If you’ve ever listened to a Federal Reserve press conference, you may have heard the terms, CPI, PCE deflator, and GDP deflator as metrics that describe inflation. In this article, we will explore the primary differences between these inflation metrics.
GDP Deflator vs. Consumer Price Index (CPI)
Gross Domestic Product (GDP) deflator is the factor multiplied to real GDP to arrive at nominal GDP. Please refer to this article to learn more about GDP. Put differently, dividing (or “deflating”) the nominal GDP by the GDP deflator will give us the real GDP.
Nominal GDP is the measure of the total expenditures of an economy at today’s prices. Therefore, nominal GDP growth is driven by both change in prices (inflation) and change in quantities produced. For example, if a country experiences 2% inflation and 2% increase in goods/services produced, the nominal GDP growth rate would be approximately 4%.
Using round numbers, if we assume 2017 nominal GDP is $100 and nominal GDP grew at 4%, then, the 2018 nominal GDP is $104. However, the real GDP, which does not incorporate change in prices, will be $102. Therefore, the GDP deflator is 2%.
If you refer to our article on GDP, you will recall that GDP is measured as:
GDP = Consumption + Investment + Government Purchases + Net Exports
GDP considers expenditures not just of consumers, but of businesses and the government. Also, GDP considers only domestic purchases since imports are subtracted from the calculation (Net Exports = Exports – Imports).
These two attributes of GDP are what drives two of the main differences between GDP deflator and CPI. There is a third difference based on how the formula for GDP deflator and CPI are calculated.
GDP deflator measures prices of purchases by consumers, government, and businesses. However, CPI measures prices of purchases by consumers only. Therefore, goods purchased by the government will factor into the GDP deflator but will not factor into the CPI
GDP deflator measures prices of domestic expenditures only since imports are subtracted out of the GDP formula. On the other hand, CPI measures the price level of expenditures that include both domestic and foreign items. For example, a consumer purchase of clothing made in China does not factor into the GDP deflator, but it does factor into the CPI
The GDP deflator assumes changing quantities in its calculation vs. the CPI which uses fixed quantities. For example, let’s assume that the CPI uses 100 units of corn in its calculation. The CPI will calculate the change in the price of corn while assuming consumers continue to purchase 100 units of corn. Now, let’s imagine that a drought wipes out 99% of corn, and the single remaining unit of corn has gone up in price by 1000%. The CPI will still assume that 100 units of corn are purchased even though 100 units of corn are no longer available for purchase. Whereas, the GDP deflator will incorporate lower units of corn purchased and increased other types of vegetables purchased as consumers shift spending into alternative products, which is referred to as substitution. While the GDP deflator method may seem more logical in this extreme example, there are pros and cons to both formula calculations. The GDP deflator method, which does allow for consumer product substitution, may not adequately reflect the decrease in quality of living if consumers must substitute products into products that they don’t want as much as the original
More Detail on the Consumer Price Index (CPI)
As mentioned above, the CPI measures prices paid by consumers on both domestic and foreign items, and it excludes purchases by businesses and governments. The CPI is calculated by the Bureau of Labor Statistics (BLS) through a detailed data collection process which we outline below:
Basic Steps in Calculation of CPI:
The BLS establishes a basket of goods to track prices on for their pricing surveys. The basket of goods is established through the Consumer Expenditure Survey, which is a survey of thousands of consumers who track their spending patterns. The data coming out of the Consumer Expenditure Survey is used to calculate the relative weightings of products (based on their importance) that will be used in the calculation of the CPI. The weightings for products are updated in January of every even numbered year
BLS data collectors call or visit thousands of stores, rental units, restaurants, etc to record prices on over 80,000 items every month
Pricing data is sent to the BLS national office for review for accuracy and to adjust the data due to changes in size, quantity, or quality of an item
The CPI is released monthly by the BLS through a press release
The most commonly used CPI measure is the CPI-U, which measures spending for all urban consumers and for all items (including food and energy). Additionally, many investors prefer to look at the core CPI, which is the CPI-U measure excluding food and energy costs since many people view food and energy costs as being rather volatile, which can skew the rest of the data.
CPI vs. Personal Consumption Expenditure (PCE) Deflator
Personal Consumption Expenditures (PCE) is the primary component in GDP as close to two-thirds of GDP consists of consumption expenditures. PCE along with CPI are the two primary metrics that individuals look at when discussing inflation. Historically, the Federal Reserve focused on CPI when evaluating inflation, but critics of the CPI believed that CPI overstated inflation since the CPI does not allow for the substitution effect. Substitution is a concept that allows for consumers to substitute lesser priced items as the price of certain items increase. Instead, the CPI assumes that consumers continue to purchase the same quantity of a good even if the price goes up substantially.
After evaluating the alternative, the Fed decided to switch their preferred inflation gauge to the PCE in 2000. In 2012, the Fed also set a formal inflation mandate of 2%, using core PCE (excluding food and energy) as their target.
Since 2000, CPI has generally outpaced PCE by around 50 bps (0.50%), but since 2008, the gap between CPI and PCE has narrowed. In the construction of the PCE, the Bureau of Economic Analysis (BEA) uses as much of the pricing data from the CPI. Therefore, the primary factor driving the difference is not the pricing data, but instead, the difference comes from the following three factors:
Formula – the CPI uses fixed quantities in the calculation of an index. For example, let’s assume the CPI says a consumer will purchase 5 bananas every month. Even if the price of bananas double, the CPI will still assume that consumers purchase 5 bananas every month. However, the PCE tries to account for the substitution effect when prices rise. In this example, the PCE may assume that consumers would reduce the quantity of bananas purchased and substitute some oranges instead
Scope – the CPI and PCE do have differences in the scope of goods and services included in their respective indices. For example, the CPI only accounts for out of pocket medical costs, but excludes medical costs consumed but not paid for out of pocket, such as Medicare, Medicaid, and employer sponsored insurance
Weighting – each of the CPI and PCE indices uses a weighting assigned to baskets of goods and services based on their relative importance to consumers. The CPI calculates these weightings based on results of the Consumer Expenditure Survey, whereas the PCE uses surveys of businesses when calculating these weightings.
The below table from the Chicago Fed quantifies some of the major differences between the CPI and PCE:
Source: Federal Reserve Bank of Chicago, Essays on Issues, Number 347