Why Past Performance Is A Good Predictor Of Future Returns In The Venture Capital Asset Class
I wrote a series of blog posts about venture capital fund performance on my personal weblog several weeks ago. One of the things that came out of that series is the fact that the top quartile of fund managers performs significantly better than the rest of the pack.
There's a saying on wall street that I've heard over the years about asset managers, "buy the 10 best and forget the rest". And that is certainly true of the venture capital asset class. If you had bought the 10 best and forgot the rest in venture over the past 20 years, you'd have netted at least 25% annual rates of return, including the ugly years from 1999 to 2003.
So that begs the question about past returns being an indication of future performance. In fact I got a lot of comments on my series of posts about this specific issue. Nik said:
I could guess some of the reasons of why past performance is an indicator of future performance. E.g. self fulfilling prophecy i.e. Sequoia has backed successful companies and I could be successful as well if I were backed by them. As a result, these firms almost get a ROFR on the best plans...But wanted to hear from you on what else there could be...
Why is it that past performance is such a strong indicator of future performance in the venture capital business?
First, it's largely true in all asset classes. Someone who has been a top performer in the hedge fund or real estate business over the past 10 years is likely to be a top performer for the next 10 years. Just like there are some who are amazing at sports or writing books or playing music, there are some people who are simply great investors. They have that special something that allows them to be better than everyone else. And like great athletes, writers, or musicians, great investors can have slumps. But if you look at the performance of great investors over the long haul, you will see that they can put up long term track records that are consistently better than the average investor in their asset class.
Second, Nik's guess is a big factor in all private markets. Successful investors create an aura of success around them and their firms. It's true of firms like Blackstone, TPG, KKR, and Providence in the buyout business and it's true of firms lik Sequoia, Kleiner Perkins, and Matrix in the venture capital business. Successful deals are associated with the firms that were the primary backers. The success of the deals builds the firm's brand. And entrepreneurs want that brand association as much as they want the money that comes with it. It doesn't give the venture firm a "right of first refusal" as Nik suggests, but it sure makes it easier to beat the competition, often with a lower price (as long as it's not too much lower).
These two reasons are not specific to venture. And they are the two most commonly cited reasons that past performance is a strong indicator of future returns. But there are a few things that are more specific to the venture capital business that play an important role as well.
If you look at the most successful venture capital firms, you will see that they are usually early stage investors who often are the first institutional money into an investment, the way Kleiner and Sequoia were in Google. Getting that place in the capital structure in an early stage investment is important because it provides access to the management and business that later round investors don't get. Look at the role Accel now has at Facebook. They are involved deeply in the strategic and business decisions that the management of Facebook makes. What comes with that kind of access is a level of knowledge and understanding about an emerging market that you just cannot get sitting on the outside looking in.
The best venture capital investors/firms use their engagement with their portfolio companies and all the entrepreneurs they meet to become smarter about the markets they invest in. Kleiner and Sequoia certainly got to see the search marketing business developing before most other investors did. Accel (and Greylock and Peter Thiel) will certainly understand how social advertising is developing better than most of us on the outside. These insights are incredibly valuable and should lead to a host of additional investments.
Another effect of this close association with portfolio companies is management talent. Look at PayPal. The PayPal alumni network is legendary, having spawned dozens of companies, many of them leading venture investments in the past five years. If you were an early and significant investor in PayPal, as Sequioa was, then you'll know all these talented entrepreneurs and ideally will be able to back them before others get the chance.
A related aspect of this management issue is the subject of "franchise entrepreneurs". Some firms will develop relationships with entrepreneurial teams that allow them to back them again and again. We have that kind of relationship with several of the teams that are in our current portfolio. There isn't any kind of "locked in relationship" like a 5 record deal in the music business. But the best firms treat their entrepreneurs well and the entrepreneurs will come back deal after deal as long as they feel they are being treated fairly (and visa versa).
So you can see that there is a virtuous cycle of knowledge and relationships that come from being one of the best venture capital firms. Good deals and good teams lead to more good deals and good teams. It's frankly hard to screw it up.
But you can screw it up. There are two common ways I have seen great firms "screw it up".
The first is getting too big. When you raise too much money in the venture capital business, two things happen. The first is you have to hire a bunch of partners to manage all the capital. And not everyone will be a world class athlete. So you end up with mediocre investors on your team. And that is a bad thing. Even more problematic is that it's hard to be the first institutional money into a deal when you are managing billions of dollars of venture capital. So other firms come in underneath you in the capital structure and take that pole position with management, and get all the relationships and insights that come with that position.
The second screwup is "style creep". It's often related to getting too big, but can be caused by other factors as well. A firm that is a great biotechnology investor may not be a great clean tech investor. A firm that is a great communications equipment investor may not be a great web applications investor. The best firms are known for certain kind of investments and stick to them. They often have partners who have specific market expertise. When firms expand their areas of interest beyond what they are good at and what they know well, it often leads to reduced performance. And their brand can get hurt. Which can have a snowball effect for the same reasons that great performance and great brand produces a virtuous cycle.
