Home > bailout, Federal Reserve, financial crisis, GSEs, public finance > Where do we go from here?

Where do we go from here?

October 2, 2009

Last week I had the opportunity to opine on this question at a lively conference on the financial crisis sponsored by the Federal Reserve Bank of Chicago and the World Bank.  Since I spoke about things I’ve been meaning to blog about for some time, I decided to post the transcript here.  Apologies that the tone is more Fed-esque than the usual posting, but here goes…

Where do we go from here?

“You never want a serious crisis to go to waste.  And what I mean by that is an opportunity to do things you think you could not do before.”  Rahm Emanuel, Feb. 2009

I would like to touch briefly on two issues in answer to the question posed for this session:  first, the integration of housing finance into the financial and regulatory mainstream; and second, the need to modernize budgetary and regulatory accounting.   I chose these topics for several reasons: they are important; they get less attention than is deserved; and I have thought quite a bit about them from both an academic and policy perspective.

For those of us who have long worried about Fannie and Freddie – their spectacular political and market power; lack of transparency; and the costs and risks to taxpayers associated with their implicit guarantee – the crisis has opened up the tantalizing possibility of rationalizing the structure and regulation of housing finance.

Yet, although it is widely agreed that the housing bubble precipitated the financial crisis and that Fannie and Freddie as too-big-to-fail institutions posed a serious systemic risk, there is still not an official, articulated, vision for how housing finance will be structured and regulated in the future.  When these issues are mentioned at all, it is outside of the broader context of any regulatory restructuring of banking and financial markets.  The omission can be seen, for instance, in a series of Treasury proposals for restructuring the financial system — both under the previous administration and this one — that are largely silent on the regulation of housing finance.  Avoidance is also evident in the current state of the Federal Housing Finance Agency (FHFA):  It has been left with the peculiar task of regulating Fannie and Freddie, now effectively themselves federal entities; and regulating the Federal Home Loan Banks, institutions that serve as wholesale liquidity providers to financial institutions (Ashcraft et. al., 2009) but that have only an indirect effect on housing finance. [1]

There are of course many proposals for housing finance reform that have been put forward by researchers working in academia, government agencies, think tanks and advocacy groups.  But the ghosts of Fannie and Freddie, and the legacy of regulatory segregation, seem to constrain the set of possibilities that are seriously under consideration. The proposals often presume a continuation of the pre-crisis model of a small number of dedicated conforming mortgage securitizers, guarantors and regulators (e.g., Government Accountability Office (2009), and Mortgage Bankers Association (2009)).  Policy options are taken to be the extent of government backing for mortgage-backed securities, with options ranging from total privatization to full federal guarantees.

A priority I would suggest going forward then, is a fundamental rethinking of the structure and regulation of housing finance, both the primary and secondary mortgages markets, including consideration of to what it extent it can be integrated into the financial and regulatory mainstream.  To quote recent comments by James Lockhart that were made in reaction to a GAO options paper, we “should consider what the secondary market should look like before considering specific institutions.”  He goes on to suggest that “one such possibility would be a market characterized by many privately owned issuers of MBS with the government providing insurance against catastrophic losses, either directly or in partnership with private companies.”

Certainly there are industrial organization and political economy reasons to favor solutions like the one Lockhart sketches that reduce market concentration and make explicit the extent of government backing.  And if that were to occur, a case can be made (although the Fed has argued against it in the past and Mr. Lockhart certainly didn’t suggest it in his comments) for reassigning the oversight responsibilities of the FHFA to a regulator with broader safety and soundness responsibilities, perhaps avoiding some of the vulnerability to political interference of a dedicated housing regulator.

Turning to my second suggestion for what urgently needs to come next, it is to modernize budgetary and regulatory accounting.  I realize this is not on everyone’s top ten fix-it list, and although it would be a stretch to claim that ill-conceived accounting rules are largely to blame for the financial crisis, I believe they have significantly exacerbated it.  Going forward, failure to correct the structural flaws in government accounting will make it more difficult for the federal government to understand, and to extricate itself from, the extensive credit market obligations it has recently assumed.  Regulatory reforms also will be less effective than they otherwise might be.

I want to elaborate on these assertions by way of several specific examples:

First, accounting rules determine the budget cost of federal financial obligations.  If budget rules cause the cost of a policy to be understated, there will be a tendency to over-rely on that policy.  An obvious example was the implicit guarantee of Fannie, Freddie and the Federal Home Loan Banks, which allowed housing policy to be executed through these entities at a zero budget cost but with a substantial hidden cost to taxpayers.  In fact this is often emphasized as contributing to the lack of political will to control their growth.

