Annual Cost-of-Living Adjustments (COLAs) are intended to protect the benefits of Social Security beneficiaries, military retirees, and other recipients of government programs from inflation.  In recent years, however, an increasing number of policymakers have said that COLAs may be growing faster than inflation.(1)  The Consumer Price Index (CPI), the measure of inflation that determines COLAs, is blamed, with some economists saying that it exaggerates the rate of inflation.(2)  

On the other hand, health care and prescription drugs, which can account for 21% to 51% of out-of-pocket senior spending (3), are growing at double-digit rates.  For example, in 2002 seniors received a COLA of only 2.6% while supplemental health insurance premiums and prescription drug spending rose by about 18%.(4)   This suggests that the CPI understates the true annual cost-of-living increase experienced by seniors.  Rapid growth in health care costs has been the case over the past five years, while COLAs are growing far more slowly.(5)  Understandably many seniors are growing increasingly concerned that they will have to give up essential items.  Some simply are choosing not to take their prescription drugs, or to go without food.

Government data indicates that the CPI used to calculate COLAs does in fact understate senior costs.  Currently, COLAs are determined by the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W).  The CPI-W surveys the change in costs of younger urban workers who are more likely to have employer-provided insurance, and to use less health care services and prescription drugs than seniors.  The CPI-W specifically excludes from its surveys the market basket of persons who receive pension income like Social Security. 

Since 1983, however, the government has maintained an unpublished, experimental seniors-only Consumer Price Index, the Consumer Price Index for Elderly Consumers (CPI-E).  Because the CPI-E gives greater weight or significance to health care costs it tends to more accurately reflect seniors' costs.  Although not used to calculate senior COLAs, over the 18 years for which the government has data, the CPI-E rose about 15% more than the CPI-W.  If the 2002 COLA were calculated using the CPI-E, seniors would have received a COLA of 3% instead of 2.6%.(6)

TREA Senior Citizens League has studied the effect of calculating COLAs using the CPI-E rather than the CPI-W.  That study found, for example, that a person who retired in 1984 with an average monthly benefit of $460 would have received about $6,321 more in benefits over the 18-year period had the CPI-E been used to calculate the COLA instead of the CPI-W.(7)

The debate over the accuracy of the CPI and COLAs achieved prominence in the mid-1990s under increasing pressure to balance the budget and the role of the CPI in determining COLAs as well as tax indexation.(8)  The Bureau of Labor Statistics (BLS) reported that the CPI "is used to adjust payments to Social Security recipients, to Federal and military retirees, and for a number of entitlement programs such as food stamps and school lunches.  An increase in the CPI increases Federal statutory obligations for these payments and programs.  In addition, individual income tax brackets and personal exemptions are adjusted for inflation using the CPI.  In this case, an increase in the CPI results in lower tax revenues.  It is estimated, for example, in fiscal year 1996, each 1% increase in the index produced a $5.7 billion increase in outlays and a $2.5 billion decline in revenues."(9) 
 
The Congressionally appointed Boskin Commission in 1996, estimated that the CPI overstated the annual cost of living by 1.1% a year and recommended changes in the way the CPI is designed and calculated. The Boskin Commission contended that the CPI does not fully take into account improved quality on many products or how consumers substitute a less expensive product when another has gone up in price.(10)

Partly in response to those recommendations, the Bureau of Labor Statistics instituted a number of changes to the way it computes the CPI.(11)  By July of 1998, the Congressional Budget Office (CBO) revised its estimates for future inflation downward by 0.7 percentage point per year based on those changes.  In addition, they revised their estimates of a budget surplus upward based on lower outlays and higher revenues.(12) 

In January 2002, the Bureau of Labor Statistics released new data based on revising expenditure weights (or the relative importance of an item) every two years rather than every 10 years as in previous years.(13)  These changes are likely to continue to reduce the rate of growth in the CPI.  A 1998 study performed by the BLS, showed that the new expenditure weight revision schedule would have lowered the measured rate of CPI growth rate by 0.17 percentage point per year.(14)

In addition, starting next year, the BLS plans to release an alternative CPI called a "superlative index" that will attempt to measure when consumers substitute less expensive items when prices go up.(15)  Tracking senior costs under these changes could be even more problematic.  For example, a senior requiring heart surgery may not be able to substitute less costly surgery or other treatment.  In like manner, there may not be a less costly or generic alternative to a brand name prescription drug. 

Yet despite the inherent problems with substitution behavior assumptions, a December 2001 report from the National Academy of Sciences makes the following recommendation, "The BLS should proceed as planned to begin publishing a superlative index with a two year lag."  The report further states, "It would be feasible and appropriate to calculate cost-of-living allowances provided for Social Security and other programs from an advance estimate of the BLS superlative index.  Any divergence between that estimate and the superlative that appears two years later could be incorporated as a correction to the Cost-of-Living Allowance provided for that year."(16)

 The changes, as well as slowing the rate of growth in the CPI, have the effective result of cutting COLAs for seniors.  A December 23, 2001, article in The New York Times states that "according to the Social Security Administration, a worker with an average wage rate who retired in 2002 at 65, then collected Social Security checks until she died at 82, would receive $318,000.  If the index were cut by 1.1 percentage points as the Boskin Commission recommended, that retiree would get $277,000 or $41,000 less over time.  For the government, of course, such a reduction in the index would mean saving billions of dollars."(17)

