somewhere kind of person, you might view this investigation as a strong signal that regulators believe that banks are healthy enough to stand the strain–and the potential fines (and perhaps damages in private litigation). If you’re not, you’re probably saying “when will it all end?” Be careful what you ask for.
A gratuitous image of a UBS branch, a block from my hotel:
In what will hopefully be the continuation of a spirited and thoughtful exchange of views on OTC derivatives markets, John Parsons and Antonio Mello at Betting the Business respond to my response to their response to my post on Gensler’s beliefs about the efficacy of central counterparties. Or something like that.
We agree on some key issues, namely, (a) the amount of margin is not the measure of the total amount of credit firms use up when moving from bilateral to cleared trades, and (b) systemic risk is the crucial issue when evaluating the effects of clearing mandates.
I think we also disagree on some issues.
Netting, for one. MP state:
The reliance on bilateral arrangements meant that exposures that offset one another at the system level were not offset. Individual market makers did not have sufficient incentives to cooperate with one another to cancel offsetting exposures. A major objective of the reform is to encourage the canceling of offsetting exposures and reducing systemic risk by substituting standardized instruments and clearing, among other things, where possible.
As I’ve written repeatedly before, multilateral netting primarily effects priority; netting effectively places derivatives counterparties in front of other creditors in the event that a firm goes bankrupt. The systemic consequences of this are ambiguous: yes, derivatives counterparties may be systemically important, but other creditors may be too, meaning that it is not obvious whether this reordering of priorities improves stability.
Insofar as the other claims are concerned, I am somewhat puzzled by the statement that “[i]ndividual market makers did not have sufficient incentives to cooperate with one another to cancel offsetting exposures.” With respect to bilateral offsetting exposures, presumably market makers relied on set-offs in bankruptcy. With respect to multilateral exposures, market participants internalized any improvements in capital efficiency from netting (or compression). They also internalized any benefits that resulted from lower risk of replacing defaulted positions with counterparty A that were offset by non-defaulted trades with B, C, D, etc. Crucially, they recognized that capturing some netting efficiencies multilaterally could undermine some bilateral netting for deals that could not be compressed or cleared because of the nature of these contracts or their liquidity; they internalized these various netting efficiencies. They internalized the benefits of customization. They would have internalized the transfer of wealth arising from the shift in priorities inherent in multilateral netting. Thus, they could trade-off these various effects and internalize the resulting costs and benefits. Maybe some costs and benefits were not internalized, so it is literally true that the incentives were not sufficient to lead to a completely efficient outcome, but I don’t consider insufficient netting to be the first order problem. Indeed, to the extent that netting transfers wealth, there can be too much netting.
I’d also note, as was recognized when the CBT was considering whether to adopt clearing in the early-20th century, that by improving the efficiency of capital utilization (by freeing up credit committed to offsetting positions with different counterparties), netting can also lead to increases in the size of net positions and net risk exposures. (This was also an objection to clearing that was raised vociferously in the aftermath of Black Friday in 1869.) This can increase systemic risks.
MP also criticize the opacity of OTC markets, and suggest that clearing would make things more transparent:
It lacked transparency. Neither the end-users nor the public authorities had adequate information about the market. This lack of information has handicapped end-users in getting good prices. The lack of transparency has hampered competition. The lack of transparency promoted unsound accounting. Opacity perverted incentives. It distorted compensations. We could go on… It also handicapped the authorities in supervising the marketplace. A major objective of the reform is creating transparency, which will then make it possible to improve the functioning of the market.
Transparency is used in different ways here. It is used to refer to price transparency and position transparency. There is another type of transparency–counterparty transparency–that MP don’t discuss but which is relevant to the issues they raise.
With respect to position transparency, this is a bad argument in favor of clearing. It is possible to achieve such transparency via reporting to repositories without the problematic risk sharing and risk pricing aspects of central clearing. Indeed, a single trade repository is required because (a) there will be multiple clearinghouses, and (b) some products will not be cleared. If it’s position visibility that you want (especially to regulators), that should be achieved via reporting to a single repository, not by having multiple CCPs.
With respect to price transparency, I think that MP overstate the lack of competition and price information, especially for the more liquid products that are likely to be traded most actively on SEFs. Most end users I’ve heard discuss the subject state that they have been able to get good price quotes from multiple dealers. ISDA recently ran an experiment which found that spreads for liquid products are extremely tight–in the .5bp-1bp range. Moreover, with respect to many standardized OTC products, end users can compare the bids and offers they get to the prices of exchange traded instruments (e.g., Eurodollar futures, crude oil futures) to determine the quality of a dealer’s quotes. End users also have the ability to utilize those products–to the extent they don’t, that means that the bundle of services provided by the dealer is more valuable to end users than the different bundle available on exchanges.
Moreover, as I’ve written before, trading in a non-anonymous way by soliciting quotes from dealers can actually be cheaper for end users than trading in an anonymous exchange or exchange-like (i.e., SEF) environment with pre-trade transparency. The verifiably uninformed are typically going to get better prices in non-anonymous settings than in anonymous ones, where spreads are widened and depths reduced in order to protect quoters against being picked off by the informed.
All of this is not to say that OTC bilateral markets were perfect. Too big to fail provided perverse incentives. Uncoordinated replacement and hedging of defaulted positions exacerbates price moves. But clearing presents its own difficulties, so just inventorying the defects of bilateral markets is not sufficient to prove the superiority of clearing. You have the Churchill democracy issue: something can be the worst thing ever, except for all alternatives that have been tried from time to time. Moreover, I don’t consider the defects that MP identify to be nearly as problematic as they do.
