|Sounding ETF Smoke Alarm|
|21 October, 2011|
[This article previously appeared on our sister site, IndexUniverse.com]
Of all the criticism of ETFs swirling around in the past year, none has created quite as much of a stir as the two white papers from the Missouri-based Ewing Marion Kauffman Foundation, the world’s largest nonprofit organisation devoted to entrepreneurship.
So, when IndexUniverse.com's Alex Ulam caught up with Robert Litan, the foundation’s research director and co-author of the two papers with Kauffman’s chief investment officer Harold Bradley, it was a chance to hear more about what Kauffman thinks is wrong with ETFs.
The subject was all the more topical given that their telephone conversation took place on the same day that Bradley testified at a Senate hearing examining ETFs and their potential risks. What’s the takeaway? While they admit they haven’t yet found a “smoking gun”, they have a John Bogle-esque aversion to all the ETF trading that takes place, and aren’t backing away from their sense that ETFs pose a danger to the funding of young companies and to the financial system itself.
Ulam: Your views seem to be getting a lot of traction in the media. There was a recent article in The New York Times about a big market uptick during end-of-the-day trading. Some analysts that the Times interviewed said the late-day gains might be linked to trading in ETFs.
Litan: We call it the “flash-up” as opposed to the “flash crash.” Last week, when the market rallied like 400 or 500 points in 20 minutes, we thought that ETFs probably had something to do with that too. The larger point here is that we don’t think it is a coincidence that we have had such extreme market volatility at a time when ETF trading and the number of ETFs have been proliferating.
We know that there are lots of other factors that can explain volatility. There have been budget crises here and in Europe, so there is a lot of background nervousness in the market. At the same time, what ETFs have done—and we pointed this out in the testimony—is make it so easy to get in and out of the market in a nanosecond. And you are not able to do that with a mutual fund. But you are able to do that with a good portion of the market with an ETF. That is why we think it is not a coincidence that we have seen so much volatility.
Ulam: How did Bradley and you get interested in the ETF issue?
Litan: Here at the foundation, our main mission is promoting entrepreneurship—promoting policies that will facilitate companies going public. Through anecdotal conversations with people in the market, we picked up chatter that companies that were about to go public were not only nervous about the cost of Sarbanes-Oxley [the 2002 Sarbanes Oxley Act], they also were nervous about the market volatility of their stocks once they went public.
And we started doing some more thinking about this, and we formulated a hypothesis, which we will admit has not been statistically validated. But we think that there is at least smoke, if not fire. ETFs have become the tail that wags the dog for small-cap stocks. And a small-cap stock is not like a big stock like Microsoft. The prices of stocks like Apple and Microsoft dictate what happens to the S&P 500, whereas with small-cap stocks, it is the other way around: It is the ETF transaction that determines the price of the underlying stock. That is because a huge amount of the activity is in the ETF and not in the stock. We think this is one of the reasons you are seeing volatility, especially in small-cap ETFs. So this is very likely why companies were not going public.
Ulam: You also have expressed concerns about ETFs and correlations between prices of different stocks.
Litan: We have looked at the very high correlations of ETFs to each other. And the movements of stocks also are increasingly correlated with each other. And this is very unusual because you would expect stock movements to be correlated when everyone is panicking. But the market is not in a panic mode right now, and yet we have very high correlations. And so, we draw an inference that one reason for these correlations has to be that people are trading in and out of the market with ETFs. That would explain why all the stock prices would be very highly correlated.
Ulam: How would you propose regulating ETFs?
Litan: There are several ideas that we have advanced. On the small-cap problem, you could do one of two things. You could first make it illegal for an ETF to have small-cap stocks in it. That is obviously a more extreme thing to do. But a less extreme thing would be to say that a small-cap ETF cannot have a small-cap stock in it unless it pays the company for the right to include it in the index. Because as it is now, you go public and anyone can put it in an index—basically you lose control of your stock. One argument is that the decision must be at least left up to the company rather than to the ETF sponsor.
Ulam: But then you would have to do that for all companies?
Litan: The small-cap problem is a small-cap problem. ETFs are not driving the price of Apple or Microsoft.
Ulam: So how do you propose regulating large-cap stocks?
Litan: Right now there are circuit breakers for individual stocks that were enacted after the flash crash. At a 10 percent price movement, there is a trading halt. We argue in the testimony for tighter circuit breakers for ETFs—maybe 3 or 4 percent, because a 10 percent movement for an ETF means a tremendous movement for a lot of the underlying stocks. So we think that applying the same 10 percent movement rule to an ETF really is not right; you should probably apply a much more restrictive rule, and that could reduce some of the extreme volatility of the ETF movement. I don’t know what else to do about this. We are not going to eliminate ETFs, but I think we can at least limit our susceptibility to ETFs’ wild swings. The circuit breakers could flip in at 3 or 4 percent.
