Introduction

A single-family mortgage is a loan secured by the underlying real estate collateral, a residential dwelling, which is being financed by the loan.¹ This report focuses on the primary market, the market where loan applicants (who, if approved, become borrowers) obtain mortgages from loan originators and lenders.² U.S. mortgages may have different maturities—the period of time that the borrower has to repay the loan, typically 15 or 30 years. Mortgages can be repaid either in fixed rate or adjustable (variable) rate installment payments over the maturity term.³ These differences influence the overall price of mortgages. For example, mortgages with shorter maturity terms are likely to have lower prices relative to those with longer maturities. Because the price of adjustable rate mortgages (ARMs) fluctuate over their maturity terms with linked interest rates, determining how expensive they are relative to fixed rate mortgages is more difficult. Furthermore, mortgage prices reflect the total cost to borrowers if their loans are repaid as scheduled and held to full maturity. The actual price paid by borrowers, however, must be calculated at the point when their mortgage obligations were terminated.

A mortgage price consists of two components—the interest rate and the upfront costs. The interest rate component is based upon an applicable risk-free rate (generally, a U.S. Treasury interest rate) and then adjusted for the borrower's default risk. The upfront component consists of fees—some fees associated with the housing purchase transaction and some linked to obtaining the mortgage that are not incorporated into the interest rate. The price or total cost of a mortgage, the annual percentage rate (APR), is the sum of the interest rate and various upfront fees expressed as a percentage. For any given APR, a trade-off exists that allows a borrower to pay either higher upfront costs to lower the interest rate or lower upfront fees for a higher interest rate. Borrowers with less cash on hand, for example, may prefer to roll more costs into the interest rate to be paid over the life of the mortgage. Some borrowers may prefer paying more costs upfront to maintain lower mortgage payments over the loan life.

Some borrowers are likely to obtain high-cost residential mortgage loans for at least five possible reasons. First, borrowers with weak credit histories face higher borrowing costs (than those with better credit histories), because lenders typically require more compensation commensurate with taking greater credit (i.e., default) risks. Second, some borrowers may not have shopped for mortgages with lower rates and fees. Third, despite applicable laws and regulations (discussed in this report) aimed at reducing them, fees associated with obtaining mortgages may not have been adequately disclosed to borrowers when they entered into the lending transaction. Fourth, borrowers may have obtained high-cost loans as a result of discrimination. Some research has found that minorities are more likely to pay rates above specified pricing thresholds (prior to controlling for some related borrower characteristics).⁴ Fifth, excessive mortgage pricing may result from forms of predatory lending.⁵ In short, mortgage loans may be expensive for a variety of reasons and subsequently contribute to housing cost burdens.⁶

Over the years, Congress has enacted legislation intended to promote transparency and reduce the proliferation of high-cost loan offerings in the primary mortgage market. The following selected laws may facilitate competitive pricing in the primary mortgage market:

• The Truth in Lending Act of 1968 (TILA; P.L. 90-301) applies to all forms of consumer credit.⁷ TILA requires covered lenders to disclose the total cost of credit, which includes the loan rate and fees, in the form of an APR.
• The Real Estate Settlement Procedures Act of 1974 (RESPA; P.L. 93-533) requires lenders, mortgage brokers, or servicers of home loans to provide borrowers with pertinent and timely disclosures regarding the nature and costs of the real estate settlement or closing costs.⁸ Settlement costs, also known as closing costs, include fees for appraisals, title searches, document preparation, and processing services—various costs associated with obtaining the mortgage itself (as opposed to costs of the residential property).
• The Home Mortgage Disclosure Act of 1975 (HMDA; P.L. 94-200) requires disclosure of mortgage origination information.⁹ The ability to observe mortgage origination patterns allows regulators to determine where further investigation of geographical discrimination, also known as redlining, may be necessary to increase credit access to creditworthy individuals and discourage discriminatory loan markup pricing practices.¹⁰
• The Fair Credit Reporting Act of 1970 regulates the activities of the consumer credit reporting industry and is implemented by the Consumer Financial Protection Bureau (CFPB). Credit reporting agencies (CRAs) are firms that prepare consumer reports based upon individuals' financial transactions history data.¹¹ CRA information can subsequently be used to compute a credit score, a metric used to predict the probability that a borrower would default on a loan. Because these third-party firms independently assess borrower credit risk, this technology promotes greater competition among lenders.¹²

Congress has also enacted legislation that imposes higher costs and legal risks on lenders when they make higher-priced loans, which arguably has the effect of discouraging practices that would result in expensive or predatory loan offerings.

• The Home Ownership Equity Protection Act of 1994 (HOEPA; P.L. 103-325) amends TILA to require additional reporting of high-cost refinance and other non-purchase loans secured by primary (owner-occupied) residences.¹³
• The Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010 (Dodd-Frank Act; P.L. 111-203) contains many consumer protection provisions, among other things. Examples include establishing the CFPB to implement and enforce federal consumer financial laws while ensuring consumers' access to financial products and services. The CFPB has regulatory authority over banking and nonbanking firms that offer consumer financial products. The Dodd-Frank Act also created the ability-to-repay (ATR) requirement, implemented by the CFPB, to establish minimum standards that all creditors must consider during the underwriting of U.S. residential mortgage originations. Lenders may reduce their potential legal liabilities by demonstrating proof of compliance with ATR via originating qualified mortgages, discussed in the last section of this report.

