Streetwise Professor

Exchange Traded Funds (“ETFs”) are attracting a lot of attention given (a) the CFTC’s clear signaling of its intention to impose limits on these funds, and (b) the reactions of fund issuers to this signal, most notably Deutsche Bank’s announcement that it was closing down some ETFs altogether rather than operate under the anticipated position limit regime.

So what’s the big deal?  As with most discussions of the issues surrounding speculation in commodities, the critics of ETFs are maddeningly vague in their criticisms, and don’t provide any evidence of a demonstrable deleterious effect.  The arguments seem to be that big must be bad.

Case in point.  On CNBC this morning I heard an oil trader say that the UNG fund (which trades natural gas futures) is a problem because it holds 30 percent of the nearby contract; routinely liquidates that position as expiration approaches; and that the liquidation of such a big position must be disruptive.

Call me Alfred E. Newman, but in response, I’m thinking: “What?  Me worry?”  This seems like something that the market is perfectly capable of diagnosing and addressing without any meat cleaver regulatory restrictions.

First, note that the operators of funds have strong incentives to minimize their disruptive effect.  If their liquidations are distorting prices, that is costing the funds money.  They would be moving prices against them–inflating the prices of what they buy and deflating the prices of what they sell.  They have an incentive to trade to minimize that impact.  After all, they have to compete for investor funds, not just with other ETFs, but an immense array of other investment vehicles.  If they become so big, or trade so badly, that their execution costs balloon, their performance relative to that of alternative investments will decline, and they will have trouble attracting investment.  So, they have an incentive to mitigate their price impact.

Second, and relatedly, to the extent that a given fund becomes so big that despite its best efforts its liquidations are messy and its execution costs are high as a result, that will serve as a natural constraint on growth.  Put differently, if these execution cost-related problems create diseconomies of scale, again in a highly competitive investment marketplace, the size of these entities will be limited naturally.  It means they have an efficient scale, and competition will constrain their ability to expand beyond that efficient scale.

All sorts of markets for all sorts of things have natural mechanisms for ensuring that firms operate at efficient scale.  Why should financial markets in general, or ETFs be any different?

Third, the CNBC critic and others who point to the large liquidations of big ETFs fail to recognize that there is a natural set of traders who are ready, willing, and able to buy what the ETFs want to sell when they roll: these are the traders who sold short in the first place in order to allow the ETFs to buy.  They almost certainly don’t want to make delivery, and are looking to liquidate/cover their expiring shorts just as the ETFs are looking to liquidate/cover their expiring longs.  That is, the ETF doesn’t need to find a new set of buyers to take its positions.  There is a natural set of buyers out there; the people who sold to them in the first place.  And wouldcha know it, they just happen to want to buy the exact same amount of stuff that the ETF wants to sell.  A match made in heaven.

The main potential problem is how to coordinate the activities of the ETFs and those they want to sell to at expiration.  If the ETF starts to sell when the natural buyers aren’t around, there can be a temporary price impact as liquidity suppliers must bridge the gap until the natural buyers turn up, and the liquidity suppliers demand compensation for the risks that they bear as a result.

But markets have myriad ways to facilitate this coordination.  Indeed, coordination is what markets are all about.

One way to facilitate coordination would be through block trading facilities.  These have been around in equity markets for a long, long time.  They have started to make inroads into futures markets, but exchange rules and government regulations have been less favorable to block trading in futures than in equity markets.  So, one response would be to put rules and regulations in place that make block trades easier.

Another way is through sunshine trading.  It’s been known for a long time (basically, since the mid-1980s when portfolio insurers used the strategy to reduce their execution costs) that a large, uninformed trader who is buying or selling for liquidity-related reasons, rather than because he has private information, can reduce execution costs by announcing his purchases or sales in advance.  If potential counterparties believe that the trader making the announcement is indeed uninformed, then (a) they will not charge transaction cost (in the form of price impact) to compensate for the possibility of trading with somebody with better information, and (b) they can make sure that they are there to take the other side of the announced trade, thereby ensuring that there’s a lot of liquidity available.

Yet another way is to develop crossing networks that don’t contribute to price discovery, but merely cross offsetting positions at some price taken from an exchange, e.g., the settlement price of NYMEX NG futures. Crossing networks are another way that uninformed traders can credibly signal that they are trading for liquidity reasons, and don’t want to have a price impact.  Stock crossing networks have been around for decades.

An alternative to a crossing network is a call market, which is essentially a one shot auction.  So, for instance, NYMEX could set up a call market in the expiring nat gas or WTI contracts.  The ETFs would submit offers to sell, and knowing that the ETF would be there, its natural counterparties could submit offers to buy.  (Since the ETFs look to roll positions, and hence want to buy the next month when selling the expiring contract, it might be desirable to have a call market in spreads.)

Still another way is to use cash settled contracts.  These contracts use “derivative pricing”–they take the price from another market, such as NYMEX, to determine the final value of the contract.  These contracts are essentially “self liquidating”; they don’t require a trade to close out the position.  The position closes out automatically by reference to another market price.

Indeed, UNG has already moved in this direction, using cash settled nat gas swaps as well as nat gas futures to achieve its exposure to natural gas.

In sum, there are many ways to mitigate price impact even of big funds.  Even if a fund controls 30 percent of the open interest in a contract and wants to liquidate, well, damn close to 100 percent of the opposite side of that open interest wants to liquidate too.  There is an almost perfect double coincidence of wants.  It’s just a matter of coordinating those traders.  There are many market mechanisms to do that, and market participants have the incentive to find the efficient mechanism–or more likely, the efficient combination of mechanisms, including efficiently scaled funds, the use of cash settled contracts, crossing networks/call auctions, block trading and sunshine trading, and stuff I haven’t thought of–to achieve this coordination and mitigate price impact.

The key thing to remember is that somebody internalizes the costs of these alleged liquidation-related disruptions.  As a result, that somebody has the incentive to use mechanisms that minimize that impact, and entrepreneurs have incentives to devise new mechanisms that further reduce the impact–if regulations or legal impediments don’t prevent that from happening.

One last thing.  The market share numbers, e.g., the 30 percent number mentioned by the CNBC talking head, don’t bother me in this context.  A big percentage translates into a big position which could pose a manipulative threat if–IF–the party holding it DOESN’T liquidate enough positions, but demands excessive deliveries.  That is, a big trader that liquidates his entire position–which is a big ETF in a nutshell–does not pose a manipulative threat.

So, let the market develop and employ the mechanisms that facilitate the accumulation and liquidation of large positions that individual investors obviously want to hold.  (Note that they are willing to pay premiums to get into some big ETFs.)  The challenge posed by ETFs is nothing new, and markets have figured out ways to address such challenges efficiently for years.   There’s no real need for the regulators to worry their pretty little heads over this one.