Getting Money Out of Politics is Simple: Public Financing
|Race||Number of candidates||Cost per candidate||Total|
|President||2||$ 1,000,000,000||$ 2,000,000,000|
|Senate||133||$ 10,000,000||$ 1,330,000,000|
|House||1740||$ 1,500,000||$ 2,610,000,000|
|Total over four years||$ 5,940,000,000|
|Total per citizen per year||$ 4.50|
Solving Tribalism with Game Theory
We need to find a more elegant solution to our tribalism. I propose we use an out of the box method from the mathematical discipline of game theory. In fact, one of my central theses is that game theory can be used to solve many intractable problems. A central premise of game theory, and you might recall it from the movie A Beautiful Mind, is that self-interest doesn’t always lead to the best outcomes. Sometimes, we have to explicitly make a commitment to cooperate, and to police it, in order to produce an outcome that is better for all of us than if we each pursued our own narrow best interest. A Beautiful Mind is about the mathematician John Nash, who invented game theory. The example from the movie is five men trying to pick up women in a bar (https://www.youtube.com/watch?v=CemLiSI5ox8).
The Prisoner’s Dilemma
The introductory case in a course on game theory is called the prisoners’ dilemma. Two prisoners are accused of a serious crime, let’s say murder, and are being questioned in separate rooms, with no means to communicate. If they both confess, if they’ll each get 10 years in prison. If neither confess, they’ll both be indicted on tax evasion and get off with only a year in prison. Clearly, they’d both be better off if they cooperated and avoided confessing. However, the FBI structures their incentives cleverly and offers them a deal. If one confesses but the other one doesn’t, the confessor gets to walk free, while the guy who tried to hold fast to the cooperative agreement will serve all 20 years. So the logic works like this; even though both sides know they’d be better off if they cooperated and neither confessed, they also know that the other guy might rat them out. If you confess, I do better off by confessing, since then I get 10 years instead of 20. If you don’t confess, then I also do better by confessing, since then I get to walk free while you serve all 20 years. Since we’re both better off by confessing no matter what the other guy does, we both confess, and we both serve 10 years like dumb criminals. That is called a Nash equilibrium, a suboptimal outcome forced on us by human tendencies to pursue our own best interest, even though we’d all be better off by cooperating.
Simple Prisoner’s Dilemma. Payoffs are relative; more positive numbers are better. Player 1’s payoffs are on the left of each box, Player 2’s payoffs are on the right of each box. You solve game theory problems simply (2×2 boxes, anyway). Since you don’t know what the other player will do, you decide which strategy you will pursue in the event of each of his strategies. If he confesses, you should also confess. But if doesn’t confess, you should still confess. Place a star next to your strategic choices in both cases. The box with two stars is the Nash equilibrium that gets chosen by default.
A good example of this is the relationship between the United States and China when it comes to climate change. The US and China would both be better off if we cooperated to phase out greenhouse gases and build a global green economy. Unfortunately, if we stop polluting but China doesn’t, they can gain short-term advantage over us, and vice versa. So instead of cooperating, human nature means that we fall into a Nash equilibrium where we both keep polluting, and we all inherit a sicker planet that ultimately may be very bad for both the US and China.
Republicans and Democrats are also locked in a prisoners’ dilemma. Both sides would be better off if we could agree on civil discourse, to compromise, to be honest and honorable, to learn from each other and work together to leverage our mutual strengths to build a better country. Each side, as much as we don’t want to admit it, does have a lot of strengths crucial to America’s greatness. For example, blue America like Silicon Valley and New York produce a lot of the technological and intellectual prowess that make America great, while red America produces the majority of our proud military and food, and both areas of America play other key roles in our chugging economy. Unfortunately, in politics, if one side behaves honorably, the other side can gain major advantage by acting dishonorably, and vice versa. So instead, both sides retreat to a Nash equilibrium where we are all worse off and acting like vile hooligans, with a discourse epitomized by Newt Gingrich, Al Sharpton, Donald Trump and the Brett Kavanaugh confirmation hearings.
To repeat a point we’ve made in earlier episodes, it is fallacious to assume that blindly pursuing our own best interest in a democracy will lead to an invisible hand of continually improving outcomes. Rather, the math of game theory shows the exact opposite happens. We get trapped in a sub-optimal Nash equilibrium, where both sides behave dishonorably, injure the other side, and tear the country’s unity apart. This can all be displayed with simple box diagrams, as above.
