The NYSE Market Close... Is It A Marketing Stunt? Should It Be Illegal?

Includes: ICE, NDAQ
by: Kurt Dew


Lately, the NYSE touts its human floor traders’ informed judgment in placing orders at the market close.

Floor traders have a monopoly - a discretionary order, the d-Quote - driven by NYSE buy/sell order imbalances, info provided to floor traders only.

Question: Is human judgment necessary or useful in exploiting this information? Could a computer do better?

Another question: Why is collective knowledge by a privileged class of trader healthy in stock markets?

In OTC markets, shared knowledge of dealer order imbalances produced lawsuits settled in the $10s of billions. People have gone to jail for collecting this information.

But there are perfect moments, and the will to choose what will bring about more perfect moments. - Mary Balogh

Every major financial market has a time (or “window” of time), an instant when liquidity is dramatically increased by customers and dealers who, by custom born of experience, have chosen to trade at that moment. In some markets, that time is “close of business.” In OTC, these times when trade prices are based on a time-based average price are called "fixings." For foreign exchange, the fix occurs at 4:00 PM London time. This article describes the basis for the forex lawsuits, shared order imbalances. Similar articles for the LIBOR fixing, 11:00 AM London time, may be found here and here. Here is a description of criminal cases that have resulted in jail time for traders sharing market imbalance information prior to the OTC fixings. In these days of round-the-clock electronic trading, the close of business has become a little artificial, since transactions in most listed securities can be placed and filled 24 hours/day. But the close of business at the two major US stock exchanges, The New York Stock Exchange [a division of Intercontinental Exchange (ICE)] and Nasdaq Inc. (NDAQ), remains a significant moment. At both exchanges, the close is a major generator of profit, since each exchange retains a monopoly on the market-at-close and the limit-at-close orders on their listed stocks.

The New York Stock Exchange, in particular, markets the close of business there as a time when human judgment is uniquely significant. Here is the NYSE's description of the essential role of the close and the necessity for human judgment. By implication, since the NYSE is the last stock exchange with a human population, it follows that the NYSE is uniquely qualified to list stocks. The close and its operation have thus evolved to become the most important competitive advantage of the NYSE.

There are three orders that are drivers of volume at the close of the NYSE: market-on-close (MOC), limit on close (LOC), and discretionary orders (d-Quotes) placed by floor brokers. Since the other exchanges have no floor brokers, questions are raised.

  • Is this d-Quote a significant reason to choose NYSE for placing current orders, instead of another exchange?
  • To what degree is human judgment essential to the d-Quote?
  • Is there reason to think that the legal d-Quote represents behavior that is condemned in other trading venues?

The answers provided by this article are: First, the d-Quote is valuable to customers because NYSE allows only floor brokers access to exchange-wide order imbalances in advance of the close. But second, this valuable imbalance information would be better exploited by computers than people. And third, the pooling of inside information available to the floor trader population is a form of market manipulation according to market regulators in other markets.

Further, the human floor broker dimension of the NYSE d-Quote most likely remains a marketing stunt. An alternative exchange that provides order imbalance information for a fee, allowing any paying market participant to exploit this data, would demonstrably outperform the NYSE humans on the floor of the NYSE. Indeed, the human dimension of the d-Quote is possibly an illusion at the NYSE itself.

Of time and trading

It is essential for every successful financial market to organize time – to use time to create moments when customers expect maximum liquidity. The sine qua non of exchange success is customer belief that an order may be placed with a reasonable likelihood that counterparties will be present in sufficient abundance that a buy order of market-appropriate size, for example, will be met with sufficient sell orders at a price reasonably close to the current market price.

This article addresses the question: “What could an electronic marketplace do to enhance the benefits that the custom of trading at the close provides to market participants?” If the close of trade is a significant source of exchange profitability, why don’t exchanges attempt to improve the attraction that the close of trade presents? Why don’t the exchanges compete?

Who trades the close?

The close has a very different significance to the two quintessential market antagonists – sell-side and buy-side traders. But like the proverbial savannah oasis, every kind of creature will meet its most vital need at the market close.

