CRS Says Moving to Chained CPI Could Increase Tax Revenues

The U.S. Bureau of Labor Statistics (BLS) publishes two important measures of inflation: the Consumer Price Index for all Urban Consumers (CPI-U) and the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W). The CPI might seem like just another economic indicator, but it could be a powerful policy lever. Because the CPI-W is used to calculate annual cost-of-living adjustments (COLAs) to Social Security retirement benefits and the CPI-U is used to calculate annual inflation adjustments to personal income tax brackets, for example, changing the basis of the automatic adjustments could substantially affect outlays and revenues, and thereby, the budget deficit.

As part of its continuing effort to improve measures of change in the cost of living, BLS has since August 2002 been publishing a supplemental measure known as the Chained Consumer Price Index for all Urban Consumers (C-CPI-U). The aim of the C-CPI-U is to produce a measure of change in consumer prices that is free of substitution bias.

One of the difficulties in estimating changes in the cost of living is that consumers often change the goods and services they purchase in response to changing relative prices. In other words, consumers will tend to buy more of those goods and services whose prices are rising slower than average and fewer of those goods and services whose prices are rising faster than average. Substitution on the part of consumers is believed to insulate them from the full effect of rising prices on maintaining their standard of living. Because the CPI-W and CPI-U do not entirely account for substitution, they overstate the impact of inflation on consumer well-being.

As a result of better reflecting consumer substitution, the C-CPI-U has typically increased to a lesser extent than either the CPI-U or CPI-W. This relationship has prompted calls for switching to the C-CPI-U when calculating automatic annual adjustments to inflation-indexed federal programs and individual tax provisions to slow growth in the budget deficit. The "Simpson-Bowles" (National Commission on Fiscal Responsibility and Reform) report included government-wide replacement of the CPI-W and CPI-U with the chained CPI, for example. The committee created by the Budget Control Act of 2011 (P.L. 112-25) may consider the approach during its upcoming deliberations.

That the CPI-W and CPI-U are not revised makes them attractive for use in calculating cost-of-living adjustments. Unlike them, the C-CPI-U is subject to two revisions after its first release. If the two indexes were replaced by the C-CPI-U, cost-of-living adjustments would either have to wait until the final number was available or rely on preliminary estimates that could change up to two years after the fact. Whether the preliminary C-CPI-U estimates might be attractive alternatives to using the final estimate depends on whether they are biased relative to the final number or if the revisions tend to be significant. If the final index tends to rise more than the preliminary estimates, it might make their use unpopular with those who would be affected. More generally, because increases in the C-CPI-U have tended to be smaller than in either the CPI-W or CPI-U, some Social Security beneficiaries, taxpayers, and individuals whose eligibility for federal programs is based on the poverty threshold (e.g., food stamps) might oppose switching to the C-CPI-U.

                            Contents

The Consumer Price Index (CPI) is probably the most important measure of inflation developed by the federal government because it is used to make automatic adjustments that affect both outlays and revenues. The Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) is the basis for adjusting Social Security retirement benefits¹ and the Consumer Price Index for All Urban Consumers (CPI-U) is the basis for adjusting personal income tax brackets² to keep up with inflation, for example. Changing the government's basis for indexing these among other federal programs and tax provisions from the CPI-W and CPI-U could have substantial effects on the budget deficit.³ (Hereafter in this report, the CPI-W and CPI-U will be referred to collectively as the standard CPI.)

Then-chairman of the Federal Reserve Board Alan Greenspan suggested in testimony before the House Budget Committee in 2004 that Congress consider replacing the standard CPI with the Chained CPI for all Urban Consumers (C-CPI-U) to make automatic cost-of-living adjustments (COLAs) to federal programs.⁴ He pointed out that, at that time, if the C-CPI-U had been used instead of the CPI-U and CPI-W over the previous 10 years, the federal debt would have been about $200 billion less. More recently, the "Simpson-Bowles" (National Commission on Fiscal Responsibility and Reform) report and the "Gang of Six" plan among others included replacing the standard CPI with the C-CPI-U for all government provisions subject to indexation.⁵ This proposal may again be considered during the deliberations of the committee created by the Budget Control Act of 2011 (P.L. 112-25).

