CHANGING PRICE LEVELS AND PRICE INDEXES

When prices change, g.n.p. expressed in terms of dollars is no longer an adequate measure of the real goods and services produced. A 100 per cent increase in real national output would mean a great rise in the national standard of living. But doubling g.n.p. merely through doubling prices is, no real economic gain at all. In order to separate “real” from merely “dollar” changes in individual and national incomes, we have to make an adjust- ment for price-level changes.

Figure 4-3 indicates the big fluctuations in prices in the United States over the last two cen- tunes. The problem of price-level changes is no minor one.

What Is a Price Index?

In 1929, a family income of $1,300 would have bought a group of goods and services called a “subsistence standard of living.” In 1933, you could have bought the same collection of goods and services for about $900. By 1967, you would have needed about $2,500.

If all prices changed in the same proportion in the same direction at the same time, measuring price-level changes would be simple. But the world What Is a Price Index? of real prices is not simple and orderly. Even in the big price rises shown in Fig. 4-3, some prices declined and others rose at very different rates. Yet, even though not all prices rose, we say that the “price level” rOse because the average of all prices rose. The price of the same market basket of goods was higher than before.

Table 4-5 shows how to calculate a simple “price index” for a “market basket” of four com- modities, to show whether their price level went up or down between 1958 and 1968. A price index is a measure of price-level changes.

With 1958 as the “base year,” the price of each item in that year is 100 per cent of itself. This is what people mean when they say, “Take 1958 as 100.” Then we compute what percentage each 1968 price is of the 1958 price. For example, eggs at 6 cents each are 120 per cent of the 1958 price. This 120 per cent is the 1968 “price rela- tive” for eggs, since it show~s the 1968 price relative to that in 1958. To find the change in the price level (the average of all four prices), we simply take the average of the four 1968 price relatives, which gives us 108 per cent for the 1968 price level. (The percentage sign is usually omitted for convenience.) Of course, most price indexes in- clude more commodities, but they are made in a generally similar way.

FIG. 4-3 Prices have fluctuated sharply throughout our
history. The big peaks have come during or after major
wars. So far, there has been no price slump since World
War II. (Source: U.S. Department of Labor.)

This simple method of calculating price- level changes may give misleading results, how- ever. It tacitly assumes that eggs, hamburger, tur- nips, and apples are equally important. A 10 per cent change in the price of hamburger influences the index exactly as much as a 10 per cent change in the price of turnips. Actually, hamburger is more important than turnips in most budgets, and it seems logical that the price of hamburger should affect the index more.

Thus, irs most price indexes, each price is “weighted” according to its importance. For ex- ample, the U.S. Bureau of Labor Statistics weights the prices in its widely-used consumers’ price (“cost-of-living”) index as follows. The statisti- cians take a “market basket” of the goods and services bought during the early 1960’s by typical urban wage earners’ families. Choosing some 400 of the most important prices, they weight each roughly in accordance with the proportion of the family’s total expenditure for that commodity. If the family bought lots of potatoes, potatoes make up a sizeable part of the weekly market basket; rent is a big item in the basket. In effect, the market-basket approach weights the price of each commodity according to the amount spent on that commodity weekly.

Having decided on the contents of the hypo- thetical market basket, the B.L.S. gets its price index by comparing the cost of the basket from one week to the next. The index now uses 1957—59 as a base period (100). So a weekly index reading of 120 means that the cost of the market basket is up 20 per cent from the 1957—59 price level for those goods and services.

Price Indexes for Different Price Lavels

Levels Since a price level is merely an average of some group of prices, we can speak of many dif- ferent price levels. One may reflect the prices paid by a group of consumers (their cost of liv- ing); another, the prices paid by wholesalers; an- other, the level of wages in manufacturing.

Figure 4-4 shows the movements of four important price indexes from 1929 through 1966. ‘What price level is most significant depends on what you are talking about. If you want to mea- sure changes in the cost of living for a particular group, the logical choice is an index of the prices of the goods and services the group buys. Given the wide diversity of price movements, it is hard to devise a single price index that will be a sig- nificant measure for the whole economy. The closest approach is an index developed by the De- partment of Commerce to eliminate the effects of price-level movements from the gross national product. This adjustment is spoken of as flating” the g.n.p. In it, first special indexes are computed for the different major sectors of g.n.p., and then these are combined to obtain a “g.n.p. deflator.” The result is a price index for all goods and services in the gross national product.

Figure 4-5 compares g.n.p. since 1870 in the actual (current) prices that prevailed each year with g.n.p. in constant (1964) prices. We often speak of the constant-price g.n.p. as “real” g.n.p., since with the elimination of price-level changes the g.n.p. index measures real changes in output of goods and services. In periods of rapidly chang- ing prices, like World War II, the money g.n.p. figures clearly give a misleading picture of how rapidly real g.n.p. is rising.