Contents

• Introduction
• Overview of the Ability-to-Repay and Qualified Mortgage Rule
• Economics of the ATR Rule
• An Economic Analysis of the Small Creditor Portfolio QM and Legislative Proposals
• 1. Portfolio Requirements
• Retained Risk and Aligning Incentives
• Three-Year Requirement
• Legislative Proposals and Policy Discussion
• 2. Lender Restrictions
• Small Creditors as a Key Source of Credit?
• Small Creditors' Sources of "Strong Incentives and Particular Ability"?
• Legislative Proposals and Policy Discussion
• 3. Loan Criteria
• Legislative Proposals and Policy Discussion
• Access to Credit and Regulatory Burden
• Concluding Thoughts

Summary

Title XIV of the Dodd-Frank Act established the ability-to-repay (ATR) requirement. Under the ATR requirement, a lender must determine based on documented and verified information that, at the time a mortgage is made, the borrower has the ability to repay the loan. Lenders that fail to comply with the ATR rule could be subject to legal liability, such as the payment of certain statutory damages.

A lender can comply with the ATR requirement in different ways, one of which is by originating a Qualified Mortgage (QM). When a lender originates a QM, it is presumed to have complied with the ATR requirement, which consequently reduces the lender's potential legal liability for its residential mortgage lending activities.

The QM rule has several compliance options that a lender can use to have a mortgage that it originates receive QM status, one of which is the Small Creditor Portfolio QM. Critics of the QM rule argue that the Small Creditor Portfolio QM, which is intended to benefit small creditors that keep loans in portfolio, should be broadened. They propose modifying the rule to make it easier for a lender to comply if it keeps the loan instead of selling it.

Proponents of an expanded portfolio QM disagree on several policy issues, but, in general, they argue that because the lender is holding the loan in its portfolio, it is exposed to the risks associated with the loan (such as the risk that the lender will not be repaid) and therefore has the incentive to ensure that the loan is safely underwritten. The lender, the argument goes, should be allowed to follow less prescriptive underwriting criteria when the mortgage is held in portfolio, and more lenders should be allowed to avail themselves of this option than is currently allowed. Critics of the expanded portfolio lending proposals counter that the incentives of holding the loan in portfolio are insufficient to protect consumers and that the existing protections in the rule are needed to ensure that the failures of the past are not repeated.

This report analyzes the policy debate related to portfolio lending and qualified mortgages, focusing on the legislation that is the subject of congressional debate: H.R. 1210, the Portfolio Lending and Mortgage Access Act; S. 1484, the Financial Regulatory Improvement Act; and S. 1491/H.R. 2642, the Community Lender Regulatory Relief and Consumer Protection Act.

The analysis in this report raises several issues:

• Economic theory supports different arguments that are made about whether a mortgage that is kept in portfolio is more or less likely to be prudently underwritten. The retained risk associated with keeping a mortgage in portfolio may provide the lender an incentive to ensure that the borrower is creditworthy, but increasing home prices and issues related to the arbitrage of capital requirements may provide incentives to keep mortgages in portfolio that may be profitable but not prudently underwritten. Empirical research has also led to conflicting results.
• The Consumer Financial Protection Bureau argues that small creditors may have "strong incentives and particular ability" to make mortgages that accurately assess a borrower's ability to repay that larger lenders may not have. It is unclear, however, what the appropriate size thresholds should be and whether other factors, such as keeping the mortgage in portfolio, can compensate for a larger lender's possibly reduced incentives and ability.
• It is unclear how significant an effect an expanded portfolio QM option would have on credit availability, as the mortgages that would qualify for the expanded option may already receive QM status under existing options. The legislative proposals may, however, have a greater effect on reducing a creditor's regulatory burden, as the lender may be able to use less cost-intensive underwriting processes of the expanded QM option. Although the burden may be reduced for the lender, it could be borne by consumers if they have fewer consumer protections under an expanded portfolio QM.

---

An Analysis of Portfolio Lending and Qualified Mortgages

Introduction

After the bursting of the housing bubble and the ensuing financial crisis, Congress passed the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act; P.L. 111-203), a broad package of reforms that touched on many aspects of the financial system. It was intended to address the problems that led to turmoil and to ensure that the financial system and the economy are better positioned to withstand future market disruptions.

While some view the Dodd-Frank Act as essential to ensure that consumers are not abused and to promote stability in the financial system, others believe the reforms have imposed an unnecessary burden on lenders—especially small lenders¹—and have made it more difficult for some consumers to receive loans, particularly mortgages. Critics have highlighted the Ability-to-Repay (ATR) and Qualified Mortgage (QM) rule, which implements a Dodd-Frank Act requirement, as a subject of particular concern. Among other things, they argue that the Small Creditor Portfolio QM, a compliance option under the rule that is intended to benefit small creditors that keep loans in portfolio, should be broadened. They propose modifying the rule to make it easier for a lender to comply if it keeps the loan instead of selling it to another institution.

Supporters of expanded portfolio lending proposals argue that because the lender is holding the loan in its portfolio, it is exposed to the risks associated with the loan (such as the risk that the lender will not be repaid) and therefore has the incentive to ensure that the loan is safely underwritten. Critics of the expanded portfolio lending proposals argue that the incentives of holding the loan in portfolio are insufficient to protect consumers and that the existing protections in the rule are essential to ensuring that the failures of the past are not repeated.

This report will briefly explain the ATR rule and will analyze the policy debate related to portfolio lending and qualified mortgages.

Overview of the Ability-to-Repay and Qualified Mortgage Rule

Title XIV of the Dodd-Frank Act established the ATR requirement. Under the ATR requirement, a lender must determine based on documented and verified information that, at the time a mortgage loan is made, the borrower has the ability to repay the loan. A lender must consider and verify certain types of information prior to originating a loan, including the applicant's income or assets, credit history, outstanding debts, and other criteria. Lenders that fail to comply could be subject to legal liability, such as the payment of certain statutory damages.²

Under the Consumer Financial Protection Bureau (CFPB) rule implementing the ATR requirement, a lender can comply with the ATR requirement in two ways. First, a lender can originate a mortgage that meets the relatively less prescriptive underwriting standards of the General ATR Option. Under the General ATR Option, the loan does not need to satisfy specified criteria, but the lender must not make a mortgage "unless the creditor makes a reasonable and good faith determination at or before consummation that the consumer will have a reasonable ability to repay the loan according to its terms."³

Second, a lender could originate a mortgage that satisfies the more prescriptive standards of the QM rule. A QM is more prescriptive than the General ATR Option in the sense that, to receive QM status, the mortgage must satisfy specific underwriting and product-feature requirements and document compliance with the requirements. When a lender originates a mortgage that receives QM status, it is presumed to have complied with the ATR requirement, which consequently reduces the lender's potential legal liability of its residential mortgage lending activities.⁴ A lender can comply with the ATR rule by making a mortgage that is not a QM and instead satisfies the General ATR Option, but the lender will not receive the additional legal protections. The definition of a QM, therefore, is important to a lender seeking to minimize its legal risk of its residential mortgage lending activities, specifically its compliance with the statutory ATR requirement. Some are concerned that, at least in the short term, few mortgages will be originated that do not meet the QM standards due to the legal protections that QMs afford lenders, even though there are other means of complying with the ATR requirement.⁵

Underwriting and product-feature requirements of the QM categories are intended to ensure that a mortgage receiving QM status satisfies certain minimum standards, with the standards intended to offer protections to borrowers. A loan that satisfies the less strict General ATR Option, in contrast, is allowed to have a balloon payment, a term in excess of 30 years, and other features that are generally prohibited in a QM so long as the lender verifies that the borrower would have the ability to repay the loan.

Economics of the ATR Rule

In the preamble to the final rule, the CFPB explained that the ATR requirement is intended to address certain market failures that policymakers identified in the housing finance system.⁶ Some attribute the expansion in mortgage credit that fueled the housing bubble to market failures that encouraged lenders to offer and borrowers to accept more credit than was socially optimal.

Market failures can occur in many instances in the housing market, such as lenders and borrowers not accounting for societal costs outside of a transaction (such as foreclosure)⁷ and borrowers' difficulty in understanding mortgage terms and accurately assessing their ability to repay given that consumers typically have limited experience buying and financing a house.⁸ According to the CFPB, the "Dodd-Frank Act and the final rule address these potential market failures through minimum underwriting requirements at origination and new liability for originators ... in cases where the standards are found not to be met."⁹ The minimum standards and the new liability are intended to encourage lenders to account for costs in addition to those that they bear when extending credit and to make more informed underwriting decisions, potentially reducing market failures and resulting in a more socially optimal amount of credit being allocated.

