Skip Main Navigation
Office of Management and Budget
President's Budget
Information &
Regulatory Affairs
Legislative Information
Agency Information

March 20, 1997



SUBJECT:  Agency Reports to Congress on Loan Portfolio Valuation

1. Purpose. This Bulletin provides guidance to Federal agencies on reporting requirements, pursuant to the Debt Collection Improvement Act of 1996 (DCIA) (See Attachment A, Section 31 001 of the DCIA (P.L. 104-134)).

2. Background. DCIA directs each Executive branch agency with current and delinquent, collateralized nontax debt (i.e., credit programs) to report to the Congress for each financing and liquidating account the following:

  • the cumulative balance of such debts outstanding;
  • the estimated net present value of such debts;
  • the estimated net proceeds that would be received by the Government, if such debts were sold; and

  • the annual administrative cost.

In addition, agencies must also report on the marketability of all nontax debt held.

Data collected in accordance with this Bulletin will be used by Congress for the consideration of the sale of loans. The DCIA instructs OMB, in consultation with Treasury, to develop standards and determine the information necessary for agencies to comply with the reporting requirement. Agencies should note that these reporting requirements are similar to, but more comprehensive than, the reporting requirements in the report language of the Treasury-Postal Appropriations Act of 1996 (P.L. 104-52) covered in OMB Bulletin No. 96-04 (See Attachment B, OMB Bulletin No. 96-04).

3. Action required. Each Executive branch agency should prepare a report on loan portfolio valuation, based upon the criteria listed below. Reports should be transmitted to the Congress no later than April 30, 1997. The final draft of the report must be reviewed by the OMB examiner(s) with primary responsibility of the applicable credit accounts prior to the transmittal of the report to the Congress.

  • Information should be reported on portfolios with outstanding balances of collateralized debt consistent with the following definitions.
  • Collateralization. Any asset pledged to the lender for the term of the loan. Assets are anything of commercial value owned by the borrower, including revenue-backed assets.
  • Outstanding balances greater than $500 million. The sum of outstanding end-of-year balances in a financing accounts, plus a liquidating account must be greater than $500 million, as of September 30, 1996.

The reporting requirement does not apply to a portfolio, in whole or in part, if one or both of the following characteristics are present.

  • Legal prohibitions. Programs with legal prohibitions against the sale of loans are excluded, but should be identified together with the legal citation.
  • Sovereign nations. Programs that extend credit to sovereign nations are excluded. If a program extends credit to both sovereign and non-sovereign borrowers, then only the non-sovereign portion of the portfolio is subject to the reporting requirement, provided that the aforementioned criteria are met.

4. Required information. At a minimum, the report must include the following for each financing and liquidating account:

  • the cumulative balance of such debts outstanding;
  • the estimated net present value of such debts (expected cash flows to the Government);
  • the estimated net proceeds that would be received by the Government, if such debts were sold (expected cash flows to the private sector); and
  • the annual administrative cost,

associated with

  • current debts outstanding (current direct loans),
  • delinquent debts (delinquent direct loans),
  • guaranteed loans (current guaranteed loans), and
  • defaulted loans previously guaranteed that result in a receivable (defaulted guaranteed loans).

The marketability of all debt (collateralized and uncollateralized, direct and guaranteed loans, current and delinquent and defaulted) must also be included in the report.

5. Methodology. The information included in the report must be calculated, in accordance with the following standards:

  • The value of the portfolio is to be assessed as of September 30, 1996.
  • The agency definition of current, delinquent, and defaulted must be included in the report.
  • Estimates of net present value and estimates of net sale proceeds must be calculated using an analysis of the remaining cash flows associated with the outstanding portfolio, exclusive of administrative expenses. The weighted average numbers of years remaining to maturity will determine the appropriate discount rate (consistent with guidance on modifications in OMB Circular No. A34, section 12.10). State this discount rate in the report.
  • Assumptions affecting the cash flows used in the net present value estimates and estimates of net sale proceeds must be stated. The estimated cash flows for private sector ownership may differ from the cash flows under Government ownership as a result of these assumptions. (See Attachment C, Analytical Perspectives, Fiscal Year 1997 Budget, Chapter 21, "Loan Valuation").
  • Assume direct loans or loan guarantees are sold to the private sector without Federal guarantee (recourse).
  • Estimate the annual administrative cost associated with administering the portfolio. Administrative costs are defined in OMB Circular No. A-11, section 33.5 (n).
  • Marketability is the speed and ease with which a loan can be bought and sold in the private sector. Describe portfolio characteristics that improve or inhibit sale of loans as well as the degree of private sector interest in purchasing such loans.

6. Inquiries. Questions concerning this Bulletin should be directed to the OMB representative(s) with primary responsibility for the review of the relevant credit accounts.

Franklin D. Raines


Attachment A

"(4)(A) Within one year after the date of enactment of the Debt Collection Improvement Act of 1996, each executive agency with current and delinquent collateralized nontax debts shall report to the Congress on the valuation of its existing portfolio of loans, notes and guarantees, and other collateralized debts based on standards developed by the Director of the Office of Management and Budget, in consultation with the Secretary of the Treasury."(B) The Director of the Office of Management and Budget shall determine what information is required to be reported to comply with subparagraph (A). At a minimum; for each financing account and for each liquidating account (as those terms are defined in sections 502(7) and 502(8), respectively, of the Federal Credit Reform Act of 1990) the following information shall be reported:

"(i) The cumulative balance of current debts outstanding, the estimated net present value of such debts, the annual administrative expenses of those debts (including, the portion of salaries and expenses that are directly related thereto), and the estimated net proceeds that would be received by the Government if such debts were sold.


