Case Study 34.1 — Stated Income, and What the Ability-to-Repay Rule Actually Changed

Type: real public regulatory and market history Sources: Tier 1 (statutes, rules, and the public record of the 2008 crisis); Tier 2 where a specific figure or date is revised or where our confidence in a detail is short of certain

Why this case study exists

Every few months a loan officer somewhere tells a borrower that non-QM lending is "what we used to call stated income." Sometimes they say it apologetically. Sometimes they say it as a selling point. Either way it is wrong, and the wrongness is not a matter of nuance — the two things differ on the single dimension that matters most, which is whether anybody verifies the borrower's ability to pay.

This case study walks the actual history: what the pre-2008 market did, what happened to it, what Congress and the Consumer Financial Protection Bureau built in response, and what the market that emerged afterward looks like. It is the argument behind §34.1 and §34.2, and it is the reason a loan officer can sell a bank statement loan with a clear conscience and cannot sell a "no-doc" loan at all.

Background: what the pre-2008 market actually offered

By the middle of the 2000s the American mortgage market had produced a family of products defined by what the lender agreed not to check. The industry named them by acronym, which is usually a sign that something has become routine:

Alongside these ran payment structures that made the monthly obligation smaller than the loan's actual economics: interest-only periods, and payment-option adjustable-rate mortgages that allowed a payment smaller than the accruing interest, so the balance grew — negative amortization. A borrower could be qualified on a payment that was, by design, not enough to service the debt.

The internal logic of the whole arrangement is worth stating plainly, because it explains the Ability-to-Repay rule's most important clause. Underwriting the borrower is expensive and slow. Underwriting the collateral is cheap and fast, and if house prices always rise, a borrower who cannot pay can always sell or refinance, and the lender is made whole either way. That premise was never written down as policy, but it was the premise, and it held right up until house prices stopped rising.

Contemporaneous industry research raised alarms about the accuracy of stated incomes well before the market broke. The Financial Crisis Inquiry Commission, the body Congress established to investigate the causes of the crisis, documented the growth of these products and quoted industry research on income exaggeration in stated-income files in its 2011 report. (Tier 1 for the Commission and its report; the specific sampling figures it cites are worth reading at the source rather than repeating from memory — see the further reading list.)

The instruments themselves were only half the mechanism. The other half was that these loans were funded through private-label securitization rather than through the agencies, so the guidelines were written by whoever was assembling the deal. Chapter 28's account of the secondary market explains why that matters: when the buyer of the loan sets the rules, the rules are exactly as strict as the buyer's appetite, and in 2005 the appetite was enormous.

The issue: what failed, and in what order

The failure sequence is documented and is worth knowing as a sequence rather than as a single event.

1. House price appreciation stalled and then reversed. The premise underneath the whole structure — that collateral would cover any borrower failure — stopped being true.
2. Borrowers who could not afford their loans could no longer refinance out of them. The exit that made an unaffordable payment survivable closed.
3. Payment resets and recasts arrived on schedule. Interest-only periods ended; option ARMs hit their recast triggers; payments stepped up on loans that had been underwritten to the initial payment. §34.8's arithmetic — a payment that jumps at recast on a balance that never went down — is not a hypothetical. It was a national event.
4. Delinquencies exceeded what the securities' models had assumed, the value of private-label mortgage securities collapsed, and the market that funded these loans ceased to exist within months.
5. Fannie Mae and Freddie Mac were placed into conservatorship in September 2008 — a Tier-1 documented event — and the private-label channel that had funded stated-income lending did not return in anything like its prior form.

The private-label securitization market's disappearance is the part loan officers most often overlook, and it matters for this chapter. Stated-income lending did not end because it was outlawed. It ended, in 2007 and 2008, because nobody would buy the loans. The law arrived afterward, to make sure it stayed ended.

What Congress and the CFPB built

The Dodd-Frank Wall Street Reform and Consumer Protection Act (2010) contains a title devoted to mortgage reform. It added to the Truth in Lending Act a requirement that a creditor make a reasonable and good-faith determination, before consummation, that the consumer has a reasonable ability to repay a residential mortgage loan according to its terms. It also created the Consumer Financial Protection Bureau and moved rulemaking authority for most of the mortgage consumer statutes to it.

The CFPB implemented the requirement through Regulation Z. The Ability-to-Repay/Qualified Mortgage rule took effect on January 10, 2014. Its architecture is the two-box picture in §34.1:

• Every covered consumer mortgage is subject to Ability-to-Repay, with a minimum set of underwriting factors the creditor must consider and a requirement that the information be verified using reasonably reliable third-party records.
• A subset of loans meeting a defined set of product-feature, points-and-fees, and underwriting criteria are Qualified Mortgages, and receive either a safe harbor or a rebuttable presumption of compliance depending on whether the loan is higher-priced.

