We Asked a Game Theorist How Democrats Should Fight Trump

It's time for some (actual) game theory.
February 3, 2017, 2:43pm

A few weeks into the Trump era, liberals are in the "really fucking on edge" stage of the grieving process. They're spending their days overwhelmed by the news of the president's executive orders, sharing conspiracy-ridden Medium posts, debating whether it's OK to punch someone if you really don't like them, donating to the ACLU, deleting Uber, and, mostly, trying to get their Democratic congresspeople to fucking FIGHT, already.

Energized by Donald Trump's anti-immigration actions, the Republican plan to repeal Obamacare, and just the general #resist vibes, left-leaning Americans have gotten increasingly vocal, taking to the streets and flooding Congress with calls to make their opposition to pretty much everything the new Republican-dominated DC is doing heard. In particular, they want congressional Democrats to signal their dissent by voting against Trump's cabinet picks and filibustering Neil Gorsuch, his Supreme Court nominee—measures that might not ultimately stop anyone from being confirmed, but would at least demonstrate that the Democrats are going to be about as obstructionist as Republicans were during the Obama administration.

But are any of these tactics actually going to make a difference? How effective can sustained protest be when it comes from a minority party that controls none of the three branches of government? Is there something Democrats should be doing instead? Maybe—to quote one of the premiere examples of an anti-Trumper going fully off the rails—it's time for some game theory.

Game theory is the study of how people or groups of people cooperate or compete to achieve whatever ends they're working toward. "Games" in this context can be actual games, like poker, but game theory has been applied to everything from biology to business. In some games, players compete; in others, they cooperate. Sometimes games are zero-sum, or situations where one side needs to lose for the other to win; others are non-zero-sum, where the players can work out mutually beneficial solutions. Politics contains a multitude of games, both ones where compromise is required and ones where only one side can emerge victorious.

To talk about how all this relates to Congress, I called up Steven Brams, a professor and game theorist at New York University. Here's how our conversation went:

VICE: When we're talking about political battles, is there any way to frame them in the language of game theory that helps us understand it? ** **Steven Brams: It seems to me that Trump is seeing all of these confrontations he's having as zero-sum games, which means when you win, I lose, so there's no possibility of agreement or compromise. That worked fine in the election, that's how he was successful in both the primaries and the general, because there are always winners and losers—but I think now that he's president, continuing this strategy of seeing every conflict as zero-sum is misplaced. I think the consequences are going to be what we've observed: His approval ratings have fallen, and he's going to be castigated by people from very different sides of the political spectrum, including Republicans.

What's an example of a situation Trump seems to think of as zero-sum but is actually a non-zero-sum game?
I think almost all international relations are non-zero-sum. Take for example an arms race—that's a non-zero-sum game in which both players can lose by spending disproportionate amounts on arms and being comparatively no better off. Whereas as a win-win solution would be reaching some kind of agreement that stabilizes the situation and puts limits on arms.

You can use serious game theory to analyze such situations. The prisoner's dilemma is a famous game in which there's a cooperative solution that is unstable and a noncooperative solution that is stable but worse for both players [i.e. if both sides cooperate, it's best for both, but not cooperating can be better for an individual]. Chicken is another game in which there's a cooperative solution, but there's a disastrous solution if you continue on a collision course. These games are difficult to play, but they're not zero-sum, because if you reach some kind of agreement, in this case an arms agreement, then I think it's pretty clear that both sides are doing better.

If you're playing a non-zero-sum game against someone who thinks that it is in fact a zero-sum game, what can you do? Just convince them that they're wrong?
That's part of the story, yes. But when one player has a zero-sum point of view and the other player does not, that defines a new game. Now you're looking at payoffs that the players think very differently about. That might have some other kind of solution.

**The Republicans seem to have really adopted a zero-sum game strategy for the Supreme Court seat left vacant by Antonin Scalia, blocking Barack Obama from filling it—and that worked out. Should Democrats reciprocate by filibustering Trump's nominee, even if that ultimately turns out just to be a gesture? Does that make sense from a game theory perspective? ** One's rational choice depends on one's goals. If the Democrats think the benefits of a fight outweigh the costs, yes. But the Republicans must consider that if they choose the nuclear option [changing the Senate rules to end filibusters on Supreme Court nominees] now, then the Democrats will be in a stronger position, when they come to power, to use the same strategy against them.

This resembles a game of chicken, in which both sides can lose—at least in the long run—by refusing to compromise, making the Senate a very unpleasant place to do business.

What about Trump's cabinet nominees? A lot of people on the left look at the past six years of congressional Republicans aggressively obstructing Obama whenever they could, and they figure Democrats should return the favor. What can game theory tell us about that kind of situation, where there's an escalating level of hostility?
Well, that would be a kind of a standoff, and nothing would get done. And that's what happened in the Obama administration. There will be no way out of it unless somebody wants to make some compromises. I think it's not just a matter of, "Look what the Republicans did to us over the past six years." I think it's a matter of Trump's taking a very zero-sum view of _all _situations. I think that's going to hurt him.

Do you have any tips for someone who is playing a game against someone who is irrational?
I don't think anybody is "irrational." We have to be careful about how we define rationality. Rationality, in its simplest form, means you choose the best and most effective means to an end. An extreme example would be if you decide to commit suicide, and you succeed in doing so, you're rational. So in a way, if Trump wants to commit political suicide—which I don't think he thinks he's doing, but he may be doing in effect—then he's being perfectly rational in creating all these confrontations. So his goals are different from others' goals, but he's being rational with respect to his goals. I think one of his goals is to continually be controversial and be in newspapers and other media all the time. These are not normally the goals of politicians.

There's a stereotype of Democratic politicians as being more willing to compromise—some would say more willing to concede—than Republicans.
I think that's Trump's view, that they're the weaker side and they'll eventually concede. That's how he's dealt with his competitors in business—he'll hold out longer and they'll cave in, which is The Art of the Deal. But I don't think it's quite that simple. I think he's going to see that.

Are Republicans better than Democrats at game theory? Is that why they keep winning?
Oh, now, I think that's a misperception. I think they have been very good at stalling things, but Obama did get some legislation through, including Obamacare, and I don't think it's going to collapse immediately. A Republican won the last election, but Obama was quite successful in his bid and did get as much accomplished as one could expect given a recalcitrant Congress.

And look, it was a very close election, and Hillary won [the popular vote] by almost 3 million votes.

Do you think that people outside of Congress can alter the rules of Congress? I'm thinking of people calling their representatives and so on—can they change outcomes?
I think they may. I think protests have a mixed record. Some protests ended the Vietnam War. I would say the same thing could happen here, but it's not going to be easy. I think the Democrats are going to have to organize well, not just in the street but in Congress and elsewhere to put pressure on elected officials.

In your capacity as a game theorist, what advice would you give Democratic politicians? Or Republican politicians, but it seems like the Democrats need the most help.
I would say, consider major changes in the electoral system so that we don't have continual confrontation in Congress and elsewhere, and so there's likely to be more compromise, more discussion across the aisle.

Get some more non-zero-sumness in there.
Yes, create new rules that induce non-zero-sum thinking.

This interview has been edited and condensed for clarity.

Follow Harry Cheadle on Twitter.