RDP 2008-06: Promoting Liquidity: Why and How? 4. Promoting Financial Infrastructure that Reduces Information Asymmetries

Given the central role that information asymmetries play in market illiquidity, an obvious way in which to address liquidity issues is to reduce these asymmetries, particularly at turning points in the economic cycle.

Here, there are a number of possibilities, including: further improving disclosure by financial institutions; improving the credit rating process; and improving settlement procedures, including facilitating the increased use of central counterparties.

4.1 Disclosure

While the amount of information disclosed by banks has increased over recent years, the level of disclosure remains, in many cases, well short of what is required. Looking at recent announcements of write-downs by international banks, it is very difficult, even for sophisticated investors, to make an assessment of whether the new asset valuations are realistic. The disclosure statements typically contain only rather general statements of valuation policies, and little specific information about particular assets or portfolios of assets.

In part, the limited disclosure reflects the fundamental difficulty of valuing some assets. But it also reflects the reluctance by financial institutions to provide information about the specifics of their portfolios for fear of revealing trading strategies or portfolio positions to their competitors and counterparties.

4.2 Credit Ratings

A second possibility is to improve the credit rating process – particularly as it relates to structured credit products – in order to rebuild confidence in the rating process, and ensure that ratings convey more complete information to investors. There are many positive aspects to ratings arrangements, including avoiding the inefficiency that can arise if each investor is required to undertake his/her own analysis. But there is little doubt that ratings arrangements can be improved. One concern that has been highlighted by recent events is that the rating agencies are paid by the issuers, rather than the investors for whom they provide information. Particularly for structured finance products, which can be designed to adhere to the rating agency's ratings criteria, the close relationship between issuers and rating agencies may distort incentives and additionally lead to structures that only just qualify for a given rating. One possible solution would be for users, rather than issuers, to pay for ratings, but the coordination or free-rider problem among investors would make such a change very difficult to achieve. A more practical modification would be to limit the degree to which rating agencies can be paid to consult on the structure of a product to be rated, acknowledging that sellers could still use their experience to attempt to structure according to ratings criteria.[3]

An issue that is at the heart of this debate is the extent to which ratings convey useful information to investors (and how investors use that information). While there is, understandably, a strong demand for simplicity, in many cases summarising all the relevant risk information in a single rating is too simplistic. Mechanisms need to be found to present investors with more complete information, without undermining the very useful role rating agencies can play in overcoming information asymmetries. This additional information could include the robustness of models typically used to rate structured finance products, and the sensitivity to external parameters, including changes in the economic environment.

One way in which ratings might become less simplistic is through the introduction of different ratings scales for different asset classes, such as structured finance products or corporate bonds. More useful still might be multi-dimensional ratings. For example, ratings could consist of both a letter rating (AAA, AA, etc) and an indicator that makes the distinction between the probability of default and the expected loss given default, or an indicator that summarises the transition probability matrix, thereby providing information about the likelihood of the asset suffering multiple notch downgrades. There have been several suggestions along these lines over the past year (see, for example, CGFS 2008b; IOSCO 2008; SEC 2008) and comments by the rating agencies (Fitch Ratings 2008; Moody's 2008). For structured finance products, these aspects of risk are much more critical than for standard corporate or government bonds which have generally been served well by a simple letter rating scale.

4.3 Market Design

A third possibility is for the trading in some derivatives and securitised assets to move from OTC markets to exchanges (see, for example, Cecchetti 2007; Alexander 2008). As discussed above, the nature of OTC markets may accentuate the problems of asymmetric information, especially at turning points, leading to sharp reductions in liquidity when conditions unexpectedly change. Several features of exchange-traded markets reduce or eliminate risks that exist in OTC markets, making them potentially more robust. One of these is that settlement typically occurs through a central counterparty. This means that instead of buyers and sellers having counterparty risk with other market participants, the risk is to a highly rated, and in many cases regulated, entity. As a result, concerns about counterparty risk which have contributed to reduced liquidity in many markets in the past year are largely obviated. Having assets traded on an exchange also increases price transparency, so that even in periods of increased uncertainty, market participants are more likely to know where the market price is and so this source of information asymmetry is avoided. The observability of the price can also reduce uncertainty elsewhere because marking assets to market is simpler, which, for example, would reduce information asymmetry about financial institutions' balance sheets.

There are other benefits of exchange-traded markets over OTC markets in that there are lower settlement and legal risks, lower transaction costs, and potentially greater liquidity through participation by a wider range of investors.

Often new financial products start out with diffuse characteristics, but over time evolve into having more standard features, making them more suited to being exchange-traded. This migration can, however, be quite difficult, requiring overcoming legal and market frictions, and the incentive that some institutions may have to retain OTC trading, where profit margins might be higher. Given these difficulties, there may be a case for regulatory policies to play a role in encouraging exchange-traded markets.

