Last year at this time, I wondered if Trump was tanking the presidency, intentionally failing to spur a GOP bloodbath at the midterms. Jury’s out on that, but this New York Times article on Trump at the midpoint (maybe?) has an interesting paragraph that might corroborate my theory.

As a result, a partisan war may be just what he wants. He has privately told associates that he is glad Democrats won the House in last month’s midterm elections, saying he thinks that guarantees his re-election because they will serve as a useful antagonist. That may be bravado, but history provides some support. Bill Clinton and Barack Obama, his Democratic predecessors, both endured even bigger midterm setbacks and went on to win re-election.

They did! That might just be meaningless noise, as Bush the Younger’s party did not lose big at the midterms and he still won. Maybe presidents get re-elected because presidents get re-elected. It only matters if Trump believes it will work, not whether it will work. Let’s say for now he believes it does.

If he does, he sure has put himself in quite the negative position this week. The government is shut down. He cannot get Congress to pay for the wall Mexico was going to pay for. His defense secretary quit with a face-melting letter of protest. He’s pulling out of Syria and maybe Afghanistan to the delight of Vladimir Putin, after bailing out Putin’s crony Oleg Deripaska and turning a blind eye to a Russian bomber base in Venezuela. He’s railing at his acting attorney general about all the prosecutions that are ensnaring his friends and business operations. He can’t find a permanent chief of staff. He’s scheming to fire the unfireable chairman of the Federal Reserve, whom he appointed. The market is having its worst crash since 2011. The market whose gains Trump consistently touted as his greatest accomplishment as president. Now he’s not tweeting about it so much.

Which leads me to again wonder aloud: If he tanked the presidency, is it possible for Trump to tank the economy? If so, why would he do it?

My first hunch was that it’s not possible. Presidents don’t get to decide how the economy does; if they could, they’d make it rain daisies every day. Clinton and Reagan had great economies, so they were great presidents. Carter and the Bushes had awful economies, so they were awful presidents. Maybe they were, maybe they weren’t, but it’s unfair to everyone else—especially those in the business sector—to give presidents sole credit.

Presidents have little control over the technological and demographic forces that shape the economy, and blind luck is a major part of it. Obama gets credit for a major turnaround in the economy, and he certainly deserves praise for keeping a steady hand on the tiller. But if you think he should get credit for the 8.4% job growth over his eight years, just think about how far down the economy was after the housing crash of 2008. You can’t give him credit for the “good luck” of showing up at the start of a major recession.

Trump, on the other hand, gives himself lots of credit for good news in his first year of office—good news that should be credited to Obama much more than him, if we’re doing that—and no blame for the crash that’s showed up at the end of this year. But let’s take a closer look at that. I think this is one case where we can actually credit the president with the state of the economy, even though it’s just about the last thing he wants credit for.

While economists fight over everything, one thing garners very little debate: the market hates tariffs. President Herbert Hoover agreed to the Smoot-Hawley Tariffs (anyone? anyone?) on October 28, 1929; the next day, the market crashed, contributing greatly to what would become the Great Depression. When FDR came in and said tariffs were “the road to ruin,” the market rebounded with three years averaging 10 percent growth.

This didn’t work.

George W. Bush enacted steel tariffs on March 5, 2002. The Dow Jones on that day was over 10,000, then began a slide down to 7,700 three months later. The tariffs cost hundreds of thousands of jobs and rang in a recession that dominated the mid-2000s. It took a major stimulus package from Obama to stem the bleeding.

Trump’s the first president in my lifetime I’d describe as a “tariffmonger.” He loves them as much as he loves cheeseburgers. On July 6, he announced a huge suite of tariffs against our partners. Two months later, for inexplicable reasons, investment magazines proclaimed the markets weren’t scared of tariffs any more. They aren’t writing that today. The market is in a freefall, having dropped more than 4,000 points since October.

We can thus hazard that tariffs are one of the actually functional levers a president has to affect the economy, especially in a negative way. There’s another lever he has too: acting like a crazy person. When the president behaves irrationally, at least as defined by the market, it goes south.

The “Nixon Shock” is a term that describes the market’s reaction to President Nixon unilaterally suspending the convertibility of the dollar into gold, freezing wages and prices, and imposing an import surcharge of 10 percent. At first it went great, with a slight uptick in the Dow Jones. But then as Nixon’s presidency unraveled, the oil market skyrocketed, and everyone realized unmooring the dollar was a catastrophically bad idea, the market crashed. From January 1973 to December 1974, the Dow Jones lost 45% of its value. The ensuing recession embroiled Nixon’s successor, Gerald Ford, in a cavalcade of dumb approaches (WIN buttons!) until finally he signed a tax and spending cut that stimulated the economy. It didn’t have to be that way. Had Nixon approached his presidency like a stable person, he might have survived Watergate and kept the market healthy. Not so much.

