Talking ETF Liquidity, Flash Crash Fallout, And More With Paul Weisbruch

by on August 31, 2010 | ETFs Mentioned:

Paul Weisbruch is the VP of ETF/Options Sales and Trading at Street One Financial, a firm that specializes in educating, evaluating, and trading ETFs, equities, and options. He recently took time out of his busy schedule to share his thoughts on the liquidity of ETFs, fallout from the “Flash Crash,” and more.

ETF Database: There seems to be a notion that an ETF needs to trade a certain amount of shares on a daily basis–usually 100,000 shares average daily volume–and/or cross a minimum asset threshold–typically $100 million, before it is “liquid” and “investable.”  Are these metrics a valid way to screen out potential ETF investments?

Paul Weisbruch: These metrics aren’t necessarily valid indicators of liquidity, but they certainly are prevalent among the investment advisor community and especially in the institutional world. And that presents a major hurdle for ETF issuers trying to raise assets. That is especially the case for some of the upstart providers, because if they can get an institution or a high end investor RIA type to invest in their funds, that would create safety and comfort in numbers–seeing some volume hit the tape and seeing some assets flow into the funds. It is often the case of the chicken and the egg: investors are waiting for the volume to appear before they invest in a particular fund.

So we usually try to explain to investors that the underlying liquidity comes via the index or just the methodology of the fund. And there are a great number of examples of funds that are very liquid from a trading standpoint–as long as executed properly–despite the fact that they don’t trade much volume. Investors are waiting for the 100,000 share mark to be crossed on an average daily basis and/or the $100 million mark as far as AUM. But crossing those thresholds really means very little for the liquidity of the ETF—again, as long as executed properly.

For those that voice concerns about crossing those thresholds–and there are many popular misconceptions in the press and in the investment advisor community–we also advise them to address the viability of the issuer themselves. For instance, if PIMCO rolls out new products it is nonsensical to wait until those ETFs reach a certain asset threshold or wait until they reach a certain volume before they can trade. And the reason it’s nonsensical is that it is PIMCO behind the product. If it is a company that is a start-up operating in the red, perhaps that would be a valid way to screen out ETFs as far as putting them on a “watch list.”

Long story short, these types of “liquidity filters” aren’t valid due to the pricing nuances of how ETFs trade. But it definitely happens and it is definitely a major obstacle in making sound decisions when it comes to building a portfolio of ETFs.

ETFdb: You mentioned that instead of looking at the trading volume it makes more sense to look at the index and the underlying holdings. Explain why that is important in connection to “spontaneous liquidity.”

PW: It is best illustrated when there is a robust, liquid index as the underlying to a newer ETF to market that has perhaps wide bid-ask spreads. No one likes wide bid-ask spreads, whether they are nickel wide or ten cents wide or, potentially, a dollar wide. But at the same time, you do not have to trade at the bid or at the ask, and that is what investors fail to realize; they tend to look at the screens and assume that what they see is what they get. They end up crossing a potentially wide bid-ask spread and it eats into their performance.

The concept of “spontaneous liquidity” that you mentioned comes out of the fact that there is an underlying index value for every product out there. Whether it is something that tracks commodity futures or an equity basket or an inverse product, there is a tangible product to that index at all times during the day because those securities are trading in the open market. An ETF may not trade for days or weeks, depending on the popularity of the product, but that index value moves in real time with the market. Regarding “spontaneous liquidity,” investors need to be cognizant of what is in the index and potentially how those individual instruments are behaving in the given time frame they are looking at. They should be trading according to that as opposed to what they see on the screen, as far as bid-asks and a potentially quieter product.

ETFdb: Explain the role that Street One Financial fills in the market; when the order to buy an ETF comes in from a client, what exactly do you do?

PW: Our objective is to access the best possible execution price at the lowest possible cost of the customer, and part of that is examining what exactly the product is. If it is a “vanilla” product like SPY or IWM or GLD, there is a lot of activity on the screen–a lot of trading volume and rather tight bid-ask spreads. In that case, it becomes a mechanism of what the market is doing and what the objective of the investment advisor is. Some investors are very comfortable, at the time of entry of their order, getting the trade executed as quickly as possible. In that case that is generally what they look at as far as their execution report.

Then there are some who do not have a strong opinion of what the market might do. They want to establish a position in an ETF and they basically lay off the trading responsibilities on us. They are trying to beat some quality metric on the execution piece of it. So if you are a buyer and we see that the futures are coming in, we are not going to just stick a limit out there and fill the order and be on with our day. Instead, we are going to try and scale that weakness and provide a lower average price for the buyer.

Again, it takes a comprehensive understanding of what is inside of the ETF and how those underlying securities are behaving in the context of the overall market. There are times where we might use limit bids or limit offers ourselves, obviously hiding the size and trying to stealthily accumulate or sell shares.  And then any combination of transacting block trades at or near what we think the underlying index value is as well. It is a combination of smart order routing and also sourcing liquidity for larger block trades to get the order done in the most intelligent manner.

ETFdb: When does it make sense to turn to someone like S1F? If I want to buy 100 shares of SPY, I’m not going to have much of a problem executing that efficiently on my own. When does it make sense to get some assistance that may add value in the execution process?

