HFT, arbitrage and flash crashes

The new Michael Lewis book, Flash Boys, which focuses on high-frequency trading, is causing a bit of a stir in the financial sector. His book is now a No 1 best seller. And various investment houses and economic commentators are complaining that he is smearing perfectly legitimate fast trading as some sort of scam that rigs markets.

Lewis reckons that high frequency trading (HFT) amounts to high-tech ‘front running’. By that he means that HFT firms, using sophisticated mathematical models and algorithms, are gaining an unfair advantage in knowledge about the movement of prices in financial markets so they can get in to buy and sell a millimetre of a second before others. Moreover these high-frequency traders are getting special privileges from investment banks to trade in ‘private exchanges and ‘dark pools’ of warehoused shares and bonds to gain more advantage over the average investor.

The defenders of HFT say that they are just doing what always has been done in investment markets, using superior techniques that make them more efficient than their rivals. They make gains because they are more clever and quicker, not because they have ‘inside knowledge’ and are ‘front running’ the market.

In one sense, high frequency trading is nothing new in financial investment – it’s just the latest technical revolution in speculative financial trading and, in particular, what is called arbitrage. That means taking advantage of slight differences in the price of the same stock or bond before others notice.

Since the beginning of financial speculation, arbitrage has existed. The most famous example is that which made the legendary banker and speculator Nathan Rothschild hugely rich from the Napoleonic wars between Britain and France. Rothschild had an agent at the Battle of Waterloo in 1815. His agent saw that Napoleon was losing and rushed back to the coast, hired a boat for a humungous sum of 2000 francs through a storm to England. On getting the news, Rothschild rushed to the London Stock Exchange and acted as though he wanted to sell British shares, giving the impression that British commander, Wellington, had lost. Everybody pitched in to sell and Rothschild quietly bought them all up before the news arrived of the British victory.

Two things come from this: speedier information gives an advantage and but also allows the ‘rigging of the market’. High frequency trading is an advance in efficiency because it is speedier, but it is open to the usual chicanery of speculative investment. HFT is just another example of speculative capital engaged in arbitrage, simultaneously buying and selling two equivalent positions, or at least instantaneously as possible. Time is risk. So reducing the time between buying and selling reduces risk to the speculator: HFT has taken that to the nth degree.

Let me quote Nasser Saber, author of three volumes on, Speculative Capital, on HFT: “HFT is the adaptation to the new circumstances of old ways… when a find places an order to buy say 100,000 shares of a stock, the order has to broadcast to reach the market. But if before it reaches the market, we can intercept it and get ahead of the trade, buying as many shares as we can, that would push the share price higher, if only by a very small amount. We then sell for a profit, maybe razor thin. If we repeat this process tens of millions of times a day, our low margin will compensated by large volume… that’s HFT in a nutshell”. (see http://dialecticsoffinance.blogspot.co.uk/).

The difference between Rothschild and high frequency traders is that the former used a man and the latter use computers and software. Yes, it is efficiency and front running together – as it always has been. That is what financial speculation is all about. But then financial speculation creates no new value, it merely redistributes existing value or produces fictitious capital way out of line with real value. As such, it is not just no use at all, but positively dangerous to productive sectors.

Those with the algorithms and the maths and the technology can gain a momentary advantage over the average punter. And of course, the average punter does not have access to the technology and also to the special ‘private exchanges’ where those ‘in the club’ can take advantage of these minute differentials at speed. Indeed, these private exchanges and dark pools of stocks were set up by the investment houses to stop the small HFT operations ‘stealing’ their arbitrage gains.

Most financial capital does not go into HFT – it’s still a small field that Lewis is highlighting in his book.  Spreads on arbitrage are continually being squeezed, so volume must continually be expanded.   Fierce competition for market share among HFT firms has been driving many into new markets, geographies and asset classes. For instance, Virtu, which uses algorithms that dart in and out of positions in milliseconds, says its profitability relies on the fact that it trades 10,000 securities and other financial products such as fixed income, currencies and commodities in 210 markets and 30 countries.  Though HFT firms will make $1.3bn before expenses this year from trading US stocks, up slightly from 2013, that expected figure will be far below the $7.2bn peak in 2009, according to estimates from the consultancy Tabb Group.

And that is where disaster brews, as speculative capital and trading is always an accident waiting to happen and now at flash speeds. Long-Term Capital Management, a hedge fund, that engaged highly-leveraged speculative investing, using a ‘foolproof’ and highly ‘efficient’ risk model called Black-Scholes, infamously collapsed in 1998, losing $3.8bn and nearly taking the 14 largest investment banks who invested in it down with it. LTCM and its investors were bailed out by public money from the Federal Reserve. And there have been several examples in recent times of sharp flash crashes in financial markets probably caused by HFT.

“At some point in time the chickens are going to come home to roost on the HFT game,” said one Goldman Sachs ‘insider’. “It’s a smart move for anyone to become more diversified in their approach to the market.” Or just not have speculative capital and a financial market at all!

6 Responses to “HFT, arbitrage and flash crashes”

  1. vallebaeza Says:

    Reblogged this on Econo Marx 21.

  2. Warren Says:

    Thanks for the clear explanation Michael. I’ve read elsewhere that during the decisive 1815 battle, Nathan Rothschild used carrier pigeons and semaphore to get information across the channel ahead of rivals. The optic fibres of their day.

  3. Karen Helveg Says:

    I saw on 60 Minutes that there also is an element of fiddling with the cable distances to core centres or some such: there is a zillionth of a minute’s difference between Boston and New York in getting the news. And that has been exploited by front running.

  4. Charles Says:

    “financial speculation … redistributes existing value or produces fictitious capital way out of line with real value. As such, it is not just no use at all, but positively dangerous to productive sectors.”

    “Dangerous” – do you mean a slump of the economy when a bubble pops, or do you refer to something else, too?

  5. Geoff Says:

    I have been reading Flash Boys, which is a well written and interesting book, following Michael’s article on the arrest of the UK trader Navinder Sarao for extradition to the US. What puzzles me is why these activities seem to be legal, and carried out by most of the US banks, but this individual has been arrested?

    “The High Court has turned down an application to vary the £5m bail conditions imposed on Navinder Sarao, the trader accused of helping cause the stock market “flash crash”.”

    Can anyone help explain why his “crime” is different from all the other high frequency traders?

  6. Mo Says:

    I like that post. There is a link between traditional investments and HFT. Investors look to arbitrage.

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 )

Google+ photo

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

Connecting to %s


%d bloggers like this: