Some interesting information was released yesterday by the Office of Thrift Supervision with respect to third-quarter results for the thrift industry. While profitability remains high (1.15% return on assets, unchanged from a year earlier, and 12.43% return on equity, down 15 basis points from a year earlier), net interest margins continue to be squeezed (2.76%, the lowest level in two years). Moreover, net interest income was down from the second quarter, with profitability being sustained only by gains in fee income and lower expenses and taxes.
As I’ve previously discussed, banks and thrifts are being squeezed by the flattening yield curve, and are turning to fee income. Inasmuch as thrifts are traditionally mortgage lenders, those that “stick to their knitting” are counting on loan origination and, for those that retain loans in portfolio, servicing fees to offset the decline in interest spreads. Expenses can only be cut to a certain extent before productivity is negatively impacted, so for most thrifts, cost-cutting is not an avenue that will yield much more benefit over the long haul.
Last Thursday, OTS Director John Reich, in a speech to the Community Bankers Association of New York State, warned lenders who have been “rolling out” new “non-traditional” mortgage loan products, such as “interest-only” and “pay option” adjustable-rate mortgages, to do so cautiously. Those lenders who are new to such products (and Reich seemed to imply that most of the industry fell into this category) should limit their concentration to a “prudent” percentage of capital until their experience with such riskier products gives them the more sophisticated management information systems and risk management expertise that such loans demand.
Reading between the lines, the OTS doesn’t want to see the thrifts it regulates respond to the decreasing demand for traditional fixed-rate mortgages and ARMs by jumping head-first into riskier mortgages in order to sustain mortgage loan volumes. It should be expected that the OTS will criticize undue concentration in such mortgages except in the case of institutions who have carved out special expertise and, even then, underwriting and other risk management practices and procedures will receive special attention.
I also think that the OTS will increasingly be looking at underwriting and risk management in the loan origination process. Even if a lender sells most of the loans it originates, and, theoretically, passes the risk of default on to the buyer of the loan, there remains an elephant lurking in the room: the risk posed to mortgage bankers from the representations and warranties made by them when they sell loans in the secondary market. Whether the loan is conventional or “non-traditional,” when a mortgage originator sells the loan, it makes extensive representations and warranties to the buyer with respect to the borrower, the property securing the loan, the mortgage instruments, the underwriting, and numerous other “qualitative” issues. In good times, when the percentage of defaults is relatively manageable, the purchasers of such loans (who customarily bundle and securitize them) tend not to look closely at these representations and warranties; however, in bad times, the holders of the loans have been known to require a second “scrubbing” of the loan files, looking for breaches of representations and warranties that will justify requiring the originator to repurchase the loan. In the last economic downturn, I personally experienced more than one instance of a mortgage lender suffering the infliction of extreme financial and emotional pain caused by buyback demands from secondary market purchasers like FNMA, FHLMC, and private institutional investors. A “pure” mortgage banker, who holds and services few loans, may think he’s passed on the risk (absent outright fraud). Sophisticated originators know better.
As interest rates continue to rise (shutting more potential borrowers out of the market), and the yield curve continues to flatten (increasing the dependence of lenders on fee income generated by loan volume), the pressure to “let slide” or even “cover over” defects in the borrower or property, or other problems uncovered in the underwriting process, will increase. When the cycle turns (as it always does) and defaults rise, those originating lenders who sacrificed sound underwriting in return for fee income will find the grim reaper knocking at their door once again, whether or not they own the loan.