The underwriting requirements and new liability in the ATR rule are viewed as working in tandem to encourage the lender to determine that the borrower will be able to repay the loan. If the lender originates a loan that satisfies the minimum underwriting standards (the General ATR Option), then the lender is exposed to a certain amount of legal liability. But if the lender originates a mortgage that meets more prescriptive requirements and is therefore deemed a QM, then the lender is exposed to less legal risk. The reduced legal risk is warranted, the CFPB argues, because the QM's minimum underwriting and product-feature standards "are judged in the rule to be enough to ensure that the lender made a reasonable and good faith determination that the borrower will be able to repay the loan."¹⁰ The CFPB notes that lenders, therefore, face a tradeoff between (1) "exert[ing] greater care in underwriting the loan" to ensure that it receives QM status and the reduced liability associated with it or (2) performing a potentially less thorough and less costly initial underwriting of the General ATR Option and having a greater exposure to liability.¹¹

In summary, the additional liability and the minimum underwriting standards enacted as part of the Dodd-Frank Act are intended to address market failures that some policymakers believed fueled the housing bubble that precipitated the financial crisis. As discussed in more detail later in this report, some argue that the QM rule is one size fits all, has led to an unnecessary constriction of credit, and has been unduly burdensome on lenders. For a more detailed overview of the ATR rule and QM compliance options, see CRS Report R43081, The Ability-to-Repay Rule: Possible Effects of the Qualified Mortgage Definition on Credit Availability and Other Selected Issues, by [author name scrubbed].

An Economic Analysis of the Small Creditor Portfolio QM and Legislative Proposals

The Dodd-Frank Act defined a qualified mortgage¹² but also provided the CFPB the authority to "prescribe regulations that revise, add to, or subtract from the criteria that define a qualified mortgage" under certain circumstances.¹³ The CFPB-issued QM regulations establish a Standard QM that meets all of the underwriting and product-feature requirements outlined in the Dodd-Frank Act. However, the QM regulations also establish several additional categories of QM that provide lenders the same presumption of compliance legal protections with the ATR requirement as a Standard QM. One of those additional categories is the Small Creditor Portfolio QM. The CFPB argues that the Small Creditor Portfolio QM is "necessary to preserve access to responsible, affordable mortgage credit for some consumers."¹⁴ It believes "that portfolio loans made by small creditors are particularly likely to be made responsibly and to be affordable for the consumer."¹⁵

To receive QM status under the Small Creditor Portfolio QM, three broad sets of criteria must be satisfied. First, the loan must be held in the originating lender's portfolio for at least three years (subject to several exceptions).¹⁶ Second, the loan must be held by a small creditor, which is defined as a lender who originated 2,000 or fewer mortgages in the previous year (excluding portfolio loans) and has less than $2 billion in assets.¹⁷ Third, the loan must meet certain underwriting and product-feature requirements.¹⁸

Compared with the Standard QM, the Small Creditor Portfolio QM has less prescriptive underwriting requirements and is intended to reduce the regulatory burden of the ATR rule for small creditors. To receive QM status under the Standard QM, a borrower must have a debt-to-income (DTI) ratio below 43% after accounting for the payments associated with the mortgage and other debt obligations. Under the Small Creditor Portfolio QM, the lender is required to consider and verify the borrower's DTI but does not have a specific numeric threshold that the borrower must be below. A lender, therefore, has the flexibility to offer a loan with a higher DTI ratio if the lender determines that the borrower is creditworthy. Because the DTI threshold is considered by some¹⁹ to be one of the factors limiting credit availability under the QM rule, relief from that threshold allows small creditors to originate mortgages that they may not otherwise have made. Though relief is provided from the DTI threshold for the Small Creditor Portfolio QM, the other underwriting and product-feature requirements for the Standard QM must still be followed.

Although the Small Creditor Portfolio QM has less prescriptive underwriting requirements than the Standard QM, the Small Creditor Portfolio QM requires a lender to hold the loan in portfolio and limits the portfolio QM option to small creditors. The CFPB justified establishing the Small Creditor Portfolio QM on the basis that even though the underwriting standards have been reduced (potentially making it easier for some borrowers to qualify), certain factors unique to portfolio loans and to small creditors provide added incentives that ensure that a lender using the Small Creditor Portfolio QM will accurately assess whether a borrower can repay the mortgage. The CFPB believes both the portfolio requirement and small creditor limitation are necessary to justify the less prescriptive underwriting standards.

Supporters of an expanded portfolio lending option contend that not all of the lender and underwriting requirements included in the Small Creditor Portfolio QM are essential to ensuring that a lender will extend credit to creditworthy borrowers and argue that a less burdensome approach can be employed. Some in Congress propose establishing an additional portfolio QM option that would have more relaxed eligibility criteria for the three major categories that serve as a framework for discussion—portfolio requirements, lender restrictions, and loan criteria. This report analyzes three legislative proposals that would establish alternative portfolio QM options.

• 1. S. 1484, the Financial Regulatory Improvement Act of 2015, was reported by the Senate Banking Committee on June 2, 2015.²⁰
• 2. S. 1491/H.R. 2642, the Community Lender Regulatory Relief and Consumer Protection Act of 2015, have been characterized as an alternative to S. 1484 because the text was initially offered unsuccessfully as an amendment to S. 1484 during the Senate Banking Committee markup.²¹
• 3. H.R. 1210, the Portfolio Lending and Mortgage Access Act, was passed by the House on November 18, 2015.

Table 1 summarizes these proposals and compares them to the Small Creditor Portfolio QM using the three-pronged framework.

The next three sections assess the three primary factors on which the legislative proposals differ compared with the status quo—portfolio requirements, lender restrictions, and loan criteria. Each section first explains how the CFPB's Small Creditor Portfolio QM addresses that factor, and then it assesses how the legislative proposals would address the issue.

1. Portfolio Requirements

The Small Creditor Portfolio QM requires a small creditor to retain the mortgage in its portfolio for three years in order for the mortgage to receive QM status. If the lender sells the mortgage before three years, the mortgage loses its status as QM, with some exceptions. The mortgage would retain its QM status if it is sold to another lender who meets the definition of small creditor. The mortgage would also maintain its QM status if it is acquired as a result of a merger of a lender with another lender or if the mortgage is sold because the sale is determined to be necessary by a regulator to revive or resolve a troubled lender.²²

Retained Risk and Aligning Incentives

By keeping the loan in portfolio, the CFPB argues, creditors have added incentive to consider whether the borrower will be able to repay the loan.²³ Keeping the loan in portfolio means that the lender retains the default risk and could be exposed to losses if the borrower does not repay. This exposure, the argument goes, would encourage creditors to provide additional scrutiny during the underwriting process, even in the absence of a legal requirement to do so. Keeping the mortgage in portfolio is intended to align "consumers' and creditors' interests regarding ability to repay."²⁴ If the lender sold the mortgage immediately after origination, as is done under an "originate-to-distribute" business model, this incentive may not exist.²⁵ According to the CFPB, the added incentive to do thorough underwriting that comes with holding the loan in portfolio is necessary to justify the less prescriptive underwriting standards of the Small Creditor Portfolio QM.

Although retaining risk may bring creditors' and consumers' interests more in line, they may not be perfectly aligned. Lenders have the incentive to be repaid and consumers have the incentive to repay, but those two things are different. Even with the loan in portfolio, the lender may have the incentive to lend to borrowers who do not have sufficient liquid sources of repayment, such as an applicant's income and money in a checking account. The ATR requirement mandates a lender focus on a borrower's liquid sources of repayment in assessing a borrower's ability to repay. When performing its underwriting, the lender typically takes a broad view of the borrower's ability to repay, considering not only liquid sources of repayment but also the value of the house that is being purchased.²⁶ Even though the home is an asset to the borrower, it is not typically a liquid source of repayment.²⁷ The lender considers the value of the house during its underwriting because, in a mortgage, the house serves as the underlying collateral. So long as house prices are rising, the house can be sold to repay the lender the amount that it is owed if the borrower does not repay, allowing the lender to profit from the initial fees it is paid for originating the loan and then recouping its principal through the home sale, a process often referred to as collateral-dependent lending. In this instance, holding the loan in portfolio may not align the lender's and borrower's interests.

