Frequent Batch Auctions
Frequent batch auctions are a proposed alternative to continuous limit order book trading in which incoming buy and sell orders are accumulated over brief, regular intervals — typically between 100 milliseconds and one second — and then matched simultaneously at a single clearing price, eliminating the first-mover advantage that governs continuous markets and potentially reducing the profitability of pure latency-based trading strategies.
In a continuous limit order book, the time at which an order arrives determines execution priority among orders at the same price. This creates a race to the exchange: traders who submit orders a microsecond earlier than competitors receive fills that slower traders miss. The commercial arms race to reduce latency — through co-location, microwave networks, and specialized hardware — represents billions of dollars in sunk costs that improve private execution speed without a corresponding improvement in the services delivered to end investors.
Frequent batch auctions, proposed most prominently by University of Chicago economists Eric Budish, Peter Cramton, and John Shim in a 2015 paper, eliminate this race by collapsing a continuous time axis into discrete batch intervals. All orders received during an interval — say, from time t to t plus 100 milliseconds — are treated as arriving simultaneously. They are matched according to price-time priority only in the trivial sense that time priority within a batch is irrelevant; price is the only meaningful dimension. The market-clearing price is the price at which the maximum volume of orders can be executed.
The theoretical benefits of frequent batch auctions are substantial. By eliminating the continuous race, the format reduces the private value of sub-millisecond speed advantages, potentially redirecting resources from latency-reduction technology toward research and analysis. The format also prevents certain forms of latency arbitrage: an arbitrageur who observes a directional signal in Chicago futures and attempts to hit stale quotes in New York equities cannot do so within the same batch because all orders, including corrective quote updates from market makers, arrive simultaneously.
Implementation challenges include defining appropriate batch intervals — too short and the auction degenerates toward continuous trading; too long and legitimate time-sensitive orders are delayed unacceptably — and managing the transition from existing infrastructure. No major U.S. exchange operates under a pure frequent batch auction model, though IEX and certain European venues have implemented partial approximations. Academic support for the concept remains strong, and it continues to inform regulatory discussions about market structure reform.