Will the real inflation rate please stand up?
I ask because we’re fast approaching the date when the Social Security Administration (SSA) will set the cost-of-living-adjustment for 2023’s Social Security checks. So it’s to be expected that, as is the case along about now in every year, retirees are wondering whether the SSA is using the correct COLA factor.
We already know that, since inflation is running at a 40-year high, next year’s COLA will be the highest in several decades. But whether it will be high enough to compensate retirees for the inflation they’re actually incurring is the subject of intense debate.
Not to bury my lead: I wouldn’t worry. The major contenders for COLA adjustment factor are each telling almost precisely the same story.
It’s helpful to begin this discussion with a review of the three primary inflation measures that are calculated each month by the federal government’s Bureau of Labor Statistics:
CPI-U. This is the headline inflation number that gets reported each month by the financial press. It stands for “Consumer Price Index for All Urban Consumers.”
CPI-W. This is the version of the CPI that the SSA by law is required to use for its COLA calculations. It is known formally as the “Consumer Price Index for All Urban Wage Earners and Clerical Workers” (CPI-W), and compared to the CPI-U it “places a slightly higher weight on food, apparel, transportation, and other goods and services,” according to the Bureau of Labor Statistics, and “a slightly lower weight on housing, medical care, and recreation.”
CPI-E. This stands for “Consumer Price Index for The Elderly,” and the SSA calculates it by taking into account the different spending habits of the typical elderly person. Many of the Social security reform proposals in Congress include a requirement that, going forward, the SSA use this version of the CPI to adjust for inflation.
The Covid-19 pandemic has introduced an additional factor in these discussions, since spending habits have changed significantly since the economy went into a lockdown in early 2020. Because the CPI’s calculation is based on pre-Covid assumptions about the relative proportions of household budgets that are spent on food, transportation, housing, and so forth, it’s possible that pandemic-induced changes in spending habits have made the CPI a poor measure of retirees’ actual inflation.
Though the SSA doesn’t calculate a separate CPI that reflects Covid-19 induced changes in our spending habits, one who has is Alberto Cavallo, a professor at the Harvard Business School. Over the 12 months through May 2020, for example, his “Covid-CPI” rose by 0.8 of a percentage point more than the CPI-U.
Little actual difference
Though there is much theoretical debate about the relative merits of these various measures, they have made little difference in recent years. Consider first the three CPI measures from the Bureau of Labor Statistics, which over the last 15 years are reporting almost identical annualized inflation rates: The CPI-U’s is 2.38%, the CPI-E’s is 2.39%, and the CPI-W’s is 2.44%.
Cavallo’s Covid-CPI does not extend back 15 years, so I have no comparison data for it over this longer period. But following the months in 2020 when the CPI was under-stating inflation, per Cavallo’s more recent measurements, the CPI in 2021 and 2022 has overstated it. The net result is that his Covid-CPI’s four-year annualized return is very similar to those of the other CPI measures.
The accompanying chart plots, for each month since early 2019, the four inflation measures’ trailing 12-month rates of change. Notice that, over the 12 months through June, the latest month for which Cavallo has calculated his Covid-CPI, it rose 8.51%, the CPI-E 8.46%, the CPI-U 9.00%, and the CPI-W 9.81%.
This means that the COLA adjustment factor that the SSA is going to use to set next year’s checks—the CPI-W—shows the highest rate of inflation of these four.
We don’t yet know the exact inflation adjustment factor that will be used for 2023, since it won’t be set until the September inflation numbers are published in early October. But, for now, each of the various inflation indices is telling a largely similar story. And, to the extent there is any difference in those indices, the one that SSA uses will result in the biggest COLA.
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at firstname.lastname@example.org.