Skip to main content

The FRED® Blog

Up in the air Air carrier capacity and use

Business or pleasure. Domestic or international. Air travel is a frequent fact of life for more and more people, and booking a flight is much easier than it used to be: A couple of clicks and you have your ticket. Unfortunately, airline logistics and operations are much more costly and complicated. Air travel technology and airport capacity have not progressed as quickly as online commerce. Crowds, delays, cancellations, and long layovers in airports are still part of the travel experience.

Deciding when or whether to fly is not always up to us, but some air travel decisions are entirely under our control. For psychological insights into the travel experience, consider the cinematic namesake of this blog post. But for some economic insights, let’s see what FRED data can tell us.

The blue line in the graph shows the number of scheduled “revenue passenger miles” each month for commercial U.S. air carriers, both domestic and international, from January 2000 through August 2017. A revenue passenger mile is equal to one paying passenger carried one mile. The red line shows the number of “available seat miles,” which is a measure of capacity. An available seat mile is equal to one available seat, occupied or not, carried one mile. The green line, which uses the right scale, shows the ratio of available seat miles to revenue passenger miles.

The graph reveals a number of interesting facts. First, albeit not too rapidly, the volume of air passenger transportation has been trending up. For businesses, online communication such as email, Skype, and Zoom have not been able to replace face-to-face meetings. For families, enhanced electronic entertainment has not replaced the full experience of visiting relatives or getting to know a new place. Second, business cycles do not appear to be a major factor of air passenger transportation: The behavior of all three series is similar in recessionary and non-recessionary periods. Although only two recessions are recorded during the sample period in the graph, one was major. Third, seasonality is a key factor. Americans travel much more during the summer than during the winter. Indeed, the volume of travel is very high in July and very low in February. Fourth, the airlines are clearly anticipating and responding to these seasonal fluctuations, as capacity (available seats) is highly synchronized with scheduled passengers, especially since 2005.

Despite this synchronization, airplanes are more crowded when there’s more travel occurring—that is, there are fewer available seats per passenger in the summer than in the winter. So, overbooked airplanes and overworked airline employees can disrupt summer travel, weather can disrupt winter travel, and all sorts of other disruptions can occur any time of year.

Last, but not least, there’s a clear trend in the seat-to-passenger ratios. The chances of having an overbooked flight have increased over time, consistent with recent and well-known unpleasant incidents.

How this graph was created: For the first two lines: Search for “passenger miles” and choose “Revenue Passenger Miles for U.S. Air Carrier Domestic and International, Scheduled Passenger Flights.” From the “Edit Graph” panel, use the “Add Line” feature to search for “seat miles” and choose “Available Seat Miles for U.S. Air Carrier Domestic and International, Scheduled Passenger Flights.” For the third line: Use the “Add Line” feature to search for and select the “Available Seat Miles…” series again. Then use the “Add data series” feature: Under “Customize data,” search for and add the “Revenue Passenger Miles…” series again; in the formula box, use the formula a/b. From the “Format” menu, select “Right” for the y-axis position for the third line, then choose the widths and colors you prefer for all the lines.

Suggested by Alexander Monge-Naranjo.

View on FRED, series used in this post: ASM, RPM

Metro area economic conditions FRED has updated its Metro Area Economic Conditions Indexes through September 2017

Last week the St. Louis Fed updated its estimates of economic growth through the third quarter of 2017 for 68 of the U.S.’s most populous metropolitan statistical areas (MSAs). Average growth across these MSAs was 2.7 percent, which is consistent with the 2.3 percent year-over-year growth in U.S. real GDP during the same period.

The data are summarized in a FRED release table called Economic Conditions Index by BEA Region (under the main category of Metro Area Economic Conditions Indexes). Growth was the fastest in the Southwest and the slowest in the Plains. Given the geographical aspects here, we can use GeoFRED to map the economic growth for these MSAs in September 2017: For example, activity declined in Detroit, Hartford, Houston, and Miami, while growth was relatively strong in San Antonio, Louisville, and San Jose.

Impact of Hurricanes on the Houston and Florida MSAs

There’s a noticeable decline in the growth rates of the MSAs affected by hurricanes Harvey and Irma last fall: The graph above shows Houston’s 0.5 percent decline in August and 1.0 percent decline in September; Miami’s 1.9 percent increase in August and 0.6 percent decline in September; and Tampa’s near 4 percent increase in August and previous months and 2.4 percent increase in September.

Economic Activity in the Eighth Federal Reserve District MSAs

Overall growth among the four largest MSAs in the Eighth Federal Reserve District continued to improve at a modest pace during the third quarter of 2017, as can be seen from the graph below. Growth in Little Rock, St. Louis, and Louisville picked up, while growth in Memphis slowed a bit. Overall third-quarter growth was fastest in Louisville at 4.3 percent, followed by Little Rock at 3.3 percent, Memphis at 2.2 percent, and St. Louis at 1.9 percent.

How these graphs where created: For the map: Go to GeoFRED and click on “Build New Map.” Under “Region Type” in the Tools menu, select “Metropolitan Statistical Area”; under “Data,” select “Economic Conditions Index.” To change legend colors, select “Colors” and chose a divergent color set that uses a bold color for both high and low values. To change the ranges, select “Edit Legend” and enter the interval values 0, 2, 4, 6.

For the first graph: Go to the Releases section on FRED and select the category “Metro Area Economic Conditions Indexes.” From this page, select “Economic Conditions Index by BEA Region.” Then, by checking the box next to their names, select the Houston-The Woordlands-Sugar Land, TX (MSA) and the Miami-Fort Lauderdale- West Palm Beach, FL (MSA) and click on “Add to Graph.” Use the slide bar below the graph to zoom in on the period of interest.