This report begins by explaining mortgage pricing fundamentals in the primary market. Next, the legislative and regulatory approaches to facilitate competitive mortgage loan pricing are discussed in two parts. First, increased market transparency resulting from mandatory mortgage interest rate and associated fee disclosures are discussed. Next, requirements that impose higher costs on lenders that offer mortgages with less affordable or desirable terms, which minimizes excessive market pricing, are discussed. In certain cases, however, lenders' avoidance of higher costs may limit credit access to some qualified low- and moderate-income borrowers.

Mortgage Pricing Fundamentals

Generally speaking, all loans, including mortgages, require some minimum compensation adjusted for additional risks. The slope of the yield curve (shown in Figure 1) consists of the interest rates of return that lenders could receive by lending to the U.S. Treasury (i.e., purchasing U.S. Treasury bonds) at different maturities at no risk. U.S. Treasury rates (yields) are considered risk-free base rates because they lack default (credit) risk and prepayment risk. The federal government has never repaid a Treasury obligation late or failed to repay, both of which would be considered defaults. Likewise, the federal government never repays its obligations early, which would be considered a prepayment. All principal and interest payment obligations are paid as scheduled to holders of Treasury securities who rely on the predictability of cash flows. Treasury rates, therefore, are considered risk-free because repayments have never been late or early. Lenders offer loan rates above the risk-free base rate to receive compensation for taking on the additional default and prepayment risks of private sector borrowers. If unable to receive additional compensation for taking on additional risks, lenders would simply purchase risk-free Treasuries and receive compensation commensurate for taking no risk over a designated time period.

Figure 1. The Treasury Yield Curve Snapshot Taken on July 1, 2018

Source: U.S. Department of Treasury, Daily Treasury Yield Curve Rates. Notes: The July 1, 2018, snapshot was chosen to avoid unusual anomalies that are likely to affect the shape of the yield curve, such as the Great Recession of 2008 or the COVID-19 pandemic.

Conventional conforming 30-year mortgages are typically risk-adjusted for prepayment and default risks that are nonexistent when lending to the U.S. federal government.¹⁴ Rather than use a comparable 30-year U.S. Treasury rate as the benchmark minimum compensation rate, which would correspond with the maturity length of a 30-year mortgage, a Treasury with a shorter maturity is frequently used to reflect the prepayment risk of most borrowers. Specifically, a 10-year Treasury bond is considered a better approximation of the prepayment risk associated with the traditional 30-year fixed rate mortgage.¹⁵ Following a decline in mortgage rates, some borrowers may repay their existing mortgages early by refinancing (i.e., paying off an old mortgage and obtaining a new one with a lower interest rate). Borrowers also prepay if they sell their homes before fully repaying their mortgages. Because borrowers are less likely to stay in the same mortgage loan for 30 years, the estimated lifespan for 30-year mortgages better aligns with that of 10-year Treasury bonds.¹⁶ In addition, the rates of 30-year mortgages and 10-year Treasury bonds tend to move in similar directions, at similar speeds and magnitudes, thus indicating possible mortgage market irregularities when these rate movements do not coincide.¹⁷

Mortgage rates are also adjusted for default risk. Specifically, additional basis points commensurate with the additional default risk of a borrower are incorporated into the loan's coupon. A basis point is one-hundredth of 1% (0.01%), meaning that 100 basis points equal 1%. A loan's coupon, which is analogous to a borrower's monthly payment, includes the annual interest rate expressed as a percentage of the face value of the loan principal from issue date until maturity as well as some additional fees spread over the life of the loan. The term risk-based pricing refers to the practice of charging borrowers with greater default propensities higher premiums to reimburse lenders for potential losses. Although higher-risk borrowers pay more for their mortgages relative to lower-risk borrowers, risk-based pricing can result in fewer credit denials and greater credit accessibility. Borrowers can pay for the default risk they pose either in the form of additional basis points or by separately purchasing mortgage insurance, usually assessed as a combination of an upfront fee and additional basis points.¹⁸

The servicing fee is an example of a fee typically included in the mortgage coupon. Following origination of a single-family mortgage, a mortgage servicer receives a monthly fee to perform various administrative tasks—collecting and remitting the principal and interest payments to the mortgage lender; managing the borrower's escrow account; processing the loan title once paid in full; and administering loss mitigation (e.g., forbearance plans) or foreclosure resolution on behalf of the lender if the borrower falls behind or fails to make full payment. A mortgage servicing right generates a servicing fee typically averaging 25 basis points (0.25% or $250 per $100,000 of an outstanding mortgage balance) per month.²⁰

Although the mortgage coupon typically quoted to a borrower includes the adjustments for prepayment and default risks plus the mortgage servicing fee, it still does not reflect the total loan price. Instead, the APR represents the total annual borrowing costs of a loan expressed as a percentage. The APR calculation includes both the mortgage coupon (spread over the life of the loan) and additional fees that are paid up front.²¹ Additional upfront charges may include loan origination fees, the costs of appraisals, title insurance, and upfront premiums charged to cover mortgage default risk. Table 1 summarizes the APR calculations—the sum of the mortgage coupon and upfront costs.

