October 16, 2009

STATUS

On September 17, 2009, the House of Representatives passed H.R. 3221, the Student Aid and Fiscal Responsibility Act, by a vote of 253-171.  The bill eliminates the Federal Family Education Loan program and shifts all student loans to a government-run system under the Direct Loan program, while creating nine new programs and increasing the federal government takeover of early education, higher education, school construction, and more.  This legislation is now being considered in the Senate.   

CBO has provided an incomplete cost analysis of H.R. 3221, and proponents of the Direct Loan program continue to argue that there is a cost savings in switching to Direct Lending.  However, without a complete and accurate cost analysis of the Direct Loan program, no such assertions can be supported. 

BACKGROUND

The federal government provides both subsidized and unsubsidized loans for higher education (both undergraduate and graduate) using two main programs-the Federal Family Education Loan (FFEL) program, and the Direct Loan (DL) program.  The FFEL loan program offers subsidized loans to students from private lenders.  The FFEL program makes it possible for borrowers to get student loans at low interest rates.  In contrast, the DL program uses the federal government as the lender to provide capital for all loans (as well as interest on subsidized loans).  Under the DL program, the Department of Education serves as the lender and funds come directly from the U.S. Treasury.

The FFEL program was created in 1966 and has provided subsidized loans to students for more than 40 years.  With over 2,000 lenders participating in the FFEL program, serving approximately 4,400 institutions, $70 billion in FFEL loans have been made available to students this year.  The DL program was created in 1993, and has only captured about 34 percent of the total loan market at its height, proving to be less efficient and responsive than the FFEL program.  With approximately 1,700 institutions currently participating in the DL program, DL has made available $22 billion in student loans this year, and many of these schools were forced into the program as the capital markets worsened last year and the private lending industry struggled.

THE HIDDEN COST OF DIRECT LENDING

When the DL program was created in 1993, proponents of the program believed that taxpayers would ultimately save money if the federal government served as the lender to students, rather than going through private lenders, who receive a subsidy to keep loan rates low for students.  However, over the years this has not proven to be the case and the program has continuously lost money.  Numerous independent studies have concluded that the DL program has not provided the promised savings and is actually paying out more in interest payments than it has taken in from borrowers. 

Historically, the DL program has paid more in interest to the Treasury than the amount of interest it has gotten back from borrowers.  A 2004 GAO report reevaluated the DL cash flow and found that the program borrowed $137 billion from the Treasury during FY 1995-2003.  Of this amount, $92 billion was outstanding as of September 20, 2003.  In FY 2003, with appropriations of $2.7 billion received, the total cash outflows of the DL program still exceeded the total inflows by $10.7 billion.  This was largely due to the fact that the interest the program paid to the Treasury was significantly higher than the interest rates collected from borrowers.  GAO also found that there were significant misestimates causing the Department of Education to claim receipts that were higher than their actual inflows.  In fact, the Department itself claims that "interest receipts are a difficult cash flow to estimate because of the complexities associated with periods during which students are not required to make interest payments." 

According to a report released by PriceWaterhouseCoopers in 2005, the DL program is not more cost effective than the FFEL program.  The report shows that current budget scoring methods do not consider important factors when comparing the costs of FFEL to DL.  For instance, the report noted that budgetary predictions of the cost of DL as opposed to that of FFEL do not take into account the impact of interest rate variability and that DL costs are more sensitive to changes.  In reality, the DL program continues to pay more out in interest than it receives and the cost to administer the program has increased almost by 600 percent. 

Proponents of the DL program make the following arguments;

• Unlike the DL program, there are several subsidies paid to private lenders and guaranty agencies within the FFEL program who serve as "middlemen."
• By statute, interest on subsidized loans does not accrue for the student while the student is in school or in deferment. In the FFEL program, the federal government pays the interest to lenders on behalf of students. In the DL program, interest is just not charged since the government is the lender.
• In the DL program, the federal government can borrow funds directly from the Treasury, making the capital cost very low. When students are charged 6.8 percent on their loans, the federal government profits from the difference.
• CBO only looks at the cash flows of the DL program when determining the "costs" of the program and neglects the costs associated with increasing the national debt or the lack of cash flow while students are in school.

COST ESTIMATES

When CBO initially scored H.R. 3221, they determined that the estimated subsidy cost shown in the budget is lower for the DL program than for the FFEL program, and that enacting the bill would yield net budgetary savings by shifting new lending from the guaranteed loan program to the direct loan program.  On balance, CBO estimated that enacting H.R. 3221 would reduce direct spending by $13.3 billion over the 2009-2013 period and $7.8 billion over the 2009-2019 period. 

However, assuming appropriation of the necessary amounts, implementing the bill would actually increase discretionary spending by at least $13.5 billion over the 2009-2019 period.  The CBO estimate reflects the bulk of the likely discretionary costs under H.R. 3221; but because they have not completed a comprehensive estimate of all effects that would be subject to appropriation action and do not take into account the cost of market risk (which reduces savings by another $33 billion) this CBO score is far from complete. 

In a letter to Senator Gregg on July 27, 2009, CBO Director Douglas Elmendorf states that the "billions in savings" in H.R. 3221 are deceptive.  According to the CBO Director, current budget scoring rules do "not include the cost to the government stemming from the risk that the cash flows may be less than the amount projected" (that is, that defaults could be higher than projected).  CBO found that after accounting for the cost of such risk ... the proposal to replace new guaranteed loans with direct loans would lead to estimated savings of about $47 billion over the 2010-2019 period-about $33 billion less than CBO's estimate under the standard credit reform treatment."  (Costs to the government stemming from the risk that the cash flows may be less than the amount projected.)

Due to this market risk assessment, CBO reported a new analysis of H.R. 3221 showing that it will actually cost taxpayers billions more than Democrats have acknowledged.  When CBO re-examined the cost of reforms to the Pell Grant program included in H.R. 3221, they found an additional $11.4 billion in spending, costs that will be passed on to taxpayers resulting in further deficit spending.  In this re-estimate, CBO used more current data to predict the number of college students expected to quality for Pell Grants and the amount of the grants under the Democrats' proposed legislation.  Their review predicted an additional $10.5 billion in direct spending over ten years, along with $900 million in additional discretionary spending. 

Given the limitations of CBO's scoring rules, the fact that most of the bill's "savings" are poured back into more entitlement spending, and the increased cost of the Pell Grant program, it is very likely that the bill will increase federal spending by $32-39 billion over ten years.

---

The Limitations of Budget Score-keeping in Comparing the Federal Student Loan Programs; http://www.studentloanfacts.org/NR/rdonlyres/65DDECF9-3020-4C6A-8C8F-B568556FEA64/2456/PWCReport.pdf.

http://www.cbo.gov/ftpdocs/102xx/doc10295/Gregg_StudentLoans__09-07-27.pdf