The Social Security benefit cut the supercommittee is most likely to make is a reduced COLA, the inflation adjustment that protects the purchasing power of recipients’ benefit checks from rising prices. Alice Rivlin, Alan Simpson, Erskine Bowles, and most pundits argue that the COLA overstates price growth facing the elderly and hence provides them a mounting windfall over time , fattening the retirement checks of the elderly at the expense of everyone who’s still working. They are wrong.

In fact, as economists have pointed out, it is likelier that the current Social Security COLA is insufficient to fully protect Social Security beneficiaries from the effects of inflation, because it doesn’t take into account the large amounts of money the elderly pay out-of-pocket for health care. It’s probably true that the chained CPI-U better accounts for price changes affecting the general population, but it simply measures the wrong market basket for the elderly. The chained CPI-U not only understates the effect of health care cost inflation on the elderly, but it may also overstate their ability to change their buying habits in response to price increases since a greater share of their incomes is spent on necessities.

New evidence that the elderly have very different consumption patterns than the general population comes from the federal Interagency Technical Working Group that just reported on a new Supplemental Poverty Measure (SPM), a new methodology for determining the extent of poverty in the United States. The study group confirmed that medical out-of-pocket expenses for the elderly are a disproportionate part of their consumption. When the study group measured poverty based on the actual spending of Americans over the age of 65, they discovered that the poverty rate jumped from 9 percent to 15.9 percent. More than 6 million Americans aged 65 or over are living in poverty, according to the SPM. The view that the elderly are doing better than everyone else and living well at the expense of the working population is contradicted by this new evidence.

No supercommittee member should be allowed to pretend that using the chained CPI to determine Social Security COLAs is a “technical” change to increase their accuracy. It is a benefit cut, pure and simple, and it will do the greatest harm to the oldest of the elderly. Under the proposed COLA, an average-wage worker retiring this year would, in 2031, receive $1,754 less in annual benefits.