Now before you get too upset, it is actually quite easy to solve a prisoners’ dilemma. In plain English, the two prisoners’ can maintain the cooperative equilibrium where neither confesses as long as three conditions are met: 1) the game is not played one time, as in the original prisoners’ dilemma, but repeatedly, where you can see what the other guy did last time and gain future benefits by continually cooperating 2) both sides agree to maintain the cooperative outcome as long as the other one does, but once their partner cheats then they will henceforth lose all trust and revert to the Nash equilibrium, and 3) both sides value the future sufficiently that the benefit of defecting once does not outweigh the benefits of maintaining cooperation. That third point is the key. Both sides must value the future enough that the benefit of defecting today does not outweigh the benefits of cooperating tomorrow. What matters here is the payoffs of the game and individual preferences: if the payoffs of cooperation are high it’s very easy to keep cooperating, while if the payoffs of cheating are high it’s going to be very hard, especially if one or both players only cares about the present. In math terms, we call this the discount rate (δ), the rate at which you discount future benefits as you consider whether it’s worth it to undertake delayed gratification. Some people have very high discount rates, meaning they place great value on future happiness and are willing to delay gratification and avoid bad behavior in the present, while others have very low discount rates, meaning they only really care about the moment. With the latter type, not surprisingly, it’s very hard to maintain a cooperative agreement, since their personal preferences make it so appealing for them to cheat. Unfortunately in America, our current president has a discount rate of nearly zero, which is why he’s cheated so much throughout his life and all but broken our spirit of cooperation.
So you solve the prisoner’s dilemma as follows. Make it a repeated game, played repeatedly after 1 unit of time forever, and discount future payoffs by δ. So now round 1 payoffs look the same, but round two payoffs look like this:
Repeated Prisoner’s dilemma after 1 time unit. All payoffs are discounted by δ. The cooperative outcome where all sides are better off is in purple, while the Nash equilibrium, where sides don’t cooperate, is in red.
You can solve this simply to determine which values of delta will yield a cooperative equilibrium, and which will yield the Nash equilibrium.
So in this situation, both sides will keep cooperating and NOT confessing (they will stay in the –δ/-δ box, the lower right box) as long as both players have a discount rate greater than .1. In plain English, if each player considers $100 tomorrow worth more than $10 today, then they will each keep cooperating and won’t wind up in the bad Nash equilibrium. However, if someone discounts the future so much that they’d rather have $10 today than $100 tomorrow, they will cheat, confess, and soon both players will be in the –10δ/-10δ box.
Sounds easy right!?
But then why are we stuck in so many Nash equilibria? Because most payoffs aren’t structured quite so favorably. If we changed the payoff structure to this:
Now, solving the same equation yields
Maintaining the cooperative equilibrium now requires that each player value $100 tomorrow more than $62.50 today. That’s still possible, but harder. If the payoffs get even better for cheating, δ may approach 1. Almost nobody has a discount rate of 1.
Some of you may have heard of the famous marshmallow experiment, which shows how discount rates do in fact vary across the population. To summarize very briefly, five year old children were offered a choice: you can have one marshmallow now or two marshmallows fifteen minutes from now if you avoid eating this marshmallow. Some kids were unable to avoid eating the one marshmallow, while other kids were able to delay gratification and were rewarded with two marshmallows. Well they followed these kids into adulthood, and they found that the kids that were able to delay gratification at age five had much better life outcomes including things like income, education, poverty, etc. So discount rates are at least partially inherent.
The Tribalist’s Dilemma
But that’s just the prisoner’s dilemma. The math of the tribalist’s dilemma is still more interesting. Unlike the prisoner’s dilemma, where each side is simply maximizing their own interest, the tribalist has an additional incentive to act like a jerk: he enjoys the pain of the other side. The tribalist is willing to take a lower current and future payoff if the tribe he hates suffers more. In other words, the tribalist hates his opponent more than he loves his country. Such a person is very hard to cooperate with. Furthermore, whereas in the simple prisoners’ dilemma both sides get stuck sharing in the misery, the math of the tribalists’ dilemma works out to end even worse. Since each side is willing to suffer if it makes the other side suffer, the fighting gets worse and worse as both sides try to get the other to capitulate completely. Tribalists would rather rule a lousy society than participate in a flourishing one with people they loathe.
Here’s the payoff structure of the tribalist’s dilemma:
Tribalist’s Dilemma, Part 1. Here, because the tribalist enjoys the pain of the other side, he doesn’t mind fighting, so the negative effects on himself are offset by his enjoyment of the other’s suffering.
In this case, maintaining the cooperative equilibrium requires a fantasy, where
Here, increasing the costs of fighting, the benefits of cooperating, or decreasing the benefits of cheating would all help encourage the cooperative equilibrium by lowering the required discount rate.