Ulam: Does the SEC seem like it is receptive to your suggestions?
Litan: I think that it is time for the SEC to be doing some fundamental research into this and also to be doing a fundamental rethink about their approach to market volatility. We just see enough smoke. I am not going to say that we have a smoking gun. I am not going to make that claim. But there is enough smoke out there that people ought to be concerned about this. Another point, which is not in our testimony, is that there are ETFs out there that are composed of derivatives.
That means that the ETF is dependent upon the soundness of the underlying counterparty. And that begins to sound like CDO [collateralised debt obligations] —and I don’t like to invoke the “CDO” word, but that is basically what CDOs were. You have CDOs and synthetic instruments. I get worried when you start building derivative on top of derivatives because ultimately all ETFs are derivatives.
Ulam: Why do you think that all ETFs are derivatives?
Litan: Because all of their value is derived from the underlying stocks. As a matter of technical precision, they are a derivative because their value is derived from something else.
Ulam: How does that make an ETF different from a mutual fund?
Litan: In that respect, they are still like a mutual fund. But a mutual fund is not like an ETF in that respect because you cannot trade it all the time. You can only trade it once a day. One of the reasons that ETFs have become so popular of course is that index mutual funds cannot be traded any more than once a day. And a lot of people want to get in and out of the market as a whole on a lot more timely basis.
Ulam: At the close of his testimony, Mr. Bradley was talking about the fail rate of ETFs. He mentioned a 7 percent fail rate for ETFs versus a 0.06 percent fail rate for general stock issues.
Litan: The data is there, and it shows rising fails. And that problem can be solved. The SEC can impose fines. And by the way, there used to be a settlement failure problem in the Treasurys market, and also in the mortgage-backed securities market. And there have been fines imposed on failures to settle in those markets, and those problems have dramatically declined. That needs to be done at the SEC where there is a three-day settlement rule. Basically if people don’t settle within three days they get fined, and that will stop that problem.
Ulam: Don’t registered market makers have six days to settle a trade, if necessary?
Litan: Whatever it is, whatever the rule is, they are failing. Let’s say they don’t have to deliver the securities for six or seven days or eight days or nine days or 10 days; it means that they got free money for six or nine days. So why not do it until you get stopped? And the chart we showed during our testimony shows that the rate is going up and up, which means that more people are doing this.
Ulam: Does that mean that they are failing completely, or that they are getting settled at a later date?
Litan: It means that there is lax enforcement of the settlement, period. And people are gaming the system. They are not settling on time, and that is the definition of a settlement failure. That gap in settlement in good times is one thing. But let’s say I am a sponsor in an ETF and I sold it at $100, but I haven’t delivered it for six days. And let’s say that there is a short squeeze and all the short-sellers on an ETF all of a sudden decide to cover their ETFs and it sends up the price of the stocks that underlie the ETFs. That means when I go out and buy the stocks that underlie the ETFs, I have to pay a lot more. I have to go out and pay $120 for the stocks that I sold for $100.
Now if I am an ETF sponsor and I am weak for any number of reasons, I could go down the tubes if I have enough obligations out there. And you could have institutions, custody banks or ETF sponsors, and they could be in trouble. That is why we have settlement rules. We don’t want the settlement lag to be out there for long, because we want to make sure that the cash and the securities get delivered in a timely fashion so we don’t have this problem of the guy not being able to make good on his promise. All I am saying is that rising settlement fails is an indication of potential systemic risk, and that is not good for the system.
Ulam: Why are we having more fails with ETFs than with regular stocks?
Litan: I think that the rules are not being enforced. Since ETF trading now has become so significant and the creation of ETFs has become a big business, I think that the more people are creating ETFs and gaming the system, the more money they can make until someone tells them to stop.
Ulam: What do you say about the fact that many investors have done very well investing in ETFs?
Litan: I am not indicting ETFs and their value to investors, because they are of value, and they are obviously of value to high-frequency traders. I am simply saying that there are larger marketwide impacts of ETFs that we ought to think about.
Ulam: So is an ETF a good investment for a regular retail investor?
Litan: On the whole, most of them are efficient ways to diversify and they certainly are advantageous from a tax point of view. And because you control the tax consequences with an ETF, they are clearly more advantageous than having a mutual fund that has a similar composition. For an individual investor, ETFs are very attractive.
What can we tell about the future based on historic, trend-following data?... More