February 14, 2011
Joint CFTC-SEC Advisory Committee
File No. 265–26 and Release 34-61358 File No. S7-02-10
On a cloudy autumn afternoon in 1870, the Chicago White Stockings, a team that would evolve into the present day hapless Chicago Cubs, played an exhibition baseball game against a hastily assembled gang of amateurs calling itself the Board of Trade Scalpers. It was a rout. In nine innings of play at Dexter Park, next door to Chicago's new stockyards, the White Stockings crushed the Scalpers by a score of 30 to 2, likely the only time scalpers on the Chicago futures exchanges were so convincingly restrained.
140 years later, almost to the day, U.S. Commodity Futures Trading Commission Chief Economist Andrei Kirilenko and several co-authors published a paper called "The Flash Crash: The Impact of High Frequency Trading on an Electronic Market," to date the definitive examination of high frequency market makers. What Kirilenko reported is deeply troubling for U.S. markets, implying structural instability, crashes and liquidity crises large and small, toxic quotes and price discovery in the public markets, and the uncertainty of any order's likely effect on prices. His breakthrough paper is a decisive empirical justification for reforming how high frequency market makers operate in today's markets.
Though he doesn't use the word, Kirilenko's study is the latest U.S. government agency report on scalpers. Perhaps the earliest was by the Federal Trade Commission (FTC) in a seven volume report on grain trading, published over six years beginning in 1920. Defining a "scalper" as a firm that "typically buys and sells in large quantities, expecting to hold the trade open only a very short time" and that "intends to be even as to quantities bought and sold at the close of the business day and is reluctant to carry a trade over night," the U.S. government's 1920 definition of scalping tracks what today's high frequency market maker firms say about themselves almost word for word.
While market makers in the futures markets have long fit the government's definition of a scalper, in U.S. equities in the recent past old-fashioned market makers carried inventory and committed capital overnight -- even several nights -- to smooth buying and selling pressures. They were required to post competitive quotes and to trade sparingly in the exchange markets. No more. They don't exist. They were too expensive and corruptible, and a series of well-intentioned reforms squeezed or priced them out, unwittingly opening the door to scalpers from the futures markets.
More than a few high frequency equities market maker firms today were founded by "locals" (scalpers) from the futures markets, particularly the Chicago futures markets. "Scalper" is not an endearment, so these firms gussy up by calling themselves "liquidity providers" or "market makers" or "principal traders" instead. The FTC observed the same in 1920, saying "There is a preference for designations that sound well. Nobody calls himself a pit scalper." Though "liquidity provider" and "market maker" are stripped of their former meaning in the equities markets, high frequency firms wear these designations as if they were a rented tux.
In his autopsy of e-mini SP 500 futures trading around the May 6, 2010 Flash Crash, Kirilenko describes how these firms destabilize markets. There's a tipping point in volatile markets when, in an instant, high frequency market makers stampede to rebalance their inventories, even cascading positions from firm to firm, while prices collapse. Kirilenko calls this "hot potato" trading, but it's an interdealer panic, a market maker fratricide. His conclusions extend to any volatile episode, because, as he wrote, high frequency market makers "did not change their trading behavior during the Flash Crash."
Just as there were few restraints on the fund group whose selling Kirilenko shows tipped the Flash Crash, there were few restraints on the high frequency market maker algorithms that bought from it. Markets crashed when high frequency market makers hit internal inventory limits and unloaded onto the next market maker, which then hit limits and unloaded onto the next one, and so on, driving the market down by almost $1 trillion dollars in a few minutes.
Kirilenko studied e-mini trading in the futures market, but the CFTC and U.S. Securities and Exchange Commission staff report on the Flash Crash showed the same behavior at work in the equities markets, doubtless from many of the same firms.
"Market Structure as a Joke"
The scalper's destabilizing practices are that it can quote as it pleases and trade as aggressively as it pleases and still carry the regulatory imprimatur and privileges of a market maker. As recently as the late 1990s, little of this was true in the equities markets. As the cash equities market automated, moved to decimals, deregulated and fragmented in the last 10 years, scalpers moved in with a soon-to-dominate business model, one they learned in the futures pits -- aggressive and often frenetic trading, keeping little or no inventory.
Instead of smoothing buy and sell pressures, as market makers in the equities markets were -- in theory -- once supposed to do, scalpers exacerbate or hide from volatility, as Kirilenko discovered in the Flash Crash. The same was true in 1920, when the FTC noted that in volatile markets scalpers "run with it, and they may accentuate an upward or downward movement that is already considerable," or even, fearing losses, that a scalper "closes out his trades when the market goes against him, and this practice can but tend to accentuate the swing."