In sum, a mortgage price consists of two components—the coupon and the upfront costs. The mortgage coupon for a 30-year fixed rate mortgage includes an adjustment for the prepayment risk of homebuyers; it typically follows a 10-year Treasury bond due to the similarities of their lifespans. A mortgage coupon is also adjusted for a borrower's default risk. The upfront cost component consists of fees—some that can be incorporated into the mortgage coupon (e.g., servicing fees) and some charged separately as upfront fees. The APR, the price or total mortgage cost, is the sum of the coupon and various upfront costs expressed as a percentage.

In practice, lenders typically provide mortgage originators with rate sheets that conceptually are derived similarly to the process described above to establish mortgage coupons. Lender rate sheets list either a set of interest rates with the various risk adjustments or a set of loan-level price adjustments to add to a designated minimum base rate.²² The various coupons or loan-level price adjustments capture differences in borrower characteristics (e.g., credit score, loan-to-value ratio, etc.); loan features (e.g., fixed rate, adjustable rate, investment property, cash-out refinance, amount, etc.); and other financial risk characteristics that would require supplemental compensation for lenders.²³

The Influence of Disclosure and Reporting Requirements on Mortgage APRs

The law of one price (LOOP) states that, under ideal conditions, identical financial securities should have identical prices.²⁹ If both borrowers (with similar risk characteristics) and lenders have access to sufficient and equivalent information, then achieving the LOOP—or at least pricing within a tolerable range—theoretically should be possible. Less price variability in the primary market not only enhances transparency of the underlying fees charged to borrowers but also enhances liquidity in the secondary mortgage market, where mortgages can be bought, sold, and used to create new tradeable securities (e.g., MBSs).³⁰

Achieving the LOOP in the primary mortgage market, however, is challenging for the following reasons. First, borrowers are not identical. Less creditworthy borrowers have different prepayment and default propensities relative to more creditworthy borrowers, and lenders require higher compensation (yields) for higher-risk borrowers. Second, because borrowers shop for mortgages less frequently than they shop for other goods and services (e.g., groceries), they are less likely to be familiar with all charges and fees associated with mortgage transactions. Third, some borrowers may not be aware that they can shop for lower mortgage rates and, in some cases, settlement fees.³¹ Fourth, the APR metric also becomes less meaningful in the context of non-traditional mortgage products with features such as balloon payments or term lengths in which an adjustable interest rate is locked.³² Furthermore, an APR in isolation is less meaningful without relevant comparisons. Even if prospective borrowers have and can understand the information pertaining to the mortgages offered to them, they are unlikely to have comparable mortgages with similar amounts and maturity lengths for sake of APR comparison. Hence, differences in borrowers' financial risk characteristics, the lack of shopping by some borrowers, and differences in mortgage products and settlement costs all limit the ability for borrowers to make relevant comparisons, thus impeding convergence to a competitive APR mortgage price.

As previously noted, Congress has enacted various legislation to enhance the disclosure of mortgage pricing information, which may incidentally reduce price variability and get closer to achieving the LOOP for some primary market segments. For example, the federal agencies that guarantee the default risk of mortgages and government-sponsored enterprises (GSEs) such as Fannie Mae and Freddie Mac have facilitated standardized single-family mortgage underwriting (i.e., loan qualification) requirements at the national level.³³ The Federal Housing Financing Agency, the primary regulator of the GSEs, has also implemented standardized data collection initiatives for single-family mortgage originations.³⁴ The Fair Credit Reporting Act, which is implemented by the CFPB, requires accurate and complete reporting of consumer payment activity to credit reporting agencies.³⁵ Standardized underwriting and improved data accuracy enhances the ability to price default risk premiums. Moreover, TILA and RESPA require certain disclosures to increase mortgage pricing transparency for borrowers. HMDA and HOEPA have reporting requirements when prices exceed certain thresholds.

Consequently, if mortgages satisfy national uniform mortgage underwriting standards, which usually make them eligible for various federal insurance programs or for purchase by federal and federally related entities, then they should have similar financial characteristics (e.g., prepayment and default risk propensities) and, therefore, exhibit less price variability. Mortgage loan prices are unlikely to converge perfectly to a single price as would be predicted by the LOOP. Instead, they are more likely to reflect, for example, regional differences in settlement costs that are less likely to be influenced by competitive forces. Nevertheless, the legislative and administrative actions discussed in this section arguably reduce price variation in the primary mortgage market particularly for prime (creditworthy) borrowers.

