We are going to be doing some modeling and simulations to understand the structure of the current PBS market. As part of this, we would like to look at as wide a range of possible designs for the validator/builder market auction as possible.
For example, we will be looking at the impact of:
second price auctions (ignoring feasibility)
sealed vs. open bids
dynamic time (clock) auctions
slot auctions
some all-pay auction variations
(we will play around with the presence of relays in the above as well).
What else should we be considering? Are there any non-standard mechanisms (e.g introducing randomness) that could maximise welfare? What is a better way to structure the market for block construction?
Why ignoring feasibility in second price? can you give some more detail of what you mean by that?
What measure of welfare are you using?
Re: clock auctions, you want to watch out here that you account for the adverse selection as information about block value is released till the very end.
I added to feasibility note mostly to point out that it’s worth studying the impact on market dynamics independent of whether the mechanism can be implemented in a credible way, although the comment should be applied to all of the studies in general.
This seems a bit confused. Feasibility of a vcg allocation brings to mind [1002.2334] Budget Feasible Mechanisms , credibility of the auctioneer is a separate issue.
I don’t think it makes any sense to look at block space allocations that are not budget feasable
Are there any non-standard mechanisms (e.g introducing randomness) that could maximise welfare?
This was more of a general question around potential novel mechanisms specific to validator/relay auctions that could maximise some welfare measure. I haven’t thought much about or have a strong opinion on the specific welfare measure to use for this work, but if you have ideas or input I would be interested in hearing more. The group that are doing the initial work on this probably have something in mind already too.