By Michael Fratantoni, Chief Economist, the Mortgage Bankers Association (MBA)
The mortgage industry in the United States changed considerably in the period following the global financial crisis (GFC). Some of these changes were the result of a new host of regulations that stemmed from the Dodd-Frank Wall Street Reform and Consumer Protection Act. This has included the implementation of the Ability-to-Repay/Qualified Mortgage (ATR/QM) rule, which has locked in more conservative underwriting and product features, and national servicing standards, which have formalized many of the loss-mitigation requirements put in place during the unprecedented level of foreclosures. Additionally, bank regulators in the US implemented—and then further refined— the Basel III capital standards, including several provisions that impacted mortgages, particularly mortgage servicing. Furthermore, the federal government pursued and obtained significant financial penalties against mortgage lenders and servicers due to mistakes made during the crisis, including seeking treble damages under the False Claims Act (FCA).
Recent trends
The net impact of these changes increased the cost and risk of mortgage lending and servicing. It is thus not surprising that many banks have stepped back from this market. Figure 1 shows that non-depositories now account for the majority of both purchase and refinance originations. Figure 2 shows a similar trend with respect to the market share for mortgage servicing. A recent collection of papers by the Federal Deposit Insurance Corporation (FDIC) {FDIC (2019)} provides a comprehensive set of explanations for why banks have become a smaller share of the mortgage market. I am going to focus here on the key reasons for the shift in mortgage servicing—despite the fact that mortgage servicing is a business in which banks have significant operational advantages, including lower cost of funds and substantially greater access to multiple sources of liquidity.
It is important to note that a growing share of mortgage originations can lead to an increase in the servicing share, as originators often have the option to retain or sell the servicing on a loan. Many non-bank originators (independent mortgage bankers, or IMBs) tend to be opportunistic—for example, retaining mortgage servicing assets (MSAs) when market values are low, selling when they are high. The lack of bank buyers of servicing in the market has depressed servicing values and thus kept more servicing with non-bank servicers. Moreover, once a lender has a sizeable servicing book, there is a benefit to trying to retain that servicing, and a borrower’s first call when seeking a refinance is often to their current servicer. Thus, the increase in servicing share can naturally lead to a higher origination share over time as well.
The issues
What were the factors that led to this marked shift in the share of mortgage servicing over time? There have been three primary drivers: the capital treatment of mortgage servicing assets, increase in servicing costs and, for Federal Housing Administration (FHA) servicing, exposure to the False Claims Act.
First, changes in capital requirements are often a powerful driver of behavior for banks. The initial implementation of the Basel III standards in the US sharply penalized the holding of mortgage servicing assets. Under the previous regime, the ratio of MSAs to Tier 1 capital was capped at 50 percent for banks and 100 percent for thrifts. With Basel III, this ratio was brought down to 10 percent. Although this cap was later relaxed somewhat in 2017 to 25 percent for banks with less than $50 billion in assets, this treatment is still quite onerous.
Moreover, in addition to lowering the cap, the risk weight on MSAs was increased from 100 to 250 percent. For many banks, this risk weight became the more impactful of the two capital provisions. Although MSAs are a Level 3 asset, and valuations can be volatile at times, it is hard to find examples of losses on MSAs leading to the failure of a bank. Thus, this treatment seems unnecessarily punitive. Banks could mitigate these increased capital charges by holding mortgage loans on their balance sheets, hence not creating MSAs. However, even for the largest banks, there are limits with respect to space on their balance sheets for mortgages, and most banks (and regulators) would prefer that they not hold long-term, fixed-rate mortgages on their balance sheets, given that most of their funding comes from short-term deposits. Moreover, originating short-term, adjustable-rate mortgages in today’s low-interest-rate environment is not a recipe for building market share.
