Biotech VC Returns: Diversified Funds vs. Healthcare-only Funds

Posted in General Venture Capital, VC-backed Biotech Returns

Over the past decade biotech investments made by diversified venture capital firms have outperformed those by healthcare-focused firms.  Its unclear why the difference exists, but the historical data suggest it’s a striking differential.

Examining the NVCA Benchmarking Database powered by Cambridge Associates, biotechnology and pharmaceutical (collectively “biotech”) investment returns by Healthcare-only VC funds vs Diversified VC funds were compared.  Healthcare-only includes firms that do biotech, pharma, medtech, diagnostics, etc but don’t do Tech deals; Diversified firms do both technology and healthcare deals.  This dataset tracks individual investments by fund type, and in this analysis includes over 2100 deals by US VCs into US companies with an initial investment between in the last decade 2002-2011 (data are as of Sept 30, 2011).  See chart below:

Surprisingly, realized returns for biotech investments by diversified funds were better by nearly 1200 basis points, a huge outperformance.  Looking at unrealized returns (deals still in progress), both were negative, highlighting the lack of markups in valuation over time for biotech in general.  But the difference still favors diversified funds.  And the combined “total value” measure (realized + unrealized returns) favors diversified funds by a remarkable 520 basis points.  These data highlight a very large performance differential.

Is this just a statistical fluke?  I guess it could be, but the numbers in each group are large enough to suggest otherwise.  There are 1401 investments included in the healthcare-only bucket, and 766 in the diversified fund bucket.  These are big numbers.  The dataset unfortunately doesn’t allow for detailed statistics on p-values and significance, but the “effect size” here looks pretty large.  Further, I have heard a few LPs with broad portfolios of venture investments confirm this outperformance of biotech deals from diversified funds.  So it’s probably real.

What are the drivers of this outperformance?  I really don’t know.  It’s certainly not that the investing teams are different: both healthcare-focused and diversified funds have biotech teams with a preponderance of PhDs/MDs who’ve been in industry or operating roles for long enough to know this isn’t an easy business.

Here are a few potential hypotheses to explain these findings:

  • Do biotech VCs from diversified funds walk from failing deals faster?  Are there differences in how reserves are allocated?  Since the scarce pool of remaining reserves in a fund are allocated across sectors, is there a more “efficient market” in a diversified fund for channeling capital towards winners and starving losers across macro cycles?  Maybe.  I suspect though that the variability in behavior across individual firms is far greater than that across fund style (meaning both types of funds are equally likely to have good/bad reserves allocation practices).
  • Are healthcare-only investors more geared to early stage deals and therefore back winners that just take longer to realize?  This could explain the difference in realized returns, but my experience is that this isn’t the case.  A large number of healthcare-only players are explicitly later-stage investors, and lots of diversified funds maintain an exclusively early stage focus.  So I’m pretty sure this doesn’t explain it either.
  • Do VCs from diversified funds do different kinds of biotech deals? Do they avoid big capital intensive plays?  This could be the case, but I don’t have the data here.  In a diversified fund, where capital efficient tech deals compete for the same fund resources as biotech, there could be a bias towards leaner, more capital efficient biotech bets.  Their portfolios could also include fewer big burn regulatory bets (source of discomfort for Tech partners) and fewer “big science project” platforms that lack product-oriented focused.  Who knows, but this might be a driver.

Whatever the historical drivers of this difference, the future could be quite different.  Although many good diversified funds have been raised in past couple years (Canaan, Flagship, Polaris, etc..), a number of well-regarded diversified funds have sadly split: Soffinova, ATV, Morganthaler, Scale, etc…  And Index has raised a dedicated LS-only fund under the brand’s umbrella.  Lastly, several smart healthcare-only funds have arrived on the scene, including Third Rock Ventures and Column Group.  This new mix will certainly change the landscape over the next decade, and perhaps change the data.  Only time will tell.

I’m still puzzled at the large difference in biotech returns by fund type.  I’d like to believe it was due to some synergy or advantage inherent to a diversified fund that makes them better at doing biotech deals (especially since Atlas is a diversified fund!).  But frankly it’s hard to explain.  While there are numerous advantages at the fund level for why diversified funds are compelling to LPs, those aren’t critical to understanding the root causes of why biotech returns are different.  Only the last hypothesis above about deal type differences seems credible, but maybe there are others hypotheses.  Maybe we can crowdsource a few more.

This entry was posted in General Venture Capital, VC-backed Biotech Returns. Bookmark the permalink.
  • http://ginkgobioworks.com/ Jason Kelly

    Seems like it could just be a statistical fluke.

    For all investments from a given VC firm, you can’t deconvolve the fund-type from any other aspect of the firm (for example, # of partners, quality of partners, size of fund, etc).  So you don’t really have 2167 data points, you have as many data points as the number of VC firms represented in the set.  It’s only when comparing between firms that you can make comparisons that deconvolve fund-type from other features of the firms.  

    In an extreme case, imagine you just compared Atlas (diversified) to Domain Associates (HC-only) – you might have 200 investments, but really you just have 2 data points to compare the effect of fund-type (overall performance of Atlas vs. Domain Associates) which wouldn’t let you conclude anything significant about the effect of fund-type. 

