A Union Contract for Two-Thirds of the Country Already Exists. America Just Stopped Signing It.

Tripling union membership in the United States would lift the median worker’s pay by 14.5%, transfer roughly $1.2tn annually from corporate revenues to household budgets, narrow racial wage gaps and reverse one-third of the rise in inequality since 1979. So reports the Economic Policy Institute — the figures were summarised this week, drawn from a new report by the institute. Union density in the US was once more than three times what it is today. We used to have what Denmark, Norway and Sweden still have. We had it for a generation after the war. Then, by deliberate policy choice, we stopped.

This is not a story about nostalgia. It is a story about the difference between a country that lets its workers bargain and a country that doesn’t — and what that difference does to a paycheque.

The mechanics are not exotic. A union is a thousand people negotiating the price of their work together instead of one at a time. Farmers do the same thing with grain elevators; no one calls the farmers socialists. EPI’s report puts the wage premium from union membership somewhere between 15% and 20%, and notes that even that is probably an underestimate, because shrinking union density drags the average down. Collective bargaining agreements also raise wages across the board for non-union workers in the same industry — a fact the EPI modelers call a “spillover,” which is a polite word for a neighbour getting a raise because the people down the hall organised one. Strip out a century of Cold War noise and a union is as American as a grain elevator. We just agreed to call the elevator patriotic and the union foreign.

So what changed? A short answer: organised business decided it would rather not have the conversation. Density was above 30% in the 1950s and is now under 10% — roughly where it sat in 2025, with the Bureau of Labor Statistics reporting the same number the EPI researchers cite. The decline tracks a sequence of specific political choices. The Taft-Hartley Act of 1947 opened the door to state “right-to-work” laws, which let workers in a unionised workplace skip paying dues while still collecting the negotiated wage — a free-rider mechanism that starves the organisation of resources while leaving it legally obliged to represent everyone in the bargaining unit. Then came the Reagan administration’s firing of more than 11,000 striking air-traffic controllers in 1981, the symbolic turning point that told corporate America the federal government would now break strikes rather than mediate them. Add in successive waves of state-level public-sector bargaining restrictions, an aggressive private-sector union-avoidance industry, and the slow grinding loss of manufacturing work that once anchored union density in whole regions. None of it was the weather. All of it was policy. Former secretary of labor Robert Reich argues in the foreword that by making it harder and harder for workers to organise and bargain collectively, the wealthy have captured an ever-larger share of income and wealth, hollowing out the American middle class in the process. The richest 0.1% now own more than five times the combined wealth of the entire bottom half of the country. That is not a coincidence; it is a balance sheet.

This is where the receipt stops being historical and starts being current. Last week’s BLS jobs report had quits softening and wage growth running below inflation in much of the private sector — exactly the pattern you get when workers have no leverage to bid up their own pay, as a recent picture of the cooling labour market makes clear. Productivity in the United States has risen roughly 2.7 times faster than typical compensation since 1979. Somebody has been catching the difference, and as billionaires’ combined wealth jumped over 31% in the past reporting year while median household budgets stayed flat, we know who. This is what an economy looks like when one side of every transaction has been organised for a generation and the other side has not. The two stories — the long arc of union decline and the recent surge in concentrated wealth at the top — are not adjacent. They are the same story.

So what gets built on the ground where that used to stand. The EPI report names specific, ordinary policy moves: passing the Protecting the Right to Organize Act to strengthen collective-bargaining rights, the Public Service Freedom to Negotiate Act to extend those rights to public-sector workers, automatic first-contract arbitration for newly unionised workplaces, and — a quieter, sharper idea — mandatory collective bargaining at any company where the CEO-to-worker pay ratio exceeds 100 to 1. That last one is worth a second read: if your chief executive makes more than a hundred times what your average worker takes home, the law presumes you have a labour-relations problem serious enough to require a formal negotiation table. Revoking state right-to-work laws and public-sector bargaining restrictions alone, the report estimates, would lift US density from 9.9% to 14.4%. None of this is rocket science. None of it requires a one-party state, a five-year plan, or a single firm nationalised. Every item on the list is a law a normal democratic legislature passes, the way it has been passed in every Nordic country since the 1930s.

The deeper question is the institutional one, and it is the one American social democrats owe the reader honestly. The reason Denmark got to flexicurity and sectoral bargaining is not that Danes are taller and better-looking; it is that they spent a century building the machinery under the policy — high union density, organised employers willing to bargain as a bloc, the Ghent system that ties union membership to actual material benefits like unemployment insurance, and a competent state capable of running universal programmes without scandal. America has some of those pieces and not others. We have the rural electric co-operatives that wired the country when private utilities would not. We have 145 million credit-union members. We have a profitable state-owned bank in North Dakota that has paid its way since 1919 and that nobody has ever called Bolshevik. We do not have the dense, federated union infrastructure that makes a sectoral bargain enforceable, and we are not going to grow it by passing a single bill in a single Congress. Policy is the visible tip. Institutions are the iceberg, and the iceberg takes longer to build than a news cycle. Anyone who tells you a child allowance, sectoral wage boards, automatic IRA-style union enrolment and a revived PRO Act get you all the way to Denmark by next quarter is selling something. Anyone who tells you that therefore none of it is worth doing is selling something worse.

The honest answer is the one EPI sketches in the numbers and that American labour organisers have been saying in plain words for forty years: union density is the single most powerful lever working people have to claim a share of the value they create, and it has been deliberately weakened for a generation by people who knew exactly what they were doing. Tripling it would not solve inequality by itself. It would solve about a third of it — $1.2tn annually in workers’ pockets, nearly $270,000 over a 35-year career — and it would put a floor under the next two-thirds that other policies could build on. The country that used to have this is the same country that could have it again. The decision, as ever, is the easy part. The institution-building is the work.