Qualified Residential Mortgages

The New York Times has a major article about the Qualified Residential Mortgage rulemaking under the Dodd-Frank Act. I think there's a lot of confusion about this ruling-making. I'm going to try and clarify a few things in this post.

What Is Actually Required by the Dodd-Frank Act.

 The Dodd-Frank Act requires that a "securitizer" retain 5 percent of the credit risk unhedged on any securitized asset unless the securitization consists solely of "qualified residential mortgages." A securitizer is basically the securitization's sponsor, although the credit risk can be shared between a sponsor and an originator of the securitized assets. 

The Dodd-Frank credit risk retention requirement, however, is not limited to mortgage securitization. It covers all types of securitizations. While this doesn't matter much for credit card securitizations, which already have skin-in-the-game requirements, it will affect auto loan securitizations, commercial mortgages, and other sundry asset classes. 

The QRM Definition

The Dodd-Frank Act does not specify how the 5% credit risk retention is to be caluculated, whether it applies to covered bonds, how hedging restrictions are to be enforced, and most importantly, it does not define "qualified residential mortgage" or QRM. Instead, it direct the OCC, the Fed, the FDIC, the SEC, the FHFA, and HUD to come up with a definition of QRM. The definition may be no broader than the CFPB's definition of "Qualified Mortgage" or QM.  In other words, all QRMs, must be QMs. The CFPB came out with its QM rulemaking last January (see here and here). 

The Dodd-Frank Act gives the regulatory ensemble a bit of guidance about how a QRM should be defined. Specifically, it requires the regulators to consider "underwriting and product features that historical loan performance data indicate result in a lower risk of default". It then gives an illustrative list of possible factors:  income and asset verification, various debt-to-income ratios, potential for payment shock, mortgage insurance and credit enhancements, and balloon payments, negative amortization, prepayment penalities, and interest-only payments.

Where Do Down Payments Fit It? 

Notably, the Dodd-Frank statutory list for factors to be considered in defining QRMs does not include down payment requirements, the feature that has everyone worked up over the QRM rulemaking. That is all to say that while the Dodd-Frank Act does not prohibit or exclude consideration of down payments from the QRM definition, down payments didn't seem to be what Congress had in mind. 

Some poorly informed (or ideologically motivated) observers have seemed to confuse QRM and QM and have criticized QM for not including a minimum down payment requirement. Frankly, it's questionable if the CFPB would even have authority to include a down payment requirement in the QM definition, which is about ability to repay. Down payments relate to likelihood of default, but they have nothing to do with ability to repay, only motiviation to repay. 

To the extent that the QM rulemaking actually ensures ability to repay, the case for an additional downpayment requirement as part of QM seems weak to me. I am a firm believer that home equity is a major factor in determining defaults, but if mortgages are made based on real ability to repay, the chances of a housing bubble are much lower, and therefore home equity is less likely to dip negative creating the walk-a-way incentive.

A much better approach than imposing a downpayment requirement as part of QRM would be to eliminate the Garn-St. Germain prohibition on due-on-sale clauses being triggered by junior mortgages. It's frankly ridiculous that a residential lender cannot make the encumbrance of a property with a junior mortgage an event of default, while a commercial lender can do so. (Indeed, the 1982 Garn-St. Germain due-on-sale prohibitions are, imho, the farthest reaching set of consumer finance regulations on the books. I can't think of anything that affects a consumer's credit profile more than death or divorce, yet neither may be treated as an event of default on a mortgage.) But that's a task for Congress; it cannot be done by regulation. (It's not clear to me what prevents other transactional responses to the problem, such as fees, rate increases, accelerated amortizations, or even triggers related to, but distinct from placement of a junior lien on a property.) 

Does Any of This Matter?

It's not clear that the credit risk retention will have much if any application to the GSEs. Fannie and Freddie currently have 100% skin-in-the-game on most MBS.  It's not clear how PMI coverage will be treated (is that hedging?) and how things like the Freddie Mac K-series (with unguaranteed subdebt) will be treated, but most mortgages in GSE pools are shouldn't be affected.  This means that what is really at stake in the QRM definition is the viability of private-label mortgage securitization. With QM in place, it's hard to imagine private-label competing with the GSEs in the conventional, conforming loan space. Instead, private-label is going to be about Jumbos and possibly about subprime competition with FHA. I'm skeptical that it will be much more than Jumbos. 