Although not as dramatic as for implicit guarantees, the rules of budget accounting cause the cost of all federal credit assistance to be systematically understated in the budget relative to the market value of those commitments (Lucas and Phaup, 2008).  By law, loans and loan guarantees are valued without any adjustment for a market risk premium, in other words, they are discounted at too low an interest rate.  The discrepancies between budget cost and economic value can be large.  For instance, in the case of student loans, our recent estimates suggest that inclusion of a risk premium increases the subsidy rate by more than 20 cents per dollar of loans originated (Lucas and Moore, 2009).

Interestingly, recognition that complying with the normal budget rules for credit would result in a severe understatement of the economic cost of the TARP led to an exemption in that legislation that allowed it to be accounted for using risk-adjusted discount rates.  Similarly, CBO reports on the cost of Fannie and Freddie adjusting for the cost of risk.  Had they not done so, the eventual transfer of the activities of Fannie and Freddie back to the private sector would likely appear to come at a significant cost to the government, making a difficult separation even more problematic. Still, this accounting bias remains in place for most credit activities, including the much expanded guarantee obligations of the FHA going forward.

A second, distinct, reason to be concerned with accounting conventions is that regulations are communicated in terms of accounting numbers, and compliance is measured against them.  Hedge accounting can be particularly problematic in this regard.  Anecdotal evidence suggests that when the two are at odds, financial institutions choose to hedge regulatory risk, not economic risk.  This results in greater risk and expenses for institutions and taxpayers than if accounting conventions were more closely aligned with economic principles.

A final example arises from the acrimonious debate over mark-to-market or fair value accounting rules.  Before the crisis, FASB rules and international accounting authorities were marching steadily towards a greater embrace of fair value concepts, and recent reports suggest that trend will likely resume.  These developments are welcomed by many economists including myself: market prices are generally the best available measure of economic value, they are forward looking and aggregate private information, and they are reasonably hard to manipulate in active securities markets.

The crisis brought a plunge in market values, a disappearance of market prices for many securities, and angry calls from bankers to abandon ruinous fair value rules.  Like many of my academic peers, my initial reaction was dismissive:  If bankers had accepted the high valuations on the upside, why the uproar on the downside?  Upon further reflection, and in a recent Carnegie Rochester paper written with John Heaton and Bob McDonald, we suggest that placing blame on fair value accounting is misplaced, but that real costs are incurred when these accounting rules interact with regulatory capital requirements.  It is the static nature of regulatory capital requirements that have not responded to the greater earnings volatility that accompanies fair value accounting, however, that may deserve the blame.  Over time capital requirements are periodically revised by bank regulators, as is FASB’s definition of capital, but the two types of regulatory actions are not coordinated.  A sensible solution to the problems caused by the interaction of more volatile market-based capital measures and a static capital requirement would be for regulators to periodically redefine the capital requirements to neutralize unwanted side effects of changes in financial accounting standards, and allow them to continue to evolve in a direction that provides better information to markets and to regulators.


Ashcraft, A., M. Bech and S. Frame, 2009, The Federal Home Loan Bank System: The Lender of Next–to-Last Resort?,” manuscript, Federal Reserve Bank of Atlanta

Government Accountability Office, 2009, “Fannie Mae and Freddie Mac, Analysis of Options for Revising the Housing Enterprises’ Long-term Structures.”

Heaton, J., Lucas, D. and R. McDonald, “Is Mark-to-Market Accounting Destabilizing? Analysis and Implications for Policy,” Journal of Monetary Economics., forthcoming. [Working paper version]

Lucas, D. and M. Phaup, 2008, “Reforming Credit Reform,” Public Budgeting and Finance, Winter.

Lucas, D. and D. Moore, “Guaranteed vs. Direct Lending: The Case of Student Loans,” in Measuring and Managing Federal Financial Risk, University of Chicago Press, forthcoming

Mortgage Bankers’ Association, 2009, “MBA’s Recommendations for the Future Government Role in the Core Secondary Mortgage Market

U.S. Dept. of Treasury, “Financial Regulatory Reform, a New Foundation” <http://www.treasury.gov/initiatives/regulatoryreform/&gt;

[1] The Agency is also without a permanent director.  James Lockhart, who was appointed by President Bush to direct the FHFA at its inception, recently resigned.

  1. October 29, 2009 at 10:44 am

    As you rightly observe, the practical, pro-cyclical problem with mark-to-market accounting for mortgage-backed securities was its impact on capital requirements. Although some criticism was directed against the accounting practice itself, the serious complaint was with it’s impact on regulatory capital. In a March 25 piece in Forbes, for example, I wrote that, “Together with similar comments from Barney Frank and the Financial Accounting Standards Board (FASB) on March 16, Bernanke’s comments hinted that regulators might avoid imposing crippling capital requirements and loan loss write-offs on the basis of dubious mark-to-market accounting for illiquid assets.” http://www.cato.org/pub_display.php?pub_id=10070

    Your comments on government accounting, for implicit and explicit loan guarantees, are excellent, clear and very important.

    Alan Reynolds
    Cato Institute

  1. No trackbacks yet.
Comments are closed.
%d bloggers like this: