This page shows textual changes in the document between the two versions indicated in the dates above. Textual matter removed in the later version is indicated with red strikethrough and textual matter added in the later version is indicated with blue.
Federal credit programs are comprised of government direct loans and loan guarantees, which are available to individuals and firms. These
Contents
Figures
Tables
Summary
The U.S. government uses direct loans and loan guarantees in a range of policy areas. More than 100 programs offering direct federal loans and private financial institution loans guaranteed by the government, known as federal credit programs, are available to individuals and firms. The credit programs support a wide range of economic activities, including home ownership, education, small business, farming, energy, infrastructure investment, and exports. At the end of fiscal year (FY) 2014FY2024, outstanding federal credit totaled $3.34.0 trillion, with direct loans at $1.08 trillion and principal amount of loan guarantees at $2.3 trillion.1
For budget formulation, the costs or profits of these government programs are estimated as prescribed by the Federal Credit Reform Act of 1990 (FCRA, P.L. 101-508). As measured by FCRA, some of these credit programs generate a profit, while others incur costs to the government. The costs of these credit programs are commonly referred to as subsidy costs. When these programs generate a profit, they are considered negative subsidy costs. Congress has periodically debated the best way to measure subsidy costs. The debate has revolved around whether the subsidy costs should be measured as prescribed by FCRA or by what is referred to as the fair value method. Subsidy costs estimates under FCRA adjust the cash outflows and inflows for the various risks a loan portfolio might face. These cash flows are also discounted using Treasury interest rates for estimating subsidy costs. One method of estimating the fair value costs of the credit programs is to use private market interest rates. Generally, private sector firms charge borrowers with government loan guarantees lower interest rates than they charge borrowers without government guarantees. Switching to fair value would be expected to increase the subsidy cost estimates of credit programs recorded in the federal budget and potentially raise the cost of borrowing. For example, the Congressional Budget Office (CBO) projects that changing the method of calculating subsidy costs estimates to the fair value method would increase the 2025 budget cost estimates for all programs by $62.7 billion. Proponents of fair value cost estimates argue that the government's cost of credit programs should reflect market risks, which are currently excluded from FCRA cost estimates. In their view, the risk posed by the borrowers should be considered as a cost to the taxpayers, because taxpayers are ultimately responsible for paying the debt of the U.S. government. Supporters of using the FCRA method argue that it is appropriate for the government to discount at the rate at which it borrows and that market risk is not the same as budgetary costs. In their view, including market risks to estimate credit subsidies includes amounts that the government will never incur. Further, adopting fair value for budget estimates does not necessarily imply that there would be a need to raise taxes or to borrow additional funds, because such costs affect only the budget projections, not the actual amount of cash flows. The Fair-Value Accounting and Budget Act has been introduced in the 119th Congress (H.R. 1388) and in the 118th Congress (H.R. 5571) to expand how subsidy costs are measured. In addition to FCRA estimates, these proposals would also require estimates based on fair value. Previously, legislation was introduced in the 114th Congress (S. 399 and H.R. 119) that would have changed the method of calculating subsidy costs to the fair value method. Similar legislative proposals passed the House in the 113th Congress but were not acted on in the Senate. FY2016 budget resolutions in the 114th Congress, S.Con.Res. 11 and H.Con.Res. 27, included provisions that addressed fair value in federal credit programs by requiring CBO to provide fair value estimates for credit programs at the request of the budget committees. S.Con.Res. 11 was adopted by the House on April 30, 2015, and by the Senate on May 5, 2015. Federal credit programs are comprised of government direct loans and loan guarantees, which are available to individuals and firms. These credit programs support a wide range of economic activities, including home ownership, education, small business,1 farming, energy,2 infrastructure investment,3 and exports. At the end of FY2024, outstanding federal credit included direct loans at $1.8 trillion and principal amount of loans guaranteed by the United States at $2.2 trillion, totaling $4.0 trillion.4 The Federal Credit Reform Act of 1990 (FCRA, P.L. 101-508) requires that estimated lifetime net costs of new loans and loan guarantees be recorded in the budget years in which the loans are disbursed.5 The costs of these credit programs, referred to as subsidy costs, are measured on a net present value (NPV) basis—which is the value of expected future cash receipts less expenditures adjusted or discounted over time using an interest rate. The interest rates used to discount the cash flows are estimated based on Treasury securities yields. The methodology used for measuring the costs of these credit programs might affect how Congress allocates the federal budget and structures the credit programs. The profitability or costs of individual credit programs had been a topic of discussion and legislation in recent Congresses. At the core of the congressional debate is whether the subsidy costs of the federal credit programs should continue to be measured with the current method. If a different method is used, the subsidy costs could vary substantially from current estimates. For example, the Fair-Value Accounting and Budget Act has been introduced in the 119th Congress (H.R. 1388) and in the 118th Congress (H.R. 5571) to expand how subsidy costs are measured and would require the Congressional Budget Office (CBO) to provide cost estimates based on both FCRA and fair value estimates. This report explains and analyzes the differences between these methodologies.The Federal Credit Reform Act of 1990 (FCRA; P.L. 101-508)2 requires that estimated lifetime net costs of new loans and loan guarantees be recorded in the budget year in which the loans are disbursed.3 The costs of these credit programs, referred to as subsidy costs, are measured on a net present value (NPV) basis—which is the value of expected future cash receipts, less expenditures adjusted or discounted,2 trillion.
Introduction
over time using an interest rate. The interest rates used to discount the cash flows are estimated based on Treasury securities yields. The methodology used for measuring the costs of these credit programs might affect how Congress allocates the federal budget and structures the credit programs.
Legislation has been introduced in the 114th Congress (including S. 399 and H.R. 119) to change how FCRA subsidy costs are measured. The profitability of individual credit programs, such as the Export-Import Bank, has also been a topic of discussion. At the core of the congressional debate is whether the subsidy costs of the federal credit programs should continue to be measured with the current method. If a different method is used, the subsidy costs could vary substantially from current estimates. Under the existing proposals, the credit programs would be measured based on fair-value accounting.
The Congressional Budget Office (CBO) provides subsidy cost estimates using interest rates equivalent to private markets to determine the subsidy costs for its fair-value subsidy cost estimates. Typically, interest rates in the private markets are higher than Treasury rates. In other words, CBO seeks to estimate what it would cost private industry to offer similar types of loans or loan guarantees.4 Switching to fair value, therefore, is estimated to increase the subsidy costs of credit programs.5
This report first provides a brief explanation of federal credit programs. Next, it examines recent legislative proposals and hearings. It assesses several policy issues Congress might consider and the benefits and challenges of remaining on FCRA cost estimates versus using fair- value estimates to determine subsidy costs. Appendix A provides a more detailed explanation of subsidy costs valuation methods. Appendix B provides an explanation of fair- value accounting requirements for the private sector, SFAS 157 (Topic 820), which would be used for implementing fair value according to some legislative proposals. Appendix C contains definitions and acronyms for certain concepts and terminology used in this report.
known as Topic 820.9 Appendix C contains definitions and acronyms for certain concepts and terminology used in this report.
Key Concepts Net Present Value (NPV)—The value of expected future cash receipts less expenditures adjusted, or discounted, over time using an interest rate. Discounted Cash Flow—A way of measuring future cash flows. For example, $100 received today is worth more than $100 received a year from now, as the $100 received today can be invested for a return that is worth more than $100 in the future. Thus, $100 received a year from now would need to be discounted to account for this forgone earning opportunity. Subsidy Costs—The FCRA defines subsidy cost as "the estimated long-term cost to the government of a direct loan or a loan guarantee … calculated on a NPV basis, excluding administrative costs."10 NPV measures the current value of all cash outflows and inflows at a discounted (interest) rate. The higher the interest rate, the steeper the cash flows in future periods will be discounted. Negative Subsidy Costs—Negative subsidy costs imply that the government is generating positive net income from the loan or loan guarantee program for budgetary purposes. These are recorded on the budget as negative outlays, meaning from a budget perspective they reduce overall spending.
nonfederalnon-Federal borrower under a contract that requires the repayment of such funds with or without interest."611 A loan guarantee is "a pledge with respect to the payment of all or part of the principal or interest on any debt obligation of a non-federalFederal borrower to a non-federal lender."7
Effective FY1992, the FCRA changed the basis of accounting for federal credit programs from cash basis to accrual basis.813 Most of the other items in the federal budget are reported on a cash basis. Before FY1992, for a given fiscal year, the budgetary cost of a direct loan or loan guarantee was the net cash flows for that fiscal year. This cash flow measure did not accurately reflect the ultimate profitability (loss) of the loan, thus. Thus, the methodology, arguably, did not accurately reflect the overall cost of the loan and loan guarantee on budget documents. Beginning with FY1992, the FCRA required that budget reports for credit programs estimate the subsidy costs (see text box) of the credit programs.