One reason besides slowing the growth in the budget deficit for changing to the C-CPI-U is that the standard CPI has not been without criticism as a measure of change in the cost of living. A true cost-of-living index would measure the change in income required for consumers to maintain a constant level of "utility" (satisfaction). But there are a number of practical complications that make constructing such an index difficult.

With a given level of income that constrains their choices, consumers decide how to spend their money based on the satisfaction the various available goods and services yield. Consumers are assumed to spend their income in such a way as to get the most satisfaction possible within the limitations of their budget. Because utility cannot be directly measured, any indicator of the cost of living must be based on what consumers actually spend. Any numerical measure that attempts to approximate changes in the cost of a given standard of living depends on a number of assumptions and has numerous practical limitations.

One of the difficulties in estimating changes in the cost of living is that consumer spending patterns change continuously. Spending patterns change because of changing tastes and preferences (e.g., for meals in restaurants rather than meals prepared at home). Spending patterns also change due to changes in relative prices. As prices change over time, consumers will tend to buy more of those goods and services whose prices are rising slower than average and fewer of those goods and services whose prices are rising faster than average. So-called substitution bias causes the standard CPI to overstate the effect of inflation on consumer well-being.⁶

The U.S. Bureau of Labor Statistics (BLS), which produces the standard CPI, has strived to construct a better measure of changes in the cost of living. Toward that end, it asked the Committee on National Statistics of the National Research Council (NRC) to convene a panel of experts to look into the conceptual, measurement, and other statistical issues that arise when constructing cost-of-living indexes.⁷ In its 2002 report, the panel endorsed continuing to use the CPI for tax indexation, which began in 1985 as required by the Economic Recovery Tax Act of 1981. It suggested that poverty thresholds (which determine eligibility for government transfer programs such as Medicaid and food stamps) might be adjusted by a fixed percentage of median income or consumption of workers, and mentioned a 1995 NRC report on measuring poverty that had suggested thresholds be adjusted by a fixed ratio of the median consumption of necessities (e.g., food and shelter). The panel expressed support for BLS's then in-development Chained Consumer Price Index for all Urban Consumers as a means of more accurately adjusting Social Security retirement benefits among other federal payments (e.g., military and civil service pensions; veterans' benefits) for real changes in the cost of living.⁸ But, there are difficulties with switching to the C-CPI-U despite it more fully accounting for substitution bias and thereby more accurately reflecting changes in the cost of living.

This report explains methodological and statistical differences between the standard CPI and the C-CPI-U. It then addresses a key impediment to moving to the C-CPI-U. The report closes with a discussion of the potential impact of such a switch on the federal budget deficit.

Methodological Differences Between the Standard CPI and the C-CPI-U

Because the standard CPI is a fixed-weight index, it does not entirely reflect ongoing changes in buying habits. As the overall level of prices changes, relative prices change as well. Consumers can to some degree change their spending patterns in response to these prices changes by buying relatively more of those goods whose prices are rising more slowly.

If these changes in consumer spending patterns have no effect on overall consumer satisfaction, then a price index based on a fixed market basket of goods and services will overstate the increase in the cost of a given standard of living. Because the standard CPI does not fully account for consumers' ability to insulate themselves from inflation by changing their spending patterns, it overestimates how much they would need to raise total spending to maintain a constant standard of living. The C-CPI-U, in contrast, is constructed in such a way as to better account for substitution bias.

The CPI-U and CPI-W

The standard CPI is a fixed-weight (Laspeyres) price index which measures the change in retail prices of an unchanging mix of goods and services purchased by consumers. To see how a fixed-weight index is calculated, consider the simple case of two time periods and two goods. In the first period, the value of the index is one. The index value in the second period is a function of the quantities in the first period and the prices in the two periods. It is a weighted sum. The first step is to calculate, for each good, the ratio of the price in the second period to the price in the first period. The ratios are then summed using expenditure shares in only the first period as weights. To see how a fixed-weight price index is calculated, see Box 1.

