Reduce, Refinance, and Rent? The Economic Incentives, Risks, and Ramifications of Housing Market Policy Options

Contents

• Introduction
• The Size of the Housing Problem
• Principal Reduction
• Risks of Principal Reduction
• Potential Impact of Principal Reduction
• Housing Market
• Consumer Spending
• Financial Sector
• Current Principal Reduction Policies
• Non-GSE Loans: The Home Affordable Modification Program's Principal Reduction Alternative and the 2012 Mortgage Settlement
• Fannie and Freddie's Position on Principal Reduction
• Legislative Proposals for Principal Reduction
• Large-Scale Refinancing
• Barriers to Refinance
• Potential Impact of Large-Scale Refinancing
• Housing Market
• Consumer Spending
• Financial Sector
• The Home Affordable Refinance Program
• Presidential and Legislative Proposals for Refinancing
• Renting Foreclosed Homes
• Barriers to Renting
• Potential Impact of Rental Program
• Housing Market
• Financial Sector
• Consumer Spending
• FHFA Pilot Program
• Legislative Proposals for Renting REO

Summary

The bursting of the housing bubble in 2006 precipitated the December 2007-June 2009 recession and a financial panic in September 2008. With the housing market seen as a locus for many of the economic problems that emerged, some Members of Congress propose intervening in the housing market as a means of improving not only the housing market itself but also the financial sector and the broader economy. Critics are concerned that further intervention could prolong the housing slump, delay recovery, and affect outcomes based on the government's preferences. Three frequently discussed proposals for the housing market are (1) reducing mortgage principal for borrowers who owe more than their homes are worth, (2) refinancing mortgages for borrowers shut out of traditional financing methods, and (3) renting out foreclosed homes.

Principal reductions have the potential to improve the housing market by minimizing disruptive defaults and foreclosures. However, by shifting the debt burden from the borrower to the lender, principal reduction may negatively impact financial institutions that would have their investments' principal balances reduced. Principal reduction, nonetheless, might improve the broader economy if it stimulates consumer spending, diverting income from debt repayment to spending on other goods and services.

Legislation introduced in the 112^th Congress to reduce mortgage principal includes H.R. 1587, H.R. 3841, H.R. 4058, and S. 2093. The Federal Housing Finance Agency (FHFA), the regulator and conservator of Fannie Mae and Freddie Mac, decided against allowing the enterprises to reduce principal for mortgages that they guarantee.

Large-scale refinancing helps borrowers who are current on mortgage payments to refinance into a new mortgage with a lower interest rate. Because refinancing generally helps borrowers who are current, it is unlikely to have a major effect on the housing market, but it may prevent some foreclosures that could occur in the absence of a refinance. In addition, refinancing has the potential to have a larger effect on the economy by stimulating consumer spending. A mortgage refinance could lower a borrower's monthly payment, freeing up more income for non-housing-related spending. Some of the additional spending of borrowers may come at the cost of the financial sector. Although some financial institutions may lose investment income from refinancing, others could benefit from the increased business associated with refinancing.

President Obama, in his 2012 State of the Union address, proposed streamlining the existing program to refinance Fannie Mae and Freddie Mac loans and establishing a new mass refinancing plan for non-Fannie Mae and non-Freddie Mac loans. Congressional proposals for large-scale refinancing of Fannie Mae and Freddie Mac loans include H.R. 363, S. 170, and S. 3085.

Renting out foreclosed homes currently held by banks and other financial institutions has the potential to stabilize housing prices by reducing the supply of homes on the "for sale" market. However, this policy depends on house prices increasing in the future such that, when the rented properties are eventually sold, they are sold in a healthier market. Unlike principal reductions and mass refinancing, renting foreclosed homes does not reduce existing homeowners' payments or increase their disposable income. Any impact on consumer spending is likely to be indirect through stabilizing house prices and preserving neighboring homeowners' equity.

FHFA implemented a pilot project to convert foreclosed homes into rentals. Legislative proposals to expand the renting of foreclosed properties include H.R. 1548, H.R. 2636, and S. 2080.

---

Reduce, Refinance, and Rent? The Economic Incentives, Risks, and Ramifications of Housing Market Policy Options

Introduction

The bursting of the U.S. housing bubble in 2006 precipitated the December 2007-June 2009 recession and a financial panic in September 2008. Falling house prices contributed to rising foreclosure rates and lower consumer spending. In addition, mortgage defaults and the rise in foreclosures hurt financial institutions that owned the loans on the foreclosed homes, triggering a much broader collapse of the financial system. As credit flows slowed, so did economic growth.

Because the housing market was a locus for many of the economic problems that emerged, some Members of Congress and other experts propose intervening in the housing market not only as a means of improving the housing market itself but also the financial sector and the broader economy. Supporters of housing market intervention argue that housing market weakness has a strong indirect negative effect on the balance sheets of households and banks, which dampens the recovery of the wider economy.¹ Skeptics, however, worry that further intervention could prolong the housing slump, delay economic recovery, and affect outcomes based on the government's preferences.

Since the housing bubble burst, the federal government has created multiple programs to aid homeowners, but their impact has been less than anticipated. Many measures have been proposed either to modify existing programs or to establish new efforts to improve the housing market. Three frequently discussed approaches are (1) reducing mortgage principal for borrowers who owe more than their homes are worth, (2) refinancing mortgages for borrowers who find themselves locked into paying high interest rates, and (3) renting out foreclosed homes.

Principal reductions aim to improve the housing market by minimizing defaults and foreclosures, thus reducing collateral damage to the economy. However, principal reduction shifts the losses of borrowers to banks and other lenders. At the same time, principal reduction might improve the wider economy if it stimulates consumer spending, allowing borrowers to divert income from debt repayment to spending on other goods and services.

Members of the 112^th Congress have introduced multiple bills that would reduce mortgage principal on certain loans. These include H.R. 1587, the Home Foreclosure Reduction Act of 2011 (Representative John Conyers et al.); H.R. 3841, the Principal Reduction Act of 2012 (Representative Maxine Waters et al.); H.R. 4058, the Bankruptcy Equity Act of 2012 (Representative Earl Blumenauer et al.); and S. 2093, the Preserving American Homeownership Act of 2012 (Senator Robert Menendez). Principal reduction is also a component of a settlement reached between several mortgage servicers and 49 state attorneys general and the federal government. The Federal Housing Finance Agency (FHFA), the regulator and conservator of Fannie Mae and Freddie Mac, decided against allowing the enterprises to reduce principal for mortgages that they guarantee. All of these initiatives are discussed more fully in "Principal Reduction" below.

A second approach, large-scale refinancing, helps borrowers who are current on their mortgage to refinance into a new mortgage with a lower interest rate. Because refinancing generally helps borrowers who are current, it is unlikely to have a major effect on the housing market, but it may prevent some foreclosures that could occur in the absence of a refinance. Some argue that refinancing would stimulate consumer spending. For example, a mortgage refinance could lower a borrower's monthly payment, freeing up more income that can be used to pay down debt or for non-housing-related spending. Some of the additional spending of borrowers may be offset by reduced income for investors. In a refinance, the previous owner of the mortgage is repaid on the loan earlier than expected and now faces reinvesting in a lower interest rate environment. Although some investors (including banks) may lose investment income from refinancing, other banks and mortgage originators benefit from the increased business associated with refinancing. Similarly, mass refinancing might cause Fannie Mae and Freddie Mac, two housing government-sponsored enterprises (GSEs), to lose investment income but gain from the reduced likelihood of default. It is unclear which effect would be larger.

In his 2012 State of the Union address, President Obama proposed streamlining the existing program to refinance Fannie Mae and Freddie Mac loans and establishing a new mass refinancing plan for non-Fannie Mae and non-Freddie Mac loans. Congressional proposals for large-scale refinancing in the 112^th Congress include H.R. 363, the Housing Opportunity and Mortgage Equity Act of 2011 (Representative Dennis Cardoza et al.); S. 170, the Helping Responsible Homeowners Act (Senator Barbara Boxer et al.); and S. 3085, the Responsible Homeowner Refinancing Act (Senator Robert Menendez et al.). These proposals are discussed more fully in "Large-Scale Refinancing" below.

A third proposal, renting out foreclosed homes currently held by banks, GSEs, and other financial institutions, has the potential to stabilize housing prices by reducing the supply of homes on the "for sale" market. Current policy encourages lenders to sell foreclosed property quickly rather than rent out vacant homes. A successful rental program depends on house prices increasing in the future such that the rented properties can eventually be sold in a healthier market. In addition, renting foreclosed homes keeps the properties in use and reduces the collateral damage to communities. Unlike principal reductions and mass refinancing, renting foreclosed homes does not reduce existing homeowners' payments or increase their disposable income. Any impact on consumer spending is likely to be indirect through favorable effects on household net worth due to stabilizing house prices and preserving neighboring homeowners' equity. The program would only prevent foreclosure if, in lieu of foreclosure, it allowed delinquent homeowners to rent out their own homes.

FHFA has started a pilot project to convert GSE foreclosed homes into rentals. Congressional proposals in the 112^th Congress to allow for renting foreclosed properties include H.R. 1548, the Right to Rent Act of 2011 (Representative Raúl Grijalva et al.); H.R. 2636, the Neighborhood Preservation Act of 2011 (Representative Gary Miller et al.); and S. 2080, the Keeping Families in their Home Act of 2012 (Senator Dean Heller). These initiatives are discussed more fully in "Renting Foreclosed Homes" below.

Table 1 summarizes the potential impact of these policies on three categories of households: delinquent borrowers, borrowers who are paying on time, and renters. Table 2 summarizes the likely impacts on house prices, financial institutions' balance sheets, and the broader economy.

The remainder of this report will further analyze principal reduction, refinance, and rental proposals. First, however, the next section provides one estimate of the size of the housing market problem.

The Size of the Housing Problem

In her September 2011 congressional testimony, Laurie Goodman, a senior managing director at Amherst Securities Group, estimated that 8.3 million to 10.4 million homeowners, approximately 15%-19% of all homeowners with a mortgage, are at risk of losing their homes in the next six years.² Goodman's estimates are shown in Table 3. Of the approximately 80 million homes in the United States, nearly 55 million have a mortgage. Goodman divides the mortgages into five categories and assigns an estimated default rate for each category to calculate the number of homes in jeopardy of transitioning to default (i.e., being more than 60 days delinquent on the mortgage).

The five categories, in order of highest to lowest default rate, are

• (1) loans that are currently non-performing because the borrowers are no longer making their monthly payments;
• (2) loans that were non-performing but have become re-performing because the borrowers have resumed payment;
• (3) loans that have always been performing but have severe negative equity, meaning the borrower owes more on their mortgage than the home is worth;
• (4) loans that are always performing and the borrowers have moderate negative equity; and
• (5) loans that are always performing and the borrowers have positive equity.

Goodman includes a "lower bound" and a "reasonable" estimated default rate for each of the loan categories based on current trends in the foreclosure rate and the pace of economic recovery.³

Goodman's analysis suggests two broad types of policy approaches for improving the housing market. First, policies could attempt to keep people in their homes by lowering the default rates. Principal reduction and large-scale refinancing are two frequently discussed proposals attempting to accomplish this goal. Principal reduction generally involves lowering the amount of the mortgage that is owed for borrowers who are delinquent, have negative equity, or are both delinquent and have negative equity. Large-scale refinance proposals generally target borrowers who are current on their loans but cannot refinance because of their negative equity. Both proposals attempt to lower monthly payments for borrowers as well as reduce default rates by stimulating the economy through reducing or redistributing the debt burden of the mortgage.

The second approach implied by Goodman's analysis is to ease the transition of foreclosed homes into the market. Rather than flooding the market with foreclosed homes and potentially further lowering house prices, some have proposed converting the foreclosed homes into rental properties. Different proposals call for financial institutions either to rent the properties directly themselves or to sell properties in bulk to investors who would rent the vacant homes.

The remainder of this report analyzes each of these three major policy proposals.