TILA-RESPA Integrated Disclosures (TRID)

A lack of transparency with respect to loan terms and settlement costs can make it difficult for consumers to make well-informed decisions when choosing mortgage products. Inadequate disclosures can make some borrowers more vulnerable to predatory, discriminatory, and fraudulent lending practices. In addition, when costs are hidden from borrowers, they become vulnerable to payment shocks that could possibly lead to financial distress or even foreclosure.

The adequate disclosure of mortgage terms is a long-standing issue that has prompted several congressional actions. For example, TILA, previously implemented by the Federal Reserve Board, requires lenders to disclose the cost of credit and repayment terms of mortgage loans before borrowers enter into any transactions. In addition, RESPA, previously implemented by the Department of Housing and Urban Development (HUD), requires standardized disclosures about the settlement or closing costs. Examples of such costs include loan origination fees or points, credit report fees, property appraisal fees, mortgage insurance fees, title insurance fees, home and flood insurance fees, recording fees, attorney fees, and escrow account deposits. RESPA also includes the following provisions: (1) providers of settlement services are required to provide good faith estimates of the settlement service costs borrowers should expect at the closing of their mortgage loans; (2) lists of the actual closing costs must be provided to borrowers at the time of closing; and (3) "referral fees" or "kickbacks" among settlement service providers are prohibited to prevent settlement fees from increasing unnecessarily.

When TILA and RESPA were implemented separately, lenders presented borrowers with separate TILA and RESPA disclosures. The Dodd-Frank Act recently transferred the primary rulemaking and enforcement authority of TILA and RESPA, along with other existing consumer protection laws, to the CFPB.³⁶ The Dodd-Frank Act also directed the agency to create a single integrated disclosure form to satisfy both TILA and RESPA disclosure requirements.³⁷ On November 20, 2013, the CFPB issued the TILA-RESPA Integrated Disclosure (TRID) final rule.³⁸

The TRID, implemented under Regulation Z, requires disclosures in connection with applying for and closing on a mortgage loan. Consumers must receive two standardized forms: the Loan Estimate form and the Closing Disclosure form. The Loan Estimate form, consisting of the good faith estimate of the loan terms and closing costs, must be delivered to the consumer no later than three business days after the consumer submits a mortgage application. The Closing Disclosure form must be received at least three days before settlement of the loan. The final costs estimates on the Closing Disclosure form may exceed the originally disclosed estimates on the Loan Estimate form only by allowable percentages (referred to as tolerances) prescribed by the CFPB.

Since the 2013 final rule, the CFPB has made various amendments to TRID. For example, the CFPB amended and clarified various mortgage disclosure provisions in 2017, including provisions pertaining to escrow closing notices, construction loans, simultaneous subordinate lien loans, tolerances for the total of payments disclosure, and other disclosure requirements.³⁹ In May 2018, the CFPB amended TRID to facilitate the ability to pass permissible costs to affected consumers under circumstances when unanticipated cost increases occurred after Closing Disclosure forms had been provided, thus mitigating the need for lenders to spread such costs to other consumers by pricing future loans with added margins.⁴⁰

The Home Mortgage Disclosure Act (HMDA)

HMDA was enacted in 1975 to assist government regulators and the private sector with the monitoring of anti-discriminatory practices.⁴¹ HMDA is implemented via Regulation C. The Dodd-Frank Act transferred HMDA rulemaking authority, which was initially given to the Federal Reserve Board, to the CFPB.⁴² Prudential regulators of depository institutions (i.e., banks and credit unions) use the HMDA data to assist with the supervision and enforcement of fair lending compliance.⁴³ The bank prudential regulators also use the HMDA data to assist with Community Reinvestment Act examinations.⁴⁴

Credit availability in less affluent and minority neighborhoods was of primary concern when HMDA was enacted. Financial institutions allegedly accepted deposits but did not make mortgage loans in certain neighborhoods. Consequently, institutions covered by HMDA were required to report home mortgage originations by geographic area, financial institution type, borrower race, sex, income, and whether the loans were for home purchase or refinance.⁴⁵ This information would show geographical patterns of mortgage originations and help regulators determine localities where further investigation of discrimination, known as redlining, was necessary. In 1989, Congress expanded HMDA to include the race, sex, and borrower income of those applicants that were rejected as well as those who were approved. This expansion allowed regulators to monitor differences in mortgage loan denial rates by income, race, and gender. The following sections discuss further amendments affecting HMDA reporting requirements.

Introduction of the HMDA Rate Spread and 2008 Modifications

While addressing mortgage credit availability concerns, regulators observed issues with respect to mortgage credit pricing. Beginning in the 1990s, credit became increasingly available for less creditworthy borrowers, particularly with wider adoption of credit scoring technology.⁴⁶ Instead of turning down loan requests for borrowers of lower credit quality, lenders began charging these borrowers higher interest rates to compensate for the additional default risks.⁴⁷ Nevertheless, the practice of charging different loan rates to different groups led regulators to question the extent that mortgage pricing patterns reflected differences in credit risk rather than discrimination.