Key Concepts Subsidy Costs—FCRA defines subsidy cost as "the estimated long-term cost to the government of a direct loan or a loan guarantee, calculated on a NPV basis, excluding administrative costs." 9 NPV measures the current value of all cash outflows and inflows at a discounted (interest) rate. The higher the interest rate, the deeper the cash flows in future periods will be discounted. Negative Subsidy Costs—Negative subsidy costs imply that the government is generating positive net income from the loan or loan guarantee program for budgetary purposes. These are recorded on the budget as negative outlays, meaning from a budget perspective they reduce overall spending. Discounted Cash Flow—A means of measuring future cash flows. For example, $100 received today is worth more than $100 received a year from now, as the $100 received today can be invested for a return that is greater than $100. Thus, $100 received a year from now would need to be discounted to account for this forgone earning opportunity. Net Present Value (NPV)—The value of expected future cash receipts less expenditures adjusted, or discounted, over time using an interest rate. |
Beginning with FY1992, the FCRA required that budget reports for credit programs estimate the subsidy costs of the credit programs. The FCRA and fair-value methods discount did not accurately reflect the overall cost of the loan and loan guarantee on budget documents. As an example, consider a loan of $100,000 that is fully repaid in 10 years. Under the pre-FCRA accounting methods, this loan would be shown as a $100,000 loss in the year it was originated. The annual payments would be shown as revenue during the repayment years.
10 14 (The fair value method also does this. The subsidy cost estimates under FCRA and fair value differ because the interest rates are not the same., as discussed in more detail below.) Some of the factors that help determine subsidy costs are the
TheGenerally, the interest rate used to discount the cash flows is estimated based on yields of Treasury securities that mature on dates comparable to those on which the loans are substantially disbursed. A two-year loan, for example, is discounted using two-year Treasury rates. Subsidy estimates under FCRA or fair value are only projections of future performance of credit programs. The actual cost of any loans or loan guarantees cannot be determined until the loans have fully matured and all payments have been received.
Government loan programs generally charge borrowers interest rates that are usually higher than Treasury rates but lower than the rates private lenders may charge. Similarly, government loan guarantees are meant to lower the borrowing costs for loans generated by private lenders by removing default risk.absorbing either partial or significant amounts of the losses in the event of default.15 The interest rate and fees charged to borrowersor insurance premiums charged by the government to borrowers arguably may not reflect the costs borne by private lenders. Private lenders might be more loss averse and have a profit motive; their.16 Their lending rates might also reflect administrative costs as well, which FCRA omits from subsidy cost estimates.
Risk is another important factor in determining subsidy cost. For example, one type of risk that is common to the estimation process is the default risk. The text box below reflects some of theother risks identified by the Export-Import Bank for its credit activities. TheThese risks may or may not reflect risks borne by other credit programs. One type of risk that is common to the estimation process under fair value or FCRA is the default risk.
Types of Risks
The risk profile for credit activities can be determined based on either an individual loan or the total portfolio. Default Risk—The risk that a borrower either is unwilling or does not have sufficient resources to make payments. In the context of the Export-Import Bank, default risk (i.e., repayment risk) is synonymous with credit risk. Country Risk—The risk that a borrower's property might be expropriated by a government. Country risk also considers the borrower's inability to pay due to war or being unable to convert domestic currency to U.S. currency. Concentration Risk—Risks from clustering of the credit portfolio by a specific industry, geographic region, or borrower.
Foreign Currency Risk—The risk of an appreciation or depreciation in the value of a foreign currency in relation to the U.S. dollar. Interest Rate Risk—There are different types of interest rate risks. The Export-Import Bank's interest rate risks arise from making fixed-rate loan commitments prior to borrowing to fund loans. The risk arises as the Export-Import Bank borrows at a higher rate than the rate it charges its customers.
Operational Risk—The risk of material losses as a result of human errors, system deficiencies, and management oversight weaknesses. |
Legislative proposals in the 114th Congress would change subsidy cost estimates to be based on a fair-value accounting standard to incorporate market risks. Fair-119th Congress (and previously in the 118th Congress) would require CBO to provide subsidy cost estimates based on fair value and FCRA. Fair value accounting, first, requires the value of assets and liabilities to be estimated based on an objective measurement of how similar types of assets and liabilities are valued in a well-functioning liquid market. If a well-functioning market does not exist, which may be the case for many federal credit programs, then fair- value accounting requires internal modeling. Internal modeling requires estimating NPV of future cash flows by using an interest rate. In certain circumstances, some credit programs might have private- market equivalents. In other instances, the credit programs would require their own internal modeling to determine fair- value subsidy cost estimates. The key difference between CBO's fair- value estimates and FCRA is the use of market interest rates instead of Treasury rates to determine subsidy cost estimates through discounted cash flow analysis. Changing the credit programsprogram's subsidy cost estimate methodology to fair value generally increases the subsidy costs.
, because a higher interest rate would be assumed in the estimate. Fair Value of Loans and Loan Guarantees
The legislative proposals in the GASB's and FASB's standards.
Fair Value of a Loan—Fair value of a loan (an asset) is the price that would be received if it were sold in a competitive market in the private sector Fair Value of a Loan Guarantee—Fair value of a loan guarantee (a liability) is the price that would have to be paid by the government for the private sector to assume the guarantee commitment. |
Another way to consider what constitutes fair value is to examine how government loan guarantees affect the borrower's interest rate. Federal loan guarantees also have the full faith and credit of the United States. In addition,, which has the power to levy taxes on the largest economy in the world to meet its obligations. As a result, investors perceive U.S. government debt as facing zero default risk, which is reflected in the high-quality credit ratings the United States receives from rating agencies allow it to issue debt and levy taxes. This allows the U.S. government to easily issue debt at relatively low interest rates, further enabling it to meet its obligations.12 Private21 Therefore, private lenders value government-guaranteed loans more than loans with no guaranteeguarantees. The difference between the value of a government-guaranteed loan and an equivalent one with no guarantee is, the fair value. This difference in valuation, could also be considered as market risk.
Private- sector lenders with or without government loan guarantees also consider both administrative costs and profit incentives in determining the value of loans. However, FCRA specifically excludes administrative costs in determining subsidy costs. Among the reasons administrative costs were excluded from FCRA may have been that Congress wanted to have direct oversight through appropriations.22 As required by some legislative proposals, if fair value for government credit programs were determined based on private markets' fair- value accounting standards, then such fair- value measurements would also include administrative costs. To be comparable, administrative costs would either need to be either removed from fair- value estimates or included in FCRA estimates. If administrative costs were included with FCRA estimates, then the difference between fair- value estimates and FCRA estimates would diminish. Among the reasons administrative costs were excluded from FCRA may have been that Congress wanted to have direct oversight through appropriations.13
Fair-value estimates have been
While lending programs in the federal government are generally required to be measured based on FCRA, specific statutes have required fair value estimates to be used in official budget estimates in limited cases in the past. For example, government-owned assets acquired through the Trouble Asset Relief Program (TARP) are statutorily required to be valued at fair value.1423 However, there are a few differences to consider in the types of assets generally owned through TARP and other credit programs.1524 With TARP, a significant portion of the activity was linked to the purchase of preferred shares to increase banks' capital levels, with the government taking an ownership interest in those firms. CBO has also used fair value to produce its forecasts for the future activities of Fannie Mae and Freddie Mac.
25
A more detailed explanation of how the cost estimates are determined based on cash method (or pre-FCRA), FCRA (Treasury rates), and fair value is provided in Appendix A. A brief overview of what is discussed in Appendix A is provided in the text box below.
To highlight the different cost estimates, consider an example in which the federal government lends $100 million. The $100 million is expected to be repaid with interest over three years, net of loan defaults. Table 1. Example of Outlays Recorded on the Federal Budget for $100 Million in Direct Loans (measured in millions)
Measured in Millions Accounting Methods 2025 2026 2027 2028 Cash $100 $-35 $-34 $-34 FCRA (Treasury Rates) $-1.6 NA NA NA Fair Value $1.3 NA NA NA |
Over the past several years, Congress has debated the most appropriate way to measure subsidy costs overall and in the context of reforming specific credit programs. Most recently, on June 17, 2015, the Joint Economic Committee held a hearing on the economic exposure of federal credit programs.18 The hearing focused on the difference between how costs of government loans and loan guarantees are measured under FCRA and the proposed fair-value method.
The Fair-Value Accounting and Budget Act (as well as H.R. 5571 in the 118th Congress and H.R. 3785 in the 117th Congress) would require CBO to provide fair value credit estimates, in addition to FCRA estimates, for loan and loan guarantees. The proposal would also require the OMB director to submit a report to the House and Senate Budget Committees on the cost of federal credit programs based on fair value estimates. As stated in the press release related to the introduction of H.R. 1388, the act "seeks to provide Congress with better oversight and information on market risk in federal lending, ensuring a more accurate assessment of costs to taxpayers."27 Arguably, there was greater interest, previously, in the topic of fair value estimates as discussed below. The Joint Economic Committee held a hearing in 2015 on the economic exposure of federal credit programs28 that focused on the difference between how costs of government loans and loan guarantees are measured under FCRA and fair value estimates.Legislation has been introduced to change from FCRA to fair-value subsidy cost estimates, as well as to request CBO to estimate2025. In subsequent years (2026-2028), if changes in the loan portfolio's performance are material, then subsidy costs would be reestimated. This table is included in Appendix A with a more detailed explanation of how the balances are calculated.
Congressional Debate
were introduced in the 113th Congress, the Budget and Accounting Transparency Act of 2014 (S. 2420 and H.R. 1872) were introduced in the 113th Congress. On April 7, 2014, the House passed H.R. 1872 by a recorded vote of 230 yeas and 165 nays-165, but the bill was not acted on by the Senate. The Honest Budget Act of 2013 (H.R. 1270) was also introduced in the 113th Congress. H.R. 1270, 113th Congress) included provisions similar to those in S. 399 and H.R. 119. These set of proposals in the 114th and 113th Congress would require fair-proposals would have required fair value determination to adhere to SFAS 157Topic 820 as promulgated by the Financial Accounting Standards Board (FASB)FASB, which sets accounting standards for the private sector.1929 The specific requirements of SFAS 157 (Topic 820) are discussed in Appendix B.
