Dark Pools Can't Keep Themselves Clean
Louis Brandeis, the father of market regulation in the U.S., famously called sunlight “the best of disinfectants.” The Securities and Exchange Commission announced Wednesday that it would apply his dictum to dark pools, the electronic alternative trading systems where the size and price of orders are hidden from other market participants. The proposed SEC rules, now open for 60 days of public comment, are aimed at revealing potential conflicts of interest between the trading systems and those who use them, and clarifying some aspects of order routing in the dark pools.
The proposed rules raise an intriguing question: Do we need more regulation for markets that are used overwhelmingly by extremely sophisticated traders, and that are themselves in competition with one another? Put another way, shouldn't the market for alternative trading systems (the market for markets, if you will) be sufficient to ensure that they operate fairly and efficiently?
The question is, of course, a variant on the more general question that we should always ask whenever regulations are proposed: Given that all regulations impose costs, are the costs worth the benefits?
Dark-pool trading is big business. By volume, it may account for about a third of equity trading in the U.S. (Bloomberg Tradebook, an alternative trading system, is owned by Bloomberg LP, the parent of Bloomberg News).
You can imagine the basic arguments for and against them, which aren't my primary concern here: Large investors especially like to be able to hide their trading strategies, while critics worry about anything happening outside the view of the ordinary investor.
The best answer to the dark pools’ critics has been that no one is forced to trade in those venues. Participants know what they're doing, and are well-placed to understand and measure whatever distinctive risks they run in using them.
Taken to its logical conclusion, this defense of dark pools would militate in favor of nonregulation. Sure, there's some information asymmetry between a trader and the trading system, an asymmetry that's probably inherent to a system that matches anonymous actors.
But sophisticated investors can compare the results across a range of roughly 45 alternative trading systems, with several different ownership structures. They should thus be able to figure out which ones are right for them.
If they think they're being misled or manipulated, they can opt out or use the leverage of their participation to demand reform. In this sense, sophisticated traders are the very model of people who don’t need the protection of regulators: They can look out for themselves. This would suggest that the costs of new regulation may outweigh the benefits.
In practice, however, there are strong reasons to think that disclosure-based regulation is desirable for dark pools, despite the sophistication of their users. The most important has to do with what markets are capable of doing on their own.
In any market with repeat players, it becomes increasingly difficult to get away with dirty tricks. Yet we nevertheless generally require market participants to engage in basic disclosures.
That’s because the basic human tendency to promote self-interest is constantly creeping into any business transaction. We play our iterated games with one another, cooperating in a virtuous circle. Yet at certain points the temptation and incentive will become great to deviate from cooperation, defecting from the cooperative norm to take a big profit. When we think that profit will be unnoticed, the temptation is still greater.
The operators of dark pools are no exception to this rather basic human impulse. Cloaked in nondisclosure, they will try to get away with giving themselves (and their favorite customers) advantages when it's sufficiently attractive and the odds of detection are low.
Markets do a pretty good job of enabling sophisticated actors to monitor one another. But markets alone aren't sufficient to solve the problem of ensuring the legitimacy of markets. Markets aren't a perpetual motion machine of self-assurance.
That's why we need governance -- and government. Even the most extreme libertarian has to acknowledge that you can't run a market absent some basic guarantees of property rights. Those rights are guaranteed by the institutions of the state. As for anarchists, they usually don't care much for property rights in the first place.
The reason for requiring dark pools to engage in disclosure, then, isn’t a moral worry about equal treatment of all market participants, the kind you might have in an ordinary regulated securities market. Rather, it's pragmatic: The markets should be able to function more efficiently if the way they operate is better understood by their participants.
Otherwise, in practical terms, you'd expect the dark pools to operate in cycles of trust followed by distrust. Yes, they should in theory compete with one another to disclose more, obviating the need for mandated disclosure. In the real world, the differential power of both the dark pools and their participants makes it very difficult to overcome the collective-action problem of voluntary disclosure.
Again, that's a key reason we have government: to provide public goods that we want as individuals but that we would otherwise have trouble coordinating ourselves to produce. The transparency of the market's operation is a public good -- and sometimes someone needs to coordinate the public to provide it.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story:
Noah Feldman at email@example.com
To contact the editor responsible for this story:
Stacey Shick at firstname.lastname@example.org