A recent structured note caught my attention because it a) offered great characteristics and b) was based on an underlying biotechnology stock. These characteristics suggest that note performance may be tied to an upcoming binary, highly volatile event Did investors pick up on this prior to putting $20M notional into this note and do they understand the implications as relates to the note?
Why are we interested?
Binary outcomes make for complex and interesting analyses. Today’s blog will focus on options and notes referencing stocks whose performance is subject to binary outcomes (e.g., Biotech companies facing drug trial results, companies with upcoming major litigation decisions, and so on); these investments require a very different kind of analysis than your everyday stock, so investors should tread carefully.
A typical stock, even a volatile one, typically gains or loses based on a shift in business and market perception over time. As an example, Apple (AAPL) was on top of the world until September 2012 — almost a year after Steve Jobs’ death — when investors became concerned about strategic direction and market share. Six months later, the stock price had fallen about 45%. There was no defined period upon which an AAPL investor (or short investor) expected a big move up or down in AAPL but it rather took some time and certain events to play out. More importantly for our discussion, on September 2012, no one knew with certainty that in 6 months the stock would be up or down some specific amount. The odds of an AAPL move from that day on September 2012 is best illustrated with the distribution curve below with future performance distributed normally around a number close to zero (technically, the forward price of AAPL would incorporate dividend and risk-free rates).
On the other hand, companies approaching a binary event, like Biotech stocks approaching a drug trial, show very different stock movements. These companies spend years and hundreds of millions in R&D developing potential blockbuster drugs, during which time they go through a number of trials with the FDA. Any one of these trials can result in success, leading to a much higher valuation as the drug moves closer to commercial success, or failure, which might wipe the company out. The end result is a binary outcome often tied to a specific day. Analysts are able to follow when trials are taking place and can predict how the stock would likely move on a positive or negative result. Investors are poised to act, en masse, the moment the news is out (or sometimes before it’s public – e.g., Martha Stewart!) so that stock price movement is dramatic. The potential outcomes are barbelled, or binary, as opposed to distributed normally around zero, as most stocks are:
As a result, implied volatility, which is a measure of future stock movements and assumes a normal distribution around zero, is virtually meaningless when binary outcomes are involved and options/derivatives and associated structured notes need to be analyzed in a very different manner!
Terms of the trade(s)
Today we will compare two similar products on high volatility stocks to show the impact when one is subject to a binary outcome.
The note I will focus on is an autocallable with a one-year term that references Vertex Pharmaceuticals (VRTX). Terms are as follows:
- 11% annualized yield paid monthly (0.9167%)
- Issuer call dates quarterly with automatic call if VRTX closes at or above initial level ($84.81)
- 45% barrier at maturity — if stock closes below $46.65, representing a 45% drop, investor receives VRTX return; otherwise investor receives return of principal
To put these terms in perspective, a similar product sold at around the same time referenced LinkedIn (LNKD) and delivered:
- 9.4% annualized yield paid monthly (0.7833%)
- Issuer call dates quarterly with automatic call if LNKD closes at or above initial level ($192.62)
- 20% barrier at maturity — if stock closes below $154.10, representing a 20% drop, investor receives LNKD return; otherwise investor receives return of principal
Terms of both notes can be found here: http://www.sec.gov/Archives/edgar/data/19617/000089109214001556/e57654_424b2.htm
Given the superior characteristics, little surprise that the VRTX structured note attracted close to $20M in assets compared with $1M for the LNKD product. Although that doesn’t make VRTX the better deal…read on.
LNKD happens to be a volatile stock with expected average movements of around 3% daily based on an implied volatility level of close to 50%.
An easy rule of thumb to find daily expected movement is to divide the volatility level by 16 — a close approximation of the square root of trading days in a year. Implied volatility is usually set by market makers through analyzing the historical volatility of the stock while taking into account the general market along with upcoming specific stock events. The implied volatility is then further adjusted based on order flow (supply and demand) and market and specific stock events. So if the volatility of the market is generally low, as in the current environment, odds are that barring any specific company event, the volatility of a single stock will be low relative to its long-run average. As an event such as earnings approach, volatility will typically rise a little as there is a higher likelihood of a one day move that is larger than typical and then it will revert to average levels.
There is obviously some chance that LNKD will decline by 20% prior to the maturity of the autocallable; however, there is no known future event pointing to this likelihood, making the 20% barrier seem pretty far away. Of course, if there is a downturn in the market the barrier can be breached, especially given LNKD’s high beta A surprise stock specific event such as Facebook entering LNKD’s space might also occur. Still, no one is currently expecting either of the above scenarios or a 20%+ decline in the next year. As a result, if someone wants to generate a 9.4% annual yield and take the risk that the barrier protecting from a 20% decline is a worthwhile asymmetric risk then I would say go for it with the caveats that they understand the effects of callables and have analyzed the note (links to previous posts).
