House Financial Services Committee Chairman Jeb Hensarling (R-Texas 5th) has an alternative fact problem. In a Wall Street Journal op-ed Hensarling alleged that "Since the CFPB’s advent, the number of banks offering free checking has drastically declined, while many bank fees have increased. Mortgage originations and auto loans have become more expensive for many Americans.”
The problem with these claims? They are verifiably false. Free checking has become more common, bank fees have plateaued after decades of steep increases, and both mortgage rates and auto loan rates have fallen. One can question how much any of these things are causally related to the CFPB, but using Hensarling's logic, the CFPB should be commended for expanding free checking and bringing down mortgage and auto loan rates. Hmmm.
Below the break I go through each of Chairman Hensarling's claims and demonstrate that each one is not only unsupported, but in fact outright contradicted by the best evidence available, general FDIC and Federal Reserve Board data.
Decline in Free Checking? Alternative Facts!
Chairman Hensarling doesn’t actually say that the CFPB’s responsible for the decline in the number of free checking accounts. Instead, he implies a connection. But that's exactly the conclusion he wants readers to reach, otherwise the observation is irrelevant to his attack on the CFPB. Three things are problematic about his claim on the number of free checking accounts.
- There’s no reliable evidence that the number of free checking accounts has actually declined. In fact, the American Bankers Association’s own numbers show that free checking has increased from from 53% to 61% of accounts. Curiously, Chairman Hensarling doesn’t give the CFPB credit for that.
- Other talking points Hensarling blames the Durbin Interchange Amendment, not the CFPB for the supposed decline in free checking. So which one is it, Durbin or the CFPB?
- It’s hard to see what the mechanism would be between the CFPB and a decline in free checking. The CFPB has done very little in respect to deposit accounts: there was (a) an enforcement action against Wells Fargo for making up fake accounts, (b) an enforcement action against RBS Citizens for failing to credit consumers with the full amount of deposits, and (c) the remittances rule required by Congress's amendments to the EFTA. There's no question that remittances rule imposed some extra costs of banks and money transmitters, but given what a small part of most banks' business remittance payments are, no one seriously suggests would affect checking account pricing. In, not least because there are many banks that don’t offer remittances, and those that do have to compete against them. Indeed, look no farther than State National Bank of Big Spring, Texas, which has sued the CFPB, alleging that it is unconstitutional, said that it stopped doing remittances because of the rule. If so, why would that affect its checking pricing, unless the 25-75 remittances it did per year were so overpriced that they could materially subsidize checking?
Increase in Bank Fees? Nope. Just Alternative Facts!
Chairman Hensarling also implies that the CFPB is responsible for "many bank fees hav[ing] increased." He doesn't specify fee type, and there's precious little data that breaks down specific bank fees. The best numbers coming from a CFPB research paper. (Ironically, Hensarling wants to kill of the CFPB’s research function. That way there’ll be no way of showing that alternative facts are, well, fiction. But if bank fees have gone up, it would primarily be in terms of overdraft, which is already over half of bank fees (again, we know this only from the CFPB’s research).
In any event, we are once again dealing with "alternative facts." The FDIC tracks total service fees for deposit accounts. That covers all types of fees (maintenance, overdraft, transaction, etc.). The chart below shows that these fees haven't increased because of the CFPB. They've been basically flat since the CFPB became effective. There's a causal story that relates to the Fed's overdraft regulation (that is now the CFPB's overdraft regulation), but it's the exact opposite story than the one Hensarling is telling.
Increased Cost of Mortgages? Nope. More Alternative Facts!
