Startups, Exits, And Ecosystem Flux: Bullish For Biotech

Posted in Biotech financing, Biotech startup advice, Exits IPOs M&As, General Venture Capital, VC-backed Biotech Returns | 4 Comments

The world is awash in cool new tech startups and poised for “A Cambrian Moment”, according to a recent special report from the Economist. For those that haven’t read it, it’s a very interesting set of articles about the trends shaping venture formation in digital media and software: the ease of starting new companies, the proliferation of accelerators, better and more transparent markets for funding like that created by AngelList, the platformization of everything, etc…

But in some ways I was more intrigued by a critical review the Economist recently posted by Daniel Isenberg where he focuses more on the “startup glut” in technology and digital media, and in particular the challenge of scaling a business when the “on-ramp” for emerging companies is so congested.

With huge winners achieving massive scale, like Alibaba and its pending IPO, Facebook, and Twitter, it’s easy to see why there’s a rush to create the next big one.  Barriers to entry are nearly zero in many cases, and speeds of product iteration are incredibly fast. This has led to the startup glut that Isenberg and many other pundits refer to – tons of startups, hyper-competitive product markets, but irrational group-think optimism around “everything is awesome”.

The nature of this evolving and exploding Tech market is in striking contrast to Biotech, where the rate of funding and new company formation have been remarkably steady for years (most of the last dozen years since the Genomics Bubble burst).

Here’s a snapshot of comparing the normalized levels of overall VC funding as well as the number of newly-backed startups (using the NVCA/PwC Moneytree data around “First Financings” as a proxy) in software and biotech:

Quarterly Normalized Trends - Software and Biotech

While biotech has been relatively constant on these funding and startup metrics, software has experienced a massive boom.  Tons of money is flowing in, and huge numbers of new startups are forming.

This dichotomy got me thinking about the relative “flux” through the two venture ecosystems.  Here’s the conceptual model: on one side, we have the supply of new startups – entrepreneurs, ideas, and capital coming together to create new ventures.  These flow into the ecosystem, and either succeed in growing, raising further capital, and creating something innovative enough that others will value, or they fail and go out of business.  If successful at creating something of value, this demand creates “exit” events as they depart the “venture” ecosystem – either IPOs where companies attract new investors to help them scale, or M&A events where they get integrated into larger businesses.  This balance of the supply and demand creates the equilibrium present in the venture ecosystem around a sector, as depicted in the figure below, and directly shapes the aggregate returns of the sector in the asset class.

Conceptual Model of Flux

As a biochemist, I didn’t get much past ECON-101, but supply and demand curves made sense to me, and what changes or imbalances can do to markets and sectors. Understanding how the respective supply of startups and exits has changed over the past few years should help better understand the future – and give a sense for the relative trends around the attractiveness of sectors to venture investments.

Biotech as an investment sector has improved dramatically in recent years: it’s a supply-constrained startup market with strengthened intrinsic demand.  As the data above suggests, the supply of new startups and funding has been largely constant (S).  But it’s also clear that the demand for biotech innovation has gone up both in terms of the continued expansion of the IPO market and robust BioPharma acquisitions of VC-backed biotechs (shift from D to D’).  In contrast, at least at the macro level, software as a market has gone the other way: a massive proliferation in the supply of startups and funding (S to S’), but relatively steady demand by the M&A and IPO markets over the past five years (D) (see NVCA Yearbook for data).

Supply Demand Curves

Is there any way to quantify this trend?  Probably not in a robust manner, but a directional measure would be to look at what I’ll call the “Startup Ecosystem Flux Ratio”, or kexit/knew, calculated simply as the number of exit events (M&A, IPO from NVCA data) vs number of new VC-backed startups (using the NVCA/PwC MoneyTree metric of “First Financings” as a proxy) in any given period of time.  While the absolute numbers are largely meaningless, the change over time in a given sector sheds light on how the supply/demand balance is shifting – so I’ve chosen to normalize both sectors’ flux ratio to zero in 2010.

Startup Ecosystem Flux Ratio

The results quantify the striking shift: innovation-rich biotech has been becoming a far more attractive sector to invest, where long term price equilibrium would be predicted to go up.  Indeed, simply looking at valuations in the recent IPO window and mezzanine financing market suggest this chart is clearly directionally correct.