As a response to collateral-dependent lending, the ATR rule requires the lender to consider and verify the borrower's ability to repay independent of the underlying value of the house. Lenders are required to focus on liquid sources of repayment as part of compliance with the rule. Although some of the underwriting requirements are less prescriptive for the Small Creditor Portfolio QM than the Standard QM, the lender is still required to assess the borrower's ability to repay independent of the underlying value of the house even though the mortgage is held in portfolio. Lenders consider the value of the house and other illiquid assets when performing their underwriting and use it to inform their decisions about whether to extend credit, but they assess a borrower's ability to repay in accordance with the ATR rule using liquid sources of repayment.

Three-Year Requirement

As mentioned previously, the Small Creditor Portfolio QM requires a small lender to retain the mortgage in its portfolio for three years in order for the mortgage to receive QM status, subject to some exceptions. If the loan is sold before three years to a lender that does not meet the small creditor definition, then the mortgage would lose its QM status, but if it is sold to anyone after three years, then the mortgage retains its QM status. One of the reasons that the CFPB cites for why the lender must keep the loan in portfolio for three years is to prevent evasion.²⁸ If a small creditor could originate a mortgage and was required to keep it in portfolio for only a short period of time prior to selling it, then the benefits associated with keeping it in portfolio would be reduced.

The CFPB also chose three years for the length requirement in order "to conform to the statute of limitations for affirmative claims for violations of the ability-to-repay rules."²⁹ If a borrower is going to initiate an affirmative claim against a lender under the ATR rule, the borrower must do so during the first three years. The borrower may cite the ATR at any point, however, as a defense against foreclosure.³⁰ By holding the mortgage in portfolio for three years, the CFPB argues, the small lender "retains all of the litigation risk for potential violations of the ability-to-repay rules except in the event of a subsequent foreclosure."³¹ As mentioned before, the possibility of litigation is one of the tools to address market failures in the mortgage market and is intended to encourage lenders to account for not only their own costs but also the costs borne by others when extending credit.

Legislative Proposals and Policy Discussion

Similar to the Small Creditor Portfolio QM, all three proposals would require the mortgage to be held in a lender's portfolio, at least initially. They differ, however, in whether the mortgage retains its QM status if it is transferred from the originating lender's portfolio and whether the mortgage must be in a lender's portfolio (either the originator's or any lender's portfolio) indefinitely or for a limited period of time. Table 2 summarizes the differences. These policy issues, as well as several other issues related to lenders' portfolios, are discussed below.

Originator's Portfolio and Three Years

The Small Creditor Portfolio QM allows (and S. 1484 would allow) a mortgage to be transferred from the originator's portfolio to another eligible entity's portfolio and retain QM status. S. 1491 and H.R. 1210, however, would require the mortgage to stay in the originator's portfolio; if it was transferred then it would lose its QM status (with some exceptions). As described below, the differences in the portfolio requirements can affect the borrower's cost of credit and the lender's incentive to perform thorough underwriting.

A mortgage that can be transferred to another entity while retaining its QM status is, generally speaking, more liquid than a mortgage that would lose its QM status when transferred. (Liquid in this context refers to the ease with which a mortgage can be bought and sold.) If the mortgage cannot be transferred without losing its QM status, the originator may charge a higher rate to a borrower to account for this lack of liquidity and for the risks associated with the lack of liquidity. This liquidity premium would potentially vary based on the severity of the restrictions, with more restrictive portfolio requirements yielding a higher liquidity premium. The Small Creditor Portfolio QM, for example, would likely face a liquidity premium because, in the first three years, the mortgage could be transferred to another small creditor only to retain QM status.³² The restriction on transferring to only small creditors may make the liquidity premium higher for the Small Creditor Portfolio QM compared to S. 1484, because S. 1484 would allow the mortgage to retain its QM status while being transferred to almost any lender. The liquidity premium could also be lower for the Small Creditor Portfolio QM, however, because after three years the mortgage would not need to be kept in portfolio to retain QM status, making it more liquid than would be the case under S. 1484. In sum, more stringent portfolio requirements may make the mortgage more illiquid and could result in a higher rate that would be expected to be charged to the borrower.

While more relaxed portfolio requirements could reduce the liquidity premium associated with the mortgage, more relaxed standards could also reduce the lender's incentive to thoroughly underwrite the mortgage. If the mortgage can be transferred without losing its QM status, there may be a weaker incentive for the originator to do thorough underwriting, because the originator knows that the mortgage can be sold, transferring both the credit risk and the liability risk to a new entity. (The liability risk also transfers to the new mortgage holder when the mortgage is sold.³³) The market, however, could respond by developing legal contracts such that, when the mortgage is transferred, the new mortgage holder could have the ability to "put back" the risk on to the originator if certain specified conditions are met.³⁴ In that case, the originator would still be exposed to the liability risk even if the originator did not hold the mortgage at the time the claim was brought by the borrower.³⁵ The possibility that the originator could be exposed to liability even if the mortgage is sold could provide added incentive to thoroughly assess the borrower's ability to repay even if the originator does not intend to keep the mortgage in portfolio. This example serves to illustrate that while the legislative proposals may have one set of portfolio requirements, it is possible that the market would evolve to reach an alternative outcome.

Evidence of Portfolio Lending's Effect on Ability to Repay

The CFPB and supporters of the legislative proposals often cite economic theory—the importance of retained risk and aligned incentives—to justify a portfolio QM.³⁶ Empirical research, however, has found conflicting results about whether loans held in portfolio are less likely to default than loans that are securitized. Researchers have suggested alternative theoretical frameworks to explain their results.

Several studies have found that loans held in portfolio were, in general, less likely to default than comparable loans that had been securitized. One study found that prime mortgages that had been securitized in private-label securities (PLS) were more likely to default than "observably similar" mortgages retained in a lender's portfolio.³⁷ Mortgages securitized by Fannie Mae and Freddie Mac, however, were less likely to default than portfolio loans, though the effect was found to be small. The same study, when analyzing subprime loans, found that subprime loans that were securitized in PLS were less likely to default than portfolio loans, though the author attributes the result to early defaults, loans that were "so risky that it may not have been possible to securitize them before they defaulted, and thus, they end up being overrepresented in lender portfolios."³⁸ The author argues that his results, especially for prime mortgages, provides evidence that lenders engage in adverse selection, that they "used private information to determine which loans to securitize" and chose to keep those that were less likely to default in their portfolios.³⁹

Other research, however, found that some types of securitized loans had lower default rates than comparable loans held in portfolio. A 2012 study found that, in the prime mortgage market, lenders "chose to sell low-default-risk (not high-default-risk) and high-prepayment-risk loans to the secondary market in the pre-crisis period, 2004-2006."⁴⁰ During this period default rates were relatively low, so lenders may not have been as concerned about default risk as they were about prepayment risk—the risk that borrowers will repay their mortgages before the full term. Keeping lower prepayment-risk loans and selling higher default-risk loans "would have been a profitable strategy [during 2004-2006] when prepayment risk driven by high refinancing activity was a bigger concern for lenders than default risk in the prime market."⁴¹ In addition to concerns about prepayment risk, lenders may have chosen to keep higher-default-rate mortgages in their portfolio because of regulatory capital requirements, which may give banks "incentives to retain riskier mortgage loans with higher expected return and securitize less risky mortgage loans, as long as both groups of loans have the same capital requirements when held on banks' balance-sheets."⁴² This is an example of regulatory arbitrage, a shift in activities in an attempt to circumvent regulation.

In the subprime market, the authors did not "find a significant difference in the default and prepayment risks between portfolio and securitized loans."⁴³ The authors also found that small lenders behaved differently than larger lenders during the period they studied. During certain years, small lenders were more likely to sell higher default risk loans than larger lenders. The authors attribute this to two different reasons. First, they argue that "small lenders are more likely to be small-town lenders who have private information about their borrowers to adversely select loans for securitization."⁴⁴ Second, "small lenders are less likely to retain higher-default-risk loans even if the loans have lower prepayment probability because of their less diversified borrower base and their small cost of gaining a damaged reputation in the secondary market."⁴⁵

In summary, different studies have found different results as to whether loans retained in portfolio are likely to have a higher or lower default rate. Among other things, the results are sensitive to the time period analyzed, whether the loans are prime or subprime, and whether the lenders are large or small. Although the CFPB and supporters of the legislative proposals emphasize the importance of retained risk to encourage lenders to originate lower default risk mortgages, the empirical research presents conflicting evidence.