Thus, it still comes down to a comparative analysis which must make comparisons along numerous dimensions; some of these comparisons may cut one way (e.g., favoring centralized clearing), others may cut the opposite way (favoring bilateral mechanisms). Moreover, when discussing policy it is important to recognize that clearing bundles many services–risk pricing, risk sharing, netting/compression, default risk management, position reporting–that may be better provided separately.
I still think that a crucial issue that relates to one of my areas of agreement with MP–that firms subject to a clearing mandate will substitute one form of credit for another–deserves further analysis. I hope that they respond more fully to my analysis of the implications of the degree of substitutibility of different forms of credit. In my view, that will have very important implications for the effects of clearing mandates.
John Parsons and Antonio Mello have a series of posts taking aim at the notion that requiring end users to post collateral on derivatives trades will impose a large burden on them that will cause them to reduce real investment. Their basic point is that OTC derivatives trades often embed an unsecured credit line, and that firms that are forced to post collateral can fund it with an explicit borrowing transaction and/or credit line. This ability to substitute one form of credit for another should dampen the impact of clearing and collateral mandates on corporate end users.
I agree with the essential point. It was one of the reasons why I long ago opined that requiring AIG to post collateral (initial margin/independent amount) likely would not have materially caused them to reduce their positions in CDS because they would have been able to borrow–and would have wanted to borrow–to fund that collateral. Less debt in one portion of the balance sheet would have made it possible to borrow more in other ways. I discussed that point more generally in my Cato Policy Analysis piece.
I further agree with Parsons’ and Mello’s conclusion that this means that looking at the total collateral that end users will be required to post is a misleading measure of the financial burden that collateral, clearing, and capital requirements will impose on end users.
That analysis is as fine as it goes, but it’s worthwhile to push it further–much further. In particular, to push it to examine the effects on systemic risk, which Parsons-Mello also (rightly) identify as the key driver of the ultimate impact for good or ill of clearing mandates.
There are a couple of cases. In the first case, firms have access to perfect substitutes for the unsecured credit embedded in many OTC derivatives trades. In the second, there are not perfect substitutes. (It seems to me that M-P believe there are pretty close substitutes–feel free to correct me if I’m wrong, guys).
In the first case, clearing mandates and required collateralization will have no appreciable effects. Firms will just substitute an economically identical form of financing/credit for the proscribed bundled credit. New labels will be slapped on old bottles.
But that means that the systemic effects of the mandates will be minimal–and perhaps non-existent. That’s because the mandates will not affect the quantity, character, and crucially, fragility of the capital structures of either end users or financial intermediaries. The same banks will end up extending the same amount of credit of the same risk characteristics to the same counterparties. It will just go by a different name.
In the second case, where there are not perfect substitutes, things are much more complicated. Firms will adjust on many margins. They will substitute different kinds of financial claims (debt and equity) for the proscribed financing embedded in derivatives: the systemic effects of this are very difficult to know. Firms will engage in less risk management, most likely: again, the systemic effects, and efficiency effects more broadly considered, are hard to predict. And yes, firms will likely reduce the scale of investment as limited pledgeable capital will be diverted from supporting real investments into liquid assets to support derivatives positions.
It is interesting to note that corporate end users–manufacturers, airlines, energy companies, and the like–have been the most outspoken critics of clearing mandates and margin requirements, while financial firms have been relatively mute on the subject. This is particularly interesting given that to address end user concerns, CFTC Chairman Gensler has emphasized that financial firms are the main targets of the mandates and margins.
To me, this makes sense in light of the foregoing analysis. Banks and other financial institutions likely can substitute other forms of unsecured credit for credit bundled in derivatives trades more readily than can end users. Banks engage in massive amounts of unsecured credit transactions (e.g., interbank funding) with other banks with which they trade derivatives and extend and receive embedded credit. (Many interbank trades are not collateralized.) A mandate that says “thou shall not extend unsecured credit via derivatives trades” is unlikely to be hugely burdensome on the banks, because they will just increase other unsecured financing to substitute for the decreased credit embedded in derivatives trades. It is plausible that end users have fewer such options.
Why might end users have poorer substitutes? Here are some conjectures.
For one, the mandates may reduce the set of firms from which end users may obtain credit. Consider a firm that is looking to hedge natural gas prices. It can do a swap with Goldman, sure, but it it can also do a trade–and get credit from–BP, or Cargill, or myriad other firms. With the mandate, the firm won’t be able to get credit from those counterparties. It may still trade with them, but it would have to finance the necessary collateral either out of its own resources or by borrowing from a bank.
This creates many potential difficulties. For instance, due to concentration limits and other factors, any given creditor may incur increasing marginal costs to extend financing to a given borrower. This means that it is often more efficient to get a given total amount of credit from multiple sources, rather than just one. Reducing the number of potential suppliers of credit can therefore raise the costs of credit.
Moreover, narrowing the sources of potential credit may expose end users to greater opportunism. It is well understood that banks obtain information about the firms that they finance which gives them an advantage over other potential sources of credit. They can exploit this information advantage opportunistically. Firms seeking credit can limit their vulnerability to such holdups by dealing with multiple creditors. Margin and clearing mandates that force end users to get credit to support derivatives positions through banks may therefore increase their vulnerability to holdup–thereby inflating the cost of trading derivatives.
I also wonder whether the credit embedded in derivatives trades is likely to represent a more robust commitment to extend finance in the future than other forms of unsecured lending. This is more conjectural, but I find it a plausible conjecture. If the suppliers of other forms of credit can more readily renege on commitments to supply liquidity to support a derivatives position, the liquidity and funding risks are greater in a clearing/margin mandate world.
