Playing Inflation Games with Grandma: The Washington Consensus and the Chained CPI

Print
Written by Dean Baker   
Wednesday, 04 April 2012 07:38

All the inside Washington types seem to agree, we should change the indexation of Social Security benefits to the chained consumer price index (CPI). This would supposedly make the annual cost-of-living adjustment (COLA) more accurate and save the government big bucks. Sounds great, right?

First of all, when all the inside Washington types agree on something, it is a good idea to hang on to your pocket books. Remember, these are the folks who thought it was great that everyone was becoming a homeowner in the middle of a housing bubble and that Alan Greenspan was the greatest central banker of all-time. In other words, inside Washington types are a group of people that mindlessly repeat the conventional wisdom and are largely incapable of original thought.

At the most simple level, the switch to a chained CPI is a way to reduce the annual COLA in Social Security by roughly 0.3 percentage points. That may sound trivial, but it is important to remember that this sum adds up over time. After ten years, this lower annual cost-of-living adjustment would imply a reduction in benefits of roughly 3 percent, after 20 years the reduction would be 6 percent, and after 30 years close to 9 percent. So this is real money.

This plan to lower the COLA raises two obvious questions. First would the new measure actually be more accurate, and second is a cut in Social Security benefits good policy?

There are some complex philosophical issues raised by a cost-of-living index but at the most basic level, the question is to what extent Social Security beneficiaries substitute between items to offset price increases. The proponents of switching to a chained index for the COLA are arguing based on research that examines the consumption patterns of the population as a whole.

The Bureau of Labor Statistics (BLS) has done research indicating that the Social Security population has qualitatively different consumption patterns than the rest of the population. This research suggests that a consumer pirce index based on the consumption patterns of the elderly would show a higher rate of inflation.

The BLS research would imply that someone who is concerned about the accuracy of the Social Security COLA might want a higher annual cost-of-living adjustment, not a lower one. Of course the BLS research is not conclusive, since BLS did not directly monitor the actual purchasing patterns of the elderly, examining the specific items they buy and the outlets where they shop. 

However, BLS could do this and construct a full elderly CPI. This would cost in the neighborhood of $10-20 million. While that may seem expensive, this index is being used to determine a COLA for $700 billion in annual spending. If the full elderly index turned out to show the same rate of inflation as the overall CPI, then there would be no need to continue to do it. However, if the rates differ, then we would continue to maintain the elderly CPI, if the interest is accuracy.

This is a simple way to distinguish between people who want an accurate COLA and people who just want to cut benefits. Those who want an accurate COLA advocate having BLS construct a full elderly CPI. People who just want to switch the indexation to a chained CPI simply want to cut benefits.

This brings up the second question as to whether a cut in Social Security benefits is a good idea. The Washington insiders like to treat this proposed cut in benefits as a small matter -- it's just 0.3 percent a year. But this adds up. If a typical beneficiary receives benefits for 20 years then the average cut in benefits is roughly 3 percent.

Is that a big deal? For almost 40 percent of beneficiaries Social Security benefits are more than 90 percent of their income. For these people, a 3 percent cut in benefits is roughly the same as a 3 percent cut in income.

If we want to know whether that is a big deal we need only turn to the Republicans who are screaming about President Obama's plan to end the Bush tax cuts. For high end earners this would raise the top marginal tax rate by 4.6 percentage points. Since most high-end earners will have most of their income taxed at lower rates, the switch to a chained CPI would probably have more impact on the income of most Social Security beneficiaries than the ending of the Bush tax cuts would have on the income of most high-end earners.

This means that if we think the Bush tax cuts are a big deal to high-end earners then the chained CPI is certainly a big deal to most seniors. It is not something to be done in the middle of the night by Washington insiders who think they know more than the rest of us.

It is important to remember in this story that most seniors are approaching retirement with very little other than their Social Security. This is the fault of the Washington insiders who were not able to see the $8 trillion housing bubble, the collapse of which wrecked the economy. It also destroyed most of the savings of most older workers, which was the equity that they had in their homes. 

In short, there is an argument that a chained CPI could provide a more accurate basis for the Social Security COLA. Anyone who really believes this argument would support having the BLS do the research that would determine whether or not it is true.

Alternatively, there are those who see the CPI as a way to cut Social Security and would like beneficiaries to believe that it is not a big deal. Whether or not something should be viewed as important is subjective. However for most beneficiaries, switching to a chained CPI would have a larger impact on their retirement income than ending the Bush tax cuts would have on the income of most high income households.

The latter has certainly been viewed as a very big deal in American politics. To be consistent, we must also view the switch to the chained CPI as a very big deal.