Stepping back, the whole QRM debate seems rather silly to me. The definition of QRM is simply irrelevant as long as the private-label securitization market is dead. The private-label market has a bunch of things holding it back. There are a host of servicing issues (compensation and monitoring), there's Basel III capital requirements for MSRs, and then the entire rep and warranty putback experience suggests that bankruptcy remoteness is just a fiction--as long as investors place value on reps and warranties, the creditworthiness of the sponsor is actually quite important. There could be a complete regulatory waiver of credit risk retention, and I don't think it would matter very much by itself in terms of restarting the private-label market. 

But this actually raises one of the puzzles about QRM. If private-label ends up being mainly a market for jumbos, the debate over down payment requirements looks kind of different. A higher downpayment requirement for a jumbo doesn't seem like such a big deal--it's not keeping someone out of the homeownership market in general, just out of certain neighborhoods (or perhaps MSAs for SF and NYC). This suggests, though, that requiring say 20% downpayments would put downward pressure on fancy urban and suburban home prices.

I would expect that to create a powerful constituency (realtors, builders, and current property owners) pushing against any kind of skin-in-the-game requirement. It would also seem to create a bi-partisan push of monied voters against skin-in-the-game, much like with the home mortgage interest deduction. It's not clear to me if this has happened yet. As it standard, the QRM rule-making is kind of stacked against the "access to credit" concern.  QRM is supposed to be a joint rule-making between the OCC, the Fed, the FDIC, the SEC, the FHFA, and HUD.  This is kind of strange, as the CFPB is the only federal agency tasked with "ensuring that all consumers have access to markets for consumer financial products and services". HUD has an access mission of sorts via FHA, but access to credit isn't part of the OCC, Fed, FDIC, SEC, or FHFA missions.  So to the extent that QRM is supposed to be balancing investor/market protection with access concerns, the access voices aren't in the room.  

Finally, it's worthwhile noting that the Dodd-Frank Act permits regulators to exempt securitizations from credit risk retention aside from QRM.  The statute separately permits "total or partial exemption of any securitization, as may be appropriate in the public interest and for the protection of investors," and specifically authorizes exemptions of the securities of federal government agencies (excluding GSEs) and municipal securities. It's not clear if the QRM exemption circumsribes the scope of the direct exemption. I would think it should, but there's a reasonable textual argument that it does not. If it doesn't, then the definition of QRM is beside the point, as regulators will have carte blanche to exempt what they want. 

All of this is to say that I'm agnotistic on the down payment requirement issue, because I think it misses the problem of second liens and won't realistically affect much beyond Jumbos. 

 

 

Comments

[mt]

After the 1929 stock market crash, legislation and regulations were adopted that established minimum equity positions in stocks (or maximum percentages of margin loans against equities). Since then, we have never had a stock market crash trigger a financial crisis. Rather, when someone's margin ratio gets too high, enough of their equities get liqudiated to repair it; that process, repeated across thousands of margin buyers, keeps market leverage under control and confines losses to the stock market, rather than having them cascade through the banking system via margin loan defaults, as happened in 1929-30.

Unfortunately leverage ratios in respect to home mortgages were not regulated for financial system safety. They were basically lowered to pander to voters and developers and bankers who were campaign contributors. 5% down, 3% down, etc. The government put FHA, Fannie and Freddie into the middle of that risk and then allowed them, even demanded them, to further increase the risk for private gain and public loss, mainly because the CEOs were well connected in DC and it was a benefit politicians could show to certain voting blocs. Had there been higher equity required, the crisis would never have happened. We should learn from the effectiveness of the stock market margin regulations and maintain sound down payment requirements on residential mortgages. That way they won't trigger another systemic crisis.

You write about opponents of your ideas being ideologically motivated. But you yourself have a clear ideological bias, so the contention seems very disingenuous.

[Adam]

mt--There are two problems in your reasoning.

First, you're mistaking my questioning of the push for mandatory down payments as part of QRM for opposition to the idea. I'm fairly agnostic, not least because I think QRM doesn't matter very much except for Jumbos. (btw, I'm curious what you think my "clear ideological bias" is here. I really don't have a definite opinion on mandating LTVs, much less on what level.)

Second, home mortgages aren't like stock. If homes were merely financial investments, then it would make sense to regulate them like other financial investments. But homes are also shelter. If I buy on margin and without recourse, I walk away when my position is out of the money. There's a walk-a-way problem for negative equity homes too, but it's not such an automatic decision. The costs to walking away from a home are much greater than walking away from stock purchased on margin.

There's a conversation to have about mandatory minimum LTVs. But QRM isn't the way to approach it.

[Brian Mayers]

I don't see any indicators of bias in your post either Adam. But there is one question I would like to ask: What do you feel mortgages are closer to in terms of manageability? I'm now a bit thoughtless...