FY2016 budget resolutions in the 114th Congress, (S.Con.Res. 11 and H.Con.Res. 27, include provisions that address the issue of) address fair value in federal credit programs by requiring CBO to provide fair value of assets and liabilities for certain federal credit programs at the request of the Chairmanchairman of the House or Senate Budget Committees. S.Con.Res. 11 was adopted by the House on April 30, 2015, and by the Senate on May 5, 2015.30 In the 113th Congress, similar provisions were included in the budget resolutions, (S.Con.Res. 11 and H.Con.Res. 25).
Policy Considerations
The S.Con.Res. 11 and H.Con.Res. 25. Legislation introduced in the 114th Congress, the Export-Import Bank Termination Act (H.R. 1605), cites the difference between fair value and Treasury-based subsidy rates (see Figure 2) as one of the reasons to terminate the Export-Import Bank, an issue Congress has debated over the past several years.
The ongoing congressional debate has focused on whether market risks should be incorporated in the subsidy cost estimates of federal credit programs. Incorporation or exclusion of market risks through fair- value accounting affects the credit programs in several ways. If market risks are incorporated through cost estimates using the fair- value method, from a budgetary perspective many of the credit programs will change from profit-making programs to loss-incurring programs. This section of the report first explains the various policy perspectives on incorporating market risks. Next, it discusses how the changes would affect the budget for the credit programs and presents a discussion on the volatility of cost estimates. Lastly, it discusses some of the policy rationales for using fair value or remaining on FCRA.
Opinions varydiffer between fair value and FCRA proponents as to whether market risks incurred by private firms should be included in the federal government's subsidy cost estimates. Proponents of fair- value treatment argue that market risks are risks borne by all investors and lenders that cannot be eliminated through diversification. Thus, the argument goes, the government's credit programs are subject to the same types of market risks as private lenders are. Therefore, costs should be estimated based on fair value. According to CBO, "the government is exposed to market risk through its credit programs because, when the economy is weak, borrowers default on their debt obligations more frequently and recoveries from defaulting borrowers are smaller. That market risk is effectively passed along to taxpayers and beneficiaries of government programs because they bear the consequences of the government's financial losses."2031 Debt not paid by the borrowers is a cost to taxpayers, because taxpayers are ultimately responsible for paying the debt of the U.S. government. Treasury rates fail to capture this risk, proponents argue, because they are considered risk-free rates and backed by the full faith and credit of the United States.21
32
Proponents of fair value also contend that current subsidy estimates imply thethat credit programs are not subject to market risks and arguethat a higher interest rate should be used to discount the cash flows to reflect the market risks. The higher interest rates used for fair- value estimates are meant to reflect the market risks that Treasury rates do not currently capture. According to CBO, "to incorporate the cost of market risk, the fair- value approach generally entails using the discount rates on expected future cash flows that private financial institutions would use. That approach effectively uses market prices to measure the cost to the public of the lower returns on federal loans and loan guarantees when the economy is weak and incomes are relatively low."22
33
Proponents of FCRA argue that current cost estimates do not understate the actual cost to the government, because market risks are not the same as budgetary costs. They argue that accounting for market risks in estimating credit subsidies includes amounts that the government will never actually incur and does not necessarily lead to additional taxes or increased borrowing. This is because market costs affect the official budget estimates but not the actual amount of federal cash flows.2334 In this view, credit programs are considered a tool of the U.S. government, not private lenders. Proponents argue that Treasury interest rates sufficiently reflect the overall risks borne by the government;, and thus, there is no need to measure market risks.35
According to FCRA supporters, basing estimates on fair- value conflicts with budgetary principles and how the budget is recorded for most other government activities. In their view, the federal budget should reflect how much revenue it receives and how much it disburses; it should not include the risk, not the premium that applies to the private sector.2436 Budgetary cost estimates based on fair value may create a mismatch between how much the Treasury pays out and what it records as a cost of the credit programs.25
37
Supporters of FCRA have also stated that the government has certain rights as a sovereign power to collect payments from borrowers that private lenders do not have. These government rights, arguably, lower the cost of lending for the government as compared with the private sector. One example is the Treasury Offset Program (TOP), which gives the federal government certain advantages over the private sector for collecting debt defaults. If delinquent nontax debt is owed by a borrowera borrower owes delinquent nontax debt to a federal or state agency, many types of government payments to the debtor can be withheld or reduced until the debt obligation is met.2638 Federal agencies are required to notify TOPTreasury if a debt is more than 120 days delinquent.27 The39 Treasury's ability to collect on debt owed to the federal government might result in higher recoveries, which could result in higher loan valuations than loans originated in the private sector.40 However, because federal credit programs are generally not driven by a profit motive, the government's incentive to maximize collections might be lower.41
A CBO comparison of 20132025 subsidy cost estimates between FCRA at $2.4 billion and the fair- value method$65.2 billion resulted in a $55.962.7 billion increase in budgetary costs under fair value (see Figure 1).42 The perspective on whether the federal government makes a profit on credit programs, arguably, is a credit program is arguably a matter of perception. For many programs, ifwhen Treasury rates are used, the government would makecosts are less with some programs making a profit, but if fair- value estimates are considered, the government would not make a profitcosts would be higher. As discussed, higher interest rates under fair value result in higher subsidy costs, which are recorded as outlays on the budget.
The increased budgetary cost estimates under fair value may or may not add to the U.S. debt, but they mustwould still be accounted for in the budget. One option is to increase the amount appropriated for credit programs., as FCRA requires the cost of loan programs to be fully paid through appropriations for any related subsidy costs.43 The increased appropriation, absent any other policy changes will, would reflect higher deficit spending from a budgetary perspective, but actual cash outlay might be less than appropriated. Alternatively, these increased costs could be required to be accounted for within the program without increasing the budget. The increased costs due to market risks could be passed on to borrowers through higher interest rates or additional fees required to offset the increased budgetary costs. Alternatively, the government could offset the increased costs by reducing the amount spent on these credit programs. One consequence of passing the increased costs under fair value might be that some borrowers are priced out of these programs.
(in billions of dollars) |
FY2025
In Billions of Dollars
|
Although federal credit programs may recordreflect lower profit or subsidy costsestimates under fair value, if the higher costs are passed ontoon to the borrowers, the federal government might actually make more profit than under estimates based on Treasury rates. In such circumstances, those that are most in need of government credit programs, arguably, might pay for costs associated with market risks through either higher fees or interest rates. Some argue that, if policymakers choose to pass on the higher budgetary costs to borrowers, the social benefits of these programs might be reduced.28 Another option is44 Others have proposed to pay for the increased subsidy costs under fair value that the government could increase tax rates or reduce spending in other program areas.29
45 The subsidy cost estimates for FY2025 and subsidy cost reestimates for prior years could be affected by changes by the President to the loans and loan guarantee programs.46
Figure 2 illustrates how the use of Treasury rates versus fair value affects CBO's cost estimates of specific programs. The three programs—student loans, Export-Import Bank, and single family mortgage insurancetypes of loan programs—housing and real estate loans,47 consumer loans, and student loans48—change from reporting either profits or lower losses profits under Treasury rates to higher losses under fair value over the 10-year period.
.
(in billions of dollars) |
In Billions of Dollars Notes: From CBO—For the Export-Import Bank, the figure shows FCRA and fair-value estimates computed from projected obligations (for direct loans), commitments (for guaranteed loans), and cash flows under current law provided by Office of Management and Budget and the agency. For student loans and guarantees of single-family mortgages, the current-law projections were prepared by CBO. To simplify the analysis, the budgetary estimates for the Department of Education are based on the obligations that CBO estimates the department will incur each year for student loans, rather than on the amount of loan disbursements (which would be the basis for official estimates). Estimates reflecting the timing of loan disbursements would differ slightly from those shown here. |
Notes: FCRA estimates were determined by either CBO, OMB, or other federal agencies. Fair value estimates were determined by CBO. Fair-value subsidy estimates, arguably, are more volatile as compared with subsidy estimates based on Treasury ratespublication/60682.
- value estimates must respond to market risks. Current legislative proposals specifically require subsidy cost estimates to be based on SFAS 157, which requires that fair-value estimates to be calculated annually.
As discussed earlier in the report, TARP cost estimates have been estimated using market risks. Market-based TARP cost estimates over the past seven yearsduring the first decade of TARP illustrate the challenge of incorporating market risks in cost estimates, although, the changes in cost estimates are not solely a result of change in market risks. The largest disparity is between theCBO's March 2009 subsidy cost estimate of $356 billion and $28 billion as of March 201549 and its $31 billion estimate50 in its final report on April 2024, a difference of $328325 billion. The significant downward revision in cost estimates is a result of higher than -than-expected repayments of principal and interest or dividend payments that Treasury has collected. As the principal is paid back, the amount of risk is reduced, which has Over the course of TARP, CBO continued to revise its cost estimates as the principal was paid back and the amount of risk was reduced, which resulted in decreased cost estimates. Further, any amount that is written off or losses realized on sales of stock no longer represent uncertainty.30
51
Cost estimates for TARP programs inherently had more inherent uncertainty than typical credit programs have. Over the life of the TARP programs, CBO identified several factors affecting its revised estimates, including changes in financial market conditions, new transactions, and timing of disbursements. Although disbursements were a factor in revised cost estimates, the changing market conditions also contributed to the increasearguably had a more significant role in the revised cost estimates.3152 To some extent, the change in default risks would have positively affected TARP estimates in later years regardless of whether it was estimated based on market risks or Treasury rates.