Importantly, this is not meant to compare the relative current attractiveness of Biotech vs Tech, but merely understand the trends within each of those sectors over the past few years.  At the market level, this analysis says Software has gotten or will get tougher to pick winners. With rates of new Tech startups vastly outpacing exits (knew>>>kexit), it suggests one of three things has to happen. Exit demand could pick up and balance the equilibrium, but more likely is either that the Tech ecosystem expansion will continue or that the aggregate loss ratios in software will have to go up (more companies will fail to generate good returns).  I suspect some of both of the latter will occur, but only time will tell. Bubbles on the supply side will either burst, or they will alter the steady state equilibrium of supply/demand for new companies and their innovations.

Back to Biotech: a few questions arise from these observations, all of which deserve entire blogs on their own:

Why is biotech startup supply so constrained?  First, there aren’t dozens of breakthrough biomedical ideas created every day; while substrate for startups is very rich, figuring out which are likely translate successfully into high impact medicines diminishes the viable number of big, attractive ideas quickly. Second, there are very few biotech venture investors still active today, and fewer still that are focused on early stage company creation.  Third, and of critical importance, it’s not easy to start biotech companies, and in most cases requires entrepreneurs with decades of apprenticeship inside larger R&D organizations – navigating the drug discovery and early development process requires experience well beyond simply advanced degrees (MD/PhD).  Seasoned veterans willing to take startup risks are relatively rare and highly sought after.  A 40-something (or older) entrepreneur with kids (responsibilities) and a solid corporate career is a very different animal than a 20-something in flipflops that likes to code along many dimensions – risk-taking, salaries, culture, type of coffee they like, etc. These three forces all conspire to constrain the number of new startups by keeping the frictional startup costs and barriers to entry relatively high.

Why has the demand for biotech innovation been so strong?  A variety of reasons.  Beyond the macro support for booming global healthcare demand, the two big buyers of equity from emerging VC-backed biotechs, the public market buyside and Big BioPharma, are both as strong as they’ve ever been. The broader biotech public markets have done well, and driven further demand for new issuances (IPOs), as discussed here in 2013 (the drivers all continue today).  The public market has also matured: dozens of biotechs have now launched products and become commercial entities (PCYC, REGN, INCY, NPSP, etc).  And the second big buyer, Big BioPharma, continues to source a significant portion of its pipeline externally (>>50%), in particular from venture-backed biotechs.

What’s likely to be the trends going forward?  This is, of course, the billion-dollar-question.  Rates of supply (new startup formation) in biotech aren’t likely to increase dramatically anytime soon, as none of the three contributors listed above are easily expanded.  After the culling of the venture lemming-herd over the past decade, LP’s appear to be taking a renewed interest in backing Life Science VCs, but this will take years to make a big impact on the sector’s startup fund flows.  On the other side, I don’t believe that demand for new, high impact medical innovation – which will feed Pharma pipelines eventually – is also likely to reduce over time.  In short, intrinsic factors around the nature of innovation favor continued strong demand forces.

So as you might expect, I’m bullish on the long-term equilibrium favoring the continued (and increasing) attractiveness of Biotech as a sector in the venture asset class: tightly-funded and supply-constrained, with exit demand outpacing investments, Biotech has the right macro fundamentals to be a good place to be investing.

Against that backdrop, we’re excited to be focused here at Atlas on increasing the supply of high impact, innovative biotech startups.


Biotech Analyst Optimism: Price Targets Post-IPO

Posted in Biotech investment themes, Exits IPOs M&As | 7 Comments

Getting high quality analyst coverage of your company as a recently minted IPO is important for communicating the rationale and excitement around your story. Thoughtful analysts can evangelize (or punish) companies they believe in (or not). But understanding the relationship between a forecasted “price target” of a stock and its current share price has always puzzled me.

Before reviewing some recent data, here’s some background: most analysts build their valuation models to reflect the disease area’s market size, share of patients addressable, price per therapy etc, and then discount these back in various ways to incorporate pipeline attrition expectations and such: NPVs with high discount rates, forward P/E ratios, etc…  These models vary in their sophistication, and there is wide heterogeneity in analyst quality.  In general, the early analysts that cover a stock are related to the investment banks that helped underwrite and manage the offering, though the linkage is now more tightly regulated post-Sarbanes-Oxley.

