Skip to main content

The FRED® Blog

A widening disconnect? Disconnected youth in America since the Great Recession

In Sierra County, California, more than 57% of 16- to 19-year-olds were neither enrolled in school, employed, nor actively looking for a job in 2015. In Nemaha County, Kansas, that figure was less than 0.17%. Outside both the labor force and the education system, these teens are known as “disconnected youth,” and this measure has gained increasing attention in recent years: The ranks of disconnected youth rose during the Great Recession, given that the percentage of disconnected youth is closely tied to local economic and social conditions. FRED has county-level data for the percentages of 16- to 19 year-olds considered disconnected from 2009 to 2015.

The maps show the percent change in disconnected youth in 2015 and in 2010. The map for 2010 shows more red, and the map for 2015 shows more blue, which indicates the percentages increased markedly in many more counties in 2010 than 2015, likely due to the residual effects of the Great Recession. According to the Measure of America, youth disconnection tends to increase along with poverty and adult disconnection—both of which were characteristic in the high-unemployment, low-output environment of 2007-09. High ratios of disconnected youth can have devastating effects on both individuals and the overall economy: Without a work or school environment, many 16- to 19-year-olds lack mentorship, resources, and engagement. In addition, fewer workers in the labor force combined with lower tax revenues and higher strains on social assistance programs can decrease the economic potential of counties and states.

The more recent data show decreases in disconnected youth in many Western counties, especially in Idaho and Utah. However, much of Georgia and Texas saw increases in disconnected youth, some up to 23%. Overall, the picture looks better in 2015 than in 2010. The average percent change in disconnected youth in 2015 was -0.31%, compared with a rise of 0.18% in 2010. This shift from an average rise to an average decline in disconnected youth is statistically significant; however, the averages are not adjusted for county populations (values corresponding to large and small counties carried the same weight) and the slight increase does not necessarily reflect a continuing trend.

Like many economic indicators, the proportion of disconnected youth has improved slightly since the Great Recession, yet still has a long way to go.

How these maps were created: In GeoFRED, click “Build New Map.” In the Tools menu, change the region type to “County” and search for “disconnected youth” in the data menu. Change the units to “Change, Percent.” In “Edit Legend,” change the number of color classes to six and adjust accordingly. To create the second map, follow these same steps, but change the date to 2010.

Suggested by Maria Hyrc.

‘Tis the season for adjustment What difference does seasonality make in the unemployment rate over time?

If you’ve spent time graphing in FRED, you’ve run across the term “seasonally adjusted.” Seasonal adjustment uses mathematical algorithms to adjust data for predictable seasonal variations using the changes between new observations and those from a year prior. When analyzing long-term trends in data such as unemployment, which tends to increase in January and June, seasonal adjustment helps economists see the cyclical trends underlying month-to-month patterns caused by school schedules, holidays, harvests, and a plethora of other factors. For more information on seasonal adjustment, see this explanation from the Federal Reserve Bank of Dallas.

The difference between the seasonally adjusted unemployment rate and the original is presented in the top graph, where subtracting the raw data from the adjusted data shows the impact of seasonality on the data over time. Overall, seasonal adjustment made a bigger difference in the unemployment rate in the 1950s than it has in the past several decades because of the recent trend toward year-round employment. The distribution of workers across various industries has shifted significantly over the past 60 years, and different industries are subject to varying degrees of seasonal variation.

The graph below shows the overall declines of employment in two U.S. industries: Agriculture, which hires more workers during harvest seasons, employed 4.4% of workers in 1970 but only 1.2% in 2012. Likewise, manufacturing employment has declined by 16.1% over the same span of time and also employs workers based on seasonal factors and fluctuating production needs. By contrast, the Bureau of Labor Statistics reports that, between 1939 and 2015, employment in the private education and health services sector, known as the least cyclical sector in the economy, grew from 4.6% to 15.6% of all nonfarm jobs. Other sectors that have grown in employment share include retail trade and financial activities, which employ most workers year-round and thus do not contribute to seasonal variation as much as the industries that have declined.