"(ii) The cumulative balance of delinquent debts, debts outstanding, the estimated net present value of such debts, the annual administrative expenses of those debts (including the portion of salaries and expenses that are directly-related thereto), and the estimated net proceeds that would be received by the Government if such debts were sold.


"(iii) The cumulative balance of guaranteed loans outstanding, the estimated net present value of such guarantees, the annual administrative expenses of such guarantees (including the portion of salaries and expenses that are directly related to such guaranteed loans), and the estimated net proceeds that would be received by the Government if such loan guarantees were sold.


"(iv) The cumulative balance of defaulted loans that were previously guaranteed and have resulted in loans receivables, the estimated net present value of such loan assets, the annual administrative expenses of such loan assets (including the portion of salaries and expenses that are directly related to such loan assets), and the estimated net proceeds that would be received by the Government if such loan assets were sold.


Attachment B

December 26, 1995



SUBJECT: New credit data requirement for credit programs

1. Purpose. This Bulletin informs Federal agencies of new credit data reporting requirements included in the report language of the Treasury-Postal Appropriations Act of 1996, P.L. 104-52. For both performing and non-performing loans, the report language requires the Administration to provide two estimates of the value of the Government's outstanding loan portfolio: (1) expected cash flows to the Government (along with a separate estimate of the administrative costs); and (2) expected cash flows to or from a private holder (if the loans were sold) for both the financing and liquidating accounts. Data collected in accordance with this report language will be used by Congress and OMB to determine whether it would be worthwhile to pursue loan asset sales. These data will be included in the FY 1997 Budget.

2. Background. Report language accompanying the Treasury-Postal Appropriations Act of 1996 instructs OMB to "direct, and coordinate with, the Federal agencies involved in credit programs to evaluate the value of their credit programs, including the cost of annual administrative expenses and develop a plan for the privatization [sic] of such credit programs." For the purposes of this analysis, "outstanding portfolio" is defined as direct loans and loan guarantees outstanding as of September 30, 1995.

The Treasury/Postal conferees believe that the private sector is willing to pay substantially more than the Government's expected value of its portfolio. The report language states that "[u]sing conservative estimates, it may be that between $20,000,000,000 to $50,000,000,000 could be realized if much of the Federal credit program was to be turned over to the private sector."

To allow for the comparison of the value of different types of loans, agencies are asked to divide the outstanding portfolios in the liquidating and financing accounts as of September 30, 1995, into two categories: Category A (substantially performing) and Category B (non-performing). Substantially performing loans are defined as loans that are current or delinquent less than 90 days. Non-performing loans are those loans delinquent for 90 days or more. For international programs, Category B loans include loans to countries that currently have a rescheduling or a debt reduction program with the Paris Club.

3. Action required. Provide the information outlined below for each loan program. To assist in the data collection, a Lotus spreadsheet ("port _yal.wk3") will be sent with this Bulletin. Data should be delivered to the OMB examiner with primary responsibility using this spreadsheet by February 9, 1996.

A. Loan categorization. Agencies are asked to categorize loans outstanding as of September 30, 1995, into either Category A (substantially performing) or Category B (non-performing). For Loan Guarantee programs, please specify the cumulative balance of defaulted loans that were previously guaranteed and have resulted in loans receivable. Agencies are requested to provide the weighted average number of years remaining to maturity for Categories A and B.1


B. Scheduled direct loan repayments. For direct loan programs, agencies are asked to report the scheduled repayments (the amounts that would be received if there were no defaults or recoveries) for each direct loan program. If scheduled repayments are not known, agencies are asked to report best estimates of scheduled repayments.

C. Expected cash flows to Government. Agencies are asked to report the loan program's expected net cash flows to (+) or from (-) the Government, excluding administrative costs. For direct loan programs, expected cash flows (primarily principal and interest payments, net of defaults and recoveries) should be positive (+). For loan guarantee programs, as a result of claim payments, expected cash flows will typically be negative (-). Note: Please separate the defaulted loan guarantees that resulted in a loan receivable from the rest of the loan guarantee cash flows.


D. Factors affecting cash flow assumptions to the Government. Agencies are asked to provide the assumptions (default rate, recovery rate, weighted-average remaining maturity, weighted-average interest rate, fees, etc.) used in estimating the expected cash flows to the Government.


E. Expected cash flows to private sector. Agencies are asked to report for each direct loan and loan guarantee program the expected net cash flows for Category A and Category B if these loans and guarantees were sold to the private sector with no Federal guarantee (recourse) on the sale transaction.


F. Factors affecting private sector cash flow assumptions. Agencies are asked to list the assumptions and conditions which could cause cash flows to a private purchaser to differ from the expected cash flows to the Government.Some examples of generic conditions may be:


  • The private sector would lose the IRS tax offset, if currently used by the agency.
  • Differences in efficiency and quality of servicing and liquidating practices.
  • The private sector can be expected to pursue non-current loans more aggressively because of the profit motive.

Some examples of conditions specific to individual loans programs may be:

  • Legal or regulatory forbearance procedures. Agency regulations may allow purchasers to take more aggressive collection steps than the Government. For example, HUD must hold assigned mortgages for at least three years and practice detailed forbearance procedures.
  • Different recovery rates on collateral pledged on loans.
  • Type and value of collateral pledged.
  • Connections to other benefits (grants).
  • Other provisions that could change the value of proceeds.

