The FRED® Blog

The demographics of the activity rate decline

Many are lamenting the record lows in the labor force participation rate (or activity rate). The debate is whether this decline is cyclical or structural. The structural view has much to do with demographic shifts as the population gets older on average, so let’s look at the rates for different age groups.

We see a strong drop in participation by young men, likely reflecting a larger share who are staying in school longer. Older men’s participation hasn’t changed much. The bulk of the overall decline comes from middle-aged men. They are the largest group and have the largest impact. But their participation has been declining throughout the sample period, so it is not a new phenomenon for them.

For women, the story is very different because of their large increase in labor force participation up until the end of the past century. Older women are still increasing their participation, but recent declines for younger women seem to mirror the declines for men. So, the recently accelerating decline in the overall participation rate may have to do with women’s participation just not increasing like it used to.

How the graphs were created: Search for “Activity Rate,” then use the tags to limit the series to “Nation,” “USA,” and then “Males” or “Females.” Select the series and then add them to the graphs. Depending on the order of the series in the search results, you may have to adjust line colors to make them consistent in the two graphs.

Suggested by Christian Zimmermann

Inflation in the dollar zone

In a recent FRED Blog post, we showed how the exchange rate regime has had an impact on inflation rates in Europe. This time, we look at the dollar zone. Indeed, several countries have adopted the U.S. dollar as legal tender, and it is startling how their inflation rates have rapidly converged toward the U.S. rate. Just look at the graph. This convergence was likely the intention of those countries: Ecuador in 2000 and El Salvador in 2001 switched to the U.S. dollar to fight against very high inflation rates. Panama had already adopted the U.S. dollar in 1904 and has had no such problems with inflation.

How this graph was created: Search for “Inflation” and the respective countries. In the case of the U.S., change the units to “Percent Change from Year Ago” to match the units of the other series. The line width for the U.S. was increased and the color changed to black.

Suggested by Christian Zimmermann

How the exchange rate regime drives inflation

This graph shows inflation rates for some of the countries that founded the euro zone. The sample period encompasses three exchange rate regimes: 1. The first is a fixed exchange rate under the Bretton Woods agreement, which allowed some adjustments but ones that were difficult to achieve. The countries’ inflation rates were similar in this period, with occasional exceptions. 2. This system collapsed in 1971 and gave way to a series of exchange rate arrangements with varying membership and success in limiting exchange rate fluctuations. The graph clearly shows that inflation rates varied considerably from one country to the next, which made it difficult to obtain relatively stable exchange rates. 3. Then came the creation of the euro in 1999. A major requirement of membership to this currency union was a low inflation rate maintained within a small range across candidate countries. The graph nicely shows the convergence in inflation rates, which has been maintained to this date.

How this graph was created: Search for “Inflation” and then limit the selection in the sidebar by choosing tags: “Nation” (under geography types) and “World Bank” (under sources).

Suggested by Christian Zimmermann

Not all books are created equal

The evolution of consumer prices is not uniform across categories, and there can be stark differences between relatively similar products. In the example above, we look at two types of books: those used for recreation and those used for education. The paths of their price indices are very different, even in the long term. While this example is rather extreme, there are plenty of others in the disaggregated CPI data. This shows that one should not use personal experience with the price of a very specific product to draw conclusions about the general price level.

How this graph was created: Search for “CPI book.” To weed out the price indices from other countries, click on USA in the left sidebar tags. Select the two series, not seasonally adjusted (as one is not available seasonally adjusted). Finally, select a right axis for one series, as they have different index years.

Suggested by Christian Zimmermann

How much money is the Fed printing?

We hear frequently that the Fed is printing money like crazy these days. This is not quite true. There are various definitions of money: For money that is being printed, one needs to look at currency in circulation, which actually counts all printed banknotes less those that have not left the Fed’s vaults. So, has the money in circulation increased like crazy since the start of the latest recession?

The currency in circulation (technically called the currency component of M1) is indeed increasing, but there is no indication that it is accelerating. To see this, we have taken the natural logarithm of the series. This means that if the slope is the same for two years, the growth rate is the same. Not taking the natural logarithm would show an illusion of acceleration, as a 1% increase in 2014 would look much bigger than a 1% increase in 1960 because the stock of currency has increased over time.

And why did it increase? One major reason is simply that the economy has grown and needs more currency to function. In the graph above, we divide the currency in circulation by nominal gross domestic product (GDP). While this ratio has indeed increased recently, it is nowhere near historical highs as some commentators seem to imply. In fact, it also seems to follow a neat U-shaped long-term trend. Thus, again, nothing special in recent years.

How these graphs were created: For the first graph, search for “currency” to find the right series. In the graph tab, expand “Create your own data transformation” and select the “Natural Log” transformation. For the second graph, undo the natural log transformation by selecting the empty transformation. Then search for GDP (not the Real one; we want a ratio of nominal series) and add it to series 1. Finally, use the data transformation “a/b” to obtain the ratio.

