Back in the 1980s, lots of us involved in the financial markets read a book called ‘Liar’s Poker’, by Michael Lewis; Lewis had started as a graduate at Salomon Brothers, at the time probably the mightiest powerhouse on Wall Street. The book was funny, perceptive and a pretty good picture of the dealing room environment of the time. Since then, Lewis has gone on to write a number of other books, and now, recently published, he has given us ‘Flash Boys’. I’ve just been on holiday for a week, and I took it with me; the first thing I have to say is - read it. I assume most readers of this column have some connection with the financial markets and the book is directly for that readership.
Whereas ‘Liar’s Poker’ is primarily a picture of a firm, here in ‘Flash Boys’ Lewis’ subject is the US stock market and the way it has been allowed to develop in recent years. His theme is the way that the fragmentation of the market, growing computer power and the constraints of information transmission have come together to create an environment open to a form of exploitation largely invisible to conventional stock market investors. Digging trenches through the Appalachians to create a dead-straight line from New York to Chicago, wrapping coils of fibre-optic cable in a box to lengthen the run, sticking your computers at the very doors of your exchange of choice to shorten it; in the new world of High Frequency Trading, all these are more important than understanding the balance sheets of quoted companies or global macro-economics. Lewis shows us how increasing (or more correctly, changing, normally by increasing) the speed of the transmission of orders from trader to exchange can significantly change the outcome of the trade. His overall contention is that the ‘ordinary’ investor is being disadvantaged - in fact, ripped off is probably not too strong a phrase for the behaviour he describes - by a relatively small group who prey on other peoples’ orders to gain their own advantage. Without rewriting the book here, what he suggests is happening is relatively simple to understand, if extremely complex actually to operate. Instead of the US stock market being on the floor of the NYSE, as was the case in the past, it is now spread over a disparate collection of electronic centres, a mixture of exchanges and proprietary dark pools (of liquidity). Those black boxes are scattered around the country, meaning that the time orders take to reach them from brokerage offices or bank dealing rooms varies from exchange to exchange. (That bit of physics I understand: fibre-optic data transmission happens at within spitting distance of the speed of light. Despite that being very fast, there is a time involved and that time is directly related to distance.)
Taxing the Transaction
The principal game seems to be for the HFT algos to put in small orders to sniff out big parcels of stock being sought or offered; then, using the fact that they have invested in swifter (ie, shorter) routes between exchanges, they can exploit the information they have gained from that effectively to front-run the large order in the market. Now, in a way, that sounds like a simple arbitrage operation - exploiting differing prices for the same item on different exchanges. But it’s not; a genuine arbitrage works to correct anomalies, and, importantly, the arbitrageur actually carries an economic risk, albeit only briefly. In the case Lewis is describing, that doesn’t happen. The HFTs' only purpose is to get into the middle of a trade which is happening anyway, to squeeze out a benefit. The effect of the operation is to inflate the price for a buyer (or deflate for a seller) - by a very small amount, it is true - while providing no benefit to the market as a whole. This is not the provision of liquidity, it’s simply adding a layer to a trade. In the crudest terms, it’s like adding a tax to the transaction. Again, I’m not going to attempt to rewrite the book here; but read it, it presents a fascinating, detailed explanation of how these ‘extra’ trades are squeezed into the order flow. There is one caveat; Lewis is unashamedly promoting the activities of one group who have established an exchange which they believe outwits the HFTs. That helps to make for a good story.
And for the LME?
Now, is any of this relevant to the LME? Well, over the last few years I have written several times about the effect of electronic trading in general and algo-trading in particular on our market. The overall picture is clearly different from equities, in that there is not a whole raft of different exchanges trading LME products, so the flipping from exchange to exchange is not an issue (although in passing, one could suggest the existence of dark pools of liquidity…..but the definition would be quite difficult). What is interesting, though, is when we think about the concepts of volume and liquidity. Despite what some - exchange operators, perhaps - may tell you, volume and liquidity are not the same. Liquidity is what markets need as their lifeblood, because if there is insufficient liquidity, then making trades becomes very difficult. Volume is the simple number of lots traded. But if what is happening - as Lewis describes in the equity market - is that a chunk of those lots traded are effectively never available to the market as a whole, then liquidity is restricted. Are HFTs operating on LME Select in a way analogous to Lewis’ book? Is some of that volume we can see reported not truly available to the market as a whole, and does that have implications for the liquidity of the market? It’s certainly been suggested to me by a number of traders recently that such is the case. Is it a problem? Well, that’s an interesting question for the Exchange and its members to consider.
‘Flash Boys’ by Michael Lewis is published in hardback by Allen Lane and also available as a Kindle download.