This blog post summarizes the book titled “Visual Guide to ETFs”, by David J. Abner

Characterization of Modern Investment Products

ETFs have democratized/equitized investing as it has made it easier for anyone to take a position in any of the various asset classes, with a small investment. There are liquid ETF’s available for Equities, Corporate Bonds, Government Bonds, Money Markets, FX, Commodities, Crypto Currencies, Asset Allocation strategies, or whatever you can think of. As of Oct 2010, the total assets are close to 9.7T dollars across 9,000 odd funds. Indeed the rise of ETFs in # and size is astounding in the last few years. Financial Advisory Business has been a great beneficiary with ETF revolution, as they are now able to advice against all sorts of asset classes. With the FinTech Robo Advisory startups, there has been a wide spread adoption of ETFs by the retail investors.

The following are the some of the points mentioned in the chapter:

  • According to Investment company institute, there are four types of products that fit into the definition of investment companies
    • ETFs
    • Mutual Funds
    • Closed-end funds
    • Unit Investment Trusts: They are a combination of several product types. They issue specific amounts of shares that are called units. They trade on exchanges but typically only to facilitate redemption of shares by investors. And the portfolios typically fixed until predetermined termination date.
  • Total AUM under MF = 100T
  • Total AUM under ETFs = 10T
  • Growth rates of CEF and UTI shows that investors are getting more savvy
  • ETF structure is an example of taking a mutual fund structure and combining it with some of the benefits of closed-end fund structure
  • Why ETFs have grown ?
    • Regulatory changes that provide significant structural advantages
    • Electronic trading
    • Marketing engines
    • Investors have been demanding a reasonably low fees and through a straight-forward structure
  • The four main reasons for ETFs popularity are
    • Transparency: No other fund product that makes its portfolio available on a daily basis
      • There are funds that are making the portfolios available on a daily basis without announcing changes before they occur
    • Exchange listing
      • Standardization
      • Intraday trading
        • Mutual funds has the benefit of never having to explain to a client the concept of how to achieve best execution
        • ETF industry has given investors the ability to manage their own executions
      • Liquidity
    • Tax Efficiency
      • The in-kind creation and redemption process is what drives the tax efficiencies of the ETF. The portfolio can be managed by the delivery and receipt of shares without trading and causing tax liabilities
      • Any creation or redemption of units in ETF do not create a taxable event.
      • In a Mutual Fund, a large redemption might impact all the long term investors as the MF manager has to trade in order to fund the redemption. This is unlike the case of ETF where the transaction is in-kind
      • The taxable events of trading the underlying equities and the actual ETF shares have been moved back to where they should be, at each particular end investor involved in the transaction
      • Lower fees
  • Players in the market
    • Retail flow
    • Institutional Flow
    • Advisor Flow
    • Arbitrage traders
    • High Frequency
  • ETFs do not do IPOs to raise assets. Every ETF launches with some amount of seed capital that the shares can be bought to the secondary market. Typically, this is a process whereby AP goes in to the market and purchases the required basket for the new ETF and processes an initial creation
  • Types of Replication methods
    • Full replication is the leading replication method for equity and commodities ETFs, which seems to be a bit contrary as investments in commodities are normally made through futures. However, since futures are not eligible for UCITS funds, ETFs do normally use swaps for the replication of commodities indices. That said, full replication is used in non-UCITS compliant precious metals funds domiciled in Switzerland which invest directly in the respective precious metal
    • Optimized replication is the most widely used replication method for bond and mixed-assets ETFs. An external link provides more clarity on this:

This strategy is used when buying every security in an index is impossible, or inefficient. Sampling strategy refers to a “sample” of holdings from the underlying index. For example, if only 80% of the securities from an underlying index are included in an ETF’s holdings, then the fund is sampling from rather than fully replicating the index. Sometimes an equity index to which an investor wants exposure might be too complex or may have too many holdings. In these cases, the ETF that tracks the index may use a sampling strategy. Instead of avoiding that index or market altogether, an ETF can use a sampling strategy to gain exposure and optimize the exposure to the index.

  • Synthetic replication is the most widely used for alternative UCITS, money market, and “other” ETFs.
  • Characteristics of MF
    • Since trading takes place within the fund, all investors are subject to the costs of trading relating to the other investors entering and exiting the fund
    • Heavy users of 12b-1 fees as they use to pay distributors for a variety of functions
    • Sec 12b-1 rule delineates distribution fees to include feed paid for marketing and selling fund shares and paying for advertising, the printing and mailing of prospectuses to new investors
    • Mutual funds do not have tax efficiency built in like many of the ETFs do
    • MF have purchases and redemptions that take place in dollar amounts
    • NAV determined at the end of the day
    • Investor does not know exactly at what NAV his units are bought or sold, during the day

Understanding How ETF Portfolios work ?