Congress responded by expanding HMDA to include rate spread information in 2002. At the time, the rate spread was defined as the difference between the APR for a 30-year (fixed or adjustable rate) mortgage loan and the rate of a U.S. Treasury security of comparable maturity. The mortgage interest rate was not chosen because it contains only the cost of the principal loan amount expressed as a percentage. By contrast, the APR includes the cost of the principal loan amount, insurance, and other upfront fees—all expressed as a percentage. The law requiring rate spread information was implemented in 2004.⁴⁸

The initial rate spread calculations, however, contained some measurement concerns. First, excessive higher-priced lending arguably may have been undercounted due to mismatched term durations. As previously discussed, the 10-year Treasury securities better tracks the durations of 30-year mortgages due to borrower prepayment behavior. By relying on 30-year Treasury rates, which are typically higher than the 10-year Treasury rates, pre-2008 rate spread calculations may have been artificially low, resulting in a higher incidence of underreporting. Second, the initial rate spreads were highly sensitive to Treasury yield curve movements. For example, short-term rates tend to rise higher than long-term rates following a sudden demand for cash by financial institutions or deteriorating macroeconomic conditions. Over the first half of 2008, declines in the 30-year Treasury rates led to wider rate spreads and greater HMDA reporting. The reported rate spreads, however, reflected yield curve rotations rather than a greater incidence of high-cost mortgage pricing.

In October 2008, the Federal Reserve amended Regulation C to change the rate spread definition to reduce the sensitivity of the rate spreads to yield curve movements, thus enhancing the reliability of the collected data.⁴⁹ The weekly Primary Mortgage Market Survey, conducted by Freddie Mac, surveys lenders on the mortgage coupon and total charges at settlement for their most common mortgage products (e.g., 30-year fixed rate, 15-year fixed rate, certain ARM products) and reports the average of these loan terms, known as the average prime offer rates (APORs).⁵⁰ These APORs replaced the use of Treasury rates to calculate HMDA rate spreads. In other words, HMDA rate spreads are defined as the difference between the APR and the APOR. The reporting thresholds were also revised. For first mortgage loans, reporting is triggered when the rate spread equals or is greater than 1.5 percentage points. For second mortgage loans, the reporting threshold equals or is greater than 3.5 percentage points. The APORs, comprised of mortgage coupons that use the 10-year Treasury as the base rate, follow mortgage market activity more closely than the 30-year Treasury rate. Going forward, the reported rate spreads would contain less variability such that any observed deviations would likely be more reliable indicators of mortgage pricing irregularities.

HMDA Reporting Requirements and Recent Modifications

Covered institutions—or those subject to HMDA reporting requirements—include depositories (i.e., banks and credit unions) and nondepository financial institutions.⁵¹ A depository financial institution is subject to HMDA reporting if it exceeds an asset-size threshold published annually in the Federal Register, has a home or branch office located in a metropolitan statistical area, has originated at least one home purchase or refinance home loan, is federally insured or federally regulated, and meets the loan-volume thresholds discussed in the next paragraph. A banking institution must also have received a positive rating from its most recent Community Reinvestment Act examination to be eligible for HMDA partial exemptions. A nondepository financial institution is subject to HMDA reporting if it satisfies the following requirements: It has a home or branch office located in a metropolitan statistical area at which the institution accepts applications from the public for covered mortgages, and it meets the loan-volume thresholds modified by the 2018 Economic Growth, Regulatory Relief, and Consumer Protection Act (EGRRCPA, P.L. 115-174).

Prior to EGRRCPA, the Dodd-Frank Act expanded HMDA reporting requirements, specifically mandating the CFPB to include additional information about applicants (e.g., credit scores, age, debt-to-income ratios); additional information about the loan features (e.g., introductory rate period, non-amortizing features); additional information about manufactured and multifamily housing; and other information.⁵² Credit history information, for example, can help detect whether any observed pricing differentials reflect differences in borrowers' financial risks or discrimination, thus capturing more information about influences on lending decisions. However, the costs associated with the collection and transmission of large amounts of digital data are lower per transaction for firms with large volumes and higher for those with lower volumes of mortgage transactions.⁵³ Lenders, therefore, prior to implementation of the Dodd-Frank Act reporting requirements, had not been required to report numerous variables used to evaluate applicants given that (1) federal regulatory agencies could obtain loan data from financial institutions they wish to examine more closely, and (2) the HMDA data are released to the public, which could compromise the privacy of individuals holding reported loans. For this reason, the public data are currently modified to protect applicant and borrower privacy.⁵⁴

In response, EGRRCPA modified the reporting thresholds to reduce the compliance costs for entities with lower origination volumes. On April 16, 2020, the CFPB issued a final rule implementing these loan-volume thresholds for reporting open- and closed-ended loans via Regulation C.⁵⁵ Open-end credit is generally defined as credit in which the creditor reasonably contemplates repeated transactions, a finance charge is imposed on the outstanding balance, and the amount of credit is replenished to the extent the outstanding balance is repaid; closed-end credit includes all credit that does not meet the definition of open-end credit.⁵⁶ A home equity line of credit is an example of an open-end credit product, which allows borrowers the flexibility to use some or all of their credit limit as well as to repay some or all of the outstanding balance over the maturity term.⁵⁷ By contrast, a traditional mortgage is an example of a closed-end credit product with pre-determined principal and interest amounts that must be repaid in regular intervals and in full by the maturity date. The previous threshold for the reporting of closed-end mortgages, which was set at 25 loans over two calendar years, has been increased to 100 loans over two calendar years. For open-end lines of credit, the permanent threshold for reporting data is set at 200 effective January 1, 2022 (following expiration of the temporary higher threshold of 500 open-end lines of credit that was increased from 100 in 2018).

