By Richard Alford, a former economist at the New York Fed. Since then, he has worked in the financial industry as a trading floor economist and strategist on both the sell side and the buy side.
The recent financial crisis has shown that the legal and regulatory steps that have been taken to provide information to parties active in the financial markets are insufficient to insure fair and efficient markets.
The credit rating agency system has serious shortcomings. Given that the theoretical economic assertions regarding market efficiency are predicated on all economic agents making rational, self-interested decisions based upon all relevant information, it is rather amazing that economists, regulators and market participants thought that a single analysis by an agent with a conflict of interest (instead of narrow self-interest) could be counted on to generate analysis that would promote and maintain market efficiency. Any surprise attached to the recent failings of the credit rating agencies should be that the system lasted so long before failing so obviously.
The rationale for single sourcing of information/credit ratings (the NRSROs) must be reappraised in light of technology and the growth of complex and bespoke instruments.
The NRSROs are too well integrated in to the regulatory structure to be done away with at this point in time. It would be unwise (given moral hazard incentives) to allow financial firms to assess their riskiness or the riskiness of paper they hold or issue without some independent analysis or check. Furthermore, potential counterparties and regulators would demand it.
Reform must bear far reaching :
• It must promote the dissemination of all relevant information in a uniform and timely fashion.
• It must also recast the roles of the SEC and the NRSROs.
• It must restrict the use of complex bespoke instruments to investors who are sophisticated enough to perform their own credit analysis,
• To avoid possible moral hazard problems, the holding of complex instruments at TBTFs must be limited.
The SEC must move past the requirement that firms simply prepare and file financial reports. It must adopt a policy of active disclosure and dissemination of financial information. It must make readily available all relevant information on plain-vanilla securities.
Reform should require that:
• All financial reports on issuers, new issues, outstanding public issues and all reports on publically traded asset back issues generated by servicers must be sent to the SEC in a timely fashion and in electronic form.
• The SEC maintain an easily accessible free public library of downloadable, easily manipulated data on all publicly traded issues be they stocks, bonds, commercial paper, or structured products.
• Users must be able to analyze a single issue or issuer, but also must be able to readily do cross-issue comparisons by either predetermined peer groups or user-defined peer groups.
• The SEC must provide price histories.
• The SEC website should provide for some statistical analysis of the data, e.g., correlations, quantitative credit scoring, portfolio risk analysis and, stress tests.
It would be reasonable to expect that numerous private firms would also provide analytic services or software to facilitate user analysis of this free data set. (The SEC might even contract the data service to private firms.). The SEC could also offer prizes or contracts with business schools or other entities with expertise and no axe to grind for quantitative analyses or analytic engines for investors to use.
These steps will not level the playing field, but the wider dispersion of the data and the reduction in the resources and costs of acquiring and processing the data should reduce market inefficiencies and the unfairness in the present system. The existence of easily accessible alternative analytic engines would not only permit a wide range of investors to inexpensively (at the margin) perform their own analyses but would also reduce the cost of entry for potential competitors. It would also serve as a check on the potential for conflicts of interest to weigh on the ratings issued by the NRSROs. Additionally, it would create some pressure on the NRSROs to revise ratings more quickly.
The active disclosure of information by the SEC will mitigate problems when the instrument being rated is a straightforward debt issue or plain-vanilla derivative. However, complex structured products would present a continuing serious challenge. The importance of self-interested research was made clear during the recent Senate hearings regarding Goldman Sachs and its structuring of synthetic CDOs. Sen. Levin repeatedly cited two internal Goldman Sachs e-mails. One referred to the referenced assets underlying a CDO that Goldman was selling as “sh***y”. Another referred to a set of assets as “lemons”.
During the same hearing, another e-mail was cited. This e-mail was from Allied Irish Bank (AIB) (if my memory is correct) to Goldman. The e-mail was a rejection by AIB of a CDO offering by Goldman. AIB not only said that it would not be an investor in the CDO, but went to call the reference assets underlying the synthetic CDO “junk”. If all of Goldman’s counterparties had done the analysis that Allied Irish had done in this case, then the losses experienced and problems we’re facing would not be nearly as large.
Given the relatively limited number of buyers (mitigating the free rider problem) and assuming the marginal cost of the required analysis would be low, perhaps the best solution would entail:
• An end any regulatory requirement to use or benefit from NRSROs ratings on complex structured products;
• Require sophisticated investors to do their own credit analysis and hence be responsible for the analysis (the buck stops here);
• Prohibit unsophisticated investors as well as pension plans and state and municipal agencies from using complex structured products;
• Require banks and TBTF firms to set aside more capital against complex structured products regardless of any internal rating;
• Limit the amount of complex structured products held by systemically important financial firms (relative to their capital and liquidity) such that a severe disruption in the relevant markets would not result in the firm becoming either capital impaired or illiquid.
With the SEC acting as a free public library of financial data or at least making the relevant financial data available inexpensively, the credit rating agencies would find themselves under increased pressure to provide accurate and timely ratings on outstanding firms and issues. The decreased cost of acquiring and processing the financial data should encourage both competition and in-house analysis. Furthermore, it should reduce any room that the rating agencies may have to adjust credit ratings to suit the issuers.
If the number of NRSROs were to increase, it would also be possible for the SEC to require issuers to periodically change the NRSRO designated to perform the rating. This would reduce any incentive that the NRSROs might have to “buy” future business with a favorable rating. Is it a perfect solution? No, but it ought to be a substantial improvement over what we have now.
The continued existence of complex structured products will remain a challenge. Returning responsibility for doing the credit analysis to the investors should reduce problems. However, there is no guarantee that analysis done by investors or third parties will keep up with increasingly complex structures devised by Wall Street. Given the moral hazard incentive, the absence of a legally definitive outside rating might also provide an avenue by which TBTF institutions could increase the riskiness of their portfolios. Consequently, it would probably be necessary to limit the size of the holdings of complex structured products relative to the capital and liquidity of the TBTFs.