These are just thoughts off the top as to how bundled and unbundled credit can differ. I suspect that there must be some difference, at least for end users, based on revealed preference grounds. This essentially turns the Parsons-Mello point on its head. If the form of credit associated with a derivatives position is a matter of indifference, why do we see a decided preference among many users for non-collateralized trades that bundle credit from the counterparty? Why are end users so vociferous in their opposition (granting that some of their vociferous arguments don’t really hold water)?
But these are the kinds of questions that have to be addressed to understand fully how clearing and collateral mandates are going to work. And these are the kinds of questions that the advocates of the mandates haven’t even acknowledged, let alone answered.
There’s another issue here that deserves special attention. Clearing and margin mandates must be met by the posting of high quality, liquid collateral such as cash and Treasury securities. Where is the increased supply of these instruments going to come from? (Manmohan Singh at IMF has written about this issue.)
Here I think that the Holmstrom-Tirole distinction between “inside” and “outside” liquidity is important. They distinguish between liquidity supplied within the corporate sector (“inside liquidity”) and liquidity supplied by households, or through the government (via taxing the household sector) (“outside liquidity”). My thoughts here are very preliminary–this is really thinking out loud–but I sense that the effects of clearing mandates on inside and outside liquidity is actually the crucial issue, and one that deserves probing analysis by knowledgeable people.
One plausible conjecture is that much of the liquidity that currently supports derivatives trading is inside liquidity, and that mandates, by requiring posting of collateral in the form of cash and high quality securities, will increase the demand for outside liquidity. Some of this may be supplied by the corporate sector as they adjust investment policies to favor more liquid assets. But a good portion will have to come from the household sector or the government sector.
This is unsettling. It is unsettling in part because it means that the efficiency of derivatives markets will be intertwined even more closely with fiscal and monetary policies that affect the supply and demand for outside liquidity. It is also unsettling because as Gary Gorton, Bengt Holmstrom, and Jean Tirole (and others) have argued, one impetus for securitization–including subprime securitization–was the need for high quality collateral, i.e., it was a response to demand for liquidity from the corporate/financial sectors. To the extent that clearing and margin mandates make the corporate and financial sectors more dependent on outside liquidity, it would tend to drive a similar dynamic.
Moreover, when systemic risk is considered, correlated/aggregate/macro shocks are the most difficult to address through the design of liquidity supply mechanisms. If we are worried about the financial system’s vulnerability to aggregate shocks–more generally, shocks that can’t be diversified away within the corporate sector–we have to think seriously about legislative and regulatory changes that may affect the reliance of the corporate/financial sector on outside liquidity.
In sum, I think it is worthwhile to think long and hard about some of the indirect effects of clearing mandates–especially their effects on capital structure and the demand for liquidity, particularly specific forms of liquidity.
To try to close the circle here, Mello and Parsons analyze a crucial issue: the effects of clearing mandates on capital structure. In a Modigliani-Miller world, this wouldn’t matter, and that might be their point. But if capital structure is relevant, legislatively mandated changes in the way companies finance themselves and the way credit is extended and used will have real effects. What’s more, some of these effects may be systemic. After all, research on systemic risk focuses on the ways that imperfections in financial markets can lead maximizing firms to choose capital structures that lead to inefficient, and often destabilizing, responses to economy-wide shocks. Therefore, to understand the systemic effects of major policy changes like mandatory clearing and collateralization of derivatives trades, we need to understand how these policies will affect financial contracting generally. I don’t think we understand even remotely what those effects will be. I am far from convinced that those effects will be salutary, or even benign. But that’s the kind of thing academics and policy makers should be focusing on.
John Parsons of MIT takes issue with my post arguing that CFTC Chairman engages in magical thinking on clearing. John claims that my criticism is based on a (non-existent) “conservation of risk” principle. I respectfully disagree. I am criticizing the arguments that Gensler has in fact advanced, whereas John is generously attributing to him arguments as to how changing rules (including clearing rules) can reduce risk that I have not seen Gensler make.
In contrast, I have written extensively that clearing affects both the allocation of risk for given economic decisions, and the amount of risk through its effects on myriad decision margins; I do so explicitly in “Do You Believe in Magic?” From the very beginning of my writing on clearing mandates, dating back to late-2008 when the issue became a salient one, I have made that point repeatedly. Indeed, my analysis of AIG–Exhibit A-to-Z in many pro-clearing mandate arguments–has emphasized this point. See here, here, and here, and my piece in the Journal of Applied Corporate Finance; I also raised it in an exchange with Jeffrey Gordon of Columbia Law School in my presentation on clearing that you can view here. This working paper, written in the fall of 2008, details a variety of ways in which clearing (via its effects on margining, netting, monitoring, and risk sharing) affects both the amount of risk exposures and the allocation of these exposures. I could go on and cite additional papers and presentations: suffice it to say, most of what I have written about clearing focuses on how clearing affects incentives and information, as compared to bilateral mechanisms, and how these changes affect the amount of risk in the system and the way that risk is allocated. That’s far more than a start, I think it’s fair to say.
John states that the effects of clearing on the amount of risk in the system is “the very heart of the original argument behind clearing which Gensler is making.” I’d like to see exactly what argument John is referring to. I’ve read a lot of what Gensler has said and written in the last three years, and I haven’t even seen that issue even at the peripheries of his arguments. Typically Gensler argues that shifting the risk to the clearinghouse eliminates interconnections and eliminates counterparty risk because the CCP stands behind trades, or because CCPs will require collateral that make sure that firms will have the resources to pay for their derivatives losses. That’s it. Nothing related to how clearing will affect incentives or information in a way that reduces the amount of risk in the system.
Consequently, I think the magic box metaphor is a very apt characterization of what he has been pushing since 2008. Or, since it’s Sunday, it’s fair to say that he is promising a miracle that is sort of the inverse of the loaves and fishes: rather than taking meager fare and feeding multitudes, he is claiming that clearing will take risk from multitudes and make it meager. There is certainly nothing as nuanced in what Gensler has said publicly as what John is attributing to him.
