Hollowing Out of the Middle Class?

We’ve heard a lot about the hollowing out of the middle class.  That is, a trend of solid job growth for middle-class Americans which turned into contraction during the Great Recession. This contraction persisted after the recession ended as more of the jobs held by the middle class moved offshore, were consolidated into fewer jobs, or were lost to productivity gains resulting from technology and innovation.

Was this hollowing out driven by the recession or is it indicative of a new norm? Chmura economists looked at this issue in 2013 and verified a clear hollowing out of the middle class, as shown in the chart below.[1] During the period from 2001 through 2011, wages were mostly stagnant, with gains exceeding inflation for only those occupations on the high end of the wage scale. Also, job growth was mainly isolated to those jobs paying the least and those paying the most (see our related post Where the Jobs Are for more).


Now, with data available through 2014 and three additional years’ distance from the end of the recession, are the jobs in the middle faring any better?

Using the same methodology as in 2013, the results demonstrate a continuation of the trends identified in the 2011 data with modest improvement. Wages grew slower than the pace of inflation for all but the 7th, 9th and 10th deciles, with deciles 2, 3, and 6 showing the slowest growth. Specifically, inflation-adjusted wages declined 3.5%, 5.0%, and 3.3% in deciles 2, 3, and 6, respectively. Similar to our findings in 2013, job growth has been the fastest in the 10th, 9th, and 1st deciles. Also, the middle class—represented by deciles 4, 6, and 7—showed negative job growth between 2004 and 2014, albeit relatively smaller declines than reported in the previous blog. 


On a more positive note, employment grew in the 5th decile, which includes occupations such as dental assistants, medical secretaries, and customer service representatives. In addition, employment grew in a number of middle class occupations between 2004 and 2014, and many are expected to continue to grow, as demonstrated in the sample selection of occupations in the table below.   

  Avg Ann
Avg Ann
Empl Growth
Interpreters and Translators 7% 4% 7
Health Technologists and Technicians, All Other 7% 2% 7
Medical Equipment Repairers 6% 3% 7
Insulation Workers, Mechanical 5% 4% 7
Occupational Therapy Aides 5% 3% 4
Medical Assistants 4% 3% 4
Industrial Machinery Mechanics 4% 2% 7
Computer User Support Specialists 4% 1% 7
Pharmacy Technicians 4% 2% 4
Physical Therapist Assistants 3% 3% 7
Medical Appliance Technicians 3% 1% 6
Machinists 1% 1% 6
Source: Chmura Economics & Analytics and JobsEQ®

Growth in high-demand occupations such as these may lead to more competitive wages for the middle-deciles that could reverse current trends. For now, the data help explain the frustration that many have felt towards slow job recovery and wage growth post-recession. Stagnant wages have not been limited to the middle; spending power declined for seven of the ten deciles. Meanwhile, the middle-wage deciles in this analysis accounted for more than 40% of all jobs where job growth has been slow, at best, for those workers over the past decade.

Research assistance for this post was provided by Johnny Constable, Claire Brunner, and Leah Deskins.

[1] This analysis was created by using over 800 occupations identified by the Bureau of Labor Statistics. We broke the occupations into ten groups based on employment and wages earned and analyzed each deciles based on job and wage growth.

Economic Impact: Jobs are being created but workers are underemployed

The unemployment rate is dropping and jobs are becoming more plentiful, but that doesn’t mean workers are in their ideal jobs.

Some people who are working are “underemployed.”

The Bureau of Labor Statistics officially defines underemployment as someone who wants to work full-time and can only find part-time work.

Arizona and California had the highest number of underemployment among states for the four quarters that ended in March, BLS data shows.

Underemployment can be measured at the state level by looking at the difference between two different jobless rates that the BLS compiles.

Arizona and California had the largest gap between the two rates, having a difference of 6.2 percentage points each. Nevada followed with 6.1 percentage point difference.

The state with the least amount of underemployment, because workers are employed part time for economic reasons, was North Dakota with a 1.8 percentage point gap.

For the U.S., the difference between the two rates that the BLS compiles was 4.4 percentage points for the four quarters that ended in March.

Virginia ranked with the 32nd gap in the nation at 3.8 percentage points.

However, BLS’s definition does not capture those who are working in an occupation below their level of qualifications. For example, it doesn't count someone with a master’s degree who is working as a retail salesperson.

There is no official measure of such underemployment by occupation.

