VC Asset Class Size and Performance
There’s a continuing discussion around the optimal size of the venture asset class that is perpetuated quarterly when the latest numbers come out around fundraising and performance. I attended one of the PWC Moneytree breakfasts last week and have been thinking through the numbers again.
It’s common knowledge that overall VC asset class 10 year performance numbers began to look much worse this year as the “positive” effect of the exits achieved during the 1999 time period began to fall off the 10 year look back period. As a result, you have asset class performance that looks like this (according to the NVCA):
The huge impact of returns in 1998 and 1999 so impacts overall trailing returns that when those values “fell off” during last year, it caused a massive shift from 14.3% overall trailing 10 returns to -3.7%.
There’s a similar spike in overall assets dedicated to the VC asset class. VC funds raised during 1995 (aka vintage year 1995 funds) raised a total of $9.8B in investable capital. To date, this capital has returned more than 6.1x, or $60.5B. As a result of the huge run up during the late 90’s, anyone and everyone began raising venture funds and there was a peak in the asset class in 2000, when there was more than $100B allocated to venture funds by limited partners.
VC funds typically invest over a period of 3-5 years and typically have a 5-7 year period when the returns are “harvested”. Thus, it’s often difficult to judge overall performance of a particular fund in the near term. But as I was looking back into the data I was fascinated by just how massive the macro impact of the vintage 1999 and 2000 funds will be.
In the chart below, I’ve married the total amount of VC dollars committed by vintage year (the black line). On the stacked bars, the green section represents realized returns; the red section is the “marked” value of the portfolio companies which are still alive and yet to see a realized return.
A couple things that came to mind as I analyzed the above data:
- The marking of value for existing portfolio companies is largely an exercise in educated guessing. Sure, firms have to follow the guidelines of FAS 157, but for many, it can be almost impossible to come up with justifiable comps. However, it’s interesting to note that for vintage years 2002-2007, the marked portfolio values are almost exactly at 1x invested capital. I have no idea if that is a result of a marking methodology to “get close” to 1x, or if its a function of a law of large numbers type effect when you look at the asset class as a whole. It will be interesting over time to see if actual performance ends up close to the marks or if the vintage years over or under perform
- In the 1999-2000-2001 vintage years cohort there was close to $200B in deployed capital. Of this total, there is only $107B in realized returns. Yet, the current marks show ~$78B in residual value. This is a simply massive amount of deployed capital that will be looking for liquidity over the course of the next 2-3 years. My guess is that, as a result, we will see quite a bit of exit activity but the resulting returns will be for far less than the marked value. The current marks are close to $0.90 on the dollar against the deployed capital. I wouldn’t be surprised if the realized returns are 25-35% worse - or $50B less in actual realized value. If so, this will impact overall asset class returns in a similar manner as the positive performance of the early and mid 90’s funds. Those in the industry trying to raise funds will be waiting for the day when the performance of the 2000 cohort will “fall off” the chart.