Was late-stage Nixon actually crazy? I don’t think so. But if enough people think U.S. leadership is unhinged, the market responds in a highly negative manner. And if there’s one thing that characterizes the White House over the past two months, it’s a lack of hinges. The last week in the stock market was a get-out-now reaction to the Trump presidency driving itself into a roadbank. There’s no oil shock, no tech bubble burst, no dust bowl, no housing collapse. There’s only one collapse, and it’s the executive branch.

I haven’t spent much time talking about game theory—that’s what this series is about, after all—because I wanted to justify to myself that it was possible to tank the economy from the Oval Office. I think I’ve done that, so let’s get to the second question: Why would Trump tank the economy? On that subject, game theory is quite clear that he has a reason to do so.

The market is governed by a game theory idea called the sentiment game. This game isn’t about determining that a stock or an investment strategy is a good idea to back on its merits. It’s about determining whether everyone else will think it’s a good idea to back it, then acting before they can do so. (But not too early. That gets you a cell next to Martha Stewart.)

The sentiment game is often described as a Keynesian beauty contest, after economist John Maynard Keynes described it thusly in his epic final book The General Theory of Employment, Interest and Money. Keynes hated the sentiment game. He portrayed it as a newspaper beauty contest where the goal was to pick the woman that readers would find most attractive. (Look, it was 1936. Don’t shoot the messenger.) Instead of choosing the model you found most attractive, you’d have to suss out what characteristics all the other readers would like, and bet on the woman who had those.

But if the other readers knew that, they’d bet on the woman they thought the average competitor would think the average competitor would think was the most attractive. That’s not the same as the average person; it’s the average of everyone playing the guess-the-most-attractive game. And if they knew that, they might think on an even higher level about which competitors knew which competitors were average, and guess how those … look, it’s turtles all the way down. The point is that if you keep going up and up the levels of who is playing the most strategic game, you eventually talk yourself into a Nash equilibrium that is far from the rational opinion you might’ve started with.

For example, you play a game where you have to guess what two-thirds of the number from 0–100 the average person will pick. A random person will pick a random number, probably 50. A first-level thinker will pick two-thirds of that, or 33. A second-level thinker, thinking everyone is smart enough to understand the game, picks two-thirds of that, or 22. A third-level thinker, thinking everyone is following along this line of thought, picks 14, and so on until everyone picks zero. That’s crazy-thinking, but it’s “strategic,” y’know?

If, hypothetically, you’re the president of the United States, and you think you should get credit for what happens in the economy, and you also think you’ve got the biggest brain in the world at understanding business, you might just believe you’ve figured out the sentiment game. I mean, really figured it out. Everybody tells you that when presidents enact tariffs, the market crashes. You know what the average person will do—sell like the building’s on fire—and you know what above average people (like you!) will do when that happens. You’ve got a very, very large brain here, so you get this.

But you don’t just listen to that part of the story. You listen to the FDR part of the story too. He comes in saying tariffs are bad and the market flies back up. You look at the Obama part of the story. He engineers a stimulus act, and the economy roars back. You look at the Ford part of the story. He signs a tax and spending cut bill, and the economy works its way northward.

None of those presidents get to save the economy unless the economy gets in trouble first. Deep down, Trump knows he didn’t save the economy when he came in; it was running just fine and it kept doing so into his presidency. The only way to deal with a “problem” like that is to create a crisis, then claim to solve it. Y’know, just like he did with DACA, with North Korea, with NAFTA. This is a tried and true method for Trump. It defines his presidency.

If a madman starts tossing around threats of tariffs and implements them without the approval of anyone else, the market is gonna come unglued.

Phase 2: President stuff.

Sure, lots of people will suffer. Whatev. That’s not his concern. What’s important is that everyone views him as a savior. So hey, a midterm crisis is great. For his re-election, a rip-roaring economy in, say, early 2020 is way better than a stable but unremarkable one right now. Just gotta make that happen. Plenty of time to do … y’know, president stuff … and … profit?

Totally gonna work. I mean, how could it not? It’s like, really smart.

Note: The day after I wrote this, Secretary of the Treasury Steve Mnuchin called all the major bank CEOs from his vacation spot in Cabo San Lucas and assured them the market had lending liquidity even though they didn’t ask, and the market reacted like it was made entirely of flame.

This is the twenty-sixth installment of a series on politics and game theory. It has covered impeachment, Russian collusion, white supremacy, abortion, guns, nuclear war, debt, the NFL, sexual harassment, the Mueller probe, taxes, Trump’s first year, the Clinton Foundation, immigration, parades, the Democrats, hope, family separation, trade wars, Trump’s endgame, the New York Times op-ed, Justice Kavanaugh, Speaker Pelosi, lame ducks, and the GOP legacy. Essays like these are in my book Game Theory in the Age of Chaos, which you can order by clicking the link.