Trust the Insiders

Once again, the institutions charged with ensuring fair elections have done a great job

By Garett Jones

Over the past week or so, America’s insiders did a great job: the media was mostly cautious about calling election results in states where the vote count was tight, judges prudently evaluated the Trump campaign’s claims of unfair vote-counting practices, and the long-standing technical process of vote-counting has proceeded smoothly even with our unprecedentedly high volume of mail-in votes. And while pre-election polls ended up underestimating support for the president, they still called the winner correctly in about 48 out of 50 states, no small accomplishment. This is what usually happens when we leave big decisions to the elites, to the insiders, in the modern United States: the results aren’t perfect, but they’re far, far better than if we just left things to the complainers on social media.

Everyone knows that insiders granted vast powers have a strong temptation to abuse those powers: hand in the cookie jar and all that. But fewer people know that modern organizations take that temptation for abuse into account and try to prevent abusive practices precisely because those organizations don’t want their reputation destroyed by bad actors. So when President Trump—someone with access to nuclear launch codes and with power over both the Internal Revenue Service and the Central Intelligence Agency—called up Rupert Murdoch to demand that Fox News retract its prediction that Trump would lose the state of Arizona, Murdoch chose to reject the demands of the most powerful person in all of human history.

Why? We can never be sure about the true motives of any one person, but by leaving the decision on Arizona to his own experts, Murdoch boosted the reputation of independent judgment at Fox News. Reputations matter. And just as Apple produces reliable, long-lasting iPhones today in order to sell more iPhones going forward—rather than just producing phones that break every 12 months in order to generate more sales—so too the mainstream media insulates its election decision desks from outside pressures in order to boost their reputation for accuracy.

One sign of that insulation: the very fact that most election decision desks didn’t call Arizona for Biden, even after Fox News and the Associated Press did. Other news orgs—and FiveThirtyEight’s Nate Silver—thought that a call on Arizona was premature when so many votes were uncounted, the vote was fairly tight and it wasn’t clear where the remaining votes in Arizona were coming from. As I was taught long ago by Nobel laureate Clive Granger, one sign you’re looking at an efficient group of forecasts is when each forecast makes its own unique mistakes, rather than all following each other off the cliff. In the real world, a good batch of forecasts is one where every forecast makes its own mistakes. That’s a lot like what’s happened at the election decision desks this year, and a sign they really are relatively insulated from each other and independent.

Likewise, our independent judiciary—the least democratic branch of our government—has shown it can assess claims of election fraud both fairly and quickly, especially when there’s no concrete evidence for widespread fraud, when the fraud is hypothetical rather than evidence-based. Judges have long terms in office—often lifelong terms—and their guaranteed salaries insulate them from the pressures to cave in this or that political direction.

And our election system, designed well in advance of the election itself, is a marvel to behold. The losing side in every election has a strong incentive to highlight every single flaw, but when the dust has settled and the paint has dried in every US presidential election since 2000, those claims of fraud and hacking and bias have had little or no real-world impact, turning out to be rounding error at most.

Consider a few examples: Despite the claims that the 2000 vote count in Florida was rigged in favor of Bush, every scholarly and expert assessment has concluded that Bush would have won under almost every vote-counting method that focused on undervotes—the so-called hanging chad votes. (A recount that had focused on overvotes—where people filled in the bubble for Gore but also wrote in “Gore”—would have led to a Gore victory, but the Gore campaign never pushed to count overvotes.) In 2004, claims that Senator John Kerry would have won Ohio if not for biased electronic voting machines helped inspire an Emmy-winning documentary—Hacking Democracy—but there’s no evidence at all that electronic voting problems were big enough to decide the 2004 election. And in 2016, fears that Russian saboteurs had hacked voting machines turned out to be entirely baseless.

Our voting system, created by technicians working far in advance and monitored at each stage by Democrats, Republicans and independents, hasn’t been perfect, but perfection isn’t the standard. The American system of voting for president, designed by elites and run mostly by insiders, has repeatedly withstood widespread claims of fraud in the past. It is doing so now and will surely do so again in the future.

I wrote my latest book, 10% Less Democracy: Why You Should Trust Elites a Little More and the Masses a Little Less, well before the 2020 election. But so far, the central argument I make is holding up well. Our elite-run institutions aren’t perfect, but wow, when we let them do their job and insulate them from short-run political pressures, they do very well. Let’s hope 2021 is the year we insulate our elites from democratic politics just 10% more.