Gabriele Grecchi
Aug 23, 2017 · 2 min read

Hi Mark! Congrats with this new great achievement.

I have a “portfolio construction” question for you. Consider me as a student and wanna be VC in the future (if I’m successful in exiting my current regenerative medicine business ^_^). Please bear in mind I have done 10 years in the asset management business doing optimal risk budgeting, macro asset allocation and portfolio construction.

Assuming you have to return 3x gross to your LPs, a $400M fund should target a $1.2B value of the underlying portfolio (realized or unrealized), which — again assuming that your average ownership at exit, after dilution, is 30% — requires your portfolio companies to have an overall enterprise value at exit equal to $4B ($400*3x/30%).

I also understand you’re going to invest in 15 companies per year, deploying around $100M annually, which means by year 5 you will have 60 companies in your portfolio, with a median total capital invested of $6.6M each.

What kind of distribution of returns you assumed when you constructed your portfolio? If, for example, by knowing you are investing very early, I would assume you will have a similar distribution: 2% Unicorn (=50x=1 company), 5% Large Win (=15x=3 companies), 18% Small Win (=3x=10 companies), 25% Save (=0.5x=15 companies), 50% Loss (0x=30 companies).

By doing the math (and excluding a lot of additional parameters we should probably include — however, they would represent a further drag on the fund performance), it seems that the overall expected value would be around $833M, a reasonable 2.2x gross on the capital invested, but still very far from the $1.2B target (3x gross).

Am I assuming a wrong distribution of outcomes than the one you thought of while doing the portfolio construction exercise?

The main idea behind my question is the following:

  1. is the market generally ready to absorb that capital (demand) and deliver the expected exit (supply)? Funds are getting bigger and bigger, and I’m always thinking the typical Power Law distribution in the VC world is an ex-post parameter, that shouldn’t be taken as an ex-ante information for portfolio construction (where outliers should be taken out of the equation and only trimmered averages of exit values should be considered while designing the optimal structure and size)
  2. how many exit of a certain type you have to have and which average ownership you should keep in your portfolio companies to deliver the 3x gross usually requested by LPs? It seems to me that the larger the funds get, the bigger and more frequent the exits should be, which is counterintuitive considering that exits will cluster in the left side of the Power Law curve, where only a few investors will be able to get those returns

Thanks in advance for your time! You’re always a great source of inspiration!!

)
    Gabriele Grecchi

    Written by

    CEO at Silk Biomaterials, INSEAD MBA. Read my books Future Health http://amzn.eu/iDLIaM9 and Capitali di Ventura http://amzn.eu/1LFrCVg