G. Administrative costs. Agencies are asked to estimate the annual expenses directly associated with administering the loans outstanding on September 30, 1995. Administrative costs should be defined consistent with OMB Circular No. A-11, Section 33.5(n).


Agencies may use Method I: their own methodology for projecting administrative expenses or Method II: the methodology contained in the Lotus spreadsheet ("port_val.wk3") which accompanies this guidance. Method II relies on per loan costs in FY 1995 for projecting future administrative costs. If using Method II, agencies must:

(1) estimate the number of loans outstanding each year until maturity;


(2) project the net growth in administrative expenses (rising wages, productivity gains, and other factors should be taken into consideration); and

(3) project the percent of substantially performing loans as of September 30, 1995, that will become non-performing each year.

Agencies may modify Method II to more accurately reflect the expected annual administrative costs of their outstanding portfolios on September 30, 1995.

4. Comparison of cash flow projections. Using the data provided by agencies, OMB will compare: (1) the net present value of the Government's expected cash flows and administrative expenses, to (2) the expected net proceeds from selling the loans to the private sector, which is the net present value of the projected cash flows if the loans were sold to the private sector. Consistent with the estimation of loan modification cost (as defined in OMB Circular No. A-34, Section 65.3), Governmental cash flows will be discounted at the "applicable interest rate (yield) in the quarter when the modification occurs." For the purposes of this analysis, present value calculations will assume the Government's loan portfolio was sold to the private sector on September 30, 1995. Calculating the values as of the end of FY 1995 will minimize the assumptions underlying this analysis, such as the size of the portfolio outstanding and appropriate discount rates.

5. Due Date. Estimates should be delivered to the OMB examiner with primary responsibility for the credit account using this spreadsheet by February 9, 1996.

Alice M. Rivilin Director

Attachment C



In compliance with the report language accompanying the Treasury-Postal Appropriations Act of 1996 (P.L. 104-52), this chapter reviews the Government-wide potential for loan asset sales. It summarizes the Government's loan asset sales in the 1970s and 1980s, describes the current budgetary scoring of loan asset sales, and enumerates the agency loan asset valuation data requested by OMB for this analysis. The following section reviews the value of Government loan assets and identifies factors to be used in reviewing potential sale opportunities. The final section describes ongoing loan asset sales, and assesses loan programs' potential for profitable sale to the private sector.

This evaluation of the Government's loan portfolios has identified a number of programs that could potentially be sold to private investors at no cost or a profit to the Treasury. Loan asset sales are not recommended on a large scale, as in prior years, in part because costs to the Government often exceed returns. This is now reflected in budgetary scoring rules enacted as part of the Federal Credit Reform-Act of 1990 (FCRA).

Previous Loan Asset Sales

1970s to mid-1980s. Federal direct loan assets were sold to the public producing over $40 billion in proceeds in the 1970s and early 1980s. In these sales, a guarantee by the selling agency (recourse) was often attached. After the sale, the loan was held privately but the risk of default in the recourse sales remained with the Government. In some cases, the Federal agency sold securities [called participation certificates (PC's) or certificates of beneficial ownership (CBO's)] that were backed by loans that the agency continued to hold and service.

1987-1990. During the late 1980s, a series of pilot loan asset sales were undertaken, resulting in the sale of nearly $25 billion of Federal loan assets. In the President's 1987 budget, the pilot program was first proposed, with the following four objectives: (1) improve credit management; (2) obtain administrative savings; (3) identify subsidies; and (4) reduce the short-term deficit.

Loan asset sales were seen as a tool for improving public sector credit management. Asset sales provided an incentive for agencies to improve loan origination and documentation, because loans could be sold at a higher price if screening and documentation met private sector standards. Sales also provided information on the condition of loan portfolios, revealing areas of improvement for servicing of agencies' remaining portfolios. Second, loan asset sales allowed the Government to reduce its administrative costs by transferring servicing and collection to the private sector. Third, nonrecourse sales of new loans provided information regarding the subsidies of Federal credit programs. The difference between face value and selling price (net of transaction costs) was the Government's subsidy cost. Finally, loan asset sales generated proceeds to increase budgetary offsetting collections in the year of the sale.

The sale of Education, HUD, USDA, and VA loan assets produced gross proceeds of $5.6 billion in 1987, $8.2 billion in 1988, and a total of $10.5 billion in 1989 and 1990. However, the cost to the Government of these sales was substantial, with sale proceeds averaging less than 90 percent of the present value of the Government's cash flows. Improved management of the unsold portfolios was expected to partially offset losses resulting from the loan asset sales.

Budgetary Scoring of Loan Sales

Prior to 1987, loan asset sales were treated as offsets to agency outlays (equivalent to loan repayments) for purposes of budget scoring. Thus, sale proceeds were permitted to offset increased spending. Despite the fact that asset sales might result in a present value loss to the Government (due to the loss of future cash flows), sale proceeds were allowed to offset spending in the year in which the sale occurred.

Amendments to Gramm-Rudman-Hollings (GRH) in 1987 allowed only those proceeds from routine and ongoing sales established prior to 1987 to offset spending, or to offset the deficit for the purposes of sequester calculations. Thus, newly proposed loan sales could not be considered as an offset to spending for budgetary scoring purposes because such sales were not viewed as reducing the structural deficit.

