Blogging on Employment and Job Markets in US Metro Areas

Real Wage Growth – Low, Mid, & High Wage Industries

The Current Employment Statistics from the BLS provide a wealth of industry-level employment and earnings data, including nominal and real wages. Using low-, mid-, and high-wage industry tiers developed by Jed Kolko, chief economist at Indeed, the graph below shows year-on-year real wage growth since January 2010. (Technical notes: within each tier, the real wage figures are weighted by the number of employed persons for each industry; the below graph shows the year-on-year percent change of the three-month moving average of that weighted average.)

year-on-year real wages growth - low, mid, and high wage industries - 2010 through September 2017

In this graph we can see that since early 2016 low-wage industries have seen higher wage growth than mid- and high-wage industries. As of the latest datapoints, for September 2017, low-wage industries show real year-on-year wage growth of 1.8%, while that of mid- and high-wage industries are 0.8% and 0.9% respectively. While wage growth was roughly equal from mid-2014 though the end of 2015, for the 2010 through 2013 period low-wage industries real wage growth not only lagged that of high-wage industries, but was negative for almost that entire four-year period. High-wage industries saw negative real wage growth only for a brief period from early 2011 through early 2012.

The below graph shows the 12-month moving average of real wages (in effect, rolling annual wages) rebased to January 2010. We can see that low-wage industries saw real wages fall until mid-2013, and recovered to 2010 levels only very recently (in mid-2016). Mid-wage industries recovered to 2010 levels in 2015. High-wage industries, on the other hand, did not see wages dip below the 2010 level at all.

Real wages rebased to January 2010 - low, mid, and high wage industries

September Metro Area Unemployment Update

The BLS recently released September unemployment figures for metro areas. This post will focus unemployment rates among the nation’s 51 largest metro areas (those that had populations of 1 million or more as of the 2000 census). For comparison, the national unemployment rate for September was 4.1%, down from 4.8% as of September 2016.

Denver had the lowest unemployment rate at 2.2%. Denver also had the lowest unemployment a year earlier at 2.8%. Three other metro area also had unemployment rates below 3% in September.

September 2017 metro areas with unemployment rates under 3%
Compared to year earlier, eleven metro areas saw their unemployment rates drop by more than 1%. Birmingham, Alabama saw the largest decrease, to 3.1% from 5.9% a year earlier (down 2.8%). The second and third largest decreases were in Tennessee metro areas, Memphis (down 2.2%) and Nashville (down 1.9%).

September 2017 metro areas with unemployment rates decrease from a year earlier greater than 1%
Only three large metro areas saw an increase in unemployment compared to a year earlier:

September 2017 metro areas with unemployment rates higher than a year earlier
Below is a table showing the September 2016 unemployment rate, September 2017 unemployment rate, and the change in unemployment rate for all 51 large metro areas. The table can be sorted by any of its columns.

Metro AreaSept 2016Sept 2017change
Birmingham, AL5.9%3.1%-2.8%
Memphis, TN5.7%3.5%-2.2%
Nashville, TN4.2%2.3%-1.9%
Orlando, FL4.7%3.2%-1.5%
Tampa, FL4.8%3.3%-1.5%
Jacksonville, FL4.8%3.4%-1.4%
Miami, FL5.3%3.9%-1.4%
Atlanta, GA5.3%4.0%-1.3%
St. Louis, MO4.7%3.5%-1.2%
Detroit, MI5.6%4.4%-1.2%
Pittsburgh, PA5.5%4.4%-1.1%
Charlotte, NC4.7%3.8%-0.9%
Houston, TX5.7%4.8%-0.9%
Providence, RI4.7%3.8%-0.9%
Kansas City, MO4.3%3.4%-0.9%
Milwaukee, WI4.3%3.4%-0.9%
Raleigh, NC4.3%3.5%-0.8%
New Orleans, LA5.8%5.0%-0.8%
Chicago, IL5.5%4.7%-0.8%
San Antonio, TX3.9%3.2%-0.7%
Hampton Roads, VA4.7%4.1%-0.6%
Dallas-Fort Worth, TX4.0%3.4%-0.6%
Minneapolis-St. Paul, MN3.5%2.9%-0.6%
Denver, CO2.8%2.2%-0.6%
Baltimore, MD4.3%3.7%-0.6%
Phoenix, AZ4.6%4.0%-0.6%
Inland Empire, CA6.0%5.4%-0.6%
Austin, TX3.4%2.9%-0.5%
San Jose, CA3.8%3.3%-0.5%
Richmond, VA4.3%3.8%-0.5%
Portland, OR4.5%4.0%-0.5%
San Diego, CA4.6%4.1%-0.5%
Los Angeles, CA5.0%4.5%-0.5%
Philadelphia, PA5.1%4.6%-0.5%
Sacramento, CA5.0%4.5%-0.5%
Hartford, CT4.6%4.1%-0.5%
San Francisco, CA3.7%3.3%-0.4%
Columbus, OH4.2%3.8%-0.4%
Oklahoma City, OK4.3%3.9%-0.4%
Cincinnati, OH4.3%3.9%-0.4%
Louisville, KY4.1%3.7%-0.4%
Las Vegas, NV5.6%5.2%-0.4%
Indianapolis, IN3.7%3.4%-0.3%
Washington, DC3.9%3.6%-0.3%
Seattle-Tacoma, WA4.4%4.2%-0.2%
New York, NY4.9%4.7%-0.2%
Buffalo, NY5.1%4.9%-0.2%
Rochester, NY4.8%4.7%-0.1%
Salt Lake City, UT2.9%3.0%0.1%
Boston, MA3.1%3.3%0.2%
Cleveland, OH5.3%5.7%0.4%
data source: Bureau of Labor Statistics

