The idea of "zero-sum" economic activity sounds sort of plausible. And it feels like there's a lot of truth to it, but when I think about it harder I find it re

The idea of "zero-sum" economic activity sounds sort of plausible. And it feels like there's a lot of truth to it, but when I think about it harder I find it really difficult to nail down what that actually means. Most of the examples given in the article are of things that you could argue provide a sort of "infrastructure" service: Legal services allow businesses to have confidence in their contracts, financial trading serves to allocate resources efficiently, advertising solves the problem of discovering new goods and services. Yes, I know that sounds like a crazy optimistic view of these industries, but how do you know which parts are useful and which aren't? Do we know it's even possible to remove the useless parts and keep the important ones? It seems to me that one interesting definition of "value of this activity" would be: How many other activities depend on it? By that particular definition, any activity that provides some sort of "infrastructure" would be particularly valuable.