Under the Federal Credit Reform Act, loan asset sales are treated as modifications that change the cost of the loan or guarantee, and are not undertaken unless budget authority has been provided for any positive subsidy cost of the sale. The 1996 Budget Resolution (Sec. 206 of H.Con. Res. 67) confirmed this budgetary treatment of loan sales under credit reform.

As modifications, the credit reform subsidy cost of a loan asset sale is the difference between the Government's currently estimated net present value (NPV) of the remaining cash flows under the terms of the existing loan contract (the "expected" value), and the net proceeds from the loan sale. The result of this calculation can be positive, negative, or zero. If the estimate is positive, i.e., the expected value to the Government is greater than the loan sale proceeds, budget authority must be provided to cover the additional subsidy cost resulting from the sale. A negative estimate would indicate that the Government is achieving a savings from the sale, and a receipt in the amount of the "negative subsidy" is generated from the sale. An estimate of zero would indicate that the modification will not change the cost to the Government, and budgetary resources would not change.

If the loan assets sold prior to 1990 had been scored under credit reform, not only would the sale proceeds not have been available to offset spending in the year of the sale, but an appropriation would have been required to cover the loan modification cost for those loans sold below the Government's expected value. By scoring loan sales as modifications, agency actions are subject to greater scrutiny by Congress and OMB. This scrutiny prevents costly sales, and encourages and gives credit to agencies for sales that save Federal resources.

The FCRA definition of subsidy cost specifically excludes administrative costs and any incidental effects on government receipts or outlays. For loan sales, this means that effects on Federal administrative costs and incidental changes to interest on the public debt are excluded from the subsidy cost calculation. For some agencies, loan sales would produce savings from reduced administrative (personnel) costs for loan servicing, management, and delinquent debt collection. Although not scorable for budgetary purposes, these savings should be considered when evaluating the total effect of a loan sale; they would lower the agency's future administrative expense requests. For other agencies, selling loan assets would relieve staff of the administrative burden of loan servicing, allowing them to be redirected to other programs. Although this would not produce savings from a reduction of personnel, it could serve to enhance the mission of the agency.

Potential for Further Loan Asset Sales

The recent success of HUD loan asset sales has sparked renewed interest in Government-wide sales. In the Treasury-Postal report language, the conferees directed OMB "... to direct, and coordinate with, the Federal agencies involved in credit programs to evaluate the value of their credit programs ... and develop a plan for the privatization of such credit programs."

In response to this request, for all direct loans, loan guarantees, and defaulted loans that were previously guaranteed and have resulted in loans receivable, OMB requested that agencies provide (1) the face value of loans outstanding as of September 30, 1995, (2) the currently expected cash flows to the Government, and (3) estimated cash flows to the private sector (if those loan assets were sold). To calculate whether the sale of these loans would result in an estimated net saving to the Treasury, for each loan portfolio when possible, the net present value of the Government's remaining cash flows were compared to the NPV of the expected cash flows to the private sector, adjusted for the private sector's administrative costs and the transaction costs of a loan asset sale. The private sector's servicing costs must be included as a factor in estimating the market value that they would be willing to pay for a given stream of cash flows, because these costs are not included in the Government's cash flows. The transactional costs take account of the difference between the gross and net proceeds to the Government. Expected cash flows to the Government were discounted at the rate on Treasury securities of comparable maturity to the remaining portfolio maturity, as required by Sec. 502(5) of the Federal Credit Reform Act. As discussed below, expected cash flows to the private sector were discounted by the appropriate private sector rate. Table 8-1 contains the Government's expected cost of new loans and its outstanding portfolio.

To allow for the comparison of the value of different types of loan assets, OMB asked agencies to divide outstanding portfolios into two categories: substantially performing and non-performing. Substantially performing loans were current or delinquent less than 90 days as of September 30, 1995. Non-performing loans were those delinquent for 90 days or more. In addition, agencies were asked to provide information on the Government's administrative costs.

This analysis is subject to severe limitations because of its broad scope, the short time period for collecting and analyzing data, financial systems constraints in obtaining data, and the difficulty of estimating private sector valuation of loan assets. As a result, this report serves only to identify loan programs which show potential for loan asset sales or which clearly cannot be sold to the private sector without substantial financial and/or public policy costs to the Government. While the analysis narrows the field of potential candidates for loan asset sales, further analysis is planned to identify the potential benefit of asset sales in the remaining programs.

Private Sector Valuation of Federal Loan Assets

The value of Federal loan assets to the Government and private sector may differ substantially. Some costs, such as subsidized interest, are valued similarly by the Government and private investors. Other non-contractual expected costs may be valued very differently. For example, given their relative efficiencies and collections tools, the Government and private investors may have dramatically different estimates of future defaults and recoveries. Before deciding to bid on Government loan assets, the private investor must adjust the expected value to the Government for the investor's higher discount rate, expected efficiency gains, and the cost of servicing loan assets.

Discount Rate

When the Government provides a loan, it contracts to receive a stream of payments of principal, interest, and often fees. It also expects to experience defaults. The "value" of this loan to a potential private sector investor is the present value of the expected net cash flow to the private sector, where the discount factor the private sector discount rate. This rate will be higher than the Government's discount rate for at least four reasons: a higher cost of funds, private sector risk aversion, desire for liquidity, and information requirements.