Principal Reduction

As a result of falling house prices, an estimated 11 million-15 million homeowners owe more on their mortgage than their home is worth.⁴ Borrowers in this situation are said to be "underwater" on their homes or in "negative equity." Negative equity may impair a borrower's labor mobility, making it more difficult to sell the house and move for a new job, or limit the ability to use the house as collateral to take out a loan for a small business.⁵ Borrowers with negative equity are also more likely to default and enter foreclosure than borrowers with positive equity, partly because borrowers with negative equity are often unable to sell their home for enough to cover the amount owed if they are unable to make their monthly payments.⁶ Foreclosures can drive down home prices, forcing some borrowers further underwater and continuing the cycle of foreclosure. Figure 1 shows the rise and fall of a measure of home prices since 1987. To address these concerns, some Members of Congress have proposed reducing the mortgage principal for underwater borrowers.

Figure 1. S&P/Case-Shiller Home Price Index, 1987 to 2011

Source: S&P/Case-Shiller Home Price U.S. National Index, Seasonally Adjusted. Notes: Index sets January 2000 equal to 100.

The amount that a borrower is underwater is often measured by the loan-to-value (LTV) ratio. The LTV ratio is the ratio of the amount that a borrower owes on the mortgage to the value of the house. An LTV of 100% means that a borrower owes the same amount as the value of the house. An LTV greater than 100% means that the borrower has negative equity, owing more than the house is worth. The more the LTV is above 100%, the more underwater the borrower is. Figure 2 shows the distribution of borrowers by the equity they have in their homes.

Figure 2. Distribution of Homeowners by Home Equity (as of First Quarter of 2012)

Source: CoreLogic.

Homeowners with a mortgage can generally be divided into three groups: (1) those with a loan that is government-insured through the Federal Housing Administration (FHA), U.S. Department of Veterans Affairs, or other government agency; (2) those with a mortgage owned or guaranteed by Fannie Mae or Freddie Mac (the GSEs); and (3) those with a privately owned mortgage. Most of the proposals discussed in this report focus on GSE loans and privately owned (non-GSE) mortgages.

A borrower with a privately owned mortgage could see principal reduced independent of any government assistance program if the mortgage holder (i.e., bank, credit union, or other investor) believed it is in its best interest. Non-GSE borrowers could also see their principal reduced through the federal government's Home Affordable Modification Program's (HAMP's) Principal Reduction Alternative (PRA). Borrowers with a GSE mortgage, on the other hand, are ineligible for principal reductions because of policies set by Fannie Mae's and Freddie Mac's regulator, the Federal Housing Finance Agency (FHFA).

Several bills have been introduced in the 112^th Congress to reduce mortgage principal on some mortgages, including H.R. 1587, the Home Foreclosure Reduction Act of 2011 (Representative John Conyers et al.); H.R. 3841, the Principal Reduction Act of 2012 (Representative Maxine Waters et al.); H.R. 4058, the Bankruptcy Equity Act of 2012 (Representative Earl Blumenauer et al.); and S. 2093, the Preserving American Homeownership Act of 2012 (Senator Robert Menendez). Principal reduction is also a component of the settlement reached between several mortgage servicers and 49 state attorneys general and the federal government. The specifics of these proposals are discussed in "Legislative Proposals for Principal Reduction" below. The next section discusses some of the factors that influence whether a financial institution would offer a borrower a principal reduction.

Risks of Principal Reduction

Some argue that principal reductions are almost always in the best interests of a borrower and a financial institution if a borrower is near foreclosure. For example, a borrower may take out a $180,000 mortgage to buy a $200,000 house (giving the borrower a loan-to-value ratio of 90%). If the borrower is having trouble making the monthly mortgage payment and the value of the house falls to $150,000, the borrower could face foreclosure. If the value of the house was greater than the amount owed, potentially the borrower could sell the home to pay off the mortgage or borrower against the equity in the house to make the monthly payment; however, because the borrower has negative equity, selling the house may not cover the mortgage.⁷ If the bank forecloses on the borrower and sells the house, it may only recover a fraction of what the house is worth, for example, $100,000.⁸ Advocates of principal reduction state that it is in the best interest of the financial institution to reduce principal by less than the amount they would lose in foreclosure.

Multiple factors not captured by the above example may explain why principal reductions are not offered more frequently.⁹ When a borrower is delinquent, he or she is likely to fall into one of four categories. The borrower

• could become current on his mortgage without any assistance (Type 1);
• could default on the loan even if a loan modification is offered (Type 2);
• could become current on the loan but only if a modification is received (Type 3); and
• does not receive a modification and ends up in foreclosure (Type 4).

When considering whether to offer a delinquent borrower a modification, the mortgage holder would prefer to only target the Type 3 borrower. No action by the mortgage holder changes the outcome for the first two types and is therefore not worth the expense associated with attempting to help.¹⁰ The challenge for the holder is in determining which of the four types of borrower any given delinquent borrower is. All borrowers have the incentive to claim to be Type 3. When offering a modification, the mortgage servicer (the financial institution that is in contact with the borrower and is tasked with making decisions about loan modifications) is therefore exposed to two different types of risk: self-cure risk and re-default risk. Self-cure risk is the risk that the servicer will offer a modification to the Type 1 borrower who would have become current without any assistance. Re-default risk is the risk that the servicer modifies a loan for the Type 2 borrower who re-defaults in spite of the modification. In addition, if it becomes known that a servicer is willing to offer a significant loan modification, then current borrowers have an incentive to default in order to become eligible for the modification. This then creates a third type of risk for the servicer, strategic default risk,¹¹ which is the incentive created for borrowers to become delinquent even though they are current on their mortgage and can afford to pay. Principal reduction could, some argue, increase strategic default risk and create moral hazard.¹²

If the objective of a financial institution is to maximize profits, then it will only offer principal reductions if it is the option that best maximizes profits. For a principal reduction to maximize profits, the expected amount gained by helping the Type 3 borrower who would have failed without the assistance must be greater than the losses associated with helping those who may self-cure, re-default, or strategically default.

Self-cure risk, re-default risk, and strategic default risk are present for all types of loan modifications, including changes to the interest rate and the balance of the loan. All borrowers, regardless of whether they "need" the modification, would benefit from decreased monthly mortgage payments.

Other loss-mitigation strategies besides principal reduction may more effectively minimize the risks to the mortgage holder. For example, principal forbearance is the removal of a portion of the principal from the amount used to calculate monthly principal and interest payments for a limited period, but the forborne amount is due at the end of the loan term, potentially with interest. Borrowers who do not need the modification may find forbearance less appealing because it changes the timing on their payments but may not necessarily reduce the amount that they must ultimately pay. Alternatively, principal reduction can be paired with other approaches to minimize risk. For example, combining principal reduction with a shared appreciation mortgage, in which the borrower agrees to share some of the future gains when the house is sold with the lender, could minimize strategic default risk. These other loss-mitigation strategies are also discussed later in this report.

Potential Impact of Principal Reduction

Housing Market

As mentioned previously, borrowers with negative equity are more likely to default and enter foreclosure than borrowers with positive equity. This can contribute to a negative spiral in which falling home prices lead to negative equity, which causes more foreclosures and a further fall in house prices. Figure 3 shows the monthly delinquency and foreclosure rates since 2001. Principal reduction could potentially interrupt this cycle by either reducing the amount of negative equity or increasing the amount of positive equity, depending on the amount of principal that is reduced and the types of borrowers a principal reduction program targets.

Figure 3. Total Delinquent and Foreclosure Percentage, by Month

Source: Figure created by CRS using data from the Mortgage Bankers Association. Notes: The Mortgage Bankers Association (MBA) is one of several organizations that reports delinquency and foreclosure data, but it does not represent all mortgages. MBA estimates that its data cover about 80% of outstanding first-lien mortgages on single-family properties. The delinquency rate includes borrowers who are more than 90 days delinquent as well as homes in the foreclosure inventory.

Principal reduction could allow borrowers experiencing "house lock," those unable to sell their home for enough to cover the balance of the mortgage, to sell their home if a principal reduction returns them to positive equity.¹³ If the goal is to stabilize the housing market, then allowing more borrowers to sell may be counterproductive. In addition, some worry that principal reductions would be a windfall for those borrowers who sell immediately and would impose unfair losses on lenders. To address these concerns, some propose reducing principal incrementally over a period of years or requiring homeowners to share part of the house price appreciation with the lender when it is sold.

Some opponents of principal reduction argue that, because negative equity is concentrated in a few states, principal reduction would therefore benefit some parts of the country at the expense of others. Figure 4 shows the percentage of homeowners with negative equity by state. In that respect, a principal reduction policy may be similar to aid for natural disasters, which flows disproportionately to some geographic regions of the country.

Figure 4. Percentage of Homeowners with Negative Equity, by State (as of First Quarter of 2012)

Source: CoreLogic. Notes: Data are unavailable for South Dakota and Vermont. Near negative equity includes borrowers with less than 5% of positive home equity.

Consumer Spending

By stabilizing the housing market, principal reduction could stimulate the economy by reducing borrowers' monthly mortgage payments, allowing them to spend more on other goods and services. However, the additional money available to consumers comes at the expense of the owners of the mortgages; involuntary principal reduction reduces the amount that the mortgage holder will receive. Although borrowers gain and investors may lose in a principal reduction, the amount is not necessarily zero-sum.¹⁴ Given the potential additional spending by borrowers and decreased spending by investors, a net spending increase could occur through multiple channels.¹⁵ First, if borrowers spend more of an additional dollar than investors would have spent of the same dollar, then principal reduction would increase spending. Second, a significant percentage of mortgage-backed securities (MBS) are held by foreign investors and the federal government. Reducing the investment income to those holders of MBS is unlikely to have a large negative impact on consumer spending in the United States.¹⁶ Third, by reducing the number of foreclosures that would have occurred and stabilizing home prices, principal reduction would help the wider economy by preserving household wealth and building consumer confidence.¹⁷ Research has found a positive correlation between household wealth and consumer spending.¹⁸

Financial Sector

Assuming that the mortgage holders will reduce principal voluntarily when it is in their best interest, then a program requiring involuntary principal reductions would not be in the best interest of the mortgage holder. It would limit their choice set and potentially restrict their ability to maximize returns. If the mortgage holder is a bank or other lender, then the revenue that is lost by forgoing foreclosure and instead facing a less profitable option could cause the lender to curtail lending. A widespread curtailment of lending could cause interest rates to rise. However, the Federal Reserve could potentially offset a possible credit contraction through open market operations or an expansion of its balance sheet.¹⁹

Although principal reductions could reduce their revenues, mortgage holders could, nonetheless, potentially benefit from a principal reduction program through two channels. First, some experts argue that mortgage servicers²⁰ face a host of competing incentives, not all of which encourage the servicer to act in the best interest of the loan holder. These incentives, as well as shortcomings in a servicer's operational infrastructure, could encourage the servicer to pursue foreclosure when the investor would be best served by a loan modification, such as principal reduction.²¹ Mortgage holders could benefit from a principal reduction program if servicers' misaligned incentives prevented the optimal number of principal reductions from occurring.

Second, mortgage holders could suffer from coordination problems.²² Foreclosure may be individually rational but collectively irrational. If mortgage holders think house prices will continue to fall, it may be in each individual mortgage holder's interest to pursue foreclosure and maximize its immediate return, but collectively—when all pursue foreclosure at the same time—each is made worse off. A large surge of foreclosures could further drive down house prices and reduce the mortgage holder's recovery when the property is eventually sold. Mandated principal reductions could alleviate the coordination problems by requiring all servicers to consider principal reduction prior to acting on foreclosure.

Current Principal Reduction Policies

Supporters of principal reduction argue that it has the potential to stabilize the housing market, support the recovery of lost household net worth, and stimulate the economy. Opponents of principal reduction, however, highlight its negative impact on financial institutions at a time when lending is still weak, and they raise issues of fairness associated with reducing the principal for some borrowers but not others.²³

This section of the report analyzes principal reduction by non-GSE financial institutions through the government's Home Affordable Modification Program (HAMP) and analyzes the GSE policy on principal reductions. As Figure 5 shows, the GSEs do not perform principal reductions, but non-GSEs—banks holding loans in their portfolio and private investors—do perform principal reductions (though, it was uncommon for private investors until 2011).²⁴ However, Figure 6 shows that GSEs and non-GSEs use principal forbearance with similar frequency.