PW: It has a lot to do with how that investment advisor currently trades. Most tend to trade through their custodian desk, whether it is a TDA or Fidelity or Schwab, and we find that custodians perform that role rather well. However, it does make sense to turn to a specialized ETF desk for orders that are time sensitive or orders that are sensitive as far as size or if you just want to lay off the trading responsibilities on a desk that is going to help you recapture basis points.

The threshold for size is probably 5,000 shares and greater where it becomes a value add if you have an ETF specialist desk working your orders for you. Most will say, “I can trade 5,000 shares of SPDRs myself.” And that is generally true, but again that goes back to the point where investment advisors are buying and selling at the point of entry. What we are trying to tell people is that you should be aiming to outperform some benchmark, whether it is the VWAP or the TWAP or just the rival price of the order.

Having said that, most of these desks are not working your orders. They are just putting them out there in the marketplace getting them filled, charging a commission, and moving on. That is where we differ as far as actually working the order with some intuition behind it–trying to anticipate what the market might do, and trying to get the best possible outcome for the customer. It seems to be from a cost standpoint that custodians tend to charge miscellaneous fees when you trade outside of their platform; 5,000 shares tends to be the sweet spot where they can recoup any of their miscellaneous costs in return for better execution quality. So for smaller lots, the value added might be there as far as better execution, but there are some costs that they may not be able to avoid just by the limitations of what custodians let you do.

ETFdb: How has the ETF industry responded post “Flash Crash” to these sudden critics of ETFs, including the mutual fund industry? What measures can investment managers take to protect their portfolios against these potentially very costly trading errors?

PW: From a rudimentary standpoint, the first change is to an investment strategy that avoids just straight stop orders. Obviously, ETFs are made to trade intraday, and if you are a technically-based manager or just a good risk manager, using stops in your portfolio process is a prudent thing to do. However, I think you need to replace all your stops with stop limits and be pretty generous with those limits, so as to actually get executed when the market might crater one way or the other. So if investors build into their expectations 10 to 25 cents away from their initial stop target and establish stop limits there, at least they will get protected and they won’t have any executions go far away from the actual NAV of the funds, which happened on May 6th.

The basic issue is that if you are being really stingy and trying to save a few pennies in a falling market, you may not get filled at all at a stop-limit and the market runs away from you. And what ends up being a few pennies turns into dollars lost. So that is probably the first thing investors should adjust as far as their trading methodologies.

The second thing goes right back to understanding the construction of ETFs and ETNs as far as the underlying index. On May 6, people were trading based off what they saw on the screens. And again the bid, the ask, and the prices that are kind of spit out as data on machines are not always accurate. There are trading errors in the market, whether on May 6th or any other day. Every day there is a bad execution and a handful of ETFs that are traded well above or well below the actual reference value of the ETF.  That kind of shows that people are not cognizant of how ETFs are priced. If someone is looking at an ETF that is based on the S&P 500 that is trading $20 below where it was trading a few minutes ago and the S&P has not moved that much, there is no reason to take action if you are a seller. You should know with confidence what that holdings are because they are published daily to the fund’s Web site and then know that it is some kind of data error.

That is what we have found people doing on May 6th; investors traded on what they thought were real prices because that is what was on the screens. They panicked and lost the concept that ETFs are not like stocks where news can come out and the supply and demand aspects of the stock rule the day. The NAVs were way out of whack from where the tangible values of the funds were for a brief amount of time, and anyone selling into that just got taken advantage of. If anything, people should have been buying knowing what is inside of the indexes and how the ETFs are priced, but probably far and few between actually did that.

ETFdb: So it seems that in your opinion, some lessons have been learned and there are some things that can be done to prevent a similar event from unfolding again.

PW: If you were trading in a smart manner, you would have put limits out there. And if you did not get filled, it is not the end of the world. Once the glitch gets cleared up the ETFs will go back to somewhat near their fair value. People did not do that; they just kept chasing and things did trade at some point where others stepped up and said “Wow, this is a really appealing trade because it is trading $10 below the NAV of the fund.” And those trades, unfortunately, did not get busted. So some investors now have a really bad experience in their memory that does not accurately reflect how ETFs are supposed to work and why they make sense as investment vehicles.

I hope they do look into what caused the meltdown across the board. However, on that day we did have inquiries with people questioning what was going on, asking for advice and saying “Should I be selling into this?” And the answer is “no.” And that answer was based on having some degree of confidence in where the index value was versus where the ETF looked like it was trading on the screen. If anything, we were telling people that they should probably be buying and be ready to hold that security, because if you immediately buy something and then try to sell it one side of those trades may be busted. And that is exactly what happened; traders who tried to buy a security at discount and then quickly flip it ended up having the buy busted and were basically short something and had a disadvantageous price the next day. For long-term holders looking to accumulate something, that was a fine opportunity to do so. But for short-term traders it was a free-for-all.

ETFdb: Very interesting thoughts—thanks for sharing with us!

For registered investment advisors, Street One Financial also offers S1F ETF Daily, a daily publication focusing on general market technical commentary, actionable ETF trade ideas, and ETF options trends/recaps. Sign up for the free subscription here.