When you trade ETFs, you have three days from the trade date to settle the trade. Simply speaking if you are the buyer, after three days you get the ETF and have to pay the cash. If you are a seller, you have three days to deliver that ETF that you own to the buyer. Unless of course, you are a market maker, or authorized participant for the ETF, in which you have six days to settle up, and not three.

Is this an issue? It depends on who you ask of course.

This topic first became an issue back in 2011, when the EW Kaufmann Foundation put out a report called Canaries in a Coalmine (we reference it below), and testified before the US Senate Banking Committee that:

Time has proven that shorter settlement periods and high levels of compliance are the best antidotes for systemic risks that might involve the failure of a very large trading party. Why would such broad indexes with supposedly instant arbitrage characteristics fail to deliver in such a significant manner? We fear that hedge funds and commercial banks may be relying on lax enforcement of settlement rules to create a cheap funding source for their trades.

A cursory analysis of trading volumes in IWM component securities indicates it would take more than 180 trading days, or more than six months, trading at 10 percent of each stock’s volume every day, to offset reported short interest in that ETF.

Repeatedly, the industry has countered that fails in ETFS don’t really matter. However this summer, FTDs in ETFs have spiked again, and the question has been raised again. We are not sure what to think; we only know we should all examine the issue more closely. Below are a few arguments that the ETF FTDs are NOT AN ISSUE, as well as a few arguments that they indeed ARE AN ISSUE. You be the judges.

NOT AN ISSUE

If you ask a market maker for the ETF they will tell you it is not an issue, as they settle virtually all trades within this 6-day window they have. They will tell you that you should ignore the very high rates of ETF FTDs (failed to delivers) for this reason.

If you ask an ETF issuer, they will tell you that it is not an issue, as it is a result of the complexity of the Create-Redeem process; the market makers need more time to settle ETF trades because of this process, which involves cooperation between the ETF issuer and the authorized participants as they make and dismantle “bouquets of flowers.”

If you ask Georgetown Professor James Angel,  he will also tell you this is a non-issue. Although he does acknowledge that there is something not exactly fair about ETF fails forcing an involuntary stock loan onto a participant.

 ABSOLUTELY AN ISSUE

If you ask Barron’s, they will tell you it is definitely an issue, which is what they already told you in this 2011 article titled Market Maker’s Edge: T+6.

MARKET MAKERS…. discovered that they can make a predictable return by delaying the purchases and selling you nonexistent exchange-traded fund shares that they will create later. The wiggle room allows the ETF crowd to engage in hedging and arbitrage strategies that help pay their overhead, using your stocks or funds. If T+6 were to disappear, some of them might disappear, too. This might result in the type of reduced liquidity and heightened volatility we see in equity markets, where robotic trading machines have displaced human middle men. So, then, why fiddle with the settlement system? Because time equals risk.

Barron’s makes the point that the T+6 loophole is a profit center for market makers:

The longer a market maker can delay buying the stocks for a creation unit, the longer he can postpone paying the “creation fee”—the associated transaction costs. Nadig says that the savings on the fee, based on a percentage of the stock price, might be minuscule—just 0.79 of a basis point, or less than 1/100th of a percentage point—per day. But it’s free, and if you are dealing with hundreds of millions of dollars, those pennies add up.

If you ask Fred Sommers, who co-authored this critical report back in 2011, Canaries in a Coalmine, he will also tell you that these fails are indeed an issue, and that they cause instability, an illusion of liquidity, and very serious systemic risk for investors. He states:

“Every fail introduces cumulative and potentially compounding liquidity risk into the orderly process of settling the $7.5 trillion of securities transactions completed each day, which could be especially dangerous during times when financial institutions are short of liquidity.”

Sommers’s point is that the extra time to settle these ETF trades could be an eternity in times of market stress. Counterparty risks at those times become especially real; just ask Bear Stearns.

Finally if you ask some in academia, they will also tell you this is an issue. This recent late 2012 paper, Exchange-Traded Funds, Fails-to-Deliver, and Market Volatility, examines ETF fails, and details how the degree of ETF fails is tied closely to the options market, and that ETF fails spike dramatically on triple witching Fridays, and cause increased volatility the following week.

Given the large dollar value of ETF trading relevant to the rest of the market, our results may contribute to an explanation of the finding that market index volatility has increased since the 2008 financial crisis (Sullivan and Xiong, 2012).

Recent research suggests that, for some ETFs, discrepancies between share price and NAV may be large and persistent (Shin and Soydemir, 2010). Deviations from NAV are partly explained by inefficiencies and time inconsistencies in the share creation and redemption process. ETF deviations from NAV may also result from either sudden changes in component stock prices or stock borrowing constraints that limit the ability of market makers or arbitrageurs to close price gaps.

The temporary market making close-out exception to SEC Rule 204T may create situations in which FTDs lead to more market volatility. Rule 204(a)(3) .“permits a borrow as well as a purchase to close out a fail to deliver position.” (SEC, 2009, page 39). This provision is relevant to our research because a significant and growing segment of ETF trading concerns so-called .“borrow-to-create.” and .“create-to-lend.” transactions.

So which is it? Are ETF Fail-To-Delivers an important issue that imposes systemic risk onto investors in the market? We all need to do more research. We tried to point out a few pro-con points of view above as a starting point for us all to study up on this trillion dollar market.

What we do know is that ETFs have become a huge portion of our market; they are responsible for half the trading volume. And investors really know precious little about how they work. They assume that if they own an ETF, they own a basket and are precisely exposed to market moves. A) This is not always the case, and B) there are risks above and beyond those market moves – namely counterparty risks – that are not fully understood.

We hope that Regulators will be ahead of the curve on this issue, for it is in times of stress in the market that severe price dislocations can and will occur, and when these dislocations happen investor confidence plummets.