It’s a monster election year. Don’t let the horse race distract you.

I have no idea how I got to my office this morning. I mean, I do know: I walked to the tube station near my house, got on a train, transferred a few stops later to another one, got off near my office and then walked in, making the briefest of stops at a coffee shop to pick up a breakfast sandwich on the way.

But that list of steps describes the limit of my knowledge. I have no idea who opened the tube station, or what it takes to keep it functioning. (Or, for that matter, why one of its turnstiles was stuck half open, bleeping a plaintive alarm about its situation to no one in particular.) I do not know how to drive a train, and certainly not how to maintain one. And I’m sure the people of London are very grateful that I have never had to consider how to dig a subway tunnel or lay a rail line.

And yet if those things had not happened in the correct order, as designed by experts and carried out by professionals, the city would shut down. This week that shutdown nearly happened, in fact, because of a transport strike that was called off at the last moment.

This is the magical thing about institutions: They exist so that complex processes can become automatic, so that large groups of people can collaborate without having to create new systems for doing so, and so people like me can rely on expertise without possessing that expertise even a tiny bit.

But because institutions often operate in the background, unnoticed, it can be difficult to pinpoint the moment when they start to break down. And, frustratingly for me, it’s even more difficult to write about incremental decline without sounding stultifyingly boring.

This has been on my mind because 2024 is set to be the biggest election year in world history, with approximately half the population set to go to the polls. The contests are undoubtedly significant: from Taiwan, which will choose a new president on Saturday, to the results of the United States’ election in November, which will have enormous consequences for the entire international order.

Yet there is a tension in how the media covers these electoral contests: following the horse race, while exciting, shouldn’t obscure the longer-running, incremental story about what is happening in institutions around the world, many of which are in a state of slow decline or accelerating rot.

In the United States, for instance, the re-election of Donald Trump would have huge consequences for geopolitics, the country’s allies and enemies, and for the United States itself. But at the same time, re-electing Joe Biden would not remove the pressure building on the international order, because the institutional gridlock in Congress would remain, as would the dysfunction in the Republican Party.

Both of these factors are already distorting U.S. foreign policy under the current administration, as revealed by the fight over continued funding for the war in Ukraine. That has huge implications for the many other countries that rely on the U.S. as an ally: even if the money keeps flowing for now, an uncertain commitment is less valuable than a certain one.

“The promise of unquestioned U.S. commitment in the world is not one that we can take for granted anymore,” Elizabeth Saunders, a Columbia University political scientist who studies U.S. foreign policy, told me this week. “Whoever wins the presidential election will just have to deal with that reality.”

Institutions work best when they rest on a strong foundation of trust, dependability and are rooted in well-established systems. To go back to my subway analogy, a tube system that can be reliably counted on every day expands the places where commuters can live, means they can buy fewer cars, and becomes a load-bearing element of daily schedules.

A system that’s frequently unavailable, by contrast, is one that cannot underpin any of those things. The whole system works less well.

And in countries around the world, political polarization, populist governments, and years of political chaos have undermined courts and other institutions, creating exactly that kind of problem. They still function, they still mostly serve their intended roles, but they can’t be counted on to the same degree. In that world, uncertainty becomes more of a factor. In that world, you need more backup plans, more individual solutions, more redundant layers of infrastructure. It’s less efficient.

“The institution has become incrementally less reliable” is not exactly clickbait. But sometimes it’s the most important story going.