Figure 5. Percentage of Loan Modifications Involving Principal Reduction

Source: OCC and OTS Mortgage Metrics Report.

Figure 6. Percentage of Loan Modifications Involving Forbearance

Source: OCC and OTS Mortgage Metrics Report.

Non-GSE Loans: The Home Affordable Modification Program's Principal Reduction Alternative and the 2012 Mortgage Settlement

Non-GSE mortgage holders, such as banks, credit unions, thrifts, and other private investors, own approximately half of all the outstanding mortgage debt in the United States.²⁵ These mortgage holders may modify the terms of the mortgage (by reducing the interest rate or the principal, for example) independent of any government program if it is in the best interest of the borrower and the mortgage holder. This section analyzes one of the federal government's major initiatives to encourage non-GSE institutions to perform principal reductions, the Home Affordable Modification Program (HAMP).²⁶ Through HAMP's Principal Reduction Alternative (PRA), the federal government offers financial incentives to investors in non-Fannie and non-Freddie loans to reduce mortgage principal.

HAMP is part of President Obama's Making Home Affordable (MHA) program, which he announced on February 18, 2009. HAMP provides financial incentives to mortgage servicers to lower eligible troubled borrowers' monthly mortgage payments to more sustainable levels.²⁷ On March 26, 2010, HAMP announced the PRA.²⁸ Under the PRA, participating servicers are required to consider reducing principal balances as part of HAMP modifications for homeowners who owe at least 115% of the value of their home. Servicers run two net present value (NPV) tests²⁹ for these borrowers: the first is the standard NPV test, and the second includes principal reduction. If the NPV of the modification is higher under the test that includes principal reduction, servicers have the option to reduce principal. However, they are not required to do so. If the principal is reduced, the amount of the principal reduction will initially be treated as principal forbearance; the forborne amount will then be forgiven in three equal amounts over a three-year period as long as the borrower remains current on his or her mortgage payments. The Administration offers financial incentives to servicers specifically for reducing principal. The PRA went into effect on October 1, 2010.³⁰ According to the Treasury Department, more than 60,000 PRA modifications were active as of January 2012. About 16,000 of these are active trial modifications, and about 44,000 are active permanent modifications.³¹

To encourage more principal reductions, the Treasury Department announced on January 27, 2012, that it would triple the incentive payments to investors.³² Investors will now receive 18 cents to 63 cents for each dollar of principal reduced, depending on the borrower's loan-to-value ratio. As discussed in the next section, Treasury has offered to expand the PRA to include Fannie Mae and Freddie Mac loans by extending the incentive payments to them, but FHFA, the regulator and conservator of Fannie and Freddie, has decided against allowing the GSEs to participate.

Some homeowners with non-Fannie and non-Freddie loans could also have their principal reduced as a result of the mortgage settlement between five banks and 49 state attorneys general and the federal government.³³ The settlement is expected to result in approximately $25 billion in monetary sanctions and relief.³⁴ Of the $25 billion, $17 billion is expected to be allocated to aiding homeowners who want to stay in their homes but cannot afford to at their current payment levels. Of the $17 billion, at least 60% must be allocated to principal reductions for borrowers who are in default or at risk of default.³⁵

Fannie and Freddie's Position on Principal Reduction

Fannie Mae and Freddie Mac do not offer principal reductions on the loans they own or guarantee, which is approximately half of the $10.3 trillion U.S. mortgage market.³⁶ The decision not to offer principal reductions was made by Fannie Mae and Freddie Mac in consultation with FHFA.³⁷ Fannie Mae and Freddie Mac loans are eligible for modification under HAMP, but do not participate in HAMP's PRA component.³⁸ GSE loans are also not covered for principal reduction under the terms of the mortgage settlement.

Congress established FHFA in 2008 in the Housing and Economic Recovery Act of 2008 (HERA; P.L. 110-289). The acting director of FHFA, Edward DeMarco, has interpreted FHFA's mandate to have three components:

First, FHFA has a statutory responsibility as conservator to preserve and conserve the assets and property of the regulated entities. Second, the Enterprises have the same mission and obligations as they did prior to the conservatorship. Therefore, FHFA must ensure that Fannie Mae and Freddie Mac maintain liquidity in the housing market during this time of economic turbulence. Third, under the Emergency Economic Stabilization Act of 2008 (EESA), FHFA has a statutory responsibility to maximize assistance for homeowners to minimize foreclosures.³⁹

In January 2012, FHFA analyzed the potential impact that principal reductions could have and decided that principal reductions are not the strategy that best fulfills the agency's mandates.⁴⁰ Shortly after FHFA's announcement, the Treasury Department offered to extend the incentive payments that are part of HAMP PRA to the GSEs.⁴¹ FHFA analyzed the Treasury Department's offer and, in July 2012, determined that the GSEs would not participate in HAMP PRA. The following sections describe the methodology for and the results of FHFA's analysis.

Methodology

Documents released by the FHFA detail attempts to determine whether implementing the HAMP PRA would result in greater savings to the GSEs and to the taxpayers than the existing foreclosure prevention options. FHFA's analysis compares the benefits of principal reduction (possibly fewer foreclosures) to the costs of offering principal reduction (Treasury's incentive payments to Fannie Mae and Freddie Mac, implementation costs, and other potential indirect costs).

FHFA's analysis begins with a general comparison of principal reduction to forbearance, a frequently cited alternative to principal reduction. FHFA then refines the analysis by adding additional assumptions and limiting criteria to create a model that can be used to simulate how the principal reduction program might actually operate.

FHFA calculated as a baseline how much the GSEs might expect to lose if no modifications were offered at all to borrowers eligible for HAMP PRA. It then used the model to estimate how much the GSEs could expect to save if they offered eligible borrowers a standard HAMP modification (which does not include principal reduction) or a modification with the possibility of principal reduction. FHFA ran two NPV tests for borrowers that would be eligible for HAMP PRA, one with principal reduction and one without. For the test including principal reduction, FHFA included Treasury's principal reduction incentive payments⁴² to the GSEs as part of the benefit. Using the results of the NPV tests, it calculated an "optimal" modification for each borrower, defined as the modification that results in the highest expected NPV. FHFA then compared the savings of the optimal modification in which principal reduction was not an option to the optimal modification in which principal reduction was an option. The difference between the two is the estimated benefit to the GSEs of offering principal reduction. To calculate the benefit to the taxpayer of offering principal reduction, FHFA subtracted out the Treasury incentive payments.

FHFA previously analyzed HAMP PRA in January 2012. The methodology used in FHFA's previous analysis⁴³ has been criticized. For example, Laurie Goodman, a Senior Managing Director at Amherst Securities, identified three criticisms of the original methodology in her testimony before the Senate Banking Subcommittee on Housing, Transportation and Community Development.⁴⁴ At least some of these criticisms appear to have been addressed in the July 2012 analysis.

First, Goodman noted that FHFA used the NPV model, a hypothetical model, in its January analysis and suggested that actual HAMP data had not been used to the fullest extent possible.⁴⁵ FHFA's July 2012 analysis, however, continues to rely on the NPV model. If the GSEs implemented the HAMP PRA, the NPV model would likely be used as part of the program. FHFA therefore decided to use the NPV model in its analysis. The NPV model is "updated over time as more information becomes available" about the performance of the modifications, especially regarding re-default rates.⁴⁶ Because HAMP results are used to calibrate the NPV model and FHFA used the NPV model in its analysis, HAMP results were used indirectly by FHFA in its analysis. FHFA used HAMP results directly in other instances, such as using its observed transition rate of current borrowers into delinquent borrowers.⁴⁷

Second, Goodman cited four technical issues as flaws in the January analysis.⁴⁸

• FHFA used state level house price indices in its January analysis, obscuring the potential variation in house prices within a state. In the July analysis, FHFA used ZIP code level indices.⁴⁹ When using ZIP code level indices, FHFA found a larger number of homeowners to be underwater, increasing the estimated number of households potentially eligible for HAMP PRA.
• FHFA's January analysis applied principal reduction to the entire portfolio of the GSEs rather than to individual loans that would be eligible for HAMP PRA. To address this concern, the July analysis attempted to determine the optimal modification for each individual loan that would be eligible for HAMP PRA.
• Goodman noted that the January analysis did not evaluate the actual HAMP program structure but a variant of it.⁵⁰ The July analysis attempted to analyze the actual HAMP structure, but because FHFA was using a model, it made some simplifying assumptions. Those assumptions are described below in the "Observations" section.
• The January analysis used the loan attributes at origination rather than the current attributes of the loan.⁵¹ The July analysis attempts to model the current attributes for eligible borrowers by lowering the credit score of borrowers by 100 points since origination and using higher debt-to-income ratios.

Third, Goodman testified that FHFA "did not consider any differentiation between loans with mortgage insurance versus loans without it," causing principal reduction to look less beneficial.⁵² Despite this criticism, FHFA's July analysis did not separately report results for loans with mortgage insurance and loans without mortgage insurance. FHFA did account for the existence of mortgage insurance on some loans because "the Treasury NPV model assumes private mortgage insurance claims are paid in full in the event of default or re-default."⁵³ Goodman argues that FHFA could design a principal forgiveness program in which only borrowers without mortgage insurance would be eligible and that the program would be beneficial to the GSEs; "there is no question in my mind that forgiveness could be implemented for part of their book of business, without implementing it on the entire book of business."⁵⁴ FHFA did not analyze a program as envisioned by Goodman.

Results

As mentioned above, FHFA performed multiple analyses with each analysis building on the previous one. By adding additional assumptions and limiting criteria, FHFA's final results were presented as describing the reality of how the programs would actually operate. FHFA numbered their different analyses 1 through 12c. This section focuses on FHFA's final six analyses, particularly numbers 11 through 12c. Table 4 shows the results of FHFA's analyses 11 through 12c.

For each analysis, Table 4 lists the assumptions used in FHFA's modeling as well as the number of loans eligible given their assumptions and the total unpaid principal balance of the eligible loans. FHFA estimated the expected losses if the eligible loans did not receive a modification. FHFA then calculated the reduction in losses if loans that were NPV positive received a Standard HAMP modification (if the loans were NPV negative, they did not receive modifications). Similarly, FHFA estimated the expected savings if eligible borrowers received their "Optimal HAMP Modification," which includes principal reduction as an option. The difference between the expected savings resulting from the Optimal HAMP Modification and the Standard HAMP is the benefit to the GSEs of offering principal reductions. FHFA calculated the expected Treasury subsidy that would be paid to the GSEs as part of the principal reduction program. To calculate the Taxpayer Benefit, FHFA subtracted the Treasury Subsidy from the Enterprise Benefit.

Analysis 11 presents FHFA's results assuming all HAMP eligible borrowers will enter into a HAMP modification.⁵⁵ Under Analysis 11, nearly 500,000 borrowers would be eligible, resulting in averted losses of approximately $3.6 billion to the GSEs and $1.0 billion to the taxpayers. However, FHFA finds in its analysis⁵⁶ that the majority of the savings—$852 million of the $952 million—would derive from borrowers who are more than a year delinquent. Of the $852 million, $606 million would come from severely underwater borrowers with LTV ratios above 140%. FHFA argues that, because past experience with HAMP indicates that the program is most effective for borrowers who are only a few months delinquent, the model's projected savings may be an overestimate. Analysis 11a removes borrowers who are more than 12 months delinquent and 11b removes borrowers more than 6 months delinquent; doing so eliminates most of the benefits found in Analysis 11.

In addition, Analysis 11 assumes that 100% of eligible borrowers would participate. FHFA notes that recent experience shows a 15% - 25% take-up rate for HAMP eligible borrowers, but FHFA hopes to improve servicer performance to achieve a take-up rate of 50%. Analyses 12a, 12b, and 12c scale Analysis 11 by 50%, 25%, and 15%, respectively, to account for lower levels of participation. While the benefits to the GSEs and taxpayers are still positive, the benefits are correspondingly lower than in Analysis 11. Notably, FHFA finds that for Analyses 12a - 12c, 85% of the optimal HAMP modifications would be principal reductions.

