The Problem with Gasoline Prices

Yesterday's report on consumer prices from the Bureau of Labor Statistics had many breathing a sigh of relief. Maybe with lower gasoline prices, inflation is back under control.

And maybe not.

If the last year is any guide, "inflation" doesn't really matter any more - energy prices drive both economic statistics and markets these days. In fact, statisticians could probably dispense with a boatload of economic reports and simply report the price of gasoline - no one would be any worse off.

But, the reporting of energy prices, and in particular gasoline prices, is becoming problematic these days if the intent is to report moderate price increases regardless of what things actually cost. If the goal of policy makers is to keep "reported inflation" low then there is a serious problem with gasoline prices - they are simply too easy to measure.

Let's take a look at what rising energy prices have done to consumer price statistics in the last few years and see if there is anything that can be done about this problem.

The All Items Index

In the last month, reversing energy prices have caused the trend of the last year to be stood on its head.

That little squiggle downward at the far right of the All Items index below is what you get when gasoline prices plunge more than 13 percent in a single month - there's more to come in next month's report as the 13 percent drop covers only about half of the total decline in gas prices seen since the early September energy sell-off.
It doesn't look like much above, but look at the chart below.

It took a while, but that year long adulterous liaison with four percent inflation appears to be over. Reported inflation has headed back to two percent just in time - some were beginning to doubt the inflation fighting mettle of the Federal Reserve, but truth be told, Fed policy had little to do with the return of two percent.
The recent taming of inflation had more to do with hedge funds exiting energy positions en masse in mid September after a number of factors aligned. If energy prices stay low, the Fed will probably get the credit - we'll see.

The Problematic Gasoline Price Index

There is a lingering concern over gasoline prices though. If they go up again, then inflation goes up too. The real problem though is not rising gasoline prices, but the ease with which these prices can be measured.

All you have to do is go to a gas station periodically and write down the numbers from the big sign out front and you can figure out the trend in prices. In the chart below it's clear that from almost four years ago, gasoline prices almost doubled before the sharp decline last month.
The charts above and below are consistent with what most consumers have experienced since 2003 - gasoline prices have risen 15 to 20 percent per year. With the year-over-year change in gas prices plunging below the zero mark in the chart below, many hope, but few believe, this is the beginning of a trend.
Again, the problem is not with the prices themselves, but the ease with which consumers can measure the change in prices. Alan Greenspan complained about this more than once - he said that everyone can measure rising gasoline prices because they fill up their tank regularly. He said it like it was a bad thing - as if consumers should not be provided with such a simple way to measure the declining purchasing power of their money.

The Example of the Owner's Equivalent Rent Index

One way that rising prices can be made easier to swallow is if they are removed completely from the statistics. Such was the case in 1983 when real home prices were removed and "equivalent rent" was substituted for housing costs incurred by homeowners.

The four year chart of home price "equivalent" costs shows a total increase of about ten percent. For some real estate markets last year, home prices rose ten percent in a single month, so clearly, the ideal way to keep reported inflation low amid rising prices is to exclude rising prices completely.
It's clear to see how well housing costs were contained in the consumer price index by looking at the chart below showing the annual change to owners' equivalent rent. There's a developing problem with this approach to containing prices - when rental demand soared earlier this year, after no one could afford to buy real estate at sky-high prices, rental demand began to increase.

This has been counter productive in the effort to keep inflation low.
But still, with this "equivalent" measure of the cost of homeownership rising less than four percent a year, the substitution approach has been largely effective.

The Example of the Medical Care Index

An excellent example of making something difficult to measure is the case of medical costs. In the statistics from the BLS, these prices rise at the rate of only a few percent a year, yet at the same time your typical $15 an hour laborer in 2006 finds that almost 40 percent of his wages go toward health insurance premiums.

That's before he or his family ever see a doctor.

In the chart below, medical care costs appear to have risen a total of less than 15 percent over a period of almost four years. While it's not clear how the BLS came to that conclusion, that seems to be the whole point - if people can't easily measure it themselves, they'll believe almost anything.
The year-over-year increases in medical care costs as shown below are going to surprise anyone who attends their open enrollment meetings next month. The Human Resources representative will probably begin the presentation the same way as they've done for the last few years, "I've got bad news, your health insurance has gone up again".
Average increases to health insurance premiums have been around ten percent for almost five years now and deductibles and co-payments have double, tripled, or more. But these are all costs that can be twisted and contorted into something that doesn't relate directly to a price paid and that's how you get the four percent "feel good" number shown above.

This all perpetuates the deception that costs are not careening out of control across the board for everything that is not shipped here from Asia.

The Solution

So, clearly the problem with gasoline prices as they relate to inflation statistics is that they are just too easy to measure, and as a result they are going to have an outsized impact on overall inflation statistics and the confidence that people have in their currency.

So what's the solution?

Some combination of how housing prices and medical costs are dealt with would seem appropriate.

Perhaps a "transportation equivalent cost" would fit the bill. After all not everyone drives a car anymore and with an increasing number of electric and alternative fuel vehicles on the road, maybe inclusion of gasoline in the consumer price index isn't the best way to measure this transportation expenses.

And the industry should lend a hand here as well - the pricing mechanism for energy is much too straightforward. The oil companies should learn a lesson from the medical care industry and somehow come up with a baffling array of alternative measures to purchase what it is that consumers need.

What's a good guess at the rate of change for this "transportation equivalent cost"?

Probably two percent.