Tripling union membership in the US would give the median worker a 14.5% raise — $7,700 a year, nearly $270,000 over a 35-year career — and shift $1.2 trillion annually from capital to labor. That’s not a think-tank wish. It’s the arithmetic in a report released Wednesday by the Economic Policy Institute, and it answers a question that’s been hiding in plain sight for decades: where did the money go?
The standard explanation is that the economy changed. Deindustrialization gutted the manufacturing base where unions lived, and the labor force is restructuring under demographic pressure and automation. Those forces were real. They’re also convenient, because they let you treat a fifty-year decline in worker power as something that simply happened — like weather, or entropy, or the invisible hand doing its inscrutable work over a long lunch. But weather doesn’t explain why union density fell from over 30% in the 1950s to 22.2% by the early 1980s to 10% today — a decline that accelerated precisely as worker productivity was growing 2.7 times faster than worker pay. The dominant factor wasn’t the invisible hand. It was specific lawmaking, maintained at the request of the employers who benefit from the arrangement.
The EPI report walks through what that architecture looks like. Right-to-work laws force unions to represent workers who don’t pay dues, quietly starving them of resources. That’s the “freedom” in “right-to-work” — the freedom to let someone else negotiate your wages and then pocket the benefit for free. Taft-Hartley’s restrictions on secondary boycotts and solidarity action severed the connective tissue between unions across industries. A National Labor Relations Board starved of funding and staff can’t process employer unfair-labor-practice charges fast enough to matter. And an entire consulting industry — hundreds of firms, billions in revenue — treats labor organizing as a risk to be managed, deployed specifically to make sure the workers who ask for a union don’t get one. Walk me through that slowly. The workers vote for representation. The employer hires a consultant to prevent it. And the law, as currently written, lets this happen. I keep being told the American labor market is the freest on earth. The freedom is doing a lot of work in that sentence, and none of it for the workers.
Sixty-eight percent of Americans view unions favorably. More than 50 million workers would join one if they could. Ten percent actually belong to one. That gap — 58 percentage points between “want a seat” and “get one” — isn’t a market outcome. It’s the output of fifty years of deliberate policy. If you ran a restaurant where 68% of customers wanted a table and only 10% got seated, you wouldn’t call it a supply-and-demand problem. You’d ask who locked the door.
The consequences are arithmetic, not ideological. The richest 0.1% of Americans now own more than five times the combined wealth of the entire bottom half of the country — a gap that has only widened as union density collapsed. Since 1979, worker productivity has increased at a pace 2.7 times faster than pay. When workers lose the power to bargain collectively, the gains from their productivity get collected by someone else. That’s not coincidence flanking a correlation. That’s one line on a chart going up and another going down, and the question is who’s catching the difference.
Triple density back to 30%, and the median worker gets that 14.5% raise. Across the whole workforce, $1.2 trillion shifts annually. The racial wage gap narrows. Health insurance coverage increases. These changes would reverse one-third of the entire rise in inequality since 1979. The spillover effects aren’t speculative — states with higher union density already invest more in public education, expand Medicaid more readily, protect voting rights more consistently.
And here’s the part that should embarrass anyone claiming the current arrangement reflects worker preference: revoking right-to-work laws and lifting restrictions on public-sector bargaining alone would push density from 9.9% to 14.4%. That’s not a revolution. That’s undoing a few provisions lobbied into existence by the specific employers who benefit from paying you less. A third of the way home, and nobody had to nationalize a single industry.
The objection that the economy has changed — that mass manufacturing is gone, the workforce dispersed across service-sector jobs and gig work — confuses a genuine logistical challenge with an impossible one. Organizing a Starbucks is harder than organizing a GM plant. The logistics are genuinely different. But the policy response to those different logistics has been, for fifty years, approximately nothing. Other countries faced the same structural shifts and built different architecture. The question isn’t whether it can be done. It’s why we keep pretending the menu has one item on it.
Germany puts workers on corporate boards and negotiates wages across entire industries through sectoral agreements covering the vast majority of its workforce. Denmark lets employers fire workers easily but cushions the fall: generous income security, active retraining, 80% of the workforce covered by agreements that set wages across whole sectors regardless of whether any individual worker signed a card. German industry did not, in fact, collapse into the sea. Danish workers are not sitting around waiting for the market to be kind to them. I’m told the German model is incompatible with American values. Remind me what the obstacle is again — is it the 80% bargaining coverage, or the part where workers have a seat at the table?
Sectoral bargaining — the mechanism that could lift bargaining coverage from 11% to 29%, from 16.5 million covered workers to over 42 million — isn’t on the near-term horizon. The institutional base it requires is the iceberg under the policy tip: organized employers willing to negotiate as blocs, high density to sustain the system, a mechanism tying benefits to membership. You can’t airlift the tip without the ice underneath, and pretending you can is the mistake reformers keep making. I’ve made it too. The honest sentence is: this is near-impossible at the federal level, and here’s the version that could pass in three states next year.
But the EPI report’s roadmap isn’t utopian. The PRO Act would strengthen organizing rights and penalize employer retaliation. The Public Service Freedom to Negotiate Act would guarantee collective bargaining for public-sector workers. Requiring annual raises for newly unionized workers, mandating bargaining at companies where the CEO-to-worker pay ratio exceeds 100 to 1 — these are specific mechanisms with specific numbers. Revoke right-to-work, fund the labor board properly, and you’ve moved density a third of the way before you’ve rebuilt a single institution.
Liz Shuler, the AFL-CIO president, put it plainly at Wednesday’s press conference: workers walk into grocery stores asking themselves when everything got so expensive. They say it over and over, big city, small town, same question. The answer, for forty-five years, has been the same one. The institutions that used to make sure productivity gains showed up in paychecks were dismantled — not by the invisible hand, but by specific laws passed at the request of the employers who benefit from the arrangement as it stands.
A 14.5% raise for the median worker. $1.2 trillion a year. One-third of rising inequality reversed. The money is there. The desire is there. Sixty-eight percent of the country is asking for the thing. Ten percent have been allowed to have it. The only thing missing is the legal architecture to let workers go collect what they’ve earned. That’s not an economic mystery. It’s a plumbing problem. And plumbing problems have plumbing solutions — if you stop pretending the pipes don’t exist because the people who profit from the leak have spent fifty years telling you the leak is structural.