«Leonard Nakamura Federal Reserve Bank of Philadelphia October 19, 2010 Durable Financial Regulation: Monitoring Financial Instruments as a ...»
WORKING PAPER NO. 10-22/R
DURABLE FINANCIAL REGULATION: MONITORING
FINANCIAL INSTRUMENTS AS A COUNTERPART TO
REGULATING FINANCIAL INSTITUTIONS
Federal Reserve Bank of Philadelphia
October 19, 2010
Durable Financial Regulation: Monitoring Financial Instruments as a
Counterpart to Regulating Financial Institutions*
October 19, 2010
Federal Reserve Bank of Philadelphia Abstract This paper sets forth a discussion framework for the information requirements of systemic financial regulation. It specifically proposes a large macro-micro database for the U.S. based on an extended version of the Flow of Funds. I argue that such a database would have been of material value to U.S. regulators in ameliorating the recent financial crisis and will be of aid in understanding the potential vulnerabilities of an innovative financial system in the future. I also argue that the data should – under strict confidentiality conditions – be made available to academic researchers investigating the detection and measurement of systemic risk.
* The views expressed here are those of the author and do not necessarily reflect those of the Federal Reserve Bank of Philadelphia or the Federal Reserve System. This paper is in preliminary form and is being circulated for purposes of discussion and comment. I would like to thank Viral Acharya, John Bell, Mitchell Berlin, Robert Bliss, John Bottega, Paul Calem, Satyajit Chatterjee, Larry Cordell, Ronel Elul, Jose Fillat, Jeff Fuehrer, Josh Gallin, Itay Goldstein, Chris Henderson, Bob Hunt, Tor Jacobson, George Kauffman, Arthur Kennickell, Bill Lang, Jamie McAndrews, Susan McIntosh, Greg Nini, Kasper Roszbach, Tom Stark, Todd Vermilyea, Larry Wall, Christian Wang, and participants in seminars at the Sveriges Riksbank and at the Federal Reserve Banks of Boston, New York, and Philadelphia, the Federal Reserve System Conference on Real-Time Policy Issues and the System Committee on Financial Structure on Regulation for many helpful comments.
1 Durable Financial Regulation: Monitoring Financial Instruments as a Counterpart to Regulating Financial Institutions I. Introduction: A Financial Regulatory Database for Durable Financial Regulation In the wake of the recent financial crisis, an effort is underway to redesign the regulation of financial institutions. As part of the new regulatory structure, a new information framework may be desirable. In particular, I here advocate a system for monitoring financial instruments as a complement to the regulation of financial institutions. If a system offinancial regulation is to be durable, it must evolve with the development of new institutions and instruments. A main purpose of this paper is to begin a dialogue on an intellectual framework for the analysis of systemic risk data collection. In particular, I discuss how to best construct a macro-micro database that links our knowledge of sectoral financial assets, liabilities, and flows to underlying micro-databases with data on individual instruments and the holdings and liabilities of individual economic actors (households, firms, states).
What is meant by such a macro-micro financial instrument database? The macro side of the database should have summary aggregate data on the nominal quantities of financial instruments and both the debtors and the current asset holders, by broad sector. I argue that this macro side is best understood as an extension of the Flow of Funds database already collected by the U.S. Federal Reserve. The micro side of the database should have micro-data samples of individual instruments and economic actors. The two sides of the database should be interconnected so that the micro-data can be interpreted as a (possibly weighted) sample of portions of the aggregates. Consider the totality of financial transactions, the economic actors who make them, and the stock of assets and liabilities for each actor that continually change as their transactions take place. The macro-micro database should capture key features of the totality so that potential systemic risks may be detected, understood, and mitigated.
In the 2007-2009 financial crisis, financial regulators were surprised both by the size of the potential losses and by the types of institutions that were affected. Regulators moved to protect investment banks, insurance companies, mutual funds, and government-sponsored enterprises, as well as traditional depository institutions. More detailed knowledge of the risks of financial instruments and the holders of these risks might have permitted regulators to move more aggressively in advance of the crisis and would have made regulators better informed once the crisis was at hand. I examine some of the risks that arose in the recent crisis and how we could have known more about them as they were arising.
The proposed database is intended to be of substantial use to supervisors in identifying risk at regulated institutions and also to help them know when financial risks are being held by unregulated financial institutions, generating new systemic risks. U.S. and European regulators 2 are already taking steps to improve data availability. Eichner et al. (2010) argue that while macro-data may be useful for discerning trends in financial risks, it is desirable to have more specialized information to further illuminate them. My framework would take a step toward facilitating this side-by-side use of macro- and more specialized data.
It is also important to have a cost-effective means of collecting and organizing the microdata, that is, the individual financial instruments, so that the evolution of the underlying risks can be followed. For this reason, we need to consider how to best use and improve existing databases, as well as how to develop new ones. Just as the Flow of Funds permits us to observe how much the sectors own and owe by broad class of instruments, so we need links across the micro databases that help us observe the distribution of individuals, firms, and agencies that own and owe, by individual instruments.
In this paper I set forth a framework in which a U.S. financial data office could be the central data-keeper for information on U.S.-originated financial instruments and could be active in making the data available to academic researchers as well as economists from regulatory agencies. Such an office has been provided for in the Dodd-Frank Act as the Office of Financial Research (OFR). The OFR would actively share data (within the limits of confidentiality) and research results on the risks of specific financial instruments with financial regulators and with risk managers within financial institutions. It would thus strengthen the ability of financial institutions to recognize and manage their own risk, conceivably reducing the burden on regulation.