Second, for all mortgage servicers, the business is much more costly than it was. As shown in Figure 3, the cost to service a performing loan has been running 2.5 to 3 times higher than was true before the GFC. Meanwhile, the cost to service a nonperforming mortgage has increased by four to five times over this same period. The implementation of the Consumer Financial Protection Bureau’s (CFPB’s) national servicing standards, which locked in many of the requirements initially imposed on the largest servicers as part of the National Mortgage Servicer Settlements reached during the peak foreclosure period, has been a major factor in these increased costs. Not surprisingly, as these costs have increased, many banks have reduced their servicing portfolios, either actively through sales or passively through runoffs. And other banks have tightened their credit profiles to reduce the risks of having large numbers of defaulted mortgages to resolve at high costs. However, tightening credit can impede a bank’s ability to lend to first-time homebuyers, who tend to have smaller down payments and shorter credit histories. As the Millennial cohort has reached peak first-time homebuyer age, the share of the market going to first-time homebuyers has increased, and this has further moved the market away from highly risk-averse banks. Some banks have focused to a greater extent on jumbo mortgages, working to build out a multi-product offering to affluent customers. However, this has been an extremely competitive segment of the market.
Figure 3: Cost to Service ($/loan)
Source: MBA’s Servicing Operations Study
Finally, for banks that had been involved in FHA lending, the period following the GFC highlighted the particular risks for servicers. If a servicer made an insurance claim on a defaulted FHA loan, and it was later found that an error had been made with respect to the origination or servicing of that loan, the servicer could be liable under the FCA for treble damages. These cases were pursued aggressively by the U.S. Department of Justice (DOJ), and many large bank servicers paid extremely large fines, often for very minor origination errors.
In the face of such disproportionate legal exposure, amidst a servicing process that was already cumbersome and expensive, some banks have decided to exit the FHA business, while others have cut back their participation. Servicing FHA loans requires compliance with an often outdated set of regulatory requirements that has not kept pace with the significant changes to mortgage-loan servicing that have occurred in the past several years. These requirements contribute to the increasing costs of FHA servicing, beyond the increase noted above for servicing as a whole. Comprehensive reform of FHA servicing standards and processes is a critical factor in ensuring broad industry participation in the FHA program, particularly for banks. Some servicing and False Claims Act reforms have been implemented recently, but it remains too early to tell if these changes will be sufficient to attract banks back to the program. Furthermore, concerns regarding the cost of FHA servicing have led many banks to halt or curtail purchases of Ginnie Mae servicing rights. Without bank support, the liquidity of the market for Ginnie Mae servicing has suffered.
Moving forward
The MBA is forecasting that there will be $2.8 trillion of mortgage originations in the US this year—the strongest level of activity since 2005. Outstanding home-mortgage debt totals $10.9 trillion. This incredibly important market requires active participation by banks and independent mortgage banks alike. The diversity of its business models is a strength, not a weakness, of the US market.
And the answer—for borrowers, for healthy markets—is to make the mortgage market a better place to do business, particularly by addressing the factors that I have raised here. While it makes sense to focus on the appropriate standards that non-bank servicers need to meet, the most effective means to bring banks back into mortgage servicing would be to address the factors that are impeding banks from more active participation in the mortgage market as a whole. Banks pivoted away from mortgage servicing because the changes outlined here made it a less attractive business. Changes to address these concerns could bring them back.
Michael Fratantoni is the Chief Economist of the Mortgage Bankers Association, 1919 M St., NW, Washington, DC 20036.
References
FDIC. 2019. “Trends in Mortgage Origination and Servicing: Nonbanks in the Post-Crisis Period.” FDIC Quarterly. Vol. 13, No. 4.
MBA and ICBA. 2018. “Re: Update and Request for Meeting on Finalization of Proposed Simplifications to the Capital Rule Proposal Pursuant to the Economic Growth and Regulatory Paperwork Reduction Act (EGRPRA) of 1996.” Letter to Chairman McWilliams, Comptroller Otting and Vice Chairman Quarles.
MBA. 2020. “Improving Default Mortgage Servicing Processes: Opportunities for Alignment and Standardization.” https://www.mba.org/advocacy-and-policy/residential-policy-issues/mortgage-servicing/servicing-government-loans
MBA. 2020. National Delinquency Survey:https://www.mba.org/news-research-and-resources/research-and-economics/single-family-research/national-delinquency-survey
MBA. 2020. FHA Servicing Issues Brief: https://www.mba.org/Documents/Policy/Issue%20Briefs/FHA%20Servicing%20Issue%20Brief.pdf
MBA. 2020. Servicing Operations Study and Forum: https://www.mba.org/news-research-and-resources/research-and-economics/single-family-research/servicing-operations-study-and-forum-for-prime-and-specialty-servicers