  • Cost of funds: The private sector's cost of funds, the base discount rate, is higher than the Government's discount rate, which is the rate on Treasury securities of comparable maturity to the outstanding loan. Therefore, even before adjusting for risk, liquidity, and information concerns, the private sector's discount rate for valuing loan assets will always be higher than the Government's.
  • Risk: Investors associate a range of risk with the Government's default estimates for loan assets. Investors are usually considered risk averse; that is, they would prefer to receive $90 with certainty than to receive $100 with a default probability distribution for which the mean is 10%. Because the amount received could be less than $90 in the latter case, private investors will require a premium to compensate for bearing this risk. The Government has no counterpart for this cost; it incurs outlays only if defaults are actually higher.
  • Liquidity: Because the market for Federal loan assets is small, it is not very liquid. Unlike U.S. Treasury securities, for example, investors cannot count on being able to sell the loan assets they purchase quickly and at an attractive price. They require a premium to compensate for this illiquidity.
  • Information: Finally, investors are not fully informed about Federal loan assets. Because the Federal programs have different rules, documentation, and collateral, investors must devote considerable time to acquiring information about potential investments. These are not standard commercial loans, and the Government's documentation methods differ significantly from private entities'. Investors look for systematic and predictable cash flows as shown by accurate historical records of the portfolio's delinquency, default, and recovery experiences, which Federal agencies are often not able to provide. Investors require a premium to compensate for these information costs.

These costs can never be eliminated, but small reductions in premiums could be expected if investors became more familiar with Federal loan assets, and if the liquidity of their market improved.

Efficiency Gains

Although the private sector will use a higher discount rate than the Government, the net cash flow to the private sector may be larger. It is generally felt that the private sector is more efficient at loan servicing and collection. For example, the loss rate on private commercial loans is much lower than the loss rate on Government business loans.

Such comparisons, however, are difficult at best. In many ways, Government loans are fundamentally different from private sector loans, and comparing the two without taking into account their differences is misleading. For example, the Government often functions as a lender of last resort and makes loans to less creditworthy borrowers than would a private lending institution. In addition, the terms of Government loans, such as lower down payments or less restrictive underwriting criteria than required in the private sector, may result in higher defaults. Finally, other characteristics of Government lenders, such as willingness to practice substantial forbearance, may increase both the administrative and default costs.

A portion of the difference in default costs is certainly due to these differences between public and private sector loans. A portion of the difference may also be due to the Government's relative inefficiency. Private firms' profit motive creates the incentive for the private sector to be more efficient than the public sector in performing common tasks. However, for some programs, the Government's superior collection tools, such as the IRS Tax Offset, may counterbalance private sector productivity gains.

Decision to Purchase Government Loan Assets

Private investors, bidding on Government loan assets, will take into account the Government's expected future losses and make assessments as to how much they can reasonably expect to lower these losses. The amount possible will depend upon how much more efficient the investors expect to be at performing the required tasks.

Whether the value of the loan asset to the private investor exceeds the value to the Government depends on: (1) the difference in discount rates; (2) the private sector's expected efficiency gains; (3) the private sector's administrative costs; and (4) the loan asset sale transaction costs. Because the private sector discount factor is higher than the Government, the value of the loan asset to a private sector investor will be lower than the value to the Government for a given set of cash flows. On the other hand, because the investor expects to reduce default costs below those of the Government, the investor's valuation of the loan asset may be greater, because expected cash flows will be higher.

Finally, for the Government to receive sale proceeds greater than its expected value, these proceeds must be greater than the Government's expected value after sale transaction costs are paid. Therefore, the private investor's productivity gains must not only cover its higher discount rate and servicing costs, but must also allow for payment of the sale transaction costs. These can be substantial, and include the cost of a financial advisor and the costs of issuance.

Identifying Loan Assets for Possible Sale

Decision Analysis

The Treasury/Postal conferees directed OMB to identify those loan assets which can be sold at a gain to the Government. Based on past sales experience and the review of the Government's current portfolio, OMB believes that a percentage may be sold at a net profit to the Government. The preliminary data recently received from the agencies will allow OMB to begin the process of identifying which sales would benefit the Government. The following framework allows for the identification of those loan programs for which sales may result in a net profit, or other type of gain, and develops a method for identifying those programs which may merit further analysis. This framework also outlines factors which would discourage private investors from offering at least the Government's expected value for certain loan assets. If loan assets are only to be sold at no cost or a "profit" to the Government, loan portfolios with these characteristics should not be considered further for loan asset sales.

Factors which May Lead to the Decision to Sell Federal Loan Assets

Efficiency in Administration of Loan Portfolio. If an agency has a dramatically expanding loan volume or a surge in defaults, it has several options for avoiding a reduction in its servicing and liquidation capabilities, including: continuing to hold the loans and requesting additional staff for administration; attempting to administer the loans with current staff levels; contracting to the private sector for servicing; requiring guaranteed lenders to work out the loans they originate; and selling the loans. With declining staff levels in many agencies, a loan sale in these cases may produce benefits for the Government apart from any immediate "profit" from the sale. That is, even if these loans are sold at the Government's expected value, the selling agency might guard against deterioration in its loan servicing and liquidation capabilities.

In programs where defaults surge briefly or are expected to grow in the future, pilot loan sales may help the Government make decisions concerning whether to add resources for program administration or sell the loans. For example, recent dramatic growth in certain Federal guarantee programs is expected to lead to large defaults in the near future. Unless steps are taken soon, current administrative resources may be insufficient to continue to adequately service these loan portfolios. Testing the market, by selling loan assets in a pilot program before the expected increase in defaults occur, could help the Government decide whether it is more efficient to add more debt collection staff and upgrade current financial systems, to contract out loan servicing, or to sell defaulted loans.