II. How Monitoring Financial Instruments Can Aid in the Regulation of Financial Institutions The task I am describing has two central pieces: One is finding out who holds these instruments and the other is measuring the risks of the instruments and of the holders. Before addressing precisely how these objectives will be achieved, let us use examples from the 2007financial crisis to consider further how this informational database will aid the systemic regulator and all financial regulators.
In this part, we make several points. (1) The database could aid in detecting buildups of systemic risk outside the regulated financial system. (2) It could reduce the opacity of institutional portfolios. (3) It could support studies of the changing risks of instruments, in particular, by permitting investigation of the actions of agents along the full life cycle of instrument creation, distribution, and servicing. (4) It could support pricing analysis that would bring financial, economic, and econometric theory to bear in the determination of potential systemic risk. (5) It could engage a broader, more creative, and potentially more objective community in systemic risk analysis. (6) It could enable regulators to observe counterparty risk and to undertake regular systemic stress testing. (7) It could be used to improve estimation of long-term relationships across variables that may be useful in identifying potential asset bubbles.
3 (1) Following the risk off balance sheet. U.S. financial institutions are regulated piecemeal. This system avoids excessive concentration of regulatory power and provides avenues of regulatory competition: Regulators who regulate efficiently can be rewarded.
However, it has weaknesses. One is that regulated financial entities may shop for weak regulators and, by finding the most complaisant regulator, may weaken the system as a whole.
Another is that financial activities and instruments may be created or moved outside the purview of regulation.
When a financial asset such as a mortgage is created, it can be created within a tightly regulated financial entity, such as a commercial bank, or a loosely regulated one, such as a mortgage subsidiary or a freestanding mortgage company. The regulatory treatment of the mortgage may depend on the form in which it is held. For example, the AAA tranche of a collateralized debt obligation may have much lower regulatory capital requirements than a mortgage loan held in portfolio. Depending on the relevant costs and benefits of the regulatory treatments and the risks of the assets, the form in which the asset is held may change, and this form may have little to do with underlying economic efficiency.
A mortgage asset can be moved off the originator’s balance sheet by placing it in a separate legal entity, such as a special investment vehicle or an entity that issues asset-backed commercial paper. This vehicle may reproduce the characteristics of a financial intermediary, by issuing short-term liabilities that are money-like while it holds long-term instruments that are subject to some risk. These entities generally have standby lines of credit issued by the financial intermediary. If the entity’s asset risk increases, the holders of its short-term liabilities may refuse to roll over the debt, creating a run on the entity and a drawdown on the standby line of credit. Thus, the risks from the vehicle can be easily transferred to the financial intermediary, while the capital of the financial intermediary may be inadequate, since these assets and liabilities were not on its books. The risks and consequences of asset-backed commercial paper are documented in Covitz et al. (2010).
This and other examples of the creation of a ―shadow banking‖ system point out the value to financial regulators of continuing to monitor financial assets after they are removed from the balance sheets of closely regulated financial intermediaries. Note that such a system would permit innovation – it would not block new institutions or new instruments from arising.
Rather, it would seek to monitor these novelties and perhaps bring them under regulation if they reach systemic importance.
As the case of AIG illustrates, the migration of systemic risk to a lightly regulated or unregulated entity contains the seeds of systemic financial crisis. To contain the 2007-09 financial crisis, the regulatory process would have had to identify that AIG was a systemic risk, ascertain the quantitative scale of those risks, and bring AIG under greater regulatory discipline, including preparing a means for unwinding AIG with minimal systemic risk.
4 (2) Another issue is opacity. An important aspect of the recent financial crisis was that many financial institutions themselves lost track of the total real estate exposure of their portfolios. This reflected, in part, the fact that regulation created incentives for opacity. An institution may change an instrument from a loan to one or more securities that it continues to hold, to reduce capital requirements, although no risk has been transferred. The resulting increase in opacity may cause an institution to lose track of its true vulnerabilities.
If financial instruments are opaque to the financial institution holding them, then they must be even more opaque to a regulator. To the extent that instruments can be finely categorized and characterized, and their risks more precisely measured, both regulators and internal risk monitors at financial institutions will be better able to avoid crises. Steps to reduce opacity are being taken to the extent that customized instruments that are not exchange traded will have higher risk capital requirements. To the extent that instruments are exchange traded, they will be easier to categorize and quantify, and measuring their risk will be easier.
To the extent that financial instruments and securities based on them are priced on exchanges, then similar customized instruments and securities may be evaluated and priced based on a model of the values of their financial characteristics. When trades or acquisitions of such assets are observed, the quality of the mark-to-model pricing may be verified.
(3) Following the full life cycle of instrument creation, distribution, and servicing. There are many facets to the creation and maintenance of financial instruments. The characteristics of the financial instrument may be influenced by many institutions and agents. Actions of one set of agents may affect instruments so as to exacerbate systemic vulnerabilities. In the losses associated with the recent financial crisis, the actions of a panoply of agents were at work in increasing the risk of mortgages. The severity of the risks of real estate finance might have been recognized earlier if regulators had been more aware of changes taking place across a variety of institutions.
The risks of mortgages were compounded because a number of the agents were subject to uncorrected weaknesses, many of them well recognized by relevant players. For example, persistent and significant upward bias in home appraisals was recognized as early as the mids (Cho and Megbolugbe, 1996). According to this analysis, some 95 percent of home purchase mortgages in the Fannie Mae mortgage database had appraisals at or above the sale price. Having been recognized, this bias was tracked through automated valuation models (AVMs) by Fannie Mae and Freddie Mac. It was well-known among those actively using these AVMs that these biases were particularly strong for refinancing and for subprime lending.