Effect on Other Business Models

A less prescriptive portfolio QM option could affect the choice of business models that lenders use to fund mortgages in the mortgage market. Lenders may be more willing to retain mortgages they originate (rather than sell them to other market participants) if the portfolio QM standards are less prescriptive. As shown in Figure 1, most mortgages originated in the past 15 years have not been held in portfolio; in the first two quarters of 2015, approximately 26% of mortgage originations were held in portfolio.

Figure 1. Mortgage Funding Channels

Source: Urban Institute, Housing Finance Policy Center, Housing Finance at a Glance, November 2015, p. 8.

An expanded portfolio QM could disadvantage mortgage bankers who follow a business model in which they do not hold the mortgages in portfolio. Some lenders, for example, engage in correspondent lending: the practice in which a lender originates a mortgage but sells it to an investor based on a prior agreement.⁴⁶ A correspondent lender does not hold the mortgage in its portfolio. A correspondent lender could face higher funding costs if a mortgage that it originates does not receive QM status but would receive QM status if held in portfolio.

Favoring portfolio lending over other business models could be one of the intended policy goals of an expanded portfolio QM because of the belief that portfolio loans better align lender and borrower incentives. As discussed below, more loans held in portfolio, however, could result in a further concentration of risk in the banking system.

Safety and Soundness Concerns

An expanded portfolio QM could encourage lenders to hold even more mortgages on their portfolios, a result that could be in conflict with regulators' attempts to ensure that lenders have diversified asset portfolios.⁴⁷ As mentioned previously, when a bank holds a mortgage in portfolio, it is also holding the risks associated with the mortgage. If mortgage default rates increase or interest rates change in unanticipated ways, the lender could suffer losses.

In addition, if a mortgage must be held in portfolio to retain QM status, it could be difficult for a lender to sell certain mortgages in the event of a downturn, which could have negative safety and soundness consequences. As the CFPB notes, "limitations on the ability of a creditor to sell loans in its portfolio may limit the creditor's ability to manage its regulatory capital levels by adjusting the value of its assets, may affect the creditor's ability to manage interest rate risk by preventing sales of seasoned loans, and may present other safety and soundness concerns."⁴⁸

The portfolio requirement could also result in a change in the relative price of mortgage and non-mortgage assets, which could be a problem during market downturns. If a bank is facing financial distress and is attempting to sell assets to satisfy its regulatory requirements, it could respond to the mortgage portfolio requirement by selling higher quality non-mortgage assets that are more liquid because they do not face the same portfolio requirements. The bank may prefer to sell the mortgage assets rather than the non-mortgage assets but could have a difficult time doing so because of the requirement that the mortgage be held in portfolio. Selling the higher quality non-mortgage assets could put the already distressed bank in a worse financial condition than it would be in if it could sell the mortgage assets.

In considering the Small Creditor Portfolio QM, the CFPB consulted with banking regulators on the safety and soundness concerns.⁴⁹ To address some of the potential issues, the CFPB decided to allow a mortgage to retain QM status if the mortgage is sold because a regulator determined the sale to be necessary to revive or resolve a troubled lender. S. 1484 and H.R. 1210 have similar provisions. S. 1484 would also require lenders that would be deemed systemically important banks⁵⁰ to have their mortgage portfolios subject to periodic review by bank regulators if their mortgage portfolios had elevated risk or an increase in delinquencies or if other criteria were met.

2. Lender Restrictions

A lender must meet the CFPB's definition of small creditor for a mortgage to receive QM status under the Small Creditor Portfolio QM. A lender is considered small if it originated 2,000 or fewer mortgages in the previous calendar year (excluding those held in portfolio) and has total assets of less than $2 billion (with the threshold to be adjusted for inflation).⁵¹

The CFPB limited the portfolio QM option to portfolio loans held by small lenders rather than to all portfolio loans because it

believes that small creditors are a unique and important source of non-conforming mortgage credit and mortgage credit in rural areas for which there is no readily available replacement, that small creditors may be particularly burdened by the litigation risk associated with the ability-to-repay rules and are particularly likely to reduce or cease mortgage lending if subjected to these rules without accommodation, and that small creditors have both strong incentives and particular ability to make these loans in a way that ensures that consumers are able to repay that may not be present for larger creditors.⁵²

This section assesses the CFPB's argument, starting with the assertion that small lenders are an important source of credit.

Small Creditors as a Key Source of Credit?

The Small Creditor Portfolio QM is just one of several types of qualified mortgage compliance options. An understanding of how it relates to the other QM options is important to understanding why it was established by the CFPB.

If a mortgage does not satisfy the Standard QM, it can also receive QM status through the government-sponsored enterprise (GSE) QM if it is eligible to be purchased by Fannie Mae or Freddie Mac, two GSEs in the housing finance system.⁵³ Mortgages that are eligible to be purchased by a GSE are called conforming mortgages because they must conform to specific standards established by law through Fannie's and Freddie's statutory charters, as well as their implementing guidance in order to be eligible to be purchased by a GSE. Some of the underwriting criteria for conforming mortgages are broader than under the Standard QM (the acceptable DTI is higher for the GSEs, for example), allowing additional mortgages to receive QM status that would not qualify through the Standard QM option. A mortgage, regardless of whether or not it is held in portfolio, could receive QM status through either the Standard QM or the GSE QM.

The CFPB argues that the Small Creditor Portfolio QM is necessary because small lenders are an important source of non-conforming credit, meaning mortgages that are ineligible to be purchased by the GSEs. The Small Creditor Portfolio QM can, therefore, be viewed as a complement to the existing compliance options. Under this line of argument, larger lenders are less likely to need an additional compliance option because they generally originate conforming mortgages that already receive QM status under the GSE QM. According to the CFPB, larger lenders are less likely to originate nonconforming loans because non-conforming loans

involve consumers or properties with unique features that make them difficult to assess using larger creditors' underwriting standards or because larger creditors are unwilling to hold the loans in portfolio. Small creditors often are willing and able to consider these consumers and properties individually and to hold the loans on their balance sheets.⁵⁴

Data cited by the CFPB in its 2013 final rule, however, presents a less clear picture. The CFPB found that "approximately one half of all nonconforming loans are originated by creditors with assets less than $2 billion and approximately one quarter are originated by creditors with total assets less than $2 billion that originate fewer than 500 first-lien mortgages annually."⁵⁵ (The CFPB's original definition used a 500 mortgage threshold, but the threshold was raised to 2,000 mortgages in the amended final rule.⁵⁶) Under the original definition of small lender, three-quarters of non-conforming loans, therefore, were originated by lenders that are not small creditors and would have been unable to take advantage of the Small Creditor Portfolio QM.

In addition to non-conforming credit, the CFPB argues that small creditors are an important source of credit in rural communities. Failure to create an additional compliance option for small creditors could, they argue, result in restricted credit access in rural areas. The CFPB notes that "small creditors are significantly more likely than larger creditors to operate offices in rural areas, and there are hundreds of counties nationwide where the only creditors are small creditors and hundreds more where larger creditors have only a limited presence."⁵⁷ The CFPB cites a study that found that in 2011 "there were 629 U.S. counties, with just over 6 million in population, where community banks operated offices, but where no noncommunity banking offices were present" and "another 639 counties where community banks operated offices but where fewer than three noncommunity banking offices were present."⁵⁸ Collectively, these 1,268 counties accounted for 16.3 million people and include more than one-third of the approximately 3,200 counties in the United States.⁵⁹

The CFPB emphasizes the importance of ensuring that non-conforming credit and credit in rural areas continues to be extended, but it addresses these concerns indirectly by providing an additional compliance option for small creditors. Alternatively, the CFPB could have provided compliance options that centered on whether the mortgage was non-conforming or originated to a borrower in a rural area rather than be dependent on the type of lender. The CFPB argues, however, that it is appropriate to focus on the small creditors because, as discussed in more detail in the next section, they believe small creditors have certain qualities and face certain incentives that make them more likely to thoroughly evaluate a borrower's ability to repay. It is unclear, however, what affect the Small Creditor Portfolio QM option has on credit availability in rural areas because the QM rule also has a separate compliance option for certain lenders operating in rural areas.⁶⁰

Small Creditors' Sources of "Strong Incentives and Particular Ability"?

