Disinflation Exception? The Special Case Of Insurance Costs
US auto and home insurance costs have come under scrutiny lately for bucking the otherwise strong disinflationary trend of the past year. We discuss the reasons behind this divergence, methodological differences in how insurance costs are calculated across various inflation indicators, and our expectations for the future. We are not particularly worried about car insurance dynamics, but home insurance1 may continue to exert upward inflationary pressures, especially in light of climate-related risks.
Auto Insurance: A Lagged Response to Input Price Pressures
US consumer price inflation peaked at 9.1% year-over-year in June 2022, but receded to 3.0% a year later. It stood at 2.9% by July 2024—the latest data available. By contrast, motor vehicle insurance inflation did not peak until April 2024, and then at an eye-popping 22.6% (Figure 1). It was still 18.6% in July. This has fueled a reacceleration in so-called “supercore” inflation (core services excluding shelter) this year, which in turn has stayed the Fed’s hand vis-a-vis rate cuts so far. The divergence is readily explainable, however, and is most likely temporary. Admittedly, its unprecedented extend has given rise to speculation about potential opportunistic pricing in this space.
We believe this is far less about opportunistic pricing than about a delayed adjustment to rising underlying replacement costs. Auto insurance policies typically follow an annual or multi-year renewal cycle. As such, premiums adjust with a lag, as shown in Figure 2. In that chart, we see the cumulative change (relative to January 2020) in the price of various CPI components: vehicle prices, repair, and insurance. If one looked simply at where those metrics stand today, the temptation to conclude opportunistic behavior would be strong. After all, motor vehicle insurance costs rose 48.8% over that period, but new vehicle prices are up “just” 24.7% and used vehicle prices are up 35.2%. However, that overlooks the fact that for more than two years through mid-2023, insurance premiums were trailing underlying replacement costs. In other words, the current “overshoot” is compensating for the prior “undershoot.”
Auto Insurance Price Increases Expected to Moderate
This lagged response is why we expect the gap to narrow going forward. This view is supported by PCE (personal consumption expenditure) inflation data, which shows sharply lower inflation rates for vehicle insurance versus the CPI (Figure 3). In the PCE, which is the Fed’s preferred inflation metric, costs are treated very differently. PCE inflation measures insurance costs on a “net” basis; i.e., premiums minus the value of received benefits. For example, if premiums rise 10%, but the value of the underlying benefit increases 5%, the PCE insurance inflation rate would be 5%.2 (In many ways, this is a better measure of true underlying costs for consumers.) By contrast, the CPI measures insurance inflation on a gross basis, i.e., 10% in this example.
There is great value in understanding the behavior of the CPI and the PCE, but over longer periods of time, they do not differ greatly. This reversion makes intuitive sense; should premiums inflation sustainably outpace underlying replacement costs by a large margin (beyond “normal” profit margins), that would create arbitrage opportunities. New insurance providers would enter the market to take advantage of that and reduce the gap in the process.
One nuance (not quite a caveat) to the disinflationary argument above is that vehicle maintenance and repair costs have been, and continue to be, on the rise. To some extent, this reflects the natural uptrend in car prices in general (as cars have become more tech-intensive). But it also illustrates the incremental shift toward electric vehicles (EVs), for which repair costs generally exceed those of traditional combustion engine cars by a large margin. A rising share of EVs in the overall fleet would put upward pressure on average repair costs, all else equal. However, the impact would likely be incremental, contained, and ultimately, temporary.
A Unique Dynamic in the Household Insurance Space
In contrast to the auto insurance CPI-PCE differential detailed above, Figure 4 shows that household insurance inflation, as measured in the PCE, has steadily outpaced premium growth in the CPI measure of tenants and household insurance.
This primarily reflects definitional differences that go well beyond the “net of benefits” concept. Tenant and household insurance in the CPI is primarily a measure of renters’ insurance, with homeowners’ insurance only partly and indirectly incorporated as part of “owners’ equivalent rent.” In other words, the CPI measure of household insurance is not representative of actual homeowners’ insurance trends.3
By contrast, the PCE measure closely tracks the producer price measure of premiums for homeowners insurance and is far more representative of trends in the real economy. Even the PCE appears to understate industry data in respect to premium increases, but this could simply reflect the delayed “rollover” of existing policies, which reset over time.
Homeowners’ Insurance Likely to Continue Exerting Upward Pressure On Inflation
It is worth looking at long-term historical data on insurance costs relative to house prices as the underlying asset. Average existing home prices have soared, rising 206% since 2000.4 Meanwhile, new home prices have increased 152% over the same period. This means the PCE measure of household insurance (up 137%) has been very well aligned with new home prices over the long run, but has significantly trailed existing home prices (Figure 5).
To be fair, home prices and insurance costs do not move in tandem over the business cycle; home prices move around a lot more than insurance costs (also illustrated in Figure 5). But the magnitude of the gap that has opened since 2020 suggests homeowners insurance could remain a source of inflationary pressures for some time.
The path forward likely involves a combination of:
a) Higher insurance premiums
b) Restated terms/reduced benefits
c) Expanded role of public insurance provisions
Market forces seem to suggest higher premiums. This, however, runs into legal restrictions on how much insurance premiums can increase from year to year. The tension helps explain why a growing number of private insurance providers are shutting down operations, especially in certain geographies such as Florida where climate change and excessive litigation puts additional strain on the insurance business (see: Insurance, Climate-Related Risks and the Rising Costs of Living). Absent higher premiums, providers can muddle through by limiting premium increases while altering contract terms to reduce benefits. This can cushion profit margins, but it leaves consumers (and, indirectly, the mortgage providers) exposed to more risk.
Alternatively, the government can step in to fill in the gap left by private providers. However, without higher premiums, this would simply shift potential losses from private to public balance sheets, and is not a long-term sustainable solution. All in all, it seems reasonable to anticipate rising home insurance premiums going forward, especially given the rise in the severity of climate-related events. If and when higher premiums occur, their impact on CPI inflation may be blunted by artificial definitional boundaries (see footnote 3), but the premium hikes could be more visible in the PCE.
The Bottom Line
Insurance costs have been an exception to the broad US disinflation trend of the past eighteen months. That is starting to change, especially in the auto insurance space, where peak inflation is now behind us. However, upward pressure in homeowners’ insurance costs is likely to persist in the near term. Longer term, climate change also poses risks of sustained inflationary pressures in this space.