The results in Table 4 are not the final FHFA results. The table only accounts for the direct benefits and the direct costs of offering principal reduction. FHFA also attempted to account for indirect costs. One of the largest potential indirect costs involves so-called strategic modifiers, which FHFA defines as "current borrowers who either claim financial hardship or who miss two consecutive mortgage payments to attempt to qualify for HAMP PRA, which includes principal forgiveness," but are able to make their payments.⁵⁷ FHFA stated that because it is difficult to accurately predict the number of strategic modifiers, it decided to calculate the number of borrowers who would need to become strategic modifiers to eliminate the benefit to the GSEs and taxpayers, rather than try to estimate the expected number of strategic modifiers. FHFA found that, using the benefits in Analyses 12a - 12c, it would take 14,000 to 126,000 strategic modifiers to eliminate the GSEs' benefits of principal reduction and 3,000 to 19,000 strategic modifiers to eliminate the taxpayer benefits. In addition to strategic modifiers, FHFA listed other indirect costs of implementing a principal reduction program, including the additional operational costs, the opportunity cost of staff resources, and the long-term implications for mortgage credit pricing (if principal reduction causes investors to believe that future mortgage contracts are not as secure, they may require higher rates and reduce credit availability).

After accounting for the direct benefits, direct costs, and indirect costs of allowing for principal reduction, FHFA concluded that principal reduction would increase GSE losses and taxpayer losses. FHFA, therefore, decided against allowing the GSEs to participate in HAMP PRA.⁵⁸

Legislative Proposals for Principal Reduction

A principal reduction policy could target all borrowers, including those underwater, those underwater and delinquent, or some other subset, such as just GSE borrowers. Eligibility in a principal reduction proposal is likely to be determined by the underlying policy goal. Proposals to improve the housing market may be most effective by targeting delinquent borrowers who could afford a lower monthly payment though they cannot afford their current payment. On the other hand, proposals to stimulate consumer spending by redistributing debt burdens may be more effective if eligibility for principal reduction is broader.

Members in the 112^th Congress have introduced legislation to reduce mortgage principal. The bills address different subsets of homeowners as well as use different methods for reducing principal. H.R. 1587, the Home Foreclosure Reduction Act of 2011 (Representative John Conyers et al.) and H.R. 4058, the Bankruptcy Equity Act of 2012 (Representative Earl Blumenauer et al.), would allow for the cramdown of mortgage debt for principal residences in bankruptcy.⁵⁹ Some believe that cramdown in bankruptcy would strengthen the negotiating position of borrowers prior to bankruptcy and would incentivize mortgage holders to modify more loans.

H.R. 3841, the Principal Reduction Act of 2012 (Representative Maxine Waters et al.), would require Fannie Mae and Freddie Mac to establish a principal reduction program for qualified mortgages. A qualified mortgage is a mortgage, regardless of whether the borrower is current or delinquent, that

• is owned or guaranteed by Fannie Mae or Freddie Mac;
• is the first mortgage on a single-family home that is the borrower's primary residence;
• was originated on or before the act would be enacted;
• has an LTV greater than 120%; and
• it has been determined that the net present value of reducing principal exceeds the net present value of foreclosing.

Under the program, qualified mortgages would see their principal reduced to an LTV not more than 90%. If a borrower in the program eventually sells the home, at least one-third of the amount that the property appreciated would be given to the GSE that owns or guarantees the mortgage. If the property eventually enters foreclosure, then the borrower must pay Fannie or Freddie the difference between the sales price at foreclosure and the amount of the outstanding principal balance before the principal reduction.

S. 2093, the Preserving American Homeownership Act of 2012 (Senator Robert Menendez), requires the director of the FHFA and the Federal Housing Commissioner to establish shared appreciation mortgage modification pilot programs for FHFA and FHA loans. To be eligible, borrowers with Fannie Mae, Freddie Mac, or FHA loans must

• be at least 60 days delinquent or at risk of imminent default on their mortgage;
• have the mortgage on their primary residence;
• be underwater on their mortgage; and
• have a documented financial hardship that prevents or will prevent them from making their payments.

Under the pilot programs, loans would be reduced to an LTV of 95% within three years by reducing the principal by one-third of the necessary amount over each of the three years; the interest rate could also be reduced if necessary to achieve affordable payments; the homeowner would have to pay the investor at most 50% of any increase in the value of the house if the homeowner refinances or sells the home. Borrowers would be eligible only if the modification results in greater cash flows to investors than other loss mitigation activities.

Large-Scale Refinancing

With mortgage rates at historic lows, some have proposed establishing a large-scale refinancing program that would help borrowers who are current on their mortgage refinance into a new mortgage with a lower interest rate. Unlike a principal reduction, the principal balance of the loan does not change with a mortgage refinancing. A borrower prepays the existing loan with a new loan, presumably with more favorable terms. Refinancing allows borrowers to potentially lower their monthly payments. To refinance, a borrower typically must be current on his or her mortgage loan.

Because refinancing helps borrowers who are current, it is unlikely to have a major effect on the housing market but may prevent some foreclosures that would occur in the absence of a refinance by lowering payments. However, refinancing has the potential to have a larger effect on the economy by stimulating consumer spending and improving household balance sheets. A mortgage refinance lowers a borrower's monthly payment, freeing up more income for non-housing related spending. Some of the additional spending of borrowers may come at the cost of the financial sector. In a refinance, the previous owner of the mortgage is repaid earlier than expected and now faces reinvesting in a lower interest rate environment. Though some investors (including banks) may lose from refinancing, other banks and mortgage originators benefit from the increased business associated with refinancing. Similarly, mass refinancing might cause Fannie Mae and Freddie Mac to lose investment income but gain from the reduced likelihood of default, though it is unclear which effect is larger.

Current refinancing policy discussions center on two approaches: (1) develop a new program to refinance non-GSE loans into government-insured loans; and (2) expand the existing Home Affordable Refinance Program (HARP) to help more homeowners refinance their GSE loans. In his State of the Union address and a subsequent fact sheet,⁶¹ President Obama endorsed both approaches. He proposed allowing borrowers who are current on their non-GSE loans to refinance through a program run by the FHA. The program would be funded through a tax on large financial institutions.⁶² President Obama also proposed streamlining HARP to allow even more borrowers with GSE guaranteed loans to refinance.

The recent mortgage settlement between five banks and 49 state attorneys general and the federal government is estimated to provide at least $3 billion to refinance underwater borrowers.⁶³

Several congressional proposals, such as S. 170, the Helping Responsible Homeowners Act (Senator Barbara Boxer et al.); H.R. 363, the Housing Opportunity and Mortgage Equity Act of 2011 (Representative Dennis Cardoza et al.); and S. 3085, the Responsible Homeowner Refinancing Act of 2012 (Senator Robert Menendez et al.), would require Fannie Mae and Freddie Mac to expand refinancing of GSE loans.⁶⁴ The specifics of these proposals are discussed in "Presidential and Legislative Proposals for Refinancing" below.

Barriers to Refinance

A mortgage is a callable loan, meaning that borrowers can pay off the balance of the loan at any time, usually without penalty. When a borrower refinances, the borrower is exercising this right to prepay by taking out a new loan and using it to pay off the previous loan.⁶⁵ A borrower can refinance with the existing lender or with a new lender. Although the existing lender has control over its own underwriting standards (the determination about who it will lend to), it does not have control over its competitors' underwriting standards. The existing lender faces the risk that a borrower will prepay the loan using a different lender and has little control over it. The value of a mortgage that a lender holds reflects this prepayment risk.

Mortgage interest rates are one of the major factors that influence a borrower's decision to refinance.⁶⁶ When interest rates fall, refinances typically increase, causing prepayment risk to rise. Although rates are at historic lows, mortgage refinances have not been as high as some would predict. Figure 7 tracks the interest rate from the Primary Mortgage Market Survey (PMMS) compiled by Freddie Mac and the Mortgage Bankers Association (MBA) Refinance Index, a measure of refinancing activity. Although interest rates have fallen significantly and the government is encouraging refinancing, the MBA Refinancing Index has not reached the peak reached in 2003 when interest rates also fell substantially.

Figure 7. Refinancing Index

Source: William Hedberg and John Krainer, Mortgage Prepayments and Changing Underwriting Standards, Federal Reserve Bank of San Francisco, July 2010, at http://www.frbsf.org/publications/economics/letter/2010/el2010-22.html.

Insufficient home equity is a significant barrier to refinancing. Traditionally, lenders' underwriting standards require borrowers to have at least 20% positive equity in their home to refinance.⁶⁷ If a borrower's home is valued at $200,000 and the borrower owes $160,000 or less on the mortgage (LTV below 80%), then the borrower is potentially eligible to refinance at a bank, credit union, or other traditional avenue. A financial institution that is refinancing a mortgage wants a borrower to have positive equity in the home to protect the value of the collateral in the event house prices fall. If a borrower has little or no equity in the home and house prices fall, then should the borrower default, the financial institution could not recover the full value of its loan by selling the house.

In addition to insufficient equity, closing costs are a potential barrier to refinance for some borrowers. Borrowers will not refinance every time interest rates fall because there are fixed costs to refinancing. A borrower may save a little each month from having a lower interest rate, but it is only worthwhile to refinance if the amount saved is greater than the cost of refinancing. A typical estimate is that interest rates have to fall by 1 to 2 percentage points below a borrower's existing rate for it to be in the borrower's best interest to refinance.⁶⁸

Large-scale refinancing proposals generally target those excluded from traditional refinancing efforts either due to insufficient equity or to high closing costs.

Potential Impact of Large-Scale Refinancing

Housing Market

Mass refinancing proposals for underwater borrowers generally target borrowers who are current on their mortgages and are therefore not necessarily borrowers in imminent danger of default. A refinancing program would impact the housing market by preventing foreclosures that would have otherwise occurred had the program not been in place by lowering borrowers' monthly payments. Any potential impact on house prices is likely to be through averted foreclosures. However, because refinancing programs target borrowers who are current, the number of averted foreclosures is likely to be limited. An analysis of a stylized large-scale refinancing program for GSE loans conducted by the Congressional Budget Office (CBO) estimated that 3.8% of homeowners that receive a refinance would have been foreclosed on in the absence of a refinance.⁶⁹ Similarly, a study by Glenn Hubbard, Chris Mayer, and Alan Boyce (Hubbard et al.) of a refinancing program for GSE loans estimates that 5% of homeowners would lose their home without a refinance.⁷⁰ However, these studies focused on refinancing for GSE loans, which tend to be of a higher credit quality than non-GSE loans. Proposals that allow non-GSE loans to refinance may have a larger impact on avoided foreclosures.

Consumer Spending

Some argue that by targeting refinancing proposals to only current borrowers, the primary motivation for mass refinancing is economic stimulus.⁷¹ In his testimony before the Senate Committee on Banking, Housing, and Urban Affairs, Mark Zandi, the chief economist and co-founder of Moody's Analytics, estimated that refinancing would save the average borrower approximately $2,500 a year.⁷² The magnitude of the potential stimulus depends on the number of borrowers that participate. Zandi calculated that 6.8 million borrowers would refinance under the President's proposals, increasing growth of gross domestic product (GDP) by 0.1 percentage points this year. CBO's analysis of a stylized refinancing program predicted fewer refinances, 2.9 million borrowers, whereas Hubbard et al. estimated 14 million refinances. The estimates of refinances vary due to the eligibility criteria of the analyzed programs and the assumptions about borrower participation.

