It was a black Monday for media titans: Tucker Carlson split from Fox amid allegations from a former producer, Abby Grossberg, that the set of Tucker Carlson Tonight was a hostile workplace for women; Don Lemon was fired by CNN just weeks after declaring that the Republican primary candidate Nikki Haley was out of her “prime” at age 51; and Jeff Shell, the CEO of NBC Universal, was ousted because of what he characterized as an “inappropriate relationship with a woman in the company” and what Hadley Gamble, the woman in question, described as sexual harassment in a complaint filed with NBC prior to Shell’s dismissal. The housecleaning may not have been entirely prompted by matters related to women and sex, but the overall effect was nevertheless redolent of the high #MeToo era, when consequences for offenders first began to materialize.  

Nevertheless, a cynical observer might ask why, years after the advent of the #MeToo movement and the collapse of so many illustrious careers on ignoble grounds—even in this same industry—incidents like these keep revealing darkly retro workplace cultures. Is it bona fide #MeToo backlash, as commentators have worried in recent months, or a failure, as one friend worriedly confided in me on Monday, of the movement to radically transform that much at all? Neither, I would say: #MeToo is, and always has been, a power struggle; in many places throughout this industry and many others, it isn’t yet won. But episodes like these—daunting as their revelations are—nevertheless suggest that progress is still being made.

[Catharine A. MacKinnon: Where #MeToo came from, and where it’s going]

It’s worth conceiving of #MeToo as a struggle for power because this frame places appropriate emphasis on who is actually in control of workplace cultures and environments, versus a more limited view that imagines #MeToo as a collection of many different conflicts between many different individuals. Each of these instances matter—the personalities and experiences matter—but the systems of corporate hierarchy that allow high-ranking company players to victimize other employees are much more crucial to ongoing workplace toxicity than any one individual, or even all of the individuals combined. The pioneers of the #MeToo movement found their greatest success in looking beyond their immediate resources for justice—in going outside the systems that had disempowered them—to vindicate their rights.

For Harvey Weinstein’s victims, that meant taking their stories out of Hollywood and into the court of public opinion—and then the courts proper, as Weinstein was criminally prosecuted and convicted. For one of Roger Ailes’s many targets at Fox News, Gretchen Carlson, it meant escaping the forced-arbitration scheme that had kept Ailes’s egregious sexual harassment a networkwide secret by filing a lawsuit in open court, which Fox later settled. In each of these cases, the right to a better workplace was fought for and won through an exercise of greater power—typically in the courts, and often in full public view.

And those victories did come with lasting endowments. Last year, the Pew Research Center found that most Americans feel that #MeToo has made accountability for workplace sex discrimination likelier; when so much of a victim’s outcry is calculated on the likelihood of a favorable outcome, that kind of terrain-changing win is precious. And then there were the people all over the country who, inspired by the movement, organized under its auspices to take power: According to researchers at Georgetown University and the University of Oregon, “between 2017 and 2021, states introduced 2,324 #MeToo-related bills and passed 286.” There was less activity on the federal level, with the exception of the Ending Forced Arbitration of Sexual Assault and Sexual Harrassment Act and the Speak Out Act, both championed by Gretchen Carlson and signed by President Joe Biden (the former bill guarantees that a victim of sexual harassment or abuse can seek relief in court rather than through a secret arbitration process; the latter limits the use of nondisclosure agreements and nondisparagement clauses to silence victims).

[Read: When the White House is a safe space]

I was at the House Judiciary Committee’s hearing on November 16, 2021, the day that several women testified about their experiences with sexual harassment and forced arbitration in support of the then-pending Ending Forced Arbitration Act. I had the sense that something momentous was happening for American workers. An employer’s ability to trap an employee in a private court of their choosing with a mere contractual clause is—and was even more so, before the bill became law—a ready aid to exploitation. The women who testified that day all said as much, under oath and on the record. The result of the bill is that employees who are sexually harassed now have access to that much more power.

But the #MeToo tale has continued long after the initial success: No law and no regulation yet has been ambitious enough to solve the problem of rich and unaccountable men. Instead the struggle to work and to live like dignified people in a civilized society has been won by smaller shifts in power in lesser victories that are still being decided, even now.

The so-called Limit, Save, Grow Act, which House Republicans passed late on Wednesday, will never become law, because its combination of a small increase in the debt limit and a freeze on much government spending has no chance of making it through the Senate. The bill is nevertheless a statement about conservative priorities—among the most important of which, the debate over the bill suggests, is putting in place work requirements for people who get government benefits. In fact, the way Republican lawmakers speak about the issue, you’d be forgiven for thinking that work requirements were essential to revving up the U.S. economy.