Best Use of Federal Staff. Selling loans could prove useful for expired programs, where no new loans are being issued, but staff time is consumed with administration of the dwindling portfolio. Loan sales may be warranted if current loan staff could be redirected to focus attention on new or different lines of business that are high priorities for the agency.

Private Sector Practices. The private sector can be expected to pursue non-current loans more aggressively because of the profit motive. This is part of the reason for the "productivity gap" referenced in the conference report requiring this study. Where Federal loans are inefficiently managed, or funds for management of these programs are declining, selling loans may capitalize on the quality of servicing and liquidating practices in the private sector, and provide a net benefit for the Government. The recent HUD sales have shown that the private sector is willing to pay more for loans where it believes that it can achieve these efficiencies in servicing.

Collateral Value. Loan sales may be warranted if collateral underlies the loan and the private sector is better at maintaining collateral value while in inventory, can dispose of it more quickly, or expects a higher collateral sale price. Collateral value was an important factor in the success of the recent HUD loan sales.

Public-private Partnerships. In this era of reinvention, loan sales foster new communication between the managers of the federally assisted credit portfolio and the private credit market. This partnership can serve to create new products or efficiencies that can be applied across all credit programs. For HUD, this has meant the creation of a user-friendly, low-cost due diligence process that, combined with the use of computer technology, has attracted a large following of investors with more than 200 bidders representing a wide spectrum of the financial markets to their loan sales. This new technology can be evaluated and the Government can assess whether it could be useful to other Federal credit programs, or for future loan sales.

Factors which May Lead to the Decision to Hold Federal Loan Assets

Small Margin for Improvement in Default Rates. When the private sector has little margin for improving on the Government's expected net defaults, it would not be profitable to sell the loan asset. This includes both programs with a low life-time default rate and seasoned loans with few remaining expected defaults. Since investors need substantial efficiency gains to overcome their higher discount rate, servicing costs, and the transaction costs of the loan sale, unless the Government expects substantial future losses, opportunities do not exist for the private sector to obtain sufficient productivity gains. A number of Federal programs have historical default rates of less than 5 percent. For example, the USDA's rural water and waste loans have expected default rates of less than one percent, leaving little room for the private sector to improve in this expected performance.

Collection Tools. The Federal Government has a variety of collection tools that are not available to the private sector. These tools include the IRS -Tax Offset, Federal Salary Offset, and the ability to withhold future benefit payments (grants) and credit. Tax Refund Offset and wage garnishment are extremely important to Collections in the Federal student loan programs. A 1987 Chemical Bank analysis of the marketability of student loans held by the Department of Education concluded that the portfolio of loans was more valuable to the Federal Government than to the private sector because of the collection tools that are only available to the Government, and because the private market would require a deep discount due to the credit risk of these loans. For defaulted student loans made under the health education assistance loan program, the Federal Government can withhold Medicare payments until a doctor's (borrower's) loan is in good standing.

For international loan programs where loans or guarantees are made to another sovereign government, the Government has a number of tools not available to the private sector. These tools include international treaties and agreements that the U.S. has signed with other nations and the ability of the U.S. to block credits from international financial institutions to debtor nations that have not honored their debt obligations to the U.S. For example the Brooke Amendment and Section 620(q) of the Foreign Assistance Act make countries ineligible for certain types of foreign assistance unless they make required payments on their related debts to the U.S.

Policy Goals of Domestic Programs. The Federal Government often uses loans as tools to implement domestic policy objectives. In these cases, the political goals of the program override the importance of individual loan performance. Because of these policy goals, Federal credit agencies often originate and service loans differently than a private financial institution. Credit review takes secondary importance to policy considerations, such as guaranteeing that credit is available for all farmers or all students.

Likewise, servicing actions in these programs and decisions on restructuring loans may be aimed at providing additional assistance to the borrower rather than at collecting funds promptly. For example, many farm loan contracts contain significant borrower rights that make servicing labor-intensive. These procedures to protect the borrower would cause private sector purchasers to discount the value of these loan assets heavily. It would be difficult for the Government to continue to achieve its public policy objectives unless its generous collection and forbearance tools were continued by the private sector. However, if the private sector continued these servicing and liquidation policies, it would be difficult to realize the productivity gains necessary for a private investor to purchase the loan assets at or above the Government's expected value.

Subsidies are not necessarily limited to beneficial loan terms and collection procedures. The Government may support the borrower by providing extensive training, counseling, and technical support. For example, SBA's Microlending Program avoids substantial defaults by providing the micro-loan borrowers with the knowledge and skills they need to successfully repay their loans. If these loans were sold, it is not clear how the borrower-Government-investor relationship would continue.

Foreign Policy Goals and Considerations. Foreign policy goals and considerations affect the provision and administration of Federal credit in that they often offer terms and conditions that are more generous than the private sector.

In some cases, foreign policy goals could also inhibit purchases by the private sector. For example, many loans are made to developing countries that are perceived too risky for the extension of private credit. This perception of risk is one of the justifications for the development of some of the Government's international credit programs. In other cases, the private sector might avoid credits that otherwise would be attractive because of foreign policy sensitivities. For example, the Department of Defense's new commercially-oriented military export credit program was created by Congress partly because the private sector has been reluctant to provide credit for military export purchases.