With the CBO's revision of inflation estimates of 0.7 percentage point in 1998, and subsequent CPI changes since that time that could add another 0.17 percentage point, the Boskin Commission's recommended 1.1 percentage point reduction appears well underway.  For several years, TREA Senior Citizens League (TSCL) has studied the effect of these changes on the CPI and senior benefits.  Similar to the Social Security Administration's example, TSCL's studies found that a person who receives an average monthly benefit of $874 in 2002 would effectively have their benefit cut by $5,356 over a ten-year period assuming a 0.8 percentage point reduction to the CPI.  By 2012, that person's benefits would be cut annually by about $1,068 or $89 per month.(18)

A study of the effect of revisions to the CPI was performed by Richard W. Johnson for the Urban Institute.  That study found that "reduced COLAs would not affect all elderly equally," warning that if the COLA formula were revised, the erosion in Social Security benefits would lead to an increase in persons falling below the poverty line, especially as the effects compound over time.  "The financial pain of reductions in Social Security COLAs," said the report, "would fall disproportionately on the elderly near the bottom of the income distribution."(19)

The claim that the CPI is inaccurate can indeed be made.  But whether the CPI exaggerates or understates inflation depends on whose market basket is under scrutiny.  The recent return of Federal budget deficits will again produce pressure to reduce expenditures and increase revenues, pointing to continued changes to the CPI.  This is also indicated as our nation weighs how it will finance the retirement of aging baby boomers. 

The new inflation index calculation methods, however, are increasingly moving from simply computing the changes of price in a fixed market basket of goods and services, to far more difficult to measure intangibles such as substitution behavior or quality differences.  In the process, the COLAs that seniors depend upon have been, and will continue to be, effectively cut.  Because this process compounds over time, TSCL is concerned that increasing numbers of seniors, especially those who have little or no other income except for Social Security, will fall below the poverty line.

In several surveys of older Americans, TSCL found a high level of concern among seniors regarding their purchasing power.  If, as predicted, health care costs continue to rise at a far faster rate than the general rate of inflation, lower-income seniors will face a grim future where essential health care becomes unaffordable and their ability to purchase other essentials is compromised.

TSCL strongly supports legislation that would create a more fair Social Security COLA by using the CPI-E rather than the current CPI-W to determine the annual increase.  TSCL supports H.R. 2035, "Consumer Price Index for Elderly Consumers Act" introduced by Representative Bernie Sanders (I-VT).  This measure directs the Bureau of Labor Statistics to make the CPI-E the permanent basis for calculating Social Security COLAs.  TSCL will seek to get a Senate counterpart bill introduced early in 2002. You can petition your Congressional representative to support H.R. 2035 by signing our online petition.

Sources:

(1) "Social Security's Cost-of-Living Adjustments: Can Reforms Protect The Most Valuable Recipients?" Eugene Steuerle, Christopher Spiro, and Adam Carasso, The Urban Institute, September 30, 1999.

(2) "Toward A More Accurate Measure Of The Cost-of-Living: Findings And Recommendations Of The CPI Commission, " Michael J. Boskin, Testimony before the Senate Budget Committee, January 30, 1997, pg. 4.

(3) "Growth In Medicare And Out-Of-Pocket Spending: Impact on Vulnerable Beneficiaries," Stephanie Maxwell, Marilyn Moon,and Misha Segal, The Urban Institute, December 2000.

(4) "Towers Perrin Survey 2002," November 6, 2001.

(5) "Propelled By Drug And Hospital Costs, Health Spending Surged in 2000," Robert Pear, The New York Times, January 8, 2002.

(6) "Bureau of Labor Statistics, October 25, 2001.

(7) "2001 CPI-E Study: Retiree With Average Monthly Benefit In 1984," Mary Johnson, The Social Security and Medicare Advisor, October 2001.

(8) "At What Price? Conceptualizing And Measuring Cost-of-Living And Price Indexes," National Academy of Sciences, December 2001.

(9) "Overview Of The 1998 Revision Of The Consumer Price Index," Bureau of Labor Statistics, August 11, 1997.

(10) "Toward A More Accurate Measure Of The Cost-Of-Living: Findings And Recommendations Of The CPI Commission," Michael J. Boskin, Testimony before the Senate Budget Committee, January 30, 1997, pg. 6.

(11) "The Distributional Implications Of Reduction In Social Security COLAs," Richard W. Johnson, The Urban Institute, 2000, and "Consumer Price Index:  Update of Boskin Commission's Estimate of Bias," GAO/GGD-00-50, February 2000.

(12) "The Economic  And Budget Outlook For Fiscal Years 1999-2008: A Preliminary Update," Congressional Budget Office, July 15, 1998 and "The Distributional Implications Of Reduction In Social Security COLAs," Richard W. Johnson, The Urban Institute, 2000.

(13) "An Economic Speedometer Gets An Overhaul," Jolie Solomon, The New York Times, December 23, 2001.

(14) "Future Schedule For Expenditure Weight Updates In The Consumer Price Index," Bureau of Labor Statistics, October 16, 2001.

(15) "An Economic Speedometer Gets An Overhaul," Jolie Solomon, The New York Times, December 23, 2001.

(16) "At What Price? Conceptualizing And Measuring Cost-of-Living And Price Indexes," National Academy of Sciences, December 2001.

(17) "An Economic Speedometer Gets An Overhaul," Jolie Solomon, The New York Times, December 23, 2001.

(18) 2002 Study: Impact of 0.8% Change Person Receiving Average Benefit of $874 in 2002," Mary Johnson, The Social Security & Medicare Advisor, January 15, 2001.

(19) "The Distributional Implications Of Reduction In Social Security COLAs," Richard W. Johnson, The Urban Institute, 2000.

March 2002