Somewhere around the second interview, the hiring manager slides a number across the table that surprises you. The job is non-union. The company has never had a collective bargaining agreement in its history. And yet the offer is noticeably better than you expected. That doesn't happen by accident.
The relationship between union density and wages in non-union sectors is one of the quieter, more consequential mechanisms in labour economics. When a significant share of an industry organises, wages tend to rise across the whole sector, including for workers who never signed a union card. Economists call this the union wage spillover or the threat effect, and it operates through two distinct channels worth understanding separately. Most people, even those who follow labour policy closely, understand only one of them.
The threat that never has to be made
The first channel is competitive pressure. It works almost like a silent auction running in the background of every HR budget meeting. Suppose manufacturing workers in a regional cluster are roughly 40% unionised. The unionised plants have negotiated a base wage well above the market floor, along with predictable shift premiums. The non-union plant down the road now faces a choice: match something close to those terms, or watch its most experienced workers walk toward the organised shops during the next hiring cycle. No strike, no picket line, no formal demand. Just the arithmetic of labour market competition.
Research going back to H. Gregg Lewis's work in the 1960s, later extended by economists including Richard Freeman and James Medoff, put the union wage premium for members at somewhere between 10% and 20% above comparable non-union workers, controlling for industry, occupation, and firm size. The spillover to non-union workers in the same industry is smaller but persistent, estimated across various studies at between 2% and 8%, depending on how concentrated the union presence is and how easily workers can switch employers.
Union density is the tide. Non-union wages are the boats. The boats don't need to be moored at the union docks to rise.
Efficiency wages and the signal you can't ignore
The second channel is subtler. It runs through employer psychology as much as market mechanics, and it is a dimension that receives far less attention in most public debates about organising than it deserves.
When unions are visibly active in an industry, non-union employers become acutely aware that unhappy workers have a credible organisational option nearby. The rational response is to raise wages preemptively, not out of generosity, but to reduce turnover, suppress organising interest, and retain the institutional knowledge that walks out the door every time a skilled worker quits. Economists call this the efficiency wage: paying above the clearing price to change worker behaviour in your favour. It is, in a sense, a protection racket that workers never have to run, because the threat does the work for them.
Consider two logistics coordinators, call them Maria and James, who joined competing freight companies in the same metro area in the same month. Maria's employer operates in a sector where roughly a third of firms have recognised unions. James's employer works in a sector where union density sits under 5%. Five years later, Maria's non-union salary has tracked closer to union-scale benchmarks at rival firms; James's has drifted further from any external anchor. Same credentials, same city, same starting point. The difference is the competitive gravitational field their employers were operating inside, one with friction and one without.
The mechanism breaks down when union density falls below a rough threshold, somewhere around 15% to 20% of an industry. Below that level, the threat becomes theoretical rather than operational. Non-union employers face little competitive pressure from the organised sector, and the efficiency wage logic weakens because workers have fewer credible alternatives. This is a large part of why the long secular decline in private-sector union membership across many industrialised economies has been accompanied by wage stagnation that extends well beyond the unionised minority. The two phenomena are not merely correlated. One is causing the other.
What people consistently get wrong
The common mistake is treating union wages and non-union wages as entirely separate systems, as though a union contract is a private arrangement that only touches the workers who signed it. It isn't, and the insistence on framing it that way has done genuine harm to public understanding of how labour markets actually function.
The spillover effects mean that union density is a kind of public good for workers across an industry, including those who actively oppose organising. A non-union worker in a 35%-unionised sector benefits from competitive pressure they didn't create and may not even acknowledge. Conversely, the worker in a 4%-unionised sector bears the cost of a weak threat effect, regardless of their personal views on organised labour. So here is the question worth sitting with: if you work in a sector where unions have largely disappeared, and your wage has stagnated, who exactly do you think has been bearing the cost of that absence?
This also means that debates about unionisation are never purely about union members. The wage floor for an entire occupational category shifts with the density of organising in that category. Nurses, warehouse pickers, software testers: wherever union presence is meaningful, it functions as barometric pressure on the whole labour market around it, raising the atmospheric weight on employers who would otherwise face none.
Historically, the periods of broadest wage growth in industrialised economies have coincided with peak union density. That parallel is not incidental.
The offer that surprised you in that interview room had a history behind it. It just didn't have a union label on it, which is precisely the point.