July JOLTS Update

Yesterday’s release of July JOLTS data shows that nonfarm job openings for were up 248,000 from from a year earlier, an increase of 4%. The below table shows opening by industry for July 2016 and July 2017:

JOLTS job openings percentage by industry July 2017 and July 2016
Comparing the distribution of job openings by industry for July 2017 to July 2016 reveals some interesting changes:

  • Professional and business services, though it still accounts for the largest percentage of job openings, is down to 17.9% from 20.8% a year earlier. The number of job openings is down 10.7% from a year earlier.
  • Heath care and social assistance as a percentage of job openings is barely changed from a year earlier.
  • Leisure and hospitality jumped from 12.9% of job openings to 14.3%
  • Despite all the well-publicized woes of the retail industry, retail job openings are up slightly from a year earlier, and account for roughly the same percentage of job openings.

June 2017 Nonfarm Payroll Update

With the release of June nonfarm payroll (NFP) numbers, it is time of an update of our chart of year-on-year employment growth:

Nonfarm payroll year-on-year growth as of June 2017
June’s NFP employment was up 2.2 million from a year earlier, an increase of 1.5%. This marks the 88th month of consecutive year-on-year growth (for comparison, the 1990s expansion saw 111 months of consecutive year-on-year growth). A year ago, the June 2016 figure was an increase of 2.5 million and 1.7% from a year earlier. While employment growth continues, the pace of growth has been slowing since the current expansion’s peak in February 2015 when nonfarm payroll employment was up by 3.1 million and 2.3% from a year earlier.
NFP employment growth for the first half of 2017 was a seasonally adjusted 1.079 million, virtually unchanged from the first half of 2016 (which saw growth of 1.081 million). At the industry level, though, there are some significant changes in employment growth. See this table showing employment growth in thousands and the percent contribution to overall employment growth by industry for the first half of both 2016 and 2017:

nonfarm payroll employment contribution by industry fist half of 2017 vs first half of 2016
Here are some industry-level observations:

  • For both the first half of 2016 and 2017, the bulk of employment growth was in Professional and business services, education and health services, Leisure and hospitality: 64.4% for 2017H1 and 63% for 2016H1.
  • Mining and logging, which is actually mostly related to oil and gas, went from losing jobs in 2016H1 to adding jobs in 2017H1.
  • Construction’s contribution to job growth in 2017H1 was almost double that of 2016H1.
  • Manufacturing went from losing jobs in 2016H1 to adding jobs in 2017H1. The gains were concentrated in these subsectors: primary metals, fabricated metal products, machinery, food manufacturing.
    Trade, transportation, and utilities contributed a mere 1.4% to employment growth in 2017H1 compared to 15.4% in 2016H1. This is due a dramatic reversal in retail employment growth: a loss of 36,000 jobs for 2017H1 compared to adding 147,000 jobs for 2016H1.
  • Information swung from contributing 1.7% to employment growth in 2016H1 to subtracting 4.0% in 2017H1. These job losses were concentrated in motion picture and sound recording, and telecommunications.
  • The contribution from local government dropped from 8.6% in 2016H1 to 5.6% in 2017H1.

Labor Share of Income, Labor Force Participation, and Wage Growth

Labor share of income and prime age labor force participation rate 1990 to 2016
The labor share of income (the blue line in the above chart) and prime age labor force participation rate (the red line) have both recently increased after long-term declines (labor share of income since the 1970s, and LFPR since the 1990s). Though it is too early to say if those long-term trends have been reversed, these are both positive developments for the labor market.