After determining eligibility to report, a financial institution must determine which reporting requirements are applicable.⁵⁸ Specifically, the HMDA data contain 48 data points (comprised of 110 fields); 26 of these data points may qualify for partial exemptions for certain depository institutions under EGRRCPA.⁵⁹ If a depository's number of originations falls below the above-mentioned loan-volume thresholds, then it qualifies for the partial exemption (although it may still voluntarily report the eligible data). Some of the data eligible for a partial exemption include reasons for denial (if applicable), the borrower's debt-to-income (DTI) ratio, credit score, interest rate, rate spread, total points and fees, and origination charges, among other things. The remaining 22 data points do not qualify for a partial exemption and must be reported. Some of the mandatory data include ethnicity, race, sex, age, income, state, county, census tract, and HOEPA status (high-cost loans discussed in the section below entitled "The Home Ownership Equity and Protection Act"), among other things.

Going forward, the partial exemptions will result in less mortgage data being reported. According to a May 2021 Government Accountability Office (GAO) report, the overall impact has been "minimal," but the partial exemptions are likely to have the greatest impact on HMDA data collected from less densely populated areas (e.g., rural areas).⁶⁰ Consequently, the ability to monitor mortgage pricing in these localities as well as the progress made with respect to the GSEs' duty-to-serve rural housing goals (established by the Housing and Economic Recovery Act of 2008 [P.L. 110-289]) may become more challenging.⁶¹ The GAO report also noted that lenders falling below the threshold for open-ended lines of credit are not required to report the relevant data points that would confirm their eligibility for the partial exemption.⁶² Amending the statute would be necessary to enhance HMDA coverage for these circumstances.

Increasing the Costs to Offer High-Cost Mortgages

Enhanced disclosures is one approach toward achieving more competitive pricing of mortgage credit. Although differences in borrowers' financial risks and variation along with regional differences in settlement costs does not allow for convergence to a single APR, greater convergence to a tighter range of prices still allows borrowers and lenders to form more accurate pricing expectations. Regulators, however, have greater data regarding creditworthy prime borrowers with similar and more predictable prepayment and default rates. The APOR used to determine HMDA rate spread eligibility is largely calculated using the mortgage terms offered to prime borrowers.⁶³ Consequently, advancement toward a more ideal LOOP outcome is more likely to occur in the primary mortgage market consisting largely of prime borrowers.

For primary mortgage markets consisting largely of borrowers with weaker or nonexistent credit histories, increasing lenders' costs to provide high-cost mortgage products may deter abusive lending practices. This approach is conceptually analogous to applying a Pigouvian tax, which raises the cost of a practice that arguably imposes a societal cost, thereby discouraging the practice.⁶⁴ This section discusses regulatory and legislative tools that impose enhanced damages (higher costs, discussed in the sections below) to lenders when offering high-cost and unaffordable mortgages.

The Home Ownership Equity and Protection Act

Although additional compensation for lenders to take elevated credit risks may be justified, the charges arguably may not be proportional to and possibly exceed the additional credit risks posed by some borrowers. Determining the appropriate amount of additional basis points lenders should charge for various gradations of elevated default risk is complicated, because fewer defaults are observed when house prices are stable or rising.⁶⁵ Likewise, determining whether excessive charges are proportional to elevated default risks or linked to unobservable factors such as inadvertent (disparate treatment) or overt discrimination—even with the use of statistical analysis—may be challenging for regulators when conducting fair lending examinations.⁶⁶

HOEPA amends TILA by imposing additional disclosure requirements on lenders when originating high-cost refinance and other non-purchase mortgages secured by their principal residences.⁶⁷ Failure to comply with all HOEPA requirements triggers enhanced remedies (e.g., monetary damages) borne by the lenders.⁶⁸ Congress initially granted implementation and rulemaking authority of HOEPA to the Federal Reserve. In 2002, the Federal Reserve made the following revisions to expand HOEPA coverage:⁶⁹

• A loan would be subject to HOEPA reporting requirements if either the APR exceeds the rate of a comparable Treasury security by more than 8 percentage points on a first mortgage or 10 percentage points on a second mortgage or if the consumer pays total points and fees exceeding the greater of 8% or a dollar amount that is adjusted annually based upon the Consumer Price Index. (This definition was modified in 2010 as discussed below.)
• If a loan satisfies certain criteria to be covered by HOEPA, a borrower must be provided with disclosures three days before the loan is closed in addition to the three-day right of rescission generally required by TILA, which means a total of six days to decide whether or not to enter into the transaction.
• Revisions to Regulation C required lenders to report HOEPA loans in HMDA, and they must also identify such loans as being subject to HOEPA requirements.⁷⁰ HOEPA, however, is implemented via Regulation Z (12 C.F.R. Part 226, Sections 31, 32, and 34).

