ETF liquidity: will it be illusory in volatile markets?

An April 5th article in the FT by Stephen Foley, “The alchemy of ETF liquidity is an illusory promise” (sorry, behind the pay wall) was pretty interesting. But the real story is an investor letter by Oaktree Capital’s Howard Marks called “Liquidity” that was cited by the article. We take a look.

Foley focuses on the use of Exchange Traded Funds (ETFs) as a proxy to buying baskets of individual assets. ETF have grown in popularity exponentially; one reason is that they trade throughout the day versus mutual funds that can be bought or sold at a single end of day price. This creates the look and feel of liquidity, but it may be just a front. The perception is especially dicey when the underlying instruments are assets that are prone to episodic liquidity like junk bonds and leveraged loans. (To this we would add leveraged ETFs, ETFs that short markets, and synthetic EFTs that don’t have to hold the underlying collateral as potential accidents waiting to happen.)

Can ETFs really be more liquid than their underlying assets? There is a mechanism embedded in ETFs to arbitrage differences between the market value of the basket and where the ETF trades. This tends to keep the two connected. But when markets are stressed, does this still hold true?

Foley cited a quarterly letter written to clients by Howard Marks of Oaktree Capital. The paper is titled simply: “Liquidity” and reflects Marks’ view on what really is liquidity with specific examples on ETFs. The letter is a must read and we strongly urge all our readers to take a look. Tell your friends too – it’s that important.

Marks notes that liquidity is a lot more complicated than volume. From the FT article:

“…Mr Marks is adamant: “No investment vehicle should promise greater liquidity than is afforded by its underlying assets. If one were to do so, what would be the source of the increase in liquidity? Because there is no such source, the incremental liquidity is usually illusory, fleeting and unreliable, and it works (like a Ponzi scheme) until markets freeze up and the promise of liquidity is tested in tough times…”

For those of our readers who traded repo or derivatives, this might recall the CDS market. A long protection trade can be viewed as similar to being short a bond and borrowing that security to the end of the CDS trade (opposite for short protection – and we know we’re glossing over the cheapest to deliver option in a CDS and some other details). But it seemed like for years the CDS market had more liquidity than the repo market ever did (and depending on when you looked, than the cash market did). But when the crisis hit, the CDS basis blew out, costing trading desks many millions on their trades. Instead of depending on the derivatives market for liquidity, the market pivoted and had to rely on the good graces of the underlying cash and repo market to hedge trades. The prices quickly reflected that. This was the subject of a Finadium paper “The Credit Default Swap vs. Repo Trade” (February, 2012.)

Marks writes that an asset might be liquid if you want to buy when everyone is selling, but if you are the seller you might have a different point of view. The answer to the question “is there liquidity” has different answers depending on the state of the market and what you are trying to do. If a market looks liquid now, will it stay that way? Often investors make decisions based on the liquidity at that moment in time, extrapolating it out to an anticipated holding period. Liquidity can tempt investors to mimic traders, thinking only in the short term. Nasty surprises can follow.

Is the answer to simply avoid anything that smacks of illiquidity? Marks isn’t taking that route. Staying only in liquid assets when an institution should be taking a long term view – where moment by moment liquidity shouldn’t be the primary driver – isn’t a sound strategy either.

Again, we urge our readers to consider the Marks paper.