For the second graph: Again, go to the release tables page on FRED. Select “Economic Conditions Index by BEA Region.” From here, select the St. Louis, MO-IL (MSA), the Little Rock-North Little Rock-Conway, AR (MSA), the Memphis, TN-MS-AR (MSA), and the Louisville/Jefferson County, KY-IN (MSA) and click on “Add to Graph”. Then adjust the dates to 2017-01-01 to 2017-09-01. From the “Edit Graph” menu, use the “Modify frequency” option to select “Quarterly” and the “Aggregation method” option to select “Average.” Repeat this step for the other three lines. From the “Format” tab, under “Graph type,” select the option “Bar.”

Suggested by Asha Bharadwaj and Charles Gascon.


Nowcasting current activity How's the economy

Forecasting, as we all know, tries to predict the future. For FRED’s purposes, that prediction is how a statistic will evolve. Nowcasting, a variant of forecasting, looks at the current state of a statistic that hasn’t yet been released because the period of coverage is not yet over. Nowcasting is one way to examine current economic activity; another was discussed in a previous post.

GDP is a popular target for nowcasting, and FRED covers the nowcasts of several Federal Reserve Banks—with the Federal Reserve Banks of Atlanta (GDPNow) and St. Louis nowcasts shown here along with the final GDP numbers released by the Bureau of Economic Analysis. To gaze into the future, focus on the very last data point for each nowcast (Q4 2017, shown here), as this is what nowcasting is all about.

The earlier data points for the nowcasts are the last estimates before the first (early) GDP release by the BEA, which is typically revised over time to create the green line. Like the BEA’s GDP numbers, the nowcasts are revised several times per month.

We see that there are disparities between the nowcasts. While they are in principle all based on the same information, estimates can differ because of different statistical methodologies and how they are revised over time. And what about the differences between the nowcasts and the final data? The BEA obviously has the advantage of access to more raw data and more time to refine the numbers.

How this graph was created: Search for “nowcast” and all the series you want should appear. Select the relevant series and click “Add to Graph.” From the “Edit Graph” menu, use the “Add Line” option to search for and select “real GDP” (use the growth rate series). Finally, start the graph in October 2011, the first data point of GDPNow.

Suggested by Christian Zimmermann.

View on FRED, series used in this post: A191RL1Q225SBEA, GDPNOW, STLENI

Advance retail sales on FRED Getting an earlier look into current economic activity

FRED recently added a set of time series from the Census Bureau known as “advance retail sales.” This data set isn’t about making technological progress in the retail sector; rather, it’s about collecting preliminary information about retail sales statistics and releasing those statistics before they’re considered definitive. This advance information may be premliminary, but it’s also quite useful: Retail sales are a large part of the economic activity in a country, and knowing how well the sector is doing is a good proxy for other economic indicators that are released much later.

The graph shows one of the advanced series (in blue) along with the history of final releases (in red). In FRED, you can save a graph in your account and choose to have the graph automatically update to include the latest data. This way you can easily monitor how a particular indicator is doing over time. The FRED dashboard is a great tool for this.

How this graph was created: Search for “advance retail” or start with the release table linked above. Choose a series. Look in the notes for the code of the corresponding historical series. From the “Edit Graph” section, open the “Add Line” tab and use the series code. Select the last year of data.

Suggested by Christian Zimmermann.

View on FRED, series used in this post: MRTSSM44X72USS, RSAFS

New reflections for the new year Some good news and some bad news about U.S. life expectancy

For many of us, it’s almost impossible to avoid at least some self-evaluation during the holidays, as we transition from one calendar year to the next. So here’s some input and, perhaps, something new to think about at the start of the year.

Let’s look at the good news first: Over the past 50 years, life expectancy in the U.S. has increased significantly, by almost 9 years, from 69.8 in 1960 to 78.8 in 2015. (The measures here are based on “life expectancy at birth,” which is the average number of years a newborn infant would live if prevailing patterns of mortality at birth were to stay the same throughout his or her life.) The graph shows that this improvement for the U.S. (thick black line) has been relatively steady over the years. To be sure, adding 9 years to a life is a big deal. A longer lifespan allows us to enjoy more years of retirement and interact with family (including children, grandchildren, and great-grandchildren) but it’s also associated with better overall health over the lifetime.

But the graph also shows two patterns that also deserve reflection. First, the gains in life expectancy have been slowing down since the early 1980s, most notably in the past five years. Second, the U.S. was right in the middle of the pack in 1960 but is now lagging far behind all other developed countries. Indeed, in 1960, the U.S. was below the U.K. and Canada, essentially on par with France and Germany, but significantly above Japan and Italy. Fast forward to 2015 and you can see the U.S. is below all these countries by a significant margin: The Japanese and Italians are expected to outlive Americans by an average of five years, and the French are expected to outlive Americans by four years.

This comparison is among developed countries with comparable economic and geographic conditions (all are developed and all are in the Northern Hemisphere), but all these countries also have a presence in the genetic makeup of the U.S. population. So, what are the culprits that have caused the U.S. to lag behind? Food? Stress? Lack of exercise? Ingesting toxins and other risky behavior? Even if we knew the precise reasons, why would conditions be so different today?

How this graph was created: Search for “Life-Expectancy” and “United States.” From the Edit Graph” menu, select “Add Line” to add each of the other six countries, each time simply typing “Life-Expectancy” and the country name. Finally, to highlight the U.S. series, choose “Format” and select 5 for the width of the line and black for its color.

Suggested by Alexander Monge-Naranjo.


Subscribe to the FRED newsletter

Follow us

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