John closes by saying:
Let’s have a good discussion about which rules of the game produce the lowest total social risk and the greatest social payoff from the operation of the OTC derivatives market. It’s a tough problem to tackle. Oversimplification will not do.
Welcome to the party! Seriously. Compare that to what I wrote at the end of the piece in Regulation, “The Clearinghouse Cure,” published in 2008 (linked above):
The nature of this analysis is inherently qualitative. It is difficult for anyone, be they academics, market participants, or regulators, to determine definitively whether a clearinghouse would improve the efficiency of the CDS market. I certainly do not claim to possess such definitive knowledge. It is troubling, however, that basic considerations relating to the economics of risk sharing and information have been almost completely absent in the public discourse over CDS clearinghouses. It is also troubling that the potential pitfalls have not been fully aired. Nor has there been an extensive comparative analysis of alternative risk-sharing mechanisms. Therefore, at the very least, this article aims to raise the quality of the debate by identifying crucial issues that have been largely ignored until now, and to challenge a consensus that threatens to engineer a fundamental transformation of the financial markets without proper regard for fundamental economic issues. Moreover, the considerations identified herein should be kept in mind when designing a CDS clearinghouse to ensure that information problems do not make this prescription worse than the disease it is intended to cure.
I’ve been calling for a good serious discussion around these issues for arguably longer than anyone, and have written extensively about many of the “hard problems.” There is indeed an oversimplification problem in this debate–and that’s precisely why I’ve been so hard on Gensler (and Geithner): because they’ve been the ones oversimplifying repeatedly and consistently, and as a result, we run the very serious danger of adopting arrangements that do not generate the greatest social payoff. If I seem combative making that point (Combative? Moi? Nobody’s ever accused me of that before!) it’s primarily a response to the repeated oversimplification by the advocates of clearing mandates.
I know I can have a constructive discussion with John Parsons, even though it is highly likely that we would disagree on crucial issues. But it would be a real discussion about real issues, and based on my extensive reading on the entire regulatory and legislative discussion leading up to Dodd-Frank and clearing mandates, that’s precisely what’s been missing in the 202 area code.
The most important recommendation in the report of the Flash Crash Advisory Board that I wrote about several days ago is that the SEC consider implementing a “Trade-at” rule in place of the current no trade-through rule. Trade-at would require a firm receiving an order to direct that order to the market offering the best quote for execution, whereas under no-trade-through, the firm could internalize the order and execute it by matching the best price currently displayed at any execution venue. Trade-at would largely cripple “dark pools” and internalization (although the Board disclaims any desire to hamper the operation of pools that facilitate block transactions)–what in the old days were referred to as “third markets.”
As I wrote in my original post, trade-at is a red herring when it comes to preventing Flash Crashes. It’s primary effect will be to alter the way securities markets work in normal times, rather than exceptional ones.
The objection to internalization is that it is a form of “cream skimming” that free rides on price discovery in public markets. This reduces liquidity in the public markets, widening spreads, reducing depth, and reducing the informativeness of prices.
Free riding is an externality that would lead to a less efficient outcome if you believe that the public market is competitive. Many of the criticisms of internalization* and recommendations that it be eliminated, or at least constrained via regulation, implicitly assume that the public markets are perfectly competitive.
This is crucial, because if these markets are not perfectly competitive, free riding can actually improve efficiency. When you have a departure from the first best (due to market power, for instance), what would be a market failure in a first best world can actually improve things. For example, usually market power is bad. Pollution is bad. But if there are no taxes or other restrictions on pollution, it can actually be better for a polluting industry to be monopolized rather than competitive. A competitive industry pollutes too much, but a monopolized industry restricts output, and therefore reduces pollution. Market power markups serve the same function as a tax on pollution. In this instance, two bads are better than one.
So when evaluating the desirability of internalization and dark markets generally, it is imperative to examine the nature of competition in public markets. If public markets exercise market power, the existence of an externality (cream skimming) is insufficient to justify eliminating it.
Sadly, this is a point virtually absent from academic analysis of the issue, and seldom raised in public policy debates either. But some people get it:
Kevin Foley, chief executive of Aqua, a block crossing platform, said: “A ‘trade at’ rule is hugely anti-competitive. We haven’t seen the SEC do something [like that] in ages.”
And the prize goes to a Russian guy (and who says I’m always slagging on Russia?):
Dmitri Galinov, head of liquidity strategy in Credit Suisse’s Advanced Execution Services unit, said that a “trade at” rule would have to be paired with reform in the SEC’s regulation of exchange fees, which are higher than fees charged by dark pools and other non-exchange venues.
Exactly right, Dmitri. As I showed in my 2002 JLEO piece, and an earlier working paper titled “Third Markets and the Second Best,” cream skimming, internalization, block markets, payment for order flow, etc., can improve welfare when exchanges exercise market power. And as I’ve also shown in my research, (a) network effects and coordination costs give exchanges market power, and (b) there is empirical evidence that exchanges do in fact exercise market power. Off-exchange trading venues–dark markets in the current parlance, third markets in the old–are a source of competition. The ability to free ride off prices in the public markets reduces the costs that these competitors incur–that is, it makes them more effective competitors and reduces exchange market power.
Indeed, there is a report in the FT that provides an excellent illustration of how exchanges exercise market power by charging supercompetitive fees: “Anger as Exchanges Raise Closing Auction Fees.” Once upon a time exchanges exercised market power and generated rents for their members by limiting the number of members: now they do it by charging supercompetitive fees.