So what would I look for if I was investing in venture funds? First, I'd look for investors who have a track record of success over a long period of time. And I mean people, not firms. Then I'd look for a firm that has a small group of such investors. I'd expect that firm to have produced top quartile returns for a series of funds. Then I'd make sure that the investors who were responsible for those returns were still actively engaged in the firm and would be making the majority of future investments. And I'd want to know that the investment strategy of the firm was not changing and that the capital being managed per partner was in the manageable range for first round/lead investments which I believe is around $50 million per partner per fund. If you find that profile in the venture business, buy that fund. Only problem is you can't get into funds that look like that. Unfortunately, I don't have an answer to that problem.
November 26, 2007 05:35 AM, By Fred Wilson
Tags: capital deal firms performance venture
Comments (15)
As an investor in venture funds, where would you place the importance, or necessity, of investing in or looking at emerging funds or managers?
Posted by AWeissman , November 26, 2007 07:36 AM
Hi Fred,
Great post!
I was in discussions with a successful enterpreneur who was looking to start a seed stage fund in India. A few VCs were discouraging him saying that being a pure seed stage investing is not great for returns.
I sent him this link (via Kedrosky)- http://vcratings.thedealblogs.com/2006/07/khoslas_law_doesnt_apply_to_bi.php
However, the key it seems is to be like USV i.e. start early but have the gunpowder to keep investing if your thesis starts turning out right.
This also means, again just like USV, you have to be thesis/expert driven to know what is likely to work or not- very early. Sometimes, you have to know this before even there is customer traction or just when it is starting because if a firm gets traction, they will in go to the brand name firms and the non-brand firms will get squeezed out.
This post by Bill Burnham best captures that type of thinking- http://billburnham.blogs.com/burnhamsbeat/2005/05/deal_flow_is_de.html
Nik
Posted by Nik , November 26, 2007 07:47 AM
Hi Fred,
Interesting post but I think your comment "First, it's largely true in all asset classes." is stretching things a little. I think the key points in the VC sector are the quality of deal flow (including desire of entrepeneur to work with proven investors) and the ability of the VC to play an ongoing role in the success (or failure) of the business. I think in asset classes where these two components don't exist the ability of past performance to predict future performance diminishes significantly.
Sam
Posted by Sam , November 26, 2007 02:40 PM
Thank you.
Posted by Lloyd Fassett , November 26, 2007 03:44 PM
"Accel (and Greylock and Peter Thiel) will certainly understand how social advertising is developing better than most of us on the outside. These insights are incredibly valuable and should lead to a host of additional investments."
Does this statement imply that VC firms that have a successful early stage deal will see a lot more deal flow in that company's sector? It would seem more intuitive that more startups would go to a different firm that didn't have a potential conflict of interest but at least had peripheral expertise.
Posted by Victor Wong , November 27, 2007 01:41 AM
Fred,
While your piece is well thought out and written, your last two sentences are by far the most important and unfortunately, for the most part, nullify the hard work and personal bandwidth you have expended creating this post. I have spent countless hours researching the networking aspect of the VC marketplace and the bottom line is quite simple. You can identify the areas that provide above market returns but at the end of the day, you aren't getting access to them. Founder's Fund is not open armed to just anyone. In fact, you have a better shot these days getting money in KPCB's doors than through Peter Thiel's.
In my opinon, as well as a number of my colleagues, the way to remotely access these types of funds is through a meticulous assessment of the successful VC's network with the hope of finding the "next generation" managers. Index or Founders Fund didn't always carry the cache they do today.
The deck is still unfortunately favoring the house in this situation however as an aside, the Sequoia/Kleiner mystique is not what it once was. This is leveling to a certain extent as many smaller and more nimble funds are starting to pull down deal flow as they have broken away from the good old boys club on Sand Hill Road. Finding them before they become stars is the challenge.
Take a look at the Forbes Midas List and check out the Capital Blog. A great read for any serious Venture minded individual.
http://www.forbes.com/lists/2007/99/biz_07midas_The-Midas-List_land.html
Posted by DJ , November 27, 2007 02:49 AM
What about the individual GP as a brand?
Won't there be turnover in leadership at these firms? Do firm demographics play a role here?
How much organizational risk must you take to find the next repeat performer?
A lot of LPs out there betting on new funds/firms to be winners. Are we fighting history? Usually, that doesn't work.
Posted by Lindel , November 27, 2007 04:23 AM
Another excellent post Fred.
Musicians, like myself, who have generated millions of dollars of sales through their work over the years, are prone to slumps. I don't know how it is in the VC world but for us there is a continual metamorphing process. It may look like you're finished when you're merely changing.
