FROM THE OFFICE OF PUBLIC AFFAIRS

Thank you for allowing me to come before you today to talk about the Treasury Department's study of the financial viability of the government-guaranteed student Federal Family Education Loan (FFEL) program. Last fall, the Office of Economic Policy at the Treasury Department was asked by the National Economic Council to consider in particular the costs and net returns to banks participating in this program under today's rules and the potential impact of the interest rate change scheduled for July 1, 1998.

My office then undertook an intensive analysis of the functioning of the FFEL program for large, for-profit lenders. Although we recognize the diversity of participants in this market, we chose this particular focus because these institutions span the markets in which other lenders operate, and represent the majority of loan origination today.

Our analysis is based on consultation with a number of large originators, holders, and guarantors of student loans, as well as several other banks that finance and follow the student loan business. We also talked with numerous outside experts, including Wall Street analysts of this industry, academic experts on the banking industry, and staff at the Federal Reserve. We relied as well on the helpful earlier analysis done by the Congressional Research Service. The basic results of our analysis are as follows:

• Under the current structure, banks earn returns well above the target rate of return that they would require to participate in the FFEL program.

• Under the interest rate change as currently scheduled, however, banks would earn returns somewhat below that target.

• There are inefficiencies associated with the mismatch of (long-term) student loan interest rates and (short-term) bank financing under the proposed formula. Therefore, joint benefits could be realized to students and lenders from moving back to a short-term index.

• An alternative rate setting formula, suggested by some, that returned the index for student loans to the short-term rate, while holding students at the same interest rate that they would face if the scheduled law change were to take place, would provide a competitive rate of return for banks and maintain bank participation in the program.

The analysis that underlies these conclusions included several steps. The first is to calculate the net income of student loan lenders. The gross income of these lenders is simply determined by the legislated interest rate. We then subtract from this net income several types of costs:

• Cost of matched funds: To finance this loan to students, lenders must raise their own capital funds; the largest cost offsetting the income from student loans is this cost of raising "matched funds". The vast majority of these funds are raised in one of two ways: by direct borrowing, which generally takes place at some markup over a short term interest rate; or securitization, whereby lenders create securities which are directly backed by their Treasury-bill denominated student loans. We obtained a number of quotes for both these types of matched funding under today's system, and the costs were very close to each other.

  After the scheduled interest rate change, however, there will be an important mismatch between the stream of payments from students (which is tied to a long term index) and the sources of bank financing (which are tied to short term interest rates). This mismatch introduces risk into lenders' portfolios, since they can't be sure that the income flowing in from students will match the required payments that banks must make to their financiers. Banks can shed this risk by "swapping" -- i.e. giving up -- their long-term income for income that is tied to a short-term index. The price for doing so is some extra "hedging cost" that is paid to the swap dealer who is willing to bear this risk for the banks. We talked with several major financial institutions, who provided estimates of the cost of the required swaps.

• Servicing and overhead costs: These consist of the expenses of running the student loan portfolio, including a share of expenses such as general management, legal, accounting, human resources, and marketing costs, plus direct costs for servicing the loan such as expenses for collections, borrower correspondence, reporting and account maintenance, and filing guarantee claims. Servicing costs may vary among lenders depending on the size of lenders and the efficiency of their operations; larger lenders probably are more efficient owing to economies of scale, while smaller lenders often sell the loans to secondary markets or contract for servicing.

• Default: The possibility of default on loans in repayment also implies some cost to lenders. This cost is small, however, because the federal government guarantees all loans, with banks bearing only two percent of the risk.

• Fees: Another cost for lenders is the one-time 0.5 percent origination fee paid by loan originators to the Department of Education. We spread the cost of this fee over the average life of a loan to obtain its annual cost. In addition to this origination fee, holders of consolidation loans must pay to the Federal Government a rebate fee, calculated on an annual basis, equal to 1.05 percent of the loan principal plus interest.

