Precidian Ups Ante With Active ETF Rebuttal

August 11, 2015

A few weeks ago, Eaton Vance took the unprecedented step of using the Freedom of Information Act to dig up a comment letter from the SEC about one of its competitors, New Jersey-based Precidian Investments.

The only reason it could possibly have for doing this was to convince investors (both in ETFs and in Eaton Vance’s stock) and fund issuers that the Precidian version of nontransparent active was dead in the water. That would leave Eaton Vance’s NextShares “exchange traded managed funds” concept the only person at the party taking dance-card reservations.

Doubling Down
The response from Precidian this morning is not to go quietly, but to double-down.

The core objections highlighted in the SEC letter surfaced by Eaton Vance were somewhat paradoxical. The SEC argued that the proposed “verified intraday indicative value” (VIIV) was too poor a proxy to allow for good market-making. However, simultaneously, too good a proxy would make it child’s play to reverse-engineer the underlying portfolio.

Similarly, the SEC was worried that the blind-trust structure proposed by Precidian created a class of investors with privileged—and abusable—information about the underlying portfolio of an actively managed fund using the system.

To me, both of these objections represented a failure by the SEC to actually understand what was being proposed, but hey, that’s on you, Precidian. It’s a filer’s job to make sure it communicates something well, and obviously, what they had “was a failure to communicate.”

Today’s re-filing seeks to clear up these misconceptions mostly by rewording and clarifying what was already in the filing. And this time, it’s bringing friends—the filing includes two papers by academics as backup to their case.

The Smartest Part Of All This

The first paper is from Ricky Alyn Cooper, a Ph.D. at the IIT Stuart School of Business, whom I’m familiar with because of his sharp work on high-frequency trading. His conclusion here is pretty clear, and entirely logical—there’s simply no way that the one-second fluctuations of an intraday portfolio value with a $20 handle is enough information to discern, really, anything about underlying holdings. At best, it will do what it’s supposed to do—give you a good idea of what you might be able to use as a hedge, based on how other securities are moving intraday.

In my opinion, it doesn’t even go far enough. The VIIC, as proposed here by Precidian, is a near-real-time estimate of the value of an ETF portfolio based not on the last trades in the underlying securities, but based on the bid/ask midpoint of those securities.

This is a vast improvement over the current process for calculating intraday net asset value, especially if the portfolio holds securities that don’t trade with lightning rapidity all day long. In the existing system, if an ETF has a stock that hasn’t traded in a few hours, that old, stale price gets propagated. For this very reason, every authorized participant I’ve ever talked to rejects INAV and simply recalculates it themselves.

More Transparency
Even better, funds using this structure will have far more transparency about how values are calculated than anything currently in the market. Here’s a juicy quote:


“Each Fund will employ two independent sources of pricing information. Each Fund will employ a verification agent and establish a computer-based protocol that will continuously compare the two data streams on a real time basis. … The specific methodology for calculating the VIIV will be disclosed on each Fund’s web site.

… if a Fund determines that the current quotations for a portfolio security are no longer reliable for purposes of calculating the VIIV, such as in a situation where an Exchange institutes a trading halt in a portfolio security or if a security is otherwise deemed to be illiquid, that fact will be immediately disclosed on the Fund’s web site, including the identity and weighting of that security in the Fund’s portfolio, and the impact of that security on VIIV calculation, including the fair value price for that security being used for the calculation of that day’s VIIV. Applicants believe that this mix of information will permit market participants to determine their own fair value of the disclosed portfolio security, and better judge the accuracy of that day’s VIIV for the Fund.”

That’s a big chunk of text, but this is actually a higher standard for fund accounting transparency than is currently employed by supposedly, fully transparent ETFs.

When China shut down half the A-share market a few weeks ago, it took significant legwork to even understand whether or not a given ETF was affected, much less with a list of affected securities. And even a boring fund that tracks non-U.S. securities faces the constant problem of its underlying securities being closed during the U.S. trading day.