Registering as equities market makers, given valuable and unique regulatory preferences and access, cozying up with exchanges desperate for business, and then let loose on the stock markets, the scalper's business model makes the stock markets structurally unstable. Scalpers make the futures markets unstable too -- the FTC observed that scalpers "themselves often create the volatility with which they are most concerned" -- but their effect is especially pronounced in the equities markets. Unlike the futures markets, prices in the stock market aren't disciplined by prices in the spot market -- it is the spot market. And unlike the futures pit markets, the public equities markets for the most part trade in fixed price/time order, with no choice over counterparties and no way to avoid scalpers who might turn to compete for the very liquidity they just extended.
The stock and futures markets differ in other ways too, if only because of the public confidence required of them. Whether the public invests in stocks has much to do with its confidence in the stability and integrity of the stock market. Whether the public buys orange juice has little to do with its confidence in the stability and integrity of the orange juice futures market. As important as the futures markets are, they aren't forums for long-term, even generational capital commitments, and the public doesn't directly invest its nest-eggs or retirement savings in them. Companies don't rely on them to raise capital for new plants, equipment and jobs.
After just the last few years of their relative dominance, these firms have had extraordinary effects on the public equities markets. Institutional volume is fleeing, while retail volume is skimmed off and shrinks. Apart from liquid stocks, spreads have increased. Ominously, exchanges are being co-opted, their traditional purpose recast as technology providers, as simple hosts to a migrant group of high frequency firms, and exchange markets, liquidity formation and price discovery are fragmenting and uncoupling.
He might have had something else in mind, but we can take it as an elegy to all of this when one senior stock exchange official was quoted in the Financial Times as saying, "Most of the world views our market structure as a joke."
A recent study found that high frequency firms post the best price at least 50% of the time in the equities markets. The study's author and high frequency firms pounced on this as strong evidence high frequency firms contribute to price discovery, more so than any other kind of firm.
The analysis and conclusion are superficial. A bid or offer has at least four dimensions. Beyond price and size, any resting bid or offer has a lifetime, and the inventory cycle or position resulting from any executed bid or offer has a lifetime. Even if it's at the best price, a bid or offer lasting a fraction of a second hasn't contributed to price discovery, and market makers use the latest technology to post and cancel thousands of bids and offers per second, even in the same stock. An executed bid or offer where the position is unwound quickly and aggressively isn't price discovery either. Kirilenko found high frequency market maker inventory or position half-lives of less than two minutes in the futures market. Some equities high frequency market makers claim as little as 11 seconds in their stocks. Market maker inventory cycles of a few seconds or minutes, enforced by aggressive trading as time or prices go against the firm, actively destabilize prices, especially so in already volatile markets.
Of a quote's four dimensions, only one has materially improved in the last 10 years. Because of decimalization, automation and deregulation, quoted spreads have improved for liquid stocks in stable markets. But quote duration is down, time-in-inventory is down, and for many stocks quote size is flat or down. This is all because, to manage costs, high frequency market maker inventory cycles are engineered down to seconds, and these firms keep their capital commitments low. High frequency firms will tell you they're like any other business except that capital is their inventory, and like any other business they make money by turning over their inventory, so they churn it as fast as they can. Frenetic trading isn't a byproduct of their strategies -- it is the strategy.
The effect of all of this is that investors looking at a quote today can't predict what the quote means. They can't tell whether the quote will be there when they submit an order against it, and they can't tell when their own buying or selling will trigger a market maker's risk threshold. If they do trigger a threshold, the market maker cartwheels from liquidity supplier to liquidity demander to compete with an investor's own liquidity needs. As it cartwheels, it can shock prices. And when events align so market makers turn as a flock, as they did in the Flash Crash, they can collapse the market.
This isn't price discovery. It's just short-term inventory management for unsupervised and risk averse scalpers.
Scalping dominates today because it's cheap, and because it's cheap scalpers can post tighter spreads. It's cheap because the scalper's liquidity is toxic. Its quotes are priced aggressively, showing tight spreads, but only for small quantities with very short lifetimes, with aggressive inventory management behind them to limit the firm's exposure. Firms fine-tune their risk models to make their inventory cycles as short as possible while preserving profits, lowering risk and costs. And since current national market regulations focus mainly on only one of a quote's four dimensions -- price -- if you're the best price, by rule everyone must do business with you, and you gain market share.