The majority of outstanding international direct loans and loan guarantees are sovereign; that is, they are direct loans or loan guarantees that are to, or guaranteed by, another sovereign government. It is unlikely that a credit to a sovereign government would have a greater value to the private sector than to the U.S. Government. Private sector creditors remember the international debt crisis of the 1980s, where a number of U.S. financial institutions lost large amounts of funds that they had extended through loans to sovereign Governments in developing countries. As a result, private creditors have since often shied away from providing sovereign credit in most developing countries. In addition, many of the sovereign direct loans that are outstanding to the U.S. Government have been rescheduled in the Paris Club, an informal group of creditor nations which agrees to extend the maturity of loans that a debtor nation could not otherwise pay on schedule. While Paris Club reschedulings are done in order to increase the eventual probability of repayment, because previously rescheduled loans are eligible for further rescheduling, the private sector often views these loans as "subordinate" to those that have not been rescheduled.

Length of Loan Term. Some Federal programs have loan terms of 40-50 years. In these cases, the private market would view the extended term as increasing the uncertainty of repayment, as well as exposing them to additional interest rate risk. These factors would cause the purchase price to be discounted heavily.

Lack of Private Interest. Several factors may lead to the lack of interest in Federal loan assets. These include: insufficient documentation; lack of collateral pledged for the loan; inadequate information regarding historical loan performance; and the lack of uniform loan characteristics. In many programs, including rural development ana farm loans, portions of existing loan origination and servicing are not automated; and, therefore, the pre-sale due diligence costs would be high. This lack of loan information and documentation would also inhibit a rating agency from arriving at an accurate rating for a pool of loans.

Legal Restrictions or Other Complications. Many international loans and guarantees have legal or contractual restrictions that would make it difficult or impossible to sell them to the private sector. For international loan guarantees to private borrowers, for example, the Government pledges its "full faith and credit" in the contract with the borrower and/or the lender. In these cases, the agency involved would first have to "buy out" the holder of the U.S. Government guarantee, by making a payment in lieu of responsibility for future claims, in order to relieve the Government of the responsibility of the original guarantee. This would be the case for the guarantee programs of OPIC, where a11 loans are made to the private sector, as well as with the non-sovereign portions of guarantees issued by Eximbank and AID.

Programs Currently Selling Loan Assets

Two programs currently sell loan assets, the Federal Housing Administration and the Veterans Administration. As with the other programs, these programs will be evaluated using the criteria above as part of the ongoing analysis of loan sales.

Federal Housing Administration. FHA has been insuring mortgages since 1934. Historically, the program default stream has been relatively predictable. However, during the 1980s weak real estate markets prompted an unexpected surge of defaults in both the single family and multifamily portfolios. By 1994, when the wave of defaults subsided, FHA owned nearly 2,400 defaulted mortgages, with unpaid principal balances of more than $7 billion, and 90,000 single family mortgages with unpaid principal balances of almost $4 billion. This volume was so large that it was compromising FHA's capacity to perform its other functions (including oversight and management of the of the insured mortgages in force), thereby making FHA more vulnerable to future losses.

In response to this large administrative burden, Congress and the Administration approved a program to sell these HUD-held mortgages in a series of competitive auctions and negotiated sales with state and local housing finance agencies. These sales have been highly successful, helping HUD reduce its inventory of mortgages, while capitalizing on the private sector's knowledge and ability to manage defaulted loans. In an improvement on earlier loan asset sales, the return to the Government was increased through use of competitive bidding and computer technology which evaluates and optimizes competing bids.

Department of Veterans Affairs. The DVA "vendee" loan program provides direct loans to veteran or non-veteran purchasers of DVA-owned real property to facilitate property sales. Nearly all of these direct loans are pooled and sold to the public with a Government guarantee (recourse). In 1995 total proceeds from DVA loan asset sales were $1,333 million.

Programs which Merit Further Analysis of Loan Asset Sale Potential

Small Business Administration. As a result of significant portfolio growth since 1990, with no change in expected default rates, the SBA can expect an increase in the number of defaulted loan guarantees which will result in loan receivables. Further analysis is needed to determine whether loan asset sales might be an effective tool for alleviating the expected pressure on SBA's servicing and liquidation offices, and for increasing recoveries. It is not clear how the private sector would value these small business loans in comparison to the Government's expected value. Unlike housing loans, which have uniform terms and collateral, SBA guaranteed loans have a wide range of purposes and varying collateral requirements.

SBA disaster loans should also be more closely reviewed to determine whether the sale of these loans assets could result in a net gain to the Government. Since these are fixed-rate subsidized loans, the value of these loans assets will be sensitive to the prevailing interest rates.

Eximbank. Project finance and short-term insurance and working capital guarantee claims are areas which merit further review. The project finance program offers financing for the U.S. export component of major overseas projects. While the typical pre-completion political risk guarantee is not likely to be attractive to the private sector, the subsequent loan or guarantee might be, particularly since the riskiest part of the project, the construction phase, will have been completed. The resulting savings in administrative expenses could be considerable, since project finance tends to be a labor intensive activity. However, these decisions would have to be made on a case-by-case basis, since each project has a different structure and therefore a different risk determination.

Eximbank extends between $4 and $5 billion each year in short-term trade finance insurance, and in working capital guarantees for small business exporters. Eximbank attempts to limit its extension of insurance and guarantees to cases where the private sector will not provide coverage. However, once claims are paid, the recovery of these claims is a task where Eximbank, as a Government agency, may not have significant advantages over the private sector. In addition, claims recovery can be relatively labor intensive, and the sale of certain claims by Eximbank, even at a loss compared to the expected value to the Government, in certain cases might be outweighed by this gain in administrative savings.

