S&P Global Ratings has updated its home price overvaluation assessment and the related Federal Housing Finance Agency Home Price Index (FHFA HPI) inputs, based on fourth-quarter 2023 data. Our assessment fell slightly to 14.3% as per capita income growth outpaced home price appreciation (HPA). However, it is still comparable to our last reading, which was an overvaluation of 15.6% at the national level based on third-quarter 2023 data (see "U.S. Home Price Overvaluation Relatively Steady, But Bumps Up A Bit," published Feb. 5, 2024). The non-seasonally adjusted All-Transactions FHFA HPI was close to flat (up 0.04%) between third- and fourth-quarter 2023, while the Purchase-Only Index decreased 0.29% on an unadjusted basis. Regional variation persists, however, and our assessment shows that about 89% of metropolitan statistical areas or divisions (which we refer to collectively as MSAs) are overvalued, consistent with our prior assessment.

We believe that the credit impact home price overvaluation could have on U.S. residential mortgage-backed securities (RMBS) will depend on the geographic distribution of the mortgage pools and the valuation dates of the underlying properties backing the loans in those pools.

Measuring Over/Undervaluation

We view home prices as overvalued or undervalued based on how much a specified region's (e.g., an MSA or a state) price-to-income (PTI) ratio is above or below its long-term average. Our regional inputs are the FHFA HPI and income per capita data (from the Bureau of Economic Analysis and the U.S. Census Bureau), which we use to compare the current PTI ratio to the 20-year average and assess over/undervaluation.

Overvaluation depends on a transaction's pool diversification and the location of the underlying mortgaged properties. Our loss severity assumptions will tend to be higher when properties are overvalued because a greater correction in home prices could occur under adverse scenarios. On April 12, 2024, we updated our over/undervaluation measures and the related FHFA HPI inputs using fourth-quarter 2023 data. We use these values because they relate to certain U.S. RMBS in our loan evaluation and estimate of loss system (LEVELS) model, which provides loan- and pool-level calculations of default likelihood (foreclosure frequency), loss given default (loss severity), and loss coverage (see "LEVELS Model For U.S. Residential Mortgage Loans," published Aug. 5, 2019). Depending on when a property valuation was performed, our indexed valuation will be slightly higher at the national level, given the FHFA HPI change between third-quarter 2023 and fourth-quarter 2023.

Overvaluation Softens

Our current nationwide overvaluation assessment fell approximately one percentage point to 14.3% due to per-capita income growth and negligible HPA gains. The fourth-quarter FHFA HPI was roughly unchanged relative to the prior quarter, with 25 states experiencing home price depreciation, compared with only two states in the previous quarter (see chart 1). The FHFA HPI has historically increased 0.86% on average in the fourth quarter.

Chart 1

Housing Is Still Overvalued

As with our February 2024 review, overvaluation remains high nationwide, with 89% of MSAs still overvalued. There is still substantial regional variation in both the number of overvalued MSAs and the extent of the overvaluation. For instance, numerous Florida MSAs (such as those including Tampa and Sarasota) are overvalued by more than 40%, while certain MSAs (including several in Northern California) remain undervalued by as much as 20% (see chart 2).

We believe U.S. home prices will continue to depend on a combination of factors, including the trajectory and stability of the 30-year fixed-rate mortgage, local housing market dynamics, and economic fundamentals. The Purchase-Only Index fell 0.1% month over month in January 2024 after rising 0.1% in December 2023 on a seasonally adjusted basis. Chart 3 shows the regional differences in home price changes.

Chart 2

Chart 3

The Most Overvalued And Undervalued MSAs

Many MSAs have overvaluations that are still much higher than the 14.3% national average. The top 10 MSAs have representation from four states: Florida (six), Texas (two), Georgia (one), and Tennessee (one). These regions have experienced relatively high population growth, which has contributed to the increase in home prices as demand has outpaced supply. For more on population dynamics, see "2024 U.S. Residential Mortgage And Housing Outlook: A Rate And Supply Story," published Jan. 17, 2024. Chart 4 shows the 10 most overvalued and undervalued MSAs, and the over/undervaluation distribution for all 399 U.S. MSAs

Chart 4

The Impact On RMBS

We believe the credit impact of FHFA HPI changes and overvaluation levels on U.S. RMBS will depend on the geographic distribution of the mortgage pools and the valuation dates of the properties backing the loans in those pools.

Our over/undervaluation measure provides information about affordability in terms of the deviation from the long-term average, which could influence how much property prices decline under some economic scenarios. To account for this when rating certain U.S. RMBS, we calculate the loss severity on a loan by applying our over/undervaluation assessment to our market value decline (MVD) assumptions (see our "Methodology And Assumptions For Rating U.S. RMBS Issued 2009 And Later" criteria published Feb. 22, 2018). Under a 'AAA' rating stress, we assume that (for a given region) 50% of the overvaluation amount of a mortgaged property will factor into the MVD, with a corresponding value of 20% at a 'B' rating level. At the national level (assuming 14.3% overvaluation), our 'AAA' MVD assumption is approximately 52%. This assumes that, under a 'AAA' rating stress, the additional decline in a property's value would reduce the liquidation proceeds by approximately 7% (compared to a market at equilibrium) and, correspondingly, increase the loss severity assumed for a given loan.

When indexing property values, we apply 50% of the cumulative upward movements and 100% of the downward movements, based on our criteria. The slightly higher property values that result from this indexation could decrease the probability of defaults and have varying effects on loss severities, depending on loan age and regional over/undervaluation.