Accounting for Inflation

Accounting for Inflation

If beauty is in the eye of the beholder, inflation is in the eye of the statistician. Because the number you end up with is dependent on the models and assumptions you choose. As the chart below shows, there have been two major revisions to how inflation is figured, one in 1983 and another in 1998. (Thanks to Barry Ritholtz at The Big Picture for this source.)

Note that using the same methodology as was used in 1983, inflation would be around 11.6% today. Before 1983, the BLS used actual home prices to account for inflation. After that time, they used something called Owners Equivalent Rent or OER. This is the theoretical price a home would rent for. There are sound reasons to use OER and equally good reasons to use actual home prices (as is done in Europe). But both methods have flaws. You just have to pick a methodology and stick with it.

And there are reasons to think that OER may not rise as it would normally do in this part of the cycle, because so many homes which cannot sell are being rented out, and rent prices might not rise as much as in past cycles.

Different ways of measuring inflation

Using actual home prices is only useful in an average sense over long periods of time. If you own a home with a 30-year mortgage you bought ten years ago, then you have not experienced price inflation for ten years. You have seen the value of your home go up, but that is not (necessarily) inflation. Your mortgage is the same. And a first-time buyer today has the potential to see a 30-50% deflation in home prices from a year ago if he is in the right area, like Florida or California.

Further, the OER tries to measure what a house would rent for. If someone pays more than that rental price, then there is some other factor at work. The Bureau of Labor Statistics suggests that this other factor is investment. If someone pays more for a house than the equivalent rental price because it is perceived as a good investment, then you are measuring apples and oranges. The OER tries to take out the investment angle.

Because the government agencies use OER, inflation was understated in the recent housing bubble. As home prices drop, OER would normally overstate inflation somewhat. If we had used actual home prices then inflation would have been overstated in the last six years, and now the CPI would be turning negative, even as gas and food are rising dramatically.

As I said, neither method is perfect. Over very long periods of time, either will give you reasonably accurate data. But over a time period as short as a few years, let alone a few months, there can be considerable "noise."

Also, notice in the chart that in 1998 the Clinton administration adopted new methodologies, among them hedonic pricing. Hedonic pricing suggests that as a product or service improves, the price for the equivalent item in today's market will fall. As an example, if we buy a computer that is twice as powerful as it was a few years ago, the statisticians assume that prices have fallen even if we pay the same for the computer.

In the same way, if in one year you had to pay extra for features like power steering or power windows in a car, and a few years later they were considered standard, then once again the price would be deemed to have gone down, as you were getting more "value" for your dollar. This is considered to be the case even if in actual dollars you paid more for the car.

Again, you can make a rational and serious economic argument for hedonic pricing. And believe me, many economists do. But those changes, along with others, have lowered the official rate of inflation. And since many government benefits are also tied to the official rate of inflation, the current methodology has lowered government expenses as well, including inflation adjustments for Social Security and pensions.

At one time, you could make a good case that the inflation numbers overstated inflation. But I am not persuaded that is the case anymore, even though many economists still argue that point. The CPI is more or less accurate ON THE AVERAGE. But that may not be the case for you. Your actual rise (or fall) in the level of your expenses may be more or less than the average.

But we do notice the increases more. The Bank Credit Analyst has a very interesting chart in its recent May issue. It shows that the high-frequency spending items like gasoline, food, education, and medical care make up 50% of the Consumer Price Index. These are items which we buy on a regular basis. And they are going up at a weighted average rate of 6.8%, a lot higher than the 4% for the CPI as a whole.

The 20% of the CPI which are low-frequency items like furniture, appliances, vehicles, and so on are actually falling at a -0.7% rate. Since OER (equivalent rent) is roughly 30% of CPI and is rising at 2.8%, even as home prices fall the overall rate is about 4%.

Our tendency to notice the price increases in more frequently purchased items more than the drop in less frequent expenditures is known as salience. What we see every day is more visible to us and is on our minds. And because the reality is that those prices are rising much faster than headline inflation, we tend to think inflation is understated.

I can look at my credit card bills and know that my restaurant bill is rising at much more than 4% a year. I do not think I am eating all that much better, and am actually eating less food in an attempt to hold down my weight. My travel expenses are up by more than the 5-7% in the BLS numbers. Those prices, and the price of turkey, are in my face constantly.