World energy prices are sliding because of oversupply, according to current news reports. At the same time, however, the government has announced that the Consumer Price Index is up because of a sizable 1.1 percent jump in gasoline prices.
What's going on?
The truth is that energy prices are down. But it is also true that the latest CPI -- because of its controversial seasonal adjustment -- transformed what was actually a small gasoline price increase of 0.5 percent into the exaggerated 1.1 percent increase.
Seasonal adjustments are intended to clarify price trends, but sometimes they have the opposite effect. Here is how that happened in September, according to Pat Jackman, a U.S. Labor Department economist who helps compile the CPI each month:
Gasoline prices generally rise during the summer, when there is greater demand because of vacation driving, and then drop during winter, when demand is lower.
To compensate for those patterns in computing the CPI, the government has developed seasonal adjustment factors for gasoline prices for each month of the year. The adjustment consists of the percentage change expected for gasoline for a given month, compared with the previous month. The change in the price of gasoline, based on pump prices, must exceed the amount of the monthly seasonal factor to be considered a legitimate increase or decrease.
In the CPI for September, the latest month for which figures are available, U.S. gasoline prices went up 0.5 percent, according to government pump price surveys. But because gasoline prices traditionally rise in September, the government factored in its seasonal adjustment of 0.6 percent. The result is that for the CPI reported, gasoline prices nationally rose 1.1 percent (0.6 plus 0.5) during September.
While gasoline prices were up a tad nationally (but not as much as suggested by the CPI), they were dropping in the Washington area. Here pump prices fell 0.1 percent in September, according to the CPI. But the government doesn't make seasonal adjustments to the CPI that is published for Washington or any other region. The adjustments are used only in the national CPI.
The change in the seasonal factor, from month to month, looked like this for 1984: January, 0.2 percent; February, 1.1 percent; March, 1 percent; April, 1 percent; May, 0.5 percent; June, 1.1 percent; July, 0.4 percent; August, 0.2 percent; September, 0.6 percent; October, 1.1 percent; November, 0.7 percent; December, 0.3 percent.
Other components of the CPI, such as food and clothing, also are adjusted seasonally, but the factors are different from those used for gasoline. Seasonal adjustments for food, for example, take into account the effect of weather on food supplies. The seasonal adjustment for clothes takes into account the effect of seasonal sales.
Seasonal adjustments have been part of the CPI since the mid-1960s, but many economists have come to oppose their use because of the distorted picture they can create. Some contend that the CPI has lost credibility because it relies on seasonal adjustments.
"The problem," said Jackman, "is that the man in the street doesn't understand seasonal adjustments and doesn't pay for them in the marketplace. So if we talk about something declining when, in fact, it increased, though perhaps not as much as usual, it turns off the man in the street because he knows he paid more this month than last month.
"And he is right."