How these graphs were created: For the first graph, search for “monthly unemployment rate U.S.” and select the seasonally adjusted series. Click “Add to Graph.” Under “Edit line 1,” add another series by searching for “monthly unemployment rate U.S.” again and selecting the not seasonally adjusted data. Click “Add.” In the formula tab, enter a-b and click “Apply.” For the second graph, search for “percent unemployment in agriculture” and select the relevant series. Click “Add Line” in the “Edit Graph” page and search for “percent unemployment in manufacturing” and select the relevant series.

Suggested by Maria Hyrc.

View on FRED, series used in this post: UNRATE, UNRATENSA, USAPEFANA, USAPEMANA

Reflections on net migration The (almost) mirror image behavior of net migration for the United States and Mexico

Some people move in, and some people move out. The difference, for each country, is its net migration. This FRED graph shows the 5-year estimates of net migration for the U.S. (solid line) and Mexico (dashed line), which include both citizens and noncitizens. Notice anything about the pattern? Net migration for the U.S. increased steadily from the mid-1960s to its peak in 1997: The largest increase was between 1992 and 1997, when it started to decrease. Net migration for Mexico is the mirror image of net migration for the U.S., albeit with some lags.

During the period 1962-2012, there were more immigrants coming into the U.S. than emigrants leaving the U.S. for other countries: i.e., positive net migration for the U.S. Moreover, during this period, U.S. net migration increased (i.e., more people immigrating to the U.S. and/or less people leaving the U.S.): It was around 959,000 people in 1962 and around 4.6 million people in 1992. In those 30 years, net migration experienced an annual increase of 5.3% per year. Between 1992 to 1997, however, the annual growth was more than twice as large, around 13%.

In Mexico, net migration was negative during 1962-2012, with more people leaving Mexico than migrating into Mexico. The evolution of Mexico’s net migration has behaved as the (almost) mirror image as that for the U.S. In Mexico, there was a steady decrease in net migration between the mid-1960s to the 2000s—the biggest drop during the 1990s—and it started recovering after 2000. However, as of 2012, net migration from Mexico was still below the value of 1962.

Most immigrants arriving in the U.S. come from Mexico, so it’s not surprising that the trends in net migration for those two countries behave similarly. A new analysis by the Pew Hispanic Center has documented a sharp decrease in net migration flow from Mexico to the U.S. As a result, net migration from Mexico has fallen to almost zero in 2011. This change has been partly driven by weaker job and housing markets in the U.S. in 2010, together with stronger border adjustment.

However, the timing of changes in net migration for Mexico and the U.S. was not exactly identical. U.S. net migration started increasing before Mexican net migration started recovering. During the 1990s, there was a sharp increase of immigrants from Asia into the U.S. Indeed, according to the Pew Research Center Projections, the share of Asians among total immigrants to the U.S. has been rising above the share of Hispanic immigrants and that is expected to increase further.

How this graph was created: Search for “Net migration for the United States,” and click on the series you want to create the first line. To add line 2 to the existing graph, click “Edit Graph” and use the “Add Line” tab to search “Net migration for Mexico.” Finally, use the “Format” tab within “Edit Graph” and select “Dash” under Line 2 line style.

Suggested by Ana Maria Santacreu.

View on FRED, series used in this post: SMPOPNETMMEX, SMPOPNETMUSA

The Great Recession’s regional effects How have different industrial compositions affected unemployment for Census regions?

Every corner of the U.S. was hard-hit by the Great Recession, but the varying makeup of local economies resulted in different effects on individual cities, states, and regions. Areas that are reliant on economic necessity—say, healthcare and waste management—may fare better during economic downturns than areas that are reliant on economic prosperity—say, tourism and construction. In 2008, the recession particularly disrupted the construction industry. According to the Bureau of Labor Statistics, the percentage of jobs lost in that industry, 19.8%, exceeded those of all other nonfarm industry supersectors.

FRED can help us see the recession’s effects on the four U.S. Census regions: Midwest, Northeast, West, and South. The annual percent change in unemployment varied most in 2008, where unemployment increased nearly 40% in the West but increased only 19.6% in the Midwest. The map below shows that some of this variation may have been the result of job losses in construction. Western states experienced declines of over 8% in the number of construction employees from 2007 to 2008. The rest of the U.S. fared better, with some Midwestern and Southern states even seeing increases in the number of construction employees.

