By Karen Shaw Petrou and Basil N. Petrou
Last Thursday, the Senate Banking Committee considered the confirmation of Brian Montgomery to be the Trump Administration’s Federal Housing Administration (FHA) Commissioner, allowing him to step back in to the shoes he filled in the George W. Bush Administration. But, times are different now – as we’ve noted before, the U.S. is far less economically equal than it was even in 2007 and the residential-mortgage market largely serves only the most creditworthy, wealthiest households.
New HMDA data show that low-and-moderate income (LMI) households went from a 28% purchase-loan share in 2015 to a 26.2% one at the end of 2016, a 6.4% drop in just one year. Even more striking is that LMI loans have fallen 28% since 2009. We draw on new Federal Reserve research to show how FHA contributes to this crisis and what should quickly be done to remedy it.
Big Banks Gone Missing
A new Federal Reserve staff paper finds that three very large banks once the crux of U.S. mortgage finance – Wells Fargo, BofA, and JPMorgan – originated 43% of FHA business in 2010, but dropped to just 5% last year. This move away from FHA is found to account for about 75% of lost LMI mortgages because LMI borrowers cannot readily substitute lost credit from large banks with loans from other lenders.
One might have thought non-banks would fill the big-bank gap. Recent data from the Urban Institute show that non-banks have increased their overall mortgage origination share from 30% in 2013 to 60% in July of 2017. Further, these same data show that non-banks now originate 76% of Ginnie Mae mortgages (which largely consist of FHA and VA loans).
However, the large drop in overall LMI lending and FHA’s critical role in this sector show that the Fed’s study is right: for FHA volume to increase to levels sufficient to reverse LMI ownership losses, big banks have to come back to the government’s table. Although non-banks are taking over from banks, they are not serving borrowers with the higher-risk profiles that generally characterize the LMI sector. The Urban Institute data cited above show that the median credit score for non-bank borrowers is 715, just a bit below the 749 median found at banks. Subprime borrowers – who are not necessarily any riskier than borrowers with these higher scores but are more often LMI, according to a new NBER paper – generally have scores below 660 with FHA agreeing to guarantee mortgages down to a 500 score (although borrowers with credit scores below 579 are required to have a 90 Percent LTV ratio).
What Would Bring Big Banks Back
The new Fed paper also looks hard at FHA-participation criteria, finding that higher premiums, indemnification risk, and servicing costs have reduced overall participation rates. Premium changes affect all lenders more or less equally, but indemnification risk and servicing costs are directly affected by large-bank rules not applicable to non-banks. Indemnification risk costs the biggest banks in terms of the direct penalties they paid – billions in recent years – under the False Claims Act and indirectly because their regulatory stress tests and operational-risk capital standards charge them up-front for this risk.
Bank rules essentially take the once-burned, twice-shy approach to FHA and thus impose a significant capital penalty for any big bank that stayed in the program. Hypothetically, indemnification risk applies equally to all lenders (FHA has in fact sought a large fine from a big non-bank). Practically, though, only large banks pay a prospective penalty for this prospective risk, making it real even if it never materializes. Free from like-kind capital rules, non-banks have nothing but future profits on the line from indemnification risk, and of course non-banks with non-profits need not fear the FHA’s reaper. If they can’t pay up, shareholders and management enjoy what they earned on the upside and leave taxpayers to clean up the mess.
The Basel III capital standards also impose a significant capital charge for mortgage-servicing assets, one reason non-banks have grown so big so fast not only in Ginnie Mae, but also across the entire government-dominated secondary mortgage market. Servicing rights vary dramatically in value, but only banks need to hedge this bet with costly regulatory capital.
What FHA Must Do
Bank regulators will not relax their penalty charges for FHA for as long as FHA retains the right to use the False Claims Act and its other powers – after all, this is a real risk and prudential regulators do not like risk. FHA on its own cannot change the other capital rules that adversely affect bank participation, but it can and should explain to the banking agencies how costly these standards have proved as a critical U.S. economic-equality consideration: sustainable, prudent lending for LMI home ownership. The banking agencies have proposed to change the mortgage-servicing capital rules, but only for smaller banks. As the Fed study above demonstrates, FHA needs big banks and big banks will not become FHA lenders as long as the cost of doing so remains prohibitive.