Politics is mostly responsible for getting us to the verge of impeaching and removing a president. While there are good reasons for initiating the process, the best legal arguments in the world won’t finish the job. As a political process, the electorate has to be convinced of the validity of the outcome or public faith in rule of law could be further eroded. Current polling suggests that slightly more people favor removal but those numbers can shift especially with the daily revelations likely to come out of a senate trial. Game theory may offer clues about how the process can be brought to a satisfying conclusion for the majority of Americans.
Conventional wisdom has it that the house of Representatives might be able to impeach Donald Trump, but the GOP dominated Senate won’t vote to remove him from office. That calculation is focused on the Constitutional reality that it takes two-thirds of the Senate to vote for removal which would require 20 GOP senators to cross the aisle to affect the deed. The arithmetic involved is what provides Trump with a firewall against removal from office. But how valid is that logic?
The math depends on a straight party line vote which would leave Trump in office by something close to the 53–47 senate split that currently favors the GOP. Some moderate and progressive prognosticators often allude to a patriotic duty for senators to examine the evidence against Trump, which is damning, and to vote accordingly. But others consistently fall back on party affiliation with an argument of the type that blood is thicker than water. Many ordinary citizens simply throw up their hands and wait for the drama to play out.
Nevertheless, these analyses, while often well intended, only scratch the surface; there’s a straight-line logic to them that brooks no possibility of variation. But the situation is completely fluid and opportunities to think in straight lines may be few and far between. A deeper look via game theory reveals something else. Something that Trump supporters should view with raised eyebrows.
The prisoner’s dilemma
Basic game theory tries to reveal how humans might act under pressure and without all of the available data and it shows that people can act erratically for self-preservation. One of the classic gambits in game theory is the Prisoner’s Dilemma in which two people accused of the same crime are separated by inquisitors for questioning. It’s a sound police tactic.
If the perpetrators are allowed to communicate, they can more easily keep their story straight. But if they are separated, neither knows what the other is saying to the authorities and there is great temptation for each to throw the partner under the bus, especially if the authorities are offering leniency to the first to offer sound evidence implicating the other. Even if there was a pact between the partners in crime not to divulge anything about the effort, the threat of a long sentence, especially if coupled with an insinuation by the authorities that they already have a confession from the other partner, is sometimes enough to get evidence.
The gambit works a fair amount of the time, and not simply with naïve criminals. For instance, the recent depositions before the House Intelligence Committee were held in private and transcript releases were delayed until related witnesses could deliver their testimony without influence from earlier proceedings. This provided an incentive for all to tell the truth or risk a perjury charge. Witnesses like Ambassador Gordon Sondland were induced to amend their testimonies once transcripts of other witnesses’ testimonies were made public. As a result, the House Intelligence Committee has a cohesive body of evidence that it has referred to the Judiciary Committee for drafting into articles of impeachment.
Foreshadowing how the senate votes
Now imagine you are a prisoner and also imagine you are seeking the best treatment from your jailer. Much as you despise your keeper, you’ll likely not make any significant opposition to him or her if you believe opposition is futile. Eventually that’s the position many prisoner’s dilemma players find themselves in. Your strategy might likely revolve around an effort to go along and get along while biding your time looking for an opportunity to make a break. This strategy could take years to play out but jailbreaks do happen.
Republican Senators are currently in a situation that mimics the above scenario. They are prisoners of the Republican Party and their jailer is none other than Donald Trump who can summon a backlash against any senator with the temerity to oppose him. The party’s preferred method of retaliation is introducing a primary challenger for any senator deemed not sufficiently loyal or supportive of the president.
There are 22 GOP senators up for re-election in 2020 and the state primary filing deadlines are all over the calendar; though a cluster of them happen in March or earlier, some happen as late as June. Many of the senators up for re-election come from safe seats in which the greatest threat for the incumbent is being removed in a primary challenge and not the general election so senators are indeed cautious. Senators who opposed Trump during the first two years of his presidency like Bob Corker of Tennessee and Jeff Flake of Arizona retired from the senate after discovering how difficult their re-elections would be in 2018 having opposed Trump.
It would seem that Trump’s hold on the Senate and therefore his acquittal is nearly secure, but this analysis omits one critical feature. If the scuttlebutt coming out of the Senate is true, most Republican senators loathe Trump in private but nonetheless support him in public. That’s just the behavior you’d expect from risk averse prisoners and it’s logical.
But all is not lost for the GOP or its senators. If any of them is thinking of a jailbreak, the impeachment trial offers the best opportunity, and the stakes are high. Preserving Trump’s presidency is one thing but arguably more important is the future of the Republican Party.
Trump’s power stems from an ability to pick off individual senators who cross him so he can almost always defeat his opposition in detail, one at a time. But impeachment and a Senate trial present a unique moment. During a Senate trial all senators are compelled to attend and to be quiet actively listening. The only defenders of the president would be Republican House members designated to defend him and at the end of the trial there would be a simple up or down vote: “aye” for removal and “nay” to acquit.
Up to the point of the vote, game theory suggests that senators and GOP members of the house in general would support their man, forcing the Democrats to make their case without their assistance. You’d expect the same in a court of law: The defense never aids the prosecution in its work.
It’s exactly at the vote that Republican senators could for the first- and only-time work in unison to rescue themselves from their jailer. But would they? At this point many senators might have a longer purview than the next presidential election. Senators are elected for 6-year terms and they can and do run for re-election repeatedly. So, an argument can be made that they have, or should have, the longer-term interests of the nation in mind. But even more fundamentally, they have history to reference–not in Watergate and Nixon’s resignation but in Herbert Hoover.
Hoover was in office less than a year when Wall Street sank in October 1929 into a sea of bad debt that initiated a significant economic downturn. Hoover’s laissez faire management of the economy turned a bad recession into the Great Depression. By the 1932 election the nation was flat on its back with 25 percent unemployment and Hoover lost to Franklin Roosevelt in a race that wasn’t close. But the important part of the story for today’s GOP is that Hoover’s maladministration poisoned the well for the Republican party for 5 election cycles, 20 years in total.
Any student of history, and we can assume that senators have more than a passing interest in it, can see the dominoes setting up for a prolonged Republican turn in the wilderness. It’s doubtful that the wilderness years could have the same result as FDR’s triumph over Hoover, for example, the 22nd amendment sets a two-term limit on presidents.
After Nixon-Ford, the Democrats were only able to hold power with Jimmy Carter for four years before Ronald Reagan served two terms and G.H.W. Bush served one. But Carter was not a great politician and it’s questionable how a better candidate would fare today assuming Trump is not removed and is instead defeated at the ballot box.
But let’s stipulate that it would be bad for the Republican brand to look the other way on Trump’s indiscretions, especially if the Democrats can put on an air-tight case.
It would be better for GOP senators to vote to remove Trump at the trial than to appear to be derelict in their duty during a televised trial and vote. There’s also insurance for senators working in unison. Each could mount the argument that they were with the president throughout the trial but that the mountain of evidence they reviewed was simply persuasive and, in the end, each voted his or her conscience. No doubt many members of the party would be upset with the result, but we’re talking about a portion of the base, which is a plurality, not a majority, of the American people.
With that logic, the GOP could affect its jailbreak and reclaim the party from extremists on the right. Some GOP senators could still lose their re-election fights, but the net effect would be to save the party, something more important. That’s why predicting a 98–2 outcome favoring removal is not out of the question though it will remain out of sight for the duration of the process.
(Cross-posted @ Denis Pombriant | Medium)