• Prepayment: Lenders offering student loans face a risk of prepayment of loans attributable to defaults, loan consolidation, and advance payments from borrowers. Prepayment reduces the life of a loan, thereby increasing the over-the-life cost of certain fixed expenses, e.g., origination support. Consequently, lenders may face a prepayment risk that results in a cost to them. Many financial experts, however, think that prepayment costs are not large for student loans.

We then subtract the sum of these costs from the return to the student loan, to obtain the net income on assets before tax. The second step in our analysis is then to compare this net income to the banks "target" rate of return, or the rate of return that banks require to continue to participate in the student loan program.

As I noted earlier, banks finance the vast majority of their student loans by raising offsetting funding, either by borrowing or by securitizing. But regulatory requirements, and generally safe and prudent bank practice, requires that some part of the loan be financed by the banks' own capital, or equity. This equity in turn, must earn a competitive rate of return to maintain investment in the bank. Thus, ultimately, the target rate of return on a student loan is determined by the share of the loan that must be financed by the banks own capital (capitalization), times the rate of return required on that equity capital. That is, the desire of banks to participate in the FFEL program will be determined by whether they can earn a competitive rate of return on the capital that they must keep to offset the loan itself.

We assume a range for both of the key variables that pin down the bank's target rate of return, the level of capitalization and the rate of return on equity. The minimum level of capitalization required by regulators for a student loans, which are a very safe asset compared to others held by banks, is on the order of 2%. However, regulators also require that, across all their assets, banks have 4-5% capitalization. Moreover, banks today have capitalization of over 7%. Our assumption for capitalization for the report is 4-5%, the overall regulatory requirement across all assets. This is well above the minimum capitalization required by banks to offset their student loans, but it is also below the average capitalization today for banks.

Our assumption for rates of return on equity capital is a range of 10-14% after-tax. The upper bound of this range is slightly above the average return on equity for the 10 largest banks over the past five years, a period of historically high returns on equity. The middle of this range represents the longer run historical average. The lower bound represents both consideration that student loans are less risky than average, so that they may require a lower return on equity capital on the margin, and the fact that these loans may function to some extent as a "loss leader" to attract later business as students borrow for other reasons. Combining our range of capitalization and rate of return on equity capital assumptions, we estimate a "target" rate of return on assets of between 0.8 and 1.15 percent.

The third step in our analysis was to compare the actual rates of return on assets for lenders to this target rate of return, over time, for several different policy scenarios. To reiterate and expand on what I said earlier, our results from doing so are as follows:

• Under the current structure, banks earn returns well above their target range, at about 1.63 percent.

• Under the scheduled interest rate change, banks would earn returns somewhat below their target range, at about 0.53 percent on average over the next five years. Such a reduction need not imply an immediate crisis in the market for guaranteed student loans, but it could be problematic for lenders in the longer term.

• There are inefficiencies associated with the mismatch of (long-term) student loan interest rates and (short-term) bank financing under the scheduled change. Therefore, joint benefits could be realized to students and lenders from moving back to a short-term index.
• We then consider an alternative rate setting formula that has been suggested by some: returning the index for student loans to the short-term rate, while holding students at the same interest rate that they would face if the scheduled law change were to take place. We find that over the next five years, "holding students harmless" in this way would result in an average rate of return on assets to banks of 0.85 percent, which is at the bottom of their range of target rates of return, and thus could maintain bank participation in the program.

• We also confirmed the significant differences in lender costs for loans in-school and in-repayment, as is recognized by the current 0.6 percent differential between the interest rate charged to students in school and in repayment. Continuing to include such a differential between the interest rates charged in school and in repayment would be an effective means of addressing the underlying cost differences in these two cases.

• Finally, we note that the uncertainties involved in this exercise point out the difficulties with regulatory determination of student loan interest rates. An alternative approach would be to use a more market-based mechanism for determining these rates. For the student loan program, this could mean using some form of auction system to determine who would receive the rights to originate student loans. The successful experience of the Health Education Assistance Loan (HEAL) program using an auction system for allocating the insurance authority for HEAL loans suggests that some form of auction could be considered for the Federal Family Education Loan program.

I hope that this summary has served to explain the basic structure of our analysis and our most important conclusions. I am happy to answer any further questions that you have about this report.