The following are some of the main points mentioned in this chapter:

  • The creation unit is the set of underlying components of an ETF that can be exchanged with the issuer for a certain quantity of ETF shares. When shares of the ETF are being redeemed, this basket is also typically used
  • Shares outstanding of an ETF represents how many shares of the ETF are trading in the marketplace at any given time. This number, multiplied by the price of the funds, gives an approximation of AUM. This number is present on a daily basis
  • Shares outstanding of a fund can be a misleading indicator
  • Creation and Redemption baskets are present
  • Shares outstanding is a misleading indicator
  • Why is creation unit different from redemption unit ?
    • If there is a corporate action in a stock that the fund holds, taking a position in a different stock. If there is time to adjust the portfolio, sometimes the fund can be delivering out the share that it needs to get rid via the redemption baskets, while taking in shares of different stocks via its creation basket. This sometime enables the fund managers to manage the portfolio without creating taxable events for shareholders
  • How to seed an ETF?
  • What happens when Authorized Participant creates or redeems ETF shares in the primary market ?
  • Three types of indicators

    • NAV: Official NAV is published once a day

    • Indicative Value: ETF issuers provide to offer a more real-time NAV. It is provided every 15 seconds. No other listed investment vehicle provides an intra day indicative value of its components based on real time prices

    • Estimated NAV: This is more high frequency component where the value is computed at every tick. This is also the ones where quants come in. For illiquid underlyings, or underlyings that trade in a different time zone, there is no real time iNAV. Hence eNAV is something that market makers have to figure out

      • Example of an international ETF. Japanese market is closed but if there is an event in the US stock market, it is better to take in to estimation that the Japanese market would correct and hence reflect in the eNAV quotes rather than replying on the basket values based on stale prices.

      ETFs and their subjective eNAVs have brought together many different types of market participants who are trying to profit from high-speed arbitrage of these products and their underlying assets. If the ETF consists of a domestic equity basket, then the perfect hedge is obviously the basket itself. The arbitrage then becomes a game of speed of execution and financing rates, and this is aided by colocation of servers and the proliferation of the high-frequency trading in the ETF product. However, if the basket of an ETF is not accessible at the same time as the ETF, as is the case with most international ETFs, the trading strategies calculate an eNAV that tries to look for a bid/off er in the market of someone with a different opinion. That is, since each eNAV in this case will be slightly different, increased trading volumes can ensue. Investors should be aware of the subjective nature of ETF valuations during times when the underlying assets are not trading and understand that some ETFs will trade at perceived premiums or discounts because of legitimate time lags in their published NAVs

Understanding the Short Interest of ETFs

The following are some of the main points mentioned in this chapter:

  • High short interest in ETFs warrants examination to understand how the arbitrage mechanism enables this to happen more cost-effectively than in a standard equity.
  • Basic formula for shorting
1
2

Number of outstanding shares + Number of open interest shares held short =
Number of open interest shares held long
  • In an ETF, the process for the stock-lending desk contains an additional option, creation and redemption
  • Basic decision to be taken - Is the cost involved in being long the ETF, and collecting the fee for lending your ETF shares outweigh the cost of being short the underlying basket to hedge your market exposure ?
  • If you are short ETF, unlike an equity, you can resort to short the basket, long the ETF shares in the primary market, to cover the short
  • The hedge fund community frequently uses ETFs as short hedges against their positions. To facilitate the stock borrows, the large trading desks typically borrow the baskets, add a fee for the use of balance sheet and their services and then lend the ETFs to the hedge funds. Since shorting is such a large part of the ETF marketplace, there is a large amount of revenue being made for the simple process of lending the ETF wrapper to provide a one trade solution
  • What are the implications for long-term investors who may own an ETF that’s being shorted ?

It’s really going to depend ultimately on the ETF in question. Now, in the case of ARKK, what you could very well see if there is sufficient demand among short-sellers for ARKK shares to be sold short, is that that could actually somewhat counter intuitively result in inflows to the fund. Within the realm of ETFs, there’s a type of activity that’s called create-to-lend. And what that means is that new shares of that ETF are being created to lend to short-sellers for purposes of being able to express their bearish view. Now, that sort of activity could put more money on Cathie Wood and her team’s plate to express long views with. So, what you see in that instance is a really unique feature, which could in certain circumstances exacerbate concerns around things like capacity, which we documented in detail in our March analysis of ARKK.

  • High short interest in certain ETFs is not a risk to the funds or their investors.
  • Why short ETFs?
    • Hedge risks borne by long positions
    • Bet against the future returns of the whole market, certain styles, or certain sectors
    • Synthetic shorting: Traders want to bet against a subset of stocks held by the ETF, but are unable to short those stocks directly.

