The Urban Institute (UI) has looked at a final mortgage rule issued by the Consumer Financial Protection Bureau (CFPB) in late 2020 which would provide a 3-year pathway to safe harbor for loans that are a rebuttable presumption or nonqualified mortgage at origination. The study, by UI analysts Karan Kaul, Laurie Goodman, and Jun Zhu found that loan performance during the first three years of the mortgage term is a better predictor of subsequent loan performance than the rate spread.
The rule requires that both rebuttable presumption and nonqualified conventional first-lien mortgages can be deemed safe harbor 36 months after origination if they are held on the originator's balance sheet. They may also be sold once and then held as a whole loan on the buyer's balance sheet for the entire three-year period. These loans cannot be 30 days delinquent more than twice nor have any 60-day delinquencies during the three-year seasoning period.
On the same day this rule was released, December 10, 2020, the CFPB finalized a major overhaul of its 2014 qualified mortgage (QM) rule, removing the 43 percent maximum debt-to-income (DTI) ratio limit from the QM definition. This made the GSE (government-sponsored enterprise) patch redundant. The agency also instituted a new rate spread cap of 225 basis points over the average prime offer rate (APOR) as the outer boundary for the QM box.
Under the new QM rule, effective no later than July 1, 2021, loans that meet mandatory QM requirements (i.e., product restrictions, limits on points and fees, and maximum 30-year terms) and with annual percentage rates less than 150 basis points above the APOR will be deemed safe harbor, regardless of debt-to-income ratio. Loans with rate spreads between 150 and 225 basis points over the APOR will be rebuttable presumption, and those with spreads of 225 basis points or more will be nonqualified mortgages, regardless of debt-to-income ratio.
The seasoning rule provides a conditional pathway for rebuttable presumption and nonqualified mortgages to become safe harbor loans. The seasoning rule applies only to first lien fixed-rate mortgages that meet product feature requirements and point and fee limitations of the general QM loan definition. In addition, loans defined as high-cost mortgages, or HOEPA (Home Ownership and Equity Protection Act) mortgages, are never eligible.
The authors sought answers to two questions:
- Is loan performance during the first three years generally predictive of long-term performance?
- How do default probabilities for loans that are safe harbor at origination compare with default rates for loans that are not safe harbor at origination but would pass the new seasoning test to become safe harbor after three years?
They placed GSE, portfolio and private label security (PLS) 30-year fixed rate mortgages originated from 1999 to 2016 in buckets based on the three categories of rate spread at origination and on loan performance during the three years following origination (i.e., a clean payment history, one or two 30-day delinquencies, one 60-day delinquency.
Loans in each bucket were then measured as to the likelihood of going 90 or more days delinquent in years four, five, and six. The red cells show the rates that happened among loans that were rebuttable presumption or nonqualified mortgages at origination but would have passed the seasoning test to become safe harbor after three years. ("N/A" indicates buckets with fewer than 500 loans.) The last column shows the number of loans in each bucket.
Among the 2013-16 portfolio originations that were not safe harbor at origination but maintained a clean three-year pay history, those with rate spreads from 150 to 225 basis points subsequently had a 3.6 percent rate of 90-day delinquency. For those with a rate spread above 225 basis, 5.0 percent went 90 or more days delinquent in years four, five, or six. The seasoning rule would deem these two loan buckets as safe harbor after three years. Note that these percentages are less than the corresponding 90-day delinquency rates for safe-harbor portfolio loans-5.3 percent for loans that went 30 days delinquent twice and 12.2 percent for loans that went 60 days delinquent once in the first three years. The comparison with safe-harbor loans with a 60-day lapse in the first three years is especially stark. That is, rebuttable presumption and nonqualified mortgages with a clean pay history in the first three years performed much better in years four, five, and six than loans that were safe harbor at origination but went 60 days delinquent once in the first three years.
The same pattern held true for 2013-16 PLS originations. Rebuttable presumption and nonqualified mortgages with clean three-year pay histories went 90 or more days delinquent in years four, five, or six at a rate of 2.5 percent, compared with 3.8 percent of safe-harbor loans with two 30-day delinquencies or 12.7 percent of safe-harbor loans with one 60-day non-current condition. It held for 2013-16 GSE originations and more broadly for historical production across the three channels as well. The authors say this indicates that three years of loan performance is a better predictor of subsequent performance than is origination spread (which is determined by origination characteristics).
Also, safe-harbor loans constituted about 99 percent of all lending from 2013 to 2016 across all three channels, with rebuttable presumption composing most of the rest. Thus, if the CFPB's seasoning rule were in effect from 2013 to 2016, it would have applied to only about 1 percent of originations.
The authors say that a three-year seasoning pathway to safe harbor could increase lending in this segment. Higher-rate-spread conventional lending, especially in the non-GSE space, is a crucial source of credit for racial and ethnic minorities, first-time homebuyers, households with limited means, or others who do not qualify for government-backed lending, including self-employed and gig-economy workers with nontraditional sources of income. It is not a panacea, as banks and other portfolio investors tend to hold relatively few of these loans, but at the margin, the pathway will expand the credit box.
While acknowledging that the above borrowers are more susceptible to being overcharged, Kaul, Goodman, and Zhu say they cannot ignore the fact that sustainable homeownership is the only viable path to wealth creation for these households. The availability of a seasoning pathway to safe harbor will likely increase lending options available to them.