USA: FHA home loans to non-permanent residents fall to near zero

In this article

Written by

Kooky

Builder of Shaka, the payment router that pays every agent their commission on closing date.

About Kooky and Shaka →

Buyers who live in the United States without permanent resident status have all but vanished from FHA home-purchase lending. Origination data from Intercontinental Exchange (ICE), shared with the trade publication HousingWire and published by it on Thursday 8 October 2026, show that non-permanent residents accounted for 5.8 per cent of FHA purchase loans before a change of rule in 2025, and for about 0.1 per cent after it.

The rule came from the Department of Housing and Urban Development (HUD), which removed the "non-permanent resident" category from its single-family FHA programmes with effect from 25 May 2025. The figures now published measure what that decision did to lending on the ground, and they show that the effect was far from even across the country. In two Utah markets, one FHA purchase loan in five or more had been going to this group of borrowers.

Intercontinental Exchange origination data as reported by HousingWire, 8 October 2026. Before: second quarter of 2024 to first quarter of 2025. After: fourth quarter of 2025 onwards.

What the ICE figures measure

The numbers are shares, not counts. ICE looked at purchase mortgages, the loans used to buy a home, and worked out what proportion of them went to borrowers recorded as non-permanent residents. HousingWire's report, written by its mortgage reporter Sarah Wolak, does not give the number of loans behind each percentage, so the data say how much of the FHA purchase business this group represented and not how many households that was.

Two periods are compared. The "before" period runs from the second quarter of 2024 to the first quarter of 2025, a full year that ends just before the rule took effect. The "after" period starts in the fourth quarter of 2025 and runs to the present. The two quarters in between, the second and third of 2025, sit in neither window. The rule took effect inside the first of them, so the comparison as published sets a full year before the change against the months well after it.

The sample is wide: the 100 largest housing markets in the United States, plus 33 border markets, 133 in all. Nationally, the share of FHA purchase originations going to non-permanent residents fell by 5.7 percentage points, from 5.8 per cent to about 0.1 per cent.

The rule behind the fall

The change was made in March 2025, when HUD issued two documents: Mortgagee Letter 2025-09 and Title I Letter 490. Together they removed the non-permanent resident category from the FHA's Single Family Title I and Title II programmes. From 25 May 2025, a borrower in that category could no longer obtain a loan under either of them.

The data therefore describe a rule doing what it was written to do, and doing it almost completely. A category that stood at 5.8 per cent of FHA purchase lending in the year before the change is measured at about a tenth of one per cent in the period after it.

One point of detail separates the official account from the data. HousingWire reports that the HUD Secretary, Scott Turner, described the share as zero. ICE's measure is about 0.1 per cent nationally, and several of the individual markets in its sample still show a fraction of a per cent. The report does not explain the remainder, and the two figures are best read side by side: zero in the Secretary's statement, close to zero in the lender data.

Where the change is largest

The national figure hides how concentrated this lending was. In Salt Lake City, non-permanent residents made up 26 per cent of FHA purchase loans in the year before the rule, a little over one in four. In Provo, also in Utah, the share was 20 per cent. Both are now below half of one per cent.

Florida supplies four of the markets HousingWire singles out. Orlando stood at 16.1 per cent, about one FHA purchase loan in six, and Lakeland at 13.6 per cent. Cape Coral and Miami were each at 10 per cent. Raleigh in North Carolina was at 11.1 per cent, and San Jose in California at about 10 per cent, the only one of the eight where the share after the rule is given as 0 per cent.

For the people who work in these markets, the size of the starting share is what matters. Where a category supplies one in four or one in five loans of a given type, its removal changes the mix of buyers a loan officer or a listing agent meets. Where it supplied roughly one in seventeen, as it did nationally, the same rule is felt much less.

Conventional loans did not fill the gap

A borrower who can no longer use an FHA loan is not shut out of the mortgage market altogether. Conventional loans remain open to lawful non-permanent residents, because Fannie Mae and Freddie Mac permit them, as HousingWire's report notes. The obvious question is whether the buyers who lost FHA access simply moved across.