As shown in the above chart, the labor share of income increased sharply after 2014, and prime age LFPR did so a year later. The unemployment rate, which has fallen steadily from 10% in late 2009, was 5.6% in December 2014, 5.0% in December 2015, and as of the latest reading (May 2017) is 4.3%.

The rate of wage growth, shown in the below chart, has recently increased, with a CAGR (compound annual growth rate) of 2.38% for the 2014 to 2016 period compared to a CAGR of 2.05% for the 2010 to 2014 period (these are nominal growth rates and do not account for inflation).

private sector wage growth 2008 to 2016
Given these changes – increase in wage growth, LFPR, labor share of income, amidst a continued decline of the unemployment rate – 2014 represents a turning point for the labor market.

The increase in wage growth from 2.05% to 2.35%, however, obscures what is going on in the different segments of the labor market. At the industry level, we can see a marked difference in wage growth over these two periods:

wage growth by industry, 2010 to 2014 vs 2014 to 2016
For the 2010 to 2014 period higher wage industries saw higher rates of wage growth. By comparison, in the 2014 to 2016 period wage growth was much more even across industries as shown by a relatively flat slope of the trend line.

The difference is even more dramatic at the occupational level:

wage growth by industry, 2010 to 2014 vs 2014 to 2016

While higher-wage occupations saw higher rates of wage growth over the 2010 to 2014, lower-wage occupations saw higher rates of growth over 2014 to 2016. Note the difference in the trend lines, and the different scales of the Y-axis (wage growth) for the two periods. This shows that the lower-wage end of the job market is finally reaping benefits from the continuing economic recovery.

May 2017 Nonfarm Payroll Update

As of May nonfarm payroll employment was 146.7 million (this is the preliminary figure from the BLS, and most likely will be adjusted in the coming months), an increase of 2.2 million from a year earlier. In percentage terms, this is an increase of 1.5% from a year earlier. May marks the 81st consecutive month of year-on-year employment growth. As can be seen in the chart below, the rate of employment growth has been declining since a peak in early 2015:

Nonfarm payroll year-on-year growth as of May 2017
Though the rate of employment growth has been slowing for over two years now, the prime age (25 to 54) labor force participation rate (LFPR) has been increasing since September 2015 (when it hit a low of 80.6%) and is now at 81.5%. This is still significantly below the prime age LFPR of the 1990s and 2000s expansions, suggesting there is still room for employment to grow (see the below FRED chart).

Prime Age (25 to 54) labor for participation rate 1990 thru May 2017

Metro Area April 2017 Nonfarm Payroll Update

Of the nation’s 51 largest metro areas (those with populations of a million or more as of the 2010 census), this table shows the large metro areas with the top 10 year-on-year growth rates in nonfarm payroll employment as of April:

Metro areas - top 10 nonfarm payroll employment growth rates - April 2017
For comparison, the national year-on-year growth rate of nonfarm payroll employment was 1.45% as of April 2017, and was 1.85% as of April 2016. All of these metro areas have seen the rate of employment growth drop from a year earlier, though their growth rates are still well above the national average. Note that Orlando, Florida has the highest rate of employment growth, and did as of April 2016, too. Also note that all of these metro areas are in the South or the West.

Of the largest metro areas, only five saw no growth or negative growth compared to a year earlier:

Metro areas with zero or negative nonfarm payroll employment growth April 2017
New Orleans is the only large metro area to have experienced negative year-on-year employment growth in both April 2017 and April 2016. The other four metro areas to post negative year-on-year employment growth as of April 2017 saw decent employment growth a year earlier. At 1.92%, Rochester, New York was above the national employment growth rate of 1.85% as of April 2016.

Here’s a table showing for all 51 large metro areas year-on-year growth for nonfarm payroll employment for April 2017 and April 2016, as well as rankings, sorted by the April 2017 growth rate:
Metro area year-on-year nonfarm payroll employment growth April 2017

Household Debt Update

The New York Fed released its Quarterly Report on Household Debt and Credit updated through the first quarter of 2017 accompanied by this press release. The NY Fed’s report includes this graph:

Total debt balance and its composition
The press release noted that “household debt reached $12.73 trillion in the first quarter of 2017 and finally surpassed its $12.68 trillion peak reached during the recession in 2008.” And that set off lots of speculation that a financial crisis could be just around the corner.

