Social Security COLA Calculations May Get Changed to CPI-E as Part of the Reform Bill. What Does it Mean for Retirees?


by Wolf Richter, Wolf Street:

Watch out for the costs of housing, medical care, and gasoline.

Social Security benefits are adjusted for inflation – the Cost of Living Adjustments or COLAs – based on the “Consumer Price Index for Urban Wage Earners and Clerical Workers” (CPI-W), released by the Bureau of Labor Statistics. By this measure, inflation was 6.9% in October.

Alas, the COLA for benefits in the year 2022 was based on the third-quarter average CPI-W, when inflation was still lower. And so the COLA for 2022 will only be 5.9%, nevertheless the highest since 1982.


As part of the efforts of reforming Social Security, there are now proposals in Congress – including a Bill by Rep. Al Lawson (D-FL), that include provisions to raise revenues – mostly focused on raising the Social Security contribution cap – and provisions to “improve” benefits, including by switching the COLA calculation from CPI-W to CPI Elderly, or CPI-E.

CPI-E is designed to reflect the purchasing habits of people 62 years and older. The weights of the items in the basket are adjusted to reflect the typical purchasing habits of the elderly.

The biggest factor in the difference is housing costs (“shelter”). It accounts for 36.8% of the weight in CPI-E but only for 32.5% in CPI-W. Housing costs have been soaring in reality, but the CPI has been slow in picking them up. But that has now started, and CPI for housing costs have started to rise and will continue to rise in 2022, and this will accelerate CPI-E more than CPI-W in 2022.

The second largest factor in the difference is medical care, where the elderly spend a lot more. And there are other major differences where the elderly spend relatively more.

In the other direction, where the elderly spend less, and where weights in the CPI-E are lower than in CPI-W, are gasoline (no more daily commutes, thank god), vehicle purchases, education, and the like.

The table below shows the major categories, accounting for about 73% of total CPI-W and 75% of CPI-E:

Relative weights
CPI-E CPI-W Points difference
Shelter 36.8% 32.5% 4.3
Medical care 12.2% 8.5% 3.7
Household furnishings & operations 5.1% 4.7% 0.4
Food at home 7.4% 7.7% -0.3
Apparel 1.9% 2.7% -0.8
Motor fuel 2.2% 3.9% -1.7
New & used vehicles 5.2% 7.1% -1.9
Education & communication services 4.2% 6.1% -1.9

The differing weights produce a different inflation reading, and proponents of CPI-E say that it produces a higher inflation reading, which would produce higher COLAs.

But this year, CPI-E is going massively in the wrong direction and the COLA for 2022 would get crushed. The CPI-E for October was 5.7% (red), while the CPI-W was 6.9% (green):

And the COLA under CPI-E for 2022 would be 4.8%, based on the average of CPI-E in Q3, compared to the actual COLA of 5.9%.

COLA calculations would be based on the average CPI-E readings in Q3 of every year. So I calculated COLAs based on CPI-E going back 22 years to 2001 (2000 Q3 CPI-E readings). Over the entire time, COLAs based on CPI-E would have averaged 2.4%, while actual COLAs averaged 2.3%.

So on average, CPI-E would be a slight improvement, and every little thing helps retirees.

In most years, the differences are not huge and go in both directions. But there were sharp divergences: in 2022 and 2009; and in the other direction in 2015 and 2016.

In 2022 (based on Q3 2021 CPIs) and in 2009 (based on Q3 2008 CPIs), gasoline prices spiked out the wazoo, and given that motor fuels weigh less in CPI-E, that index didn’t move as much as CPI-W. and the COLAs based on CPI-E would have been substantially smaller.

Also in 2022, soaring housing costs were not picked up by the CPIs in Q3 (though that has now started), and this pushed down the CPIs in Q3. Given that shelter weighs so much more in CPI-E than in CPI-W, it pushed down the CPI-E further than CPI-W. Hence the massive difference of 1.1 percentage points in the COLAs for 2022:

  • 2009 actual COLA: 5.8%; CPI-E COLA: 5.1%
  • 2022 actual COLA: 5.9%; CPI-E COLA: 4.8%

The opposite happened in 2016, based on Q3 2015 CPIs, and in 2017, based on Q3 2016 CPIs, when gasoline prices plunged following the collapse in the price of crude oil as part of the Great American Oil Bust. Note that COLAs do not go negative, but bottom out at 0%:

  • 2016 actual COLA: 0%; CPI-E COLA: 0.6%
  • 2017 actual COLA: 0.3%; CPI-E COLA: 1.5%

Switching to CPI-E this year would have meant an even colder shower for the purchasing power of Social Security benefits.

But if gasoline prices ever plunge again, then CPI-E COLAs would look better. And when the housing cost surges get picked up by the CPIs in Q3 2022, then the CPI-E COLA for 2023 is going to look better.

Overall, over the next two decades, a shift to CPI-E will likely produce slightly higher COLAs on average, and every little bit helps for folks living on a fixed income and struggling with surging inflation. But there will be years with nasty surprises for CPI-E COLAs, and years where things turn out better for CPI-E COLAs.

Overall, the shift to CPI-E would not change much, and my advice would still hold: If you don’t have a big nest egg, work for as long as possible after receiving benefits – even a part-time gig helps – because the actual increases in your costs of living will outrun the COLAs as you get older.

Read More @