What’s a sell-side trader? A sell-side trader is a market participant that sees the evolving price, volume, resting order book, and recent transaction data as information of value in placing their orders. Sell-side traders are in the business of seeking an information advantage over other market participants, in order to profit for these other participants’ shortage of valuable knowledge of current supply and demand conditions.

Other market participants are buy-side traders. By buy-side trader, I mean a market participant that takes the market price to be the best available price; placing orders with the objective of buying or selling at the current market price. Buy-side traders may be thought of as generating orders as part of a broader operation. For example, a hedge fund may have conviction about the long-term performance of a particular security. For this customer, a good fill is important, but not the primary reason for the transaction. Buy-side traders see the transaction as a way of collecting profit made through their other activities.

How the sell-side and the buy-side see the close

The sell-side sees the close as a target-rich environment. It is then that valuable information may be exploited against the greatest volume and therefore, for the greatest profit. Most sell-side traders are also brokers, since the brokers’ order flow is one valuable source of information to sell-side traders. Useful information about the price at the close is a very tasty morsel since every price advantage is multiplied by the close’s greater volume.

The buy-side, correctly for the most part, sees the close as the time when sell-side sharpers are in more direct competition with each other than usual. This intra-sell-side competition works to the advantage of the buy-side, narrowing bid/ask spreads and increasing liquidity. In a market where there is no collusion among sell-side players (contrary to popular belief, sell-side collusion is largely absent in markets with extraordinary volume) sell-side games are likely to be buried in the general rush to get orders filled.

For example, the sell-side ability to play games that take advantage of broker colocation with exchange computers is curtailed at the close since closing orders are all filled at one location. This reduces the effectiveness of any high frequency trading (HFT) games that depend on prior knowledge of bids and offers.

Trade in advance of the close

A classic buy-side way of exploiting the liquidity available at the close is to place a market order to be filled at the closing price. The historical sturdiness of this form of order demonstrates that both buy-side and sell-side believe they gain something from the existence of trading at the close.

The expected benefits to both traders and customers are well known. On the sell-side, traders that successfully collect large amounts of customer orders in advance of the close have a limited idea of what prices will do at the close. A trader possessing a very large volume of buy orders at the close has reason to believe that if she places a buy order for the house account prior to the close, the house will most likely profit from a closing price higher than the current market price. And so, the sell-side will “trade its book” by placing orders of its own in advance of the close.

The buy-side gains in two ways. First, the buy-side order at the close is hidden until executed. So, other than the routine front-running a well-informed customer expects from her own broker, she has little to fear from being scalped by the sell-side at the close. Second, buy-side order execution risk is shifted from the buy-side trader to her broker. When the at-the-close orders are dumped into the market, a fill is guaranteed. Thus there will be a buy/sell imbalance, and some buy-side orders won’t get filled.

If, for example, there are more market orders to sell than to buy, and a broker is left holding the bag with an unfilled sell order, that broker will be wise to provide the buy-side customer with a sale at the closing price, filled by the firm itself, which will inevitably be laid off at a loss. This is the down-side of being a sell-side market maker.

Could a computer improve the efficiency of d-Quote execution at an exchange close?

Of course. If a buy-side trader knows two NYSE floor brokers, one who uses her judgment and acts upon market imbalances at a speed consistent with the time it takes her to form an opinion and place an order, and another who programs her order book to execute on the order imbalance at the last moment at her computer’s volition; the buy-side trader would choose the floor broker with the algorithm, since the computer can wait longer to identify imbalances at the close and react more predictably and swiftly.

With this thought in mind, I make the following wager. I bet most NYSE floor brokers already use these algorithms and make no judgments themselves at the close. The notion that d-Quotes require a person, and more obviously, the need for that person to be in Manhattan, rather than New Jersey, is a marketing ploy counter to the customer’s interest, unless that person’s role is completely without substance. To serve a well-informed buy-side customer, the floor broker will use a computer-driven execution algorithm to optimize customer gains from the imbalance.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.