Skip to main content

The FRED® Blog

Measuring inflation expectations, part II

In the previous blog post, we looked at using survey data to measure inflation expectations. Now we consider market-based measures. The graph shows various measures of the breakeven inflation rate, which is computed as the difference in returns of constant-maturity Treasury bills: one being the traditional bill and the other being the inflation-indexed bill. If we look at 10-year Treasury bills, we can evaluate what the markets think the average yearly inflation rate will be over the next 10 years. With such a long horizon, it makes less sense to compare these expectations to realized inflation. But this graph still includes a segment to signal the Fed’s 2% inflation target announced on January 25, 2012, since the purpose of that announcement was to anchor inflation expectations.

How this graph was created: Search for “breakeven inflation” and many series will be shown. Here, we used those with a monthly frequency. For the segment, choose “Add a Series” but select “Trend line” from the pull-down menu. Once that’s added, change the initial date to “2012-01-25” and use “2” for both start and end values.

Suggested by Christian Zimmermann

View on FRED, series used in this post: T10YIEM, T20YIEM, T30YIEM, T5YIEM, T7YIEM

Measuring inflation expectations, part I

An important element of monetary policymaking, as well as financial market pricing, is the level of inflation that people expect. There is no direct measure of each individual’s inflation expectations, but we can infer expectations at the aggregate level. One way we do this is to simply ask some people what they think. The Surveys of Consumers, an initiative of Thomson Reuters and the University of Michigan, ask many questions to evaluate consumer sentiment, and one question is what the inflation rate will be over the next year. The average answer is shown in blue in the graph, with the actual inflation rate shown in red. Note that this is not an entirely fair comparison: For any particular date, the blue line shows expectations over the next 12 months and the red line shows actual inflation over the past 12 months.

We also added a short green segment: This is the Fed’s 2% inflation target, announced on January 25, 2012. We see it falls between expectations and realizations.

If you want to learn more about inflation expectations, take a look at this recent Economic Synopses essay.

How this graph was created: Search for “inflation expectations” and the Michigan Survey should be your first choice. Then add the series “CPI” to your graph, making sure to change the units to “Percent Change from Year Ago.” Finally, for the green segment, choose “Add a Series” but select “Trend line” from the pull-down menu. Once that’s added, change the initial date to “2012-01-25” and use “2” for both start and end values.

Suggested by Christian Zimmermann

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

The PPI

The producer price index (PPI) is one of the oldest continuously recorded statistics in the United States. While the better-known consumer price index (CPI) computes the price of a basket of consumer goods in retail stores, the PPI looks at raw materials, intermediate goods, and goods ready to be shipped. In fact, it was previously known as the wholesale price index (WPI). FRED offers the PPI in all sorts of decompositions, over 10,000 series in total.

In the graph, we compare the CPI with the PPI. Notice that the PPI appears to be more volatile, at least in recent years, and the two indexes tracked each other much better before the 1980s than since. In particular, the PPI has increased much less than the CPI and has seen some dramatic drops.

How this graph was created: Search for PPI, and the index for all commodities will likely be your first choice. Add that series to the graph. Then add the CPI by searching for it in the “Graph” tab through the “Add Data Series” panel.

Suggested by Christian Zimmermann

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

SNAP!

FRED recently added SNAP data. SNAP stands for “Supplemental Nutrition Assistance Program”—more commonly referred to as food stamps. FRED has data at the state and county level on the number of benefit recipients for the past 30 years or so. In the graph here, we chose five states with similar population size. A few points stand out: 1) It’s remarkable how little the state rankings have changed over the years. The rise of Arizona is likely due to a relative population increase. 2) Apparent seasonal fluctuations in some states disappear after a few years. 3) Since 2000, the numbers of recipients has more than doubled; in Massachusetts, it quadrupled. This rise has occurred through both booms and recessions and cannot be explained by population increases.

How this graph was created: Search for SNAP, then restrict the results by selecting the “monthly” and “states” tags in the sidebar. Then choose your states and click “Add to Graph.”

Suggested by Christian Zimmermann

View on FRED, series used in this post: BRAZ04M647NCEN, BRIN18M647NCEN, BRMA25M647NCEN, BRTN47M647NCEN, BRWA53M647NCEN

The demographics of the labor force participation rate

There is much lamenting about the decline in the labor force participation rate. As we recently discussed on this blog, while the rate decreased quickly during the previous recession and its recovery, the overall decline began several years before. This decline indicates there must be more than cyclical or even policy-related forces at work. One likely candidate is demographics. In the graph above, the proportion of the U.S. population 25 to 54 years of age follows a pattern similar to that of the labor force participation rate over the past 10 years. Why look at this 25-54 age range? Because this group has the highest labor force participation rate. So, if the share of this age group is declining, the aggregate labor force participation rate is likely to decline as well.

How this graph was created: For the first line, search for “population 25-54” and select “Civilian noninstitutional population—25-54 years.” To create the ratio, add the “Civilian noninstitutional population” series via the “Add Data Series” option: When you add this series, be sure to select “Modify existing series” for series 1. Then use the “Create your own data transformation” option using the formula a/b*100 so that the result is expressed in percentages. For the second line, simply add the civilian labor force participation rate as series 2.

Suggested by Christian Zimmermann

View on FRED, series used in this post: CIVPART, CNP16OV, LNU00000060


Subscribe to our newsletter


Follow us

Twitter logo Google Plus logo Facebook logo YouTube logo LinkedIn logo
Back to Top