# Dunbar’s Number

One of my favorite ideas in social science is Dunbar’s number: the cognitive limit to the number of relationships our brains can handle. It’s something like 150. That’s about the number of people our ancestors might have shared their tribe with 20,000 years ago.

Our sense of social propriety is tuned to dealing with people within our circle. Economics often seems counter-intuitive because it’s largely about how to interact with people outside that circle.

Here’s the example I use in class:

You’ve got a date tonight. You stop at a florist to pick out a bouquet and start wondering if maybe chocolate would be a better gift. Dark chocolate or milk? Or maybe something else. You think back to your economics classes and realize that if your date had \$20 cash, they could buy this bouquet if it’s what they really wanted, or chocolates if that’s preferred. So when you knock on the door, instead of offering a bouquet, you hold out a crisp \$20 bill.

What happens next?

If you aren’t dating an economist, you get the door slammed in your face.

So you run back to the florist, buy the flowers, run back, and nobody answers the door. Your date probably went to the bar with friends. You call a cab. When it pulls up to your door, the fare is \$20. You just spent \$20 on these flowers. You try to pay for your fare with the flowers.

What happens next?

The driver refuses and insists on cash.

So what’s better, flowers or cash? Is your date irrational, or the cab driver? Neither. Both are rational within the context they’re acting in. The driver is a stranger and market rules are appropriate. In this context, \$20 is worth more than \$20 worth of flowers. Maybe the cab driver wants flowers, but cash gives them the option to buy whatever best meets their needs.

You and your date were trying to cease being strangers. The cab driver is outside of Dunbar’s number, but your date would have (could have) been inside that inner circle. At that point, the signaling value of the flowers would trump the economic value of the cash.

Economics has a lot to tell us about how to behave with those inside and outside of our Dunbar’s number. But that dividing line calls for different rules on either side: the rules of family and neighbors on one side, and the rules of the market on the other.

I’m thinking about Dunbar’s number because I just finished a recent episode of EconTalk where they talked about a classic example from behavioral economics: An Israeli daycare, tired of late pickups started charging fines to late parents. Ironically, this resulted in an increase in late pickups that persisted even when the policy was reversed.

The daycare example is often trotted out to say “see! Sometimes adding an incentive backfires! Raising the price from \$0 to \$x increased the quantity of lateness demanded. People are irrational!” Of course it only takes about 5 seconds of thinking to realize that we aren’t holding all else equal here. As usual, there’s a lot of interesting stuff hidden in the ceteris paribus assumption.

The more sophisticated interpretation of this example is that attaching a price shifted parent’s interpretations of the norms. In my language: the inclusion of money in that interaction shifted the rules of behavior from those of neighbors to those of strangers.

(Roberts brought up an important point I hadn’t considered with this example: the price was too low. Prices communicate information about how onerous it is to produce a product, and that price told parents “it’s not a big deal if you’re late…”)

More generally, when we’re looking at some social scientific question, Dunbar’s number demarcates a point separating the assumptions we can make about sharing and monitoring–whether it’s about the practicability of communism (the real kind, not the kind with mass murder), corporate bureaucracy and firm size, or the tenability of informal institutions.