Making Better Markets

by: Nemo Incognito

This is a pretty wonky post and in the spirit of giving credit where credit is due, I thought it would be good to highlight two very good books on the topic.

Firstly, the excellent Restoring Financial Stability by the staff of NYU Stern. This is a very thorough and detailed account of what went wrong and what to do about it. Though I don’t agree with everything they suggest there are some great suggestions. I will stick to the issue of market structure for OTC derivatives though the book has numerous other useful suggestions regarding shadow banking, capital adequacy requirements and the like. The other significantly more basic book is Trading and Exchanges which is a useful guide to how trading actually works. When studying economics in university I noted that detail in modeling and description of economic systems is often sacrificed for the sake of mathematical elegance. Books like this make one think much more fundamentally about how exchange works.

One thing that is often sorely ignored in most analysis in the media and even in economics programs is how important market design is in determining the final outcomes of clearing prices, volatility and liquidity. The moronic “just let the market sort it out” approach often does work out well enough but should not be considered an article of religious faith as it is for some rusted-on Chicago schoolers. As Paul Krugman noted, people of this view of things are having a hard time maintaining credibility after the events of the last few years.

So with that in mind, how should markets be designed and what should they optimize? Trading and Exchanges provides a quick and dirty sum of the literature and outlines two key interests: those of the participants that cause the markets to exist in the first place and the wider public’s interest in market transparency, information value of accurate prices and so forth. One additional note made in Trading and Exchanges is this:

Third, public policy should support the interests of protit-motivated traders only when necessary to pursue the first two objectives. Since dealers help make markets liquid and since informed traders help make prices informative, these profit-motivated traders play important roles in our markets that we should support. However, we should support them only for the sake of the benefits they provide other traders and the economy as a whole. We should not favor them when we can obtain liquidity and informative prices more cheaply elsewhere. For example, public policy should not support dealers to the exclusion of public traders who are equally willing to provide liquidity. Likewise, public policy should not allow informed traders special access to information that could as easily be granted to all traders.

The problem that has emerged in CDS markets is that those who originally had the need to manage their credit risk were banks, who, coincidentally, also set up the CDS market and acted as dealers. It is no small wonder that the market became an institution that was by the banks, for the banks and yet which pushed hard and fast to trade with everyone else. In fact, the organization representing the banks fought so hard to ensure that the privileged information and market data stayed strictly on their side and out of the way of regulators and other market participants. Gillian Tett outlines the objectives of ISDA pretty clearly here in Fool’s Gold:

Then, around the same time, Mark Brickell, a member of Hancock’s team [in the JP Morgan CDS business], received a surprising overture from a Washington-based organization, the Group of Thirty. Brickell had joined J.P. Morgan straight out of Harvard Business School around the same time as Hancock and was a true believer in the wonders of swaps. Hancock had selected Brickell to act as his key point man in lobbying regulatory officials and politicians, who had been asking so much of late about the derivatives business. Brickell was ideally suited for the job. He was an intense man, with passionate libertarian views, who had entertained the idea of pursuing a career in politics. Unlike most swaps traders, he therefore relished dealing with politicians, and he was so enthusiastic about lobbying that by 1992 he held the post of chairman of the ISDA. He was Hancock’s and the industry’s Rottweiler in fending off regulatory concerns.

With these kinds of conflicts of interest and lobbying capacity it should be no small wonder that CDS markets stayed OTC, even if they should have been more transparent and better intermediated from a welfare economics point of view. All the internal crossing of transactions and lack of pricing security helped dealers in terms of bid/offer while ensuring that as hubs of trading they could adequately lay off their own risk, or at least they thought they could until the failure of Bear Stearns and then Lehman brothers exposed the serious externality problem of this market: the various “hubs” of the network do not know what one another is doing. As a result, it is easy for excessive risk to build up in a particular counterparty like AIG and the network is poorly equipped to deal with hub failure. This leads the staff of NYU stern to support at the very least centralized clearing:

The rationale for regulatory intervention in getting credit derivatives to trade through a centralized clearinghouse is twofold. First, when one party trades with another, the two parties currently do not internalize the counterparty risk externality they impose on others: That is, they do not recognize that by requiring both sides/parties to post sufficiently high margins, they reduce the uncertainty that their defaults create losses to other parties connected to them. Second, large players benefit substantially from the relatively opaque nature of OTC contracts—both in terms of low trade execution costs and access to privileged information contained in order flow. Hence, large players are unlikely to coordinate, if left to private incentives, to eliminate the risk externality by moving to alternative platforms that feature centralized trading and greater transparency. To us, the case for a centralized clearinghouse—a single counterparty for all trades—is thus clear.


While some would claim that there is no guarantee a clearinghouse can’t fail, it is hard to argue with history on this point:

Since the inception of derivatives trading (1848), no clearing corporation has ever gone bankrupt. Nowadays, clearing corporations are large, and some clear for several exchanges—for example, the Options Clearing Corporation (OCC) and the Chicago Mercantile Exchange (CME).