But it can be estimated by comparing the educational skills  of residents in a region to the education achievements required by occupations employed by industries in the same region.

Looking at all 381 metropolitan statistical areas, the three MSAs with the largest surpluses of high-skilled workers are Barnstable, Mass., with a 13.2 percentage point surplus; Washington, D.C. with a 12.5 percentage point surplus; and San Francisco, with an 11.5 percentage point surplus.

Some of those are desirable areas to live that attract many high-skilled residents. Some of those are college towns where a lot of graduates choose to stay.

The three MSAs with the largest deficits of high-skilled workers are Hanford-Corcoran, Calif., with a 16.3 percentage point deficit; Hinesville, Ga., with a 14.7 percentage point deficit; and Cumberland, Md., with a 14.6 percentage point deficit.

In Virginia, the Richmond MSA is ranked 78th nationally with a 1.3 percentage point surplus of high-skilled workers, indicating that underemployment here is not as much of an issue as it is in other regions.

The Hampton Roads MSA has a 3.4 percentage point deficit of high-skilled workers (ranked 177th), indicating a less severe issue of underemployment by occupation.

Viewing underemployment across the nation shows that even as the unemployment rate continues to drop, underemployment will vary by metropolitan area.

Underemployment in the United States

Many students who graduated during the Great Recession and over the last few years were unable to find jobs for which they were trained. The so-called underemployed workers are employed in an occupation below their level of qualification. For example, a graduate with a Bachelor’s Degree in economics who is waiting tables or working at a retail store is considered underemployed. Chmura calculates a proxy for underemployment by comparing educational attainment supply and demand in a given labor market at various skill levels.

Some metropolitan statistical areas (MSAs) around the country have a higher percentage of underemployed than others.  MSAs in Massachusetts, the District of Columbia, and California top the list of regions that possess a surplus of high-skilled workers in the latest update to Chmura Economics & Analytics’ underemployment dataset.

Interactive FeatureUnderemployment is a useful supplement to other indicators of labor market health. The traditional measure of unemployment from the Bureau of Labor Statistics does not distinguish between workers who are employed in a position aligned with their skills and education. Workers who are underemployed and not necessarily contributing as much as they could to the labor market, represent potential lost productivity, wages, and tax revenue for the region.

High underemployment in a region may also be a positive measure, reflecting the desire of workers to live in a particular area (like the scenic Cape Cod waterfront of Barnstable Town, Massachusetts) and/or higher standards for occupations in certain regions (such as for computer occupations in San Francisco).

Chmura’s underemployment proxies for MSAs, along with more detailed methodology and definitions, are available on our website and at the county, MSA, and state levels within JobsEQ®.

Research assistance for this post was provided by Patrick Clapp.

Using Job Postings to Measure Employment Demand

An article in the Harvard Business Review recently touched on Why Job Postings Don’t Equal Jobs, explaining that these data should be considered unreliable when trying to estimate job demand under various circumstances. Specifically:

  • Professional-type jobs are more likely to be posted online
  • Companies often advertise the same job multiple times, and
  • For job boards that require payment to post openings, firms may post more openings when there is a discount offered, whether or not they currently need those workers.

A few additional concerns were not mentioned in that article:

  • Some jobs are posted for legal reasons, such as firms sponsoring foreign workers for permanent residence (green card). Firms have to “test” the labor market by advertising those jobs even though they have hired a foreign worker already and have no intention of hiring someone else. Most of these cases are professional jobs as well.
  • The methods used to collect and clean online postings and estimate trends over time can be problematic.

The methods used to obtain and clean job postings data are varied and are typically closely guarded. For an objective review of several providers of these data, see the Vendor Product Review:  A Consumer’s Guide to Real-time Labor Market Information. Vendors scrape and spider job boards automatically and manually, code results into anywhere from 5 to 70 data elements, and deduplicate 60 to 90 percent of job ads. Based on around 4 million job postings daily, and assuming ads are only duplicates (not triplicates, etc.), that could mean anywhere from 1.2 to 1.8 million job postings are thrown out as duplicates every day.

Methods for analyzing the postings range from keyword searches to natural language processing and text analytics, but small details in methodology can have outsized effects on what gets counted. Take, for example, the difference between searching job postings on Indeed.com for registered nurses using different keywords such as “rn,” “registered nurse,” or “registered nurses.”

