Bid and Ask Taxes: Laboratory Evidence and Market Design Questions
Summary
The document distinguishes a bid or ask tax from a transaction tax: the charge applies when a limit order is submitted, even if the order never trades. It describes two laboratory double-auction studies in which participants traded against experimenter-set supply and demand. One study varied those conditions between rounds, while the other held them steady for periods of time.
Both studies reportedly found that the tax reduced trading volume and a laboratory measure of market efficiency. One also found that prices continued to converge toward the experimenter-defined competitive equilibrium. The author cautions that this measure of efficiency may not translate directly to real markets and asks for field evidence or financial engineering simulations. The response notes that some European venues charge fees on unexecuted orders after an order-to-trade threshold, and raises a design problem: how should fees distinguish a long-lived cancelled order from many rapidly entered and cancelled orders? The discussion offers no empirical market estimates or simulation results beyond the cited laboratory findings.
Key ideas
- A bid or ask tax charges for submitting a limit order even when it does not execute.
- Two cited laboratory studies found lower volume and lower measured efficiency under the tax.
- In one study, prices still approached the competitive equilibrium despite the tax.
- Laboratory outcomes may not predict market-wide effects in live financial markets.
- Fee design must address differences between long-lived orders and rapid order churn.
Tags
Full text
# Quantitative before/after or financial engineering studies of a bid or ask tax? # Quantitative before/after or financial engineering studies of a bid or ask tax? Has anyone in the quantitative finance or financial engineering community studied the effects of a bid or ask tax with actual or simulated data? If so, what were the quantitative results or predictions for the market? The kind of data I am curious about could arise as part of a before/after study of a market adopting such a tax, or it could arise as part of a financial engineering simulation comparing various form of possible but as yet unimplemented market taxes. Definitions: A bid or ask tax is different from a transactions tax or stamp duty in that it is payable even on unexecuted limit orders. In other words, submitting a buy order or sell order will create a tax cost even if the order is not accepted by a counterparty for execution. For example, if the market is trading at $20-$20.01, and you send in a limit order to buy at a lower price (e.g., $19, $20) and because the market moved up the limit order expires untraded at the end of the day -- you still owe the tax. The tax might be small per limit order but would add up across participants, especially because it is owed even if someone doesn't successfully transact. One research issue has been how it would affect behavior. I am aware of two quantitative studies from the experimental economics literature: Jamison and Plott, “Costly offers & the equilibration properties of the multiple unit double auction under conditions of unpredictable shifts of demand & supply” Journal of Economic Behavior and Organization, V. 32, (1997) and Noussair, Robin, and Ruffieux, "The effect of transaction costs on double auction markets" Journal of Economic Behavior and Organization, V.36, (1998) The studies are similar. In a human subject laboratory, supply and demand curves are established by secret limit orders from the experimenter to the subjects and the subjects are allowed to trade with each other on a computerized system. Bids, asks, and trades on the laboratory exchange are recorded. The studies differ in that Jamison and Plott moved around the supply and demand parameters between repetitions of the experiment whereas Noussair, et.al. held the parameters constant for a while. In both studies, the addition of the bid and ask tax lowered transaction volume and reduced a common measure of market efficiency that is measurable in laboratory markets [but perhaps not in the field]. Noussair also reports that even with the lower volumes and efficiency, prices still converge to the competitive equilibrium. The competitive equilibrium in the lab is determined by the intersection of the supply and demand curves set by the experimenter. While these laboratory results are interesting, I am curious whether there are any quantitative results from financial markets or financial engineering studies seeking to advise governments or exchanges about policy. Has anyone in the quantitative finance or financial engineering community studied the effects of a bid or ask tax with actual or simulated data? ## Answer by lehalle (score 1) https://quant.stackexchange.com/a/3292 It seems that you speak about a "quote tax". In Europe some markets now ask for fees for orders even if they are not executed if you exceed an order to trade ratio to prevent them from cpu-harrassing flows. Let s wait and see what will append. The effect should be similar to the One of a tax. The real question is: how can you make the difference between an order inserted, not executed, but cancelled after 3 hours and a sequence of 60 orders inserted and cancelled after 3 seconds, 60 times more fees/taxes?
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.