Although a refinancing increases the amount of disposable income for a borrower, it reduces the potential income for the investor holding the mortgage. Borrowers are more likely to refinance when it is in their best interest, such as when interest rates fall. When a borrower refinances, the remaining amount of principal that is owed is returned to the mortgage holder, requiring the mortgage holder to reinvest at a time when rates are low. Similar to what happens in a principal reduction, the gains to borrowers and the losses to investors in mass refinancing are not necessarily zero-sum.⁷³ If borrowers spend more domestically of an additional dollar than investors would have spent of the same dollar, then refinancing would increase aggregate spending and support the economic recovery.⁷⁴ In addition, a significant percentage of mortgage-backed securities (MBS) are held by foreign investors and the federal government. Reducing the investment income to those holders of MBS is unlikely to have a large negative impact on consumer spending in the United States. See Figure 8 for a breakdown of the major investors in agency MBS by type of investor. Agency MBS include securities guaranteed by Fannie Mae, Freddie Mac, and Ginnie Mae.⁷⁵ The Federal Reserve Bank of New York estimates that every dollar that a borrower's monthly payment is reduced by a refinancing would generate nearly 50 cents of additional spending.⁷⁶

Some argue that it is unfair to target economic stimulus through a relatively small number of homeowners who happen to meet certain eligibility criteria.⁷⁷ They suggest that if there is going to be fiscal stimulus, it should be broader based and discussed more openly and directly, not necessarily within the framework of housing policy.

Figure 8. Investors in Agency MBS (as of Fourth Quarter of 2011)

Source: Data from the Federal Reserve System, Flow of Funds Accounts of the United States, Fourth 2011, Table L. 210, at http://www.federalreserve.gov/releases/z1/Current/z1.pdf. Chart modeled off of Joseph Tracy and Joshua Wright, Why Mortgage Refinancing Is Not a Zero-Sum Game, Federal Reserve Bank of New York, January 11, 2012, at http://libertystreeteconomics.newyorkfed.org/2012/01/why-mortgage-refinancing-is-not-a-zero-sum-game.html. Note: Percentages have been rounded.

Financial Sector

Even if the benefits to borrowers of large-scale refinancing are greater than the losses to the mortgage holders, mortgage holders are still taking a loss. As Figure 8 shows, a large-scale refinancing of agency MBS would not only impact large institutional investors, but would also impact mutual funds and individuals' savings in public and private pension funds, which hold approximately 18.6% of agency MBS. The Investment Company Institute (ICI), the national association of U.S. investment companies, estimates that 64% of mutual fund-owning households had annual incomes of less than $100,000 in 2010.⁷⁸ However, as mentioned earlier, the value of the MBS held by investors already factors in prepayment risk. A mass refinancing proposal would only impose a cost on investors if refinances were greater than anticipated. In addition, some banks and other institutions may benefit from the increased business associated with refinancing.

The Home Affordable Refinance Program

To help borrowers with GSE loans refinance even if they have little or no equity, the Obama Administration created HARP as part of its Making Home Affordable program.⁷⁹ To be eligible a borrower must

• have a mortgage owned or guaranteed by Fannie Mae or Freddie Mac;
• have a mortgage on a single-family home;
• owe more than 80% of the value of the home on the mortgage;
• be current on mortgage payments with no late payment in the past six months and no more than one late payment in the past 12 months;
• have the ability to make the new payments; and
• have had the mortgage sold to Fannie Mae or Freddie Mac by May 2009.⁸⁰

Fannie Mae and Freddie Mac do not make loans themselves but buy loans from lenders. Through HARP, the GSEs agree to purchase a new loan from the originator if the borrower meets the eligibility criteria. The GSEs only refinance a borrower through HARP if the borrower's loan is already guaranteed by the GSEs. A refinance, therefore, does not add additional credit risk to the GSEs because they already own the credit risk of the borrower.⁸¹ If a refinance lowers a borrower's monthly payments and makes it less likely that the borrower will default, then a refinance could lower the GSEs' credit risk.

The Administration originally estimated that HARP would aid between 4 million and 5 million borrowers, but the program has refinanced approximately 998,000 mortgages as of November 2011.⁸² Most of the beneficiaries of HARP are those with some positive equity or only moderate negative equity, not those who are deeply underwater. Of those who refinanced through HARP, 91% had LTV ratios between 80% and 105%. Fewer than 9% had LTV ratios over 105%.

Experts have identified multiple factors that may be limiting the reach of HARP.⁸³ The GSEs charge additional fees, called loan level price adjustments, to some borrowers to compensate for the additional risk that they might pose based on their credit characteristics.⁸⁴ Loan level price adjustments, as well as other closing costs associated with getting a mortgage, such as an appraisal, may require more upfront expenses than a borrower can afford. In addition, HARP allows for more streamlined refinancing if performed through a borrower's existing servicer rather than a different servicer. This could potentially reduce competition and increase rates faced by borrowers. There are also questions about the capacity of originators to handle the increased refinance applications.⁸⁵ Also, HARP is a voluntary program; an eligible borrower needs to find a lender willing to offer them a new loan.

To address some of these concerns, FHFA announced changes to HARP in October 2011.⁸⁶ Previously, HARP eligibility was restricted to borrowers with LTV ratios below 125%, but the cap has been removed under HARP 2.0.⁸⁷ The GSEs have also agreed to eliminate or reduce some of the loan level price adjustments that were charged to borrowers. They will also attempt to reduce closing costs through greater use of automated valuation models in place of property appraisals. Fannie and Freddie are incentivizing lenders to refinance homeowners by waiving certain representations and warranties made on the original loans. Representations and warranties are assurances that lenders make to Fannie and Freddie about the quality of a loan when they are selling the loan to the GSEs. If it is later determined that the loan does not meet the criteria that the lender claimed the loan met, then the lender may be required to repurchase the loan.⁸⁸ By waiving the representations and warranties against the original loan, Fannie and Freddie are allowing the lender to re-underwrite the loan to ensure that it meets the agreed upon standards.⁸⁹

Presidential and Legislative Proposals for Refinancing

In his 2012 State of the Union Address, President Obama proposed allowing some non-GSE borrowers to refinance through a new program to be run by the FHA and to streamline HARP to allow more borrowers with GSE loans to refinance. In a subsequent fact sheet,⁹⁰ the Administration outlined the proposed eligibility criteria for non-GSE borrowers. Any borrower with a mortgage not guaranteed by the GSEs would be eligible if the borrower

• has been current on the mortgage for the past six months and has not missed more than one payment in the previous six months;
• has a current FICO score greater than 580;
• has a mortgage no larger than the current FHA loan limit in his or her area;⁹¹
• is refinancing the loan on the borrower's owner-occupied, single-family primary residence;
• is employed or can otherwise prove that he or she can afford the new mortgage; and
• meets certain loan-to-value limits on the new mortgage.

The application process would be streamlined by eliminating the need for a new appraisal or a tax return to prove primary residency. The $5 billion-$10 billion cost of the program would be paid for through a fee on large financial institutions. A separate fund, independent of FHA's existing Mutual Mortgage Insurance (MMI) Fund, would be created for the new refinancing program. The Administration proposal would require enactment of legislation.

The GSE component of the Administration's plan would further streamline HARP beyond the changes made in October 2011. The Administration proposes reducing closing costs by eliminating manual appraisals entirely. The GSEs would be required to use automated valuation models or other appraisal alternatives if they are unable to use automated valuation models due to lack of recent comparable sales. The Administration proposal would also allow lenders who do not currently service a borrower's mortgage to have access to the same streamlined underwriting that the current service is allowed to use. This could increase competition and potentially lower costs to borrowers. The plan would make the streamlined refinancing available to all GSE borrowers, even those with significant equity in their homes. Currently, borrowers with sufficient equity can more easily refinance through traditional channels, but they cannot take advantage of the lower costs and streamlined process associated with HARP.

Underwater borrowers who participate in either the FHA refinancing program or HARP would have the option to apply their savings from refinancing to either lower their monthly payments or keep their monthly payments at the same level but build equity more quickly in their home by shortening the term of the loan. Those who choose to build equity more quickly would have their closing costs covered by the GSEs or FHA.⁹²

As mentioned previously, refinancing GSE loans does not add additional credit risk to the GSEs because they already own the credit risk of the borrower defaulting. However, the FHA refinancing program would require the government to assume additional risk. The loans that are eligible for the FHA program would not be guaranteed by the government prior to the refinancing but would be insured after the refinance.

Multiple proposals have been introduced in the 112^th Congress to mandate greater refinancing of GSE loans. S. 170, the Helping Responsible Homeowners Act (Senator Barbara Boxer et al.), and H.R. 363, the Housing Opportunity and Mortgage Equity Act of 2011 (Representative Dennis Cardoza et al.), require Fannie Mae and Freddie Mac to establish refinancing programs for qualified mortgages that they own or guarantee. For both bills, a qualified mortgage is a first mortgage on a one- to four-family dwelling that is the principal residence of the borrower and is owned or guaranteed by Fannie Mae or Freddie Mac. Both also put certain restrictions on the GSEs' ability to charge additional fees besides the standard guarantee fees and place a cap on the interest rate that a borrower can be charged. Under both proposals, the refinancing program would expire one year after the enactment of the bills; S. 170 gives FHFA the option to extend the program.

S. 170 and H.R. 363 also differ in several ways. S. 170 requires borrowers to be current on their mortgages to participate, whereas H.R. 363 does not have such a restriction. H.R. 363 prohibits the mortgage servicer that is refinancing the loan from charging the borrower any fee for refinancing, but it instructs the GSEs to pay the servicer up to $1,000 for each qualified mortgage that it refinances. S. 170 does not put limits on the fees that the servicer may charge and does not pay the servicer for the refinancing. S. 170 and H.R. 363 are similar to HARP in that they target GSE loans for refinancing, but both bills put greater emphasis on reducing the costs associated with refinancing.

S. 3085, the Responsible Homeowner Refinancing Act of 2012 (Senator Robert Menendez et al.), would expand borrower eligibility and streamline the application process for HARP. S. 3085 would change HARP eligibility to include borrowers with more than 20% equity as well as extend the date of HARP eligibility by one year to May 31, 2010. The bill would also prohibit loan level price adjustment fees and other up-front fees and eliminate appraisal costs for borrowers. S. 3085 would attempt to increase the competition among servicers by allowing the same streamlined refinancing process to apply to a different servicer as to a borrower's existing servicer. In addition, the bill would impose penalties on junior lien holders and mortgage insurers if they take action to prevent an eligible borrower from refinancing.

Renting Foreclosed Homes

Federal Reserve Chairman Ben Bernanke and other policy experts state that the housing market is experiencing serious imbalances. Overbuilding leading up to the bursting of the housing bubble, the subsequent increase in foreclosures, and the fall in household formation⁹³ have led to an excess supply of homes in the market. Economic theory predicts that prices will continue to fall until the market clears. Falling house prices can contribute to a negative spiral in which falling prices lead to more foreclosures and prices falling even further.

The foreclosed homes that financial institutions assume control over are called real estate-owned (REO) properties. As of December 2011, single-family REOs valued at $11.6 billion, an estimated 79,000 units, are held by FDIC-insured institutions.⁹⁴ In addition, Fannie Mae and Freddie Mac held 179,063 REO properties at the end of 2011,⁹⁵ and FHA held 31,046 properties as of January 31, 2012.⁹⁶ According to some estimates, a third of REO inventory is from private-label securities.⁹⁷

Because of the uncertainty surrounding the foreclosure process in the wake of investigations into the robo-signing allegations,⁹⁸ many large banks slowed down the rate at which they initiated foreclosure proceedings. With the recent mortgage settlement involving five of the banks that engaged in robo-signing, some expect the rate of foreclosures to increase as the backlog of foreclosures reaches the market.⁹⁹ Figure 9 shows the shadow inventory, the approximately 1.5 million borrowers who are seriously delinquent and in danger of having their home become part of the REO inventory. In a January 2012 speech, William Dudley, the president and chief executive officer of the Federal Reserve Bank of New York, estimated that the flow of properties into REO could be as high as 1.8 million per year in 2012 and 2013, up from 1.1 million in 2011 and 600,000 in 2010.¹⁰⁰

Figure 9. Shadow Inventory

Source: CoreLogic.