Current legislative proposals require fair-valuespecifically require subsidy cost estimates to adhere to accounting standard Topic 820. be based on GASB standard 3100—Fair Value Measurement, issued in February 2015.53 GASB's fair value standard is similar to the standard promulgated by FASB, Topic 820.54 Similar to FCRA (Treasury rates), GASB and Topic 820 require fair value estimates be calculated annually.
Topic 820 specifically states that if an insufficient market exists to determine fair value, then internal modeling is required (see Appendix B). Insufficient markets to determine fair value might imply that risks borne by a specific credit program are high or that there is a lack of private-firm interestinterest from private firms or that the government has crowded out private activity. If markets freeze for one or more asset class, then the interest rate to fund such transactions in the private markets would rise. In such circumstances, fair- value estimates based on internal modeling would reflect the higher discount rate. Thus, the cost of credit programs is likely to be even higher when markets are volatile. Although FCRA Because FCRA relies on internal modeling and does not use fair- value estimates even when the equivalent fair- value markets might exist, neither is it subject to market volatility, as it relies on internal modeling.
its estimates are not as volatile. For certain programs, when a market that has been previously liquid becomes less liquid or freezes, under fair value, estimated subsidy costs are likely to be higher. If the markets become liquid again in the future, the estimated cost of credit programs could see substantial change. The volatility of cost estimates under such market conditions might make it challenging for policymakers and agencies to estimate costs and thus evaluate programs accurately. Even under normal circumstances, government agencies may need time to develop more accurate financial models that estimate subsidy costs based on fair value.
Choosing between FCRA and fair value may depend in part on one's underlying view of the purpose of federal credit programs. Fair- value proponents argue that the transaction price the government pays to purchase private- sector goods, —such as electricity, military hardware, computers, and buildings, —is at a fair- value price and that federal credit programs should be as well. The transfer of federal funds through loans enables the borrower to purchase goods at fair value; therefore. Therefore, subsidy cost estimates should reflect that.3255 A corollary to this line of reasoning is grounded in the idea that government is expected to compensate property owners at fair value—not at a value determined by the government—when its power of eminent domain is used to acquire properties.33
56
Fair- value supporters also argue that federal credit programs could crowd out private- market activity if federal credit programs are competing with private lending at below-market rates. Further, federal credit programs could encourage higher debt levels by lowering borrowing costs and increase market risks by encouraging excessive risk-taking.3457 The more the government becomes involved in the credit marketmarkets, the greater the concentration of various risks, including market risks on the government's balance sheet.35
58
Proponents of FCRA argue that adjusting for private- sector risks could be a significant barrier to the federal government taking action in markets that involve risks. It might discourage or undermine support for the federal government to take action in programs such as student loans, mortgage guarantees, and many others. In addition, supporters of FCRA state that the effective functioning of federal credit programs can reduce existing market failures economically through, potentially improving equity and efficiency. Equity addresses the issues of distribution of income, consumption, and wealth. Efficiency focuses on how private financial intermediaries should allocate capital to its most productive uses. Despite private financial intermediation, market imperfections may exist when social costs and benefits are not equal to private costs and benefits. According to the Office of Management and Budget (OMB)previous OMB documents, certain private- market imperfections, —such as "information failures, monitoring problems, limited ability to secure resources, insufficient competition, externalities, and financial market instability," "—might justify federal intervention.36
59
Supporters of federal credit programs suggest that trying to capture the increased costs may affect the availability of government credit negatively and that fair- value estimates ignore the societal benefits of the credit programs.3760 There are many reasons why an individual or business may need to access credit programs. One reason might be lower borrowing costs; another. Another might be the inability to obtain credit from the private market.
Appendix A. Comparison of Valuation Methods
This appendix explains theprovides a comparison of three different methods of estimating the federal credit programs' costs—: cash, FCRA, and fair value. The amount disbursed—that is, cash lent to borrowers and received as repayments of principal and interest, —is the same under each method. The amount recorded as costs (outlays) for the budget is different under each method.
For example, consider a $100 million portfolio of direct loans made by the government. The first method, the cash method, does not recognize the lifetime costs of the loan at inception. It records the costs of the loan each year based on cash flows (see Table A-1). The cash method recognizes the cash outlay of $100 million as costs to the government in 20152025 (the current year), and in subsequent years, the cash method recognizes the principal and interest payments as negative cash outlays.
By contrast, both FCRA and fair- value methods recognize the lifetime costs (subsidy costs) of the loan at inception. FCRA defines subsidy costs as "the estimated long-term cost to the Government of a direct loan or loan guarantee, … calculated on a net present value basis,38 excluding administrative costs."3961 In Table A-1, the $100 million direct loan has an estimated negative subsidy cost (positive return) of $1.6 million based on FCRA estimates, whereas the subsidy cost to the government is $1.3 million based on fair- value estimates. Unlike the cash method, neither FCRA nor fair value generally records budget outlays in subsequent years once the loans have been substantially disbursed. If the estimated value of the loan portfolio changes, then subsidy costs would also need to be reestimated. The reestimated costs are incorporated into the budget for future years.40
62
Table A-1. Outlays Recorded on the Federal Budget for $100 Million
Measured in Millions Accounting Methods 2025 2026 2027 2028 Cash $100 $-35 $-34 $-34 Treasury Rates $-1.6 NA NA NA Fair Value $1.3 NA NA NA Notes: Positive numbers reflect cash outlays (i.e., costs) to the government, and negative numbers reflect cash receipts (i.e., negative costs). For budgetary purposes, subsidy cost estimates based on Treasury rates or fair value are recognized only in 2025. In subsequent years (2026-2028), if changes in the loan portfolio's performance are material, then subsidy costs would be reestimated.
Source: Adopted from Congressional Budget Office (CBO), How CBO Produces Fair value Estimates of the Cost of Federal Credit Programs: A Primer, July 2018, pp. 4-5, https://www.cbo.gov/publication/53886.
Table A-1. Outlays Recorded on the Federal Budget for $100 Million
(measured in millions)
Accounting Methods |
2015 |
2016 |
2017 |
2018 |
Cash |
$100 |
$-35 |
$-34 |
$-34 |
Treasury Rates |
$-1.6 |
NA |
NA |
NA |
Fair Value |
$1.3 |
NA |
NA |
NA |
Source: Adopted from Congressional Budget Office (CBO), Fair Value Accounting for Federal Credit Programs, March 2012, p. 9, http://www.cbo.gov/sites/default/files/03-05-FairValue_Brief.pdf.
Notes: Positive numbers reflect cash outlays (costs) to the government and negative numbers reflect cash receipts (negative costs). For budgetary purposes, subsidy costs estimates based on Treasury rates or fair value are recognized only in 2015. In subsequent years 2016-2018, the subsidy costs estimates based on Treasury rates or fair value are not reestimated or recognized unless the loan portfolio's performance expectations change substantially.
The rest of this appendix discusses in more detail how the outlays in Table A-1 are determined for the same loan under each of these approaches. It begins with the cash method, —which was used prior to the enactment of FCRA; —explains how credit programs' costs are currently determined under FCRA;, and lastly explains the proposed fair- value method.
Cash Method
The federal government generally records revenue (i.e., cash receipts) and spending (i.e., cash outlays) on a cash basis. The difference between revenue and spending results in either a budget deficit or a surplus. As shown in Table A-2, under the cash method, disbursements are recorded as outlays, and interest and principal repayments are recorded as cash inflows (or negative outlays), in the respective fiscal years in which the transaction occurs.41
63
Table A-2 provides an example of how federal direct loans4264 and repayments with interest would be recorded under the cash method over three years. The positive numbers reflect cash expenditure by the government, and negative numbers reflect cash receipts. In FY2015FY2025, the example assumes that $100 million in direct loans is recorded as cash outlays. Over the next three years (FY2016-FY2018FY2026-FY2028), the principal with interest at 3% is repaid to the government, while while recognizing $1 million in expected loan defaults is anticipated each year.
each year.
Year |
2015 |
2016 |
2017 |
2018 |
|
$100 |
$0 |
$0 |
$0 |
|
$0 |
$-3 |
$-2 |
$-1 |
|
$0 |
$-33 |
$-33 |
$-34 |
|
$0 |
$1 |
$1 |
$1 |
|
$100 |
$-35 |
$-34 |
$-34 |
Source: Adopted from CBO, Fair Value Accounting for Federal Credit Programs, March 2012, p. 9, http://www.cbo.gov/sites/default/files/03-05-FairValue_Brief.pdf.
Notes: Positive numbers reflect cash outlays (costs) to the government and negative numbers reflect cash inflows (negative costs).
a.
The $100 million in the first year is considered an expense to the government.
Measured in Millions Year 2025 2026 2027 2028 $100 $0 $0 $0 $0 $-3 $-2 $-1 $0 $-33 $-33 $-34 $0 $1 $1 $1 $100 $-35 $-34 $-34 Notes: Positive numbers reflect cash outlays (costs) to the government, and negative numbers reflect cash inflows (negative costs). b.
Table A-2. Disbursements and Receipts Under the Cash Method
A. Disbursementsa
B. Scheduled Interest Paymentsb
C. Scheduled Principal Repaymentsc
D. Default Lossesd
E. Cash Flows for the Federal Government (A+B+C+D)e
Source: Adopted from CBO, How CBO Produces Fair value Estimates of the Cost of Federal Credit Programs: A Primer, July 2018, pp. 4-5, https://www.cbo.gov/publication/53886.