The wars over securities market structure drag on. The criticism of dark pools and internalization is nothing new. If you believe that public exchanges don’t exercise market power, then you won’t like dark pools and internalization. If you believe that exchanges do exercise market power, the case isn’t obvious, and you have a strong basis to conclude that free riding is the lesser of two evils. I’m of the belief that exchanges do exercise market power, and as a result I am highly skeptical of regulations intended to clamp down on internalization and dark pools.
Too often, however, this crucial issue of competition has been overlooked. Discussing the pros and cons of internalization without coming to grips with exchange market power is to ignore the crucial issue. It’s good to see that at least some are making the point.
* Kind of ironic that internalization creates an externality!
Gary has a magic box, called a “clearinghouse.” You put counterparty risk in the box, and it disappears!:
One of the central goals of Dodd-Frank is to redirect much of the OTC derivatives market through exchanges, electronic trading platforms and central clearinghouses that stand behind dealings in derivative contracts.
“What that means to the American public is it’s less likely that the taxpayers stand behind that transaction,” [Gary Gensler] said.
“Instead the clearinghouse stands behind it, and I think that’s what Congress recognized. Moving as much of this into clearinghouses lowers interconnectedness, but also says a clearinghouse stands there, not the taxpayer.”
Uhm, but what stands behind the clearinghouse? Where does that risk go? Well, financial institutions—including many major banks—actually stand behind clearinghouses. Meaning that if counterparty losses are big enough, the banks bear a big chunk of the loss (so things are still interconnected). Or, if the banks limit the amount of loss they are willing to bear, the clearinghouse can fail. Who wears the loss in that event?
Gensler might argue that collateralization in clearinghouses makes it highly unlikely the banks will bear any loss, and even less likely that the clearinghouse will fail. Let’s say that collateralization is indeed extremely robust, so that the risk of clearinghouse failure is vanishingly small. Does that mean that everything is great?
Well no, really. The risk still hasn’t disappeared. Given the positions that firms take*, collateralization (and netting, another commonly touted benefit of clearing) affects the distribution of losses arising from the bankruptcy of a derivatives trading firm, not the total amount of the losses. If collateralization is so robust that derivatives counterparties always get paid, that means that the entire loss arising from a bankruptcy falls on the firm’s other creditors. So the magician may point to the magic derivatives clearinghouse box and say: “See! No losses! They disappeared!” But they are somewhere, behind the curtain, borne by somebody else; collateralization changes the priority of claims, not their magnitude.** Most likely, the loss is shifted to unsecured creditors—who may include banks backed by taxpayers, or owners of money market mutual funds who may run if it looks like their funds will break the buck, which can also lead to a Federal bailout (like what happened in September, 2008).
There’s also the issue of where the collateral comes from. Some of it may come in the form of loans extended by banks that used to extend credit via derivatives trades; like in a real-world magic trick, the credit didn’t disappear, it was just shifted somewhere else. Some collateral may come from firms foregoing higher yielding, but less liquid investments: this is a real cost as too many resources are tied up in liquid assets and too little invested in more productive projects.
There are other things going on behind the curtain. Rather than enter into expensive collateral derivatives trades, firms may decide to hedge less, forcing their claimants (including lenders, employees, suppliers, and shareholders) to bear more risk. Again, the risk doesn’t disappear–it leaves through a trapdoor and winds up someplace else.
And collateral can have perverse effects in a crisis. The pressure of collateral calls can force firms to liquidate positions, exacerbating price moves. Or firms may engage in fire sales of other assets to meet the calls, causing price shocks to spill over from the derivatives market to other markets. How’s that for an interconnection?
Gary also has a magic transparency trick, a visibility cloak if you will. Exactly the opposite of Harry Potter’s invisibility cloak, it makes the invisible visible:
“I also think transparency helps,” he said, looking back to the widespread uncertainty during the crisis of the value of assets on the books of giants such as American International Group.
“The law now says they have to be valued daily and valuations have to be shared between the counter-parties and if they run a clearinghouse, the clearinghouse has to share it with the public.”
Indeed, this cloak is so magical it can make visible things that don’t even exist. For, in situations like AIG, the problem was that the values of the assets were uncertain because there were no trades: these things weren’t marked to market, they were marked to a guess—or to a self-serving price that was hard to challenge because there were no real prices to compare it to. Many of the contracts that Gensler would like to see cleared trade little, if at all—and no trades, no prices. It is a magical cloak indeed that can make prices and valuations visible when there are no trades to produce actual prices.
Magical thinking about clearing, counterparty risk, and transparency is extremely dangerous. It creates very real risks. Those who believe that things like clearing and transparency are magical solutions, and who do not grapple with the complex realities and difficult trade-offs are all too likely to impose rules and regulations that make things worse, not better. Sad to say, I think that’s happening in spades, both here in the US and in Europe–and that’s not a card trick. The consequences are pretty sobering to contemplate.
* The amount of margin will affect the sizes of positions and the structure of positions that firms take. Netting will also affect positions, because (a) netting can actually economize on collateral, and (b) the amount of collateral savings can depend on how extensive cross-margining and portfolio margining are. It is difficult to know what the net effect of clearing mandates will be on the trades that firms make, and the composition of the firms that make them. I mention one possible effect below–some hedgers may exit the market, or hedge less. Indeed, the potential effects are so extensive it is difficult even to list them. Making it costlier to shift risks via derivatives transactions affects not just risk management decisions, but capital structure decisions (hedging really being just a capital structure choice), investment decisions (the amount and composition of investment), management compensation decisions, and on and on. All I want to emphasize in the main discussion is that one major effect of collateralization is to affect claim priority, given the total aggregate size of claims against a bankrupt firm. It secures some claims–which reduces the priority of other claims.
