Simply put, inflation is what happens when the cost of living goes up due to a rise in the prices of common goods and services. People worry about inflation because it threatens to erode their purchasing power: when everything costs more than it used to, someone whose income hasn’t gone up is able to buy less than before.  

Of course, employers can give people raises to adjust for inflation, and the government can make adjustments to Social Security checks. But how big should the adjustment be? To answer that question, we need to be able to measure inflation, not just be aware of it qualitatively. But if inflation is essentially a rise in cost of living, then to measure inflation we must first get a good handle on the cost of living. This is where the consumer price index comes in.  

The consumer price index, or CPI, is a cost-of-living indicator. It measures the total cost of goods and services purchased by a typical consumer within a country and thus is one way of defining the national price level—the price of “things in general.” The CPI allows economists and consumers to see both how much purchasing power a dollar has and how that power varies from one year or era to another.  

The consumer price index is based on the overall cost of a fixed “basket” of goods and services bought by a typical consumer, relative to the price of the same basket in some base year. By including a broad range of thousands of goods and services in the fixed basket, the CPI can obtain an accurate estimate of the cost of living. The CPI tracks the dollar cost of the basket, but it is reported as an index that equals 100 for the base year. For example, if prices are rising at a steady pace of around 2 percent per year, the CPI would be 98 for the year before the index year and 102 for the year after the base year. 

Constructing the CPI

Each month, the Bureau of Labor Statistics publishes an updated CPI. While in practice this is a labor-intensive task that requires the consideration of thousands of items and prices, in theory computing the CPI is a simple four-step process: 

  1. The fixed basket of goods and services is defined. This requires figuring out where the typical consumer spends their money. The Bureau of Labor Statistics surveys consumers to gather this information. 

  1. The price of every item in the fixed basket is found. Some of the data is gathered electronically, but other data may require a researcher to visit retail outlets in person. Since the same basket of goods and services is used across a number of time periods to determine changes in the CPI, the price of every item in the fixed basket must be found for every point in time. 

  1. The cost of the fixed basket of goods and services must be calculated for each time period. If the same item appears in the basket several times, the total contribution of that item is found by multiplying the quantity times the price. 

  1. A base year is chosen. The price of the fixed basket of goods and services for each comparison year is then divided by the price of the fixed basket of goods in the base year. The result is multiplied by 100 to obtain an index of the cost of living for each comparison year, relative to the base year. So the formula is: 

\(\text{Consumer Price Index} = (\text{Comparison year basket price}/\text{Base year basket price}) \times100\)

For example, let's compute the CPI for Country B. In this simplified example, consumers in Country B only purchase bananas and backrubs. The first step is to fix the basket of goods. The typical consumer in Country B purchases 5 bananas and 2 backrubs in a given period of time, so our fixed basket is 5 bananas and 2 backrubs. The second step is to find the prices of these items for each time period. This data is reported in the table.  

A table that tracks the prices of bananas and backrubs over time. At time 1, bananas cost $1 and backrubs cost $6. At time 2, bananas cost $2 and backrubs cost $7. At time 3, bananas cost $3 and backrubs cost $8.  

The third step is to compute the basket's cost for each time period. In time period 1, the fixed basket costs \((5\times$1)+(2\times$6)=$17\)

In time period 2, the fixed basket costs \((5\times$2)+(2\times$7)=$24\)

In time period 3, the fixed basket costs \((5\times$3)+(2\times$8)=$31\)

The fourth step is to choose a base year and compute the CPI. Since any year can serve as the base year, let's choose time period 1.

The CPI for time period 1 is \(($17/$17) \times100 =100\)

The CPI for time period 2 is \(($24/$17) \times100 =141\).

The CPI for time period 3 is \(($31/$17) \times100 =182\).

Since the price of the goods and services that comprise the fixed basket increased from time period 1 to time period 3, the CPI also increased. This shows that the cost of living increased across this time period. 

Problems with the CPI

While the CPI is a convenient way to compute the cost of living and the relative price level across time, because it is based on a fixed basket of goods, it does not provide a completely accurate estimate of the cost of living. Three problems with the CPI deserve mention: the substitution bias, the introduction of new items, and changes in quality. Let's examine each of these in detail. 

The first problem with the CPI is substitution bias. As the prices of goods and services change from one year to the next, they do not all change by the same amount. The number of specific items that consumers purchase changes depending upon the relative prices of items in the fixed basket: to the extent consumers can easily do so, they substitute away from products whose prices have risen more, toward products whose prices have risen less. For example, if the price of backrubs in Country B jumped to $20 in time period 4 while the cost of bananas remained fixed at $3, consumers would likely purchase more bananas and fewer backrubs. But since the quantities of the items in the basket are fixed, the CPI calculation does not reflect this consumer behavior.  

The second problem with the CPI is the introduction of new items. As time goes on, new items enter into the basket of goods and services purchased by the typical consumer. For example, if in time period 4 consumers in Country B began to purchase books, this would need to be included in an accurate estimate of the cost of living. But since the CPI uses only a fixed basket of goods, the introduction of a new product cannot be reflected. Instead, the new items, books, are left out of the calculation in order to keep time period 4 comparable with the earlier time periods. 

The third problem with the CPI is that changes in the quality of goods and services are not handled well. When an item in the fixed basket of goods used to compute the CPI increases or decreases in quality, the value and desirability of the item changes. For example, if backrubs in time period 4 suddenly were to become much more satisfying than in earlier time periods, but the price of backrubs did not change, then the cost of living would remain the same while the standard of living increased. This change would not be reflected in the CPI from one year to the next. While the Bureau of Labor Statistics attempts to correct this problem by adjusting the price of goods in the calculations, in reality this remains a major problem for the CPI.