The CFPB argues that small lenders "have both strong incentives and particular ability to make" portfolio mortgages that consumers are able to repay, but the strong incentives and particular ability "may not be present for larger lenders."⁶¹ The CFPB cites three factors unique to small lenders that, it argues, allow small creditors "to make more accurate assessments of consumers' ability to repay than larger creditors."⁶²

First, a small lender's size, the CFPB argues, provides added incentive to accurately assess a consumer's ability to repay. When the loan is held in portfolio, the lender is exposed to the risk, and that risk "represents a proportionally greater risk to a small creditor than to a larger one."⁶³ For any given mortgage, a small bank has a smaller base across which to spread any losses, so, the argument goes, a small lender would have more incentive to only originate mortgages that it expects consumers to be able to repay. Small size could, alternatively, mean that a small lender has less to lose and could restart as a new business if its investments fail to pay off. This is likely less of a concern for a bank or credit union that has to go through the process of receiving a charter than a non-bank entity that would face lower start-up costs.

Second, the CFPB argues that small creditors are more likely to use a relationship-based lending model that may yield a more accurate assessment of the borrower. A relationship-based lending model is one in which lenders develop close familiarity with their respective customer bases. Relationship banking allows institutions to capture lending risks that are unique, infrequent, and localized. Underwriting in the relationship banking model often relies on "qualitative information gained from personal relationships between creditors and consumers."⁶⁴ Qualitative or "soft" information, such as a consumer's character, "may be difficult to quantify, verify, and communicate through the normal transmission channels of a banking organisation."⁶⁵ The relationship banking model emphasizes maintaining extended relationships with consumers, offering not just a mortgage but also holding their deposits and offering other services.

Relationship-based lending can be contrasted with a "transactional banking" or high-volume lending business model that is often associated with larger lenders. These lenders are more likely to rely on automated underwriting methodologies that emphasize hard data—credit scores, income, and other quantifiable characteristics—that often cannot capture atypical lending risks, such as the uniqueness of a particular property that is collateralizing a mortgage loan in a rural area.⁶⁶

The CFPB believes that lenders using the relationship-based model "may be better able to assess ability to repay because they are more likely to base underwriting decisions on local knowledge and nonstandard data and less likely to rely on standardized underwriting criteria."⁶⁷ The CFPB cites several studies that found that lenders using soft information in their underwriting had lower default rates on mortgages than lenders that did not.⁶⁸ Overreliance on soft information, however, could leave lenders susceptible to their biases.⁶⁹

Third, small lenders' ties to their communities, the CFPB argues, provide added incentive to thoroughly underwrite their mortgages for the borrower's ability to repay. Small lenders are likely to provide financial services within a circumscribed geographical area, allowing them to "have specialized knowledge of the community in which they operate" and potentially "have a more comprehensive understanding of their customers' financial circumstances."⁷⁰ A more comprehensive understanding would likely improve a lender's capacity to accurately assess a consumer's ability to repay. In addition, the emphasis on maintaining long-standing relationships with consumers and limiting their business to a specified geographic area, the CFPB argues, makes small lenders more concerned about their reputational risk.⁷¹

These three factors, the CFPB argues, provide small creditors with a better ability to assess a borrower's creditworthiness and offer strong incentives for the lender to internalize the costs of a borrower's default.

Legislative Proposals and Policy Discussion

Each of the three legislative proposals would expand lender eligibility for a portfolio QM compared with the Small Creditor Portfolio QM but to different degrees. Table 3 summarizes the different lender restrictions, dividing the requirements into size requirements, origination requirements, and other criteria that a lender would have to satisfy. As can be seen below, S. 1484 and H.R. 1210 would allow a lender of any size to take advantage of the portfolio QM, whereas S. 1491 and the Small Creditor Portfolio QM have size and origination requirements that a lender must satisfy.

As mentioned in the previous section, the CFPB argues that it is important to limit the portfolio QM to small lenders because small lenders have a particular ability and incentive to accurately evaluate a borrower's creditworthiness. Those proposals that have less prescriptive lender restrictions rely less on size limitations to provide protections to borrowers but instead emphasize that the portfolio requirements and loan criteria encourage an accurate assessment of ability to repay. Several policy issues related to lender restrictions are discussed below.

Expanded Eligibility: Banks by Asset Size

To provide context for the different asset thresholds contained in the Small Creditor Portfolio QM and the legislative proposals, Figure 1 shows the number of different banks by asset size. The number of small banks has decreased over time, with much of the decrease driven by the drop in the number of banks with less than $100 million in assets. Although the number of small banks has fallen, the vast majority of banks had less than $1 billion in assets, and just 107 banks had more than $10 billion in assets in 2014. Small banks make up the majority of institutions, but Figure 2 shows that most assets are held by banks over $10 billion.

Figure 2. Number of Banking Institutions, 1984-2014

Source: Figure created by CRS using data from the FDIC.

Figure 3. Share of Industry Assets by Different Categories, 1984-2014

Source: Figure created by CRS using data from the FDIC.

The CFPB's lender restrictions are stricter than those of the legislative proposals. The CFPB uses both an asset limit and origination threshold in its definition of small because it "believes that both elements of the threshold play independent and important roles."⁷² The CFPB contends the origination threshold is the best way of restricting the portfolio QM to small lenders that use the relationship-based business model that it believes is important. The CFPB intends the asset limit to prevent a large lender that has a low origination level and does not use the relationship-based business model from participating. Both factors impose limits on which lenders can participate. The CFPB estimates that "approximately 95 percent of creditors with less than $500 million in assets, approximately 74 percent of creditors with assets between $500 million and $1 billion, and approximately 50 percent of creditors with assets between $1 billion and $2 billion" would meet its original definition of small, which has since been expanded.⁷³

S. 1491 would allow larger lenders—those with up to $10 billion in assets—to participate if the lender satisfied certain criteria. It is unclear how the CFPB would implement and define the other criteria, so it is unclear how many additional lenders would be eligible. Using $10 billion as a benchmark provides the upper bound for lender eligibility under S. 1491. Approximately 98.2% of banks⁷⁴ have less than $10 billion in assets,⁷⁵ but banks with less than $10 billion in assets account for only 18.7% of the banking system's total assets.⁷⁶ H.R. 1210 and S. 1484 have no asset or origination threshold for participating, and thus many additional lenders would be eligible to participate if the proposals were enacted.

The CFPB had some discretion in setting the thresholds that it used, and the CFPB's rule implementing the Small Creditor Portfolio QM explained why it chose to limit eligibility to the smaller lenders that it did. The CFPB notes that "a creditor with assets between $1 billion and $2 billion has, on average, 16 branches, 252 employees, and operations in 5 counties. In contrast, a creditor with between $2 billion and $10 billion in assets has, on average, 34 branches, 532 employees, and operations in 12 counties."⁷⁷ As the staff and geographic reach of the lender increases, the CFPB argues, "it becomes less and less likely that a creditor will engage in relationship lending or use qualitative or local knowledge in its underwriting. In addition, as an institution adds staff and branches, it is more likely from a systems perspective to handle compliance functions."⁷⁸

Although the CFPB emphasizes the correlation between institution size and relationship lending, it is unclear why the CFPB originally chose the $2 billion in assets and 500 mortgages originated as the thresholds. In fact, the CFPB expanded the origination threshold to 2,000 mortgages per year (excluding mortgages held in portfolio).⁷⁹ Although larger banks may be less likely to use a relationship lending business model, those who support eliminating size restrictions for portfolio QM in S. 1484 and H.R. 1210 would deem the quality of the mortgages they originate to be sufficiently high to warrant receiving QM status. The next section provides evidence on default rates and other performance measures for lenders of different sizes.

