There are many who think the structure of work is rapidly changing. The “gig economy,” they argue, is replacing the old economy; robots and artificial intelligence are displacing workers faster than ever. “The Future of Work” has become an angst-prone staple at conferences across the globe.

It’s always possible that this time is different, and, in fact, some notable differences are in play. But after reading a new paper by the Economic Policy Institute’s Larry Mishel on the economics of labor at Uber, I’m convinced that the new thinking needs a rethink. The challenges facing the future of work are, in fact, steep, but they’re largely the same ones facing the present (and the past): Too many workers have too little bargaining clout, and they are exploited in jobs that lack a robust set of labor protections. In other words, this is neither a technology problem nor an evolution-of-work problem. It’s a political problem — one that could be solved by supportive policy.

To be clear, the structure of work is changing, but that has always been the case. It does so glacially — Mishel shows that gig, or on-demand, workers represent a tiny share of the workforce — and it’s not changing any faster now than it has in the past. Neither is the pace at which technology is displacing workers speeding up; if anything, based on disappointing productivity growth, it surprisingly appears to have slowed.

Mishel’s paper zooms in on a specific question: how much are Uber drivers paid? But in deriving the answer, his work yields important insights into the problem and solution of low-wage work in the United States.

The punchline, paycheck-wise, is that Uber workers make, on average, $10.87 an hour in compensation, or “the income drivers get after deducting Uber fees and vehicle expenses and the mandatory extra Social Security/Medicare taxes that self-employed drivers must pay.” (The fact that Uber drivers, along with most on-demand workers, are self-employed contractors is an important part of the problem.) Their wage, net of any benefits, is $9.21 an hour, which puts them at the 10th percentile of the wage distribution and below the minimum wage in many of the places they drive.

Moreover, Mishel points out that Uber drivers usually work part time (17 hours per week, on average) and part year (an average of three months per year). “This means that an Uber driver provides roughly 12.5 percent as much ’employment,’ or total hours of work in a year, as a full-time, full-year worker.” Thus, as a share of full-time equivalent work (e.g., where two half-timers are counted as one full-timer), Uber is 0.07 percent (seven-one-hundredths of a percent) and the gig economy as a whole comprises 0.1 percent. Even if he’s off by a factor of five, the gig economy is a tiny share of the whole.

This is all valuable perspective, especially if you’re walking around thinking these arrangements account for significant shares of our workforce. But it doesn’t mean these new services are unimportant or unwelcomed (I’m an avid Lyft user). While relatively unregulated firms such as Uber unfairly compete with heavily regulated taxis, most analysts, like most riders, will tell you they are introducing much-needed competition into that sector.

It does mean that the gig economy is far too small to be having much effect on the structure or nature of work or pay in the United States. In fact, on-demand workers typically have other jobs and use gig work to supplement their income.

They, like the many more millions of regular workers, need much stronger labor market protections. Fortunately, another EPI economist, Heidi Shierholz, laid out that agenda in some detail in a recent piece for the Brookings Institution’s Hamilton Project.

First, being a regular “W-2 employee” should be the default status. There are too many workers, including in the on-demand space, misclassified as contractors, which automatically pulls them out from under existing protections (arguments that this would limit the flexibility of their hours are wrong). The real value of the minimum-wage and the overtime-salary threshold must rise. We need fair scheduling laws, paycheck transparency laws, and laws that limit the use of noncompete practices (like the noncompete rules at Jimmy John’s sandwich shops, which they were shamed into dropping).

The use of mandatory arbitration, in which, as a condition of employment, workers sign away their right to pursue workplace disputes (e.g., wage theft, racial or gender discrimination) on a collective or class basis, and which is far more common than you think, must be stopped. Immigrant workers should have full labor rights. And all of the above must be enforced by an activist Labor Department that is unequivocally on the side of workers.

I’m sure that gig work will grow from its tiny share, which underscores the problem of the increasing arm’s length between workers and their employers. Thus, all these protections must be extended to them, even when they are legitimate contractors.

This fissuring of the workplace pushes back on the “uber” solution to weak bargaining power: more collective bargaining. We cannot give up on some form of unionization, as that is an obvious and potent antidote to both the fissured workplace and the much larger share of low- and mid-wage employees without union representation.

The future of work will be what we, not technology, make of it. Technology and work arrangements have been and always will be changing, usually in ways we can’t foresee. What we can do is ensure that workers with inherently low bargaining power have the protections they need to significantly improve the quality of their jobs, whatever they are. Granted, that’s an old, unsexy problem, one for which there’s no app. But it’s real and pressing nevertheless.