The ICE data suggest that, as a group, they did not. Across all loan types, non-permanent residents' share of purchase mortgages fell from 5.6 per cent to 3.4 per cent, a drop of 2.2 percentage points. Conventional prime conforming lending to the group stayed relatively stable, according to the report. It did not rise to make up for the FHA loans that were no longer being written.

Put together, the two findings describe a group that still buys with conventional financing at much the same rate as before, and has lost most of the purchases it used to finance through the FHA. The data do not show what became of the households concerned: whether they are renting for longer, buying with other kinds of finance that the report does not break out, or not buying.

List prices in the same markets

HousingWire set the lending figures beside its own list-price data for the same places. For seven of the eight markets, all but Miami, it gives the range in which the median list price of single-family homes sat during the "before" period and during the "after" period.

The picture is mixed. In Provo the top of the range came down from US$877,400 to US$699,950, and in Raleigh and Cape Coral both ends of the range are lower than they were. In Salt Lake City, the market with the largest starting share of this lending, the bottom of the range barely moved while the top fell from US$795,000 to US$742,450. In Lakeland the range edged up at both ends, and in San Jose the top stayed at US$1.70 million.

These are list prices, what sellers ask, and the two periods differ in many ways besides one lending rule. The figures show what happened in these markets over the same months. They do not show that the rule caused it, and the HousingWire figures as reported do not isolate its effect on prices.

What HUD and lenders say

The views reported by HousingWire fall into two groups, and they describe the same result from different sides.

For HUD, the outcome is the purpose. Secretary Scott Turner, as reported by HousingWire, said the department is preventing people without legal status and non-residents from using taxpayer-backed programmes at the expense of American homebuyers. That is his characterisation of the policy and of the borrowers it concerns.

From the lending side, HousingWire spoke to two home loan specialists at Churchill Mortgage. Michael Brown, who works in Tennessee, told the publication that the policy has clearly affected families who would have qualified before May 2025. Many of them, he said, now use conventional loans. Isaac Schultz, his colleague in Michigan, said that the fall in this lending was, for better or for worse, the main result the policy change was meant to produce.

None of the three disputes the direction of the data. They differ on how to describe it: as protection of a public programme for American buyers in the Secretary's account, as lost eligibility for particular families in Mr Brown's, and as the intended effect of the rule in Mr Schultz's.

Reading the figures with care

Several limits apply. The first is the source: the data are ICE's, supplied to one trade publication, and the full dataset behind the percentages is not set out in the report. The second is the sample, which covers the largest markets and a set of border markets, not every county in the country. The third is that shares can fall for more than one reason. A share is a fraction, and it moves when the total changes as well as when the group itself does.

It also helps to be clear about who is counted. The figures concern borrowers, people taking out a mortgage to buy a home in the United States, sorted by residency status. They say nothing about buyers who pay without a mortgage, and nothing about overseas owners selling American property, who meet a separate federal regime. Under the withholding rules published by the IRS, a buyer must generally withhold 15 per cent of the amount realised when a foreign person disposes of a US real property interest, with no withholding where that amount is US$300,000 or less and the buyer will use the property as a residence. That is a tax rule on sales. The HUD change is a rule on who may borrow, and the two operate independently.

For agents and loan officers in the eight markets named, the practical reading is narrow and concrete. A buyer who is a lawful non-permanent resident is no longer an FHA buyer, and has not been since 25 May 2025. Conventional financing remains open to that buyer under Fannie Mae and Freddie Mac's rules, on conventional terms, and whether a given household qualifies depends on its own file. The research behind this article found no further change to the rule and no scheduled follow-up release of the data.

Kooky, from Shaka

Kooky edits Agents Estate and builds Shaka, the payment router he made for real estate professionals. One payment comes in, and every agent, agency and party in the deal receives their signed share on closing date.

About Kooky and Shaka →