Tonight's topic is trading illiquid stocks. I regard a stock or bond as illiquid if dollar volume traded is low relative to the market capitalization of the security. Why would I want to own such a security? Neglect. There are few eyes analyzing a security that cannot be sourced in size. Thus, if you have a strong opinion on the security, and it is neglected, you have the opportunity for some strong returns.
In the portfolios I manage for clients, roughly 10% of the assets are illiquid, as I define it. I have generally had good results with neglected stocks, but many stocks are neglected for a reason, typically a management team that is unfair to outside passive minority shareholders. Who wants to put cash into a "roach motel?"
In dealing with illiquid stocks, I typically use discretionary reserve orders. A discretionary order displays the price at which you want to buy or sell, and what degree of latitude you are willing to compromise on the price. Here's an example:
Bid: 10.40 Ask: 10.50
A discretionary order might say, "Buy 1000 shares @ 10.39, with 0.06 latitude." Any sell order flashing through at 10.45 or lower would trigger the discretionary buy order.
A reserve order says, "I want to sell a large number of shares, but I only want to display a small number, to avoid scaring the market." It could be "Buy 1000 shares, show 100 @ 10.40." If someone comes in with a larger order than 100 @ 10.40, it will execute up to 1000 shares. Makes you look small to everyone but the market makers and specialists, who know how much you really want to do.
And that is a difficulty, and I learned it the hard way — I tried a trade offering 20,000 shares and showing 100 on a NASDAQ stock, and the market for the stock went down on what was an otherwise calm day. Other retail investors couldn't see my size, but the market maker could, and he adjusted his bid down considerably. So now I break up my trades, and don't exceed a certain normal size for the liquidity of a given stock.
A discretionary reserve order combines the two. For less liquid stocks, I use it frequently, because it erases some of the advantages that high frequency traders have. My example would be: "Buy 1000 shares, show 100 @ 10.39, with 0.06 latitude."
So for an illiquid stock I was buying lately, I was doing the following (using the same market as before):
- Buy 1000 shares, show 100 @ 10.40, and,
- Buy 1000 shares, show 100 @ 10.39, with 0.06 latitude
The idea is to be ready to buy the sell orders that flash through the market at 10.45 or less, while being ready to buy shares cheap if someone decides to lift the bid. The combination allowed me to buy 20% of the volume across two days, and not budge the stock price — it declined until the seller exhausted himself, after which, the stock started to rise, and I could get nothing more done. I got 80% of the position on and will wait for a better opportunity for the balance.
Moving from Micro to Macro
So I have my own techniques for coping with market complexity. I don't complain about high frequency trading [HFT]; I fight back. I don't trade much, so trading is not a core way in which I add value. That said, I do what I can to minimize the total costs of trading, which is a balance between paying up and getting the stock, and not paying up and the stock gets away from you.
As this article points out, HFT is the result of squeezing market makers and specialists through decimalization and a few other regulatory changes. Intermediaries lost margins and large traders lost the ability to do large trades without revealing their intentions to the world. Thus in a free market, alternative execution sites come into existence. Providing liquidity deserves some sort of reward, because it is good. HFT is one way for a liquidity provider to get paid, even if some of the tactics are questionable.
The competition to trade the same stocks in different venues led to complexity, such that data exchanged across execution venues must all run on the same clock. If that is not so, it can lead to the "Flash Crash," or the recent shutdown at NASDAQ, which ostensibly happened to avoid a repeat of the "Flash Crash."
Complex systems require some degree of cooperation or regulation in order to operate well. I have suggested that we don't need microsecond liquidity to make things work. Liquidity once per second, or even once per minute would do. Having one central order book for each stock also would do it, together with tiny fees on orders that get cancelled.
Willingness to offer liquidity deserves a reward, and the more so for orders of size and duration. They act similarly to what market intermediaries do, so let us come to a market system where such offers of liquidity, whether to buy or sell, get properly rewarded. That might bring stability to the markets, together with a reduction in complexity.