The standard CPI is, strictly speaking, a modified fixed-weight price index. That is, the market basket of goods and services is periodically changed to keep it up to date. Until recently, however, those updates occurred only about once every 10 years. With the release of CPI data for January 2002, the market basket was updated to reflect spending patterns reported in the Consumer Expenditure (CE) Survey for the 1999-2000 period. Since then, BLS has updated the expenditure weights every two years.⁹ With the release of the January 2010 CPI, the weights were updated to reflect spending patterns in the 2007-2008 period. Despite this more frequent updating of the market basket, the standard CPI continues to be subject to the substitution bias that is inherent in a fixed-weight index.

 

The C-CPI-U

In an effort to get an estimate of the effect of consumer substitution on the standard CPI, BLS introduced a supplemental measure known as the Chained Consumer Price Index for all Urban Consumers (C-CPI-U). It does not replace either the CPI-W or CPI-U, and has not to date affected any current indexing provisions of federal programs.

The aim of the C-CPI-U is to produce a measure of change in consumer prices that is free of upper-level substitution bias. Upper-level substitution refers to consumers changing their spending between broad categories in the market basket (e.g., buying more chicken and less fish due an increase in the price of fish compared with chicken from one month to the next).¹⁰

The final release of the C-CPI-U is calculated using a Törnqvist index formula that relies on consumer expenditure data for the current and prior months as a means of accounting for any substitution across categories made by consumers in response to changes in relative prices. In other words, "the final version of the C-CPI-U is based on actual consumer behavior, rather than assumptions about consumer substitution."¹¹ BLS estimated that the decrease in cost-of-living growth due to accounting for upper-level substitution may be 0.3 percentage points. To see how a Törnqvist price index is calculated, see Box 2.

 

The initial and interim releases of the C-CPI-U are based on the same expenditure weights used for the CPI-U but a geometric mean formula is used in aggregating the basic indexes to create the upper-level indexes in the initial and interim releases.¹² In contrast with the Laspeyres index in which the quantities are held constant in both periods, the geometric mean index formula holds expenditure shares (price times quantity) constant. It assumes a particular consumer response to the change in relative prices. That means that if the price of a good rises, the quantity consumed implicitly falls. Some research has suggested that the geometric mean based price index may have a negative substitution bias if consumers are assumed to respond to changes in relative prices more than they actually do.¹³ To see how a geometric mean index is calculated, see Box 3.

The initial and interim releases of the C-CPI-U are further adjusted based on historical differences between geometric mean and Törnqvist indexes. This is done so that the initial and interim release are closer to the final index number. BLS is continuing to study methods to further narrow the gap between preliminary and final index numbers.¹⁴

 

     Table 1.Number of Months After Reference Month That Data Are Released

Data for the C-CPI-U are now available beginning with December 1999. Final data are available through the end of 2009, and interim data are available through the end of 2010. As shown in Table 2, most of the time the increase in the final C-CPI-U has been smaller than for the increase in the standard CPI.

                 Table 2.The C-CPI-U, the CPI-U, and the CPI-W

 

Figure 1.The CPI-U and the C-CPI-U

 

 

 

Policy Considerations

The CPI is important, not only as an economic indicator, but also because it has significant implications for the budget through the indexing of some tax provisions and federal programs. If the CPI overstates the effect of inflation on consumers, then Social Security benefits are rising more rapidly than necessary to preserve the living standards of beneficiaries and income tax brackets are rising more than necessary to avoid "bracket creep."¹⁵

If the C-CPI-U is a better measure of changes in the true cost of living, and the goal of indexing is strictly to reflect changes in the cost of living, then the C-CPI-U might be considered as a measure on which to base those adjustments. As previously discussed, however, a major complication of switching to the C-CPI-U is that final data are not available for up to two years after the reference period. The Social Security cost-of-living adjustment (COLA) payable in January 2011 was based on the average percentage increase between third quarter 2010 and 2008 CPI data (because 2008 was the last year in which a COLA was effective).¹⁶ Final C-CPI-U data for the third quarter of 2010 will not become available until February 2012, however. Such a long time lag might make the final C-CPI-U number a poor candidate as an index for automatic adjustments.