This chapter is very interesting for even a ETF veteran, as the author clearly explains the mechanics of shorting in the context of ETFs, that creates very unique situation that sometimes is exploited by traders who are looking to short hard to borrow stocks

Trading and Liquidity of the ETF Markets

The following are some of the main points mentioned in this chapter

  • Displayed lot sizes have lost relevance, with the advent of non displayed order size types. Also with the unique structure available, it is easy for a large client to walk up to the primary market and get the liquidity directly from the fund via an authorized participant
  • Many reasons to trade ETFs
    • Believer in the growth story and hence wants to go long the basket of stocks
    • EFT fits a specific trading strategy. Could be high frequency strategy, merger arb or any strategy specific to ETF
  • The average trading volume of ETF is a misleading indicator as one can directly approach the primary market and create baskets.
  • Bloomberg publishes implied volume indicator for each of the ETFs based on the underlying liquidity of the basket
  • ETF’s liquidity has several components such as
    • Underlying Basket Liquidity
    • ETF Average trading volume
    • Related Derivative volumes
    • Correlated trading vehicles
  • ETF basket implied liquidity scale developed by the author

The takeaway from this chapter is that formal measurement metrics in the secondary market are not very useful in the ETF space. One needs to take in to consideration the implied volume indicators or crunch one’s own metrics that takes in to consideration both the primary market characteristics as well as characteristics of the underlying basket of stocks.

ETFs and the flash crash

This chapter goes in to explaining some of the factors mentioned in the SEC report on flash crash in May 2010. Here is a blurb from wiki, that summarizes the event

On May 6, 2010, U.S. stock markets opened and the Dow was down, and trended that way for most of the day on worries about the debt crisis in Greece. At 2:42 p.m., with the Dow down more than 300 points for the day, the equity market began to fall rapidly, dropping an additional 600 points in 5 minutes for a loss of nearly 1,000 points for the day by 2:47 p.m. Twenty minutes later, by 3:07 p.m., the market had regained most of the 600-point drop

What is the role of ETFs in this period ? What happened to ETFs in this period ? These questions are concisely answered in the chapter.

  • ETF trading volumes exploded on May 6 and May 7. This confirms the usefulness of ETFs in various kinds of requirements such covering short trades, hedging derivatives
  • ETF market makers paused trading during the distress period causing price volatilities
  • Market makers started putting in ‘stub quotes’, that were responsible for even more wider spreads and higher volatility among underlying equities
  • NYSE invoking LRP triggers

LRPs are intended to act as a ‘speed bump’ and to dampen volatility in a given stock by temporarily converting from an automated market to a manual auction market when a price movement of sufficient size is reached. In such a case, trading on NYSE in that stock will ‘go slow’ and automatic executions will cease for a period ranging from a fraction of a second to a minute or two to allow the Designated Market Maker to solicit and/or contribute additional liquidity before returning to an automated market.

Even though the author provides many aspects that are popularly quoted as factors, there is a lot of debate on what exactly happened. This article says

Other reports point to Stub Quotes as a factor in the Flash Crash. This is not supported the data. Nearly all Stub Quote Trades, which are trades executed at stub quotes, occur after the market bottomed and therefore stub quotes could not have been a cause of the crash.

We have noticed that trade executions in a stock immediately after it hits an LRP are often at substantially lower prices than bid prices on other markets. It was this reason why we began our flash crash by first investigating LRP’s role in the flash crash. However we found that the number of stocks switching to Slow Mode did not become significant until well after the market had already bottomed.

How to execute orders in ETFs ?

This chapter walks the reader through some of the common ways to obtain ETF shares. At first blush, it might seem that there is one way, i.e. approach a trading exchange and buy it as there is no difference in buying an equity share or an ETF. However the unique structure of ETF gives rise to many ways of seeking liquidity

  • Using an electronic trading system to trade the ETF using limit or market orders
  • Calling a block liquidity provider and asking for a risk market
  • Calling a liquidity aggregator and asking for a market
  • Using an algorithm on either the basket or the ETF
  • NAV based execution

Some of the other points mentioned in this chapter are

  • The growth of ETF business and the equitizing of a wide variety of asset classes have happened at a faster rate than banks have been able to modify their internal relationships and trading systems
  • A lot of newer ETFs do not fit easily into the pricing and trading for standard equities. Those that contain currencies or bonds require additional tweaks to trading and risk systems that were primarily designed to handle equity ETFs. To fill the niche left open by structural deficiencies, specialized ETF desks have grown
  • ETF portfolios have typically two types of cash - cash from dividends and cash from corporate action
  • Typically orders increase around end of market as buyers want to value their portfolios based on closing NAV
  • Liquidity aggregator acts as a middleman between client and broker-dealer/etf market making firms/banks