The lack of a coherent Strategic Petroleum Reserve strategery has apparently been an issue for quite awhile. John McCormack points me to something he wrote for Regulation in 1996:
Owners of above-ground crude oil and refined product inventories have an important role to play. It was just such a role that the Strategic Petroleum Reserve (SPR) was meant to serve but has not. While it has accumulated a crude-oil stockpile of well over half a billion barrels, the SPR has never had a coherent, systematic strategy for selling crude oil during periods when prices spiked upwards. Only small amounts were sold from the stockpile during the 1990-91 Persian Gulf War, and then only after prices had begun to decline from their peak. For the stock- pile to have been of any use during periods of distress, SPR managers needed to have a coherent strategy in place and should have communicated their intentions beforehand. Part of the Energy Department’s reluctance to sell barrels from the SPR resulted from fears that they would be reducing the stockpile before the shortage was greatest. They could have avoided this problem by doing time swaps that would not permanently reduce the stockpile.
Given that a Cato publication and a Carl Levin-chaired committee came to identical conclusions, I think it’s safe to say that SPR has not operated according to sound economic principles.
The SPSoI report also has a lot of stuff on manipulation, including a lot of quotes/cites from me before I was SWP 🙂 I talked extensively with one of the staffers who authored the report while he was preparing it. I haven’t read it in years and on re-reading was kind of taken aback to see how much of what I told him made it into the final report. So, even if I do say so myself, it’s a pretty good primer on manipulation generally, and oil market manipulation more particularly.
Numerous futures markets heavyweights, including CME, ICE, and BlackRock are up in arms over a CFTC rule that would require delisting of a futures contract that had more than 15 percent of its volume traded outside the exchange’s central order book. Their opposition–anger, really–is justified.
The Commission’s logic is that it wants to protect the price discovery process in centralized markets. It recognizes that block trades and other off-exchange trades can mitigate price impact on large transactions, e.g., in hedging trades. Verifiably uninformed traders have an incentive to trade off exchange because this allows them to avoid paying the adverse selection-related costs incurred in an anonymous market; it can also reduce execution risk. Loss of some uninformed order flow reduces liquidity on the centralized market, and can reduce the informativeness of the prices on that market.
Quantifying the impact of all these changes is impossible. Much of the effect is to transfer wealth from one group of uninformed traders (those who trade on the centralized market) to another (those who can avail themselves of the block market or other off-exchange trading venues.) In my JLEO “Market Macrostructure” paper I demonstrate that the net effect can be positive, i.e., the benefit to the latter can exceed the loss suffered by the former. The cost of a decline in price informativeness cannot be quantified. Period. Moreover, even here things are ambiguous. A decline in the amount of uninformed trading on the central market tends to reduce the amount of informed trading–that’s why price informativeness declines. But some informed trading is rent seeking–it involves spending real resources to acquire information in order to achieve a transfer of wealth. This is inefficient. Reducing this activity actually improves welfare.
In other words, nobody knows how much block trading/off-exchange trading activity is the right amount. Least of all the CFTC.
One thing is for sure, though. The marginal cost of block trading doesn’t become infinite when block trading volume is 15 percent of the total. But that is what the CFTC rule implicitly assumes. Moreover, even if the marginal cost (including all relevant costs and benefits) of block trading is positive when it totals 15 percent of volume, shutting down the centralized market if block volume exceeds 15 percent makes no sense at all. A market with “too much” block trading activity still contributes to price discovery. It still provides a cheaper place to trade than no centralized market at all. Maybe its value isn’t maximized when there is too much block trading, but shutting it down sure does minimize its value. It’s like cutting of your nose to spite your face.
Presumably the CFTC believes that no market will ever be shut down. That any exchange facing delisting of a contract will change the block rules (e.g., increasing the minimum block size) in order to reduce block trading and avoid a shutdown.
But in adopting this change-your-rules-or-the-contract-gets-it strategy, the CFTC presumes that it knows the right amount of block trading. It knows no such thing. Indeed, the basis for its threshold is purely arbitrary–a joke, actually:
In the second largest category, involving 128 contracts from all asset classes which included contracts with large and small open interest, the average amount of off-exchange trading over the three-month period ranged from 0% to 15%. The Commission believes that this second category of contracts, where there was actual centralized market trading to observe, provides a reasonable basis for establishing a minimum centralized market trading requirement. Accordingly, from this second category the Commission took the upper range of the maximum average amount of off-exchange trading, and proposes that a maximum of 15% of total trading volume of a contract would be an allowable amount of off-exchange trading in order to protect the price discovery process of trading on the centralized market.
So, the staff looks at some historical numbers for historical off-exchange trading activity (including, apparently, exchange for physicals and exchange for swaps, etc.), the Commission picks a number at the top of the range, and declares it “reasonable” to apply this number to all markets going forward. That’s not analysis: that’s arithmetic. It involves no serious consideration of any of the relevant issues and trade-offs relating to on- and off-exchange trading.
The opposition of the exchanges is particularly interesting here. Exchanges internalize many of the effects of their block trading and EFP rules. To the extent that block trading reduces liquidity on the centralized market, it reduces the derived demand for trading on that market–and hence reduces the fees an exchange can charge and the quantity of trades there, which combine to reduce the exchange’s revenue from that source. It can collect revenues from block trades; these revenues reflect the benefits that block trading provides. Through the effects of its rules on these different revenue streams, the exchange internalizes, albeit perhaps imperfectly, the costs and benefits of those rules on execution costs and liquidity for alternative ways of completing trades. (This is different, I would add, from internalization or third markets, where a third party “skims” uninformed order flow and does not internalize the effects on the liquidity of a central market.) Thus, the exchange is in a much better position to know, and has a stronger incentive to know, the liquidity effects of block rules than does the CFTC. It can balance and has the incentive to balance these effects to craft an efficient rule. This argues strongly in favor of letting the exchanges make their own rules.