Preliminary C-CPI-U estimates might be an attractive alternative to using the final C-CPI-U. If there is a tendency for the final index to rise more than the initial or interim indexes, it might make the preliminary indexes unpopular with those who would be affected. Speaking more generally, basing future COLAs on any version of the C-CPI-U could generate opposition from some Social Security beneficiaries, taxpayers, and those whose eligibility for federal programs is based on poverty thresholds because the index has tended to rise less than either the CPI-U or the CPI-W as they are now calculated.

The elderly and their advocates are likely to be among those who oppose to changing the current basis for indexation. If the prices Social Security beneficiaries and federal and military pensioners face increase at an above-average rate, switching to the C-CPI-U might not enable this population to maintain its standard of living. The elderly spend more than the population at large on health care, and prices for these services generally have increased at an above-average rate.¹⁷ An experimental fixed-weight index for those aged 62 and older (CPI-E) computed by BLS has increased 0.27 percentage points faster than the CPI-W from the CPI-E's inception in December 1982 to December 2010. However, it is difficult to gauge whether the cost of living among the elderly actually increases more quickly than among younger persons due to rising health care prices because BLS may underestimate the rate of improvement in the quality of these services. If this is the case then all versions of the CPI -- but especially the CPI-E -- overstate increases in the cost of living.¹⁸

In its regularly issued report on spending and revenue options to reduce the deficit, the U.S. Congressional Budget Office (CBO) includes estimates of the effects on the budget if policymakers substitute the C-CPI-U for the standard CPI.¹⁹ The estimates assume that, in the future, the C-CPI-U will increase 0.25% more slowly each year between 2012 and 2021 than the CPI that is now used for indexation. Using preliminary estimates of the C-CPI-U as described in a technical appendix to Using a Different Measure of Inflation for Indexing Federal Programs and the Tax Code,²⁰ CBO projected that a switch from the CPI-U to the C-CPI-U could yield a cumulative increase in revenues between 2010 and 2021 of $71.8 billion in the case of indexation of various provisions of the tax code.²¹ In the case of Social Security COLAs, CBO projected that a switch from the CPI-W to the C-CPI-U would result in a cumulative decline in outlays of $112.0 billion between 2012 and 2021. In the case of COLAs for federal and military pensions, switching the index could result in a cumulative decline in outlays of $24.0 billion between 2012 and 2021.

Author Contact Information

This report was originally authored by Brian Cashell (retired).

² The CPI-U is derived from the average expenditures of households beyond those in the CPI-W population. It includes salaried workers (e.g., professionals and managers), part-time and part-year workers, the self-employed, the unemployed, and households with no one in the labor force (e.g., retirees). The CPI-U covers about 87% of the U.S. population. Publication of the CPI-U began in 1978.

³ Other than Social Security retirement benefits, federal programs adjusted by the CPI-W (U.S. city average, all items) include federal civilian, military, and railroad pensions; veterans' benefits; and Supplemental Security Income. In addition to adjusting income tax brackets, other provisions in the individual tax code that are adjusted by the CPI-U (U.S. city average, all items) include the standard deduction, personal exemption, and Earned Income Tax Credit among other credits.

⁴ Testimony of Alan Greenspan before the Committee on the Budget, U.S. House of Representatives, February 25, 2004. Available on the Federal Reserve Board website at http://www.federalreserve.gov/boarddocs/testimony/2004/20040225/default.htm.

⁵ Adam Rosenberg and Marc Goldwein, Measuring Up: the Case for the Chained CPI, Moment of Truth Project, Washington, DC, May 11, 2011.

⁶ See, for example, Toward a More Accurate Measure of the Cost of Living, Final Report to the Senate Finance Committee from the Advisory Commission to Study the Consumer Price Index, December 4, 1996. The publication is commonly known as the Boskin Report after its chairman.

⁷ National Research Council, At What Price? Conceptualizing and Measuring Cost-of-Living and Price Indexes, ed. Charles Schultze and Christopher Mackie (Washington DC: National Academy Press, 2002).

