Wednesday, June 22, 2011

Fudging the CPI Numbers

As part of the federal budget talks, there is a proposal on the table to alter the way that the consumer price index (CPI) is calculated. Essentially, the proposed method would try to take into account the fact that consumers often trade down (e.g., go from steak to hamburger) when prices rise.

If passed, the alteration would make the 'headline' inflation number smaller.

Why is this on the table as part of the budget debate? Because a smaller inflation number would lower federal payouts (such as social security) that include cost of living adjustments. Viola! An instant $200 billion in budget savings.

This would not be the first time that the CPI has been dumbed down. There have been multiple changes to the methodology over the past couple of decades. The weird (criminal?) thing is that when the goverment changes the method, they do not go back and alter the historical series. Those looking at historical CPI data are not comparing apples to apples (the same is true for unemployment, GDP, and other measures). If we were measuring the CPI the same way as in 1980, the headline inflation number would be nearly triple the currently reported level.

How such a practice is viewed as legitimate and is tolerated is beyond me. If I had tried to manage measurement systems like this during my industry days, then I would surely have been fired.

Make sure you understand the dynamic here. The federal government is printing money, which undermines the value of the dollar. Government officials are then supressing the metric that is supposed to reflect the dollar's value, effectively under-reporting reporting the inflationary consequences of their activities.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.