iShares’ Claims Make Me Nervous

July 11, 2013

ETFs offer a window into the behavior and liquidity of an underlying asset class—but nothing more.


[This article previously appeared on our sister site,]


Which financial innovation introduced greater tradability to a range of financial market assets, was acclaimed as representing the “democratisation of capital”, was widely adopted as a new technology and was held to have brought down companies’ cost of finance?

For the answer, read down a little further.

But first, a short digression. iShares, the world’s largest issuer of ETFs, recently published an open letter on its website. The letter appeared at the tail end of a bad month for ETFs (June 2013 witnessed the largest redemptions from the sector since early 2010).

In the view of iShares’ global head, Mark Wiedman, “more and more, ETFs are becoming the true market”.

ETFs increasingly allow investors to gauge the true market price for a particular investment exposure, Wiedman argues.

In the open letter he says that “in a rapidly moving market, the reported prices of individual underlying securities may become stale,” before concluding that “the ETF price can become the true price for that market, and the underlying assets may eventually catch up with any gap between the two (called a ‘premium’ or ‘discount’)."

According to iShares, the ETF is effectively taking the lead in pricing the stocks or bonds in the underlying index, not merely acting as a passive, “pass-through” vehicle to reflect the prices of the constituents.

You can see where iShares (and other ETF proponents who have made similar arguments) are coming from.

There’s ample evidence of ETFs in a sense “becoming” the market: from the huge volumes and wafer-thin dealing spreads of the SPDR S&P 500 fund, the most traded single security in the world, to the recent boom in junk bond funds, where you can trade in and out of the index basket (via an ETF) at a hundredth of the cost of dealing in the underlying bonds themselves.

iShares, it appears, judging by recent statements, has big ambitions for its ETFs to be seen as more than just tradeable index funds. The firm appears to want its funds to combat futures and other derivatives markets as well as to take on, to some extent, the role played by individual stocks.

All the same, Wiedman’s comments make me uneasy. They remind me of the point at which another financial innovation went wrong.

That new force in the financial markets was securitisation, a technique that gained the plaudits I mentioned in the opening paragraph. It was indeed hailed as a democratising force, one that enhanced the tradeability of previously illiquid assets and which brought down access costs to the financial markets. There’s an excellent short history of the topic on the Economist website.



In the basic form of securitisation, multiple assets of a similar type were pooled and the payouts from the pool were distributed in sequence to different categories of investor.

For the first two decades of its existence, securitisation volumes grew steadily. Banks loved to securitise and sell off loans that they previously would have had to hold on their books until maturity. The technique, blessed by regulators, allowed financial institutions to manage their capital resources much more efficiently.

But then things started to go wrong. Around the turn of the millennium, the originators of securitised loan deals, aided and abetted by the main ratings agencies, began to argue that the simple fact of securitising could enhance the average credit quality of the underlying assets.

A pool of, say, 10,000 mortgage loans to a bunch of Americans with poor credit ratings, shaky employment prospects, who have put down no deposits and who have overpaid for their properties is, if you think about it, a pile of junk.

But according to the securitisers, the simple act of “tranching” the payouts from the loans—allowing the investors in the top tranche of the securitised pool to be paid first, and those in other tranches later, in sequence—was now sufficient to allow the top credit rating, AAA, to be awarded to the first part of the pool—even if the underlying assets were still those dodgy home loans.

This alchemy could be repeated almost ad infinitum: securitisations of securitisations of securitisations spawned more and more AAA-rated bonds from lower-quality loans. That is, until the music stopped in 2007/08 and many bonds with the top credit rating (a stamp of quality that should have meant investors were immune to credit-related losses) were suddenly trading at thirty or forty cents in the dollar.

And this is what makes me nervous about the claim that ETFs are becoming the market. Clearly, ETFs allow investors of all types to trade into and out of a variety of asset classes at the click of a button. Often they do this at apparently low cost.

But in my view, it’s a potentially dangerous mistake to draw the conclusion that ETFs are actually enhancing the liquidity of the underlying market, even if you can see that average ETF bid-offer spreads may be a small fraction of the average cost of trading in the underlying basket.

After all, the costs of executing the trade in the underlying basket are what the ETFs’ authorised participants will have to pay if they want to create or redeem units in the ETF. And it’s that upper limit (the creation cost) and lower limit (the redemption cost) at which investors will very soon find themselves trading in the secondary market when times get tough, as we’ve witnessed in recent weeks.

Ultimately, the liquidity of the ETF is no more and no less than the liquidity of the underlying asset being tracked—which may be generally very good (in indices of large-cap shares, like the Euro STOXX 50 and the S&P 500) or, from time to time, extremely patchy and unreliable (in lower-grade corporate and many emerging market bonds).

Securitisation offers an instructive story of another financial innovation that brought measurable benefits for a while. But it went disastrously wrong once those involved thought they could improve the overall credit quality of a pool of assets, rather than merely redirecting the payouts from the assets to their new owners.

Although ETFs are involved in liquidity, rather than credit transformation, for me the claim that ETFs are doing anything other than reflecting the behaviour of the underlying asset class is to take a similar step in a potentially dangerous direction.



Find your next ETF