In 2008, the Federal Reserve amended Regulation Z to apply HOEPA rules to all mortgage lenders—including non-bank entities that do not have primary prudential regulators.⁷¹ The amended rule also prohibited lenders from making loans based upon the home value without regard for the borrower's ability to repay the loan from income and assets. It also required verification of income and assets for determining repayment ability. Further, higher-cost loans may not have prepayment penalties lasting for more than two years, and prepayment penalties are not allowed for loans in which the monthly payment can change during the initial four years. Finally, escrow accounts for property taxes and homeowners' insurance must be established for all first-lien mortgages. These additional protections apply to all higher-priced loans; that is, first-lien mortgages with APRs 1.5% above the applicable APOR or subordinate-lien mortgages with APRs 3.5% above the applicable APOR.

The Dodd-Frank Act strengthened consumer protections for high-cost mortgages by revising HOEPA's coverage tests and providing further restrictions on loan terms, which were implemented by the CFPB.⁷² When the rule became effective in 2014, some of the requirements and restrictions included the following:

• A loan is subject to HOEPA reporting requirements if (1) the APR exceeds the applicable APOR by more than 6.5% on a first mortgage or 8.5% on a second mortgage, (2) the consumer pays total points and fees exceeding 5% of the total loan amount, or (3) the consumer pays total points and fees exceeding the lesser of 8% or $1,000 for loans less than $20,000 (with the dollar amounts adjusted annually for inflation).⁷³
• A loan is subject to HOEPA reporting requirements if it has a prepayment penalty extending beyond 36 months for closing or the prepayment penalty exceeds 2% of the amount prepaid.
• Balloon payments are generally banned except under certain circumstances.
• The calculation of points and fees that apply to high-cost loans was revised to include most compensation paid to a loan originator (i.e., origination, underwriting, and broker's fees); prepayment penalties; and the amount of the upfront mortgage insurance premium in excess of the Federal Housing Administration (FHA) upfront premium.
• Before making a high-cost mortgage, creditors are required to confirm that a borrower received counseling on the advisability of the mortgage by a federally certified or approved homeownership counselor.⁷⁴

Since passage of HOEPA, mortgages that would be covered by the law have made up a small share of the mortgage market and are concentrated among very few lenders.⁷⁵ HOEPA lending declined markedly after new regulations were implemented to amend the definition of high-cost mortgage to cover more types of loans.⁷⁶ Stated differently, greater disclosure requirements and legal risks significantly raise lenders' potential costs, thereby reducing the likelihood of mortgage originations with features that could exacerbate borrowers' housing cost burdens.

The Ability-to-Repay Requirement and Qualified Mortgage Rule

On January 10, 2013, the CFPB released a final rule implementing the ability-to-repay (ATR) requirement of the Dodd-Frank Act, which became effective on January 10, 2014.⁷⁷ The Dodd-Frank Act requires lenders to verify borrowers' ATR with documentation. As in the case of HOEPA violations, the Dodd-Frank Act provides enhanced damages for violation of the ATR requirement.⁷⁸ This section discusses how mortgage pricing is used to determine compliance with the ATR rule.

The ATR requires a lender to make a reasonable good faith determination of the consumer's ability to repay the loan according to its terms. Before making a residential mortgage loan to a consumer, a lender must consider and verify with documentation eight underwriting criteria for the borrower: (1) current or reasonably expected income or assets; (2) current employment status; (3) monthly payments of principal and interest on the primary mortgage lien; (4) monthly payment on any junior mortgage lien; (5) monthly payment for mortgage-related obligations (e.g., property taxes, homeowner association fees); (6) any additional debt obligations (e.g., automobile, credit card, education); (7) monthly DTI ratio or residual income; and (8) credit history.

The 2014 final rule provided multiple ways for a loan originator to comply with the ATR requirements,⁷⁹ one of which is by originating a qualified mortgage (QM). A general QM must meet certain product feature and underwriting requirements. The mortgage must fully amortize, meaning that the borrower's payments must be applied toward paying down a portion of the principal loan balance over time. A general QM cannot have a balloon or large principal payment due at the end of the loan. Furthermore, a general QM loan cannot negatively amortize, meaning that its principal loan balance cannot increase over time.

Consequently, the QM rule imposes on lenders additional legal risks that may translate into higher costs for improper underwriting, which could adversely affect borrowers' financial situations. Specifically, a general QM receives a rebuttable presumption of compliance with ATR. In other words, a borrower may still be able to prevail in court if the information, which had been presented to the lender during the mortgage application and origination process, would have indicated that the borrower's residual income was insufficient to meet living expenses after paying the mortgage and other debts. If, however, additional underwriting requirements (described below) are met, a lender originating a general QM can receive safe harbor legal protection, meaning that a borrower would not be able to assert that the originator (and any subsequent secondary-market purchaser) failed to comply with any of the required underwriting criteria.

