A call to periodic auction markets

Professor Budish,

I’ve read your papers on auction markets with increasing conviction that you have a fundamental understanding that when it comes to the exchange of value for value, between market participants, a human timescale might be a reasonable goal for all manner of exchanges.

I’ve long held that the speed at which equity, futures and even foreign exchange markets are executed is one which benefits no one but the exchange agents and society none at all. When it comes to trading anything, value for value, that such transfers need only ever occur at human conscious frequencies. I’m curious as to what your opinion of such a position might be.

When humans transact in any exchange, outside that of the financial markets, they do so at humans speeds. When you buy a car, it make take days or hours. When you buy a coffee, a minute perhaps. When you sell an item on eBay or Craigslist, hours to days. And even, when, as an investor, you move to purchase or sell shares in an ETF or equity, you do so with hourly, daily, if not weekly deliberation. Humans trade at human speeds.

Why so must we be told that the financial markets must transact at microsecond frequencies? Are not all of these transactions done for human purposes? What overarching body dictates that when it comes to Wall Street, that we must abandon our human heritage and be forced to do business at speeds which, frankly, only benefit Wall Street?

Your theories on auction technologies applied to markets mirrors beliefs I’ve been developing for some time, years and years. Specifically, that periodic auctions can allow fair value to be exchanged, that price discovery can be done at human understandable frequencies and that the need for technology, beyond that required to execute the auctions, record the transactions and perform the clearing, is not required.

By embracing auctions executed periodically at daily, hourly or if necessary, minutely frequencies, we can cleanse the markets of those whose sole industry is the extraction of, as you say, “rent”, from the machine that is the market exchange. Like grist from a mill, traders sap friction produced income which benefits only them. True investors, those whose intentions are to act with future purpose, have no need to feed these parasitic traders. Periodic auctions can eliminate such so called jobs in the finance sector; sending those, with no doubt high intellect, back into society where they might enjoin society benefiting occupations.

My core tenet is just this, at what cost to society does the furious frequency of the finance industry operate? What benefit, across all of society, is provided by this sector? And is it not in society’s best interest to guide the practice of exchange, value for value, toward a more reasonable frequency? Using a periodic auction, I believe, is the key.

My ideas are radical. I admit. I hope however, you might have insight that may parallel them, however slightly.

Many thanks for any reply.

One thought on “A call to periodic auction markets

  1. Periodic Auction Exchange (PAE)


    *Price good for duration.

    Following a periodic orderbook matching event (OBME), the matched price shall be tradable until some time prior to the next matching event. That is, if the OBME arrives at the price of X, then that price shall be tradable, buyer and sellers submitting orders into the book at that price, will be matched and executed at that price, for some amount of time prior to the next matching event. Such orders are effectively market orders which may sit in the book at the going price until opposite side market orders are submitted.

    Rational: The price is set for the period. That’s the going price. Why restrict trading from occurring at the agreed upon price. If trading can occur without disrupting the operation of the periodic auction, then there should be no need to block it. “The price for a loaf of bread is $3.00 — all day long. Tomorrow it may change. But for today, it is $3.00.”

    *Open order book duration

    A market’s order book will remain open for entry until some time prior to the next OBME. As one of the goals of the periodic auction exchange is to reduce intentional market manipulation, Suspending order entry or cancellation, some time prior to the matching event, is necessary to ensure a low order churn prior to matching.

    Rational: There is an increase in information that may be available the near time gets to the OBME. This information may be asymmetrical in delivery, aggressive agents may acquire or access and use this information in a manner that detracts from the goal of the periodic auction exchange of an equitable trading environment.

    *Order book transparency

    Dark or lit?
    Dark would be more equitable to investors.
    Lit would be more lucrative and inviting to traders.

    *Spread limit (maximum / minimum order price)

    Each order submitted to the PAE will be augmented with a spread limit. This is the maximum spread the trader will allow their limit price to drift during the matching event. Once the match price is calculated, the midpoint, difference between the best bid and best offer determines the spread. Half of the spread is then added or subtracted from the top of the book prices. To each order in the book the midpoint price is tested. If the midprice is within the order limit price plus (minus) the order’s spread limit then the order is elected to participate in the matching event.

    Rational: traders should be able to set the the maximum spread they would pay on top or less than their limit price. If the match price is outside this boundary, such orders should fail to join the matching event. “I want to buy at 100.00, but I’ll accept a sale at a max of 100.05”

    *Sweep limit

    The sweep limit is the volume adjusted depth of book to which orders will be examined for inclusion in the current matching event. If the top of the book price is set to X, then orders of lower bid prices down to X – Y will be tested to join the matching event. This number is volume adjusted to allow for a deeper book sweep on low volume, and a more shallow sweep the greater the volume.

    Rational: on low volume, investors who want to get their orders executed, may elect to post low bids (high offers) but with increased spread limits. Without the means to sweep deeper into the book, to some calculated level, such orders would not get executed.

    *OBME examples

    Example Zebra:

    Trader A : books order to buy 1 share @ 99.95
    Trader B : books order to sell 2 shares @ 100.05

    At the periodic orderbook matching event…

    The top of the book is scanned and the bid and offer prices are determined:
    • bid 99.95
    • offer 100.05

    The sizes for these two prices are reconciled:
    • 1 share is common for both sides.
    • Matching price is calculated
    ( 1 x 99.95 + 1 x 100.05 ) / 2 = 100.00
    • The spread limit for each order is checked
    • As the spread limit for each order was greater or equal to 0.05, then the match was accepted.
    • The trade price is set and the cross for both orders is executed.

    Example Yarrow:

    Trader A : books order to buy 1 share @ 99.95 spread limit 0.05
    Trader B : books order to sell 2 shares @ 100.05 spread limit 0.10
    Trader C : books order to buy 1 share @ 99.90 spread limit 0.10

    At the OBME:

    The order to sell for trader B is divided between the two bid prices based on the price and spread limits:

    A: spread limit order set to 100.00 (+1)
    B: spread limit order set to 100.00 (-1)
    B: spread limit order set to 99.975 (-1)
    C: spread limit order set to 99.975 (+1)

    1 trade at 100.00
    1 trade at 99.975

    Trade price set to : 100.00 + 99.975 / 2 = 99.9875

    Because 99.9875 is within all the traders’ spread limits, all trades are executed.

    Example Xena:
    Trader A : books order to buy 1 share @ 99.95 spread limit 0.05
    Trader B : books order to sell 2 shares @ 100.05 spread limit 0.10
    Trader C : books order to buy 3 share @ 99.90 spread limit 0.10
    Trader D : books order to sell 2 shares @ 100.15 spread limit 0.10


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s