The last time I checked, mortgages were a helluva lot cheaper than they have been for decades. The CFPB has done major work in the mortgage space: the QM ability-to-repay rulemaking, the Reg X servicing rulemaking, the TILA-RESPA integrated disclosures, and various enforcement actions. But there's no obvious causal effect on mortgage interest rates that anyone has teased out. (There's one badly done study that points to a limited impact; I've addressed the methodological flaws in detail in an earlier post). Look at the data on interest rates for the standard American mortgage, the 30-year fixed-rate mortgage. The CFPB doesn't seem to be having any effect on the rates. Now you will notice that mortgage rates spiked in May and June of 2013. But at the time no one thought that had anything to do with the CFPB's rulemaking that was finalized in January 2013. Instead, it was driven by an expectation of a change in Federal Reserve monetary policy. And after the rule went into effect in January 2014, rates have fallen. Let me be clear--I am not making a causal claim. But Hensarling certainly is, and the data do not support his claim.
Increased Cost of Auto Loans? Nope. Even More Alternative Facts!
Let’s take a look at auto loans. The Fed has data on auto loan pricing. Guess what. Auto loans have gotten cheaper since the CFPB came on the scene. That’s true both in absolute terms and if one backs out the cost of funds (I used Fed Funds, perhaps a 5 year constant maturity Treasury would be better, but there’s no directional difference here). In absolute terms a 48 month auto loan was at 5.89% when the CFPB went operational in Q3 2011. It was at 4.69% when the CFPB issued its indirect auto lending guidance at the end of Q1 in 2013. And it was at 4.45% at the end of 2016. The numbers excluding cost of funds are 5.81% in Q1 2011, 4.55% in Q3 2013, and 3.91% in Q4 2016. The same holds true for 60-month car loans from commercial banks. The story isn’t quite as strong for new car loans from auto finance companies (excluding cost of funds): 4.37% in Q3 2011, 4.65% in Q1, 2013, and 4.57% in Q3 2016. In other words, no real change, but lending volumes have increased, up 12% between Q1 2013 and Q3 2016, suggesting that lenders have not be constrained by regulations. Moreover, increasing volume with steady prices would indicate that if demand were held constant, that prices would in fact be down. Look at the charts below for more details. The numbers don’t lie. Jeb Hensarling is pushing a huge regulatory reform based on, um, “alternative facts.”
(The first chart is unadjusted rates from Federal Reserve Statistical Release G.19. The second chart is adjusted rates that take the G.19 data and back out the end of quarter Fed Funds rate using FRED data. I did a shorter duration because I'm lazy and things before 2008 don't really matter here.)
Banking Killing and Job Killing? Huh?
This argument truly puzzles me. Here’s what Chairman Hensarling says, “With consumer protection outside the democratic process, consumers were harmed by a reduction in competition. With fewer lenders serving fewer borrowers, fewer businesses employed fewer workers.” Huh? First, what on earth does consumer lending have to do with businesses employing workers? The CFPB doesn’t have anything to do with business credit, so I don’t see what the second part of the claim, “fewer businesses employed fewer workers”. And as far as the first part of the claim, the “fewer lenders serving fewer borrowers”, where is the evidence is there that the CFPB has resulted in a reduction in competition? Yes, the number of banks has declined over the past several years, but it’s done so at the exact same steady rate of 316 banks per year that it has done for the last twenty-six years. (The chart gives the R^2 figure that reports the degree to which the y-axis variable here (number of depositories) is predicted by the x-axis variable (year). Over 95% of the change is predicted simply by passage of time, not regulation, market conditions, etc. That's pretty amazing.) To suggest that the CFPB is killing the banking industry is ludicrous in the face of this evidence. There’s just no evidence that can be used to even infer, much less confirm a causal link to the CFPB (or any other number of things that affect banks’ propensity to lend). There’s no evidence of the CFPB resulting in a material decline in the number of financial institutions, and indeed, it’s pretty hard to point to any particular CFPB regulation that would have driven anyone but a bad apple out of business. In any case, even if the number of lenders declined, there’s still plenty of lenders in the US and no shortage of lendable funds. It’s all well and good to throw out claims like “job killing,” but this is as wacky and frankly irresponsible as the Obamacare Death Panels claims. Alternative facty, fact, facts...
Bottom line: Jeb Hensarling's claims about the CFPB are based on a set of utterly concocted alternative facts. This is not the way we should be making policy.