One relatively new product, which in many cases has become fairly standardised and thus suited to being exchange-traded, is the CDS. However, to date, attempts by several exchanges to list credit default derivatives have been unsuccessful.[4] One guide for how credit derivatives could evolve is the development of interest rate derivatives, which have a longer history. As Figure 2 shows, OTC markets in these derivatives grew much more rapidly through the 1990s than the exchange-traded markets. This partly reflected the fact that interest rate derivatives were still evolving reasonably quickly and there was considerable innovation. In contrast, in more recent times – as the products have become more standardised – the two market types have seen similar growth rates.[5]

Figure 2: Interest Rate Derivatives

A transition from OTC to exchange-traded markets is obviously not universally possible, nor desirable, given the customised features of many financial assets. For these assets, improvements in clearing and settlement procedures can bring some of the benefits that come from exchange-based trading. In particular, it is important that the post-trade arrangements encourage the matching and clearing of trades on the trade date, or as soon as is practicably possible. The establishment of the Depository Trust Clearing Corporation's (DTCC's) Trade Information Warehouse in the United States has been a useful step forward in this regard, particularly for credit derivatives. Moreover, the use of central counterparties need not be restricted to exchange-traded markets. Indeed, there is a strong case for the use of such arrangements for a variety of OTC markets. On this front there have been some positive developments in recent months. DTCC and the Clearing Corporation (CCorp) have agreed to provide central counterparty services for some OTC credit derivatives, using DTCC's Trade Information Warehouse and the central counterparty services of CCorp.[6] There is a good chance that a central counterparty will become a feature for some OTC credit derivatives; at a recent meeting hosted by the New York Fed, industry participants and regulators agreed to support a central counterparty for CDS (see FRBNY 2008). However, there are notable challenges to overcome in developing a functional central counterparty, not the least of which is determining how to value bespoke credit derivatives in order to set margins.

One means of facilitating more products to trade on exchanges, and also directly reducing information asymmetries, is to increase the standardisation of the structure of various financial assets. Increased standardisation can concentrate liquidity, making the market more robust to shocks that would otherwise tend to cause liquidity to dry up. For RMBS, one possibility is for an exchange or another entity to set and monitor ‘qualifying’ standards, with RMBS that meet these standards being traded on an exchange. It is also possible to imagine continuous disclosure requirements being placed on the entity managing the underlying assets.[7] In a sense, such arrangements would make the processes and infrastructure for trading of a variety of structured debt products more like those currently widely used for equities.

4.4 The Way Forward

There is little doubt that further steps along the lines discussed above could, and should, be taken to reduce existing information asymmetries and to improve market infrastructure. The main challenge is to develop arrangements that work not just in good times, but in bad times as well. Particular attention needs to be paid to ensuring that the integrity of information and the smooth functioning of infrastructure are not impaired when credit conditions or market sentiment deteriorates. Simply developing arrangements that add to the amount of information in good times, but that do not hold up in turbulent conditions may actually increase the probability of systemic liquidity problems.

It is, however, important to be realistic about what can be achieved in this dimension. The recent market strains are the end result of a long boom in the financial sector, underpinned by generally favourable economic conditions. During that boom – as has been the case in almost all preceding booms – investors and institutions simply did not pay enough attention to counterparty risks and the information that was available, applying an overly optimistic lens when looking to the future. This inherent excess optimism during the boom, followed by a period of pessimism when the risk built up during the boom materialises, is endemic and drives the procyclicality of the financial system. It means that simply providing more information and improving market infrastructures is unlikely to be enough to address the liquidity problems that can emerge at the end of a long boom.

One consequence of this is that financial institutions and policy-makers need to consider other ways of reducing the probability of such problems emerging and dealing with them when they do emerge. These issues are addressed in the following sections.


See IOSCO (2008) for a proposal along these lines. [3]

Attempts by several exchanges in the United States (CME, CBOT and CBOE) and Europe (Eurex) to list credit derivatives have been unsuccessful because of a lack of support from market participants. [4]

The levels of outstanding derivatives in OTC and exchange-traded markets cannot be directly compared as exchanges have netting whereas the outstanding value in OTC markets is a gross figure. [5]

Initially CCorp will act as a central counterparty for US index trades, but it has plans to expand to cover other CDS products. The announcement is available at <http://www.clearingcorp.com/press/pressreleases/20080528-dtcc-cds.html>. See also Alexander (2008). For earlier discussion, see Ledrut and Upper (2007). [6]

While not advocating a move to exchange trading, the American Securitization Forum has recently proposed standardising disclosure for RMBS to facilitate comparison of different securities and publishing monthly information on the performance of RMBS loan pools. See ASF (2008). [7]