So while centralized clearing appears to be a layup in reducing systematic risk it does nothing to solve the more mundane problems of this market being by the banks, for the banks and thus having significantly higher trading costs and lower transparency than other products. The key issue here is whether the market participants want more transparency and if so to what degree. The tradeoff is pretty clear here:

Unlike with regulators, unfettered transparency of prices, volumes, and positions to the public at large involves a trade-off. On the positive side, greater transparency helps limit the risk externality created by counterparty credit risk. On the negative side, transparency may be onerous for institutions since (1) it may reveal their trading strategies, and (2) it may reduce their inclination to trade, and thereby also affect liquidity in an adverse manner. Moreover, there are costs to collecting and processing information, which are likely higher in more diffuse markets such as the swaps market and lower in more concentrated markets such as the CDS market, where there are only around 25 key players.

With respect to (1) there is obviously a tradeoff here. 13F’s are not particularly well loved by the investment community, especially once positions get to be of a certain size and scope such that the information could be used by competitors. As a result, many firms with big positions use total return swaps and the like to mask their economic interest behind a custodian or broker, adding to trading costs. With respect to (2) there are particular problems with products that are highly convex in their payoffs: if you are a broker dealer trading CDS whose payoffs are very asymmetric it is hard to provide extensive liquidity as a market maker when some market participants may be better informed than you and the product has serious gap risk. This is particularly so with high yield credits and the like, much unlike equities or most commodities which have pretty symmetric risk profiles in the short run. This is not without precedent:

This [real time price disclosure] is a feature of most markets and is now a feature of the corporate bond market, which was hitherto entirely OTC but now has trade-level disclosure to TRACE. For example, the CDS market for corporate bonds is a natural OTC market for which a TRACE-like system seems appropriate. Real-time public transparency of prices and volumes at the trade level is unambiguously beneficial for small trades (since it ensures smoother revelation of information into prices and more orderly liquidation of positions, which both lower volatility) but involves trade-offs for large trades (since revealing the volume may reveal who is trading). As a solution for large trades, the TRACE system reports the trade price, but does not reveal the volume. The cost of revealing the volume of a large trade, which makes it easier for market participants to determine the parties to the trade, needs to be traded off against the benefit of revealing the volume of a large trade, which aids the dissemination of information into prices.


And TRACE is not without its problems. By not disclosing any volume data at all it makes it very hard to make markets “in size” given the risk of being stuck with a large illiquid position. This problem has been outlined here: while spreads collapsed the liquidity of the market suffered due to dealers having to fight harder for less margin on the same gap risk for the bonds. One logical answer to this would be to provide volume data infrequently – maybe once a day – so that dealers taking on a position could know what their liquidity risk is. Optionally, the 13F disclosure requirements could be withdrawn and made private information to a regulator so that more frequent volume data would not make it easy to work out who the parties to a trade were.

Once the prospect of real time price and volume disclosure seems reasonable, the logical question is whether to move these products to exchanges. There are a lot of advantages to this, including all those of clearinghouses as well as the greater information content available to non-participants or regulators. The problems are as below:

One significant inconvenience of exchange trading of derivatives is that the contracts need to be quite standardized to permit a large number of traders to be trading the same instrument. Standardization would be a challenge for many OTC products, like CDO tranches. However, OTC credit default swaps are already highly standardized with regard to maturities and other terms that are mostly specified by selecting standard options in an International Swaps and Derivatives Association (ISDA) agreement. A bigger issue is resistance from large players to move trading from OTC markets to centralized exchanges, because they benefit from lack of transparency of OTC markets and would likely be required to post higher collateral to clearinghouses and exchanges.

Given the standardization of contracts as they currently stand that does not seem to be a major problem. The second by now should be seen as a good thing: credit derivative are market risk and should be provisioned against accordingly rather than becoming a tool of regulatory arbitrage and the “shadow banking system” – a parallel universe where economically equivalent risks do not require the same capital charges.

And What about Other Derivatives?


The final issue is what to do about customers like oil companies who use derivatives as a true hedge and who do not want to post margin. This was considered to be the obvious out for the OTC world by the banks despite the howls of protest at the CFTC which have not softened much. Gary Gensler is right – specific oversight is necessary and there is no reason it cannot be done by the CFTC. A customer of a bank, say, Exxon (NYSE:XOM), that wants to hedge its exposure to oil can readily do so by having two things with a bank: a trade made by their trading entity or a shell operated by the bank but held by them and a credit agreement with the bank such that the bank will post collateral into the account to make margin calls. The loan itself would have all the usual covenants, bells and whistles and yet would allow the bank to enter into vanilla trades as a market maker and make loans like a bank, neither activity of which Obama/Volcker seem inclined to stamp out while making sure their customer does not face a cash crunch due to a timing mismatch.

In Summary


Putting most OTC derivatives onto exchanges would not be particularly hard and having a centralized clearing mechanism for others would be easy too. The important thing to keep sight of here is that the people who currently run these markets have every interest to stop this happening. Please, do get angry about this. It may be obscure but it is going to be a big part of ensuring that we never have to go to bed one morning wondering whether the banking system will be there the next day.

About this article:

Expand
Want to share your opinion on this article? Add a comment.
Disagree with this article? .
To report a factual error in this article, click here