Defaulted Guaranteed Loans and Non-Performing Loans. In addition to reviewing the SBA and Eximbank programs specified above and analyzing more closely the ongoing HUD and VA sales, the potential for selling defaulted guaranteed loans and non-performing loan assets across agencies will be examined more closely. These loans offer a special case where efficiency gains could be large. If credit programs have low recovery expectations on such loans, then the private sector may be willing to offer a price higher than the Government's expected value.

Programs where Loan Asset Sales Are Not Suitable

Departments of Agriculture and Interior. Many of these loan programs have significant borrower rights that would transfer with the loan upon sale and would cause the proceeds to be significantly discounted. They also often have loan terms of over 35 years. This increase in uncertainty would be reflected in the sale price of the loans. Finally, since most of the existing servicing is not automated, due diligence costs would high.

Departments of Education and Health and Human Services (HHS). The Department of Education sold most of its low-interest college housing and higher education facilities loans at a discount in the late 1980s. A small number of the facilities and housing loans it still holds are in default, and many of these schools have negotiated payment arrangements with the Department or are in formal foreclosure proceedings, which would preclude sale of their loans. Those housing and facilities loans that are in good standing have extended maturities and low expected default rates, which makes it unlikely that the private sector would be willing to offer a price higher than the Government's expected value. Direct student loans and defaulted guaranteed loans at Education and HHS are uncollateralized loans, which the market would discount highly because of the uncertainty of collection. The Departments also make considerable use of Federal Government collection tools which are not available to the private sector. For example, HHS' health evaluation assistance loan program can bar defaulters from participating in the Medicare program.

Department of Veterans Affairs. To collect on defaults in the DVA housing loan programs, some loans are paid through allotment of DVA benefit checks, which may not be paid to private loan holders. In addition, DVA offers beneficial financing not available in the private sector to help it dispose of Federal property, and DVA has invested in loss mitigation staff dedicated to pursuing loan reinstatements and alternatives to foreclosure.

Department of Defense. Most outstanding Foreign Military Financing loans are to countries with which official U.S. diplomatic relations are particularly important and sensitive. This argues against selling these loans, particularly in cases such as Israel, where military sales are an important component of the relationship. In addition, because of the sensitive military nature of FMF loans, the private sector is likely to be much less willing to acquire such loans.

USDA Foreign Agricultural Service. The primary aim of the Public Law 480 program is to provide humanitarian food assistance through low-interest, long term (30+ year) loans. Borrowers typically have low credit ratings. Given the high default risk, lack of collateral, and deep interest subsidies, it is unlikely that private investors would be interested in purchasing P.L. 480 loan assets.

U.S. Agency for International Development (USAID). USAID's Economic Assistance Loans (EAL) and Housing Guarantees (HG) are unlikely to be of greater value to the private sector. EALs are: (1) sovereign and primarily in very low-income developing countries; (2) highly concessional; and (3) long term (30-40 year maturities). Because the private sector does not have the leverage of the Brooke Amendment sanctions, diplomatic pressure, or significant foreign aid funds, nor the ability to use its experience in operating in developed financial markets in most EAL countries, it is unlikely to have efficiency gains in collecting on these loans. HG guarantees are primarily sovereign loans and are similar to EALs, except that they have market interest rates. Therefore, for the reasons discussed above, they are unlikely to be attractive to the private sector.

Overseas Private Investment Corporation (OPIC). Partly through its agreements with the Governments of the countries in which it does business, OPIC has historically had low credit losses. In addition, each guarantee contract carries the full faith and credit of the U.S. Government. Given the relatively low expected cost to the Government of these guarantees, and the relatively high price that the beneficiaries are likely to ask in order to give up the security of the guarantee, it is unlikely that such buy outs could be undertaken at a savings to the Government. Because the full faith and credit contractual issue discussed above does not apply to direct loans, it might be more feasible to sell outstanding loans to the private sector. However, direct 16ans tend to be smaller, more risky transactions, and therefore are not likely to be attractive to the private sector.

Eximbank. The majority of Eximbank's outstanding portfolio (about 75 percent) consists of sovereign credits, which are very unlikely to be attractive to the private actor because of memories in the private sector of the international debt crisis and the possibility of rescheduling of sovereign credits. Similarly, while Eximbank's "traditional" non-sovereign credits are not subject to Paris Club rescheduling, they are still concentrated 'm riskier developing countries, and are therefore subject to the same perception of risk in the private sector.

Next Steps

In the coming months, OMB will work with credit agencies and Treasury to form a task force to evaluate loan portfolio management. The goal of the task force will be to review options for improving the quality and efficiency of current practices. Options will include: achieving efficency gains through upgrading financial systems; increasing the staff of servicing and liquidation offices; contracting to the private sector; requiring guaranteed lenders to work out the loans they originate; and loan asset sales. In considering opportunities for loan asset sales, the task force will refine the framework outlined in this chapter. Similar to the credit performance measures framework discussed in Chapter 8, this framework will be used to develop decision criteria which can be applied to all loan portfolios in order to identify programs that show potential for loan asset sales.

1 Consistent with guidance on modifications in OMB Circular No. A-34, Section 65.3, the average years remaining to maturity will determine the appropriate discount rate for net present value calculations of Governmental cash flows.