The relatively small increase in unemployment in the Midwest may be a result of its more agriculturally based economy, which wasn’t initially hit as drastically as the shock-prone industries of tourism and construction, which are more prominent in the West. The years 2009 and 2010 saw less-significant differences in unemployment changes among regions. But data from 2011 show the continued comparative resilience of the Midwest, where the decrease in unemployment was over 11%, compared with the other regions’ 4% to 6% range. This variation in rates of recovery may also be attributable to industrial makeup: Research from the St. Louis Fed has analyzed the construction industry’s contribution to the slow overall recovery and compared it with the manufacturing industry. Research from the Minneapolis Fed notes that the construction industry took longer to recover from the recession than other industries in part because of workers’ reluctance to return to construction jobs.

How this graph was created: In FRED, search “unemployment rate census region.” Check the boxes next to the seasonally adjusted series for each of the four regions and click “Add to Graph.” Next, click “Edit Graph” and change the units to “Percent Change.” Click “Copy to all.” Modify the frequency of each line to “Annual.” Next, click the “Format” tab and change the graph type to “Bar.” Finally, adjust the dates on the graph to show from 2007-01-01 to 2011-01-01.

How this map was created: In GeoFRED, select “Build New Map.” In the tools menu, change the region type to “State” and search for “construction” in the data menu. Click “All Employees: Construction” and select “Not Seasonally Adjusted, Annual, Thousands of Persons.” Change the units to “Percent Change” and set the date as 2008.

Suggested by Maria Hyrc.

View on FRED, series used in this post: CMWRUR, CNERUR, CSOUUR, CWSTUR

Cost of living and per capita incomes in U.S. cities

A recent post introduced regional price parities (RPPs) and their applications at the state level. These RPP data are also available for metropolitan statistical areas (MSAs), which include a principal city and its surrounding area. So we can conduct a similar analysis of price levels and adjusted incomes at the metro level. Data are indexed such that the population-weighted national average is equal to 100. Of the 349 MSAs in the data, 94 fall within 5% of the national average for cost of living.

Consistent with the state-level data, the most expensive cities are heavily concentrated in the Northeast and on the West Coast. Of the 34 MSAs with a cost of living more than 5% higher than the national average, 32 are in either of those two regions. Honolulu, Hawaii (not pictured above), is the priciest MSA, with a cost of living over 24% above the national average. Inland cities are substantially cheaper, especially in the Midwest and the South. In Rome, Georgia, the least costly MSA, the cost of living is about 20 percent lower than the national average.

One implication of these regional cost of living differences is that a dollar in one city isn’t necessarily the same as a dollar in another: Average per capita personal income nationwide is about $43,996. In terms of purchasing power, the equivalent income in St. Louis, Missouri, is below $40,000 due to the relatively low cost of living. Meanwhile, in comparatively expensive New York, New York, the equivalent income is almost $54,000. In other words, as the cost of living goes up, it takes more dollars to buy the same basket of goods and services. Hence, using the RPPs to adjust per capita personal income for cost of living yields a more accurate measure of how much the average person in a given city can consume.

As seen on the next map, real (cost-of-living-adjusted) per capita personal income is more broadly dispersed geographically than the RPPs: 40 MSAs across 32 different states have a real per capita personal income more than 10% greater than the national average. Midland, Texas, has the highest income, at almost $96,000 (adjusted dollars) per person.

Given that we derive real per capita personal income by dividing nominal income by cost of living, one might expect that the most costly cities have the lowest adjusted incomes. While that is sometimes true, it’s not always the case. For example, San Jose and San Francisco, California, and Bridgeport, Connecticut, are each in the top ten for both most expensive cities and highest-earning cities. Similarly, McAllen, Texas, has the lowest real per capita personal income of any MSA, about $27,000, despite being one of the 20 cheapest cities in the country.

How these maps were created: From GeoFRED, click “Build New Map,” open the tool bar in the top left corner, and select “Choose Data.” For the first map, under “Region Type” choose “Metropolitan Statistical Area” and under “Data” search for “Regional Price Parities: All Items.” For the second map, still at the MSA level, search under “Data” for “Real Per Capita Personal Income.” For each map, use the “Edit Legend” tool bar to edit as you see fit. Note: To view the state-level data, simply change the “Region Type” to “State.”

Suggested by Andrew Spewak and Charles Gascon.



Subscribe to the FRED newsletter


Follow us

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