Frankly, I’d love this VIIV standard to be the norm in ETFs, not just for nontransparent funds.

Fact And Fiction: Arbitrage And The Blind Trust

The second academic piece in the new filing is from Craig Lewis at Vanderbilt, who’s hardly a punter when it comes to evaluating pricing and arbitrage. The piece serves more as a layman’s explanation of how ETF arbitrage works and how the blind-trust structure is actually designed to work, which is probably necessary given the SEC’s lack of understanding the last time around.

The most critical piece is his explanation of how all market participants will interact with the ETFs, and it sheds light on a lot of common misunderstandings about how ETF arbitrage actually works.

Right now, in a traditional ETF, authorized participants make money by arbitraging out the difference in price between what the market will pay for ETF shares and what they can buy up the underlying basket for and deliver to the ETF issuer.

For a simple ETF with a few dozen stocks, that math is easy: If the ETF is trading for even a tiny bit more than the basket of stocks is worth, they buy up the basket in its entirety, sell-short a creation-unit’s worth of ETF shares (often 50,000 shares) and at the end of the day, deliver the stocks, knowing they’ll get shares to make good on the short-sale of the ETF itself.

In the proposed structure, it’s actually even easier. If the ETF is trading above its fair value (now made much easier to determine by the VIIV, at least for U.S.-listed underlying securities), they can sell-short those shares and simply hand the blind trust over the cash proceeds of the sale, knowing that they’ll get new ETF shares to make good on the short-sale in the same way they always do.

Buying The Blind Trust
Instead of buying dozens or hundreds of stocks, they just press a button and buy more of the blind trust. The blind trust, in turn, goes and buys whatever stocks the actual ETF issuer wants that day.

That part’s relatively easy. The part I think most folks miss is that you don’t even have to be an AP to help out in this process. As the paper points out, regular-old market makers can open an account with the blind trust too, and simply use it as a perfect hedge for the value of the ETF shares.

And critically important, nobody has access to any abusable information here. The blind trust only buys and sells pro-rata slices of a portfolio. They don’t tell anyone what that portfolio is—that’s the “blind” part.

The AP and other market makers get to interact with the blind trust to do arbitrage trading, and the whole market gets to see the value of both the ETF shares and the VIIV.

My Point (And I Do Have One) …

The structures proposed by Eaton Vance and Precidian are both complex. They each engage in structural engineering to build a bridge between the old-world of stock-picking active managers and the modern, millisecond-trading world of index-based ETFs.

To be perfectly clear, I remain skeptical that either one of these structures is actually solving a problem individual investors actually have. After all, the latest SPIVA data from S&P suggest active managers are doing just as bad a job as ever versus low-cost indexing.

But this isn’t about whether I believe active management works. It’s about whether it’s possible for nontransparent active to live inside the ETF wrapper.

Neither of the proposed structures is perfect. Precidian adds complicating layers to the arbitrage process. Eaton Vance asks the brokerage industry and the entire investing world to adopt an entirely new way of trading. It’s the price you pay for letting your portfolio manager keep his cards hidden.

But in my opinion, the Precidian structure just works.

Yes, the blind structure complicates the arbitrage mechanism, but the market has a really good way of dealing with that—prices and spreads. And the market also has a great way of telling fund managers whether they like their products—they buy them.

We should let the market decide. If the institutional community likes what they see, and can establish the blind-trust arbitrage linkages effectively, prices will stay in line and spreads will be tight. If I’m wrong, and there’s huge pent-up demand for active management, well, investors will pile in and prove me wrong.

And if we can get VIIV as the new standard for intraday pricing? That strikes me as a win for everyone.


Dave Nadig is the director of exchange-traded funds at FactSet Research Systems Inc. You can reach Dave at [email protected], or on Twitter @DaveNadig.

Find your next ETF

Reset All