Anti-Rivalrous Goods

Uri -

In economics there's a traditional distinction between two types of goods. "Rivalrous goods" are anything where my consumption of the good prevents consumption by somebody else: if I eat an orange then there's one less orange for you; if I sit down at a restaurant there's one less space for somebody else, etc. Most traditional physical goods are rivalrous.

The second category of goods is called "non-rivalrous": my usage of the good doesn't stop anyone else from using it. The classic example for some reason is missile defense: I can "consume" protection from my country's missile defense system without reducing protection for anyone else.

It struck me the other day that there's an obvious missing category here: anti-rivalrous goods, where I actively benefit from other people consuming the same thing. It turns out somebody else beat me to it by a couple of decades, and gave it the same name, so well done Steven Weber in 2004. But since I never learned about anti-rivalrous goods in school, I suspect at least some other readers might not have either.

The example Weber gave was free and open source software: the more people use it, the more powerful it becomes for others. But I think anti-rivalrous goods are incredibly common, and worth paying attention to. For example:

  • I want to watch movies and TV shows that my friends have also watched. We saw this phenomenon strongly with Boppenheimer: many people went to the cinema to watch at least one movie they weren't especially interested in, just because at some point their value from seeing "the movie everyone else is currently watching" outweighed their substantive preference over which movie to watch.
  • Jane Jacobs' "Eyes on the Street" theory held that when more people are out on their stoops then the street feels safer, and therefore more people go out on their stoops (etc).

On the last example, obviously at some point you have too many people out on the street and it starts to feel crowded: in other words, at some point you transition into a rivalrous good again.

I suspect that actually many goods have a rivalrousness s-curve: things start anti-rivalrous, become non-rivalrous in the middle, then become rivalrous again at the end.