The United Auto Workers strike against Detroit’s Big Three—Ford, General Motors, and Stellantis—escalated into its third week on Friday. Workers at two additional plants operated by Ford and GM walked off the job, taking the number of union members striking for better pay and benefits to more than 25,000.
The dispute looks unlikely to end soon. As they try to understand where things are headed, economists, philosophers, labor experts, business professors, and a handful of boutique consulting firms see a juicy opportunity to put a 100-year-old economic theory into practice. Guys. It’s time for some game theory.
For those who learned it from memes, not in school, game theory is the “science of strategic thinking,” says Kevin Zollman, a professor of philosophy and social decision science at Carnegie Mellon University. It uses mathematics to model and predict human behavior when two or more people or parties have potential conflicting interests. Game theory has been used to plan out business negotiations, auctions, and poker strategies, and even to guide parenting decisions. (Zollman literally wrote the book on that last one.)
The current UAW strikes provide plenty for game theorists to chew on. Bargaining is a classic game theory problem, Zollman says, and a strike is a great example of a high-stakes contest between two players with different but definable interests. Adding to the intrigue—and game theoretic fun—is that this is the first time the union is moving against all three major US automakers at the same time. Even more interesting: The UAW is using escalating and targeted walkouts at specific facilities, dangling the threat of a wider strike if negotiations don’t go their way. The big game being played with the US industrial economy has four players, each with their own set of priorities and incentives.
No matter who you’re rooting for—automakers, workers, the UAW, or bystanders in the wider auto supply chain—game theory can provide a way to think through their strategies, and maybe even predict the future.
The first step is generally to figure out what’s at stake—the value or resources that would be created or lost in a future agreement. Barry Nalebuff, a businessman and professor at the Yale School of Management, likes to keep things folksy and calls that “the pie.”
Nalebuff is using the auto strike to illustrate the basics of game theory negotiation tactics to his students. He starts by breaking down what’s at stake for each player. The automakers’ meaty contribution to the pie begins with the sales they might lose because of the strike’s production shutdowns and the future cost of wage and benefit hikes. Nalebuff adds on ripple effects such as damage to their dealerships, which might run out of stock to sell, and the cost of losing customers to non-union automakers, such as Tesla and Hyundai.
The UAW and its members add to the pie the cost of striking from lost wages and withdrawals from the union’s strike fund, set up to support workers during their fight.
Everyone involved in the strike should be calculating what’s at stake to figure out what sort of offers they’d be willing to accept or make—in other words, to resolve the whole dispute. “Absent calculating the pie, I would be negotiating with my eyes closed,” Nalebuff says.
Perhaps that sounds too abstract to be much help in the boiler room atmosphere of Detroit C-suites right now. But for more than 20 years, Marc Robinson went through this kind of strategic exercise inside General Motors. Robinson, who is now an outside consultant and writes about game theory for business contexts, says he used it to advise the automaker through some 100 decisions, including supplier negotiations, policy shifts like trade deals—and, yes, labor negotiations, including the 2019 UAW strike against GM that saw workers walk out of 50 plants for over a month.
Classic game theory scenarios involve two players, like the widely taught prisoner’s dilemma, but Robinson says GM’s process began with getting the company’s experts and executives in the room to identify everyone who might be affected by the game—in this case that’s the automakers, the UAW, and even different political factions inside the unions and 2024 presidential candidates.
The next step involves the group mapping out the different “levers” each of those players can pull, meaning the four or five moves they could possibly make. For example, the UAW could escalate further and tell workers to walk out of more plants, or instead choose not to expand the strike. One automaker could decide to raise wages, or decide to stand firm. Once the forest of all the players’ levers has been mapped out, the group systematically thinks through which are most likely to get pulled.
In the end, Robinson’s groups produced a one-page document that clarified what they were willing to give up, and what they really, really wanted to avoid. The process helps an organization be brutally honest about the risks it faces, he says. “Then they can say, ‘Well, what do we do about that?’” Robinson says.
Corporate game theory can get even more complicated, to the point where computers, not whiteboards, have to be used. Gerry Sullivan’s Priiva, based in Canada, uses game-theory-based algorithms to point businesses toward decisions on things like where to roll out new products or how to approach a negotiation. In a couple hundred cases, he says, the firm’s game theory analysis has correctly predicted outcomes over 80 percent of the time.
Ford declined to comment about its strike strategy, and neither General Motors nor Stellantis responded to requests for comment. The UAW didn’t respond to WIRED’s questions about game theory either, but in texts leaked first to the The Detroit News last week, the group’s communications director said, “If we can keep them [automakers] wounded for months, they don’t know what to do … This is recurring reputations damage and operation chaos.”
In fact, some of the UAW’s moves already match a classic game theory strategy called “tit for tat,” says Art Wheaton, a professor who directs the labor studies program at Cornell University’s ILR School. On Friday the union announced it would expand its strike to more Ford and General Motors facilities—but none belonging to Stellantis. The multinational company behind Chrysler and Peugeot had made significant changes to its proposal and would be spared, UAW president Shawn Fain said.
In “tit for tat,” one player reflects another player’s moves. If Player A—Stellantis—cooperates, Player B—the UAW—cooperates too. But if the other automakers don’t play along, the union won’t play either, allowing the workers to play each company off each other. “To use a spaghetti Western analogy, it’s the good, the bad, and the ugly,” says Wheaton.
So what does game theory actually say about how this auto strike will end? Robinson, the consultant, has thought through each player, their moves, and what they want. The UAW, and particularly Fain, its new and fiery leader, will need to prove it’s put in the work to get the best deal possible. Meanwhile, because the UAW’s gradual strike strategy isn’t as painful as it could be for either the workers or the automakers, it could drag on for some time.
Robinson is ready to call it: The whole thing will wrap up between Halloween and Christmas, with more labor-friendly Ford finding a finish line first. That would make for a long autumn. Or at least, that’s the theory.
If you buy something using links in our stories, we may earn a commission. This helps support our journalism. Learn more. Please also consider subscribing to WIRED.
