Rethinking Markets

I want to talk about dark pools of liquidity, but to do so, I need to first talk about financial markets and how prices work. Bear with me for the background. The dark pools post is coming.

Prices in financial markets are not like prices in other kinds of markets. When you buy gas, you see prices posted for the various grades of gasoline. These prices are for buying gas; theoretically speaking, this is: a) a posted, fixed price – as opposed to a barter or negotiated price); and b) this is a market-taking rather than a market-making price – the price is for something that you can only buy, but not sell. You can only take what is offered, not make your own market.

By contrast, financial prices (and I’m going to use stock prices and financial prices interchangeably, though they are not necessarily interchangeable) are negotiated prices for a continuous auction. There is no single price for a futures contract, a stock issue, or a derivative. Instead prices consist of four main elements: bids, offers, bid quantities, and offer quantities. Buyers provide a price and quantity at which they would be willing to purchase a stock (a bid), while sellers provide a price and quantity at which they would be willing to sell (an offer). When these bids and offers “match” – if a buyer is willing to meet a sellers offer, or a seller is willing to meet a buyers bid – a trade will take place, until there are no longer any bids or offers at that price. If the bids and offers are not matched, there will be no trade, and the current “price” is understood to be the currently-best bids and offers.

For example, if there is a bid to buy 100 shares of Intel stock (INTC) at a price of twenty, and an offer to sell 50 shares at a price of 21, the current “price” is “twenty bid at twenty-one offer,” or 20-21. If the potential buyer and seller decide to stay at their respective prices, no trade will occur; the market will simply stay 20-21 for as long as they hold out. If, however, the buyer decides to raise her bid to 21, a trade will take place: she will buy the 50 shares that the seller was willing to offer, and have 50 shares left over still to buy. Those 50 shares left over would become the currently-best bid. The price would move to “21 bid.” If the next best offer were 22, the price would be 21-22. If the next best offer were 23, it would become 21-23. This process continues throughout the trading session, with the market constantly fluctuating depending on the bids and offers available to trade.

In this fashion, the last reported price is a useful piece of information, but much less than complete. More complete is to know the current best bid, best offer, and the quantities associated with them.

Even more complete would be to know the entire list of bids and offers behind the currently-best ones.

Why is that? Well, you could imagine that you think you’re looking at a market like the snaily one here:

If this were the case, perhaps you might say that there is mildly more shares on the buy side than the sell side, or upward pressure on the price. If you were a seller, deciding whether or not to take the best bid available (in the hopes that the share prices were falling, or simply trying to sell your existing shares at the best possible price), you might think you’re making a decision based on a relatively small bid. My willingness to buy or sell at a particular price is contingent not just on my own absolute feelings about the stock price, but also on how much I think I can get for it – which is a question about not just my intentions, but others’ intentions as well.

Because, what if you knew that the market looked like this snaily-cat one?

If this were the case, where there is relatively strong resistance at 19 (lots of people wanting to buy, or a single big buyer at 19), your decision-making process might be different. And knowing the intentions of buyers and sellers not yet in the marketplace makes a huge difference.

Each trade involves a trade-off, between being first and getting the price you want on the one hand, and signaling your intentions to the rest of the world on the other. Since most financial markets work with some version of a ‘First in, First out’ algorithm (or FIFO, meaning that at the same price, earlier orders get filled before later orders), making your intentions visible early gets you a good spot in the FIFO sequence, ensuring a better price.

But, any time you signal that you want to buy or sell something, other people can react to your desires, potentially changing their own in the process. This might get you a worse price, if you ‘tell’ other people you are a buyer, they might raise their offers as sellers. And vice versa. If you know that someone is itching to buy 10,000 shares of stock you own, you may decide to sell at a higher price. And so, better/faster price vs. signaling your intentions and getting a worse price.

How much of the total buy and sell orders are shown varies dramatically by market. Some exchanges only reveal the size at the best bid and offer, others show their book 10 or 20 prices deep. Here’s the trading screen from the Swiss Exchange:
Here you can see bids, offers, and quantities well above and below the current market price. Other places don’t show any bids and offers.

And it’s between transparency and non-transparency that the arguments about dark pools of liquidity. Dark pools are alternative trading systems that report trades but not bids, offers, and quantities. They are trying to solve the problem of large orders moving the markets too much in their execution. But they are doing it badly.