Exchanges don’t directly internalize any benefits flowing from more informative prices. Nobody does–including the CFTC. The rent seeking argument I mentioned above means that these benefits may not even be positive–they may be negative, once the costs of information are taken into account. Imposing an entirely arbitrary standard based on a calculation that a fifth grader could make, and which cannot–cannot–quantify even approximately the benefits and costs at issue is an act of regulatory overstretch that is far too characteristic of this Commission. It denies those who have some information about some relevant costs and benefits–the exchanges, with regard to the effects of the rules on trading costs–from taking that information into account when establishing their rules. It substitutes its own judgment for what is the “right” amount of off-exchange trading when that judgment is predicated on complete ignorance.
The Commission has a lot on its plate. The Commissioners, with some justice, complain about their crushing work load repeatedly. Here’s a suggestion: they can lighten their burden by avoiding dealing with issues about which the Commission completely lacks the information necessary to craft constructive regulations. The off-exchange trading rule is a perfect example of that. One can have nice philosophical arguments about these issues, but a rational regulatory triage strategy would relegate such debates to academics and focus on more pressing matters about which it is possible to get enough information and knowledge to craft sensible rules.
Today’s WSJ online has a short piece quoting Daniel Yergin counseling against tapping the Strategic Petroleum Reserve in response to Mideast turmoil. Yergin argues that it is desirable to keep the stockpile untouched as a hedge against the possibility of worse disruptions in the future. Insofar as commercial storage is concerned, economic theory highlights two factors that are relevant in determining the optimal timing of adding to or drawing from inventories. My forthcoming book goes into these factors in detail.
The first is that it is optimal to draw down on storage in response to adverse supply shocks that are expected to be transitory. This means that current supply is below expected future supply, making it less desirable to hold inventories to meet future consumption needs. Well-functioning markets move things from where they are relatively abundant to where they are relatively scarce. A temporary supply shortfall means that the present has become more scarce, relative to the future, making it desirable to shift some consumption from the future to the present. When some inventories are available, this intertemporal shift can be achieved by drawing down on stocks.
In contrast, supply shocks that are expected to persist should have minimal impacts on storage decisions as such shocks do not affect supply today relative to what is expected in the future. Since scarcity today relative to future scarcity isn’t affected by a highly persistent shock, there is little or no need to change inventory to move consumption from the future to the present.
The second factor is uncertainty about fundamentals. Inventories are in part a hedge against uncertainty about future demand and supply conditions, a buffer against adverse supply or demand changes in the future. An increase in uncertainty about future supply increases the precautionary demand, leading to an increase in inventory holdings. This is the factor that Yergin emphasizes, and which I analyze in detail in chapter 5 of the forthcoming book.
In the current situation these considerations are in conflict. It is likely that the effect of the effect of the loss of Libyan production will be relatively transitory as this production will be brought back on line when the situation, which appears to be reaching a climax, stabilizes, and because other suppliers, notably Saudi Arabia, have pledged to expand output. This would tend to cause a decline in inventories. On the other hand, the risk of future supply disruptions—not excluding major disruptions in Saudi Arabia—has increased dramatically. This would tend to lead to an increase in inventories.
Movements in the forward curve suggest that the first effect is dominating. The contango in WTI has declined and Brent has also moved into a backwardation. The WTI curve has become slightly humped at the short end. Such moves are associated with declines in inventories. That said, the effect of the increased uncertainty has probably led to smaller stock drawdowns.
The foregoing relates to commercial storage decisions. SPR is not operated on a commercial basis. That said, it would probably make sense to take commercial considerations into account when making storage decisions, if only to reduce the cost to the taxpayer. The fact that SPR is not operated on commercial principles, but is a political animal with difficult to predict decisions likely has some feedback effects on commercial storage decisions; with respect to politics, some, such as MA Rep. Markey, are calling for the SPR to release supplies into the market. Commercial storers have to take into account their expectations about SPR actions when deciding how to make their own storage decisions.
It would seem to be very hard to predict what SPR will do given its political-military nature. Clarifying SPR’s operational methodology would assist private parties to make rational storage decisions that would assist in reducing price volatility.
The Joint CFTC-SEC Advisory Committee on Emerging Regulatory Issues has issued its findings regarding the Flash Crash and potential regulatory remedies to such events. Overall it is a solid effort that makes a serious attempt to grapple with the thorny issues arising out of the May 6 events, and changes in financial market structure more generally.
My primary criticism is that the report assigns far too much importance to the speed of modern, electronic markets dominated by high frequency traders that are the primary liquidity suppliers. This over-emphasis is pronounced in section I, titled “Volatility.”
The report confirms what has been pretty evident from the start, namely, that liquidity was eroding before May 6, and then virtually disappeared for a short period of time on that date. But it does not provide any reason to believe that this was something that was driven particularly by high speed electronic trading. The economics of market making have been the same for a long time under a variety of trading technologies. These economics are dominated by considerations related to information–and especially information asymmetry–and risk. These considerations are present in every trading and technology environment. There are, moreover, informational and risk conditions in which liquidity suppliers leave the market in droves. This happened in old-fashioned face-to-face markets: witness the fact that on 19 October, 1987 most futures market locals left the floor–or were pulled off by their clearing firms. The Flash Crash just demonstrates that these factors are relevant in electronic markets too.
By focusing on high speed computer trading, and suggesting that it is fundamentally different from liquidity supply in the old days, the report distracts attention from the more fundamental causes of crashes. Yes, HFT is new, but HFT market makers are responding to the same economic factors in similar ways to their old school floor trader counterparts. The main novelty of HFT is that high frequency traders trade across markets, rather than in one or two. The report would have done better to focus how to best mitigate the age old plagues of market making in an electronic environment.
