Job growth in past decade was in temp and contract

But it's bigger than Uber.

An Uber driver passes the Barclay's Center as he drives toward the Brooklyn Bridge, in New York, Jan. 30, 2015. New research shows that the rise in independent contractor and on-call jobs -- sometimes referred to as on-demand and gig economy work -- between 2005 and 2015 accounts for the entire net gain in jobs in that span.

If you believe the Silicon Valley sloganeers, we are in a “gig economy,” where work consists of a series of short-term jobs coordinated through a mobile app. That is both the prediction of tech executives and futurists and the great fear of labor activists.

But anyone who cares about the future of work in the United States shouldn’t focus too narrowly on the novelty of people making extra money using their mobile phones. There’s a bigger shift underway. That’s a key implication of new research that indicates the proportion of U.S. workers who don’t have traditional jobs — who instead work as independent contractors, through temporary services or on-call — has soared in the past decade. They account for vastly more U.S. workers than the likes of Uber alone.


Most remarkably, the number of Americans using these alternate work arrangements rose 9.4 million from 2005 to 2015. That was greater than the rise in overall employment, meaning there was a small net decline in the number of workers with conventional jobs.

That, in turn, raises still bigger questions about how employers have succeeded at shifting much the burden of providing social insurance onto workers and what technological and economic forces are driving the shift.

Labor economists Lawrence F. Katz of Harvard and Alan B. Krueger of Princeton found that the percentage of workers in “alternative work arrangements” — including working for temporary help agencies, as independent contractors, for contract firms or on-call — was 15.8 percent in the fall of 2015, up from 10.1 percent a decade earlier. (Only 0.5 percent of all workers did so through “online intermediaries,” and most of those appear to have been Uber drivers.)

And the shift from conventional jobs to these more distant employer-employee relationships accelerated in the past decade. By contrast, from 1995 to 2005, the proportion had edged up only slightly, to 10.1 percent from 9.3 percent. (The figures are based on a person’s main job, so someone with a full-time position who does freelance work on the side would count as a conventional employee.)


This change in behavior has profound implications on social insurance. More so than in many advanced countries, U.S. employers carry a lot of the burden of protecting their workers from the things that can go wrong in life. They frequently provide health insurance and paid medical leave for employees who become ill.

They pay for workers’ compensation insurance for people who are injured on the job, and unemployment insurance benefits for those who are laid off. They help fund their workers’ existence after retirement, at one time through pensions, now more commonly through 401(k) plans.

Perhaps most significant of all, the implicit contract between an employer and an employee is that there is a relatively high bar for firing the employee if business slumps. If the economy turns down or business slows, a contract worker is, as a rule, far more likely to be out of a job than a conventional employee.

It’s true that the Affordable Care Act has made health insurance more easily within reach for independent contractors, for example, and temporary services firms can offer retirement benefits and workers’ comp. But overall, there’s little doubt that workers in these nonconventional work arrangements carry some of the burden of protecting themselves from misfortune that employers traditionally have carried.

That makes the question of why the shift has happened particularly important.

You could imagine a world in which more workers become independent contractors voluntarily, trading the social insurance functions of traditional employers for higher pay and greater flexibility. If the period from 2005 to 2015 had been one when workers had a lot of power in the job market, that might even be plausible.


It wasn’t. The unemployment rate was above 7 percent for nearly half of the period, from the end of 2008 to late 2013. Employers had the upper hand. That suggests it’s more likely that employers were driving the shift to these alternate arrangements.

Katz and Krueger raise the possibility that something has changed beyond the weak job market of the past several years. And that’s technology.

When people working as a team need extensive experience working together, it can be tricky to contract out the work. But when there are clear, simple measurements of how successful each person is, and a company can monitor it, the employer now has flexibility.

“New technologies may allow some things to be shipped out and standardized and easily monitored,” Katz said. “Call center workers can be at home. Independent truck drivers can be monitored for the efficiency of their routes. Monitoring makes contracting more feasible.”

So Uber alone may not be a major force reshaping the nature of work. But the same technologies that made it possible could be making employers more interested in building a workforce of nonemployees. A weak job market has probably given them more ability to make it a reality.

A big question for the next decade is whether this was a one-time shift or whether it will continue in the years ahead, even with a tighter labor market. The answer may determine if the employer-provided social insurance that was a staple of the 20th-century U.S. economy will remain there in the 21st.