U.S. Economy

Why More Jobs Doesn't Mean More Growth

The sectors adding the most workers aren’t among the most productive.

Baumols disease at work.

Photographer: Leon Neal/AFP/Getty Images

How can the U.S. economy keep creating jobs and still grow so slowly? One explanation can be found in the latest employment data: The sectors adding the most workers are among the less productive.

The Labor Department’s monthly survey for May suggests that employers were still in a hiring mood. They added an estimated 138,000 jobs -- less than expected but still enough to push down the unemployment rate, which declined slightly to a 16-year low of 4.3 percent (albeit due to a drop in the number of people actively seeking work -- a requirement for being counted as unemployed). In all, nonfarm payrolls have expanded by more than 16 million jobs since the end of 2009.

Impressive as the employment growth may be, it won’t do as much as it could for the economy unless those workers start producing a lot more goods and services. So will they? To get a sense, I took 14 industry sectors and divided them into three productivity groups, based on a very rough estimate of how much their output per hour increased during the decade through 2016. 1 I then looked at job gains in each group.

The results aren’t encouraging. The high-productivity-growth group -- which included sectors such as manufacturing and information -- saw by far the smallest job gains, only 0.4 percent in the year through May. It’s still more than a million jobs away from recouping its losses in the last recession. The groups with slow and middling productivity growth -- including sectors such as education, hospitality and construction -- gained 1.8 percent and 2.6 percent, respectively. Here’s a chart showing each group’s cumulative gain (or loss) since December 2007:

Employment by Productivity-Growth Group

Cumulative percent gain or loss since December 2007

Sources: Bureau of Economic Analysis, Bureau of Labor Statistics, author's calculations

Employment is also shifting toward sectors with lower levels of output per hour, a phenomenon that should reduce measures of economy-wide productivity growth. Here’s a chart focusing on sectors grouped by level of productivity:

Employment by Productivity Group

Cumulative gain or loss since December 2007

Sources: Bureau of Economic Analysis, Bureau of Labor Statistics, author's calculations

The trends are troubling, because producing more for each hour worked is crucial to boosting wages and increasing living standards in the longer run. The shift toward lower-productivity jobs also supports a theory posited by the late economist William Baumol: that as some sectors figure out how to make do with fewer workers, overall employment will inevitably gravitate toward those that can’t or don’t (which also helps explain why costs keep growing in the latter). Manufacturing might ultimately need hardly any workers at all, but we’ll pretty much always need the same number of people to perform a live concert.

To be sure, measuring productivity is a tricky business. If the quality of entertainment, health care or houses has increased more than the economic data reflect, workers in those sectors might be a lot more productive than we recognize. Also, past gains might not be the best guide to the future: Driverless cars and trucks, for example, promise to be a game changer in the transportation sector.

Still, it’s worth considering what can be done to boost productivity in services. As economist Victoria Bateman has noted, various barriers to trade and to the free flow of labor -- such as poorly crafted immigration policies and excessive occupational licensing -- stifle the competition needed to catalyze gains. Removing such obstacles might not have an effect on all kinds of jobs, but every bit helps.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
  1. To rank the sectors, I took real annual gross output data from the Bureau of Economic Analysis and divided by an estimate of annual hours derived from Bureau of Labor Statistics data. This is far from ideal, but gives a rough idea.

    The rankings, from high to low productivity growth during the decade ended in 2016: information, durable-goods manufacturing, retail trade, mining and logging, nondurable-goods manufacturing, professional and business services, financial activities, education and health services, wholesale trade, other services, leisure and hospitality, transportation and warehousing, construction, utilities.

    And from high to low average annual productivity level: utilities, financial activities, nondurable-goods manufacturing, information, mining and logging, durable-goods manufacturing, wholesale trade, transportation and warehousing, construction, professional and business services, other services, education and health services, retail trade, leisure and hospitality.

To contact the author of this story:
Mark Whitehouse at mwhitehouse1@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net

Before it's here, it's on the Bloomberg Terminal.