Why increasing conforming loan limits makes housing goals more challenging

On Aug. 25 the Federal Housing Finance Agency published a proposed rule establishing affordable housing goals for Fannie Mae and Freddie Mac for 2022-24. These goals were originally established by statute in 1992, with revisions in 2008.

They call for the FHFA director to set annual goals for the minimum percentages of the government-sponsored entities’ single-family mortgage acquisitions for home purchase mortgages for (1) low-income families, (2) very low-income families, (3) families in low-income areas, and (4) refinance mortgages for low-income families. Here “low-income” refers to families with incomes no greater than 80 percent of Area Median Income and “very-low income” refers to families with incomes no greater than 50 percent of AMI.

There are also goals for the minimum numbers of units in multifamily properties for low-income families, very low-income families, and low-income families in small, 5 to 50 unit properties.

On Nov. 30, the FHFA announced the 2022 conforming loan limits for mortgages on single-family mortgages. For one-unit properties in most areas, this limit will be increased from the current level of $548,250 to $647,200 for 2022, an increase of 18%. This will be the largest increase since 1981. Higher limits apply in high-cost areas and for two- to four-unit properties. If we applied this 18.0 percent increase to the conforming loan limits for 2020, the GSEs would have been able to purchase loans with an unpaid principal balance greater than $510,240 up to a maximum of $602,500, which we refer to as “2020 jumbo mortgages.”

This increase in the conforming loan limits would have the greatest impact on the two property types for which the proposed goals would increase the most —low-income home purchase goals, from the current 24 percent to 28 percent for 2022-24, and low-income refinance mortgages, from the current 21 percent to 26 percent for 2022-24.

It is highly doubtful that any of the 2020 jumbo mortgages would have qualified for either of the low-income goals. But they would have been added to the denominator in calculating goal performance, thereby reducing such performance on both of these goals. The question then is how many such mortgages were originated in 2020. The GSE data submitted to FHFA does not include such mortgages, since they were not eligible for GSE purchase. The only source for such information is the Home Mortgage Disclosure Act data, which is released annually by the Consumer Financial Protection Bureau.

To take a simplified example, suppose that 2,000 conforming conventional mortgages were originated in 2020, and 400, or 20%, were for low-income families. And suppose that there were 200 additional 2020 jumbo mortgages, none of which were for low-income families. Then 400 of the 2,200 mortgages were for low-income families, for market performance of 18.2%. Thus the increase in the conforming loan limit would have reduced market performance by 1.8 percentage points.

To continue the example, suppose that a GSE buys 50% of the 2,000 originations, including 50% of the low-income mortgages, for performance excluding the conforming jumbo mortgages of 200/1,000, also 20%, the same as for the market as a whole. Then assume that the GSE buys 60% of the 200 additional conforming jumbo mortgages, which would yield performance of 200/1,120 mortgages, or 17.9%, a drop of 2.1 percentage points. Thus in this example the impact of the conforming jumbo mortgages would be greater for GSE performance than for market performance, causing GSE performance to fall short of market performance.

This is a simplified example, but it lays out the framework that FHFA could follow based on HMDA data on mortgage originations and GSE purchases in 2020.

Both GSEs discussed this effect in their comments on the proposed housing goals. Fannie Mae stated that “this dynamic would exert additional downward pressure on Fannie Mae’s low-income home purchase goal borrower share.” And Freddie Mac said the "anticipated expansion of conforming loan limits could introduce a dilutive effect on the housing goals results.”

FHFA did not have an analysis of this issue in the proposed housing goals rule for 2022-24. A definitive analysis, based on 2020 HMDA data, would yield answers to this question of the effect of making these higher-UPB mortgages eligible for GSE purchase would have on market performance on these goals, and better enable formulation of the final goals for 2022-24.