On the Sia blog, Zach Herbert wrote about some of the thinking behind Sia’s two-token cryptoeconomic model. It’s good and worth a read. It prompted me to float some thoughts on revenue-sharing tokens.
Note: I haven’t thought too deeply about Sia’s particular use case so my thoughts/questions here may not be applicable. I’m just speaking about revenue-sharing tokens in general.
It seems to me that revenue-sharing tokens have strange consequences in public protocols and raise the question of what is reasonable/fair compensation for protocol developers.
Founders (and revenue-sharing token holders) are incentivized to maximize network usage. This aligns very well with network participants in the early days of the network where growth is generally beneficial for everyone. Over time, however, alignment seems to diverge.
At some point, for some/most network participants, the network is big enough and diverse enough that it meets their needs and further network growth has diminishing or zero returns. At this point, participants become much more interested in the efficiency of each transaction.
But there is nothing that they, nor any other stakeholder (e.g. miners), can do to drive down the "tax" that they have to pay on every transaction. The revenue-sharing fee is independent of the cost-of-goods and so runs the risk of seeming punitive.
And short of a network revolt, even if revenue-sharing token holders could reduce their fees, they are not incentivized to do so.
In this respect, from the perspective of network participants, single-token models seem to compensate founders more fairly: founders get paid as the network grows but not beyond that. Extracting fees from a network in a steady state smells like rent-seeking.
BTW, I also somewhat disagree with the criticisms of single-token models in the article. I think vesting schedules for founders removes the desire to pump the token in the short term and models such as those suggested in Kyle Samani’s New Models for Utility Tokens can attenuate the velocity problem.