FY2016FY2026 interest is determined as 3% of the $100 million principal,: $100 million x 3% = $3 million. In FY2017FY2027, as $33 million was repaid the previous year, the interest is calculated on the remaining balance,: $66 million x 3%= $2 million. In FY2018FY2028, as an additional $33 million was repaid in the previous year (total of $66 million repaid), interest is determined based on $33 million x 3% = $1million.
c.
c. Repayment of the original loan.
d.
d. Default losses are the amount that is not expected to be repaid by the borrower.
e.
e. The $100 million in direct loans at inception is considered an expenditure. In subsequent years, the amount recorded reflects the portion of the loan principal repaid, net of interest payments and defaults. For FY2016FY2026, the amount received by the government is as follows: interest payment $3 million + $33 million principal repayment - $1 million default, resultsresulting in the government receiving $35 million.
The cash method did not accurately measure the estimated long-term costs of the credit programs to the government. Before FCRA, $100 million in direct loans had the same effect on the budget estimates in the first year as $100 million in federal grants did. Direct loans had a significantly smaller impact than did grants on the budget outlays over the term of the loan, but budget estimates under the cash method did not account for the cash the government received in subsequent years for principal repayment and interest. In other words, the costs to the government are exaggerated when loans are disbursed, because the cash flows over the life of the loan are not considered for the respective year's budget. Similarly, revenues received by the government are exaggerated in subsequent years as the initial cash outflow is not taken into consideration. If the loan matured outside of the budget windows, the loan could appear to have a cost even if it were profitable to the government.
By contrast, loan guarantees appeared to produce income for the government, because borrowers were required to pay a premiumpremiums or origination feefees at inception without any cash outlays by the government. In subsequent years, if the borrower defaulted, then cash outlays by the government were recorded as an expenseexpenses. The cash method favored loan guarantees over direct loans, because it appeared to reflect lower costs. Even when direct loans might have been the most feasible way for Congress to achieve its policy objectives, the more favorable budgetary treatment of loan guarantees, arguably, made it difficult for Congress to justify direct loans over loan guarantees. The cash method also has no discounting for net present valueNPV determination. If interest payments were greater than loan defaults, then all loans appear to be moneymakers. For these reasons, cash -basis accounting made it challenging to measure the actual cost of the credit programs and led Congress to enact FCRA.
to the enactment of FCRA. FCRA (Treasury Rates)
Unlike the cash method, FCRA recognizes subsidy costs of the loans at inception. Currently, subsidy costs are estimated as the Net Present Value (NPV)NPV of future cash flows using a method prescribed in FCRA. NPV helps analyze financial information by making loan inflows and outflows over the life of the loan comparable. Subsidy costs are estimated using Treasury interest rates at the time loans or loan guarantees are substantially disbursed.4365 The interest rate charged to the borrowers varies, but they are generally higher than Treasury rates. The difference between the two interest rates resultscould result in credit programs generally reporting positive net cash flows even after considering loan lossesif loan losses are minimized. Positive net cash flow is called negative subsidy. In the private sector, these gains would be equivalent to firms making profits. FCRA estimates exclude administrative costs when determining subsidy costs, giving Congress flexibility to have greater input on administrative costs through appropriations, but reducing subsidy cost estimates if viewed in isolation.
.
The discount rate applied to determine the NPV of disbursed loans is based on the average interest rate of Treasury securities of similar maturity dates as the loans. The NPV of cash flows takes into consideration both cash outflows (e.g., disbursements, adjustments for defaults, delinquencies, and interest subsidies) and inflows (e.g., repayments of principal and interest, prepayments, fees, and penalties).44
66
Table A-3 provides an example of how subsidy costs of federal direct loans45 are determined using Treasury rates. The Treasury discount rates are higher in later years to account for various risk factors. Generally, the longer the repayment terms of the loan, the higher the interest rate to account for the risks. To determine the subsidy costs, the net cash flows for each year are discounted based on the Treasury rate applicable to that year (see Table A-3, row G). The sum of these cash flows determines the subsidy costs (see Table A-3, row H).
row H).
Table A-3. FCRA (Treasury Rate) Subsidy Costs
Measured in Millions Year 2025 2026 2027 2028 $100 $ 0 $0 $0 $0 $-3 $-2 $-1 $0 $-33 $-33 $-34 $ 0 $1 $1 $1 $100 $-35 $-34 $-34 0 0.25 0.50 1.00 $100 $-34.9 $-33.7 $-33.0 $-1.6 NA NA NA Notes: Positive numbers reflect cash outlays (costs) to the government, whereas negative numbers reflect cash receipts (negative costs). For budgetary purposes, subsidy cost estimates based on Treasury rates or fair value are recognized only in 2025. In subsequent years (2026-2028), the subsidy costs estimates based on Treasury rates or fair value are not reestimated or recognized unless the loan portfolio's performance expectations change substantially.
A. Disbursementsa
B. Scheduled Interest Paymentsb
C. Scheduled Principal Paymentsc
D. Default Lossesd
E. Cash Flows Before Discountinge (A+B+C+D)
F. Treasury Discount Rate (percentage per annum)f
G. Discounted Cash Flow Based on Treasury Ratesg (row E discounted based on Treasury rates)
H. Subsidy Costs Sum of row G ($100-$34.9-$33.7-$33.0) = $-1.6
Source: Adopted from Wendy Kiska et al., How CBO Produces Fair value Estimates of the Cost of Federal Credit Programs: A Primer, CBO, July 2018, pp. 4-5, https://www.cbo.gov/publication/53886.
Table A-3. FCRA (Treasury Rate) Subsidy Costs
(measured in millions)
Year |
2015 |
2016 |
2017 |
2018 |
|
$100 |
$ 0 |
$0 |
$0 |
|
$0 |
$-3 |
$-2 |
$-1 |
|
$0 |
$-33 |
$-33 |
$-34 |
|
$ 0 |
$1 |
$1 |
$1 |
(Sum of rows A+B+C+D) |
$100 |
$-35 |
$-34 |
$-34 |
|
0 |
0.25 |
0.50 |
1.00 |
(row E discounted based on treasury rates) |
$100 |
$-34.9 |
$-33.7 |
$-33.0 |
H. Subsidy Costs Sum of row G ($100-$34.9-$33.7-$33.0)= $-1.6 |
$-1.6 |
NA |
NA |
NA |
Source: Adopted from CBO, Fair Value Accounting for Federal Credit Programs, March 2012, p. 9, http://www.cbo.gov/sites/default/files/03-05-FairValue_Brief.pdf.
Notes: Positive numbers reflect cash outlays (costs) to the government whereas negative numbers reflect cash receipts (negative costs). For budgetary purposes, subsidy costs estimates based on Treasury rates or fair value are recognized only in 2015. In subsequent years 2016-2018, the subsidy costs estimates based on Treasury rates or fair value are not reestimated or recognized unless the loan portfolio's performance expectations change substantially.
a.
The amount disbursed by the government, $100 million.
b.
Interest payments received per year at 3% of principal, as Table A-3 reflects the interest payments decline in subsequent years as a portion of the principal is repaid. For example, in FY2016 interest is determined as 3% of the $100 million principal, $100 million x 3% = $3 million. In FY2017 as $33 million was repaid in the previous year, the interest is calculated for the remaining balance, $66 million x 3%= $2 million. In FY2018 as an additional $33 million is repaid in the previous year (total of $66 million repaid) interest is determined based on $33 million x3%=$1million.
c.
Repayment of the original loan.
d.
Default Losses are the amount that is not expected to be repaid by the borrower.
e.
For each respective year, row E reflects the amount disbursed or the amount expected to be received by the government.
f.
Treasury discount rates for each year. Generally, the longer the repayment terms of a loan, the higher the interest rate that is charged to the borrower to account for various risks borne by the lender.
g.
The balance from row E discounted based on Treasury rates, Net Present Value (NPV). For example, as the Treasury discount rate is 0.25% in FY2016, the discounted cash flow rate is determined by 1÷1.0025=0.998. Thus, the discounted cash flow is $35 million x 0.998 = $34.9 million, which is remitted to the Treasury. FY2017 discounted rate is determined by 1÷1.005^2 = 0.990. The discounted cash flow for FY2017 is $34 million x 0.99 = $33.7 million. FY2018 cash flows are determined similar to FY2016 and FY2017.
The same Treasury discount rate tables are used across all federal credit programs for loans of similar maturity dates. Use of Treasury rates creates a certain degree of consistency in how credit programs subsidy costs are determined and increases comparability across programs. The same discount rate tables are also used for budget formulation and financial reporting by the federal government, arguably, creating further consistency in how financial transactions within the government are measured.46
67
After the enactment of FCRA, the Office and Management Budget (OMB)OMB created new guidelines for determining subsidy costs. Congress codified several of the OMB guidelines through the Balanced Budget Act of 1997.4768 One of those requirements is for government agencies to have a data-driven decisionmakingdecision making model, which includes statistical models to predict loan defaults and other deviations from loan contracts. Further, prior to making loans or issuing loan guarantees, the agencies are required to regularly analyze and provide estimated subsidy costs to obtain budgetary approval.48
69 Fair Value
Treasury discount rates are the interest rates at which the federal government borrows. Fair- value rates are closer to the interest rates that private lenders charge borrowers. Interest rates that private lenders charge are higher than Treasury rates. Higher interest rates result in cash flows being discounted more in later years. Which interest rates are used to determine the subsidy costs is a significant driver in whether most credit programs show an annual gain from their lending activities.
determining the cost of federal credit programs.