Job Trends from Indeed.com

Source: Indeed.com

This simple search raises a few questions:

  • Which keyword or collection of search terms best represents job postings for registered nurses?
  • Do the keyword results change by region? 
  • In another field, how might different data providers distinguish between R, the statistical programming language, and H.R. (Human Resources) or R&D (Research & Development)?

The answers to these types of questions will likely vary by data provider and should be considered before relying on the data for analysis.

Many providers make a concerted effort to improve collection, parsing, and deduplication methods; however, significant changes in methodologies can cause additional confusion and inconsistency in job advertisement data if used in analysis over time. Changes to the deduplication methodology used by The Conference Board, for example, resulted in revisions lowering estimates by about 460,000 jobs for every month in the series. The overall curves were fairly consistent, showing similar shape and trends, but anyone relying on the actual levels for measuring or forecasting employment demand could find old estimates too high by hundreds of thousands of jobs.

Impact of revisions, HWOL data series

In summary, the use of online job postings data to glean labor market information is promising, but there are a number of concerns that suggest these data are not sufficient replacements for traditional labor market data.

Research assistance for this post was provided by Patrick Clapp.

Economic Impact: When will the Fed raise rates?

After six years of an essentially zero percent federal funds rate target, it looks like rates will begin increasing soon.

The timing of that rate increase is based on the current and future strength of the economy.

However, we get clues about when the rate increase will occur from speeches and interviews of voting members of the Federal Open Market Committee, which is the Federal Reserve’s policy-making committee.

Federal Reserve Chair Janet Yellen reaffirmed in a speech about 10 days ago that she believes it will be appropriate to raise rates this year.

“If the economy continues to improve as I expect, I think it will be appropriate at some point this year to take the initial step to raise the federal funds rate target and begin the process of normalizing monetary policy,” she said in her speech.

But she is not the only voting member.

For instance, Fed Governor Daniel Tarullo takes an opposite view of when to raise interest rates.  “In my view, it likely will not be appropriate to begin raising the federal funds rate until sometime in early 2016,” he said in a presentation May 4.

Richmond Fed President Jeffrey M. Lacker, a voting member this year of the FOMC, was quoted by Reuters last week saying, “What I’ve said is that a case might be strong in June. I still think that’s possible. But as I said . . .I haven’t made up my mind yet about June.”

The upcoming rate increase is good news.

It means that the FOMC members believe the economy is strong enough to continue growing with higher interest rates. Yet, policy makers also are concerned that a premature rate increase could dampen the recovery if it is still not strong enough.

The rate hike also is great news for savers. When the Fed raises the federal funds rate target, which is the rate that banks use when they borrow from each other on an overnight basis, banks then increase the rate they pay depositors.

The iMoneyNet money fund average, the seven-day average yield over all taxable money market funds, is currently  0.02 percent in the nation. Late in 2008, when the federal funds rate target was 0.50 percent, the money fund average was 1.22 percent.

Some analysts believe that the federal funds rate target will eventually get back to a more normal rate of 3 percent over the next two years. If historical relationships hold true, savers will see a money fund average around 3 percent as well. This would certainly help retirees who are on a fixed budget.

On the other hand, borrowers will find that it costs more to get a home mortgage or to use a credit card.

The interest rate on many loans is tied to either the prime lending rate or LIBOR. The prime rate is currently at 3.25 percent and the one-month LIBOR is 0.18 percent.

In 2008, we saw how quickly the prime and LIBOR rates fell when the Fed dropped the federal funds rate.

While the federal funds rate target stood at 3 percent in February 2008, the prime rate was 6 percent and the one-month LIBOR rate was 3.14 percent. It dramatically changed by November, when the federal funds rate target was 0.50 percent and the prime rate was 4 percent and the 1-month LIBOR rate was 1.44 percent.

Longer-term interest rates also typically rise with increases in the federal funds rate, but are more dependent on inflation expectations.

The average rate for a 30-year fixed mortgage was 3.87 percent as of Thursday, up from 3.84 percent a week earlier and matching the level at the end of 2014, mortgage lender Freddie Mac said. The average 15-year rate increased to 3.11 percent from 3.05 percent.

A forecast from Chmura Economics & Analytics expects the 30-year fixed mortgage rate to rise to 6 percent by the end of 2016.

For now, it looks like higher interest rates are still possible by year’s end.

As many Fed officials say, the exact time of liftoff is data dependent. But it’s important to track FOMC member comments because not everyone interprets the data the same way.