Sometimes You Have To Beat A Dead (Exchange Traded) Horse

| About: iShares iBoxx (HYG)

Summary

Sometimes you have to beat a dead horse.

Common sense has officially been abandoned and left for dead in the ETF market.

Once upon a time, there was a guy named Mark Adams.

Sometimes you have to beat a dead horse.

And it's not because you necessarily want to. Rather, it's because you're looking out for the greater good. That's the case with ETFs.

For whatever reason, there are still droves of money managers and investors who don't seem to understand the problem. Take this guy, for instance:

That's Jose Garcia-Zarate. Jose is Senior ETF Analyst for Morningstar's European Passive Fund Research and he doesn't believe in me or my Labradors. Here's an excerpt from a recent piece on etfstrategy.com.uk (this is a lenghty block quote, but bear with me):

A new study from investment research house Morningstar, has found that high yield bond exchange-traded funds play a significant role in relieving market stress during volatile periods. The report was produced to challenge claims that ETFs may actually play a damaging role in the functioning of bond markets.

The study found that ETF trading was significantly more frequent in the secondary market compared to the primary market: the median secondary/primary market ratio for the iShares iBoxx $ High Yield Corporate Bond ETF (NYSEARCA:HYG) ranged between 5 and 8. This indicates that between 5 and 8 US dollars were traded between holders of existing ETF shares in the secondary market before the ETF manager had to purchase or sell a single US dollar worth of high-yield bonds in the primary market to meet creations or redemptions from the fund. During the same period, the median ratio for SPDR Barclays High Yield Bond ETF ranged between 3 and 5.

It also revealed that in times of market stress in the underlying high-yield bond market, secondary/primary market ratios for the ETFs spiked considerably above their median range values. This indicated that heavy trading of ETFs in these periods had been largely netted off between holders of existing ETF shares, without unduly impacting the liquidity of the underlying high-yield bond market.

"Our analysis suggests that far from being agents of instability, high-yield bond ETFs have acted as a safety valve, allowing investors to express their investment views without unduly affecting the underlying market," added Garcia-Zarate. "The bulk of secondary trading is regularly netted off between buyers and sellers of existing ETF shares.

Got all that?

Ok, so it should be immediately apparent what's wrong with that argument. But in case it isn't, allow me to give you an example of that very same flawed reasoning. I'm going to tell you a story now. I hope you enjoy it.

Once upon a time there was a guy named Mark Adams. Mark was no dummy. He has a Bachelor of Laws degree and two Master of Laws degrees (Securities Law and Business Law) from Osgoode Hall Law School to go along with an MBA from the University of Toronto's Rotman School of Business. He began his career at DBRS in the 1990s, was an original member of the ratings agency's structured finance group, and quickly became an important advocate for both asset-backed commercial paper and the use of CDO technology in Canada.

In a note from 2005 entitled "Credit Derivatives, CDOs, and the Canadian Asset-Backed Commercial Paper Market," Adams and his team made the laughable claim that "transactions in Canadian conduits [were] structured to be conservative" and recommended that Canadian banks increase their participation in the market for structured credit products.

Adams was also an early defender of DBRS' policy on rating asset-backed commercial paper that did not enjoy global-style liquidity protection (basically a no-questions-asked guarantee bestowed upon a special purpose vehicle that guaranteed payment to investors in the event the paper issued by the vehicle became illiquid).

In 2001, he famously penned a 19-page note that carried the rather unfortunate title "Canadian Liquidity Facilities: The Preferred Alternative." In it, Adams laid out the case for the use of contingent liquidity provisions (as the name implies, these were not no-questions-asked), justified his employer's treatment of the facilities, and in effect cleared away much of the concern that surrounded the deals. While the benefit of hindsight always makes it easy to pick apart a position which history has proven incorrect, the precision with which Adams described the meltdown that would occur five-and-a-half years later and the confidence with which he summarily dismissed the possibility of such an event occurring is difficult to ignore.

Critics of Canadian-style liquidity "suggest a situation involving a timing mismatch and a difficulty finding purchasers for new asset-backed commercial paper," Adams said, adding that in this hypothetical case, "the conditions for a market disruption under the liquidity agreement have not been met" (if there was a so-called "market disruption event," banks would pay out investors rather than the conduit, or special purpose vehicle that issued the notes or, commercial paper).

He continued: "Assuming the underlying assets have a long-term duration, investors are then left with assets that only pay out over a longer period of time and the [conduit] cannot provide investors with timely payment." This of course is exactly what happened in August of 2007, but Adams essentially claimed that such an event was well nigh impossible: "The practical reality is that the market has not experienced these conditions individually, let alone in combination."

He was willing, however, to entertain the notion (if only to tear it down) of such a series of events unfolding and he listed several ways the Canadian conduits might "handle" the situation. To begin with, Adams directed readers' attention to a few sentences which appeared earlier in his note: Should conditions deteriorate in the manner described, "all of the points made in the paragraph above" apply, he said.

Adams must have believed readers had very short memories because the "points" to which he was referring were prefaced with the following sentence: "First, on the maturity of the asset-backed commercial paper, the [notes are] almost always rolled over, provided purchasers are available". Of course one of the central assumptions in the adverse scenario is that purchasers are not available, thus any suggestions which assume the presence of buyers were simply not applicable to a discussion of the worst case scenario. That said, one of the arguments here is at least worth mentioning because it introduces the notion of "credit enhancement" as it relates to the CDO structures backing the paper.

If we assume -- as Adams did -- that there would always be at least some buyers in the market, then any funding gap that might arise would in all likelihood be manageable. More specifically, most of the principal owed to investors at maturity would be paid for by issuing new notes to buyers and any remaining amount due (the interest rate owed to investors for instance) could be funded by liquidating a small portion of the assets backing the paper. "It will be an unusual case where the [underlying] assets are not sufficiently liquid to support the [conduit's] funding costs," Adams said in 2001. Put simply: some portion of the underlying assets would almost always be liquid and that portion would almost always be sufficient to cover any amount that could not be funded by issuing new notes.