Finally, there is a fourth case:
Tribalist’s Dilemma, Part 2. In this case, the pain of fighting is worse than the pain of capitulating. In that case, each side will keep trying to make the country worse until their opponent capitulates. In this HYPOTHETICAL case, the Capulets will capitulate, since submission is less miserable for them.
It is more complex to calculate the necessary discount rates to maintain a cooperative equilibrium here, since the strategies are more complex. Please paste your solutions below!
Now before we lose all hope, the math still says it’s still possible to prevail. Game theory shows us four strategies to break out of the worst Nash equilibrium of the tribalist’s dilemma, to move our society toward a cooperative cycle.
Solutions to the Tribalist’s Dilemma
- Both sides meet in secret and move simultaneously
- Change the payoffs of cooperating
- Separate truly “irreconcilable tribalists” from the rest, and appeal to the latter honorably. The irreconcilables are likely a small percentage
- Have an outside player enforce the cooperative equilibrium
Option One: both sides must move from cheating to cooperating simultaneously. If one side move first, it’s hard to keep the other side from cheating. Instead, leaders of both sides must meet secretly and emerge with an agreement, like Richard Nixon and Mao Zedong in 1972 or Tip O’Neil and Ronald Reagan in 1982.
Another strategy to solve the tribalist’s dilemma, our second one, involves changing the payoffs, especially the payoffs of cooperating. If you do the math, you find a surprising outcome; making the pain of constant fighting even more painful actually doesn’t solve the problem. Punishing Trump for being tribal is not going to make the situation better. Instead, it just increases the chance that one side will capitulate, the side for whom submission is less miserable. That’s exactly the strategy that Trump and McConnell are pursuing. On the other hand, increasing the payoffs for cooperating has enormous benefits if you want a cooperative outcome, not surprisingly. If we rewarded those on our side who help forge compromise, rather than mounting a radical primary challenge against them, we could have a much better country. Instead, we do the opposite. When Republican Jeff Flake was asked about the one-week delay he forced to get an FBI investigation of Kavanaugh, he confirmed exactly this intuition. Ronald Reagan said: “The person who agrees with me 80% of the time is a friend and an ally, not a 20% traitor.” People don’t even agree with their spouses 80% of the time; I suspect the average is much lower. It’s unreasonable to expect your political allies to agree with you more than your spouse does in a country of 330 million people, and yet, too many Americans do, and so they reduce the payoffs of cooperating to nearly zero.
Along this line, we could also just pass laws that are good for the whole country to show the benefits of cooperation. But again, that would diminish the power of those who enjoy tribalism, so they do all they can to squash good laws. That’s why McConnell acts the way he does.
Our third strategy to defeat the tribalist’s dilemma, is to differentiate the true troublemakers from the rest, and appeal to them intelligently. How would you do get through to this sizeable group? You could consider: acquiescence, anger, shaming, propaganda, rebuttal, or better solutions. Let’s immediately dismiss acquiescence, since then our opponent wins without a fight. But anger will make them defensive. Propaganda won’t work either, because then we stoop to the President’s level, and he’s better at it. It backfired with Kavanaugh. What about facts? Attempting to simply rebut his lies will confuse the patriotic persuadables; how do they know whom to believe? The neuroscience of debunking myths shows that people adhere to a false belief unless they find a more credible and comfortable belief to replace it with.
So we have to replace the lies with a more credible belief that is sometimes hard to explain quickly. Therefore, there remains only one way to peel off the sizable chunk of the other side who are voting with tribalists. persuadables—winning the debate on each issue with honorable, understandable, and actionable solutions. We haven’t really tried this yet. What would this have looked like with Kavanaugh? Democrats could have figured out whatever Flake and Collins needed or wanted…and given it to them. Collins was certainly open on the abortion issue, and Flake on checking Trump. At the end of the day, from Flake and Collins perspective, they were damned if they did and damned if they don’t; they were boxed in by tribalism, so they retreated to their tribe. I want to be clear, I would have voted differently, but that was their perspective.
So solutions are the key to winning the persuadables on the other side. That’s why this podcast is about solutions. It is the media’s job to educate the public on solutions, but right now they’re largely failing, caught up in the outrage trap. That’s why Trump stokes the outrage trap: because it keeps the game on his turf.
Is this game fair? Not even close. But as Tom Friedman counseled us, “Fairness is not on the menu.” We should not expect an existential fight against an amoral man to be fair.
