Dear Starknet community,
I’m opening this topic to discuss everything economics on Starknet, following the great article & Video from Noam Nisan.
First of all, I think it would be great to have articles explaining Starknet tokenomics, the same way the great @ilia wrote a few detailing Starknet decentralization.
Now going on to the questions/remarks I have:
I think it was very much needed to define the goal of the platform, and to maximize social welfare is I think the best way to describe this goal. Now very naively, I think I would have defined what this means based on the participants rather than the system. In the article, part 2.2, how can we maximize welfare is based on the system: users and operators are defined, and social welfare will be maximized based on this two-sided market. Of course, it is the most important part of the system, but I feel this narrows the scope of the analysis too much, as many other parts are included in the system. If I were to define what maximizing social welfare means for Starknet, it would be maximizing the value provided to the world (same as in the article), but applied to every community member, and this includes of course users & operators, but also any other economic agent that can for instance just hold STRK, use it as collateral in DeFi, etc. This might not make a big difference right now, but in the later discussion it will.
On the microeconomics part, it is very well detailed, and I would love to have more details on the path chosen to price fees on Starknet, as there are only hints on the different possibilities for now. I think detailing the MEV part would also make a lot of sense since it plays a huge part in the overall social welfare maximization.
On the macroeconomics part, I’m not sure I’m 100% convinced that inflation is the best way to cover the fixed cost of capital immobilization for validators. The assumption made at the beginning that the system is static is a big one. In a dynamic system, inflation doesn’t equally share the fixed costs of the infra on token holders, as we tend to understand in the article. In fact, it flows, and the people the closest to the new-minted money are the ones benefiting the most of it. It is a well-detailed matter in traditional economics, but to clarify in the case of Starknet, new STRK is created and sent to the validator, which can use it by buying blockspace for instance. The validator entering the blockspace market with this new supply of tokens will be able to buy it at the current price but will drive the price of it higher, as the demand increased. Thus token holders end up with a STRK less valuable, as it can buy less blockspace. With this solution, the users pay just the marginal cost (happy), the operators are the closest to the source of inflation (happy), and the token holders are the ones paying for the fixed costs, in a very unequal way because inflation in a dynamic system is flowing (somewhat unhappy). Now if we decide to make the user pay for the fixed cost, we just end up with a different set of happy/unhappy people, with users somewhat unhappy, operators happy, and token holders happy. That’s why I think defining the maximization of social welfare using only operators & users might be misleading, as we end up finding the best solution that puts the burden on a third group of people, that wasn’t part of the initial equation. That being said, opting for the solution “the users are paying for the fixed costs” doesn’t justify the very existence of STRK, as ETH or USD could be used instead.
I also think more details on the particular case of Starknet would be great. It is an L2, so there are now provers & sequencers, which adds more layers to the basic analysis. There’s also maybe less decentralization needed since sequencers are only providing liveness and not trustlessness, so this would change the whole target of inflation, and maybe the subject of liquid staking and so on that would be very different from L1.
I couldn’t tag Noam, he doesn’t look on the forum, hope to have him soon here!
