Federal Reserve Economic Data

The FRED® Blog

Posts tagged with: "CPIAUCSL"

View this series on FRED

What’s real about wages?

A look at the increases and decreases in wages

People have been talking about the evolution of wages. Some say they’re increasing, others say they’re decreasing. Who’s right? As is so often the case in economics, it depends. First let’s look at the graph above, which has four different indicators for wages. Three of them show a clear and steady upward trend. But one of them—the green line, which shows median weekly earnings—is starkly different. It could be because the median is different from the mean if the distribution of wages skews strongly at the top. Or it could be that people work less per week. Or it could be that it’s a real measure, whereas the others are nominal.

The second graph corrects for this bias. The three nominal series are now real, after being divided by the consumer price index so that general price increases aren’t reflected in the wage. Now all four series evolve along basically the same path. It’s clear that decreases can be frequent and sometimes long lasting. It’s also clear there’s a lot of variability, which means one should really wait for a good amount of data before reaching for any conclusions.

How these graphs were created: For the first graph, search FRED for “wage” and pick the four series. Limit the time period to the past 10 years. From the “Edit Graph” section, choose “Index” for the units with the default of 100 at the end of the last recession. Then click on “Apply to all.” For the second graph, add the CPI to each of the three nominal series, apply formula a/b, and again choose “Index” for the units.

Suggested by Christian Zimmermann.

View on FRED, series used in this post: CES0500000003, CPIAUCSL, ECIWAG, LES1252881600Q, USAHOUREAQISMEI

New tax code = new price index = new tax bracket adjustments

How we measure price changes can affect our income tax

The new Tax Cuts and Jobs Act changed both personal and corporate income taxes. Much of the discussion has focused on the changes in the tax rates, but there’s another change in this law that has an effect on personal income tax. A household’s tax obligation depends on the income bracket for their earned income. For example, a married couple earning $77,000 faces a tax rate of 12% on all income over $19,050. The next bracket starts at $77,400. While the new tax code decreased the tax rate, another change that has received less attention deals with how the tax brackets are adjusted over time. Economists argue that the income brackets should be adjusted when prices change. If the household in this example received a 3 percent cost-of-living increase, their new income level would be $79,310. Despite their inflation-adjusted income level not changing, this household would now fall in the next income bracket and face a 22% tax rate on income over $77,400. If the tax brackets were adjusted for the increased inflation, the marginal rate would not change. This is referred to as indexing the tax brackets.

The Act continues to adjust the tax brackets for price changes, as did the prior personal income tax legislation. The difference is the price index used to adjust those tax brackets. The previous tax code used the CPI-U measure to adjust the brackets, while the new code uses the C-CPI-U measure. The first measure, the consumer price index for urban consumers, assumes you purchase the same quantity of each good in the CPI basket over time. In contrast, the C-CPI-U, or chained consumer price index for urban consumers, recognizes that individuals can shift their expenditure patterns toward cheaper goods in the same expenditure category. For example, suppose chicken prices increase while turkey prices remain unchanged. The CPI-U implicitly assumes individuals continue to buy the same quantity of chicken. In contrast, the C-CPI-U recognizes that consumers will see the price change and substitute turkey for chicken. As a result of this expenditure shift, the price change is less. In the graph, these two price indexes are shown for the period 2000 to 2017 and indexed so that both equal 100 in 2000. The prices in the C-CPI-U index increase at a slower rate because individuals can shift their purchases toward cheaper substitute goods.

The choice of CPI measure has consequences. As shown in the graph, C-CPI-U implies smaller price increases than the CPI-U. This means the tax brackets will have smaller adjustments and more individuals will fall into the higher tax brackets and pay more taxes. If social security payments are linked to the C-CPI-U, then social security payments will increase at a slower rate. In the popular press, some have pointed out that this change in the price index has resulted in an implicit increase in tax revenue, which is true. But from an economic perspective, the primary concern is which index more accurately measures the change in the prices of goods that individuals purchase.

How this graph was created: Search for “cpi” and select “Consumer Price Index for All Urban Consumers: All Items.” From the “Edit Graph” menu, click “Add Line” and enter “chained cpi” in the search box. Select “Chained Consumer Price Index for all Urban Consumers: All items” and click “Add Data Series.” Change the units to “Index (Scale value to 100 for chosen date),” change the date to 2000-01-01, and click “Copy to all.” Finally, adjust the date range to start at 2000-01-01.

Suggested by Daniel Eubanks and Don Schlagenhauf.

View on FRED, series used in this post: CPIAUCSL, SUUR0000SA0

Healthy inflation?

Inflation in the healthcare industry vs. general CPI

Some components of the consumer price index have consistently, over several decades, risen faster than the rest. This blog recently discussed education as one such component. The components of the CPI devoted to medical care have also seen faster price increases than the rest of the basket. Going back as far as the series are available, since 1948, the price of medical care has grown at an average annual rate of 5.3% while the entire basket, headline CPI, has grown at an average annual rate of 3.5%. In the past 20 years, in the regime of stable inflation, headline CPI has grown at an average annual rate of 2.2%, whereas the price level of medical care has grown at an average annual rate of 3.6%—about 70% faster.

The graph above shows the two time series. Besides the difference in their levels, it’s also notable how much less cyclical medical care inflation is. Although overall CPI inflation dips during recessions, medical care inflation stays steady.

The implication of these two features is far reaching: It’s symptomatic of the increasing share of income the U.S. spends on medical care. Beyond macro trends, the features of these two series themselves have policy implications. Indeed, indexing government healthcare budgets to overall CPI rather than medical care prices has implications for spending in real terms. This gap could also widen during recessions, when government help may be most in demand.

The CPI is intended to measure the price of goods consumers purchase directly, and therefore the medical care subset is actually measuring only the prices of out-of-pocket expenses. For healthcare, however, there’s a great deal of other spending going on. And the inflation rate of that spending is something a policymaker might need to know. Luckily, the BEA puts together a more holistic price index for healthcare spending—the health expenditures price index—which we add in the graph below. Although the history of this series is shorter, this measure of healthcare prices is still rising considerably faster than headline CPI: In 2001-2013, this measure of healthcare inflation rose almost 4% per year, whereas headline CPI rose 2.3% in this period and the other healthcare CPI rose 3.9%.

How these graphs were created: For the first graph, search for “Consumer Price Index Medical.” In the “Edit Graph” tab, convert the units to “Percent Change from a Year Ago.” Then use the “Add Line” feature to search for “Consumer Price Index All Items.” Add this line and again check that its units are the same. (FRED does this automatically, but it doesn’t hurt to check.) These series are both also available as chained indices, but for a shorter period. For the second graph, add to the first another line by searching for “Health Expenditures Price Blended Account.” Then restrict the period to show the entirety of the new line.

Suggested by David Wiczer.

View on FRED, series used in this post: CPIAUCSL, CPIMEDSL, HLTHSCPIBLEND


Subscribe to the FRED newsletter


Follow us

Back to Top