Fair Combinatorial Auction for Blockchain Trade Intents: Being Fair without Knowing What is Fair

Abstract

We study blockchain trade-intent auctions, which currently intermediate about USD 10 billion in trades each month. These auctions are combinatorial because executing multiple trade intents jointly generates additional efficiencies. However, the auctioneer cannot observe what each trader would have received had its order been auctioned individually and hence cannot determine how these efficiencies should be shared. We compare the two dominant mechanisms - batch auctions and simultaneous individual auctions - and introduce a novel definition of fairness applicable to combinatorial auctions. We then propose a fair combinatorial auction that endogenously constructs a fairness benchmark from individual bids and a counterfactual mechanism. Whether fairness guarantees arise in equilibrium depends on the counterfactual: all traders receive more in the equilibrium of the fair combinatorial auction than in the equilibrium of the counterfactual mechanism when the counterfactual is simultaneous first-price auctions, but that may not be the case if the counterfactual is simultaneous second-price auctions.