Read the QM product-feature restrictions and you are reading a list of the previous decade's failures, itemized: no negative amortization, no interest-only, no balloon payments outside narrow exceptions, no terms exceeding thirty years, and a cap on points and fees. Each prohibition is a scar, which is the argument Chapter 2 makes about the entire rulebook.

The QM definition has been revised since. The original General QM definition included a debt-to-income limit together with a prescribed methodology, and a temporary category (widely called the GSE Patch) treated loans eligible for purchase by Fannie Mae or Freddie Mac as qualified mortgages while those enterprises remained in conservatorship. In 2020 the Bureau replaced the General QM's DTI limit with a price-based threshold keyed to the loan's annual percentage rate relative to the average prime offer rate, and separately created a Seasoned QM category for loans that perform for a defined period. Compliance dates for those amendments were adjusted more than once.

Tier 2 discipline applies to every date and threshold in the preceding paragraph. The architecture is stable; the parameters are not. Read the current rule text before you rely on any of it, and ask your compliance department which version of the General QM definition your institution is applying.

What it shows: the four differences that actually matter

Put the pre-2008 product family next to the modern non-QM product family and the differences are not subtle.

There is a fifth difference that is easy to miss and is arguably the most important: the direction of the documentation burden. Pre-2008 stated-income products reduced documentation to speed origination. Modern non-QM programs frequently require more paper than an agency file — twelve or twenty-four months of statements, all pages, with every non-revenue deposit explained; a licensed preparer's letter; a business license verification; heavier reserves. The forms are unfamiliar, not absent.

Outcome: a market rebuilt on a different footing

A private, non-agency market for loans outside the QM definition re-formed over the second half of the 2010s. It funds the products in this chapter. It is materially smaller than the pre-crisis private-label market and it is built on a different premise — that the loans in the pool were underwritten to a documented repayment analysis, because if they were not, the securities inherit a legal defect and not merely a credit one.

That market has also demonstrated that it is thinner than the agency market, which is §34.9's central point stated as history rather than as pricing. In the spring of 2020, when credit markets seized during the early weeks of the COVID-19 disruption, non-QM origination largely stopped: lenders pulled programs, locks were repriced or withdrawn, and borrowers in process discovered their loan no longer existed. Agency lending continued throughout. Nothing about the borrowers changed. The buyers changed. That is what a thin investor base means, and it is why the premium exists.

The lesson

Ability-to-Repay did not outlaw lending to people with complicated incomes. It outlawed lending without checking.

That is the whole distinction, and a loan officer who holds it firmly can do three things others cannot. They can sell a bank statement loan without apology, because it verifies income from a third-party record. They can recognize a proposal that is not lawful — "no income verification" on an owner-occupied purchase — from a single sentence in a voicemail. And they can explain to a borrower who lived through 2008 why this is not that, in language the borrower will believe, because it is true.

One more lesson sits underneath. The pre-crisis market did not decide to make bad loans. It made loans on a premise — that collateral value would cover borrower failure — that was invisible because nobody had to state it. The clause in the Ability-to-Repay rule that says the creditor must consider income or assets other than the value of the dwelling securing the loan exists to make that premise unavailable forever. When you are looking at a file and thinking "there's a lot of equity here," you are standing exactly where that clause was aimed.

Discussion questions

1. Stated-income lending ended in the market roughly two years before the rule that prohibited it took effect. What does that sequence suggest about the relative power of investor appetite and regulation in this industry? Does it change how you read Chapter 14's guidelines-versus-overlays distinction?

2. The QM product-feature restrictions read as a list of prior failures. Choose one — negative amortization, interest-only, balloon payments, or the points-and-fees cap — and argue both sides: what legitimate borrower need did it serve, and why was it restricted anyway?

3. A borrower says, "My parents lost their house in 2008 on one of those loans where nobody checked anything. Is this the same thing?" Write your answer in under ninety seconds of speech, using the four differences in the table above.

4. The Bureau replaced the General QM's debt-to-income limit with a price-based threshold. What is the theory of that substitution — what does a loan's price supposedly tell you that a ratio does not? Name one way that theory could fail.

5. In spring 2020 non-QM programs disappeared for weeks while agency lending continued. If you had six non-QM files in process at that moment, what would you have done on day one, and what would you build into your process now to make that day survivable?

6. This case study argues that modern non-QM often requires more documentation than an agency file. Why, then, does the phrase "no-doc" persist in the market's vocabulary — and what does its persistence cost borrowers?