Implications of a 'Chained' CPI
Publication Date: September 2011
Publisher(s): Center for Retirement Research at Boston College
Keywords: savings and consumption
Coverage: United States
Two prominent commissions recently proposed introducing a “chained” consumer price index (CPI) to adjust Social Security benefits, other government benefits, and the brackets in the federal income tax each year. The argument is that a chained CPI would be more accurate since it reflects the extent to which people substitute one item for another in the face of a price increase. 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. Although this provision was not included in the initial package of cuts to raise the debt limit, it will almost certainly be considered by the Congressional Joint Select Committee on Deficit Reduction, which has been assigned the task of identifying an additional $1.5 trillion in cuts over 10 years by the end of November. Therefore, it is important to understand how a chained CPI would work and how it would affect Social Security beneficiaries.
This brief proceeds as follows. The first section describes what the CPI is intended to measure and how it is constructed. The second section summarizes progress to date in accounting for substitution and the goal and mechanics of creating a chained CPI. The third section discusses one of the assumptions underlying the improved accuracy of a chained CPI – namely, that the current index fully reflects the increase in prices faced by beneficiaries. This assumption may not hold, since the CPI-E, which re-weights components to reflect the market basket of the elderly, has risen more rapidly than the current index. The fourth section explores another underlying assumption of the chained CPI – namely, that everybody has an equal ability to substitute when relative prices change. To the extent that the low-income elderly lack this flexibility, adopting a chained version of the CPI would understate the impact of inflation on their welfare.
The final section concludes that, under current circumstances, moving to a chained index should be viewed as a cut in benefits. The cut could impact the poor, who are less likely to be able to shift their spending patterns in response to price changes, and the oldest old as they see the effects compounding over time. The adverse impacts can be mitigated by one-time adjustments around age 85, as suggested by both commissions. But, in the current context, moving to a chained CPI is much more than a technical correction.