Brown, Chasing the Same Signals

If you believe that books should be written by people who know how to write, you’ll have a tough time with Brian R. Brown’s Chasing the Same Signals: How Black-Box Trading Influences Stock Markets from Wall Street to Shanghai (Wiley, 2010). The book suffers from a lack of structure, it is repetitive, and it has numerous factual and typographical errors.

Nonetheless, it is a useful book for those who are trying to peer into quant black boxes and, even more, for those who are interested in the interrelationships among market structure, algorithmic order execution, and liquidity. Brown introduces the reader to the world of high frequency trading and to some principles that inform hedge fund strategies. He does this from the vantage point of an insider who worked for Morgan Stanley as director of pan-Asia systematic trading and for Trout Trading Management, researching and managing statistical arbitrage strategies.

Brown claims that the market data metrics most commonly monitored in quantitative strategies are volatility, spread, and volume. In searching for price anomalies quants look for deviations from normal metric readings. An increase in volatility in an individual stock, usually viewed as the result of shifting supply and demand and/or price uncertainty, is taken to be a signal of increased risk. But on the level of market structure, Brown writes, there is another cause of greater volatility—higher transaction costs, which discourage short-term speculation and hence suck liquidity out of the markets.

The bid-ask spread also provides important information to quantitative traders. Traders are accustomed to shifts in the spread during the course of the day depending on the level of trading activity, but “an unusual movement in the spread can denote the beginning of a price rally or a reversal.” (p. 111) Spreads can also dramatically widen out en masse during market crises as market participants worry about underlying risks and the cost of providing liquidity. A review of the tape during the flash crash speaks volumes on this score. Brown writes (well before this particular event): “Algorithms are designed to minimize market impact but they largely depend on forecasting liquidity. When liquidity is weak, algorithms may respond unpredictably to the adversity. Price impacts can be rapid and severe.” (p. 103)

Consider, for instance, the practice known as pinging the book. High-frequency traders submit orders to an ECN; if they are not filled within 60-80 milliseconds the orders are cancelled. There’s a sleazy side to this practice, but let’s not go down that path. The point is that high-frequency traders are testing the waters, searching for liquidity. If they don’t find it, they don’t provide it. They are not magnanimous.

Yet Brown claims that “the essence of a black-box firm is to be a liquidity provider. . . . Before the financial crisis, the black-box influence on the world’s equity markets was observed with historical lows in volatility, dispersion, and spreads. The frictional conditions for long-term investors had never been better.” (p. 176) For instance, statarb and market-neutral strategies dampened down market volatility and dispersion (and hence over time became less profitable). But we can’t conflate plain vanilla hedge-fund strategies with high frequency strategies. They have very different risk profiles. And they will impact markets in very different ways.

Brown’s book is not a model of tight reasoning. Its strength is that it offers up hypotheses from various vantage points that might improve our perception, perhaps even regulation, of the brave new world of the financial markets.