“Wall Street Turns to ETFs to Sidestep Illiquidity in Bond Market” ran a recent Bloomberg headline. Because the ability to trade individual bonds is degrading, the story reported, investors are piling record dollars into global bond ETFs.
Hmmm. So Wall Street, in its wonderful way, has invented a magic product wrapper that makes bond illiquidity problems disappear. Sort of like a financial version of Harry Potter's cloak, I guess. But investors should probably keep in mind that Harry was still there, even though you couldn't see him.
Meanwhile, the Federal Reserve is worried that thus-hidden bond market problems will reappear to haunt bond ETF investors: “As ETFs may appear to offer greater liquidity than the markets in which they transact, their growth heightens the potential for a forced sale in the underlying markets,” said a Fed monetary policy report released in February.
Well known investor Howard Marks agrees: “The ETF can't be more liquid than the underlying.” So does activist investor Carl Icahn: “phantom liquidity,” he said, is “a keg of dynamite.”
So what explains this mismatch of blasé and fear? How can huge investment flows be predicated on exactly the feature that many say is the product's Achilles' Heel? I think it's that one side understands how ETFs really work, and one side doesn't.
So let's do a quick review.
HOW ETFS WORK
Shares in an ETF are created when an “authorized participant” or “AP” — a financial institution in a contractual relationship with an ETF — delivers a basket of securities to the ETF in the proportions specified by the index or formula that the ETF tracks. For example, an AP working with the SPDR S&P 500 ETF (SPY) could deliver all the stocks in the S&P 500, in the current proportions of that index today, to that ETF in exchange for a new SPY share. That AP then can sell that share into the market via an exchange, hold it in inventory, or perhaps sell to an existing client who has placed an order.
Crucially, APs can also do the reverse: turn an ETF share into the fund in exchange for the basket of securities it holds. We'll see why that's so important in a moment, but for now just note that individual shareholders can't do this, only APs can; nor can individual holders redeem for cash, as they can with open end mutual funds. (And note that this is a core reason why ETF expense ratios are lower than most mutual funds: the infrastructure needed to support retail shareholder redemptions is relatively expensive, and it's a cost ETFs don't have to bear).
But what retail investors can do, of course, is sell (and buy) ETF shares on an exchange.
So when you go to cash out of an ETF, you'll always get the NAV right? Well, if the traded price and NAV are essentially the same, yes. But what exactly keeps those values close to each other? In other words, what protects investors from having to sell at a big discount to the fund's NAV … as, for example, so often happens with other kinds of traded closed end funds?
The answer is that the APs are expected (but not required) to keep traded prices in line with the NAV by exploiting an arbitrage opportunity that, in theory, should be present when the price of the ETF gets out of whack with the value of the securities it holds. This is the lynchpin of the entire ETF structure.
To illustrate: if an AP sees that shares of an ETF are trading at a price that's “too cheap” — maybe $99 when the value of the underlying basket of securities is $100 — it can expect to make money by buying a share for $99, redeeming it for the basket of securities the ETF holds, and selling that basket of securities into the market for $100. The resulting buying pressure should push the traded price of the ETF shares back to NAV.
Elegant, indeed. But never forget that the APs act purely out of profit motive. They don't have to do anything at all to “fix” an ETF's traded price. If there's not a clean and clear way to execute on that theoretical arbitrage opportunity to make money, they will sit on the sidelines and let the price go wherever. And note that if even one security in the ETF basket can't be quickly unloaded at a known price, there is no longer a clear arbitrage. So illiquidity in the underlying market means that the arbitrage mechanism won't work; and that means that ETF pricing won't, either.
Doubt it? Look at last August 24, when trading halts in just eight S&P stocks cascaded into stoppages of 42% of all U.S. equity ETFs. On the same day, fully one-fifth of all equity ETFs experienced price movements of 20% or more, even though just 4% of individual stocks did. Case in point: the very popular iShares Select Dividend ETF (DVY) experienced losses at much greater percentage price swings than any of the individual stocks it held.
FEATURE OR BUG?
So now let's reconsider once again that key marketing argument for ETFs: that they are “more liquid” than the underlying components of the fund. Is this a feature, or a bug?
Well, if you're a quick-twitch hedge fund, it's a feature. Why fool around with the clumsiness of transacting in many securities when you can instantly clear the lot? Indeed, the ETF structure, initially conceived as a convenient and cheap indexing tool for average investors, has indeed become a primary weapon of the Billions crowd … a fact well evidenced by the fact that the SPY has about 500 times the annual turnover of its underlying components. If that doesn't show that it's primarily a trading vehicle, not a long term investment tool, I don't know what does.
The good news is that so long as all these guys are furiously taking positions against each other, there should be a healthy market into which average investors can sell, too. So more supposed proof that ETFs are “liquid,” and not to be worried about.
But that's not the idea of the product structure, and those conditions, as we saw last August, can disappear in an instant. More to the point, hedge funds might well be nimble enough to exit their positions before a bout of illiquidity and mispricing sets in, but average investors aren't. And if the SPY, the world's largest ETF, can suddenly experience stoppages and wild price swings because a handful of stocks are temporarily halted, imagine what might happen when the underlying is radically less liquid than S&P stocks.
Instead of a hot market driven by nanosecond investors, the reliability of the pricing of ETFs is meant to derive from the arbitrage trade built into the structure: assets in the ETFs should be liquid enough for the APs to perform their pricing function. With the bond market, that just is not true… as the Wall Street machinery, in a profound irony, proudly touts as exactly the reason investors should pile in.
So maybe Harry's cloak isn't quite the right analogy. Maybe the Wizard's curtain is better. For when bond prices do come under pressure — whether by Fed action, a credit event, or a surprise whiff of inflation — Dorothy's going to pull it aside. And the little old market inside suddenly won't appear up to the miracles we're expecting.
Bob Rice is managing director of Tangent Capital in New York. He is also the bestselling author of “Three Moves Ahead” (Jossey Boss, 2008) and “The Alternative Answer,”(HarperCollins, 2013). Bob appears frequently on various television and radio programs and is a contributing editor for InvestmentNews. He can be reached via ricepartners.com