CBO Letter To Sen. Gregg Questions SAFRA Savings Figure (Updated)

CBO has posted a letter from Doug Elmendorf, CBO Director, to Sen. Judd Gregg of New Hampshire highlighting several shortcomings to CBO's recent scoring of the Student Aid and Fiscal Responsibility Act. 

Here were the two items that Elmendorf believes should be adjusted:

• First, he knocked off $7 billion from the $87 billion estimate based on administrative costs of the program which are categorized as discretionary expenses (bold type is my own):

"In its July 24, 2009, cost estimate for H.R. 3221 (the Student Aid and Fiscal Responsibility Act of 2009, as approved by the House Committee on Education and Labor), which would
incorporate the President’s proposal, CBO estimated that replacing new guarantees of student loans with direct lending would yield gross savings in federal direct (or mandatory) spending of about $87 billion over the 2010–2019 period.2 (Mandatory spending is governed by existing provisions of law and does not require future appropriations.) About $7 billion of those savings would represent a reduction in the administrative costs of the guaranteed loan program, which are recorded in the budget as mandatory spending. In contrast, most of the administrative costs for the direct loan program are funded in appropriation bills and recorded as discretionary spending. Thus, of the $87 billion reduction in direct spending, roughly $7 billion would be offset by an increase in future appropriations for administrative costs, for an estimated net reduction in federal costs from the President’s proposal of about $80 billion over the 2010–2019 period."

• Second, Elmendorf argues that a more appropriate method of measuring the subsidy difference between FFELP and Direct Lending would incorporate a market-based risk adjustment or the risk that taxpayers are taking on in making unsecured loans to young borrowers:

"Alternative estimates of the value of the programs’ subsidies that might better reflect the costs they represent for the government would incorporate the estimated cost of the market risk that taxpayers bear through such lending—a cost analogous to the higher returns that private investors expect for making risky investments."

What is the impact of these changes?

• The net impact of the using risk-adjusted discount rates, would significantly reduce projected savings to $47 billion or $33 billion less than the $80 billion CBO had originally estimated:

"Applying a set of risk-adjusted discount rates to the cash flows from the government’s student loans would raise the subsidy rates for both student loan programs, but the rate for the direct loan program would increase by more than the rate for the guaranteed loan program because of differences in the timing and riskiness of the estimated cash flows. CBO estimates that if projected savings for the President’s proposal were calculated using risk-adjusted discount rates, those savings would be $47 billion over the 2010–2019 period—a difference of $33 billion relative to CBO’s cost estimate for H.R. 3221 issued on July 24."

• Elmendorf goes on to note that this risk-adjusted approach was recently incorporated into TARP budget estimates: 

"Nevertheless, CBO believes that risk-adjusted subsidy rates provide useful information about the cost of federal programs in terms of the value of the economic resources that are devoted to those programs. The Congress adopted the approach of incorporating the cost of market risk into budget estimates for the 2009 enactment of the Troubled Asset Relief Program (TARP). That approach requires that the costs of assets purchased under the program be estimated using a present-value approach that, except for its requirement of an adjustment for the cost of market risk, is similar to the way loans and loan guarantees are evaluated under the Federal Credit Reform Act."

The question now is what impact, if any, this new revelation from the CBO director (who recently injected himself into the healthcare debate recently too)  will have on the SAFRA legislation, once it gets to the House floor, which could be as early as this week or may be postponed until after the August recess. 

The other thing that perplexes me a bit about the CBO budget estimates is that it uses the "old FFELP" model as a point of comparison assuming private capital is coming from these lenders to fund the program (bold type is my own). 

"In the guaranteed student loan program, the federal government faces a more complicated set of payments. It does not disburse a principal amount (loans are disbursed by private lenders) but instead receives some up-front fees, makes a stream of subsidy payments (known as special-allowance payments) to lenders, partially compensates lenders for loans that go into default, and pays certain borrower benefits, in addition to various other receipts and payments."

The ECASLA financing programs, in which the Department of Education purchases loans or participation interests in FFELP loans is estimated to account for 50% of FFELP loans for 2008-09 (we will have a definitive figure on this in about 3 weeks).  This program has been extended for 2009-10.  The word on the street is that CBO is working on cost and savings estimates for the Student Loan Community proposal, so stay tuned. 

-----------------------------------------

Updated (7/28/2009)

It hasn't taken the Democrats long to respond to this letter.  Here is the press release from Rep. George Miller's office ("Republican's Try to Cook Books On Historic Student Aid Bill").  Here is the lead quote provided by Miller:

“It’s clear that Republicans didn’t like the truth – that our legislation generates almost $90 billion that could be used to help students, families, and taxpayers – so they shamelessly decided to have a little fun with the numbers,” said U.S. Rep. George Miller (D-CA), the chairman of the House Education and Labor Committee and the author of the legislation. “This is nothing more than a desperate attempt to confuse the public and manipulate a clear determination by CBO that switching to the Direct Loan program is the most sound, fiscally responsible policy decision we could make for families and taxpayers. This is yet another predictable political gimmick from a party that is proving that they will do or say anything – no matter how misleading – to block reforms that will make a tremendous difference in improving the lives of America’s families.”

Miller also made the point that I had made earlier that the alternative in CBO report reflects the "old FFELP" model and not the current market realities:

This alternative estimate ignores current student loan market conditions, under which the federal government is currently supporting 60 percent of all federal student lending. This estimate assumes normal credit market conditions, under which the federally guaranteed student loan program functions entirely independently, as it used to. It does not reflect today’s reality: that the federally-guaranteed student loan program is on life support. The federal government, through both the Ensuring Continued Access to Student Loans Act, and the Direct Loan program, is now financing 60 percent of all federal student loan activity. This current ECASLA-supported, federally-guaranteed student loan program is very similar to the DL program – it generates the same cash outflow upfront, but the funds are sent to lenders who then give it to borrowers. If this alternative estimate was based on this current reality, it would likely show a higher market-risk adjusted subsidy rate for the Federal Family Education Loan Program – again reflecting that the program is on life support.

Let's see what happens when the Student Loan Community proposal is scored...

-----------------------------------

Related articles:

• Bloomberg:  "Obama's student loan plan may save less than expected"
• Inside Higher Ed:  "Alternative price tag for Obama student loan plan"
• Higher Ed Watch Blog:  "More scare tactics from student loan industry and friends"
• Reuters:  "U.S. Rep. Miller defends student loan bill"