Table A-4 provides an example of how subsidy costs of federal direct loans might be determined using private- market interest rates. Similar to FCRA, subsidy costs of the loans would be determined at inception. To determine subsidy costs, net cash flows for each year are discounted based on the market rates applicable to that year (see Table A-4, row G). The sum of these cash flows determines the subsidy costs (see Table A-4, row H). Similar to Treasury discount rates, market rates are higher in later years to account for various risk factors. In this example, the switch from Treasury rates to market rates reduces the discounted value of inflows (row G) in FY2016-2018FY2026-FY2028 enough to switch the program from a negative subsidy to a positive subsidy.
Table A-4. Fair Measured in Millions Year 2025 2026 2027 2028 $100 $0 $0 $0 $0 $-3 $-2 $-1 $0 $-33 $-33 $-34 $0 $1 $1 $1 $100 $-35 $-34 $-34 0 1.75 2.00 2.50 $100 $-34.4 $-32.7 $-31.6 $1.3 NA NA NA Notes: Positive numbers reflect cash outlays (costs) to the government, whereas negative numbers reflect cash receipts (negative costs). For budgetary purposes, subsidy cost estimates based on Treasury rates or fair value are recognized only in 2025. In subsequent years (2026-2028), the subsidy cost estimates based on Treasury rates or fair value are not reestimated or recognized unless the loan portfolio's performance expectations change materially.- Value Subsidy Costs
A. Disbursementsa
B. Scheduled Interest Paymentsb
C. Scheduled Principal Paymentsc
D. Default Lossesd
E. Cash Flows before discountinge (A+B+C+D)
F. Fair value Discount Rate (percent per annum)f
G. Discounted Cash Flow Based on Fair Valueg (row E discounted based on fair value)
H. Subsidy Costs Sum of row G. ($100-$34.4-$32.7-$31.6) = -$1.3h
Source: Adopted from Wendy Kiska et al., How CBO Produces Fair value Estimates of the Cost of Federal Credit Programs: A Primer, CBO, July 2018, pp. 4-5, https://www.cbo.gov/publication/53886.
Value Subsidy Costs
(measured in millions)
Year |
2015 |
2016 |
2017 |
2018 |
|
$100 |
$0 |
$0 |
$0 |
|
$0 |
$-3 |
$-2 |
$-1 |
|
$0 |
$-33 |
$-33 |
$-34 |
|
$0 |
$1 |
$1 |
$1 |
(Sum of rows A+B+C+D) |
$100 |
$-35 |
$-34 |
$-34 |
|
0 |
1.75 |
2.00 |
2.50 |
(Row E discounted based on fair value) |
$100 |
$-34.4 |
$-32.7 |
$-31.6 |
H. Subsidy Costs
|
$1.3 |
NA |
NA |
NA |
Source: Adopted from CBO, Fair Value Accounting for Federal Credit Programs, March 2012, p. 9, http://www.cbo.gov/sites/default/files/03-05-FairValue_Brief.pdf.
Notes: Positive numbers reflect cash outlays (costs) to the government whereas negative numbers reflect cash receipts (negative costs). For budgetary purposes, subsidy costs estimates based on Treasury rates or fair value are recognized only in 2015. In subsequent years 2016-2018, the subsidy costs estimates based on Treasury rates or fair value are not reestimated or recognized unless the loan portfolio's performance expectations change substantially.
a.
The amount disbursed by the government, $100 million.
b.
Interest payments received per year at 3% of principal, as Table A-4 reflects the interest payments decline in subsequent years as a portion of the principal is repaid. For example, in FY2016 interest is determined as 3% of the $100 million principal, $100 million x 3% = $3 million. In FY2017 as $33 million was repaid in the previous year, the interest is calculated for the remaining balance, $66 million x 3%= $2 million. In FY2018 as an additional $33 million is repaid in the previous year (total of $66 million repaid) interest is determined based on $33 million x3%=$1million.
c.
Repayment of the original loan.
d.
Default Losses are the amount that is not expected to be repaid by the borrower.
e.
For each respective year, row E reflects the amount disbursed or the amount expected to be received by the government.
f.
Fair-Value discount rates for each year. Generally, the longer the repayment terms of a loan, the higher the interest rate that is charged to the borrower to account for various risks borne by the lender.
g.
The balance from row E discounted based on fair-value rates, Net Present Value (NPV). As an example, since the fair-value rate is 1.75% in FY2016, the discounted cash flow rate is determined by 1÷1.0175=0.983. Thus, the discounted cash flow is $35 million x 0.983 = $34.4 million, which is remitted to the treasury. FY2017 discounted rate is determined by 1÷1.0200^2 = 0.961. The discounted cash flow for FY2017 is $34 million x 0.961 = $32.7 million. FY2018 cash flows are determined similar to FY2016 and FY2017.
h.
As in this example, when there is a negative cash flow, it is a budgetary expense. Thus in the table the cost is recorded as a positive number reflecting the budgetary cost.
Unlike FCRA requirements in which Treasury discount rates apply across all federal credit programs, fair value takes into consideration the various risks that are unique to each type of loan portfolio. Consequently, various credit programs and loan categories within each program would use different market rates. Further, the credit programs interest rates will no longer be the same rates used for budget formulation and financial reporting.
Legislative proposals such as S. 399 and H.R. 119 require that fair market values be determined as specified by Statement of Financial Accounting Standard 157 or its successor, currently identified as Topic 820. In 2009, the Financial Accounting Standards Board (FASB) adopted a new codification system and reclassified previous accounting standards.49 Accounting Standards Codification Topic 820 has superseded Statement of Financial Accounting Standards 157 (SFAS 157) for fair-value accounting.
The legislative proposals in the 119th Congress (H.R. 1388) and in the 118th Congress (H.R. 5571) would require fair value subsidy cost estimates to be based on GASB's "Fair Value Measurement and Application," issued in February 2015. This standard70 is similar to the standard promulgated by Financial Accounting Standard Board (FASB), called Topic 820.71 Currently, GASB's fair value standards are applicable only to state and local governments that adopt those standards, not to federal government entities. Similarly, FASB's standard on fair value measurement, Topic 820, is applicable to private sector companies, not to federal government entities. Definitions of fair value of a loan and fair value of a loan guarantee below are explained in the context of FASB's Topic 820, updated in June 2022. They could be considered to more reflective of current economic circumstances.72 H.R. 1388 would require fair value standards to be applied in addition to FCRA when measuring costs of federal credit programs.Topic 820 defines fair value
Appendix B.
Topic 820
50 73 Market participants are is defined as buyers and sellers that are (1) not related parties;, (2) knowledgeable about the asset or liability and the transaction;, and (3) willing and able to enter into the transaction. In simpler terms, fair value might be described as how much someone, who is not influenced by any other incentives, is willing to pay for an asset or receive for a liability. Alternatively, it is the entry or exit price at which a willing buyer orand seller would agree to exchange an asset or liability.51
74
Topic 820 does not specify how to develop a fair- value model. It does not have specific rules on how to identify active markets, nor does it specify which model or how many models to use when valuing assets or liabilities.5275 Topic 820 specifies disclosure objectives but does not specify disclosure requirements.
Federal financial reporting standards are promulgated by the Federal Accounting Standards Advisory Board (FASAB). Should Congress adopt legislative proposals requiring fair value be adopted by Congress, FASAB would need to update the federal government's accounting standards in conjunction with OMB. Currently, Statement of Federal Financial Accounting Standards (SFFAS) 2 specifies how to account for loans and loan guarantees. SFFAS 2It was customized to meet the requirements of FCRA. As further explained below, should private markets not exist for a specific loan program, FASAB would need to develop accounting standards that are specific to the needs of the federal government. Alternatively, Congress could require FASAB to create accounting standards that are similar to SFAS 157Topic 820 for federal credit programs, without specifically requiring SFAS 157Topic 820. The remainder of this section provides an overview of the two main parts of Topic 820,: levels 1 through -3 fair- value hierarchy and disclosure objectives.
Three Levels of Fair- Value Hierarchy
Topic 820 does create a 3creates a three-level fair- value hierarchy based on the type of information (inputs) available to determine the fair value. In essence, based on the available information, level 1 is the most preferred, whereas level 3 is the least preferred because it is the most subjective. At level 3, Topic 820 allows the management substantial discretion in how fair value is determined and interpreted.53
76
Level 1—RepresentsThis represents the most objective fair- value estimate, and it includes assets and liabilities that have an active market.
Inputs for determining the fair value for level 1 are measured based on active independent markets (e.g., equity that is actively traded on a major exchange, such as the New York Stock Exchange [NYSE] or Chicago Board of Exchange [CBOE]). The markets must be active and trade in identical assets or liabilities regularly.
Level 2—Assets and liabilities at Levellevel 2 have no active markets, but they have either a parallel market or alternate means to estimate fair value.
Inputs for determining the fair value for level 2 are observable data elements, but they are not from an active independent market as with level 1. Quotes from similar but not identical assets or liabilities that were recently traded or transferred between external entities might be used to determine fair value. Other observable inputs, such as interest rates and default rates, may also be used. Unlike level 3 inputs, level 2 inputs are not influenced or provided by the reporting entity, but rather. Rather, they are all external and independent.