To address the imbalances in the housing market, some propose temporarily allowing financial institutions to rent the foreclosed homes or to sell the homes to investors who agree to rent the vacant units. Historically banks and the GSEs have been encouraged by their regulators to sell their REO property quickly and have not performed extensive property management services, though they are allowed to rent REOs for some limited periods.¹⁰¹ Some proposals focus on renting properties as a way to help foreclosed homeowners through lease back or right-to-rent. Others do not address the foreclosed homeowners but focus on rentals as a way to stabilize neighborhoods and home prices. Given the imbalances in the owner-occupied housing market, supporters of a rental program believe expanding the amount of time that REOs can be rented is justified. Critics worry that if house prices do not increase while the homes are being temporarily rented, then the policy may serve to delay recovery by flooding a weak market with additional homes a few years in the future. Critics would rather allow the market to bottom out now rather than risk prolonging the housing slump.

Barriers to Renting

In a housing white paper prepared for Congress, the Federal Reserve outlined three obstacles to implementing an REO-to-rental program as well as possible policy solutions. First, as mentioned previously, regulators have discouraged financial institutions from holding REO properties.¹⁰² Institutions are currently allowed to temporarily rent REO if doing so maximizes the potential return of the asset. On April 5, 2012, the Federal Reserve issued guidance to financial institutions that it regulates to "clarify supervisory expectations regarding rental of residential REO properties by such organizations while such circumstances continue."¹⁰³ The guidance explains to regulated REO holders how they may rent the properties within "the existing statutory and regulatory holding-period limits."¹⁰⁴

Second, if the REO program involves sales to investors, then investors would need to purchase a large number of properties in close geographic proximity to each other to be profitable on a large scale. Many of the components of property management, such as processing payments, advertising for renters, and developing relationships with local contractors, benefit from economies of scale—the more houses a landlord has in a given area the more profitable it is. But if an investor must assemble a critical mass of properties one property at a time, then the investor must bear carrying costs, such as maintenance and taxes, without receiving any revenues.

To facilitate bulk transactions, the Federal Reserve notes that REO holders could auction their existing inventory in a given area to an investor as well as the future stream of REOs in an area. However, the Federal Reserve paper states that bulk sales can yield lower recoveries than individual sales in some cases because each property cannot be allocated to the highest individual bidder when sold in bulk. But it may take longer to sell properties individually, during which the properties deteriorate.¹⁰⁵ Others predict that bulk investors will pay a premium for the opportunity to buy a large quantity at once rather than pay the expense of individually collecting properties.¹⁰⁶

Third, to amass a sufficient number of properties, investors may need to obtain financing. However, financing for purchases of bulk pools of REO properties is not something that lenders traditionally offer. For example, Freddie Mac only allows investors to finance up to 4 homes and Fannie Mae sets the limit at 10 homes.¹⁰⁷ Without more accessible credit, there may be a limited number of investors able to participate.

In addition, not all communities are suited for a REO-to-rental program. The program may be most effective by targeting communities in which the number of REO properties is high, but the current demand among investors for one-at-a-time purchases is weak. Figure 10 shows the communities with the most GSE and FHA REO properties.

Figure 10. Top 10 Metropolitan Statistical Areas for Freddie Mac, Fannie Mae, and FHA REO (as of December 26, 2011)

Source: Created by CRS from FHFA data.

Potential Impact of Rental Program

Housing Market

A REO-to-rental program would prevent foreclosed homes from flooding the "for-sale" market and instead temporarily shift them to the "rental" market. Foreclosed homes negatively impact house prices in several ways. First, foreclosed homes and other forms of distressed sales (such as a short sale) typically sell at a discount compared with non-distressed sales,¹⁰⁸ potentially lowering the appraisal value for neighboring homes. Second, foreclosures increase the supply of homes that are on the market, which typically drives down home prices. Lower home prices may benefit those who are looking to purchase homes, but they hurt those who currently own their homes. A Goldman Sachs research paper estimated that if 475,000 vacant REOs were converted to rentals, then house prices would rise by an additional 0.5% in the first year of the program and by 1% in the second year.¹⁰⁹ If the program increases house prices in the for-sale market, then prices are expected to fall in the rental market. The Federal Reserve white paper on housing notes that the current movement in relative prices—prices that are falling or flat in the for-sale market but rising in the rental market—provides evidence that it may be appropriate to transition REO properties to rentals in some areas.¹¹⁰ Figure 11 shows the trend in prices for rent compared with sales.

Figure 11. Rent-Versus-Own Prices

Source: U.S. Census Bureau, at http://www.census.gov/hhes/www/housing/hvs/historic/index.html.

Financial Sector

The major REO-to-rental policy proposals, discussed in "Legislative Proposals for Renting REO" below, have two variations: (1) allow banks, GSEs, and other institutions to rent properties directly themselves or (2) sell the properties in bulk to investors who will rent them. In both options, REO is rented, but how and when the assets are eventually sold will impact the balance sheet of the institutions.

If the REO properties are immediately sold in bulk to investors, then the institutions can have a relatively good idea of what price the properties can be sold given current trends, though some factors on the margin may influence the final price, such as whether auction methods are used or if financing is provided to facilitate the sale. However, if the institutions rent the properties for several years prior to eventually selling, then the direction of future house prices will significantly impact the price for which the properties are eventually sold. Supporters of policies that encourage institutions to rent for several years before selling are likely expecting future house prices to increase so that eventual sales occur in a more stable market. However, future asset prices are inherently unpredictable. If future house prices fall, then institutions would lose by postponing the sale of REO properties.

Consumer Spending

Unlike principal reductions and mass refinancing, renting foreclosed homes does not reduce existing homeowners' payments or increase their disposal income. Any impact on consumer spending is likely to be indirect through stabilizing house prices and preserving nearby homeowners' equity and net worth.

FHFA Pilot Program

In August 2011, FHFA, in consultation with the Treasury Department and the Department of Housing and Urban Development (HUD), submitted a request for information on ideas for asset disposition.¹¹¹ After receiving more than 4,000 responses, FHFA announced the first property sales under its Real Estate-Owned (REO) Initiative on February 27, 2012.¹¹² Under the REO Initiative, Fannie Mae and Freddie Mac REO properties in the hardest-hit metropolitan areas are to be sold to qualified investors who agree to rent the properties for a specified number of years. Table 5 summarizes the Fannie Mae REO in the first transaction.

Most of the properties in the first transaction (91.2%) are single-family homes or condominiums. However, most of the single-family homes and condominiums are already rented with fewer than 10% vacant. Because of their desire to see vacant homes rented out, some commentators expressed disappointment that a larger percentage of the homes were not vacant.¹¹³ Others argue that by starting with properties with existing tenants, FHFA can attract more investors for the first sale; by having existing renters, investors are ensured of a future stream of income and may therefore face less uncertainty in attempting to price this new asset class.¹¹⁴

It is unclear if FHFA will allow Fannie Mae (and possibly Freddie Mac in potential future sales) to retain an ownership stake in the properties or if they will be true sales. It is also unclear if the GSEs will provide financing to investors and how the properties will be sold, whether to a single investor, multiple investors by metropolitan area, or some other method.¹¹⁵

Legislative Proposals for Renting REO

Multiple proposals have been introduced in the 112^th Congress to establish REO rental programs. Some proposals focus on renting foreclosed properties to any renter, while others focus on allowing delinquent homeowners to stay in their home but pay rent. H.R. 2636, the Neighborhood Preservation Act of 2011 (Representative Gary Miller et al.), and S. 2080, the Keeping Families in their Home Act of 2012 (Senator Dean Heller), temporarily authorize depository institutions, depository institution holding companies,¹¹⁶ Fannie Mae, and Freddie Mac to lease foreclosed property held by those institutions for up to five years. Section 2 of H.R. 2636 makes clear that one of the goals of the proposed REO program is to increase the options available to financial institutions to maximize the value of their assets. By temporarily renting a home before selling it, the institution could eventually sell the property into a more stable market and receive a higher return.

H.R. 1548, the Right to Rent Act of 2011 (Representative Raúl Grijalva et al.), gives eligible borrowers who are delinquent on their mortgage and subject to foreclosure proceedings the option to rent their home at a fair market rent for up to five years. To be eligible, a delinquent borrower must have a mortgage originated by July 1, 2007, on a single-family property that has been used as the borrower's principal residence for at least two years prior to the foreclosure filing. In addition, the borrower's home must have had a purchase price less than the median purchase price for residences in the metropolitan statistical area (or state if not in an MSA) at the time of the purchase for the borrower to be eligible. The fair market rent will be determined by the court and adjusted annually in line with the consumer price index.

Acknowledgments

Matthew Kurlanzik, former CRS research associate, co-authored this report.

