I've enjoyed shorting biotech stocks over the last 12 years. In the latter half of those years the companies have generally been small. They have a strange tendency to crowd in the $300 to $500 million range.
There were some memorable larger ones in the last 6 years: Telik (TELK), Neurocrine (NBIX), MannKind (MNKD), Human Genome (HGSI). I missed a couple of big ones, too: Elan (ELN), Sequenom (SQNM), Dendreon (DNDN) were notable $1bn+ market cap shorts I failed to see (at least I wasn't long!).
However, the very best shorts (on a risk/return basis) are usually smaller. This is notable because as a hedge fund, you generally want size and liquidity in shorts. Many funds simply won't even trade stocks if they're above a certain market cap or if there is limited volume.
Mesoblast is a $2.5 billion biotech company that I think is worthless. With 95% downside and limited upside, the short Mesoblast position is one of the largest in my hedge fund(s). There is a catch, of sorts. Mesoblast is largely traded in Australia with the ticker MSB. There are ADRs - OTCPK:MEOBF and OTCPK:MBLTY - but they're not very liquid. MSB AU is very liquid however, and I suggest you find a broker which allows you to trade in Australia if you do not have one. All major international banks should allow it, and for the "home gamer", Interactive Brokers may allow you to do so as well (not sure!).
This idea is part of a broader theme for me: International stocks are less well "arbitraged" than United States stocks. In fact, my funds are largely focused on international stocks, where many US hedge funds won't play. Retail investors should do the same. The internet has allowed for us to do work on these companeis just as easily as we can in the United States. Other than a perception of a barrier, a (sometimes) language barrier and a (sometimes) time zone barrier, there are few reasons not to trade international stocks, and a lot of reasons to!
Mesoblast largely has one product, "Revascor", which are mesenchymal precursor cells used for Congestive Heart Failure. There is one trial that supports the valuation for Mesoblast a 60-subject Phase II study that showed Revascor patients did a lot better than control group patients. I will attempt to show that the results of this study cannot be relied upon and the phase III study that will fully interrogate the ability for Revascor to do what Mesoblast thinks it will do, will fail.
One of the salient results of the Phase II Revascor study was the mortality benefit seen at 6 months with Revascor. 4.4% of Revascor patients died at 6 months and 13.3% of placebo of the patients died at 6 months. That sounds good at first blush but these results are a bit abnormal.
Our team reviewed quite a few CHF studies and the 13.3% death rate for placebo at 6 months was very abnormal. The average mortality at 6 months in the studies we reviewed was 4.17%. So, our meta-analysis shows two things: The Revascor arm had mortality that was about normal, exactly what we see across many studies. The control group had a very strange high death rate. In fact, it is the highest 6-month mortality rate for a CHF study that we can find. It is not statistically reliable to compare outcomes across trials (the aggregate patient population could be much sicker), but it was telling to see these mortality levels are highly unusual if you assume the null hypothesis.
So why did one group do so much better than the other? In clinical trials, baseline imbalances are a good predictor of outcome for patients, especially in CHF, where ejection fraction, patient age, NYHA class all are reflective of patient disposition. The smaller the trial, the more likely that the baseline characteristics of patients were imbalanced. The more unblinded the trial, (this trial was single-blind), the more likely that investigator bias creeps into the results. The more the control and intervention arms are different, the more likely protocol bias skews the results. Yep, that happened here, too!
The mechanism of action of MSCs is highly dubious. It is widely acknowledged that these cells do not differentiate, downstream, into other cell types. This is not how these cells work. Even Mesoblast's investigators feel this way. So how do they work? They might recruit revascularization factors to the site of injury. They might send important signals to the real stem cells in the body. They might do nothing (this is my position).
Cephalon (CEPH) licensed Revascor. This means the $2.5 billion value you are paying when you buy a share of Mesoblast is not for 100% of Revascor, but for 50% (or less) or Revascor. So this drug is worth $5 billion according to the market. There aren't many Phase II drugs worth that much. I can only think of one or two that have had that much value ascribed to them in biotech history. Given the specious nature of this asset, I suspect there is little upside left to Mesoblast shares.
A lot of friends ask me why I'm short Mesoblast if there is no "catalyst" coming. In our office, catalyst is a four-letter word. Timing catalysts is a funny subject. I ask folks we interview when the best time to short a stock that you think is worthless is. Answers vary from "the day before the news that proves it is worthless" to "short a little bit everyday" to "wait for the top" and they get worse from there. None of those answers are too great. I don't have the best answer, either.
In general, the job of a stock speculator/fund manager is to short stocks that are very overvalued and buy stocks that are very undervalued. Timing is a tricky game, and I think Mesoblast has "topped". We've been good at picking tops lately, but our display of that skill might be fleeting.
Sizing is the more important element and it is independent of timing. Timing a trade maximizes your IRR. Sizing a trade is a risk question. We can't maximize our IRR on Mesoblast because we don't know when it will go down 95%. But we are reasonably sure we won't get destroyed shorting it, so as long as the IRR is good (about +90% annualized if it goes down that much in the next 12 months, +40% in 24 months, +260% in 6 months, etc), it's worth doing now. That's a relatively mathematical answer. The best trade is the one where you know it will go down and you know when. That is when you "go for the jugular" as Soros says.
I think we know this will go down (and not up) but we don't know when. That's okay. In general, shorts like this have a way of surprising you. A financial crisis here, a clinical hold there, and you reach your price target sometimes for reasons you didn't think would occur.
Shorting overvalued stocks often comes with free synthetic puts -- the overvalued nature of the stock allows for any negative surprise at all (even a commentary from Seeking Alpha) to move the arbitrage closer to fair value. My old boss had a way of saying this I particularly like: "expensive stocks go down, cheap stocks go up". It's a very "Tiger" mantra, but it works if you're good at spotting the very cheapest and the very most expensive stocks.