⁸ Annual cost-of-living adjustments were linked to changes in the CPI in the 1972 amendments to the Social Security Act. The only CPI available at that time was the CPI-W. The National Research Council did not recommend switching from the CPI-W to the CPI-U for indexing Social Security payments because the two indexes have changed at about the same rate over time. It also concluded that there was no rationale for switching from the CPI-W to the special experimental index for the elderly (CPI-E) that Congress requested the BLS to construct. The CPI-E is discussed in more detail at the end of this report.

⁹ The U.S. Bureau of the Census conducts the CE Survey for BLS. "Collection and processing the annual data consumes the greater part of a year, meaning that expenditure data introduced into the CPI at the beginning of year t can pertain to year t-2 at the latest. . . . Recognized international best practice calls only for revising expenditure weights at least every five years, and more frequently if there is high inflation or evidence of rapid changes in consumption patterns." John S. Greenless and Elliot Williams, Reconsideration of Weighting and Updating Procedures in the US CPI, U.S. Bureau of Labor Statistics, Working Paper 431, Washington, DC, October 2009, p. 5.

¹⁰ In 1999, BLS began applying a geometric mean formula when creating basic indexes within which goods are relatively close substitutes to account for lower-level substitution in the standard CPI. Lower-level substitution refers to consumers changing their spending within narrow categories in the market basket (e.g., buying more Muenster than Swiss cheese due to a relative increase in the price of Swiss). BLS estimated that the actual decrease in CPI growth due to use of the geometric mean to calculate most lower level indexes may be 0.28 percentage points per year. (See John S. Greenlees and Robert B. McClelland, "Addressing Misconceptions About the Consumer Price Index," Monthly Labor Review, August 2008.) BLS uses the geometric mean to create most basic indexes, but there are some exceptions. The lower level category of hospital services is one. BLS judged that consumers are unlikely to substitute between the services performed at hospitals in response to relative price changes.

¹¹ David S. Johnson, Stephen B. Reed, and Kenneth J. Stewart, "Price Measurement in the United States: A Decade After the Boskin Report," Monthly Labor Review, May 2006, p. 12.

¹² As previously noted, BLS has since 1999 used the geometric mean in developing the basic indexes of the standard CPI. It is not used it in their aggregation.

¹³ See Matthew D. Shapiro and David W. Wilcox, "Alternative Strategies for Aggregating Prices in the CPI," Federal Reserve Bank of St. Louis Review, May/June 1997.

¹⁴ For additional methodological information on the standard CPI and C-CPI-U see the chapter on the CPI in the BLS Handbook of Methods, updated June 2007 (http://www.bls.gov/opub/hom/pdf/homch17.pdf); and for the C-CPI-U in particular see Robert Cage, John Greenlees, and Patrick Jackman, Introducing the Chained Consumer Price Index, paper presented at the Seventh Meeting of the International Working Group on Price Indexes, Paris, FR, May 2003, and Frequently Asked Questions on the Chained Consumer Price Index for All Urban Consumers, http://www.bls.gov/cpi/cpisupqa.htm.

¹⁵ With progressive tax rates, as incomes rise with inflation, more income is subject to higher tax rates. If the CPI-U overstates inflation, then the tax brackets are rising more than they need to in order to avoid increasing taxes on incomes that are simply keeping up with rising prices.

¹⁶ Because there was no increase in the CPI-W, Social recipients did not receive a COLA in 2011.

¹⁷ For more information see CRS Report RS20060, A Separate Consumer Price Index for the Elderly?

¹⁸ U.S. Congressional Budget Office, Using a Different Measure of Inflation for Indexing Federal Programs and the Tax Code, Economic and Budget Issue Brief, Washington, DC, February 24, 2010.

¹⁹ U.S. Congressional Budget Office, Reducing the Deficit: Spending and Revenue Options, Washington, DC, March 2011.

²⁰http://www.cbo.gov/ftpdocs/112xx/doc11256/WebAppendix.pdf

²¹ They include the amounts of the personal and dependent exemptions as well as the standard deductions; the levels of incomes that separate individual income tax brackets from one another; the amount of the annual gift tax exemption; and the income thresholds and phaseouts for the earned income tax credit and the child tax credit among other credits.