SSA Announces New COLA

Yesterday, the Social Security Administration announced that beneficiaries will be receiving a 1.5 percent Cost-of-Living Adjustment (COLA) this year. Although this update is below the 1.7 percent increase provided this year and the 3.6 percent increase provided for 2012, it is reflective of the relatively low inflation experienced over the past year.

Some groups and commentators have suggested that this COLA is too small relative to growing living costs, a claim that does not account for the fact that while some prices have gone up faster than 1.5 percent (for example meat prices rose by about 2 percent), others have grown more slowly or fallen (for example, gasoline prices have fallen by about 2 percent).

Other groups have used the COLA announcement as an opportunity to blast efforts to adopt a more accurate measure of inflation, or to call for indexing Social Security with an experimental senior-specific index known is the CPI-E. Because seniors spend more on health care and housing, they argued, their COLA should be measured with a senior specific index.

As we have pointed out before, however, the case for the CPI-E is quite weak. Among our concerns with adopting the CPI-E are:

• The CPI-E is highly experimental. The Bureau of Labor Statistics, who publishes the CPI-E, has explained it is not ready for prime time.
• The CPI-E suffers from substitution bias, like the traditional CPI, and also suffers from a much more significant small-sample bias.
• The CPI-E does not appropriately account for things like mail-order shopping or for senior discounts.
• The CPI-E likely overstates the impact of growing health prices by conflating increases in value with increases in price.
• The CPI-E misstates the impact of housing on cost-of-living by imputing rental value for homeowners. 80 percent of seniors are homeowners and 70 percent of them have paid of their mortgages (compared to 60 percent and 17 percent of non-seniors). When the value of their homes go up, the CPI-E measure this as if they face higher rental prices.

For these reasons and more, the CBO has questioned whether the elderly actually face higher inflation than the general population at all. And if they do, almost all evidence suggest it is by less than the CPI-E would suggest.

Adopting the CPI-E would also raise questions of fairness. What should happen to the non-elderly portion of the Social Security population? Should they get a smaller COLA? Should each inflation-indexed program and provision in the budget and tax code get its own index? Why should we adjust for a 0.2 percent claimed difference in inflation between seniors and non-seniors but ignore the 0.9 percent difference between New York and Detroit?

Ultimately, well-intentioned efforts to make government benefits more accurate and fair through the CPI-E would do the opposite, making them less accurate and less fair. Instead, policymakers should use the most accurate measure of economy-wide inflation available -- the chained CPI -- and use that measure to meet the stated goal of inflation indexing. It is our more accurate because, as our paper "Measuring Up" explained:

Moving to the chained CPI would address [upper-level substitution] bias by using a superlative index that updates expenditure weights and formulas in order to address consumer response to substitution between categories. As the Congressional Budget Office (CBO) explains, the chained CPI "attempts to fully account for the effects of economic substitution on changes in the cost of living... [It] provides an unbiased estimate of changes in the cost of living from one month to the next by using market baskets from both months, thus 'chaining' the two months together."

Other targeted concerns should be met with targeted reforms, not across-the-board mismeasurements. For example, the risk of outliving ones savings in very old age could be addressed with a benefit bump-up for the very old. And the risk of catastrophically high health care costs among a small group of seniors – which appears to be responsible for much of the difference in health spending between seniors and the population at large -- could be addressed through cost-sharing reform that limits out of pocket costs.

Those focused on the COLA as a vehicle for making policy often forget what it is there for: to reflect cost-of-living increases over the previous year. The chained CPI best lives up to that goal.