Posted December 23, 2012
Adopting a technical fix for measuring inflation could save upwards of $200 billion over a decade
By Loren Adler and Shai Akabas
The U.S. government currently loses tens of billions of dollars each year by inaccurately measuring inflation. There is a relatively simple fix, yet its inclusion in the fiscal cliff negotiations is generating significant ire from some in the blogosphere, including Ezra Klein.
First, a brief crash course to explain what the “chained-CPI” is, why the Bureau of Labor Statistics (BLS) believes that it more accurately captures the rate of inflation, and why this matters for the federal budget. (For a more-detailed explanation, see this good primer from the Moment of Truth.)
Cost of Living Adjustments (COLAs) for many federal benefit programs, tax brackets, and other thresholds are indexed to versions of the Consumer Price Index (CPI), which measures price changes in a market basket of goods. Maintaining purchasing power by accounting for inflation is essential for all of these government functions, but the current CPI metric does so inaccurately. BLS has found that current calculations of the CPI overstate inflation (e.g., when the prices a consumer faces rise by 2 percent, the CPI measures roughly a 2.3 percent increase). When this happens year after year, the error is compounded.
This miscalculation is the product of a simple omission: The formula used to calculate the CPI does not account for changes in consumer spending habits because the measurement assumes that consumers will purchase the same categories of goods and services in the same quantities regardless of the prices. In reality, consumers will lessen the effect of a significant price increase in one good by consuming more of a comparable but cheaper good in its place.
In order to correctly measure the cost-of-living increases that the average consumer faces, an inflation calculation must account for actual consumer buying patterns among goods and services. A measurement called the “chained-CPI” does just that, and therefore, is a more accurate measurement of inflation. Switching to that improved metric would modestly reduce the growth of government benefit payments and increase tax revenues (as more income would remain in higher tax brackets). Over the course of a decade, this change could save upwards of $200 billion.
Despite general acknowledgement that the chained-CPI is more accurate, some journalists and advocacy groups charge that the adjustment would be a stealth Social Security cut and tax increase. Even more perplexingly, some argue that if the intent is to correct a technical problem, the federal government should make the correction and then compensate the losers. This makes little sense.
When policymakers close a loophole in the tax code that a certain group was unfairly taking advantage of, the federal government doesn’t compensate them for not having the loophole anymore.
Fixing how we measure inflation would repair a flaw in federal law that unintentionally leads to larger growth in benefits and lower taxes. If policymakers want to increase certain entitlement benefits or cut taxes, they should do so with explicit legislation, not through maintaining loopholes.
A legitimate concern about the adjustment, however, is that certain groups – particularly those with very low incomes and the elderly – may be unduly affected by the chained-CPI since their consumption patterns are not as flexible as the average American’s. While there is no perfect way to craft exceptions, policymakers should attempt to do so where possible.
For example, given that beneficiaries of Supplemental Security Income (SSI) comprise the nation’s poorest elderly and disabled, and that specific complexities cause the chained-CPI switch to disproportionately affect SSI benefits, that program should receive special consideration.
Additionally, the oldest seniors spend a substantially greater than average proportion of their income on medical expenses – which rise in cost faster than most other goods – a situation that is not accounted for by the chained-CPI. This difference, though, is focused in the latest years of life (one-fourth of Medicare spending is on people in their last year of life).
In part to handle this issue, both Domenici-Rivlin and Simpson-Bowles recommended a modest phased-in benefit increase for seniors over 80 (equal to 5 percent of the average Social Security benefit) in order to progressively help out those who most need the additional assistance. In fact, if combined with the CPI change, most older, low-income beneficiaries would be better off than they are today.
Read the Center on Budget and Policy Priorities’ analysis for an explanation of why this method of boosting benefits for the oldest seniors is a better method of dealing with this problem.
If policymakers can’t find a way to adopt this technical fix for measuring inflation on a bipartisan basis, hopes are slim for agreement on broader and more substantial policy changes to curb our long-run debt problem. We should welcome debates over whether certain entitlement benefits should be increased or tax rates cut en route to addressing our fiscal issues; in fact, Domenici-Rivlin proposed some of both, cutting marginal tax rates across the board and increasing Social Security benefits for those least well-off. But those questions should be answered separate and apart from making this overdue adjustment to accurately account for inflation in our 21st century economy.
Economic Policy Project