    Of course the upside of diversified funds is you get to invest in photo-sharing apps ;-)

  • Lee C

    Thanks, Bruce. In terms of “types of deals,” could it also be about types of products? For therapeutics, is there any clear data on which types of funds put out more ‘me too’ drugs versus innovative mechanisms/targets?

  • David Henkel-Wallace

    I conçur with Jason that the raw data are so muddy that it may be a statistical fluke.  BUT, on a qualitative basis (and as someone who has raised investment from both diversified and sector funds): I suspect the diversified funds have a better ability to discriminate on non-science issues.

    In other words: a life science investor, if he/she believes in the science and the indication, will have a positive bias when evaluating price, team, market etc.  In a diversified fund, a (say) IT partner can weigh in and say, “well my cousin had that disease and the treatment was nothing like what the company claims is the current standard” (paraphrase of actual statement an IT partner said while making a partner presentation to a diversified fund).  This can derail a good deal, or can provide dispassionate opinion on a marginal deal.  Also the diversified funds will simply tend to have more people to look at the deal.

    All firms are different, and I’ve had the opposite too (ran into an IT partner who was frustrated that the team didn’t offer us a TS).  And deals where all the partners weighed in and others where few did.  But in the aggregate I suspect this is why diversified funds would do better.

  • http://twitter.com/raman_minhas Raman Minhas

    Thanks, Bruce, this is a really interesting question. Maybe it comes down to the risk profiles of diversified versus specialist funds?

    As you point out, both types of funds have healthcare teams that are populated by specialists. Both teams are also likely to have a rational and emotional response to any given investment opportunity, given they have considerable experience and knowledge of the space.

    However, within the specialist funds, oversight is comprised of other specialists who have usually spent their working lives in the healthcare space. That’s what they’re intimately familiar with and feel most comfortable with. Compare this to a diversified fund. Here, a healthcare investment opportunity has to pass an industry specialist hurdle AND a generalist hurdle – Is this the best we can do with our funds, given we have opportunities outside healthcare? Just the share act of having this external viewpoint creates discipline.

    An external viewpoint also means a diversified fund is less likely to get caught in a specialist “bubble” for any given popular healthcare technology. By all means, it pays to have specialist insight and knowledge to evaluate opportunities, but being entirely focused on one sector can lead to missing the obvious that generalists would see.

    So, by way of having an external viewpoint and discipline, I think your points 1 and 3 could go a long way to explaining the difference.

     

  • LifeSciVC

    Jason,
    Thanks for your comment. That isn’t the way the analysis works. Every fund reports their returns (or paper value) for deals to Cambridge Associates. Each of those individual returns are data points. The A-round investor has a different return than the B-round etc… All those data points are pooled for funds that are diversified or healthcare-only. Those pools have gross mean IRRs. So 2000 data points, if we assume each company has ~4 investors on average, is about 500 biotech companies. That’s a reasonable dataset for most analyses. B

    Bruce L. Booth, D.Phil.

    ATLAS VENTURE | 25 First Street, Suite 303 | Cambridge, MA 02141
    bruce@atlasventure.com | P: +1 (617) 588-2636 | C: +1 (917) 302 8571
    @LifeSciVC | Blog: http://www.lifescivc.com | EA: Ann – agiambertone@atlasventure.com

  • Pat B

    Perhaps performance is heavily influenced by sector and timing. In other words, perhaps the diversified fund’s performance was dragged down by their  healthcare investees performance? To help complete this analysis, it would be useful to look at IT (and Cleantech?) specialist VC performance in addition to the Healthcare specialists…

  • Knuss

    Aren’t the data consistent with the view that the biotech venture model is broken?
    While pure healthcare funds are motivated to continue with the broken model and invest in weaker deals, disciplined diversified funds are not compelled to lower their investment standards…

  • Christian Quaade

    Interesting analysis. A couple of questions, which also related to your 2011 Nature Biotechnology (NB) article:
    – Is this data set directly comparable to the set from the NB article? (same period, same set of data points)
    – Does a set of data exist for European biotech VC investments? I.e., for EU-based VCs investing in European biotechs, or for any VC (home country irrelevant) investing in European biotechs.

    There is a lot of talk about European VCs not performing well in the Biotech arena. And LPs saying that good Biotech VC investments can only happen in the US, due a “different culture”. It could be interesting to know if this is just popular perception, i.e., if VC investments in Europe are doing as well as its US counter part — at least relative to other sectors.

    Any input would be appreciated.

  • LifeSciVC

    Christian,

    The data is the same source, but I believe updated for some more recent time points.
    I don’t have access to run the EU vs US cut, but its an interesting one.
    B

    Bruce L. Booth, D.Phil.

    ATLAS VENTURE | 25 First Street, Suite 303 | Cambridge, MA 02141
    bruce@atlasventure.com | P: +1 (617) 588-2636 | C: +1 (917) 302 8571
    @LifeSciVC | Blog: http://www.lifescivc.com | EA: Ann – agiambertone@atlasventure.com