While the 2017 Q1 dollar figure is indeed higher than the peak reached in 2008, household debt as a percentage of GDP at 66.9% is considerably below the 87.1% peak of 2008.

house hold debt as a percentage of GDP, 2003Q1 thru 2017Q1

In other words, households are not nearly as leveraged as they were in the run-up to the financial crisis.

The NY Fed’s report contains a wealth of data on household debt and credit. One thing I found interesting is how the composition of household debt has changed over the past several years. Altering the Fed’s above graph to show percentage of household debt by category gives us this:

composition of household debt, percentages
While mortgage debt remains the largest component, as a percentage of household debt it is now at 67.8%, down from over 73% in 2008, and below where it was in 2003. Student loans now account for over 10% of household debt, over three times its percentage in 2003.

table of components of household debt

Update: at the request of reader “MetroGuy”, I’ve added this chart showing household debt as a percentage of personal income, rather than as a percentage of GDP:

household debt as a percentage of personal income, 2003Q1 thru 2017Q1

The pattern is much the same, with the latest figure household debt as a percentage of personal income (77.5% as of 2017 Q1) down considerably from the peak (103.9% in 2009 Q1) and now back down to 2003 levels.

Metro Area Unemployment and Labor Force Changes

As a follow up to my recent post on March unemployment rates for large metro areas, this post will focus the change in labor force and unemployment rates. The labor force and unemployment figures shown are taken form this BLS news release. As in the previous post, this post will focus on the 51 largest metro areas, those with populations of 1 million or more as of the 2010 census.

Before looking at the metro area level data, a quick look at the national level: the national unemployment rate was 4.6% in March, down from 5.1% in March 2016. Over that twelve month period the civilian labor force increased by 1.06 million. Among the 51 largest metro areas as a group, the labor force increased by 1.13 million. This means that outside of those large metro areas, the labor force shrank by 70,000.

When looking at changes in labor force and unemployment, keep in mind that a decrease in the unemployment rate means that the percentage change in the labor force is greater than the percentage change in the number of unemployed persons. If the labor force decreased and the unemployment rate fell (as experienced by several metro areas over the past year – see below), the number of unemployed persons decreased by more than the labor force.

The below table shows the labor force change and unemployment rates for the twenty metro areas (of the 51 largest metro areas) that saw the largest decreases in unemployment rates from a year earlier:

labor force changes for large metro areas with decrease in unemployment rates March 2017 vs March 2016
The Chicago metro area, which had the largest unemployment rate decrease, saw its labor force decrease by 1.7%. St Louis and Pittsburgh also saw relatively large decreases in their labor forces (1.9% and 0.9% respectively). Only one metro area saw a labor force decrease larger than that of Chicago or St Louis – Rochester, NY (not shown in the above table, but included in the table at the end of this post) with a 3% decrease. In terms of number of persons, Chicago’s labor force decrease was the largest. The metro areas with notable labor force increases are Phoenix (3.7%), Seattle-Tacoma (2.2%), Denver (2.1%), Portland, Oregon (1.7%), and Hartford (1.7%).

Here’s a table showing the metro areas where the March unemployment rate was unchanged from or higher than a year earlier:

labor force and unemployment changes March 2017 vs March 2016
With the exception of Cleveland, these metro areas all saw significant labor force increases. Notably, although the four Texas metro areas (Austin, Dallas-Fort Worth, Houston, San Antonio) all saw their unemployment rates increase from a year earlier, as a group they account for a labor force increase of over 238,000, which is over 20% of the national increase. It is further worth nothing that the Dallas-Fort Worth metro area’s labor force increase is the largest of any metro area (though not in percentage terms) and accounts for over 12% of the national increase.

Below is a table showing labor force and employment changes over the year ending in March for all 51 large metro areas in order of the percent change in labor force. Though there are a few exceptions, we see the labor force growing in warm weather and western metro areas, and shrinking in cold weather metro areas.

Changes in labor force and unemployment rates for large metro areas March 2017 vs March 2016

Where New Grads Can Find a Roof and a Paycheck

Jed Kolko (Chief Economist at Indeed) and Ralph McLaughlin (Chief Economist at Trulia) look at job opportunities for new graduates and housing affordability:

The bad news is that the local markets with the most opportunities for young grads are among the least affordable. The good news is that some lower-cost markets also offer numerous opportunities for recent grads, though not as many as the priciest markets. While there’s no place that offers the magic combination of extensive job opportunities and easily affordable housing (and if that place existed, it probably wouldn’t stay affordable for long), we found six metros where you can spend a bit less on housing without giving up too much on the job options.

Click the graphic below to read the whole post:

Where New Grads Can Find a Roof and a Paycheck