The Government Accountability Office just dealt Republicans a blow in the long-running debate over whether government lending programs make or lose money for taxpayers.
To little notice, the agency released an 83-page report at the end of last month siding definitively against Republicans such as Senate Budget Committee Chairman Mike Enzi, who wants to change the way credit programs are accounted for in the budget.
At stake is the future of the government’s roughly $9.4 trillion portfolio of government loans and credit guarantees. The loan book includes student loans made by the Department of Education, the mortgage-backed securities guaranteed by Fannie Mae and Freddie Mac, and the financing offered by the Export-Import Bank, among many other programs.
All those programs are accounted for in the federal budget using a method unique to the government, one that Republicans in recent years have argued ignores the true costs for Uncle Sam to extend or back loans. Many prominent experts have agreed that it needs to be changed, including the nonpartisan Congressional Budget Office and an outside group of financial economists.
However, the GAO concluded, in a unusually definitive statement, that the alternative favored by Republicans is “not appropriate for use” in budgeting and that introducing it would only create headaches for government budgeteers.
Using the current accounting method, the federal government is expected to make hundreds of billions of dollars on its $1.2 trillion student loan portfolio, which has led Democrats such as Sen. Elizabeth Warren, D-Mass., to argue that Uncle Sam is profiting off students. Under the change the Republican want, however, the government would lose almost as much, implying that there is a big subsidy embedded in student loans. Similarly, the controversial Export-Import Bank, which proponents say makes profits for taxpayers, would be shown to receive subsidies.
At issue is the inherent difficulty of budgeting for the cost of loans.
With a spending program, the accounting is easy. To find out how much it’s spending, the agency simply counts up how much money goes out the door each year and subtracts any associated revenue, a method known as cash-flow accounting.
Credit programs are harder. A Department of Education student loan, for example, would appear to lose thousands of dollars in the year it was disbursed and then make profits in the years it was repaid, under cash-flow accounting.
To solve that problem, Congress in 1990 passed a bill to switch the government to a form of what is known as accrual accounting.
Here’s how it works: When the government makes a loan, it tallies up all the expected future payments, accounting for potential losses from, for example, student borrowers failing to repay their debt. Then it calculates a present value of those future expected payments, under the economic principle that a dollar today is worth more than a dollar in the future. It does so by dividing the payments using a “discount rate,” namely the interest rate on a Treasury security of the same duration of the loan in question.
To determine how much the loan cost, the government then subtracts that present value of future repayments from the initial loan amount.
Where the controversy comes in is that the government uses the interest rate on Treasuries for the discount rate. Treasury securities are generally considered risk-free and consequently offer lower interest rates, and using them for discount rates misses a key source of risk, the CBO and other experts have argued.
Specifically, the budgeting misses “market risk,” the risk that the government might lose money at a particularly bad time.
Take a student loan: If a student goes to a bank for a college loan, he will be charged a higher rate. In part, that rate reflects the possibility that he could default during a banking panic, just when the bank is desperate for incoming loan interest revenue to pay off depositors. If the government were backing the loan, however, there is no such risk — the government is not going to become insolvent in a panic, because it can always raise taxes.
The CBO has argued that, as a result, the 1990 credit reform budgeting misses the true cost of capital involved in government loans. To correct for the problem, the government should switch to “fair-value” accounting, in which a typical market interest rate, rather than a Treasury interest rate, would be used to discount cash flows.
Because the GAO has sided against fair-value accounting, liberals’ defense of the 1990 credit reform accounting has better intellectual backing. It’s “now viewed as an issue, I think, where there are arguments on both sides and that authoritative experts disagree,” said Richard Kogan, a budget expert at the Center for Budget and Policy Priorities and former Obama budget official. The center, a left-leaning Washington think tank, has criticized the idea of fair-value accounting for years.
The intellectual positioning could prove important in future years. Last year, Republicans in the House and Senate voted in their budget resolutions to implement fair-value accounting for major credit programs. Enzi remains a “big fan” of fair-value accounting, a spokesman said.
The real-world effect of such a change would be to endanger many programs’ funding, Kogan said. “When there are budget pressures, the programs that now appear to be more expensive than they previously appeared would be more likely to be subject to downward squeezes.”
Yet the report, which was requested by Sen. Chris Coons, D-Del., has not swayed some of the most prominent proponents of fair-value accounting.
“It doesn’t take a degree in finance to understand that there is a subsidy involved if the government is making risky student loans and mortgages on terms that no private financial institution could possibly offer without going bankrupt,” said Deborah Lucas, a professor of finance at the Massachusetts Institute of Technology who worked on budgeting for credit programs at the CBO.
The GAO interviewed a number of experts to conclude that fair-value accounting wouldn’t be more accurate and would be difficult technically to implement. But what was new and original in the report is that researchers looked through 14 years’ worth of agencies’ credit subsidy estimates, and then compared them to the actual payments made. They found that the estimates, made using the credit reform methodology, ended up very close to the actual cash amounts the Treasury had to pay out.
Experts interviewed by the Washington Examiner said the GAO’s comparison of actual payments to estimates was beside the point, because the real argument is at a conceptual level. The questions are whether taxpayers are being fairly compensated for the risk they are taking on and whether borrowers are getting subsidized when the government offers credit.
Lucas said the review “has no bearing on the question at hand, which is whether credit risk should be treated as costly to taxpayers and the government.”
Donald Marron, a former acting director of the CBO now at the Urban Institute, said the difference was that defenders of the current approach are focused on the fiscal effects of the budget, whereas fair-value accounting would identify who was receiving subsidies. With fair-value accounting, “you could do a head-to-head comparison between subsidized lending and a grant program on the same terms because you could see how much taxpayers were giving up to help whoever it was we were trying to help, whereas under the current system that gets distorted,” said Marron, who favors a third method of accounting to identify costs and subsidies.
That distortion is one of the key conservative complaints of the current approach. Indeed, while government spending growth has slowed in recent years, credit growth has not. Along with its analysis of different budgeting approaches, the GAO reported that government lending and credit guarantee programs have doubled since 2008.
