Case Study 1 — The 2013 Duration Change: When an Insurance Fund Ran Out of Room

Tier 1 / Tier 2 mix. The statutes, agencies, and policy instruments named here are real and public. Specific dollar amounts and capital ratios are reported figures that you should verify in HUD's Annual Report to Congress on the Financial Status of the FHA Mutual Mortgage Insurance Fund and in the mortgagee letters themselves. No figure in this case study should be quoted to a borrower without verification.

Why this case

Section 16.5 taught a rule that decides the Linden Street comparison, costs a 3.5%-down borrower tens of thousands of dollars over a loan's life, and is invisible on every disclosure the borrower signs: annual MIP duration is set by the loan-to-value at origination and never revisited.

That rule is not ancient. It is not a law of nature. It was adopted in 2013, in response to a specific and documented problem with a specific balance sheet, and the rule that preceded it looked much more like conventional mortgage insurance. Understanding how the rule came to exist is the single best inoculation against the habit this book keeps warning about: treating a current guideline value as a permanent fact.

Background: what an insurance fund actually is

FHA does not lend. It insures. Every FHA borrower pays an upfront premium and an annual premium into the Mutual Mortgage Insurance Fund, and every time an FHA-insured loan goes to foreclosure and the lender files a claim, the fund pays. The fund is, in the most literal sense, a pool of borrower premiums standing behind lender losses.

Congress requires that fund to hold capital. The National Housing Act establishes a statutory minimum capital ratio of 2 percent — the fund's economic value expressed as a percentage of its insurance-in-force. HUD reports on the fund's condition annually, in the Annual Report to Congress on the Financial Status of the FHA Mutual Mortgage Insurance Fund, supported by an independent actuarial review. Both documents are public and both are worth an hour of your time.

Two things follow from that structure, and they explain almost everything about FHA policy volatility:

1. FHA cannot price loan-by-loan for risk the way a private insurer does. Its premiums are set administratively and apply broadly. It cannot charge one borrower more because their score is lower.
2. When the fund's capital position deteriorates, FHA's available levers are blunt: raise premiums, extend how long premiums are collected, or tighten eligibility. All three fall on future borrowers, because the terms of existing loans cannot be changed.

Hold that second point. It is the whole case.

The issue: a countercyclical insurer in a crisis

FHA's market share is countercyclical by design. When private capital is abundant and lending standards are loose, borrowers who can get conventional financing take it, and FHA's share shrinks. When private capital withdraws, FHA is what is left.

In the years after 2007, private capital withdrew at a speed and scale without modern precedent. The subprime market ceased to exist. Private mortgage insurers tightened drastically or stopped writing at high loan-to-value entirely. Conventional guidelines contracted. And FHA — which had been a small share of the purchase market during the boom — expanded to a multiple of its pre-crisis volume, essentially overnight, in the worst house-price environment since the 1930s.

Two things happened at once:

The books FHA insured in 2007 and 2008 performed badly. They were underwritten before the crisis, on values set at the top of the market, to borrowers whose employment situations then deteriorated. Claims on those vintages ran well above expectation.

The books FHA insured in 2009 onward performed well — arguably among the best-performing vintages in the agency's history, because they were underwritten at post-crash values, under tightened standards, to borrowers who then experienced a long expansion.

But the good vintages take years to earn premium, and the bad vintages file claims immediately. The MMI Fund's capital ratio fell below the statutory 2 percent minimum and stayed there for several years. In the FY2012 actuarial review the fund's economic value was reported as negative.

In September 2013, FHA drew on Treasury for the first time in its history — a mandatory appropriation widely reported at approximately \$1.7 billion — to cover projected losses in the MMI Fund. (Verify the exact figure and the mechanism in HUD's FY2013 Annual Report to Congress.)

Note carefully what that draw was and was not. It was not a bailout of borrowers, and it was not a default. It was a statutorily automatic transfer that occurs when the fund's reserves are projected to be insufficient. But the political and institutional reaction to it was intense, and the pressure on FHA to restore the fund was correspondingly intense.

What FHA did

Between 2010 and 2013, FHA raised the annual mortgage insurance premium repeatedly. Each increase arrived as a mortgagee letter, effective for case numbers assigned on or after a stated date — which is precisely why §16.2 insists that you check the mortgagee letters and check the case-number date.

