Issues with volatility ETNs
Volatility exchange-traded notes (ETNs) provide an easy access for investors to constant maturity VIX futures portfolios which, in turn, provide a long exposure to equity market volatility as measured by the VIX index. One of the most popular products, the TVIX ETN, has gained significant media attention since late February 2012, after the issuer (Credit Suisse) had discontinued creating new shares. As a consequence, positive premiums started building up and creating a market distortion that resulted in significant losses for retail investors in late March 2012, when the issuer implied that issuing new shares might resume.
In a forthcoming paper, we conclude that the distortion was created by factors specific to ETNs, with no relation to the particular exposure to a volatility index. The main factors suggested by the academic literature are the inefficient share creation process and the speculative motive of uninformed, return-chasing investors (see Wright et al (2010) and Diavatopoulos et al (2011)). The share creation process of ETNs differs a lot from that of exchange traded funds (ETFs). ETFs are generally characterized by a transparent and fluid share creation process which ensures that the price of the ETF remains close to the indicative net asset value (NAV). As any other exchange-traded product, the prices of ETFs are determined by the corresponding supply and demand. Thus, the price may deviate below or above the NAV. The indicative NAV is published intra-day and can be compared to the price of the ETF almost in real time. If an ETF appears to be undervalued compared to the NAV, then an arbitrageur can buy ETF units, redeem them at the custodian bank for the underlying securities and sell them on the market realizing a profit. Alternatively, if an ETF is overvalued, an arbitrageur can buy the underlying securities on the market, redeem them for ETF units, and sell the ETF units on the market realizing a profit.
As long as this mechanism is not limited by any regulatory or liquidity constraints, the price of an ETF remains close to its NAV. These two simple strategies are however not applicable in the case of ETNs (see Wright et al 2010) for which the creation process is controlled solely by the issuer. Thus, suspension of share creation in times of a significant surge in demand leads to accumulation of positive premium especially if the security may become unavailable for borrowing limiting short-selling activities.
As far as TVIX is concerned, short-selling intensified at the end of February and throughout March 2012 reaching volumes in excess of 4 times those of December 2011 which, however, proved insufficient to suppress the accumulation of positive premium. A reason why TVIX could have been impossible to short-sell even more intensively is the inability of investors to borrow the security in order to sell short. This intuition has been confirmed with practitioners from the industry and is also supported by opinions expressed in the public domain. Further on, there were indications for a short-squeeze as sophisticated market participants tried to short-sell TVIX but had to cover the short positions in an increasing market. Pengelly (2012) reports that, according to market participants, the short-squeeze could have been caused by algorithmic trading.
More generally, investors in volatility exchange-traded products (both ETNs and ETFs) need to be aware of two aspects. First, the underlying that the product is tracking does not correspond to the actual volatility index (e.g. VIX) but to a systematic strategy investing into volatility index futures. There could be substantial associated roll-over costs arising because of the contango of the futures term structure. Second, since the volatility index itself is derived from option prices, the degree to which it is representative of the volatility of the corresponding equity market index depends on the efficiency of the index option market. Generally, this might be of lesser concern for the US but should a volatility futures and ETN markets develop rapidly elsewhere, investors may be at risk if the corresponding option-derived volatility index is not based on a solid index option market and may, therefore, be not fully representative of the dynamics of the corresponding market index volatility.
Diavatopoulos, D., J. Felton, and C. Wright, (2011), The Indicative Value–Price Puzzle in ETNs: Liquidity Constraints, Information Signaling, or an Ineffective System for Share Creation?, The Journal of Investing, 20(3): 25-39.
Wright, C., D. Diavatopoulos, and J. Felton, (2010), Exchange Traded Notes: An Introduction, Journal of Investing, 19, pp. 27-37.
Pengelly, M. (2012), The ETN that grew too fast, Risk Magazine, available online at http://www.risk.net/risk-magazine/feature/2170072/etn-grew-fast.
 See, for example, the blogs “The TVIX short-squeeze” (http://soberlook.com/2012/03/tvix-short-squeeze.html) and also “Another ETN you should not be buying” (http://seekingalpha.com/article/768291-another-etn-you-shouldn-t-be-buying).EmailSharePrint