Measuring the Costs of Federal Student Loans, Fairly
A recent GAO report shows how the costs of federally-financed loans have consistently been under-estimated over time. But it fails to recognize that bank-based student loans would have cost the federal government much, much more.
The Government Accountability Office (GAO) recently was asked by House Republicans to examine the costs of federally-financed direct student loans. The just-released GAO report accurately notes that using federal funds to finance student loans have cost a lot more than was initially estimated. But the report ignores the fact that federal costs of student loans would have been much higher over time if the shift away from bank-financed loans had not occurred.
First, some context. When Direct Loans were first enacted early in the Clinton administration, a major argument was that relying on federal capital would save the government money compared to the traditional reliance on using private capital to finance student loans. While some said this meant the government would ‘profit’ from financing student loans, advocates’ broader intent was to use the savings gained from shifting to federal financing to reduce the deficit or to free up funds to pay for other purposes.
The Clinton administration wanted to use the savings for enacting a national service plan and to allow student borrowers to repay their loans based on their income after they left school. The Obama administration made a similar argument when it pushed for a total shift to federal financing in 2010 and used the additional savings to help pay for the shift to Obamacare.
So how did things work out? The GAO report shows how and why costs were mis-estimated over the past 25 years. The bottom line of the GAO Report is that federally-financed loans have cost substantial sums over the past quarter century. The report shows that the federal government spent nearly $200 billion for interest subsidies, defaults and other costs from 1997 to 2021. This contrasts to the initial CBO and other cost estimates which indicated the government would save more than $100 billion over time from shifting to a system in which the federal government provided the initial capital rather than subsidize banks and other private holders for the use of their money. Thus, there has been a swing of more than $300 billion from the ‘profit’ projected in the 1990s to the actual costs of federally-financed student loans over a 25-year period.
The GAO report also makes the very useful distinction between cost increases from programmatic changes in the legislation once it was enacted compared to inaccurate projections of program participation, interest rates, and the incidence of default. It estimates that roughly 40 percent of the $300 billion change in overall costs were due to programmatic changes such as which students were eligible to borrow, how much they could borrow, and what they would be charged. The biggest of these programmatic changes is the pause in repayments over the past two years due to COVID which increased costs by an estimated $100 billion.
The remaining 60 percent in re-estimates are related to mis-estimates of assumptions about program participation, default rates and interest rates. This reflects the fact that estimating future interest rates is an inherently difficult task. Forecasting college costs and their impact on the need to borrow is just as difficult.
The Relative Costs of Federally-Financed and Bank-Based Student Loans
But the GAO report fails to focus on two key aspects of federal student loans which tell a much different story. One is that the nature of the federal credit budgeting rules that were adopted in 1990 budget legislation which require the government change its accounting rules to estimate the present value of the future costs of student loans.
The problem here is that the estimates of the course of future interest and default rates inevitably will be wrong to some significant extent. The 1990 legislation recognizes this inherent challenge by requiring CBO to re-estimate the costs of past cohorts of borrowers when it estimates the costs entailed in making loans to the new cohort of borrowers each year. A look at various CBO estimates over time shows that re-estimates of the costs of previous cohorts of loans often exceed the costs estimated for the new cohort of borrowers.
But the biggest missing component in the GAO report is that it doesn’t examine how much more federal costs of student loans would have been if the federal government had continued to rely on banks and other private entities for student loan capital. Indeed, the GAO report and some other similar efforts leave the impression that costs would have been lower if bank-based lending had continued to be the basic structure for federal student loans.
But the fact is that the federal costs of student loans over the past twenty-five years would have been much higher if the direct loan legislation had not been adopted in 1993 and then modified in 2010 when full scale direct lending was adopted. The fact is that it is not unreasonable to assume that the costs of student loans would have been twice as high or more if we had continued to rely on private lenders to provide the capital for student loans.
This can best be seen by how much the pause in payments over the past few years would have cost under privately financed system. With direct loans, the cost of the pause is what it would cost the government to borrow to replace the interest and principal not collected on these loans during the pause. But the federal cost of the pause if private entities hold the loans would have been much higher because the government would be obligated to pay lenders statutory interest rates (which are much higher than the government’s cost of money). In addition, the government also would have been obligated to pay off the remaining principal and accrued interest on defaulted loans rather than writing off these bad loans. Taken together, the cost of the pause for privately-financed loans could easily have been at least 50 percent higher than the costs under a federally-financed system.
The key point is that while it is hard to estimate the long-term costs of student loans when the loan is initially made, we can know with confidence that the costs of using federal funds to provide the capital will be lower than if the government is obligated to pay private sector lenders for the use of their money for student loans.