Level 3—Fair-For this level, fair value estimates for assets and liabilities are based on agenciesan agency's own internal assumption about how the market would price the asset or liability. There are no observable (market) data on the asset or liability, but data may include data such as interest rates and default rates.
Inputs for determining the fair value for level 3 are provided by the reporting entity. Inputs might include the entity's estimation of future cash flows, earnings, revenue, or sales as well as other information.
Unlike Topic 820, the majority of the accounting standards as promulgated by FASB are considered rules -based. Topic 820 is considered principles -based and does not include specific rules on how agenciesentities should adopt fair- value measures.
Further, most federal direct loans and loan guarantees issued by the government are held to maturity. By contrast, private lenders often sell their loan portfolios to third parties at fair value. Fair value considers the price someone is willing to pay to buy the asset or to assume the liability.54 Similar to how the federal government currently holds loans and loan guarantees to maturity, certain investments in state and local governments and private entities are also held to maturity. Topic 820 as promulgated by FASB for the private sector requires that investments held to maturity be valued at amortized cost.5577 Similarly, Statement Number 72 promulgated by the Governmental Accounting Standards Board (GASB) also requires that investments held to maturity be valued at amortized cost.56
78 Disclosure Requirements
Topic 820 states that disclosure in financial statements should help users of financial statements understand the valuation technique and inputs used to develop fair- value measurements. Disclosures should communicate the effects on net assets of the entity due to fair- value measurements.5779 If an asset or a liability is measured at a different level than how it was reported in the prior period, the circumstances that led to the change should be disclosed. For levels 2 and 3, any changes to the approach (e.g., market, income, or the cost) and any changes to inputs should be disclosed. The reason for the changes in the methodology should also be sufficiently explained.
Credit risks and counterparty credit risks are required to be disclosed. Credit risks in the private sector take into consideration any changes to the reporting entity's credit ratings and how it affects the fair value of its assets and liabilities. Counterparty credit risks account for the borrower's ability to meet its debt obligation. For federal credit programs, arguably, it willit may be difficult to gauge the credit risks of individual programs. Each credit program needs to be evaluated based on the overall credit risk of the U.S. government, because credit programs are backed by the full faith and credit of the United States. The credit risk of the United States is generally assumed to be zero. Counterparty risks for each credit program would need to be individually evaluated, as each program might have unique factors that are applicable to its portfolio. Many of these types of disclosures are not relevant under FCRA.
Appendix C. Key Definitions and Acronyms
The following definitions are described in the context of how they are discussed in this report; they. They are not meant to be precise legal or finance definitions.
Amortized Cost—Amortization refers toThe increase or decrease in investments or debt as a result of interest and principal payments.
Default—The failure of a borrower to make interest and principal payments.
Direct LoansLoan—FCRA defines direct loansloan as "a disbursement of funds by the Government to a non-Federal borrower under a contract that requires the repayment of such funds with or without interest." 58
80
Discounted Cash Flows—Measures theThe value of future cash flows based on a discounted rate (i.e., interest rate).
Efficiency—How private financial intermediaries should allocate capital to its most productive uses.
Equity—The issues of distribution of income, consumption, and wealth.
Federal GrantsGrant—A financial asset transferred or awarded by the federal government to an eligible state, municipality, or an individual, andindividual that is not expected to be repaid.
Loan GuaranteesGuarantee—FCRA defines loan guarantee as "any guarantee, insurance, or other pledge with respect to the payment of all or a part of the principal or interest on any debt obligation of a non-Federal borrower to a non-Federal lender, but does not include the insurance of deposits, shares, or other withdrawable accounts in financial institutions." 59
81
Negative Subsidy Costs—Opposite of subsidy costs, which implyAn instance where the current value of all cash inflows is greater than for all cash outflow. This implies that for budgetary purposes the government is generating income from the loan or loan guarantee programs, see. See Subsidy Costs.
Net Present Value (NPV)—Measures theThe current value of all cash outflows and inflows at a discounted rate (or interest rate). It is a way to measure against competing investment choices.60
Subsidy Costs—The estimated long-term cost to the government of a direct loan or loan guarantee, calculated on a net present basis, excluding administrative costs.
Frequently Used Abbreviations
CBO |
Congressional Budget Office |
FCRA |
Federal Credit Reform Act of 1990 or federal credit programs defined by the act. |
FV |
Fair Value |
GSE |
Government Sponsored Enterprise (GSE) |
NPV |
Net Present Value |
OMB |
Office of Management and Budget |
Author Contact Information
Acknowledgments
[author name scrubbed] and Sean Hoskins provided valuable help in the writing of this report.
1. |
Executive Office of the President, Office of Management and Budget (OMB), Fiscal Year (FY) 2016 Analytical Perspectives of the U.S. Government, February 2, 2015, p. 323, https://www.whitehouse.gov/omb/budget/Analytical_Perspectives. |
|||||||||||||||||||||||
2. | ||||||||||||||||||||||||
3. |
2 U.S.C. §661c. |
|||||||||||||||||||||||
4. | calculated on a net present basis, excluding administrative costs.
Marc Labonte, Darryl Getter, and Anthony Cilluffo provided valuable help in writing this report. For loan programs related to small business, see CRS Report R47909, Small Business Administration (SBA) Business Loan Program Rule Changes in the 118th Congress: Background and Summary, by Anthony A. Cilluffo. For a few of the loans programs related to the energy sector, see CRS Insight IN11984, Inflation Reduction Act of 2022 (IRA): Department of Energy Loan Guarantee Programs, by Phillip Brown. There are many types of loan programs funded by the federal government related infrastructure. For a discussion of few of these loans, see CRS Report R46892, Infrastructure Investment and Jobs Act (IIJA): Drinking Water and Wastewater Infrastructure, by Elena H. Humphreys and Jonathan L. Ramseur; and CRS Report R48097, Energy and Water Development: FY2025 Appropriations, by Mark Holt and Anna E. Normand. Department of the Treasury, Financial Report of the United States Government: Fiscal Year 2024, January 16, 2025, pp. 86, 88, https://www.fiscal.treasury.gov/reports-statements/financial-report/current-report.html. 2 U.S.C. §661c. Wendy Kiska, Estimates of the Cost of Federal Credit Programs in 2025, CBO, August 2024, https://www.cbo.gov/publication/60682. | |||||||||||||||||||||||
5. |
| |||||||||||||||||||||||
6. |
2 U.S.C. §661a(1). |
|||||||||||||||||||||||
7. |
2 U.S.C. §661a(3). |
|||||||||||||||||||||||
8. | 2025.
Topic 820's predecessor was referred to as Statement of Financial Accounting Standards 157 (SFAS 157). Financial Accounting Standards Board (FASB), "Fair Value Measurement (Topic 820)," June 2022, https://www.fasb.org/page/ShowPdf?path=ASU%202022-03.pdf. 2 U.S.C. §661a(5)(A); P.L. 101-508. 2 U.S.C. §661a(1). 2 U.S.C. §661a(3). | |||||||||||||||||||||||
9. |
2 U.S.C. §661a(5)(A); P.L. 101-508. |
|||||||||||||||||||||||
10. |
The Balanced Budget Act of 1997 (P.L. 105-33) was enacted, portions of which amended the Federal Credit Reform Act of 1990 (FCRA; P.L. 101-508) to make technical changes, including codifying several guidelines developed by OMB over the previous years. |
|||||||||||||||||||||||
11. |
Export-Import Bank of the United States, Export Import Bank of the United States Annual Report 2014, November 2014, pp. 68-69, http://www.exim.gov/news/reports/annual-reports. |
|||||||||||||||||||||||
12. |
The sovereign credit rating of the United States remains one of the highest in the world despite Standard & Poor's lowering the rating to the second-highest rank within their ranking system in 2011. Lisa M. Schineller and John B. Chambers, United States of America 'AA+/A-1' Ratings Affirmed; Outlook Remains Stable, Standard & Poor's Rating Services, June 10, 2015, http://www.standardandpoors.com/spf/upload/Events_US/US_SV71_Webcast_art4.pdf. |
|||||||||||||||||||||||
13. |
The Balanced Budget Act of 1997 (P.L. 105-33) amended the FCRA to make technical changes, including codifying several guidelines developed by the Office of Management and Budget over the previous years. Wendy Kiska et al., Public-Private Risk Sharing in Federal Credit Programs, CBO, September 2024, https://www.cbo.gov/system/files/2024-09/59408-Public-Private-Risk-Sharing.pdf. Government Accountability Office (GAO), Credit Reform: Current Method to Estimate Credit Subsidy Costs Is More Appropriate for Budget Estimates Than a Fair Value Approach, GAO-16-41, January 2016, https://www.gao.gov/assets/gao-16-41.pdf. Export-Import Bank of the United States, 2024 EXIM Annual Report, November 15, 2024, p. 54, https://img.exim.gov/s3fs-public/reports/annual/2024/FY2024+Final.pdf. Export-Import Bank, 2024 EXIM Annual Report. Governmental Accounting Standards Board (GASB), "Fair Value Measurement," February 2015, https://gars.gasb.org/3154072/2147483431. FASB, "Fair Value Measurement (Topic 820)." The sovereign credit rating of the United States remains one of the highest in the world despite the credit rating being downgraded by all three major credit rating agencies to the second-highest rank within their ranking system to Aa1. Peter G Peterson Foundation, "Moody's Downgrade of U.S. Credit Rating Highlights Risks of Rising National Debt," May 19, 2025, https://www.pgpf.org/article/moodys-downgraded-its-us-credit-rating-and-warns-that-recent-policy-decisions-will-worsen-fiscal-outlook/. | |||||||||||||||||||||||
12 U.S.C. §5232. | ||||||||||||||||||||||||
The Troubled Asset Relief Program (TARP) was created as part of the Emergency Economic Stabilization Act of 2008 (P.L. 110-343). For more information, see CRS Report R41427, Troubled Asset Relief Program (TARP): Implementation and Status, by | ||||||||||||||||||||||||
16. | Baird Webel.