Examining ETF Trading Strategies

This chapter deals with trading a large block of ETF in the market. The following are some of the points mentioned in the chapter:

  • Trading a large block in a low-volume ETF by calling up the liquidity provider
  • An ETF trading in U.S markets with holdings in India, is trading at prices that are the result of all the various market participants making estimates of where the underlying basket will be trading when the underlying market next opens. It is a price discovery vehicle
  • Executing daily rebalances in a low-volume ETF via preagreed trading price that is a fixed spread away from estimated NAV
  • There is no liquidity provider that is trading an ETF without a calculated eNAV
  • Way to assess the impact of large block trade in single currency ETF
  • Way to assess the impact of large block trade in Bond ETF
  • Way to assess the impact of large block trade in currency basket ETF
  • Three metrics to think about in the trading strategy: the level at which you want to execute a trade, the estimated NAV, current bid/offer prevailing in the market

Product Labeling and Structural Impact

This chapter is useful for someone who wants to more clarity on the type of fund structures available with in the ETP space.

  • Bulk of the ETF assets are registered under Investment Company Act of 1940
  • ETFs are sometimes called 40 Act funds, which means they are registered under the Investment Company Act of 1940.
  • A variety of issues or concerns about ETFs have arisen
    • Concerns about transparency and reporting standards
    • Counterparty concerns
    • Impact of methods of achieving leverage
    • Managerial conflicts of interest
    • Increased trading volatility
    • Possible systemic risk of the products

The chapter has a nice table that explains the various types of products in ETP space.

In the last decade or so, there are now many more additions to the above table, but the above one is a good starting point

The following are some of the points mentioned in the chapter:

  • ETF is an investment vehicle that combines key features of traditional mutual funds and individual stocks. ETFs are open ended securities
  • ETVs are open ended trusts or partnership units and often includes commodity and currency trusts
  • ETNs are senior unsecured and unsubordinated debt obligations of the issuer. They are designed to track the total return of the underlying index or other benchmark, minus investor fees. ETNs are different from ETFs because they are subject to the creditworthiness of the issuer.
  • The need of the hour is a detailed labeling system that addresses active/passive , levered/unlevered, swap/optimal/full replication etc ?
  • Chinese ETFs - Premium demanded is a genuine compensation for the counterparty credit risk that arises by using swaps for replication. It cannot be arbitraged away
  • Just because there is a premium doesn’t mean there is an arbitrage opportunity. It all depends on the accessibility of the underlying. Presence of the premium does not always indicate any problem in the fund.
  • ETF that tracks commodity futures - There are nuances that one needs to take care of such as contango and backwardation impact on the ETF
  • ETFs NAVs can trade at a huge premium when the underlying market is closed. Happens when there are restrictions applied on underlying stocks and the underlying stocks become very sought after, but can only be gained exposure via ETFs units from a fund via Authorized participants
  • ETF prices can trade at premium because there are genuine issues in getting access to hedging the underlying basket
  • ETF can forever be trading at a discount to the benchmark if there are genuine issues in getting access to hedging the index constituents
  • There could be divergence in performance between ETF and its benchmark when ETF fund manager has to constantly roll the futures contracts and has to incur costs.

Where to find ETF information ?

This section lists out all the important players in the market that provide ETF data. Reuters is missing from the list - This shows the immense competition between Bloomberg and Reuters, that one of them does not even acknowledge the other, when reporting on the competition. Reuters/Refinitiv/LSEG has an extensive offering, Lipper Fund Analtyics for the last 20 years and has very deep history of raw fund data as well as many derived analytics. The firms mentioned in this section are

  • Bloomberg
  • ETFDb
  • Morningstar
  • Index Universe
  • XTF
  • Markit: Most trading in European markets is done via OTC and hence Markit has an advantage
  • DTC
  • Netik

Institutional investors and traders require lot of additional data such as detailed basket data, forward-looking ETF dividend predictions, creation and redemption fees per ETF, excluded asset breakdowns and many other fields.

This section is sort of outdated as the book was written almost a decade ago. There are many new players in the market now, offering a variety of data that relates to ETFs.

I think for someone trying to understand the space, a quick look at ETFDb screener will be useful as it serves an indication of the various aspects that make ETF different from any other type of fund.

Takeaway

Even though the book was written in 2013, there are many useful aspects that are very well explained by the author. Given that the author has been a market maker in the ETF space, almost everything given in the book is rooted in reality and is not some theoretical discussion. Of course there is a heavy reliance on BBG screens to illustrate various concepts. Even if you do not have access to any ready made market data terminal, there are many free resources on the web that you can use to verify/ investigate further, the various idiosyncrasies of ETF market. I think this is a very good book for anyone who wants to detailed understanding of ETFs.