Exchange traded funds (ETFs) part two. We discuss the hidden risks of concentrated ETF holdings in Apple and Microsoft and the exclusive club of authorized participants, most of whom had to be bailed out in the Financial Crisis. We look at the exciting new bitcoin futures ETF and explore why that succeed where previous attempts had failed, how the regulators are thinking about bitcoin ETFs, and when we can expect a bitcoin spot ETP.
.png)
Hey friends -
We're continuing our exploration of exchange-traded funds this week! Check out last week's letter and listen to the podcast. Even with the in-depth dive last week, a lot of ETF history got left on the cutting room floor. I didn't want to let it go to waste and posted it on Twitter later in the week. You can read that full history here.
Special thanks to CryptoMondays Brooklyn for inviting me to speak and hosting another great event! We had a lot of fun discussing The Great Wealth Transfer and what that will mean for crypto as digital-native millennials and zoomers come into wealth for the first time. If you're in NYC, I highly recommend joining. We get together every Monday at 7pm at The Williamsburg Hotel with invited speakers from across the ecosystem. It's a good group with varying experience in crypto. And there's free food and booze to boot!
In this week's letter:
Total read time: 11 minutes, 48 seconds.
With over 4000 public listed companies in the US, you'd probably expect the ETF industry to be highly diverse. Some ETFs like the SPDR would cover well-publicized indices of the 500 or so largest companies. Other ETFs might cover the long tail of much smaller companies that have the potential for massive growth and to become the next trillion-dollar company.
In one sense that's true. There are over 1800 ETFs focused on public companies that can give investors exposure to just about every corner of the market. In another, very important sense, the industry's highly concentrated.
Almost every ETF owns Apple and Microsoft. And in truly astonishing quantities.
The five largest ETFs each manage more than $200 billion. All of them own Apple and Microsoft. Of the 22 ETFs that manage more than $50 billion, over one-third own the tech giants. When they do, it's not just in small amounts. It's almost always the number 1 and number 2 holdings totaling tens of billions of dollars. Take the top five ETFs for example:
That's an almost $200 billion exposure to just Apple and Microsoft.
Where the ETF analysis gets even more interesting is as you go further down the league tables into smaller ETFs. Even though Apple and Microsoft are massive companies, they're growing revenue at double-digit rates and showing up in growth ETFs as well as value ETFs. For the most part, unless an ETF is focused on small companies or non-US companies, it often owns Apple and Microsoft.
This remarkable concentration in just two companies is driven by their astonishing size coupled with their remarkable growth rates. ETFs looking to mirror the market find that Apple and Microsoft account for around 12% of the total value of the S&P 500. ETFs looking for more rapid growth find that the companies are among the fastest-growing and - critically - dollars can be invested at a capacity that often isn't possible in smaller companies.
Now imagine if one of these companies had an accounting hiccup or, even worse, was prosecuted for fraud. To be clear, there are no indications of fraud, but the reverberations would be tremendous. Far beyond impacting the shareholders that directly invest in the company, most passive investors would also see significant losses due to their exposure via ETFs. Such passive investors account for 45% of assets invested in the stock market today. Those losses would likely lead to contagion as investors began questioning whether other large, technology-focused companies like Amazon were also in trouble.
The probability of something like this happening is small, but it isn't zero. Worldcom was a $186 billion company before the fraud was uncovered and it filed for bankruptcy in 2002. Whereas a large event like Worldcom was mostly isolated to the direct shareholders, such an event today would likely be systemic because of the concentration in ETF holdings. That's a new type of risk for the market, one we haven't encountered before and against which most investors are poorly positioned.
But it's not the only concentration risk in ETFs. There's another lurking in the primary market.
As we saw in last week's letter, Authorized Participants engage directly with the issuer in the primary market to create and redeem ETF shares.
Unlike the secondary market which facilitates trades among millions of retail investors, the primary market is limited to major financial players who have the capital, sophistication, and technology to participate. As a result, it is far more concentrated.
If we look at the headline statistics, the market looks reasonably diverse.
There's an average of 27 authorized participants per ETF of whom 5 are active, where active means they created or redeemed ETF shares in the last 12 months. The problem with averages is that they can be misleading. We often assume that distributions are normal, but the distribution could be in fact be barbell-shaped with lots of small ETFs and lots of big ETFs but not a lot in the middle. In a barbell distribution, averages don't do us much good.
When we look at the number of authorized participants by the size of the ETF, we find that there are more authorized participants for large ETFs and fewer for small ETFs. This is a good example of being suspicious of how data is presented. We see little-to-no change in the number of authorized participants as the ETF size scales from $50 million to $500 million, but then we see a 2x jump for the $500+ million group.
That's a weird cutoff. We know from earlier that 22 ETFs manage more than $50 billion. In fact, there are over 550 ETFs that manage more than $500 million and their size ranges almost three orders of magnitude up to just shy of $500 billion. That's not even remotely a reasonable comparison to the one order of magnitude grouping of ETFs from $50 million to $500 million.
Why the misleading data? The data above is sourced from BlackRock. BlackRock owns iShares, the largest ETF provider by quite a stretch. They have a significant incentive to publish data demonstrating that all's fine in the world of ETFs.
With suspicions reasonably aroused, let's turn our attention to the individual Authorized Participants themselves. To little surprise, the world of Authorized Participants is highly concentrated.
Just four authorized participants account for 60% of ETF creation and redemption. The BlackRock report from which the chart comes continues with an analysis of how the authorized participants performed during the market volatility in March 2020 as COVID-19 lockdowns began. That discussion is based on data from an Investment Company Institute report on the same topic. From the report:
[r]ather than stepping away in a crisis period as some have predicted, more APs, on average, stepped up to facilitate the higher level of primary market activity in the March 2020 stress period compared to the March 2019 normal period.
Any other conclusion would have been shocking. Authorized participants make money from arbitrating the difference in price between the ETF and the underlying securities (see last week’s letter). Volatile markets are good for authorized participants - they make more money!
A more prescient question would be to ask how authorized participants are likely to perform in the event of another financial crisis. While we can only guess at the answer, it's helpful to remember that the four largest authorized participants all had to be bailed out in 2009. And the fifth and sixth largest. And two of the next four largest.
A full 79% of authorized participant activity can be accounted for by just 8 firms, all of whom had to be bailed out in 2009. If that's not concentration risk I don't know what is.
I take that back. The $678 billion Fixed Income ETF market. 82% of the activity is spread across just 3 firms - Bank of America, JP Morgan, Goldman.
Exchange-Traded Funds are part of a larger group of financial products known as Exchange Traded Products, ETPs. While ETFs are issued as Investment Companies under the '40 Act and therefore limited to holding securities, ETPs can invest in a much larger universe of assets. The difference may seem trivial but can be highly material.
Take gold for instance. There are seemingly many gold ETFs. If you google it, you'll find results for "Investing in Gold: 3 Largest ETFs" and "Gold ETF List - ETF Database."
The problem is, none of them are actually ETFs. Just read the regulatory filings for the products, such as those for SPDR Gold Shares (ticker: GLD), the largest commodity "ETF":
GLD is not an investment company registered under the Investment Company Act of 1940 (the “1940 Act”) and is not subject to regulation under the Commodity Exchange Act of 1936 (the “CEA”). As a result, shareholders of the Trust do not have the protections associated with ownership of shares in an investment company registered under the 1940 Act or the protections afforded by the CEA.
Most of the protections investors get under the '40 Act focus on disclosures by the ETF issuer. They have to publicly publish the value of the ETF daily, a graph of the difference between the price of the ETF and the underlying securities, and related information that would make it difficult for the issuer to manipulate the price(s) of either the ETF or underlying securities without ETF shareholders becoming aware. The CEA plays a similar role for futures, a financial product where you agree today to buy or sell a commodity or security at a predetermined price at a specified time in the future.
There is a tremendous variety of ETPs that give investors exposure to just about every corner of the financial world. Commodity ETPs for oil, gold, and natural gas; currency ETPs for take-your-pick of the government currency; and even "volatile" ETPs so investors can bet on how volatile markets will be.
Until recently, there has not been a bitcoin ETP. That changed on October 19th.
The first Bitcoin ETP was proposed over eight years ago. It was rejected by the Securities and Exchange Commission (SEC), the primary regulator, as have been all subsequent applications. The November 12th, 2021 rejection of the VanEck Bitcoin Trust provides good insight into why.
The regulator stated that the ETP application failed to demonstrate that it was “'designed to prevent fraudulent and manipulative acts and practices” and “to protect investors and the public interest,'" two requirements taken from the Securities Exchange Act of 1934. In particular, the regulator noted that :
…it is essential for an exchange listing a derivative securities product [such as the Bitcoin ETP] to enter into a surveillance sharing agreement with markets trading the underlying assets for the listing exchange to have the ability to obtain information necessary to detect, investigate, and deter fraud and market manipulation, as well as violations of exchange rules and applicable federal securities laws and rules.
Such "surveillance sharing agreements" and the related technology infrastructure does not exist today.
The rejection is notable for two reasons. First, much of the information that needs to be collected under a "surveillance sharing agreement" is not collected by the exchanges where bitcoin trades today. Because bitcoin is a bearer asset like cash, buyers and sellers can exchange the asset without intermediaries. That limits the opportunity to collect information about the individuals who own bitcoin unless it is an artificially imposed requirement.
Second, the regulator quoted from the Securities Exchange Act, not from the '40 Act. That's because the proposed financial product would hold bitcoin directly. Bitcoin is a commodity, not a security. Only financial products that hold securities are issued under the '40 Act.
What did the regulator approve? An almost bitcoin ETF issued under the '40 Act.
I say "almost" because the ETF does not actually hold bitcoin - it holds bitcoin futures. Those futures trade on the Chicago Mercantile Exchange and are regulated by a different entity, the Commodities & Futures Trading Commission (CFTC). This is where the story gets rather odd.
In October, the SEC approved a bitcoin ETF that holds bitcoin futures based on the bitcoin markets. One month later, the SEC rejected a bitcoin ETP because it couldn't detect if there was fraud in the very same bitcoin markets. That's weird.
What we're watching play out is in part a turf war. The CFTC, which regulates futures, green-lighted financial products that reference the bitcoin market. The SEC, which regulates securities, has not. By restricting their approval to just an ETF that holds futures, the SEC can point the finger at the CFTC if fraud emerges in the bitcoin markets.
That cynical view has a strong counterpoint. The '40 Act protections are real and meaningful for investors. Bitcoin markets are relatively new and opaque. By restricting bitcoin products to just those that qualify for the '40 Act protections, the SEC is limiting the potential damage done to investors. This may turn into a wait-and-see approach where, if successful, a bitcoin ETP that actually holds bitcoin gets approved.
The counter-counterpoint is that futures ETFs often have high fees, may not track the price of the underlying asset as well, and can get out of wack when the underlying futures contracts expire (since futures are an agreement to buy or sell predetermined price at a specified time in the future, they have to be renewed).
The approved ProShares Bitcoin Strategy ETF certainly has high fees, about 5x the industry average. Time will tell how well it tracks the price of bitcoin and how it manages through the expiration of the underlying futures contracts. The SEC's made a bet that it will work out and that the high fees are justified given the additional investor protections offered by the '40.
Time will tell if the SEC got it right.
Great for brunch or even in the evening when you want to get the night started but can't get off the couch.
2.0oz Rye Whiskey
1.0oz Cold Espresso or Cold Brew Concentrate
0.5oz Brown Sugar Syrup
0.75oz Cream
2 dashes Chocolate Bitters
Pour just the rye, coffee, and sugar syrup into the shaker. Add ice until it comes up over the top of the liquid. Shake for ~20 seconds until the outside of the shaker is frosted. Strain into a rocks or cordial glass. Discard the ice. Add the cream and bitters to the now-empty shaker. Shake for 20 seconds or longer to whip. Pour over the rounded back of a spoon to float the cream on top of the drink. Enjoy, complete with a milk mustache.
I love a good coffee cocktail. A group of us were prepping an epic Friendsgiving feast and collectively hit a wall halfway through cooking. The Rye Cappuccino was a much-needed kick. The cocktail's a bit like a mochaccino met a Mexican hot chocolate. The spice from the rye plays nicely with the coffee and chocolate, while the sugar sweetens it up just enough so it's not savory. By mixing the chocolate bitters with the milk and floating it on top, it hits your nose much more prominently. A warm version of this might just have to make its way into the snowy winter's day rotation.
Cheers,
Jared
