I am increasingly drawn to the proposition that financial markets and capitalism are not soul-mates. They are more like fractious siblings competing for parental attention, or narcissistic partners in a rocky marriage, each needing the other but each fearful of the other’s shortcomings.
There is a close tie between them which can’t be totally severed. Traders in financial markets need real economic activity and growth in order to have financial contracts – stock, loans, options, derivatives, insurance guarantees, etc. - to buy and sell. Finance has no social purpose other than gambling unless there are related economic transactions to use the money which is bought and sold in financial markets.
And, reciprocally, productive capitalism needs cash and loans. You can’t grow an economy without equity investment and credit. That was one of the big lessons learned by the Communists: government extractions, planned expenditures, and socialized property can’t replicate the wealth creation potential of equity and credit markets responding to entrepreneurial initiatives.
But, trading in financial markets has its own gods and its own rules and they are not always wise and reasonable, tolerating “irrational exuberance” and other forms of hubris and ignorance.
For example, former Chairman of the US Federal Reserve System Paul Volcker has recommended a separation of proprietary trading for their own profit from service-oriented financial intermediation for large financial institutions, the very ones whose trading operations tanked the global credit system in the fall of 2008.
So recently, as I read of “high frequency trading” being used by those same firms, I said to myself a la Ronald Reagan: ”There they go again.”
Big Wall Street trading houses are using new algorithms and super fast computers to spot price anomalies before other investors can even blink. This technology, available to a privileged few, gives them the equivalent of inside information in trading. Is such trading a public good so that its profits are respectable?
High frequency trading uses algorithms to trade at ultra-fast speeds, often 1,000 times faster than the blink of a human eye. Software programs decide when, how and where to trade financial instruments without human intervention.
Tobin Harshaw writes “powerful algorithms – algos in industry parlance – execute millions of trades in second and scan dozens of public and private marketplaces simultaneously. They can spot trends before others investors can blink, changing orders and strategies within milliseconds.”
With so many such lightening fast trades, the average order size is falling and the number of separate trade orders is rising. On the NYSE the average order amount fell 67% from $19,400 in 2005 to only $6,400 recently. On the NASDAQ, the average order size dropped from $44,600 to $14,400 over those 5 years.
High frequency trading permits firms with massive computer capacity to track momentum moves in the prices of financial assets, buying and selling quickly to capitalize on the likely very small movement of a price up or down towards an expected equilibrium price.
High frequency traders can issue and the cancel orders almost simultaneously, confusing others as to what prices are real. And, placing orders first can get a tiny premium – but tiny premiums aggregated over millions of trades become handsome profits. Tradebot in Kansas City said in 2008 that, on average, it holds a stock for 11 seconds. Many high frequency traders end each day with no position in the market, having sold off all buys. They thus provide no sustaining support to the prices of the financial instruments in which they trade. They assume no equity risk of enterprise. They are merely parasitic profiteers and not owners.
Computers detecting via their flash orders that there is an interest in a stock issue tiny orders at a price below a buyers limit price. If that tiny offer to sell is accepted, the computer issues additional tiny offers at incrementally higher prices until the buyer’s limit is reached. Now knowing what the buyer’s limit price is, the computer offers a large lot for sale just under that limit, taking for itself the margin between the lower price – which the buyer would have loved to get – and the limit price.
Up to half of all stock market volume consists of such algorithmic trades. The company Tradeworx runs computers that buy and sell 80 million shares a day. Some of the biggest high frequency traders make billions of trades a day. Tradeworx computers get prices from exchanges, decide how to trade, complete a risk analysis, and generate a buy or sell order in 20 microseconds. Last month the US firm of Algo Technologies unveiled a system that can trade in 16 microseconds.
On May 6th, the mysterious “flash crash” in the NYSE may well have been both caused by and exacerbated by high frequency trading. During the ten minutes of price collapse, some 19 billion shares were traded. When prices began to plunge, operators of the Tradeworx computers entered a command to sell everything and shut down. Several of their competitors issued similar commands to their computers. Trading went into meltdown, sending chills through the financial world.
With the advent of high frequency trading, technology has created a new financial market place, one perhaps less fair to those without access to massive computing power. The new market demands new thinking as to its ways of operating responsibly in support of the economy.
There are other objections to high frequency trading.
First, it provides low quality liquidity. Unlike NYSE specialists who were required to maintain a fair and orderly market, high frequency traders can drop out at any time. Recently the Financial Times reported that among one thousand orders, some 99% were quickly cancelled.
SEC Chair Mary Shapiro spoke in early June on the “illusion” of liquidity that happens when trading is too fast and automated. She noted that in periods of high volatility investors can find it difficult to buy and sell shares. In times of market stress, liquidity can be ephemeral as retail investors become scared and cautious and flee to higher quality assets. So it is debatable that high frequency traders are proving liquidity to the general public. If they are not providing a needed intangible asset for market capitalism, what do thay do that deserves compensation?
Michael Goldstein, professor of finance at Babson College, has suggested that “speed limits” might be necessary on the highways of institutional trading. The response from high frequency traders is that “speed reduces risk” by more quickly finding equilibrium prices. But with such reflexive trading, when does the market “equilibrium” price reflect the actual value of the underlying financial asset?
Second, high frequency trading can generate false trading signals. A spike in such trading can attract momentum investors when there is really no substantive demand for the security. Such a spike then attracts options traders who start to build positions. These positions then attract risk arbitrage traders who believe there is potential news that could affect the stock to exploit.
High frequency trading determines overall market direction without fundamental or technical reasons. The defense of high frequency trading that it accelerates finding equilibrium prices strikes me as questionable. Trades executed in microseconds respond not so much to foundational reasons for valuing economic activity but more to “friction” and marginal “transaction costs” in the movement of information in financial markets. Momentary equilibrium prices that reflect miniscule adjustments don’t strike me as that important for markets to serve the needs of capital allocation in society at large.
When word of high frequency trading appeared in the New York Times on July 24th, 2009, people reacted. Two of the three US equity exchanges said they would end “flash orders” by the end of September. This came after Mary Shapiro, chair of the US Securities and Exchange Commission, said the Commission would clamp down on flash trading. After the May 6th “Flash Crash” the SEC is introducing “circuit breakers” to stop trading on the NYSE when prices move too quickly, too fast.
It is hard for me to square high frequency trading and its profitability with the equilibrium needs of a fair market that will produce prices with integrity to permit capitalism to avoid irrationality. That being so, rules and regulations to limit the market freedoms of traders may very well be in order so that private interest can’t too easily trump the common good.