Evidence of Small Lenders' Assessment of Ability to Repay

In assessing eligibility requirements for a portfolio QM option, one factor to consider is the performance of mortgages originated by lenders of different sizes. In November 2012, former Federal Reserve governor and community banker Elizabeth Duke cited the better performance of community banks' mortgage portfolios after the bursting of the housing bubble as evidence of "the responsible lending practices of community banks"⁸⁰:

Over the last several years as mortgage delinquencies reached record levels, the serious delinquency rate of mortgages held by community banks did not go much over 4 percent, far lower than the serious delinquency rates that climbed to almost 22 percent for subprime, fixed-rate loans and more than 46 percent for subprime, variable-rate loans. In fact, over the last several years, on average, mortgages held by community banks outperformed even fixed-rate, prime loans, the best performing mortgage category.⁸¹

In addition, research from the Federal Reserve Bank of Dallas, presented in Figure 4, shows the delinquency rate of mortgages held in portfolio by banks of different sizes in the aftermath of the bursting of the housing bubble. Banks of all sizes had similar delinquency rates at the beginning of 2008, but the rates diverged over the next four years, just as argued by Governor Duke. The larger banks experienced higher delinquency rates, whereas smaller banks saw less of an increase.

Figure 4. Delinquency Rates for Mortgages Held in Portfolio by Banks of Different Sizes

Source: Jeffrey W. Gunther and Kelly Klemme, Community Banks Withstand the Storm, Federal Reserve Bank of Dallas, 2012, at http://www.dallasfed.org/microsites/fed/annual/2012/e1/1201e1.cfm. Notes: The Federal Reserve Bank of Dallas notes that, for the figure, "Data for commercial banks. Residential real estate loans are closed-end, first-lien, one- to four-family mortgages. Asset size is based on the total assets (expressed in June 2012 prices) of a U.S. banking organization (holding company, when applicable). Noncurrent loans are loans past due 90 days or more and loans no longer accruing interest. Noncurrent loan rates have been adjusted to exclude loans rebooked from Government National Mortgage Association securitizations."

Performance during the most recent crisis, however, may not be indicative of future performance. Others highlight the fact that small institutions failed at a higher percentage during the Savings and Loan crisis in part because of mortgage-related losses, evidence that small institutions are capable of making significant mistakes that could lead to their failure.⁸²

Large Banks as Source of Liquidity

As previously explained in the portfolio requirements section, a mortgage that retains its QM status when transferred to a different lender's portfolio would have a lower liquidity premium (and potentially a lower rate for the borrower, all else equal) but may provide the originator with less incentive to perform thorough underwriting. Similarly, if the portfolio QM option is available to more types of lenders, such as large banks, a portfolio mortgage would have a lower liquidity premium (because there would be a larger pool of eligible purchasers of the mortgage), but the incentives to perform thorough underwriting that are associated with being small might be lost.

Some argue that the added liquidity injected in to the market by allowing large banks to purchase mortgages, especially during a market downturn, would justify large banks' eligibility for the portfolio QM. S. 1484 would allow large banks to purchase mortgages that they did not originate, but if the mortgage was held in portfolio, the mortgage would receive QM status. H.R. 1210 would allow a large bank to originate a mortgage and have it receive QM status if the mortgage was retained in portfolio, but it would not allow a large bank to use the portfolio QM option for a mortgage that it acquired and did not originate. S. 1491 would not allow banks over $10 billion in assets to use the portfolio QM.

Whether a large bank would actually provide support to the mortgage market during a downturn, however, depends on the nature of the particular episode. Figure 5 shows the change in mortgage portfolios relative to 2006 for all banks and the 10 largest banks (based on their mortgage portfolios) in any given year. As can be seen below, the banking system reduced the total amount of mortgages that were held in portfolio, and the largest banks shrank their mortgage portfolios by a similar amount. During this period, the largest banks, in aggregate, were not providing disproportionate support or drag on the market but were generally moving with the market. Past actions are not, however, necessarily predictive of future performance.

Figure 5. Cumulative Percentage Change in Mortgage Portfolio for All Banks and the 10 Largest Banks

Source: Figure created by CRS using data from the FDIC. Notes: The figure compares the mortgage portfolios of the 10 largest banks in any given year and all banks in a given year to 2006. Data for 2015 is as of the end of the first quarter; other years are based on the end of the fourth quarter.

3. Loan Criteria

The CFPB's Small Creditor Portfolio QM has the same product-feature requirements as the Standard QM—no negative amortization or terms longer than 30 years—but has less prescriptive underwriting standards. The main difference in underwriting standards is that the Small Creditor Portfolio QM does not have a numeric debt-to-income (DTI) threshold that must be met, unlike the Standard QM, which requires a borrower to have a DTI below 43%. Lenders are still required "to consider the consumer's debt-to-income ratio ... and to verify the underlying information" used to calculate the DTI under the Small Creditor Portfolio QM, but "there is no numeric limit."⁸³ The CFPB removes the numeric DTI limit for portfolio loans offered by small lenders because it argues that the factors unique to portfolio loans and to small lenders (as explained earlier in this report) provide added incentives that ensure that a lender using the Small Creditor Portfolio QM accurately assesses a borrower's creditworthiness. Though the 43% limit is removed, the CFPB argues that lenders should still have to consider DTI because "consideration of debt-to-income ... is fundamental to any determination of ability to repay."⁸⁴ The less prescriptive underwriting standards are intended to reduce the burden on small creditors and allow for greater credit availability.

Legislative Proposals and Policy Discussion

The legislative proposals have significant variation in how prescriptive they would be in setting underwriting and product-feature requirements for a loan that would be eligible for a portfolio QM option. Table 4 summarizes the requirements. H.R. 1210 has the fewest requirements, and S. 1491 has the most prescriptive requirements. In addition, H.R. 1210 and S. 1484 generally would limit the ability of the CFPB to add additional loan criteria beyond what is in the bills, but S. 1491 would allow the CFPB to add additional loan criteria for a mortgage to qualify for portfolio QM status.

Lender Discretion and Prescriptive Standards

The CFPB intended the minimum underwriting standards and the possibility of a consumer bringing a lawsuit to work in tandem to protect consumers. A consumer can bring a suit claiming the lender failed to comply with the ATR rule, but the borrower's likelihood of success is partially dependent on the stringency of the loan criteria and the ease with which the lender can document that it complied with the requirement. If a lender must satisfy prescriptive or difficult to document criteria to have a loan receive QM status, then a borrower is likely to be more successful in showing that the lender failed to comply with those detailed standards. If a lender must satisfy less prescriptive criteria, such as simply documenting that the loan is currently held in portfolio, then it is likely to be more difficult for the borrower to win the suit.⁸⁵ The more likely a borrower is to win, the argument goes, the more incentive a lender has to thoroughly underwrite the loan and address the market failures that are at the root of the ATR rule. Alternatively, the more prescriptive standards that may make it more likely for a borrower to win a suit could dissuade lenders from extending credit to borrowers who are not clearly within the required underwriting standards. Making the loan criteria easier to satisfy could reduce the burden on lenders and allow them to expand credit but could reduce consumer protections by making it harder for a borrower to win a suit. This highlights the tradeoff between consumer protection and access to credit when establishing minimum underwriting standards.

One of the areas that has received particular attention is the treatment of borrowers who have atypical financial situations, such as self-employed individuals or seasonal employees. The CFPB has standards and definitions for terms such as income and debt in its Appendix Q that lenders use to ensure compliance with the ATR rule, and Appendix Q has specific information related to how to address certain atypical financial situations. Some industry groups argue, however, that Appendix Q is too complex. They contend that lenders are less likely to extend credit to atypical borrowers whom they otherwise deem creditworthy because of the complexity of Appendix Q and of the uncertainty lenders face as to whether they have appropriately applied its requirements to atypical borrowers.⁸⁶

Supporters of the expanded portfolio QM options argue that the less prescriptive standards and easier to document criteria in their proposals would expand credit availability, especially for those with atypical financial situations. They contend that the lender should be given the discretion to set its own underwriting standards and product-feature standards because it holds the risk on its balance sheet. The ability to set their own standards, the argument goes, could allow lenders to extend loans to more types of borrowers.

Supporters disagree, however, with how much discretion the lender should be given. H.R. 1210, for example, provides lenders with the most discretion. It would not mandate any specific underwriting requirements and, for product-feature requirements, would limit only prepayment penalties. S. 1491 is at the other extreme, requiring the lender to satisfy prescriptive standards that are similar to those found in the CFPB's Small Creditor Portfolio QM. Supporters of the underwriting and product-feature requirements in S. 1491 and in the Small Creditor Portfolio QM argue that the criteria are essential to ensuring that the lender performs a thorough assessment of the borrower's ability to repay.

Access to Credit and Regulatory Burden

Having assessed the three major parts individually—portfolio requirements, lender restrictions, and loan criteria—for the different proposals, what would the collective effect on credit availability and on regulatory burden be if an expanded portfolio QM were adopted? The effect on credit availability would depend on many different factors, including which expanded portfolio QM proposal (if any) was implemented. Even if the broadest and least prescriptive proposal were enacted, it is unclear how significant an effect it would have on credit availability. As a recent Government Accountability Office (GAO) report explains, "agencies, market participants, and observers estimated that the QM ... regulations would not have a significant effect initially because many loans made in recent years already met QM ... criteria before these regulations were promulgated."⁸⁷ Based on GAO's analysis, an expanded QM option may not result in a significant expansion in credit availability, as many of the mortgages that would receive QM status under the new QM option would also receive QM status under the previously available options. Similarly, a December 2015 study by the Federal Reserve found "evidence that some market outcomes were affected by the new rules, but the estimated magnitudes of the responses are small."⁸⁸ Others argue that QM's prescriptive criteria are preventing some lenders from lending to creditworthy borrowers. A portfolio QM, they argue, could increase credit availability by offering lenders more flexibility.

Although GAO and the Federal Reserve found that the QM rule has had a small effect initially on credit availability, the Federal Reserve argues that the QM rule "could be more binding in the future."⁸⁹ An expanded portfolio QM's longer-term effects on credit availability, therefore, are unclear. Market participants expect mortgage credit conditions to loosen as more time passes from the bursting of the housing bubble,⁹⁰ meaning that lenders may want to expand credit availability, but it is unclear whether they would do so because of the restrictions in the QM rule. In addition, the Temporary GSE QM category—which, in general, has less prescriptive underwriting criteria than the Standard QM category—is scheduled to expire in 2021, which may restrain credit after its expiration for borrowers on the margins of creditworthiness. An expanded portfolio QM option may have a greater effect on credit availability at that time.

Although the effect on credit availability may be minimal, an expanded portfolio QM option that has less prescriptive underwriting standards could have a greater impact on regulatory burden. As with many policy discussions related to regulatory burden, reducing the burden borne by the lender could reduce the protections available to consumers.

As mentioned before, lenders point to the prescriptive underwriting standards found in the CFPB's Appendix Q as source of their unduly burdensome compliance costs. The less prescriptive portfolio options would not require lenders to underwrite using Appendix Q. So even if a mortgage would have received QM status under an alternative QM option, the fact that the less prescriptive portfolio QM option is available would mean that a lender would not necessarily have to perform as cost-intensive of an evaluation. The effect on credit availability may be small in this example, but the reduction in regulatory burden may be more meaningful.

Concluding Thoughts

The CFPB and many in Congress generally agree that if a lender holds a mortgage in portfolio, then the incentives of the lender and borrower are more likely to be aligned than if the lender sold the mortgage. There is disagreement, however, on the policy implications of whether the incentive effects alone are sufficient or whether other types of restrictions are needed to ensure prudent lending and consumer protection. In establishing the Small Creditor Portfolio QM, the CFPB argued that because borrower and lender incentives were more correctly aligned, the underwriting criteria for the Small Creditor Portfolio QM could be less prescriptive than for the Standard QM, but the compliance option would be limited to small creditors.⁹¹ The Small Creditor Portfolio QM needs to contain certain lender restrictions and loan criteria rather than just apply to all portfolio loans offered by all lenders, the CFPB contends, because lender and borrower incentives are not perfectly aligned. The possibility of collateral dependent lending and other factors could result in a lender keeping mortgages in portfolio that may be profitable but not prudently underwritten. Although the lender restrictions and loan criteria are intended to ensure consumers are adequately protected, they may unnecessarily restrict credit and be unduly burdensome to lenders.

The different legislative proposals that would expand the portfolio QM option vary in their implicit responses to the CFPB. S. 1491 is the most similar to the Small Creditor Portfolio QM in that it would require similar loan criteria to be followed for a mortgage to qualify for the expanded QM but it would allow larger lenders to be eligible.

H.R. 1210 would have the least prescriptive loan criteria and lender requirements. Nearly any mortgage originated by any depository institution would be eligible for the expanded portfolio QM so long as it would be held in the originating lender's portfolio.

S. 1484 would also have more expansive lender eligibility than the current CFPB rule allows and less prescriptive underwriting criteria, but it would require more documentation of a borrower's financial status than H.R. 1210 would require.

The key questions that policymakers face in choosing which, if any, of the legislative proposals to support include the following:

• Are borrower and lender incentives sufficiently aligned so that consumers will be adequately protected and loans will be prudently underwritten when the lender keeps the mortgage in portfolio?
• Is it necessary to limit the portfolio QM option to small lenders because the relationship banking model that they often employ is believed to result in a more accurate assessment of a borrower's ability to repay? If so, lenders of what size?
• Should any type of mortgage held in portfolio—including balloon or interest-only mortgages—be eligible? Should the lender be required to follow certain prescribed underwriting procedures or to document a borrower's financial status?
• Could an expanded portfolio QM reduce lenders' regulatory burden while ensuring that consumers have sufficient protections?

The answers to these and other questions, which have been explored in this report, may help policymakers in evaluating the legislation.