Suggested by Christian Zimmermann

Quits and layoffs

Consider times when employment has declined. What are the causes? An employment decline can come from fewer hires, more layoffs, and people quitting their jobs. But these factors can interact in complex ways. Indeed, the Job Openings and Labor Turnover release from the Bureau of Labor Statistics shows that hires went down and layoffs went up during the past two recessions. But quits went down, not up; in fact, the decrease in quits partially counteracted the impact the other two factors had on employment, even to the point of entirely canceling the increase in layoffs. This makes perfect sense: The incentive to quit a job is lower when there are fewer opportunities.

The graph also highlights that layoffs came back down quickly after the most recent recession to the lowest levels in this sample. So, the sluggishness of hiring is to blame for the slow recovery in the labor market.

How this graph was created: Go to the Job Openings and Labor Turnover release, select the three series (Rate, Seasonally Adjusted), and click on “add to graph.”

Suggested by Christian Zimmermann

When population drops

The population of a country can decrease for various reasons: fewer people are born, more people die, or migration out of the country is large enough to counter the usual population growth. Japan currently shows no population growth because of their balance between fertility and mortality. For the countries shown in the graph here, the story is mostly demographic and thus economic. Bulgaria, Moldova, and Romania are poor countries from Eastern Europe. In the early 1990s, it became possible to emigrate from these countries to look for better opportunities. Many residents chose to do so, and this trend continues to this day. For Greece and Portugal, the story is different. For example, the economic turmoil in recent years prompted a sufficient number of locals to leave for jobs elsewhere, which also led to a reduction in population. The data sample available in FRED shows that this happened twice before in Portugal.

How this graph was created: Search for total population and the respective countries. Convert units to “Percent Change from Year Ago” (if frequency is not annual) or “Percent Change.”

Suggested by Christian Zimmermann

M2 velocity and inflation

It is quite common to see arguments that if M2 velocity (the M2/nominal GDP ratio) is low, it must be that inflation is high. While M2 velocity is currently at historical lows, inflation is clearly not high. Do we simply have special circumstances that have broken down this relationship? Is there such a relationship in the first place? Let us look at the data:

Eyeballing the graph, we see no clear relationship between these variables. There is a better alternative than line graphs to eyeball correlations, though: scatter plots. For each quarter, CPI inflation is plotted on one axis (horizontal) and M2 velocity is plotted on the other (vertical):

Not much of a relationship can be found here. If anything, there is a slight upward slope, indicating that higher M2 velocity is associated with higher inflation, although this would not be statistically significant.

How these graphs were created: Search for M2 velocity, then add CPI. Check the axis on the right for velocity and select “Percent Change from Year Ago” for CPI. This gives you the first graph. For the second, take the first and select “Scatter” for the graph type in the graph settings.

Suggested by Christian Zimmermann

Finding old inflation data

A recent FRED Blog post showed that individual products provide an incomplete understanding of overall inflation, but sometimes individual products are all you have. For example, before 1913 there was no official CPI (and the CPI wasn’t even seasonally adjusted until 1948). But specific prices from the past do exist. The NBER Macrohistory Database gathers a variety of historical sources, including newspapers, to create data series on prices. The graph shows some of these series. Again, it becomes pretty clear pretty quickly that tracking these individual prices doesn’t allow for a well-defined picture of the evolution of the general price level. You need to compose an index with a broad base of products for that.

The NBER Macrohistory Database does have a few price indexes, including one for wholesale prices that uses the series shown in this graph and one for general prices that is cobbled together from available sources, including wage data. The quality and scope of this slice of economic history certainly don’t match the standards of the current CPI.

How this graph was created: Search for and select the NBER Macrohistory Database, select the tag “price” in the left bar, and choose the various series you want to see. It may require searching more than a screenful to find the series used in this graph.

Suggested by Christian Zimmermann

The speed of Internet adoption

FRED recently added Internet usage data from the World Bank. The Internet was initially available only to the richest households who could afford both a computer and the connection. It has democratized considerably since, although the poorest still cannot afford it. The Internet was invented in the U.S., so it’s no surprise that its use became widespread in this country before it did elsewhere. The graph shows, however, that other countries have been catching up and even overtaking the U.S. It also shows that China and India are developing rapidly. At some point in the future, the Internet will be like refrigerators and televisions: Everyone will have access to it, except those who purposefully abstain from it.

How this graph was created: Search for “Internet” and the country name to find the series and then add it to the graph.

Suggested by Christian Zimmermann


Recently Viewed Series


Subscribe to our newsletter for updates on published research, data news, and latest econ information.
Name:   Email:  
Twitter logo Google Plus logo Facebook logo YouTube logo LinkedIn logo