Case Study 1 — When FHA Made Its Insurance Permanent

A real, public case. The FHA's mortgage insurance premium structure and duration rules have been changed several times; this account describes the 2013 change and its logic. Current factors and duration rules must be verified with HUD.

Background

The FHA's insurance fund is not a taxpayer appropriation in the ordinary case. It is a mutual mortgage insurance fund — premiums paid by FHA borrowers pay claims on FHA loans that default. The statute requires the fund to hold capital reserves against expected losses, expressed as a ratio.

Through the crisis and its aftermath, that fund came under severe strain. FHA's market share rose sharply — precisely because it was serving borrowers the conventional market had stopped serving, which is what FHA is for — and the loans it insured in the mid-2000s performed badly. By the early 2010s the fund's capital ratio had fallen below its statutory minimum, and FHA required a mandatory appropriation from Treasury.

The agency's response was a series of premium changes, and one of them altered the product permanently.

The operating issue

Before 2013, FHA's annual mortgage insurance premium terminated. The rules varied over time, but the general structure was familiar: once the loan reached a specified loan-to-value and had been held for a minimum period, the annual premium stopped. It behaved, roughly, like conventional private mortgage insurance.

That structure created a specific problem for the insurance fund, and it is worth understanding because it is a genuine actuarial argument rather than a revenue grab.

FHA loans do not default early. Default risk on a mortgage is not front-loaded the way intuition suggests. It builds, peaks several years in, and persists. A borrower who reaches 78% loan-to-value has usually done so through amortization and appreciation over roughly a decade — and remains capable of defaulting for the twenty years after that.

So under the old rules, FHA collected premiums during the period of rising risk and stopped collecting them at roughly the point the loan entered a long tail of continuing risk. The insurance remained in force. The premiums did not.

What happened

In 2013 FHA changed the duration rule. For most new loans with terms greater than 15 years:

(As described at the time. Verify current rules with HUD.)

The threshold is the interesting part. It is set at 90%, not 80%, and it is measured at origination — not as the loan amortizes. A borrower who puts 3.5% down can never reach the 11-year category, no matter how much they pay down or how much the property appreciates. The category was determined the day the loan closed.

This means the standard FHA loan — 3.5% down, 96.5% LTV — carries mortgage insurance for 360 payments.

What it shows

1. A single duration rule can dominate every other difference between two programs. This is the finding that makes §5.8 the most important section in Chapter 5. On the Linden Street file, FHA wins on rate, wins on monthly payment, and wins on cash required at closing:

And loses, decisively, on the only figure that is not visible in a quote:

A loan officer who quotes monthly payments and stops has given the borrower the worse answer while being perfectly accurate about every number they said.

2. The escape hatch is a refinance, which is not free. The standard advice is that an FHA borrower should refinance into a conventional loan once they have 20% equity, thereby ending the MIP. That is correct and it is not costless: it requires a new loan at whatever rate exists then, new closing costs, a new appraisal, and requalification with whatever credit and income the borrower has at that point. An exit that depends on future market conditions is an assumption, not a plan — the same category of assumption Chapter 2 identified in the 1920s balloon and the 2000s teaser ARM.

3. It was not a bad decision. The fund had a real actuarial problem and the premium structure did not match the risk profile of the insurance. The change worked; the fund's position improved substantially in subsequent years. This case is not an indictment of FHA — it is an illustration that a program's cost structure is set by the program's own solvency needs, and those needs are not about your borrower.

4. The threshold at 90% and measured at origination is the part practitioners miss. Two things follow that are worth knowing cold. A borrower putting 10% or more down on an FHA loan gets the 11-year category. And a borrower who puts 3.5% down and later reaches 50% LTV is still in the life-of-loan category, because the test was applied once, at closing.

The outcome for the practitioner

Always price conventional against FHA. Never guess. §5.10's Rule 1 exists because of this case. The comparison turns on a feature that does not appear in a rate quote or a payment, and it moves with the borrower's score, down payment, and horizon.

Quote the total cost of the mortgage insurance, not the monthly amount. Two numbers, side by side: \$24,218.86 and \$62,374.40. Borrowers understand this instantly, and almost nobody shows it to them.

Ask the horizon question. If the borrower will refinance or sell in four years, the \$38,155.54 is largely theoretical and the \$5,775 of down-payment relief is real. If this is the house they intend to raise a family in, the reverse. The right answer is a fact about the borrower, which is the argument of this entire chapter.

And know the 10%-down FHA case. A borrower with 10% available who is using FHA for credit reasons rather than cash reasons should be shown the 11-year category explicitly. It is a genuinely different product and many loan officers do not know the threshold exists.

Discussion questions

1. FHA's actuarial argument was that premiums stopped just as the long tail of risk began. Evaluate it. What would you need to know to decide whether the change was correctly calibrated rather than merely directionally right?

2. The duration category is set at origination and never revisited. Argue for that design — it has real administrative virtues — then argue against it from the borrower's perspective.

3. The standard escape is a refinance into conventional at 20% equity. List every assumption that plan depends on, and name which of them failed for borrowers between 2021 and 2024.

4. This case shows a program cost driven by the program's solvency rather than by the individual borrower's risk. Identify one other place in mortgage lending where a borrower pays for something determined by pool-level rather than loan-level considerations. (Chapter 28 is a hint.)

5. Write the two sentences you would add to your standard FHA-versus-conventional presentation as a result of reading this. Then write the one question you would ask the borrower.