How Uber's 'Invisible' Workforce Could Affect Your Taxes

Gig workers are hard to count, distorting our view of the US economy.
Image may contain Building Office Building Plant and Grass
The Federal Reserve, Washington, D.C.ANDREW CABALLERO-REYNOLDS/Getty Images

The “gig economy” is hardly new, but there’s still a yawning gap between the attention it receives and our understanding of how it is---or isn’t---altering the nature of work in America. It may be a Bay Area joke that everyone is either working in the valley or for Task Rabbit, and Uber may be the world’s most valuable startup, but there may be dozens of Apple executives who are personally worth more than Ikea paid to acquire TaskRabbit. The gig economy may be disruptive, but it is still in its infancy.

We know startlingly little about the interplay between online-facilitated gigs (Uber, TaskRabbit, GrubHub, Airbnb hosting) and traditional employment. That’s because we analyze employment in the US using 20th century definitions, and data that government agencies struggle to understand on limited budgets. The result is that we know something has changed, but we don’t know how or how much.

It’s more than just a statistical question. Our understanding of employment affects government spending, Federal Reserve policy around interest rates, and public sentiment about whether we are doing well or badly. One of the Fed’s mandates is to design monetary policy to maximize employment.

Thursday, President Trump nominated Jerome Powell to be the next Fed chair. Confusion about the gig economy will complicate the central questions of his term, including the appropriate level of interest rates, how fast wages should be increasing, or the possible effects of tax reform. Current Fed Chair Janet Yellen recently confessed that the economy isn’t working as the Fed anticipated: inflation is lower, and wage growth slower, than projected, given how low unemployment has fallen. The gaps in our understanding of the morphing nature of work are not academic. They are compromising our collective ability to understand and address the economic challenges of our day.

Friday, the government pegged the unemployment rate at 4.1 percent, the lowest in 17 years. It is even lower for some groups, such as those with a college degree, and higher for others, such as teens without a high-school diploma. But every group is doing far better than in the years following the 2008-2009 financial crisis.

Given such robust numbers, you’d think that the public mood would be bright; it is not. Every major poll shows significant majorities believe the country is headed in the wrong direction, even as many say they are feeling better about the economy. One recent poll found just over half of Americans said the economy is decent while nearly two-thirds said the country is dangerously off-course.

How do we account for such statistics? You could say that pocketbook concerns are now less acute than political ones. Or you could point to slow wage growth, which has left many Americans feeling that they are falling behind. You could also highlight that the share of American adults who say they are working or looking for a job---less than 63 percent---remains far lower than in previous decades.

To add to the complexity, our understanding of the workforce remains grounded in the 20th century. We view the labor pool as workers employed by a company on a payroll. And indeed, the vast majority of American workers are indeed employed by an employer paying a salary. There is a category for self-employed, but for budget and methodology reasons, the Census Bureau (which provides some material for the Bureau of Labor Statistics employment reports) has had a difficult time accurately measuring what it calls “alternative or contingent employment arrangements.”

Read through the pages of data released with each monthly employment report, and you won’t find any category for “gig” work. (You might find a report from 2005 on alternative and contingent employment.) As the statisticians at the BLS will remind us, gig work pre-dated smartphones; actors, construction workers, musicians, caterers have long gone from job to job without full-time employment status. And those statisticians have always had categories for self-employed.

But today’s gig economy workers are more likely to be “1099” workers with incomes from disparate sources than self-employed “Schedule C” workers. The latter are somewhat easier to track and categorize. Gig workers may never even consider themselves “self-employed,” which means that they may still be largely invisible.

Take the Uber driver who says (as many have to me) that they lost their job X months or years ago and are driving as a way to get themselves through until a more permanent job materializes. Statistically, those drivers are a conundrum. They’re working, many of them full-time or more. But they may not consider themselves “employed” when answering a government survey. And if they don’t say they’re actively looking for work, they’re not “unemployed,” either. Instead, they’re not part of the labor force, which might help explain why labor force participation is so low. And yet, the Uber driver without a “job” is still making money, complicating the economic effects.

Numbers are hard to come by. The Census Bureau in May did a new study of alternative work arrangements, but the results have been, according to the government website, “delayed indefinitely.” That leaves the picture murky. The McKinsey Global Institute in October 2016 estimated that more than 20 percent of the workforce in the US and Europe engaged in some former of gig work, either part-time, full-time, or in conjunction with a full-time job. On the other hand, two noted economists, Alan Krueger and Lawrence Katz, in early 2016 calculated that only about half a percent of the US workforce had gig jobs through “online intermediaries” like Uber in 2015. That would translate into fewer than a million jobs. But we know that Uber reported having 300,000 US drivers in 2015, which would suggest that there are more than a million online gig workers. (In total, the economists say many millions of Americans primarily work in gigs.) The confusion does not help policymakers.

The result is that we have one economic picture generated by official jobs reports that seems to do well capturing the dynamics of the 160-million-person workforce that we can measure, and which leaves us rather in the dark about another 100 million or so people who are doing something but who aren’t showing up in the labor force. Reduce that number by the very aged or infirm, and there still are many tens of millions of people living their lives, spending money, not starving, not homeless, and even thriving who are all but missing from our collective dashboard.

It should, therefore, come as little surprise that this thing we call “the economy” isn’t working as economic models say it should. There is little unemployment and little wage growth and not as much inflation. That isn’t supposed to happen, and it is making policymakers nervous and uncertain.

It would help if the government increased budgets to answer these questions about the invisible labor force, the gig economy and the changing nature of work, but that isn’t going to happen anytime soon. In the absence of that, we would do best to acknowledge more publicly that much is happening and changing too quickly to understand in real time. Until we get a clearer picture, none of it is going to make much sense.