Exchange-regulated firms with longer term inventory models couldn't compete against the scalper's lower costs and tighter spreads, so over the last 10 years exchanges deregulated and became hosts for the scalpers, even cloning themselves to make new hosts. Deregulated, toxic quotes flourished and crowded out more stable quotes. U.S. exchanges today are interchangeable because scalpers determine market structure through their increasingly toxic quotes. Rather than a diverse ecosystem of market centers, with systemic resilience in that diversity, our deregulated markets are inbreds relying on the same high frequency market maker firms trading the same toxic scalper models.
As has been said, an insight from the Flash Crash is that "volume is not liquidity." A further insight is that, batted about by scalper inventory microcycles, published quotes don't represent genuine liquidity either. A best bid today isn't a bid to own shares at a price, or even a traditional dealer or market maker's attempt to provide liquidity. As much as 50% of the time it's just a firm trying to scalp a few basis points as quickly as possible. When that scalper's bid is executed, it then becomes an unexploded competitive liquidity demand, with the timer set, as Kirilenko found, at about two minutes, or even less. These destabilizing trade practices are a fundamental structural instability behind the Flash Crash.
At the November 5, 2010 meeting of the Joint CFTC-SEC Joint Advisory Committee on Emerging Regulatory Issues, former SEC Chairman David Ruder wondered whether to increase costs for high frequency market makers. He didn't suggest taxing them, as with a transaction tax. He asked whether they should face affirmative quoting obligations, "whether there should be some requirements that they provide liquidity." But along with affirmative quoting obligations, they should also face negative obligations, once common regulations limiting their ability to exacerbate price movements. Without negative obligations, affirmative quoting obligations make quotes still more toxic -- firms required to provide liquidity will trade even more aggressively to manage inventory.
Negative obligations will prevent scalper fratricides, and stop high frequency market maker firms from unloading inventory onto the firms behind them. Without that kind of "hot potato" trading, the volume sensitive algorithm that tipped into the Flash Crash would not have descended into a lethal feedback loop as it traded against cart wheeling toxic quotes. The simplest negative obligations will extend market maker inventory cycles, preventing these firms from flipping into a liquidity crisis, as they did in the Flash Crash.
In the equities markets, inventory microcycles are another reason to restore the public's priority at a price. The public's orders used to trade first at a price because professionals had time, place and information advantages over the public, but that regulation was eliminated as the exchanges deregulated. Professionals still have these advantages over the public, and today they are for sale to as many professionals as can afford them. Apart from questions of access and fairness, a frequently overlooked structural advantage of public priority is that the public's inventory cycle is longer than that of market makers. Restoring the public's priority in price queues both limits the reach of unexploded liquidity demands and recognizes the public's many disadvantages to professional scalpers.
The equity market reforms and deregulation of the last 10 to 15 years happened for good reasons -- monopoly profits were flowing to intermediaries and exchanges, intermediaries were taking advantage of their customers, innovation was being strangled by entrenched interests -- and it was time for reform. As many of us hoped, new participants, technology and business models sprang up. Nobody wants to undo that progress. By analyzing trade and position data from the futures market, Kirilenko's breakthrough was to show how a dominant class of these business models can be disruptive, and how these models can be destabilizing enough to create systemic risk as an inherent consequence of their design. On a gross basis, these models can be checked by circuit breakers or price limits, and this is one reason price limits are standard in the futures markets. In the equities markets, these models must be checked even before they trigger circuit breakers or price limits because the equities markets are profoundly different from the futures markets. The simplest way to check these models is to put reasonable restraints and obligations on them.
A basic function of any market is to produce a quote. The scalper's toxic quotes, thousands of them a second, are a hoax on our equities markets. No one planned it. It happened as an unanticipated consequence of well-meaning reforms to a flawed system. There is no competitive solution to this problem within current regulations so long as quote price is a routing table's first regulatory imperative. Competition simply forces exchanges to publish more and faster toxic quotes, as market power continues to shift from the exchanges to the scalpers.
Finally, some have pointed out that regulation didn't work in the market break of 1987, when old-fashioned specialists and market makers shirked their responsibilities and hid from the market, and regulation won't work today. Regulation didn't stop them from shirking their responsibilities, but regulation didn't excuse them either, and regulation didn't encourage them to automate a deadlier game than hide-and-seek -- intermediaries didn't exacerbate the 1987 market break by playing market maker "hot potato," a feat unique to the Flash Crash. And the logic of "regulation didn't prevent 1987, so regulation won't work" is a civic novelty. Should we apply that logic to drunk driving, or to any other misbehavior?
Of course not. Please regulate them.
R. T. Leuchtkafer