conversations with tylerCOWEN: If you think about economics — what you’ve taught yourself — what do you think of as the central ideas relevant for cryptoeconomics, which is a term I think you coined?BUTERIN: I would say that cryptoeconomics, first of all, is economics. It’s not like we’re inventing some completely different parallel society with a different parallel economy and different rules, but it is economics specialized to a particular set of circumstances. Then you have to ask yourself, “Well, what describes those circumstances?” There are a few major parts to that answer.
First of all, whatever mechanisms you have in cryptoeconomics land have to be fully specified exactly — not exactly to the standards of a court judge, but exactly to the standards of a computer programmer. What that means is that there’s a lot of things that you can’t do.
For example, you can’t say in cryptoeconomics, “It’s illegal to bribe people,” because there’s really no simple way to define what a bribe is. If someone really wants to bribe someone else, he can just go and do that outside of the protocol, and the protocol would have no way to tell.
Whatever your rules are for rewarding, penalizing inside of the mechanism, they have to be specified as a piece of Solidity code, Viper code, whatever programming language you’re using in that set. That’s a much tighter constraint than policymakers writing laws have.
Another one is, of course, that all of the actors are anonymous, and what that means in practice is that you cannot drag people’s utility down below zero. If I have 70 ether, and I put that 70 ether into a mechanism, the worst thing you can do to me is you can take away that 70 ether.
You cannot throw me in jail. You cannot socially ostracize me so I can’t earn any money again because I can always just switch identities. But to the extent that I’m willing to lock that ether up and make it vulnerable to a mechanism, then you have the ability to motivate me to that extent.
Cryptoeconomics is basically taking economics with those particular constraints and then adding together insights from fields that are fairly close by to the cryptocurrency space — particularly cryptography, information theory, math, and distributed systems, including all of the research around consensus algorithms, hash functions, signatures, zero-knowledge proofs, and what we know about all of those primitives.
Basically we’re trying to figure out, given these constraints and given these building blocks, what kind of systems and what kind of mechanisms can we design to achieve the properties that we want? And under what kinds of assumptions do those properties hold?
COWEN: What’s the contribution or understanding you wish economics could give you that it hasn’t given you yet? What problem do you want it to solve for you?
BUTERIN: I think that once you start getting into the weeds, and you try to move beyond very simple things like proof of work, the simplest versions of proof of stake, and the tricks that we’ve known about since 2009 or 2013, then you get into this situation where you realize the kinds of problems that you run into when you try to incentivize behavior.
Particularly when you run into things like what I call speaker/listener fault equivalence, which is basically this idea that if the protocol requires Alice to send a message to Bob, and the protocol detects that Bob never showed a message from Alice, then the protocol knows that either it’s because Alice forgot to speak or it’s because Bob pretended not to listen.
Then, under those kinds of conditions, trying to figure out what is the right way to motivate Alice and motivate Bob while preserving all sorts of different properties. These are questions that are very close to questions that game theorists, mechanism designers, people in the economics community have already had a lot of things to say about. There definitely are results from the existing fields of economics and game theory that have helped significantly in that regard.
Another whole aspect is also the broader and social question of how will blockchain-based systems actually end up affecting society?
What is a blockchain’s unique competitive advantage? What are things that you can do with blockchains that you can do only with more difficulty without them? Trying to go from there and zero in and try to figure out exactly what industries and what kind of applications in those industries should be going after.
Do the things at the application layer that we’re trying to build even make sense? Should we modify them in some way to have them make more sense? I do think that people from the economics community have a lot to say about that.
Alex Tabarrok, for example, had that very nice post on Marginal Revolution a couple of weeks back, where he talked about blockchains in terms of being an alternative to platform monopolies, and I thought that was a very good insight.
I thought that the general idea that a blockchain is one of the few tools that allows you to credibly commit to not turning into a monopolistic jerk is actually a really interesting insight and one that I have arrived at myself, but it definitely helps to have it be crystallized in that way.