The Initial 43% Debt-to-Income Requirement for QM Status

Limiting the borrower's DTI ratio to 43% had initially been one of the additional underwriting requirements for a loan to receive QM status under the 2014 final rule.⁸⁰ Mortgages with DTIs exceeding 43% would still qualify as QMs if they meet the eligibility requirements to be insured or guaranteed by FHA, U.S. Department of Agriculture (USDA), or Veterans' Administration and meet permanent QM standards established by each of those agencies. These agencies adopted their own QM definitions, which excluded product features they considered would impede repayment from borrowers they predominantly serve—but they did not limit DTIs to 43%.⁸¹ Mortgages with DTIs exceeding 43% could still qualify under the Temporary GSE QM (or QM patch) category. In addition, the CFPB provided originators with Appendix Q, which was a list of required guidelines to verify borrowers' incomes and debt obligations for mortgages with DTIs above 43% or ineligible for sale to the GSEs or for federal mortgage insurance.

In a January 2019 assessment report of the ATR, the CFPB found that the DTI cap of 43% may have restricted credit access.⁸² For example, borrowers with mortgages who had demonstrated the ability to repay their loans—but with DTIs exceeding 43%—experienced reductions in credit access when attempting to refinance, suggesting a lender preference for safe harbor legal protection. Many originators, also reporting Appendix Q to be unclear and complex in practice, had limited themselves to making QM loans only to avoid exposure to potential liability and litigation risks.⁸³ Consequently, the CFPB found that borrowers who applied for loans eligible for purchase or guarantee by one of the GSEs or federal agencies were less affected by the QM rule. Therefore, the Temporary GSE QM patch, along with the QM status received when mortgages are federally guaranteed, grew more important for accessing credit.⁸⁴ In other words, a policy tool designed to discourage the production of potentially harmful products may still be regressive, meaning in this case that low- and moderate-income borrowers may bear more of the costs in the form of reduced credit access because lenders receive greater legal protection when originating loans to higher-income borrowers.⁸⁵

Revisiting and Expansion of the QM Definitions

The CFPB has since amended the QM definitions to possibly achieve a better balance between ensuring consumers' ATR and access to affordable mortgage credit. Using pricing as a proxy for affordability, the CFPB released a final rule in December 2020 with the following revisions to the 2014 final rule:⁸⁶

• All general QM loans must still meet certain underwriting and product feature requirements as well as limits on points and fees. The 43% DTI ratio requirement, however, was removed and replaced with requirements that are based on the mortgage pricing, which reflects the credit quality of borrowers.
• When a lender issues a general QM, it creates a presumption that the lender has complied with its ATR responsibilities, reducing the lender's legal exposure. The level of protection afforded a lender is linked to the loan's pricing. General QMs, with rate spreads—previously defined as the difference between the APR and APOR—that do not exceed 1.5 percentage points (or 3.5 percentage points or more for subordinate-lien mortgages) qualify for safe harbor legal protection.⁸⁷ These thresholds are the same as in the 2014 final rule.
• For a rebuttable presumption general QM, the difference between APR and APOR for first-lien mortgages may exceed 1.5 percentage points, but it is limited to 2.25 percentage points under the revised final rule. Under a rebuttable presumption, a borrower can argue that the lender violated the ATR rule if the information presented to the lender during the loan application and origination processes would have indicated that the borrower's residual income was insufficient to meet living expenses after paying the mortgage and other debts.
• The 2020 final rule removed Appendix Q and granted safe harbor to creditors using other verification standards specified by the CFPB.⁸⁸
• The 2020 final rule increased the loan amount thresholds that the APR could be above the APOR for certain circumstances. For example, the thresholds for first-lien loans with smaller balances and subordinate-liens, which tend to have higher APRs, were increased to be eligible for QM status. A first-lien secured by a manufactured home (as defined under HUD regulations that establish construction and safety standards) may also be eligible for QM status.

The Temporary GSE QM granted QM status to mortgages eligible for purchase by the GSEs until January 10, 2021, or when they would exit conservatorship, whichever would have occurred earlier.⁸⁹ On October 20, 2020, the sunset date for the Temporary GSE QM definition was replaced with a provision stating that it would be extended until the mandatory compliance date of the final rule that amends the General QM definition, which was initially July 1, 2021.⁹⁰

Despite support for the 2020 amendments from some in the mortgage industry,⁹¹ the CFPB issued a final rule on April 30, 2021, to delay the mandatory compliance date of the 2020 final rule from July 1, 2021, to October 1, 2022.⁹² The termination date of the Temporary GSE QM was also extended to October 1, 2022.⁹³ The CFPB also issued a new "seasoned QM" rule.⁹⁴ Under this rule, certain non-QM mortgages could become QMs or certain rebuttable presumption QMs could become safe harbor QMs after a lender has held them in its own portfolio for a certain amount of time and the loan met certain performance expectations over the period.