Democracies and Long-Term Games

Asked about whether he thinks democracy works, Dominic Cummings recently said:  “It’s like asking questions about 250 BC in 100 BC about the Roman Republic. Well, the Roman Republic has lasted for a century so far. You certainly can’t say now that democracy has proved to work.”

I tend to agree with Cummings. In school, we spend quite some time discussing the differences between autocracies and democracies. However, we do not cover how the former arises from the latter. How does a pre-autocracy look like, and are we becoming one?

In particular, we do not cover nearly enough how we can prevent our current democracies from devolving into autocracies.

To do this, we have two instruments.

1. The first tool is called pre-mortems. It consists of imagining that the risk we are afraid of materializes in the future and then asking ourselves what the most likely pathway to that is. For example, let’s imagine that in ten years, our country became an autocracy. What might have happened? And what can we do today to prevent that?

2. The second tool is to observe how such transitions happened in the past in other countries. For example, what are some democracies that transitioned to autocracies? What mistakes did those countries commit? Is our country committing the same mistakes now?

The key to both exercises is to avoid thinking, “How likely is it to happen?” and instead focus on “If I don’t take care of this risk, eventually, it will materialize.”

Survivorship bias

It’s easy to think, “Our country is a democracy; we are better than other countries that are autocracies.”

But guess what? They also thought they were a democratic country when they were already pre-autocratic.

It’s a bit like that data point that 93% of US drivers believe they are better drivers than average. Until you get into an accident, you will think you are a better driver than average.

• Do you drive slow? You will think that that makes you a good driver.

• Do you drive fast? You will think that that makes you a good driver.

• Did you drive recklessly and almost got into an accident but managed to emergency brake at the last second? You will think that your great reflexes make you a good driver.

So, how can you know how good a driver you really are?

You can check whether your driving looks like the driving of the drivers that crashed – but not on the day they crashed, on the days before that. If so, I’m sorry, but you might be at risk of a crash, too.

Problems grow the size they need for us to acknowledge them

If we want to avoid large problems, we shall tackle them before they grow large enough to justify taking action.

This means to regularly practice the exercises described above. And not just in the context of our country’s governance but in all aspects of our lives.

Let’s imagine we executed our current strategy yet failed. What might be the most likely cause, and what can we do to prevent that?

How do others with the same goal as ours fail? And what can we do about it?

What does pre-failure look like? And how can we avoid that?

Contrary to common belief, thinking about these questions doesn’t bring anxiety. Instead, it is precisely what enables us to sleep well (because we know we took care of the most significant risks).

Notes

This essay was written as a chapter of my 2024 book “Winning Long-Term Games,” but got cut out at the editing stage. You can get the full book and audiobook here.