Then, in 2013, FHA changed something structurally different from a premium increase. It changed how long the premium is collected.

Mortgagee Letter 2013-04, applicable to case numbers assigned on or after June 3, 2013, revised the annual MIP cancellation policy. (Verify the letter number and effective date at HUD.)

Read the "before" column again, because it is what most experienced originators still half-remember. FHA's old cancellation rule looked very much like the Homeowners Protection Act rule that governs conventional MI. A 3.5%-down FHA borrower on a 30-year loan reached 78% of original value somewhere in the second decade, and their MIP stopped. Under the post-2013 rule, that same borrower pays for 360 months.

The 2013 change is therefore not a tweak. It is a categorical difference in the product.

What it shows

1. A "guideline" and a "permanent feature" are not the same thing, and the difference is not visible from inside a single transaction. A loan officer working in 2012 could have told a borrower, accurately, that FHA mortgage insurance comes off at 78%. A loan officer telling a borrower the same thing in 2014 would have been badly wrong. Nothing about the borrower, the property, or the program's name changed. A mortgagee letter changed.

2. The lever fell on new borrowers, because it had to. Existing FHA borrowers kept their cancellation terms — the contract had already been made. All of the burden of restoring the fund fell on people who had not yet applied. That is not a criticism of FHA; it is a structural feature of any insurer that cannot reprice its existing book. But it means that the cost of a fund's bad years is paid, mechanically, by the next cohort of buyers, who are disproportionately first-time buyers with small down payments.

3. Premiums are reversible. Duration was not. This is the most useful observation in the case, and it is the one most originators have never noticed. FHA reduced the annual MIP factor by 50 basis points in January 2015 (Mortgagee Letter 2015-01), announced a further reduction in January 2017 that was suspended before it took effect, and reduced the factor again by 30 basis points in 2023. The fund's capital ratio recovered above the statutory 2 percent minimum in FY2015 and has risen substantially since. (Verify all of these at HUD.)

Through every one of those reductions, the duration policy stayed. The premium came back down. The end date did not come back.

4. Structural rules outlive the conditions that produced them. The 2013 duration policy was adopted when the MMI Fund was in the worst condition of its history. The fund is no longer in that condition. The policy remains. This is not unusual and it is not sinister — an insurer that has just been through a near-death experience does not lightly restore the feature that made the near-death experience worse — but it means that a rule's original justification is a poor guide to whether the rule is currently in force. Check the rule, not the reasoning.

Outcome

FHA's insurance fund recovered. Premiums fell. The duration policy in §16.5 is the surviving artifact of the crisis, and it is now the single most consequential structural difference between FHA and conventional financing for a low-down-payment borrower.

For your practice, the outcome is a habit rather than a fact:

• Never quote a premium, a factor, a threshold, or a duration from memory or from a training deck. Check the handbook, then check the mortgagee letters issued since the handbook's revision date.
• Know the case-number date on every FHA file you have in process, because it is the date that determines which version of the rules applies when a policy changes mid-pipeline.
• Tell borrowers the duration rule out loud, before they choose the program. They will not learn it from the Loan Estimate.

Discussion questions

1. FHA's premiums are set administratively and apply broadly rather than being priced borrower by borrower. §16.10 showed that this is exactly why FHA is cheaper than conventional for a 641-score borrower. Is the same feature also what forced FHA to use blunt instruments in 2013? Argue both sides.

2. The 2013 change fell entirely on future borrowers, because existing loan terms could not be altered. Identify one other place in this book where a rule change could only be applied prospectively, and describe who bore the cost there.

3. A borrower asks you in 2013, two weeks before the effective date, whether they should rush to get a case number assigned before June 3. What are the professional and compliance considerations in how you answer? What would you actually say?

4. The premium reductions of 2015 and 2023 restored the price but not the end date. Explain, in dollars, why a borrower who cares only about their monthly payment would perceive those reductions as a full restoration, and why they would be wrong.

5. Suppose you had joined the industry in 2011 and learned FHA under the pre-2013 cancellation rule. What specific habit would have caught the 2013 change before it embarrassed you in front of a borrower? Write the habit as a step in a checklist.

6. This case study is written from public documents — mortgagee letters, an annual report to Congress, and an independent actuarial review — all of which are free. What does it say about the profession that most working originators have never opened any of them?