The report discusses the advisability of imposing obligations on market makers, but wisely demurs from recommending this. It recognizes the difficulties, though it does not acknowledge the most important problem, namely, that obligations are costly and will affect adversely liquidity supply in non-crisis periods.
The report also correctly lauds the CME’s stop functionality, and recommends it be adopted in other markets as well. It makes constructive and reasonable recommendations about how this functionality can be tweaked to make it more robust. In brief, it recommends what could be described a “reload” feature that would kick in if the initial application of the stop functionality failed to stem a crash.
One particularly interesting part of the report relates to “Preferencing, Internalization, and Routing Protocols.” It is particularly interesting because it highlights an issue that I analyzed in my 2005 article “The Thirty Years War.” In that article, I noted that the SEC had deliberately chosen an “information and linkages” approach rather than a central limit order book (CLOB) approach to securities market structure. I further noted that the rules adopted by the SEC did not preserve time priority across trading venues, and did not protect all limit orders in all books against execution on another venue at an inferior price to those not at best of book. These were problematic features of the SEC RegNMS, and I was critical of both (especially the fact that only limit orders at the top of the book are protected).
The report argues that the lack of time priority inherent in the “no trade through” rule has encouraged internalization of orders, and that the lack of integration across books at prices away from the inside market can exacerbate price movements. It therefore recommends that the SEC study the costs and benefits of changing these aspects of RegNMS. Specifically, it recommends study of the imposition of a “trade at” rule which would effectively restore time priority.
I agree with the recommendation, but think that these issues are somewhat of a red herring with respect to the Flash Crash. This is especially true of the “trade through” rule. The report authors argue that the current rules favor internalization; internalizers supply liquidity; but internalizers stopped providing liquidity during the Flash Crash, leading to an additional flow of orders to public markets that stressed an already liquidity constrained situation. Imposing a “trade at” rule that would make it impossible for internalizers to match quotes standing in the public market would encourage additional liquidity supply in the public markets.
This is no doubt true, but the report doesn’t analyze the counterfactual, i.e., how things would have worked under the trade at rule. Yes, there likely would have been less internalization. Thus, in normal circumstances, more liquidity would be supplied via public quotes and less via internalization. But we saw during the Flash Crash that public market makers fled and as a result quotes disappeared. Internalizers provided less liquidity at the same time. But isn’t it highly likely that the liquidity suppliers that entered the public markets during “normal” times to replace internalizers would have also fled the market in conditions like those that provoked the Crash? I think so. The report doesn’t consider this counterfactual, so it cannot support a different conclusion.
That is, I think that it is accurate to say that a “trade at” rule would shift liquidity from internalizers to public quotes during normal times, but that this does not imply that liquidity would be higher overall (across all sources of liquidity supply) during stressed periods. The quotes that appear in public markets because quote matching becomes harder would be highly likely to disappear in Flash Crash-like conditions because it is particularly costly to provide firm quotes under those conditions–that’s why the quotes disappear. Thus, it is worthwhile to have a debate/study about the wisdom of the trade at rule, but that debate study should recognize that just because internalizers fled when the market went crazy doesn’t mean that the new public quoters attracted to the markets when internalizers are hobbled will stick around.
A look at the futures market bolsters this skepticism that internalization was a major contributor to the Crash itself. The index futures markets are not fragmented, and are not subject to internalization. The quote matching problem doesn’t exist there. At all. But it suffered serious liquidity withdrawals on 6 May. Thus, in my view, internalization is a red herring with respect to Flash Crashes. It is an important issue, but its relevance is primarily during “normal” times and that should be the focus of any study.
I was also intrigued by the report’s discussion of peak load pricing and the problems of system capacity constraints driven in part by high cancellation rates for HFTs. This subject was, in fact, the subject of the first substantive post on SWP.* There I puzzled as to why exchanges didn’t set traffic-sensitive price schedules for access to exchange matching computers. I noted that the lack of pricing of capacity could lead to overconsumption of computer resources, lags in getting orders executed, and even shutdowns of the trading system. Interesting to see that 5 years after I wrote that piece the issue is getting some attention. I’m still puzzling, and look forward to seeing how the exchanges and regulators deal with this issue. (The report mentions the difficulty of pricing cancellations in a fragmented market structure. This is another manifestation of the US equity market-centric mindset of the report’s authors. That isn’t an issue in futures markets, or foreign markets that are less fragmented. The puzzle is not explained by market fragmentation.)
In sum, the report moves the ball forward some. I think it gets sidetracked on some issues, but its focus on the need to encourage liquidity supply during stressed times is salutary. A lot of work remains to be done to see just how that can be accomplished in an efficient way, however.
And speaking selfishly, that’s a good thing. For the report confirms something I wrote in the conclusion of “The Thirty Years War”:
Therefore, the proposed rules [Reg NMS] are not the final battle in a ThirtyYears War. I fully expect that in 2075, some professor will write an article about the latest clash in an ongoing Hundred Years War over securities market structure regulation.
These market structure issues are thorny indeed, and will not be resolved any time soon. I am sure they will not be resolved by early-2013, when my next book, Financial Market Macrostructure: The Organization of Securities and Derivatives Markets should appear. Which should be good for sales 🙂
*The post was titled “VOLT”, a neoacronym (which might be a neologism) for “Value of Lost Trade.” The idea was that in figuring out how much generating capacity is needed in an electricity market, it is conventional to assign a “value of lost load” (VOLL) to the consumption of power lost when system capacity is reached. Similarly, when deciding how much trading capacity to build, it would be advisable to determine the value of a trade lost or delayed due to a shortage of system capacity.