As it happens, able-bodied people ages 18 to 49 and without dependents already have to meet certain work requirements in order to be eligible for Supplemental Nutrition Assistance Program payments. The new bill would, for no obviously discernible reason, raise the age threshold for those requirements to 56. More substantively, the bill would subject similarly defined Medicaid recipients to new work requirements, including rules that recipients spend 80 hours a month either working, doing community service, or participating in a designated work program (or some combination of the latter two activities).

The argument for these requirements is straightforward: Threatening people with the loss of benefits will encourage them to get off their butt and get a job, which will be good for their well-being and better for the economy. That’s why the section of the Limit, Save, Grow Act that deals with work requirements is headlined “Grow the Economy.” Advocates of these requirements frequently point to what happened after 1996, when work requirements were attached to what had been Aid to Families With Dependent Children (AFDC) benefits: Welfare rolls shrank dramatically, supposedly because recipients poured into the workforce and found gainful employment.

[Read: 20 years since welfare ‘reform’]

The logic here seems neat and clean. And work requirements also seem morally just to numerous voters. “I don’t think many people think it’s right to be paying billions of dollars to allow people to sit at home,” House Majority Leader Steve Scalise said recently, “and not work when everybody’s looking for workers.” The only problem: No good evidence exists that work requirements of the kind that are in this bill will have a noticeable impact on employment; plenty of evidence exists that what they’ll do instead is cause many thousands of people to lose their benefits despite being perfectly eligible for them.

To start with, most people on government benefits who realistically can work already do so. They may not work 80 hours every month, because they work in seasonal employment or can’t get enough hours, or for health reasons. But a study by researchers at the Brookings Institution’s Hamilton Project found that only 29 percent of SNAP and Medicaid recipients did not work over a two-year period. This should not be surprising: Few people who can work are going to choose a to subsistence existence on food stamps and Medicaid. That’s why the Congressional Budget Office’s evaluation of the new bill’s impact is that “the employment status of and hours worked by Medicaid recipients would be unchanged.” And it’s why, when Arkansas put work requirements in place for Medicaid recipients ages 30 to 49 in 2018, the program did not increase employment in the group, according to a 2020 study.

What Arkansas’s work requirements did do, though, was result in a lot of people losing their health insurance. In fact, 18,000 people fell off Medicaid in the months immediately after the requirements went into effect (and before they were halted by a federal judge), and the uninsured rate among Arkansans in that same age group rose by four percentage points. Many of those who lost their benefits were either working or were eligible for exemptions and should never have been kicked off the rolls.

That is similar to the situation today with Temporary Assistance for Needy Families, the welfare program that succeeded AFDC. Millions fewer people are on assistance rolls today than was the case in the mid-1990s—but most of that reduction has been because of a sharp decline in the percentage of eligible people receiving benefits, rather than a big drop in the actual number of people who need help.

So why do work requirements result in eligible people losing their benefits? One explanation, according to a 2019 paper from the nonprofit organization Ideas42, is that they create a lot of new paperwork for benefit recipients: If you have a job, you have to document it. If you don’t, you have to document your hours of community service or time in a work program, or, in some cases, your history of looking for a job. Then you have to submit all of the documentation in a form that the government considers acceptable. In Arkansas, Medicaid recipients initially had to navigate a complicated and confusing website to enter all of this information, which meant you needed both internet access and solid web skills. You also needed to complete your submission by the fifth day of each month, or you lost your health insurance. Oh, and the website shut down every night from 9 p.m. to 7 a.m.

[Read: Why states want certain Americans to work for Medicaid]

Work-requirement advocates are fond of quoting 2 Thessalonians: “He who will not work shall not eat.” In practice, that amounts to “He who is bad at documenting his work history or his community service will not eat.”

No doubt, some number of able-bodied slackers are unfairly mooching off the system. But public policy is a matter of benefits and costs: Although the main benefit of work requirements is that they make gaming the system harder, the heavy cost is that they result in lots of people who genuinely need government assistance, and are legitimately eligible for it, losing their benefits. By any reasonable evaluation, that cost far outweighs the modest benefits.

For conservative Republicans skeptical of welfare to begin with and eager to find ways to cut government spending, the fact that work requirements reduce caseloads may be a desired outcome, even if it means tossing eligible people off the rolls. If, however, the goal is to reduce the number of people on welfare, work requirements are an inefficient, needlessly complex, and ultimately unjust way to do it.

The superficial appeal of work requirements is easy enough to understand. But policy makers need to acknowledge this simple reality: Work requirements just don’t work.