The story is much different for VRTX. Judging from the pricing and activity of the options along with a quick review of what VRTX does, it is obvious that a large binary event is on the company’s horizon. Unlike LNKD and despite its very high volatility and assumed beta, VRTX movement will not really be correlated with the market because the upcoming event will generally overshadow most anything else.
When a large definable and binary outcome resides in the near future, volatility becomes irrelevant. There is no point in looking at past performance and calculating volatility because the impact of the upcoming event will dwarf whatever learnings may be gleaned by past smaller movements. Instead, the upcoming event should be analyzed to determine potential future valuations and probabilities.
Thus, if I told you that if VRTX is successful in their findings the stock will increase by 50% and if unsuccessful it will decrease 50%, it would be of no relevance whether Vertex tends to trade up/down 1%, 2%, or 3% each day – you would only care and be focused on that large upcoming move. If you had some insight into the company and had an opinion about their future success (or lack thereof), you would simply buy an option that gave you the best bang for the buck based on the $ amount. (For example, if the stock was trading at $100 and the $145 strike was $1, you might prefer buying those options rather than the $100 option for $20. If you bought ten (10) of the $145 strike you would spend $10 and make $40; if you bought one (1) of the $100 strike you would spend $20 and make $30.)
So instead of the volatility level dictating the $ value of the option, the $ value dictates the volatility level. And more importantly, as mentioned earlier, there is not a standard distribution of price outcomes but more of a barbell expectation, which further complicates both the value of a series of options and the models used to price both vanilla and more sophisticated options such as autocallables.
Let’s take a quick look at the series of options available on VRTX. Options that mature up to and including May 2014 show volatility of approximately 50% or below. Options that mature in June, however, show a volatility of 80%, and ones that mature in July show a volatility of over 90%! Reading the options tea leaves, it seems like news is expected to come out sometime in June/July, and as a result, the stock is expected to see a massive price shift. Just how massive? The stock is currently trading at approximately 84; July options show a lot of open interest on the 70 line and the 50 and 52 line, meaning investors believe these levels are in play. On the upside, there is interest on the 110 and 120 line. To summarize, the option open interest seems to point to a belief that the stock can move 50% in either direction based on a specific upcoming event this summer.
I would like to stress that I did not actually research VRTX to figure out what sort of event will occur or when. I am simply illustrating how we can deduce a potential event due to a large jump in volatility at a specific time, along with potential movement and probabilities, based on looking to see where investors of the options are trading.
Now investors in the VRTX autocallable have taken an asymmetric risk in the face of an upcoming binary event. Was this a good idea? Lets review: The first call date for the investor is on May 21st, which is just a few days after the May expiration. Through that period, the investor will receive a little less than 1% a month in interest (.9167%). Interestingly enough, the market is not expecting an event through the first auto-call date. Typically, when an event is on the horizon, barring rumors and leaks, the associated stocks will tend to have very low actual volatility as people are simply positioning for the large expected upcoming move. Thus, the stock should hover near the issue price and as a result, there is close to a 50% chance that this security will be called on the first call date.
If the product does make it through the first call date (i.e., on the first call date, the stock was trading below the price at issuance), it will be dramatically impacted by the news that the options market seems to expect prior to the second call date. Given the binary nature of the expected announcement and provided there is no delay, the investor will either:
- Be autocalled at the second call date (if the stock is up following the announcement) and have generated roughly 5.5% in coupons over the half year period, or
- Not be autocalled (if the stock is down following the announcement) and be concerned about a potential significant loss – if the stock is down following the news, it will most likely be very down. The investor will likely get the next two coupons and indeed earn the 11% yield, but will face the real prospect that the 45% barrier will be breached at maturity, exposing the investor to a significant loss
This investment involves a realistic upside cap of about 5.5%, no matter how well VRTX does, and very significant exposure to VRTX downside should the clinical trials fail, in which case the 45% barrier won’t look so far away (remember stock is trading at 84 and there is significant option interest at 50!). Said another way, almost all scenarios in which more than 5.5% in coupons are paid out involve a very steep decline in stock price, exposing the investor to significant potential losses.
The question I would pose to any would-be investor is: why does this asymmetric return profile make sense? At a minimum, one would have to be convinced that the prospect of the trial succeeding outweighs the prospect of it failing to a considerable degree – and if one believes that, then why not just invest in the stock or an upside call?
We’ve been focusing thus far on the investor’s perspective – as an aside, I would be interested in learning how the bank ran its models and put on the hedge for this transaction. To all you quants out there: feel free to chime in on impressions of how you would have gone about appropriately valuing the trade and how you would hedge this from the issuer’s perspective – are the characteristics not good enough? Too good?
In summary, there is significant risk involved with this position that investors should understand before entering into this trade. Autocallables and Reverse Convertibles are popular products headlining enhanced yields with barriers that seem far away, at least at issuance. However, as we point out, it might merit staying away from these types of asymmetric payouts when a known binary event is lurking within the time frame of the note.