Four times a year the Federal Reserve comes out with a Senior Loan Officer Opinion Survey, an effort to see how bankers view financial trends.
The reports get a lot of attention but if you look carefully you’ll see that there are very good reasons to regard such surveys with caution.
The Federal Deposit Insurance Corporation (FDIC) says it insures deposits at more than 6,300 banks and savings associations and yet the Federal Reserve interviews only “69 domestic banks and 23 U.S. branches and agencies of foreign banks.” That’s not much of a sample and to make sure that it’s even less relevant the footnotes tell us that “unless otherwise indicated, this document refers to reports from domestic banks in the survey.”
Why survey “23 U.S. branches and agencies of foreign banks” and then not tell us the results? Maybe there are big differences in the responses when compared with domestic banks, but who knows when the results are treated like state secrets.
With the Fed survey we’re looking at the views of 69 domestic banks out of some 6,300 depository institutions and from the survey the Fed is attempting to spot banking trends. Alternatively, in the Internet era why can’t the Fed use e-mail and send out the same questions to 3,000 banks and get a more legitimate and useful set of responses? Given the cost of email, maybe even monthly reports are possible….
I bring this up because on November 2nd the Fed published its last survey of the year and when it comes to mortgages the results are widely quoted. Unfortunately, when quoting the opinion of a few anonymous bankers it’s rarely mentioned that the survey is hardly extensive and may miss what’s really going on. Not only that, but without names and data how can we tell that the Fed is reading the survey results correctly?
“Banks,” said the Fed, “reported having eased lending standards on loans eligible for purchase by the government-sponsored enterprises (GSE) and on qualified mortgage (QM) loans over the past three months on net.”
Translation: If this conclusion is correct then banks are saying it’s marginally easier to get an FHA, VA, or conventional mortgage as well as portfolio loans which meet “qualified mortgage” standards.
In fact, that’s not what the Fed found. What the survey actually says is this:
“Modest net fractions of banks indicated that they had eased underwriting standards on loans eligible for purchase by the government-sponsored enterprises (known as GSE-eligible mortgage loans) and on “qualified” but not GSE-eligible mortgage loans. In contrast, modest net fractions of banks tightened standards on government residential mortgages.
Meanwhile, the vast majority of banks continued to report that they do not extend home-purchase loans to subprime borrowers. On the demand side, modest net fractions of banks reported weaker demand across most categories of home-purchase loans. On balance, lending standards were reportedly little changed for home equity lines of credit, and demand for such loans strengthened.”
Reading the Fed findings is like trying to parse a speech from the Kremlin. If you read it quickly and at face value you’ll see that it doesn’t say the same thing as when read carefully and with caution. Here are some examples:
No Numbers, No Data
“Modest net fractions of banks” is a meaningless measure. We don’t know whether one happy lender relaxed lending standards or 31 lenders came over to the good side. There is no definition for whatever it is that “modest net fractions” might mean.
Moreover, it’s hard to imagine that underwriting standards have “eased” for FHA mortgages when some of our biggest lenders are largely out of the FHA business.
“JPMorgan Chase may be the second-largest mortgage lender in the nation,” reports Diana Olick with CNBC, “but when it comes to government-backed, low down-payment Federal Housing Administration loans, it’s not even in the top 100. Onerous regulations and the constant threat of litigation have all but stalled originations of these loans, which were originally designed to help first-time homebuyers with lower credit scores and less cash.”
“In the second quarter,” explains National Mortgage News, “JPMorgan Chase originated just 340 FHA loans, compared with 19,111 FHA loans in the second quarter of 2013. Meanwhile, the bank’s overall home lending business is booming. JPMorgan originated $29.3 billion of home loans in the second quarter, up 74 percent from a year earlier.”
Wells Fargo originated almost 75 percent fewer FHA loans between 2013 and 2014, according to Inside Mortgage Finance.
How, exactly, are standards “easing” for FHA borrowers at this huge bank when far-fewer loans from that program are being issued?
The Fed’s results do not seem especially related to reality. When it comes to commercial and industrial (C&I) financing, the Fed explains that “the few banks that reported having eased either their standards or terms on C&I loans predominantly pointed to more-aggressive competition from other banks or nonbank lenders as an important reason.”
That makes sense, so why not a parallel statement regarding mortgages? Does it matter if a few selected Fed lenders have “eased” underwriting standards if in fact they’re rapidly losing market share?
“The market share of large banks and their affiliated mortgage companies declined from 53.8 percent to 38.3 percent” between 2010 and 2014, according to the Consumer Financial Protection Bureau. Given their reduced importance, rather than polling banks perhaps the Fed should survey nonbanks and credit unions to find out what’s really happening in the mortgage marketplace.
To confuse matters further, the Fed also reports that “modest net fractions of banks tightened standards on government residential mortgages.” Again, we don’t know what a “modest net fraction” means.
Less Money Down?
Next we find out that bankers “eased underwriting standards on loans eligible for purchase by the government-sponsored enterprises (known as GSE-eligible mortgage loans) and on ‘qualified’ but not GSE-eligible mortgage loans.” This sounds like nonsense but what the Fed is trying to say is that lenders are more likely to approve applications for VA, FHA, and conventional mortgages. At the same time, they may be less willing to offer portfolio loans — loans originated and kept by lenders — which meet “qualified mortgage” (QM) standards. This is a big deal because portfolio loans that meet QM standards under Dodd-Frank rules can have terms which are attractive to borrowers who do not want FHA, VA, or conventional mortgages.
We don’t know what percentage of bankers “eased” their underwriting standards and we don’t know what “eased” means. Is the Fed saying that lenders are willing to accept applications from borrowers with lower credit scores? Less money down? We just don’t know.
The term “layering” does not appear in the Fed survey. Layering is a situation where lenders require additional terms for FHA, VA, and conventional borrowers beyond program requirements. For instance, instead of demanding a 43 percent debt-to-income ratio, to avoid iffy borrowers a lender might only accept applications for borrowers who have a 41 percent DTI. If the Fed reported less layering then we might be able to agree that lending standards had eased.
Meanwhile, says the Fed, “the vast majority of banks continued to report that they do not extend home-purchase loans to subprime borrowers.” Surely in such circumstances lending standards have not eased so where are subprime borrowers getting their loans, if anywhere?
The Fed’s Senior Loan Officer Opinion Survey offers the illusion of information without actually saying anything concrete. It’s a neat trick, but wouldn’t it be better to provide data that could be used by lenders and consumers to better understand the marketplace?