The fourth strategy to maintain the cooperative equilibrium in the tribalist’s dilemma is to have a competent and well-resourced enforcement mechanism for cooperation. A perfect case study is General Petraeus’ Iraq Surge of 2007. It shows all of our four strategies. In Iraq in 2007, the civil war between Sunnis and Shiites was completely out of control. The tribal animus that America had ignored when it went in was creating hell on earth. A full blown insurgency was decimating the country. American soldiers were constantly being blown up by IEDs. Death squads on both sides were committing horrible atrocities every day; the Sunnis were beheading their enemies, while Shiites drilled holes in children’s heads. And yet, when General Petraeus showed up with a competent strategy of protecting the population, capturing or killing the worst leaders of both sides, and investing in economic development to give all Iraqis a better life and a cooperative situation, the Civil War ended within 3 years (at least until President Obama pulled the troops out in 2011 against the judgment of most military leaders). The data document how Iraq went from more than 1000 weekly attacks when Petraeus took over in early 2007 to fewer than 300 by the end of the year, and nearly zero when America pulled out.
Indeed, the Petraeus model of 2007 is the perfect model for America today, because it’s analogous yet more severe. Fortunately, nobody is being beheaded in America right now, although the hatred is almost as bad and some communities suffer way too much violence. Yet Petraeus took this tribal war, and brought about prosperous peace within three years. He employed all four strategies. First, he ensured that Sunnis and Shiites moved simultaneously, by protecting the entire population from the death squads rather than only one tribe. Second, he severely increased the payoffs for cooperation, as he promoted one successful economic project after another, and put the angry and out-of-work young people back to work. Third, he explicitly pursued only what he called the irreconcilables relentlessly (killing or capturing them), and amazingly, they turned out to be a very small percentage of the insurgency. Most of the population turned out to just be low-level insurgents choosing between bad options: fighting or hiding. This stands in contrast to Hillary Clinton’s great mistake in the 2016 election, when she assumed that many Trump voters were “deplorable and irredeemable.” Petraeus also granted amnesty to people who had done horrible things, who had American or Iraqi blood on their hands. This was hard to do, but it was necessary, because they were not in fact irreconcilable, and they had a role to play in rebuilding Iraq. They would have fought to the death otherwise. Petraeus correctly assessed that the regular folk were willing to buy in to a peaceful Iraq concentrated on economic development for all Iraqis. And fourth, he used the resources and guns of the United States to enforce this on all actors simultaneously, and he did so competently, in a way his predecessor Generals did not. Petraeus’ role as the enforcer of the agreement was indeed so crucial, that as soon as America pulled out of Iraq, and the competent enforcer of the cooperative equilibrium was gone, old wounds began festering, and eventually the downward spiral of tribalism began again, as the Shiite Prime Minister exacted revenge on the Sunnis, until the Sunnis laid down their weapons and welcomed ISIS.
So who could play the Petraeus Role in America today? As usual, I offer four options. First, a President like Teddy Roosevelt could play the honest broker, as we saw in the Episode about Teddy and John McCain. Or as we saw in the Skowronek episode, after Trump we have a once in a generation chance for a new ideological regime, and we must not squander it. Second, the media, for once in their lives, could play the role of honest broker and enforcer, rather than shamelessly fanning the flames of tribalism for ratings. Third, a moderate third political party could emerge between the Bernie Sanders socialists of the left and the Trump Republicans of the right to represent the many Americans who love their country more than their party, and actually know a little science and economics. Indeed, many Americans today aren’t so much voting for their own party as they are against the worst of the other. There isn’t a lot to love in American politics right now, but there is a lot of hate. A third party in the middle could change that and give concrete representation to those who believe in cooperation. Yes, our two-party system makes that hard right now, but solutions we’ll discuss in the season finale, such as electoral reform and campaign finance reform could change that. We’ll talk about those solutions in our season finale. And fourth, a critical mass of citizens in each party could be the enforcer if they line up behind moderate politicians like Jeff Flake and Joe Lieberman, rather than driving them out of their party.
Will this be easy? No; this is not a time for sunshine, rainbows, and easy solutions. It took overwhelming force, money, and casualties before Iraqis were willing to reconcile. And yet if Shiites can work and live with Sunnis who may have done terrible things to them, then so can Americans. I’d rather not wait until a Romeo and Juliet scale tragedy to do so. We will have to move simultaneously, using some of the strategies I mentioned earlier.
Other solutions, such as electoral reform, or ending gerrymandering or improving America’s education system or reducing wealth inequality, will also make ending the tribalism trap easier, which in turn will make the other problems easier to solve. Indeed, it would move us from the Nash equilibrium to the cooperative equilibrium and we would create a virtuous circle upwards instead of a downward spiral of tribalism.
Tune in to the season finale as we discuss those actionable first steps.
We need to move out of our downward spiral and into this virtuous circle.