Budget Blog

The federal government makes and guarantees hundreds of billions of dollars in loans every year. These credit programs serve a wide variety of public policy purposes, ranging from making homeownership or attending college more affordable to making it easier for small businesses to expand. In recent years, a debate has emerged over the best way to account for these loans and loan guarantees, with some critics contending that official estimates understate the true costs. And while the critics have some valuable points to make about the broader societal costs and benefits of federal credit programs, the simple fact is that the current method of accounting is already extremely accurate and the proposed change would deliberately yield incorrect estimates of actual fiscal costs. This shortcoming of the proposed change in accounting is unavoidable and should give enormous pause to any advocate of fiscal responsibility.

Current estimates of federal loans and loan guarantees already take into account the chance that a borrower will default, or pay late, or pay early. And they also already take into account the fact that a dollar repaid tomorrow is not the same as a dollar repaid today. And most importantly, the methods used today have proven themselves to be extremely accurate in terms of true fiscal effect. The Office of Management and Budget found that, since 1992, the initial cost estimates of all credit programs differed from their actual cost by less than 1 percent of the face value of all the loans and guarantees.

The objection, then, is not that the current method is inaccurate. Rather, critics want to incorporate a non-budgetary cost into official estimates. Critics contend that official cost estimates should include something called “market risk.” Market risk is different from the chance of a default, which the current estimates already incorporate fully. Market risk is the risk that remains in a portfolio of investments even if those investments are perfectly diversified. In other words, it is the risk that broad, underlying changes in the market will affect all of your investments simultaneously and in the same direction. Making loans, whether to homebuyers, businesses, or students, exposes the federal budget to this market risk. And since most people prefer certainty to risk, critics of the current method argue that this market risk should show up as a cost in budget estimates and projections. Advocates of this approach call their method “fair-value accounting.”

The biggest problem, of course, is that while market risk may carry societal “costs” in a broad sense, those are not the same as budget costs. There are many costs and benefits of federal policies that are not captured in the simple number that describes how much money will be spent or raised by those policies. If the main purpose of a budget is to give decision-makers an accurate view of how much money is going out and how much money is coming in, then “fair-value” estimates will undermine that purpose. To put it as clearly as possible, a “fair-value” estimate of federal credit programs will not equal the actual dollars those programs will cost.

While this debate may seem technical and arcane, it has important implications for pressing policy decisions before Congress right now. For instance, at the end of the fiscal year, the Export-Import Bank is due to be reauthorized. A “fair-value” estimate of the bank’s credit activities would imply that it loses money when, in fact, it returns a profit to the US Treasury. The “fair value” approach might reflect a broader definition of “costs” than the current method, but it also gives a false impression of the actual fiscal effects.

There are other objections to “fair-value” accounting as well. For example, many aspects of the federal budget are subject to market risk, including tax receipts and Social Security obligations. If we apply “fair-value” methods only to credit programs, it will be impossible to make apples-to-apples comparisons between different parts of the budget. Others have also argued that it might not be appropriate to assume that governments are “risk-averse” in the same way that individuals are.

Even if these other objections didn’t exist, the simple fact that “fair-value” estimates do not match actual budgetary effects should be enough to reject its use as an official measure. The federal budget still faces a long-term structural gap that must eventually be addressed and accurate budget estimates are absolutely necessary to develop appropriate policy responses. It is certainly appropriate to consider “market risk” and other non-budgetary costs when evaluating legislative proposals, but incorporating non-budgetary costs into a budget would be a mistake.

For more on this subject:

A July 2012 issue brief from the Congressional Research Service provides an excellent overview of the main issues pertaining to federal credit programs, including a discussion of “fair value” accounting.

In June of 2013, the Center on Budget and Policy Priorities released an analysis of H.R. 1872—a bill requiring official budget estimates to use “fair-value” methods—that provides a good overview of the subject, as well as a summary of the main objections to both the bill, in particular, and the method in general.

Dr. Deborah Lucas, a professor at MIT and formerly at the Congressional Budget Office, wrote an issue brief on the topic while she was at CBO. Dr. Lucas is one of the preeminent supporters of the idea of “fair value” accounting. Her issue brief is entitled “Fair-Value Accounting for Federal Credit Programs,” and was released in March 2012.

The Center for American Progress, which opposes “fair value” accounting, penned an issue brief entitled, “An Unfair Value for Taxpayers,” on the subject, released in February 2012.

The Office of Management and Budget has also written on the topic of “fair value” accounting, most recently, in the president’s fiscal year 2015 budget submission.

Finally, a scholarly article from New York University law professor David Kamin addresses many of the arguments in favor of “fair value” accounting and offers his take on the shortcomings with each one. That article is entitled, “Risky Returns: Accounting for Risk in the Federal Budget.”