An eagle-eyed reader caught a mistake in my column last Thursday, gently suggesting that price indexes apparently must be confusing even to experienced economists, let alone laypeople.
I wrote that the Fed Funds rate was 3.9 percent when adjusted for inflation. I should have said it was about 2 percent given that the nominal Fed Funds rate was 5.25 percent and most inflation measures were running more than 3 percent.
Even economists (and their editors) make typos and fail to catch their mistakes. Still, the reader offered a timely suggestion that I take the opportunity to explain the tabulation and uses of price indexes.
Two measures of July’s inflation came out this week. On Tuesday, the Producer Price Index showed a one-month increase of 0.1 percent, less than expected. Stock and bond prices rose smartly as traders decided that the small increase in prices would avert further Federal Reserve tightening of the money supply.
On Wednesday, the U.S. Bureau of Labor Statistics released the better-known Consumer Price Index. It showed retail prices had risen 0.4 percent in the same month measured by the PPI. This was twice June’s CPI increase.
But, since most of the gain came in one category, energy prices, financial markets took the news in stride. As students learn in introductory classes, increases in the price of one or two products do not constitute inflation.
Inflation is frequently defined as an increase in the general price level. Deflation, which many feared in the United States five years ago, is a decrease in the general price level. Price indexes are a systematic way to estimate changes in general prices.
Tabulating an index is fairly straightforward: Decide what set of prices you wish to monitor — those affecting consumers, businesses or all goods and services in the economy. Select a representative subset of items whose prices you can verify periodically. Go out, check all these prices and calculate the total value of your “market basket” of goods and services.
A month, quarter or year later, go out and recheck prices for the same items. Calculate what the basket costs now. Compare that sum to the starting point. To make comparisons easy, construct a scale that equals 100 at the time you start. If the total cost of all the items increased 4 percent between your initial survey and the subsequent one, then the index is now at 104.
This change of 4 points in the index equals 4 percent annual inflation if the interval between checking prices was one year. But such a 4-point move equals the percentage change in prices only for the first move from the starting point of 100. If the index stands at 250, a 4 percent increase in prices would change the index by 10 points to 260.
The current CPI measures prices compared with prices during the base period of 1982 through 1984, when the index stood at 100. In June 2006, it was at 202.9 and increased to 203.5 in July. This 0.6 point increase in the index indicates about a 0.3 percent increase in general prices.
July’s index of 203.5 means that consumers pay more than twice as much for a given set of goods as they did, on average, in the 1982-1984 period. One also can say that prices have increased more than 100 percent in two decades.
An older CPI that stood at 100 in 1967 is now at 609.6. Thus, a set of items worth $10 when I got out of high school would cost about $61 now.
The difference between the point change in the index number and the percentage change in prices confuses many. It is not the only source of confusion, however. Going from monthly to annual rates also is complicated.
Many consumers understand that a 1.5 percent monthly interest rate on a credit card results in an effective annual percentage rate of 19.6 percent — more than the 18 percent one gets by multiplying 1.5 percent per month times 12 months.
In the same way, a 0.3 percent increase in prices over one month would exceed 3.6 percent if the gain continued for a whole year. The same compounding formula used to calculate effective interest rates shows that a 0.3 percent increase each month over the preceding month would total 3.66 percent after a year.
The difference is small when inflation is low but becomes more important when inflation is high. The 40 percent monthly inflation Brazil had in early 1992 implied an annual rate of nearly 5,700 percent, not 480 percent. Thankfully, this difference has not had any practical importance in our country for 25 years.
One also can look at CPI changes over longer intervals. If you compare the July 2006 level of 203.5 to the 195.4 in July 2005, you can tell that consumer prices rose 4.1 percent in a year. Even so, such year-over-year calculations are not very helpful in determining whether inflation is accelerating or slowing in the short run.
Changes in what consumers buy also pose knotty problems. Vinyl albums and bell-bottom jeans were important in 1967, but double espressos were rare and iPods hadn’t been invented. I’ll need another column to explain that complication.
© 2006 Edward Lotterman
Chanarambie Consulting, Inc.