Michael Falkenheim and Mitchell Remy, Fannie Mae and Freddie Mac's Housing Goals, CBO, November 2024, https://www.cbo.gov/publication/60978. | |||||||||||||||||||||||
17. |
CBO, Fair Value Accounting for Federal Credit Programs, March 2012, p. 9, http://www.cbo.gov/sites/default/files/03-05-FairValue_Brief.pdf. |
|||||||||||||||||||||||
18. |
Office of Rep. Ralph Norman, "Rep. Norman Reintroduces Budget Transparency and Reform Package," press release, February 14, 2025, https://norman.house.gov/news/documentsingle.aspx?DocumentID=2040. | |||||||||||||||||||||||
The Financial Accounting Standards Board (FASB) developed Generally Accepted Accounting Principles (GAAP) for nongovernmental entities. The Federal Accounting Standards Advisory Board (FASAB) developed GAAP for federal government entities. | ||||||||||||||||||||||||
20. |
A concurrent resolutions is "an agreement between the House and Senate establishing various enforceable levels of spending and revenue. It is not submitted to the President for signature and therefore is not considered law." CRS Report R48284, The Congressional Budget Resolution: Frequently Asked Questions, by Tori Gorman. | |||||||||||||||||||||||
32.
|
| Deborah Lucas, "Introduction," in Measuring and Managing Federal Financial Risk (University of Chicago Press, 2010), p. 2. | ||||||||||||||||||||||
22. |
| |||||||||||||||||||||||
Richard Kogan | ||||||||||||||||||||||||
24. |
GAO, Credit Reform. | |||||||||||||||||||||||
25. |
Richard Kogan, Paul Van de Water, and James Horney, House Bill Would Artificially Inflate Cost of Federal Credit Programs, Center on Budget and Policy Priorities, June 18, 2013, at http://www.cbpp.org/research/house-bill-would-artificially-inflate-cost-of-federal-credit-programs. |
|||||||||||||||||||||||
26. | Fair value approach would make the subsidy cost of federal credit programs appear more costly for budget estimates than providing the same funds through a grant.
Kogan et al., "House Bill Would Artificially Inflate Cost." | |||||||||||||||||||||||
The Internal Revenue Service Department of Education, "Collections on Defaulted Loans," https://studentaid.gov/manage-loans/default/collections. Department of Education, "Student Loan Forgiveness," https://studentaid.gov/manage-loans/forgiveness-cancellation. Kiska, Estimates of the Cost of Federal Credit Programs in 2025. 2 U.S.C. §661(c). Board of Governors of the Federal Reserve System, Economic Well-Being of the U.S. Households in 2023, May 2024, https://www.federalreserve.gov/publications/2024-economic-well-being-of-us-households-in-2023-higher-education-student-loans.htm. Josh Bivens, "There Will Be Pain," Economic Policy Institute, February 13, 2025, https://www.epi.org/publication/tcja-extensions-2025/#epi-toc-4. The White House, "Restoring Public Service Loan Forgiveness," March 2025, https://www.whitehouse.gov/presidential-actions/2025/03/restoring-public-service-loan-forgiveness/. For information related some of the home loan programs, see CRS Report RS20530, FHA-Insured Home Loans: An Overview, by Katie Jones; and CRS Report R48253, Transportation, Housing and Urban Development, and Related Agencies (THUD) Appropriations for FY2025, coordinated by Maggie McCarty. For additional information on student loan programs, see CRS In Focus IF12828, Direct Loan Program Student Loans: Loan Types and Limits, by Alexandra Hegji; and CRS Report R43571, Federal Student Loan Forgiveness and Loan Repayment Programs, coordinated by Alexandra Hegji. CBO, Trouble Asset Relief Program, April 2017, https://www.cbo.gov/publication/24884. Avi Lerner and Zunara Naeem, Final Report on the Trouble Asset Relief Program, CBO, April 2024, p. 2, https://www.cbo.gov/system/files/2024-04/59919-TARP.pdf. | ||||||||||||||||||||||||
28. |
David Kamin, "Risky Returns: Accounting for Risk in the Federal Budget," Indiana Law Journal, vol. 88, no. 2 (2013). |
|||||||||||||||||||||||
29. |
Ibid. |
|||||||||||||||||||||||
30. | CBO, Report on the Troubled Asset Relief Program, March 2015, p. 2, https://www.cbo.gov/sites/default/files/114th-congress-2015-2016/reports/50034-TARP.pdf. | |||||||||||||||||||||||
CBO, A Preliminary Analysis of the President's Budget and an Update of CBO's Budget and Economic Outlook, March 2009, pp. 8-9, https://www.cbo.gov/sites/default/files/111th-congress-2009-2010/reports/03-20-presidentbudget.pdf. | ||||||||||||||||||||||||
32. |
Deborah Lucas, "Introduction" in Measuring and Managing Federal Financial Risk (University of Chicago Press, 2010), p. 4. |
|||||||||||||||||||||||
33. |
GASB, "Fair Value Measurement." FASB, "Fair Value Measurement (Topic 820)." Lucas, "Introduction," p. 4. | |||||||||||||||||||||||
34. |
| |||||||||||||||||||||||
35. | ," Brookings Institution, 2016, https://www.brookings.edu/wp-content/uploads/2016/03/lucastextspring16bpea.pdf.
| |||||||||||||||||||||||
36. |
| |||||||||||||||||||||||
37. | Kogan et al., "House Bill Would Artificially Inflate Cost." 2 U.S.C. §661a(5)(A). | |||||||||||||||||||||||
38. |
Net Present Value (NPV, net basis) measures the current value of all cash outflows and inflows at a discounted (interest) rate. It is a way to measure against competing investment choices over different time horizons. |
|||||||||||||||||||||||
39. |
2 U.S.C. §661a (5)(A). |
|||||||||||||||||||||||
40. | Congressional Budget Office, Estimating the Value of Subsidies for Federal Loans and Loan Guarantees, August 2004, p. 3, | |||||||||||||||||||||||
41. | Neill Perry and Puja Seam, "Accrual Accounting for Federal Credit Programs | |||||||||||||||||||||||
42. |
| |||||||||||||||||||||||
43. | .
The federal government borrows at the Treasury interest rates to finance its operations when revenue ( 2 U.S.C. §661. | |||||||||||||||||||||||
44. |
2 U.S.C. §661. |
|||||||||||||||||||||||
45. |
Direct loans are defined by FCRA as "a disbursement of funds by the Government to a non-Federal borrower under a contract that requires the repayment of such funds with or without interest." 2 U.S.C. §661a (1). |
|||||||||||||||||||||||
46. | Anthony Curcio and Roger Von Elm, "Fair Value in Federal Credit | |||||||||||||||||||||||
47. | ||||||||||||||||||||||||
48. | GASB, "Fair Value Measurement." FASB, "Fair Value Measurement (Topic 820)," https://asc.fasb.org/820/showallinonepage. FASB, "Fair Value Measurement (Topic 820)." FASB, "Glossary Term Usage," May 2025. | |||||||||||||||||||||||
49. |
| |||||||||||||||||||||||
50. |
Financial Accounting Standards Board, Fair Value Measurement (Topic 820), Norwalk, CT, May 2011, https://asc.fasb.org/imageRoot/00/7534500.pdf. |
|||||||||||||||||||||||
51. |
Anthony Curcio and Roger Von Elm, Fair Value in Federal Credit, Summit llc and Clifton Larson Allen, Washington, DC, April 2015, http://cdn2.hubspot.net/hubfs/355318/FCMF_Knowledge_Base/Federal_Credit_Fair_Value_White_Paper.pdf?t=1433429114841. |
|||||||||||||||||||||||
52. |
| |||||||||||||||||||||||
53. |
CCH - Commerce Clearing House - Accounting Research Manager, Overview and Scope of Topic 820, May 29, 2015, http://www.accountingresearchmanager.com. |
|||||||||||||||||||||||
54. |
Roger Von Elm and Anthony Curcio, What Agencies Need to Know About Fair Value Budgeting and How to Plan for Its Potential , Association of Government Accountants, April 13, 2015, http://www.agacgfm.org/AGA/Topics/2015/041315/fair_value_budgeting_13april_topics.pdf. |
|||||||||||||||||||||||
55. | CCH, "Fair Value Measurement—Measurement of Fair Value." | |||||||||||||||||||||||
56. |
| |||||||||||||||||||||||
57. | "Standards and Guidance," https://fasab.gov/accounting-standards/.
| |||||||||||||||||||||||
58. |
2 U.S.C. §661a(1). |
|||||||||||||||||||||||
59. |
2 U.S.C. §661a(3). |
|||||||||||||||||||||||
60. |
John Downes and Jordan Elliot Goodman, Dictionary of Finance and Investment Terms, 7th ed. (Hauppauge, NY: Barron's Educational Series, Inc., 2006), p. 456. |
2 U.S.C. §661a(1).
2 U.S.C. §661a(3).