Changes To Inflation Formula Will Clip Social Security Benefits, Study Says

Last Updated Sep 7, 2011 1:54 PM EDT

Two of the high-level commissions asked to develop budget-balancing strategies proposed what seemed to be a technical change in calculating Social Security benefits -- swapping out the general CPI-U, which we know as the headline measure of inflation, in favor of C-CPI-U (the extra C is for "chained"). They also recommended it for indexing tax brackets, among other things. The net effect is to slow down the rate of increase of Social Security benefits for a rising cost of living.

It would save a huge projected amount -- eliminating one-quarter of forecasted deficits. The change has some economic merit, says a new study from the respected Center for Retirement Research at Boston College, but needs to be studied fully, because the new inflation measure may not deliver to seniors the income protection they need in retirement, especially on the costs of health care.

There's no argument, at least from me, that the United States needs to cut expenditures, toward preserving our long-term economic viability. And social programs are one way to get there. Social Security and Medicare offer lots of possibilities; their long-term nature allows small changes in direction today that can build up into big numbers over time.

The move to change the statistic on which cost of living increases are based is one of those. And politically, it's very efficient. Because this proposed change likely to slip through without much comment or understanding of its nature by the people who are, in effect, losing part of their Social Security benefits, it's the sort of backdoor adjustment we will be seeing a lot of in the next few years.

From the CRR's report:

The chained CPI is projected to rise about 0.3 percentage points per year more slowly than the current index. Thus, the change would result in lower cost-of-living adjustments (COLAs) for Social Security beneficiaries and for federal civilian and military retirees, and would also lead to an increase in federal taxes.
(The CRR's paper explains how the Consumer Price Index is assembled -- gathering price data from around the country on 80,000 different items. Interesting stuff for you statistics fans.)

Before 1972, Social Security benefits were not automatically indexed to increases in the cost of living, but after inflation started to kick up in 1966, Congress decided to put in systematic raises. (It's a good thing they did, because between 1966 and 1981 -- the years of the highest recent inflation -- the cost of living rose by about three times. That was a terrible punishment on people on fixed incomes.)

The bottom line is that using the alternative chained CPI will allow slower growth in Social Security benefits. Both the Bowles-Simpson panel (called the National Commission on Fiscal Responsibility and Reform) which President Obama convened to come up with bipartisan ideas on cutting the government's budget deficit, and The Bipartisan Policy Center's Debt Reduction Task Force were in favor of slipping in the chained CPI. Said the BPC: "This is a technical change that will be applied in all government programs that use COLAs, including the indexation of tax brackets." Bowles-Simpson characterized the chained index as "a more accurate measure of inflation."

What is this about? I'm oversimplifying, because I am no statistician, but the conventional CPI measures the cost of living by comparing the prices on the same quantities of goods from period to period -- it assumes we are buying the same basket of food, clothes, computers, etc.

The chained method allows for changes in the basket. Say the price of imported beer goes from $9 a carton to $12. You switch to domestic, for $6. The old CPI would count the cost of this month's beer as $12, while the chained method thinks it's just $6.

Here is the CRR's interpretation of the long-term effects (emphasis mine):

...[S]witching to the [chained] CPI-U was estimated to reduce Social Security's 75-year actuarial deficit by 0.5 percent of taxable payrolls. Since the deficit is about 2 percent of taxable payrolls, this change alone would eliminate about one-quarter of the long-range deficit.
It may seem surprising that such a small change could eliminate such a large portion of the deficit. Af­ter all, 0.3 percent less in a COLA applied to the aver­age monthly benefit of $1,200 amounts to only about $4 per month. But while shifting to a chained CPI involves a relatively small change for young retirees, it results in a substantial benefit cut as retirees age. A COLA that is 0.3 percentage points lower per year would produce a monthly benefit that is about 6.5 per­cent lower by the time a retiree reaches 85.
While it may be realistic to model inflation with the "domestic for imported sorts of substitutions -- really, if the price of something gets to a point where you can't afford it, you shouldn't buy it -- the CRR says it doesn't take into account the details of the cost of living of seniors. Maybe they have more time to shop for bargains, but they spend a lot more of their incomes on health care -- 13 percent for those over 65, versus five percent for younger consumers. In fact, the BLS's special index for the elderly has risen about 0.3 percent faster than the general CPI over the last 30 years.

The CRR also makes a qualitative assessment of the cost of living of the people this affects most:

Low-income elderly are not decid­ing [between] whether to buy a watch or a bracelet. They spend most of their income on essential amounts of neces­sities, like housing, food, health care, and transporta­tion. If the price of gasoline doubles, they cannot mitigate the impact on their total costs simply by driving less. They are already consuming close to the minimum. Most likely, they will drive a little less and then cut spending on housing, food, and health care. With little ability to respond to price changes, the poor have no mechanism to offset the full brunt of a price increase.
There is a rich body of thought -- some of it valid economic research, and some conspiracy theory -- on the measurement of prices. I have no problem with adjusting the way inflation is passed through in social program benefits and tax rates -- we all have to share the burden -- as long as it's fair. By that I mean it doesn't harm the people who can't defend themselves in a political forum, such as the elderly poor.