HFT: the backlash continues

By Buttonwood

MICHAEL Lewis's book "Flash Boys" on high frequency trading (reviewed here) has attracted some vigorous criticism from within the industry but it has raised awareness of a topic that only seems to crop up when the market has a conniption like the flash crash of 2010. But it is worth thinking about the issue in more detail; what is the function of a stockmarket? It is surely threefold - to raise long-term capital for companies, to allocate capital to the most promising businesses and to provide a vehicle for savings. It is hard to see how high-frequency trading helps in any of those functions. The traders are not providing long-term capital, they could not care less about the relative merits of the companies concerned and they do not provide a vehicle for small savers.

What they do provide, the defenders of the industry will argue, is liqudiity. A more liquid market makes investors more willing to take part, because they will be able to sell more easily; in turn, this lowers the cost of capital for companies. HFT enthusiasts point to lower spreads as a result.

The Lewis book counters this argument by saying that spreads are not the only factor. Yes, you can sell a small amount of your holding at a low spread. But the HFTs then move the price against you; your first 100 shares go for $10 but the rest for $9.90. The market impact of the deal costs you more than the spread would have done. Unfortunately, Mr Lewis doesn't put any numbers on this to prove his point.

However, a paper from Lin Tong of the University of Iowa comes to the rescue. She finds that

An increase in HF traders’ participation rate is associated with higher trading costs for institutional investors

The effect could be $10,000 a day per institution. The problem is that

First of all, the liquidity provided by HFT may be illusory and may disappear when institutional investors most need it. Moreover, the large order sizes and potentially high information content make institutional trades most vulnerable to HFT strategies such as front running. Such strategies can dramatically increase the price drifts and market impact during the execution of a large order

It can hardly be surprising that investors suffer some kind of cost. After all, HFTs would not operate if they did not expect to make a profit, and that money has to come from somewhere. Some argue that the flaw in Lewis's book is that HFTs do not make money out of the retail investor but from the big institutions (see this review as an example). But while Lewis certainly could have been more clear on this point, I don't think it is a fatal flaw. After all, many institutions (pension funds, mutual funds, ETFs) represent the aggregated savings of retail investors. Secondly, these institutions are the ones that provide the long-term capital for companies (one of the stockmarket's main functions, as already stated) and thus we do not need to discourage them.

Other criticisms of HFTs follow up on this point. In a speech to the Fed in Atlanta, Joseph Stiglitz argues that

if sophisticated market players can devise algorithms that extract information from the patterns of trades, it can be profitable. But their profits come at the expense of someone else. And among those at whose expense it may come can be those who have spent resources to obtain information about the real economy. These market players can be thought of as stealing the information rents that otherwise would have gone to those who had invested in information. But if the returns to investing in information are reduced, the market will become less informative. Better “nanosecond” price discovery comes at the expense of a market in which prices reflect less well the underlying fundamentals. As a result, resources will not be allocated as efficiently as they otherwise would be.

Mr Stiglitz also tackles the issue of liquidity.

what matters of course is not the average thickness of the market (or even the average volatility of price) but the value of these variables at the relevant times—when I might want to sell the asset. If, for instance, high frequency trading (or capital market liberalization) leads to more volatility in those circumstances when I might want to sell my assets, then it leads to less liquidity in the relevant sense. By the same token, if in such times, the thickness of the market declines because high frequency traders leave the market, the fact that on average markets are thicker is of little relevance

In other words, HFTs are the kind of people who are willing to lend you an umbrella only when it isn't raining. A similar argument emerges from a paper from Yesha Yadav of Vanderbilt University who argues that

Algorithmic traders have powerful advantages over more fundamental traders, most notably, in their ability to decode how an informed trader is likely to transact – and to then get to this trade before anyone else. Algorithmic traders can use speed and technology to consistently “front-run” the fundamental trader. By free-riding on the intelligence of others, algorithmic traders save themselves time and money – and take home the winnings at the same time. Faced with diminishing gains, informed traders are left with little incentive to invest in research and analysis. Where informed actors see their gains systematically reduced or wiped out by swifter algorithmic traders, investing in deeper information and analysis makes little business sense. With markets growing ever more short-term in focus, traders are likely to see their gains coming from more short-term trading than in longer term investments. If long-term trading promises a poor trade-off, actors of all stripes might look to invest their limited capital in less research intensive markets.

This is the key point. A focus on short-term liquidity is not just irrelevant to the provision of long-term capital, it may impede the provision of such capital. To return to Stiglitz he points out that

Those making real decisions, e.g. about how much to invest in a steel mill, are clearly unlikely to be affected by these variations in prices within a nanosecond. In that sense, they are fundamentally irrelevant for real resource allocations. That is why we have argued that the real focus of HFT is the negative-sum game of trading with a fixed set of assets.

Indeed, this could serve as the epitaph for the direction of economic development over the last 30 years; an obsession with the financial markets at the expense of the real economy they are supposed to serve.