Raymond's Weblog^

Conversation as a mini economy

tldr: the movement of large conversations is a bit like trades in a market, where speech is an implicit bid on the value of what you’re saying

If you already find group conversation easy and comfortable then this post will be at best an interesting novelty. But somehow I keep meeting people who are much better at intuiting economics than at intuiting how to participate in large groups.

What’s up with big conversations

Here are some behaviours you might have observed in big groups and been baffled by:

This is obviously partly to do with what’s being said in the moment, but also not entirely. It can seem like over the course of a conversation, some people get shut out. Why?

The basic mechanism

I think what’s going on is that over the course of the conversation, the group is collectively keeping a kind of scorecard for how good people’s contributions are.

You start off with some number of credits, for outside reasons like what the context is or how you joined the conversation - basically people’s rough estimate of how good your contributions will be.

Then, whenever you try to speak, you are implicitly gambling some of those credits on this contribution being valuable. The longer you talk, the more credits you’re burning. But if the contribution actually is valuable, you get more credits back.

If you run out of credits, that’s when people start talking over you, or just not listening very hard to what you’re saying. Once you get lots of credits, that’s when you start getting to make big speeches.

This is basically a good system. It’s a good way to make sure that conversational time gets spent on valuable things, in a fairly distributed way. The downsides are:

I claim that in practice, this last point covers all the above points. Once you know how to play the game, you can just be a bit more strategic, and then you won’t get talked over, and you won’t make long points until you have enough credit banked.

Conflicts

Most of the time, I don’t think people are actually trying to be dicks. Taking the most charitable lens, I think things like interrupting and ignoring should be viewed as prosocial behaviour - trying to make sure the conversation goes in a good direction.

It is also a risk: if you interrupt someone who was saying something that you thought was pointless, but then they interrupt back, keep talking, and say something that the group agrees was valuable, then you lose a lot of credit.

Analogously, when Person A tries to open a line of conversation, Person B’s contribution to that line is a form of investment. If Person B’s contribution leads somewhere good, both people get credit for it.

In fact, Person B can invest in many different ways:

Much like the real economy, conversations are not a series of narrow trades where some people lose and others profit. Conflict is still an important mechanism, for the same reason that efficient markets include a lot of shorting.

Sensible investment in conversation

So what should you actually do with all this? How should you approach large group conversations with all this in mind? Here are some thoughts:

The above strategy is, I believe, generally better than both saying whatever comes into your mind and never saying anything, especially for anyone who’s more comfortable with economics than large groups.

(Bonus: Why groups suck at making decisions)

Groups are often weirdly bad at coming to agreements about what to do, where to eat, and so on: in fact, some people really resist even making suggestions or advocating for actually doing anything.

Why? Because people are concave optimisers! Making a decision for the group is a really large bet. Sure, you’d get a lot of credit if you could pick a movie everyone wanted to watch, but do you really want that much credit?

Even if you’re pretty confident, making a choice for a large group requires staking a lot, and most people would prefer many small investments. It’s just the prudent choice (for an individual, at least).