Footnotes

1.	For more on the current economic recovery, see CRS Report R41332, Economic Recovery: Sustaining U.S. Economic Growth in a Post-Crisis Economy, by [author name scrubbed].
2.	Testimony of Ms. Laurie F. Goodman, a senior managing director at Amherst Securities Group, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, Subcommittee on Housing, Transportation and Community Development, New Ideas to Address the Glut of Foreclosed Properties, hearing, 112th Cong., 1st sess., September 20, 2011, available at http://banking.senate.gov.
3.	The "reasonable" and "lower bound" default rates are estimated default rates based on Goodman's projections about how the different categories of mortgages will perform. The "lower bound" estimate is a more conservative estimate that assumes fewer defaults in the future.
4.	Using data from CoreLogic, William Dudley of the Federal Reserve Bank of New York estimated 11 million underwater borrowers in a January 2012 speech. See http://www.newyorkfed.org/newsevents/speeches/2012/dud120106.html. In his February 2012 congressional testimony, Mark Zandi estimated 14.6 million underwater borrowers using data from Equifax. See http://banking.senate.gov/public/index.cfm?FuseAction=Files.View&FileStore_id=65e8490f-7a21-43f8-a24b-7f7a856ece38.
5.	See Testimony of Mark Zandi, chief economist at Moody's Analytics, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, New Ideas for Refinancing and Restructuring Mortgage Loans, 112th Cong., 1st sess., September 14, 2011. However, other experts find minimal relationship between negative equity and homeowner mobility; see Sam Schulhofer-Wohl, "Negative Equity Does Not Reduce Homeowners' Mobility," Federal Reserve Bank of Minneapolis Working Paper Series, December 2010.
6.	One study of borrowers in Massachusetts with negative equity in the 1990s found that more than 90% retained their homes, suggesting that opportunistic default may be overstated in some cases; see Christopher L. Foote, Kristopher Gerardi, and Paul S. Willen, "Negative Equity and Foreclosure: Theory and Evidence," Federal Reserve Bank of Boston, Working Paper No. 08-3, available at http://www.bos.frb.org/economic/ppdp/2008/ppdp0803.htm. The current housing downturn has seen steeper declines in house prices, potentially making the implications of negative equity more severe than the period studied by Foote et al.
7.	In this example, a borrower may attempt to arrange a short sale, but that option may not always be available. A short sale is a sale in which the mortgage holder allows the borrower to sell the house for an amount that is less than the amount owed and does not require the borrower to pay the difference.
8.	RealtyTrac, a publisher of one of the largest databases of foreclosures, estimates that the average price of a foreclosure-related sale is approximately 29% less than a non-foreclosure sale. See http://www.realtytrac.com/content/foreclosure-market-report/q4-and-year-end-2011-us-foreclosure-sales-report-7060. There may also be additional costs associated with foreclosure, such as maintaining the property, that reduce the returns on a foreclosure.
9.	The following analysis of the principal reduction decision is based on Manuel Adelino, Kristopher Gerardi, and Paul S. Willen, Why Don't Lenders Renegotiate More Home Mortgages? Redefaults, Self-Cures, and Securitization, Federal Reserve Bank of Boston, Public Policy Discussion Paper No. 09-4, July 2009.
10.	The Type 4 borrower is a borrower that is determined to be net present value negative and therefore not considered a good candidate for a modification. A borrower that is net present value negative is one in which the servicer determines that the expected gains from no modification are likely to be greater than the gains associated with a modification. See https://www.hmpadmin.com/portal/programs/docs/hamp_servicer/npvmodeldocumentationv403.pdf.
11.	Strategic default behavior is characterized by Fannie Mae as "borrowers who walk-away and had the capacity to pay or did not complete a workout alternative in good faith." See http://www.fanniemae.com/portal/about-us/media/corporate-news/2010/5071.html.
12.	Moral hazard refers to the phenomenon in which actors take on more risk because they do not bear the full consequences.
13.	For more on house lock, see the Council of Economic Advisers, Economic Report of the President, February 2012, at http://www.whitehouse.gov/sites/default/files/microsites/ERP_2012_Complete.pdf.
14.	A situation that is zero-sum is one in which the gains to one party are exactly offset by losses to another.
15.	The argument related to principal reduction potentially increasing consumer spending is modeled off of similar analysis for mortgage refinances; see Joseph Tracy and Joshua Wright, Why Mortgage Refinancing Is Not a Zero-Sum Game, Federal Reserve Bank of New York, January 11, 2012, at http://libertystreeteconomics.newyorkfed.org/2012/01/why-mortgage-refinancing-is-not-a-zero-sum-game.html.
16.	Approximately 31.2% of all Fannie Mae, Freddie Mac, and Ginnie Mae mortgage-backed securities are held by the U.S. Treasury, the Federal Reserve System, foreign investors, and Fannie Mae and Freddie Mac. See the Federal Reserve System, Flow of Funds Accounts of the United States, Fourth 2011, Table L. 210, http://www.federalreserve.gov/releases/z1/Current/z1.pdf.
17.	The Federal Reserve Bank of New York analysis also discusses the impact that foreclosures have on credit history and the prolonged drag this could be on the economy.
18.	For more on the household wealth effect, see Dean M. Maki and Michael G. Palumbo, Disentangling the Wealth Effect: A Cohort Analysis of Household Saving in the 1990s, The Federal Reserve System, April 2001, at http://www.federalreserve.gov/pubs/feds/2001/200121/200121pap.pdf.
19.	See CRS Report RL30354, Monetary Policy and the Federal Reserve: Current Policy and Conditions, by [author name scrubbed].
20.	The owner of a mortgage loan or mortgage-backed security typically hires a mortgage servicer to act on its behalf. When loans are current, a mortgage servicer collects payments from borrowers and forwards them to the mortgage holders. If the borrower becomes delinquent, a servicer may offer the borrower an option that could allow the borrower to stay in his or her home, or the servicer may pursue foreclosure.
21.	See CRS Report R42041, National Mortgage Servicing Standards: Legislation in the 112th Congress, by [author name scrubbed]. For examples of shortcomings in servicers' operational infrastructure, such as determining the return of a short sale versus a foreclosure sale, see Kate Berry, "Banks Face Tough Choices Unloading REO Properties," American Banker, February 23, 2012.
22.	For more on coordination problems related to bankruptcy and distressed assets, see David Smith and Per Stromberg, "Maximizing the Value of Distressed Assets: Bankruptcy Law and the Efficient Reorganization of Firms," in Systemic Financial Crises, eds. Patrick Honohan and Luc Laeven (Cambridge: Cambridge University Press, 2005). Note that the article focuses on firms rather than individuals.
23.	See Senator Bob Corker, "Obama Administration's Principal Write-down Proposal for Underwater Home Mortgages is 'Terrible Public Policy,' Forces Tennesseans to Pay for Reckless Housing Practices in Other States," press release, January 30, 2012, at http://www.corker.senate.gov.
24.	See CRS Report RL34386, Could Securitization Obstruct Voluntary Loan Modifications and Payment Freezes?, by [author name scrubbed].
25.	Using data from the Federal Reserve System on mortgage debt outstanding, non-GSE and non-government insured mortgages account for approximately 43% of mortgage debt outstanding at the end of September 2011. See http://www.federalreserve.gov/econresdata/releases/mortoutstand/current.htm.
26.	There are other government programs involving principal reduction besides HAMP, such as the FHA Short Refinance, but HAMP is the largest program and is the focus of the next section. For more on federal government programs to reduce mortgage principal, see CRS Report R40210, Preserving Homeownership: Foreclosure Prevention Initiatives, by [author name scrubbed].
27.	See Making Home Affordable website, at http://www.makinghomeaffordable.gov.
28.	This section was prepared using material from CRS Report R40210, Preserving Homeownership: Foreclosure Prevention Initiatives, by [author name scrubbed].
29.	"The NPV Model compares the expected discounted cash flows associated with the modification of a loan – considering probabilities of default – under two scenarios: the loan is modified according to HAMP terms and the loan is not modified... A loan that is NPV 'positive' – where the value of the probability-weighted mod cash flows exceed the value of the probability-weighted no-mod cash flows – is considered to be a good candidate for modification." See Steve Holden, Therese Scharlemann, and Austin Kelly et al., The HAMP NPV Model: Development and Early Performance, Federal Housing Finance Agency, Working Paper 11-1, July 2011, p. 3, at http://www.fhfa.gov/webfiles/21680/REE_HAMP_07-22-11_FINAL.pdf.
30.	Detailed guidelines on the Principal Reduction Alternative were released in Supplemental Directive 10-05 on June 3, 2010. These guidelines are available at https://www.hmpadmin.com/portal/docs/hamp_servicer/sd1005.pdf.
31.	U.S. Department of the Treasury, Making Home Affordable Program Servicer Performance Report Through January 2012, at http://www.treasury.gov/initiatives/financial-stability/results/MHA-Reports/Documents/Jan%202012%20MHA%20Report_WITH_SERVICER_ASSESSMENTS_FINAL.PDF.
32.	See http://www.treasury.gov/connect/blog/Pages/Expanding-our-efforts-to-help-more-homeowners-and-strengthen-hard-hit-communities.aspx.
33.	The five banks are Ally/GMAC, Bank of America, Citi, JPMorgan Chase, and Wells Fargo. For more information, see http://www.nationalmortgagesettlement.com/.
34.	For an overview of the settlement see http://www.atg.wa.gov/uploadedFiles/Home/About_the_Office/Cases/National_Mortgage_Settlement/National_Settlement_Executive_Summary.pdf.
35.	The remaining funds are allocated among programs to refinance borrowers, other forms of homeowner assistance (such as facilitating short sales), to compensate individuals who were improperly foreclosed on, and the participating states' foreclosure relief and housing programs.
36.	See The Federal Reserve System, Flow of Funds Accounts of the United States, First Quarter 2012, Table L. 218, at http://www.federalreserve.gov/releases/z1/Current/z1.pdf.
37.	In private-label securities and mortgages held in banks' portfolios, the investor who holds the mortgage (or its agent, the mortgage servicer) ultimately makes determinations about loan modifications. For GSE loans, Fannie Mae and Freddie Mac, which are the entities that bear the credit risk and set guidance for the servicers, make the determination about loan modifications. Loan modifications attempt to reduce the risk of default and are therefore made by those who bear the default risk.
38.	Since entering into conservatorship, Fannie Mae and Freddie Mac have performed approximately 1.8 million home retention actions, which include repayment plans, forbearance, and other approaches. See Federal Housing Finance Agency, Foreclosure Prevention and Refinance Report November 2011, at http://www.fhfa.gov/webfiles/23123/Nov_2011_Foreclosure_Prev_Rpt.pdf.
39.	See http://www.fhfa.gov/webfiles/23056/PrincipalForgivenessltr12312.pdf. This CRS report does not attempt to interpret FHFA's mandate or comment on whether FHFA is interpreting it correctly or incorrectly.
40.	See http://www.fhfa.gov/webfiles/23056/PrincipalForgivenessltr12312.pdf.
41.	Tim Massad, Expanding our efforts to help more homeowners and strengthen hard-hit communities, Department of the Treasury, January 27, 2012, at http://www.treasury.gov/connect/blog/Pages/Expanding-our-efforts-to-help-more-homeowners-and-strengthen-hard-hit-communities.aspx.
42.	Treasury will subsidize the GSEs for reducing principal on eligible loans that are less than six months delinquent. Treasury will pay 30 cents per dollar of principal forgiven down to 140% LTV, plus 45 cents per dollar of principal forgiven from 140% down to 115% LTV, plus 63 cents per dollar of principal forgiven from 115% down to 105% LTV. See Federal Housing Finance Agency, FHFA Technical Analysis of Principal Forgiveness, July 31, 2012, p. 6, at http://www.fhfa.gov/webfiles/24111/PF_TechApp73112.pdf.
43.	Federal Housing Finance Agency, January 20, 2012, at http://www.fhfa.gov/webfiles/23056/PrincipalForgivenessltr12312.pdf.
44.	U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, Subcommittee on Housing, Transportation and Community Development, Strengthening the Housing Market and Minimizing Losses to Taxpayers, Testimony of Laurie Goodman, Senior Managing Director at Amherst Securities, 112th Cong., 2nd sess., March 15, 2012, available at http://banking.senate.gov/public/index.cfm?FuseAction=Files.View&FileStore_id=0f96e0ff-8500-41a5-a0f2-0139d0df2e07. Henceforth referred to as Testimony of Laurie Goodman, March 15, 2012.
45.	Ibid., p. 8.
46.	Making Home Affordable, Home Affordable Modification Program: Base Net Present Value (NPV) Model v5.0 Model Documentation, p. 26, at https://www.hmpadmin.com/portal/programs/docs/hamp_servicer/npvmodeldocumentationv50.pdf.
47.	Federal Housing Finance Agency, FHFA Technical Analysis of Principal Forgiveness, July 31, 2012, p. 11, at http://www.fhfa.gov/webfiles/24111/PF_TechApp73112.pdf.
48.	Testimony of Laurie Goodman, March 15, 2012.
49.	Federal Housing Finance Agency, FHFA Technical Analysis of Principal Forgiveness, July 31, 2012, p. 8, at http://www.fhfa.gov/webfiles/24111/PF_TechApp73112.pdf.
50.	Testimony of Laurie Goodman, March 15, 2012.
51.	Ibid.
