A Proper Accounting: The Real Cost of Government Loans and Credit GuaranteesPublished: December 05, 2012 in Knowledge@Wharton
While all eyes are turned to the U.S. government's enormous debt, few have given equal attention to the massive costs and risks embedded in another of the government's financial functions: its role as lender rather than borrower.
Flaws in the way the government accounts for its loans and credit guarantees understate the costs that taxpayers are bearing with student loans and other credit programs totaling more than $2.5 trillion, plus more than $5 trillion in mortgages backed by the federally owned companies Fannie Mae and Freddie Mac. In fact, a proper accounting -- like that required of most businesses -- would make the government's budget deficit even larger than the officially reported amount.
Those are the conclusions of the Financial Economists Roundtable (FER), a group of prominent academics that since 1993 has produced an annual statement based on financial analysis of a critical policy issue. In October, the FER published this year's statement, Accounting for the Cost of Government Credit Assistance, the product of its July meeting in Ontario, Canada. The FER is administered by the Wharton Financial Institutions Center.
"The federal government is the world's largest financial institution, but policymakers and [government] managers are handicapped by an accounting system that is seriously deficient," says Deborah J. Lucas, finance professor at MIT's Sloan School of Management and author of the FER statement. "The accounting standards that the government sets for private financial institutions require far greater transparency than the rules that it imposes on itself."
Student loans and mortgages backed by the Federal Housing Administration, among more than 100 other lending programs, contain potential losses that are much more costly than what current accounting suggests, according to the FER. Fannie and Freddie are "a recent and costly example where the government treated its implicit [loan] guarantees as having no cost," Lucas says. Those companies, private but operating as government-sponsored enterprises, collapsed in 2008 and were taken over by the federal government, costing taxpayers upwards of $120 billion. The markets had long assumed the government would stand behind the two firms, though it was not legally required to do so, and it frequently disavowed any intention to back them up.
Buying Low, Booking High
The FER statement, signed by 33 financial economists, faults the government's use of yields on Treasury securities as the "discount rate," an interest rate that determines the value in today's dollars of income to be received in the future. Treasury yields, says the FER, fail to account for many of the risks government loan programs face, making that uncertain future income look more valuable than it really is. "A number of risks simply don't get priced in the rate on Treasury bills," says Wharton finance professor Richard J. Herring, executive director of the FER, co-director of the Wharton Financial Institutions Center and a signer of the statement.
Illustrating this process, the FER says that in 2009, the Treasury Department paid $220 billion for bundles of mortgage-backed securities purchased in the open market, and then reported a $5 billion gain by booking them at higher values produced by the lower discount rates justified by government accounting rules. "It is the same kind of deceitful accounting that would take place if you were to take over General Electric and discount all the cash flows at the Treasury rate," Herring says. Doing that would make it look as if you had added value to GE overnight, even though you had actually done nothing except destroy value through government interference, he observes.
A 'Budgetary Illusion'
While a better assessment of the cost of government guarantees might not have prevented problems like the collapse of Fannie and Freddie, it would have raised a red flag to prevent government lending and loan guarantee programs from becoming so large, thus reducing losses borne by taxpayers, Herring points out. The rapid growth of lending programs, such as those for students, is easier for Congress to authorize when the costs appear smaller than they really are, the FER statement argues. "The failure to fully account for the cost of risk has several significant consequences," the statement says. "For one, it has sometimes resulted in the budgetary illusion that government credit programs reduce the government deficit."
For example, a group of lending programs that includes student loans and mortgages guaranteed by the Federal Housing Administration are projected by the Congressional Budget Office to reduce the federal deficit by about $45 billion in 2013, while an accurate accounting would show them likely to cost taxpayers $11 billion, according to the FER.
Improper accounting also encourages the government to use loans and loan guarantees when it sometimes would be better policy to offer grants with a similar cost to taxpayers. "One possible reason for the increasing reliance on credit assistance over grant assistance for higher education is that in the budget, student loans look like they are saving the government money," Lucas says. In fact, these loans are more costly than the accounting suggests, she adds.
What Discount Rate?
As is true for the private sector, money that will flow into government coffers in the future is worth less than money on hand today. One hundred dollars to be received a year from now is not worth as much as $100 today, because the $100 could be invested to grow larger over the next year. The smaller the discount rate -- the closer it is to zero -- the greater the present value of future income. So, with loans, whether government or private, the issue is: What discount rate should be used?
Current government accounting rules require the use of Treasury yields, which are lower than yields of other types of securities, such as corporate bonds. Treasury yields are lower because investors believe the government's promise to pay back its lenders is ironclad -- much stronger than the assurance investors get, for example, from car manufacturers, airlines or other corporate borrowers. Treasuries are often called "riskless," because there is virtually no risk the government will default.
The FER argues that Treasury yields are inappropriate for the discount rate on government loans because those loans involve various risks that Treasury yields don't reflect -- market risk, prepayment risk and liquidity risk. The lower discount rate provided by Treasuries makes the future income look more valuable than it is. Using the wrong discount rate, says Lucas, is like using the price of gold to assess the value of a block of lead -- it just doesn't work. "If you use the price of Treasury securities to try to assign a price to a risky loan, you get nonsense."
But if government accountants don't use Treasury yields for the discount rate, what should they use? "The appropriate methodology for evaluating the cost of government credit assistance is to use the same discount rates that are used by private businesses and investors who might provide such assistance," according to the FER. In many cases, those rates would be higher than Treasury rates, reducing the present value of future income.
Each type of government loan or guarantee will require identifying the best methodology for setting the discount rate, Lucas says, noting that private lenders do this routinely. "There are standard ways to go about doing this. Private-sector financial institutions are responsible for reporting fair values [of loans and guarantees], so there is an entire infrastructure for providing these values...." Government accountants, Herring adds, could use as a guide the values that private markets place on loans similar to those issued by the government.
But it will not be easy to get the government to change its accounting practices in the ways the FER recommends. A bill to do so, H.R. 3581, passed in the House early this year but is stalled in the Senate. With so much concern about the size of the deficit, Herring notes, Washington may be reluctant to reform an accounting system that, by inflating the value of government assets and understating the cost of government guarantees, makes the deficit look smaller than it really is.