In the event that all of the underlying assets were illiquid and a shortfall existed between the amount raised from issuing new paper and the amount needed to pay off investors whose paper was maturing, Adams contended that the conduits could tap the "credit enhancement" mechanisms built into the structure of the deals to cover the difference. Such mechanisms included "overcollateralization" (i.e. when the value of the assets backing the paper exceeds the face value of the notes issued) and letters of credit. Here's Adams again: "If … the assets are insufficiently liquid, the [conduit] would likely be able to draw on the credit enhancement for the transaction, including any interest rate spread, asset liquidity from overcollateralization or a letter of credit for the transaction." As S&P noted in 2002, however, credit enhancements were designed to "sustain fundamental credit quality," not to ensure payments to investors at maturity. "It is conventional to use liquidity facilities to deal with liquidity risks, and to use other sources of adequately sized credit enhancement to absorb credit risks," S&P said at the time.

Even if we leave aside S&P's criticism, Adams's argument still assumes there are some investors available. Tapping a deal's credit enhancement and/or selling the liquid portion of the underlying assets will by definition be insufficient to cover the entirety of investors' principal payments at maturity, so one of two things must be the case if the notes are to be paid in full: Either some new paper is sold or market conditions meet the criteria of a "general disruption" and the liquidity providers intervene to pay off the investors.

Recall however that the real issue at stake is the possibility that the pool of buyers dries up in a market where the conditions do not justify the use of the term "general disruption." Despite the fact that this is indeed the scenario Adams was attempting to address, he seemed to have found the idea of it so difficult to accept that he simply refused to consider it in a straightforward manner. In other words, each and every alternative he discussed either explicitly or implicitly assumed some level of demand for the problem debt. This point will become clear shortly.

Once Adams was satisfied that readers had given adequate consideration to the possibility that in a challenging market, the liquid portion of the underlying assets could be sold-off and the credit enhancement structures could be tapped, he attempted to address what might happen should those measures be insufficient to remedy a shortfall. First, Adams said, it is rarely ever the case that there are literally no investors. More common is a scenario wherein sellers simply do not want to pay the premiums demanded by the interested buyers. "The point is not whether investors can be found but what spread or interest rate has to be offered to attract investors," Adams asserted, adding that in "an efficient market, the [conduit] will be able to obtain investors, albeit at a higher cost of funds."

It is important to take a step back here and note that at this juncture, Adams had failed twice to address what he claimed to be addressing, namely a scenario where "there is a difficulty finding investors." In his first attempt, he simply referred readers right back to the preceding paragraph in his note which contained arguments that depended on the existence of buyers and then, in his second attempt to address the issue, he immediately dismissed the feasibility of the problem by emphasizing what a "rare" occurrence such an event would be.

Do you know what happened to Mark? Well, let me tell you.

He founded Edenbrook Hill Capital which, with the help of broker-dealer Dundee Securities, launched Skeena Capital Trust (named after a river in British Columbia) in 2006.

A little over a year later, Skeena was left staring into the abyss when, on August 15, 2007, the conduit was able to sell $176 million in notes to investors. The problem: $249 million worth of paper had matured and needed to be paid back. The trust was able to convince three of its liquidity providers (ABN Amro, HSBC, and Bank of Nova Scotia) to advance around $40 million, leaving it $33 million short of the amount needed to pay off the maturing notes. No procedure existed for paying out noteholders at maturity unless 100% of the paper was funded. In other words, the $33 million shortfall effectively negated the $216 million in available funds. Because all of the money wasn't there, no one got paid.

Do you see where Mark went wrong in his reasoning? When asked to explain how the market would function if there was no liquidity, each and every explanation he gave, no matter how tortured and lengthy, implicitly assumed there would be some semblance of liquidity. He simply refused to address the real question.

The bottom line was that the conduits (special purpose vehicles) that issued the commercial paper were perpetually borrowing short to lend long. It was a horrific case of maturity mismatch. If the paper stopped rolling, that was it. Game over.

Morningstar's Garcia-Zarate is making the exact same mistake. Note what he says: "The bulk of secondary trading is regularly netted off between buyers and sellers of existing ETF shares."

Yes, it is. Here's the diagram:

(Graphic: Barclays)

And here's etfstrategy.com summarizing:

This indicates that between 5 and 8 US dollars were traded between holders of existing ETF shares in the secondary market before the ETF manager had to purchase or sell a single US dollar worth of high-yield bonds in the primary market to meet creations or redemptions from the fund.

But the question is: "what happens when that's not the case?" That's the crux of the whole argument. You can't very well answer that question by effectively saying: "well, that's always been the case."

It had also "always been the case" that home prices went up.

In the past, the answer was that Wall Street would inventory the bonds. Now you can argue that Wall Street will be the first ones running for the door when things go awry all you want, but the numbers speak for themselves. Here's yet another look at dealer inventories:

(Chart: Goldman)

Sorry, but there's no one home and there are even fewer people home today than there were just three years ago. And trading costs are going up…

(Chart: Goldman)

...as ETF AUM continues to climb…

Click to enlarge

(Chart: Deutsche Bank)

In the summary of the Morningstar study excerpted above, it says that "between 5 and 8 US dollars were traded between holders of existing ETF shares in the secondary market before the ETF manager had to purchase or sell a single US dollar worth of high-yield bonds."

To that I say: "Yes, but for the thousandth time: who is going to buy the bond if it ultimately has to be sold?!"

Maybe Mark Adams knows.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.