Author Contact Information

Footnotes

1.	For more, see CRS Report R43999, An Analysis of the Regulatory Burden on Small Banks, by [author name scrubbed] and [author name scrubbed].
2.	Consumer Financial Protection Bureau (CFPB), "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act," 78 Federal Register 6416, January 30, 2013.
3.	15 U.S.C. §1639c.
4.	There are two different types of presumption of compliance: rebuttable presumption and safe harbor. See CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act," 78 Federal Register 6587, January 30, 2013.
5.	CFPB, Prepared remarks of Richard Cordray at a meeting of the Credit Union National Association, February 27, 2013, at http://www.consumerfinance.gov/speeches/prepared-remarks-of-richard-cordray-at-a-meeting-of-the-credit-union-national-association/. For a preliminary analysis of the effect of the QM rule on originations, see Bing Bai, Data Show Surprisingly Little Impact of New Mortgage Rules, Urban Institute, August 21, 2014, at http://www.urban.org/urban-wire/data-show-surprisingly-little-impact-new-mortgage-rules.
6.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6559, January 30, 2013.
7.	For example, see William Rogers and William Winter, "The Impact of Foreclosures on Neighboring Housing Sales," American Real Estate Society, vol. 31, no. 4 (2009), pp. 455-479.
8.	Brian Bucks and Karen Pence, "Do Borrowers Know Their Mortgage Terms?," Journal of Urban Economics, vol. 64, no. 2 (September 2008), pp. 218-233.
9.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6560, January 30, 2013.
10.	Ibid.
11.	Ibid.
12.	15 U.S.C. §1639c.
13.	15 U.S.C. §1639c.
14.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6624, January 30, 2013.
15.	Ibid., p. 6657.
16.	12 C.F.R. §1026.43.
17.	The CFPB expanded its definition of small credit in an amendment to the ATR rule. See CFPB, "Amendments Relating to Small Creditors and Rural or Underserved Areas Under the Truth in Lending Act (Regulation Z)," 80 Federal Register 59943, October 2, 2015.
18.	12 C.F.R. §1026.43.
19.	For example, see statement of Jack Hartings, Independent Community Bankers Association, January 14, 2014, at https://www.icba.org/files/ICBASites/PDFs/test011414.pdf.
20.	The text of the bill was among the financial regulatory changes included in the FY2016 Financial Services and General Government Appropriations Act (S. 1910), reported by the Senate Appropriations Committee on July 30, 2015. It was not included in the Consolidated Appropriations Act, 2016 (P.L. 114-113) that was signed into law on December 18, 2015.
21.	Peter Schroeder, "Senate Dems Unveil Competing Dodd-Frank Tweaks," The Hill, May 19, 2015.
22.	12 C.F.R. §1026.43(e)(5)(ii).
23.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6634, January 30, 2013.
24.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35483, June 12, 2013.
25.	The incentives of the originate-to-distribute model may vary depending on whether the mortgage is sold to an entity affiliated with the seller or to an unaffiliated entity. For an overview of the originate-to-distribute model, see Vitaly M. Bord and Joao A. C. Santos, The Rise of the Originate-to-Distribute Model and the Role of Banks in Financial Intermediation, New York Federal Reserve Bank, July 2012, at https://www.newyorkfed.org/medialibrary/media/research/epr/12v18n2/1207bord.pdf. Additional ways have been developed to potentially align incentives, such as representations and warranties. For more, see Laurie Goodman and Jun Zhu, Reps and Warrants: Lessons from the GSEs Experience, Urban Institute, October 24, 2013, at http://www.urban.org/research/publication/reps-and-warrants-lessons-gses-experience.
26.	In the underwriting process, a lender will consider the size of the loan relative to the value of the house in what is called the loan-to-value or LTV ratio.
27.	In some cases, a borrower may rent the home and earn income, but the example above is not focusing on that narrower circumstance.
28.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6623, January 30, 2013.
29.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6657, January 30, 2013.
30.	Ibid.
31.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35486, June 12, 2013.
32.	The CFPB highlights liquidity issues in its final rule. See CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35478, June 12, 2013.
33.	See 15 U.S.C. §1641.
34.	Similar contracts exist in the mortgage finance system. For more on put backs, see Laurie Goodman and Jun Zhu, Reps and Warrants: Lessons from the GSEs Experience, Urban Institute, October 24, 2013, at http://www.urban.org/research/publication/reps-and-warrants-lessons-gses-experience.
35.	This would potentially involve a two-step process. First, the borrower would bring a suit against the assignee. Second, the assigned would sue the originator for a breach of contract. As a result, both would have some liability risk.
36.	For example, see Rep. Andy Barr, "Barr Introduces Legislation to Help Homebuyers, Prevent Bailouts," press release, February 27, 2015, at https://barr.house.gov/media-center/press-releases/barr-introduces-legislation-to-help-homebuyers-prevent-bailouts.
37.	Ronel Elul, Securitization and Mortgage Default, Federal Reserve Bank of Philadelphia, March 2015.
38.	Ibid., p. 18
39.	Ibid., p. 18.
40.	Sumit Agarwal, Yan Chang, and Adbullah Yavas, "Adverse Selection in Mortgage Securitization," Journal of Financial Economics, vol. 105 (2012), p. 641.
41.	Ibid., p. 658.
42.	Ibid., p. 643.
43.	Ibid., p. 641.
44.	Ibid., p. 656.
45.	Ibid., p. 656.
46.	CFPB, Policy Guidance on Supervisory and Enforcement Considerations Relevant to Mortgage Brokers Transitioning to Mini-Correspondent Lenders, July 2014, pp. 5-6.
47.	FDIC, "The Savings and Loan Crisis and Its Relationship to Banking," at https://www.fdic.gov/bank/historical/history/167_188.pdf.
48.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35486, June 12, 2013.
49.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35486, June 12, 2013.
50.	S. 1484 would establish a new process for determining which banks are deemed systemically important. For more information, see CRS Report R42150, Systemically Important or "Too Big to Fail" Financial Institutions, by [author name scrubbed].
51.	12 C.F.R. §1026.35.
52.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6656, January 30, 2013.
53.	The temporary GSE QM option will expire when Fannie Mae and Freddie Mac are removed from conservatorship (or receivership, if the entities are placed into receivership). The temporary QM option would also expire if none of those events occurs within seven years after the CFPB's rule becomes effective (January 2021).
54.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6624, January 30, 2013.
55.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35438, June 12, 2013.
56.	CFPB, "Amendments Relating to Small Creditors and Rural or Underserved Areas Under the Truth in Lending Act (Regulation Z)," 80 Federal Register 59943, October 2, 2015.
57.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35478, June 12, 2013.
58.	Federal Deposit Insurance Corporation (FDIC), FDIC Community Banking Study, December 2012, pp. 3-5, at https://www.fdic.gov/regulations/resources/cbi/report/cbi-full.pdf.
59.	Ibid.
60.	See CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act," 78 Federal Register 6556, January 30, 2013.
61.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35486, June 12, 2013.
62.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6624, January 30, 2013.
63.	Ibid., p. 6656.
64.	Ibid., p. 6634.
65.	CFPB quote of Allen N. Berger and Gregory F. Udell, Small Business Credit Availability and Relationship Lending: The Importance of Bank Organisational Structure, Economic Journal (2002). See CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6634, January 30, 2013.
66.	For more information on automated underwriting, see Wayne Passmore and Roger Sparks, "The Effect of Automated Underwriting on the Profitability of Mortgage Securitization," Federal Reserve Board, at http://www.federalreserve.gov/pubs/feds/1997/199719/199719abs.html.
67.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6656, January 30, 2013.
68.	The CFPB cites Sumit Agarwal et al., "The Role of Soft Information in a Dynamic Contract Setting: Evidence from the Home Equity Market," Journal of Money, Credit and Banking, October 2011; and O. Emre Ergungor and Stephanie Moulton, "Beyond the Transaction: Depository Institutions and Reduced Mortgage Default for Low-Income Homebuyers," Federal Reserve Bank of Cleveland Working Paper 11-15, August 2011.
69.	Behavioral economics offers examples of mechanisms that may lead individuals to make suboptimal choices.
70.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6652, January 30, 2013.
71.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6634, January 30, 2013.
72.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35486, June 12, 2013.
73.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35486, June 12, 2013.
74.	This section provides data on banks for illustrative purposes to show how many additional banks would be eligible, but some proposals would apply to a broader set of institutions including nonbanks.
75.	As of June 30, 2015, 112 of the 6,357 total banks had assets in excess of $10 billion, according to the FDIC.
76.	As of June 30, 2015, banks with less than $10 billion in assets collectively held $2.96 billion in assets of the industry's $15.9 billion in total assets.
77.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35486, June 12, 2013.
78.	Ibid., p. 35486.
79.	CFPB, "Amendments Relating to Small Creditors and Rural or Underserved Areas Under the Truth in Lending Act (Regulation Z)," 80 Federal Register 59943, October 2, 2015.
80.	Gov. Elizabeth A. Duke, Community Banks and Mortgage Lending, Federal Reserve Board of Governors, November 9, 2012, at http://www.federalreserve.gov/newsevents/speech/duke20121109a.htm.
81.	Ibid.
82.	FDIC, The Savings and Loan Crisis and Its Relationship to Banking, at https://www.fdic.gov/bank/historical/history/167_188.pdf.
83.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 35487, June 12, 2013.
84.	Ibid.
85.	This may not be true if the prescriptive standards are themselves bright lines that are easy to document.
86.	For example, see Housing Policy Council, Financial Services Roundtable, "Docket No. CFPB-2013-00120; RIN 3170-AA37; Amendments to the 2013 Mortgage Rules Under the Real Estate Settlement Procedure Act (Regulation X) and the Truth in Lending Act (Regulation Z)—Second Letter," May 24, 2013, at http://fsroundtable.org/wp-content/uploads/2014/01/Questions-and-Clarifications-on-Appendix-Q-05-24-13.pdf; and Kate Berry, "Why Wells Fargo Says Its Portfolio Lending Is So Safe," American Banker, May 11, 2015, at http://www.americanbanker.com/news/consumer-finance/why-wells-fargo-says-its-portfolio-lending-is-so-safe-1074266-1.html.
87.	U.S. Government Accountability Office, Mortgage Reforms: Actions Needed to Help Assess Effects of New Regulations, GAO-15-185, June 2015, p. 25, at http://www.gao.gov/assets/680/670979.pdf. For an overview of additional research of the effects of the QM rule on credit availability, see CRS Report R43081, The Ability-to-Repay Rule: Possible Effects of the Qualified Mortgage Definition on Credit Availability and Other Selected Issues, by [author name scrubbed].
88.	Neil Bhutta and Daniel Ringo, Effects of the Ability to Repay and Qualified Mortgage Rules on the Mortgage Market, Board of Governors of the Federal Reserve System, FEDS Notes, December 29, 2015, at http://www.federalreserve.gov/econresdata/notes/feds-notes/2015/effects-of-the-ability-to-repay-and-qualified-mortgage-rules-on-the-mortgage-market-20151229.html.
89.	Ibid.
90.	For example, see mortgage origination projections at Urban Institute, Housing Finance Policy Center, Housing Finance at a Glance, November 2015, p. 12.
91.	CFPB, "Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)," 78 Federal Register 6624, January 30, 2013.