Analysts have to adjust their price targets upon any “new” news: a positive clinical readout or other good event (and a risk-related discount on the value of a stock has been removed), they typically go up, and vice versa.  Since there aren’t real revenues and earnings to go off of, a lot of “future value” and sentiment is baked into these price targets.

To try and understand the relationship of price targets and current prices in biotech today, I examined data from ~60 or so therapeutic VC-backed biotechs that went public since January 2013, and have at least three analysts covering them with price targets.

The striking, although not surprising, summary data conclusion is that the differential between the average analyst 12-month price target and the current stock price is often quite considerable: for all the IPOs since January 2013, its 90%.  So the mean target price is nearly double the current price.

Below is the breakdown of various IPO cohorts plotted against the ratio of mean analyst target price versus current stock price (TP/P ratio as of 8/27/2014) , compared to a basket of larger Biotechs (Gilead, Celgene, Biogen, Alexion, Vertex, Biomarin, Pharmacyclics, and Amgen).

Analyst optimism_Biotech_Aug2014

A few observations:

  • Most of the biotech offerings since June 2013 have a 12-month price target that is 2x higher than the current price, with a few companies (Max) having mean targets as high as 3x higher than the current price.
  • As you might expect, the ratio for more seasoned larger cap biotech companies is small – on average just 11% above the current stock price.  More coverage, deeper understanding, less price target differential – a rather obvious point, but nice to see in the data.
  • Interestingly, even over rather short time periods of 2011-2014, a meaningful trend exists that the longer a stock “seasons” as a public company the closer the target price to current price ratio becomes.

Why is this last point the case? 

Could be a few things.  Either analysts get smarter on these stocks over time (and adjust their forecasts appropriately up or down), or companies’ stocks perform and approach their price targets, the latter being a function of the market valuing the company “more in line” with the analyst’s forecast.  I don’t have the longitudinal data to understand how analyst price targets have changed over time, but given the outperformance of biotech in general over the past three years, I suspect that, while analysts undoubtedly get smarter every day watching these stories progress, it is probably a reflection of both.

How does all this compare to the past? 

A 2006 paper by Mark Bradshaw of Harvard and Lawrence Brown of Georgia State reviewing analyst price target performance is of interest (here).  They reviewed 100,000 12-month price targets by analysts across all industries from 1997-2002.  I realize this is an old sample set and not specific to biotech, but their findings were interesting.  The aggregate dataset across the period showed that the ratio of target price to current price was rather tight around 135% (i.e., targets were 35% higher than current price).  Further, 24% of stocks hit their target at the end of the forecast horizon, and 45% hit the price target at some point during the 12-month period.  Analysts were, therefore, more optimistic on average than they otherwise should be, and most were unable to persistently perform well in forecasting.

Reflecting back then on the Biotech IPO dataset, one thing is very clear: analyst price targets are much higher than for more seasoned stocks.  A ~100% premium on average is well above the large Bradshaw dataset (which was during the first bubble, btw), and well above larger cap Biotech stories.  This premium presumably reflects several things:

  • Analysts are overly optimistic about “shiny new toys” (as we all are, especially VCs), and adjust their forecasts over time as management teams and their drug candidates perform.  It is fair to say, however, that a good analyst who has done their homework knows a company and its drug portfolio far better than many public investors (especially retail investors lacking institutional support).
  • Given the short trading histories, the overall market has had less time to find an equilibrium price point accounting for expectations of performance, a more fulsome understanding of the drug candidates, and a “permission to believe” that what a management team is saying is truly likely to happen.  All these things affect market sentiment
  • Lastly, these stocks are very illiquid and presumably trade below their fair value because of it – the concept of the illiquidity discount – which keeps lots of potential investors out of the market for their shares.  The average trading volumes, even after the lock-up expires, remain very thin until “big event” days when huge amounts of shares can move.  This makes price targets and market equilibrium concepts challenging.

I’ll close with a thought experiment as an optimist: if, in line with historic data, 45% of the current IPO class hits their price targets at some point over the next 12-months, which implies a doubling of many recent biotech IPO’s stock prices, that would certainly be quite the year ahead for the biotech market.  There are, of course, other possible futures, but that one is particularly intriguing.