new site graphic2

Pain With the Consumer Price Index

The Drive to Cut Social Security Benefits


There is a full-fledged drive to cut Social Security benefits by lowering the annual cost-of-living adjustment for people already receiving benefits. The plan involves changing the index for calculating the cost of living. The new index, which is known as the "chained consumer price index" (CCPI) typically shows a rate of inflation 0.3 percentage points less than the CPI currently used to adjust benefits.

A reduction of 0.3 percent in benefits may seem small, but this will accumulate through time. After being retired 10 years, benefits will be almost 3.0 percent lower with the CCPI. After 20 years the loss will be near 6 percent, and after 30 years the reduction in benefits will be close to 9 percent. This is a serious loss of income for seniors, the vast majority of whom rely on Social Security for most of their income.

The justification for the change in the benefit formula is that the CCPI takes account of the substitutions that consumers make in response to changing prices. The classic story is that if the price of beef rises and the price of chicken doesn’t, people will buy more chicken and less beef. The CCPI takes this switching from beef to chicken into account in calculating inflation. The current CPI does not.

While there is an argument for taking account of this sort of substitution in the index, there are two important issues that arise when evaluating the cost of living of seniors. First, their consumption patterns differ substantially from the rest of the population. They consume more health care and fewer computers.

The Bureau of Labor Statistics (BLS) has constructed an experimental index that tracks the consumption patterns of the elderly. This index actually has shown a somewhat higher rate of inflation than the CPI currently used to adjust benefits. In other words, it implies that the current cost-of-living adjustment is too low, not too high.

The other problem is that it is not clear that the elderly would be as likely to substitute in response to price changes as the rest of the population. There are three reasons for this. First, many of the items consumed by the elderly don’t lend themselves well to substitution. If the price of heart surgery goes up, people are unlikely to substitute other medical care. Health care and shelter together account for almost half of the consumption basket in the elderly index.

Second, they tend to be a less-mobile population. This means that if responding to a change in prices means traveling further to shop, the elderly might be less capable of doing this than the rest of the population.

Finally, older people may just be more set in their ways. If they have been eating beef twice a week for 40 years, they may continue to eat beef, even if the price does rise.

At this point we don’t know what a full elderly index that included substitution would show about the cost of living for the elderly. However, if the point of changing the indexation formula for Social Security is to make the indexation more accurate, then it would seem that we would want to find out.

In other words, if the people who claim to want a more accurate cost-of-living adjustment are being honest, then they should be calling for the BLS to construct a full elderly index. This index would then be used for adjusting Social Security benefits. At this point, we don’t know if this index will show a higher or lower rate of inflation. We just know that it will be more accurate.

In the push to cut benefits, many have claims that "all economists agree" that we should switch to the Social Security adjustment to the CCPI and thereby lower benefits. While the claim is not true, it is also worth pointing out that "all economists" have a very bad track record.

"All economists" missed the $8 trillion housing bubble that wrecked the U.S. economy as well as the bubbles whose collapse did similar damage to the European economies. "All economists" thought that the stock bubble of the 90s would just keep inflating indefinitely. In fact those wanting to invest Social Security money in the stock market effectively assumed that price-to-earnings ratios would rise into the hundreds in the decades ahead.

"All economists" even have a very bad track record on this exact issue. Back in the 90s there was an effort to reduce the annual cost-of-living adjustment for Social Security by 1.1 percentage points based on the report of a commission chaired by Michael Boskin, the chief economist for the first president Bush. At that time "all economists" lined up behind the report, agreeing with the Boskin commission that the CPI overstated inflation by 1.1 percentage point.

This effort was defeated in Congress. Remarkably, all the economists who accepted the Boskin commission’s claim that the CPI overstated inflation by 1.1 percentage point then continued to use it as though it was an accurate measure of inflation. (According to estimates from the Boskin commission, changes in the CPI reduced the overstatement by about 0.3 percentage points.) In other words, when there was a political reason to claim the CPI was overstated, "all economists" were willing to rise to the occasion. But when that reason disappeared, they ignored what they previously asserted.

Based on this track record, the public should view "all economists" as people who are either not very good at their work or not very honest, or perhaps both. They should not be looking to them for guidance in policy debates.