Now that we have lost our "factory" - the craftspeople are bollixed as they find a new way to the market, which hasn't reappeared as of yet. This is a monumental shift as a class of creatives and as individuals.
For those who survive it will yield more inspiration than ever before. For some of us, like myself, it has opened up new creative vehicles to go with the songwriting and performing - my startup muzlink.com.
Coming into a relational basis with the VCs has brought me to this post where your insights into your world have only made my current recordings better by showing me the universality of creative endeavours.
Money and Music seem to have much more in common creatively than any of us had presumed.
further...
Posted by Brad Parker , November 27, 2007 09:26 AM
Hi Fred,
I have been looking at some of the data that you have posted. Are you sure that what you are saying is based on the data - for instance have you taken the top quartile performers each three years and compared their performance for the three years after that and the three years after that - if you do that how many "stars fall to earth and how many remain?
Great posts - love your stuff
Cheers
Paul
Posted by Paul Higgins , November 27, 2007 06:16 PM
if you look at the performance of great investors over the long haul
Whoa Fred, hold on here. I don't think this works to answer the past performance = future performance. I don't think you can look at the top 25% of firms - which by definition are successful - and then go back in time. Of course their past performance is better because you have screened for that initially.
I'm not up on the stats analysis required here, but a simple approach is to take the firms you think are predictably better, then take their *next* investment and see if those are more likely to succeed than average.
I would think you'd find that success breeds success because the past performers have better connections and buzz for their clients, but this is different than the notion they apply a better analysis or formula.
Posted by Joe Hunkins , November 30, 2007 02:31 PM
While I agree with just about everything in the post, my intuition tells me that in some respects younger VCs ("newbies") have certain advantages over older, established performers. Many web2.0 services such as social networking and new media are consumed almost exclusively by people under 40 (and in all likelihood, under 30) . With these services so integrated into daily life of the under 40 (or 30) web-generation, I can't help but think this gives younger VCs an experiential advantage over older, established performers.
Posted by ben siscovick , November 30, 2007 05:10 PM
One quick comment on the finance side of things. Curiously, you are reiterating one of the biggest (and most costly) misconceptions about the world of investing. It's not true at all that "a top performer in the past 10 years is likely to be a top performer in the next 10 years" in hedge funds or mutual funds. That may precisely be what sets these asset classes apart from real estate and VC, where bigger information asymmetries between market participants can be exploited (the alluded-to ROFR on business plans). In fact, it makes more money to systematically go short funds that outperform the markets! Hedge funds in particular, across all strategies (long/short equity, global macro, bonds-only, event-driven, whatever), have almost only made money through levering up "betas", i.e., systematic returns from risky assets. This means that the past top funds are only likely to outperform in the future as long as their underlying beta exposure performs as well as in the past. For long/short equity, that's stocks. For event-driven or risk arbitrage, it's fund flows into merger activities. Simple as that. So yeah, Wall Street is now slowly unlearning the "buy the best, drop the rest" formula as it was as naive a system as Newtonian mechanics describing planetary movements before Einstein's general theory of relativity came along.
Posted by Mario , December 22, 2007 01:01 PM
The first assumption is incorrect -- ie, that investors who have outperformed for the past 10 years are more likely to outperform for the next ten years. See Burton Malkiel's A Random Walk Down Wall Street, which looks into mutual fund performance (the only publicly available returns). Anecdotally, Julian Robertson's Tiger Management lost more money in 1999-2000 than it had made in the previous 20 years. This is true in real estate (ie, in late 80s and early 90s, almost all of the largest REITs went bankrupt). And this is true in team sports (think of Pro Bowl-caliber players with the Patriots and Colts who left for other teams).
Posted by Sujay , December 24, 2007 02:15 PM
there's truth to both sides of the argument here and i think what it takes to sort out the good from the bad is both an appreciation for what makes a team/firm successful coupled with ongoing vigilance about style drift and excessive growth in AUM, etc.
as with many alternative, talent-based investment areas, its important to net out true alpha from the results through a return attribution analysis. i dont think there's a conclusive way to ascertain this but there are methods that yield some degree of accuracy. the surest one is how a manager has performed in tough times.
ultimately, the problem is how do you get on board once others have picked up on what you have. a smart manager will close the gates before they get too big for their own good. i suggest a healthy dose of networking to build bridges with the best.
Posted by pendolino , February 7, 2008 05:39 PM
Great post, Fred. I wonder how the performance of regional VC firms stack up against that of sector VC firms. Here in Cincinnati (I'm in the airport heading back to NYC as I read this), a lot of our shops have a regional bent. Therefore, performance relative to other firms in general is weighted or lifted by the local economy. Though USV is a sector-oriented firm, you still cull most of your deals from NYC, it seems, so aren't you also affected by the New York tech economy? Aren't there a lot of external factors here?
Posted by Nate Westheimer , November 26, 2007 07:04 AM