52.	Ibid. Goodman argued that principal reduction is less likely to be NPV positive for loans with mortgage insurance. If the GSEs reduce principal for a borrower they also reduce the amount that the mortgage insurer will compensate the GSEs for in the event of foreclosures, thereby reducing the potential benefits of principal reduction.
53.	Federal Housing Finance Agency, FHFA Technical Analysis of Principal Forgiveness, July 31, 2012, p. 4, at http://www.fhfa.gov/webfiles/24111/PF_TechApp73112.pdf.
54.	Testimony of Laurie Goodman, March 15, 2012.
55.	Analysis 11 adjusts the borrowers' debt-to-income (DTI) ratios to reflect the DTI distribution of loans that received HAMP modifications; Analysis 10 (not shown) sets DTIs to the minimum of 45% or origination DTI capped at 60%. Analysis 10 found lower savings for the GSEs and taxpayers than Analysis 11.
56.	See Federal Housing Finance Agency, FHFA Technical Analysis of Principal Forgiveness, July 31, 2012, p. 15, at http://www.fhfa.gov/webfiles/24111/PF_TechApp73112.pdf.
57.	Federal Housing Finance Agency, FHFA Technical Analysis of Principal Forgiveness, July 31, 2012, p. 16, at http://www.fhfa.gov/webfiles/24111/PF_TechApp73112.pdf.
58.	In addition to FHFA, Fannie Mae and Freddie Mac performed their own analyses of principal reduction. The analyses are not described in this report. For more see Fannie Mae, Fannie Mae's Analysis Regarding Principal Forgiveness and Treasury's HAMP Principal Reduction Alternative (HAMP PRA) Program, July 31, 2012, at http://www.fhfa.gov/webfiles/24107/PF_FannieMae73112.pdf and Freddie Mac, Estimates of Economic and Operational Effects of Increased HAMP Principal Reduction Incentives on Freddie Mac, July 31, 2012, p. 1, at http://www.fhfa.gov/webfiles/24109/PF_FreddieMac73112.pdf.
59.	For an analysis of the legal issues surrounding cramdown and of previous cramdown legislation, see CRS Report RL34301, The Primary Residence Exception: Legislative Proposals in the 111th Congress to Amend the Bankruptcy Code to Allow the Strip Down of Certain Home Mortgages, by [author name scrubbed].
60.	For more on closing costs, see The Federal Reserve Board, A Consumer's Guide to Mortgage Refinancings, at http://www.federalreserve.gov/pubs/refinancings/default.htm#cost.
61.	See http://www.whitehouse.gov/the-press-office/2012/02/01/fact-sheet-president-obama-s-plan-help-responsible-homeowners-and-heal-h.
62.	The cost of the plan would be offset by using a portion of the proposed Financial Crisis Responsibility Fee.
63.	The five banks are Ally/GMAC, Bank of America, Citi, JPMorgan Chase, and Wells Fargo. For more information, see http://www.nationalmortgagesettlement.com/.
64.	Similarly, H.R. 3733, the Affordable Mortgage for Homeowners Act of 2011 (Representative Alcee Hastings et al.) requires Fannie Mae and Freddie Mac to lower the interest rates on GSE loans but does not do so through a refinance.
65.	A refinance involves taking out a new loan to pay off the existing loan. A modification, by contrast, changes the terms of the existing loan.
66.	See Sumit Agarwal, John C. Driscoll, and David Laibson, "Optimal Mortgage Refinancing: A Closed Form Solution," NBER Working Paper Series, October 2007.
67.	If a borrower has private mortgage insurance, the borrower may be able to refinance if the LTV is above 80%.
68.	See Sumit Agarwal, John C. Driscoll, and David Laibson, "Optimal Mortgage Refinancing: A Closed Form Solution," NBER Working Paper Series, October 2007.
69.	Mitchell Remy, Deborah Lucas, and Damien Moore, An Evaluation of Large-Scale Mortgage Refinancing Programs, Congressional Budget Office, Working Paper 2011-4, September 2011, at http://www.cbo.gov/sites/default/files/cbofiles/attachments/09-07-2011-Large-Scale_Refinancing_Program.pdf.
70.	Alan Boyce, Glenn Hubbard, and Chris Mayer et al., Streamlined Refinancings for up to 14 Million Borrowers, January 18, 2012, http://www4.gsb.columbia.edu/null/download?&exclusive=filemgr.download&file_id=739308.
71.	See Testimony of the Honorable Phillip L. Swagel, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, State of the Housing Market: Removing Barriers to Economic Recovery, hearing, 112th Cong., 2nd sess., February 9, 2012.
72.	See Testimony of Dr. Mark Zandi, chief economist of Moody's Analytics, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, State of the Housing Market: Removing Barriers to Economic Recover , hearing, 112th Cong., 2nd sess., February 9, 2012.
73.	See Joseph Tracy and Joshua Wright, Why Mortgage Refinancing Is Not a Zero-Sum Game, Federal Reserve Bank of New York, January 11, 2012, at http://libertystreeteconomics.newyorkfed.org/2012/01/why-mortgage-refinancing-is-not-a-zero-sum-game.html.
74.	For more on the spending decisions across income levels, see Karen E. Dynan, Jonathan Skinner, and Stephen P. Zeldes, Do the Rich Save More?, Federal Reserve Board, November 2000, at http://www.federalreserve.gov/pubs/feds/2000/200052/200052pap.pdf.
75.	Ginnie Mae guarantees mortgage-backed securities backed by federally insured or guaranteed loans. See http://www.ginniemae.gov/about/about.asp?Section=About.
76.	See Joseph Tracy and Joshua Wright, Why Mortgage Refinancing Is Not a Zero-Sum Game, Federal Reserve Bank of New York, January 11, 2012, at http://libertystreeteconomics.newyorkfed.org/2012/01/why-mortgage-refinancing-is-not-a-zero-sum-game.html.
77.	See Testimony of the Honorable Phillip L. Swagel, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, State of the Housing Market: Removing Barriers to Economic Recovery, hearing, 112th Cong., 2nd sess., February 9, 2012.
78.	See Investment Company Institute, Profile of Mutual Fund Shareholders, 2010, February 2011, p. 105, at http://www.ici.org/pdf/rpt_11_profiles.pdf.
79.	For more on HARP, see CRS Report R40210, Preserving Homeownership: Foreclosure Prevention Initiatives, by [author name scrubbed]. HARP is not the only government refinance program. FHA also established the FHA Streamline Refinance Program. However, HARP is the largest of the refinance programs and the subject of several proposals. Therefore it is the focus of this section.
80.	See http://www.makinghomeaffordable.gov/programs/lower-rates/Pages/harp.aspx.
81.	HARP is limited to GSE borrowers to prevent the GSEs from assuming additional credit risk.
82.	See Federal Housing Finance Agency, Foreclosure Prevention and Refinance Report November 2011, at http://www.fhfa.gov/webfiles/23123/Nov_2011_Foreclosure_Prev_Rpt.pdf.
83.	The list of barriers to refinancing is not exhaustive but highlights what some experts have identified as major factors in HARP performing below what was expected by the Administration.
84.	For more on loan level price adjustments, see https://www.efanniemae.com/sf/refmaterials/llpa/pdf/llpamatrixrefi.pdf.
85.	See Kate Berry, "Bank of America Says It Can't Process All Refinance Applications," American Banker, February 8, 2012, at http://www.americanbanker.com/issues/177_27/bank-of-america-harp-refinance-applications-1046498-1.html.
86.	See http://www.fhfa.gov/webfiles/22721/HARP_release_102411_Final.pdf.
87.	However, as mentioned previously, most HARP participants had an LTV below 105 and would therefore be unaffected by eliminating the cap.
88.	FHFA has filed lawsuits against at least 17 lenders in cases related to put-back claims, which are lawsuits related to potential violations of representations and warranties or other underwriting violations. See http://www.fhfa.gov/Default.aspx?Page=110.
89.	The waiving of certain representations and warranties applies only to refinances through the same servicer and not through different servicers. See Amherst Securities Group LP, HARP: Program Changes and Their Implications, October 24, 2011.
90.	The President's plan also includes other components such as a Homeowner Bill of Rights and expanding forbearance for the unemployed, but this analysis focuses on the refinancing proposals. See http://www.whitehouse.gov/the-press-office/2012/02/01/fact-sheet-president-obama-s-plan-help-responsible-homeowners-and-heal-h.
91.	To find the FHA Mortgage Limits in a particular area, see https://entp.hud.gov/idapp/html/hicostlook.cfm.
92.	To encourage borrowers to refinance into shorter-term mortgages, FHFA eliminated certain risk-based fees for borrowers who chose that option under HARP 2.0. See http://www.fhfa.gov/webfiles/22721/HARP_release_102411_Final.pdf.
93.	For more on household formation, see Robert Denk and David Crowe, "Pent-Up Housing Demand: The Household Formation That Didn't Happen - Yet," February 2, 2011, at http://www.nahb.org/generic.aspx?genericContentID=152243&channelID=311.
94.	The Federal Deposit Insurance Corporation (FDIC) reports the dollar value of REO and not the number of properties. See http://www2.fdic.gov/qbp/index.asp, and Lewis S. Ranieri, Kenneth T. Rosen, and Andrea Lepico et al., Options for REO: The Private Sector Solution to the Foreclosure Problem, Rosen Consulting Group and Ranieri Partners Management LLC, February 2012.
95.	See Fannie's Form 10-K at http://fanniemae.com/resources/file/ir/pdf/quarterly-annual-results/2011/10k_2011.pdf and Freddie's Form 10-K at http://www.freddiemac.com/investors/er/pdf/10k_030912.pdf.
96.	See Office of Risk Analysis and Regulatory Affairs, Federal Housing Administration, Department of Housing and Urban Development, Monthly Report to the FHA Commissioner on FHA Business Activity, January 2012, at http://portal.hud.gov/hudportal/documents/huddoc?id=12jan.pdf.
97.	See Lewis S. Ranieri, Kenneth T. Rosen, and Andrea Lepico et al., Options for REO: The Private Sector Solution to the Foreclosure Problem, Rosen Consulting Group and Ranieri Partners Management LLC, February 2012.
98.	See CRS Report R41491, "Robo-Signing" and Other Alleged Documentation Problems in Judicial and Nonjudicial Foreclosure Processes, by [author name scrubbed].
99.	See Prashant Gopal and John Gittelsohn, "Foreclosure Deal to Spur New Wave of U.S. Home Seizures, Help Heal Market," Bloomberg, February 10, 2012, at http://www.bloomberg.com/news/2012-02-09/foreclosure-deal-to-spur-new-wave-of-u-s-home-seizures-help-heal-market.html.
100.	William C. Dudley, Housing and the Economic Recovery, Federal Reserve Bank of New York, January 6, 2012, at http://www.newyorkfed.org/newsevents/speeches/2012/dud120106.html.
101.	The Federal Reserve Board of Governors, The U.S. Housing Market: Current Conditions and Policy Considerations, white paper, January 4, 2012, at http://federalreserve.gov/publications/other-reports/files/housing-white-paper-20120104.pdf.
102.	For FDIC guidance to financial institutions on their handling of REO, see http://www.fdic.gov/news/news/financial/2008/fil08062a.html. Also, see 12 U.S.C. § 29.
103.	The Federal Reserve Board of Governors, The U.S. Housing Market: Current Conditions and Policy Considerations, white paper, at January 4, 2012, http://federalreserve.gov/publications/other-reports/files/housing-white-paper-20120104.pdf.
104.	See http://www.federalreserve.gov/newsevents/press/bcreg/bcreg20120405a1.pdf.
105.	For more on REO disposition issues, see Testimony of Laurie S. Goodman, a senior managing director at Amherst Securities Group, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, Subcommittee on Housing, Transportation and Community Development, Strengthening the Housing Market and Minimizing Losses to Taxpayers, hearing, 112th Cong., 2nd sess., March 15, 2012. Available at http://banking.senate.gov/.
106.	Testimony of Laurie S. Goodman, a senior managing director at Amherst Securities Group, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, Subcommittee on Housing, Transportation and Community Development, Strengthening the Housing Market and Minimizing Losses to Taxpayers, hearing, 112th Cong., 2nd sess., March 15, 2012, available at http://banking.senate.gov/.
107.	Testimony of Ms. Laurie F. Goodman, a senior managing director at Amherst Securities Group, in U.S. Congress, Senate Committee on Banking, Housing, and Urban Affairs, Subcommittee on Housing, Transportation and Community Development, New Ideas to Address the Glut of Foreclosed Properties, hearing, 112th Cong., 1st sess., September 20, 2011, available at http://banking.senate.gov.
108.	RealtyTrac estimates that the average price of a foreclosure-related sale is approximately 29% less than a non-foreclosure sale. See http://www.realtytrac.com/content/foreclosure-market-report/q4-and-year-end-2011-us-foreclosure-sales-report-7060.
109.	Alex Phillips and Hui Shan, US Daily: Thoughts on a Federal REO-to-Rental Program, Goldman Sachs Research, January 5, 2012.
110.	The Federal Reserve Board of Governors, The U.S. Housing Market: Current Conditions and Policy Considerations, white paper, January 4, 2012, p. 9, at http://federalreserve.gov/publications/other-reports/files/housing-white-paper-20120104.pdf.
111.	See http://www.fhfa.gov/webfiles/22366/RFIFinal081011.pdf.
112.	See http://www.fhfa.gov/webfiles/23403/REOPR22712F.pdf.
113.	Kerri Panchuk, "Bulk REO pilot plan disappoints: Capital Economics," HousingWire, at http://www.housingwire.com/article/bulk-reo-pilot-plan-disappoints-capital-economics.
114.	Suzy Khimm, "Can we fix housing by turning foreclosures into rentals?," The Washington Post, at http://www.washingtonpost.com/blogs/ezra-klein/post/can-we-fix-housing-by-turning-foreclosures-into-rentals/2012/02/28/gIQAENTsgR_blog.html.
115.	HUD was also part of the initial Request for Information (RFI) on REO sales in August 2011. It is unclear if HUD will convert FHA REO properties into rentals in a manner similar to FHFA.
116.	For more on depository institutions and their holding companies, see CRS Report R41339, The Dodd-Frank Wall Street Reform and Consumer Protection Act: Titles III and VI, Regulation of Depository Institutions and Depository Institution Holding Companies, by [author name scrubbed].