How can allocators make the most of today’s venture capital investments? Ted Seides hosts a conversation with industry LPs to reveal their strategies.
How can allocators make the best of venture capital investments? Today’s market is full of opportunities for growth, but not every institution is set up to benefit the same way. Learning how successful strategies work under different circumstances is the key to forming your own approach.
Capital Allocators host Ted Seides held this live webinar as part of The Institute’s Conversations, Ideas & Opportunities series on December 10, 2020.
[Steven Kim, Partner, Investment Strategy & Risk Management, Verdis Investment Management]
The size of your pocketbook allows you to invest anywhere you want. Why did you decide to go for early stage?
We’re very data driven in our investment thesis. We use data a lot to understand the underlying probability structure of the asset classes we invest in. When we look at early-stage venture, it’s by far the most attractive stage to invest venture capital, and venture capital is, by far, the most attractive asset class to invest in. It’s also investing at the most favorable valuations. The other plus is that you see venture capital performing in these times of stress, recessions and disruption: if you look at the data, venture capital has performed very well in those time periods.
So you have the über sub-asset class within the über asset class — yet there’s also wide dispersion returns within that early stage, so did you think about just investigating as many funds as you can find? And, if not, how have you participated within that sort of sub-asset class area?
I think when you invest in early stage venture, portfolio construction is everything. It’s really important to understand what you want to create as far as the portfolio and the strategy that you want to deploy.
If you look at early-stage venture returns from 2017 to the first quarter of 2020, when you look at the Burgiss data the mean net returns are at 11.61%; the median for that time period is a negative 10 basis points. That gives you a sense of the dispersion of returns between the median and the mean — it’s really high.
It’s the only asset class where you see that. You see the mean and median returns very close to each other in most asset classes, but in early-stage venture there’s a wide dispersion.
If you’re highly diversified, and you invested in all those managers from 2017 to the first quarter of 2020 you would have received an 11.6% return. If you picked investments, you have a high probability of that minus 10 basis returns instead, because that is the typical return of a venture investor. If you look at the data, press releases, journalists, they typically give you the median venture return, which is not very attractive. Over the years it’s only been 6%, so it doesn’t look like a very good asset class, but the mean returns have been very strong.
So how do you approach the space?
You make an assumption that the baseline return for early-stage venture is random; therefore, you want to harvest the mean return as much as you can. Anything above that is a bonus. So, if you’re looking for skill, or you’re biasing your selection by geography or by sector, that will give you the opportunity to beat that mean return. But if you assume that you’re only going to get that random mean return, as long as you’re diversified it’s a highly attractive return.
How do you think about risk?
I think venture risk is all about how you build that portfolio. If you look historically at what a typical investor is going to get in venture capital, it’s roughly around the 6% return. That means half of them have been below 6%. But the average return of the asset class is very strong, so you really have to think about how you’re constructing the portfolio and whether you want to take that kind of risk.
That leads to this narrative of being in the right funds. In order to beat that median return, the 50/50 return, you have to be in the right funds: that’s been the narrative in venture. But if you think about it from the standpoint of being highly diversified, that’s probably easier than trying to claw into the right funds.
A lot of these ‘right funds’ make you invest in their whole product portfolio, so you have to be really careful about what stage you’re going to be invested in. The return and risk profiles differ by stage and venture. But if a great, top-ten investor is making you invest in their India fund and their late-stage fund and their co-investment fund, the majority of your capital is going to be deployed there versus their early-stage fund, which is much smaller and has much more difficulty and you’re diluting your return.
I think one of the asset classes that’s really interesting if you’re really worried about downside risk is life sciences. Life sciences is a sector in itself in venture capital and has a completely different risk and return profile than general tech investing. The number of outliers that get formed in life sciences is ten times higher than it is in the other sectors. You don’t have halo deals like the Ubers and the Facebooks of the world, but you have a lot of outliers that exit in that 500 million-2 billion valuation.
How many different relationships do you have in that portfolio?
We make 20 or 25 pre-seed, seed and early A investments in a three-year period, so our cadence is really high to capture that mean return. We typically target 20% of all first check institutional deals to be in, so when you get 20%, we think there’s roughly 2000 start-ups that form every single vintage year. If you can allocate to 20% you have a 95% probability of getting into that mean return and getting into those outliers.
And how do you decide which 20% to invest in? And why not make it 40% and try to get to all of them?
We’d like to get 40%, but we don’t have enough capital to do that and the cadence is so strong. The more you do, the higher your probability of capturing the mean, so it does pay to do more. What we try to do is selection bias; some 80% of outliers come out of California, 12% come out of New York and the balance come out of the rest of the country, so you want to bias your selection to California.
The other point is that the expected value of California outliers is four times what they are everywhere else in the country. The average value of an outlier captured in California is eight billion dollars. The average value of an outlier captured in New York, which is the second best, is only two billion, so you have to capture four outliers in New York to get the same value as one outlier in California. That’s the selection bias, and that gives you the opportunity or probability of exceeding that mean return, which is the baseline you’re using when you’re thinking about diversification and investing in early-stage venture.
[Kevin Tunick, Managing Director & Vice President, UNC Management Company]
You’ve got the mature endowment portfolio of venture managers — what’s the structure of that portfolio?
I started there around 2008 and said to the Board that my goal in venture is to invest in the top ten firms that are out there. I also said my conundrum was I couldn’t name ten firms at that time that I thought belong there, so our goal was to find the firms that we believe would end up filling those empty slots in the top ten. That’s where we’ve gone. So we’ve got a fairly concentrated portfolio, with 11 core managers and I don’t see that number growing.
We haven’t seen markets quite like this since the early 2000s, where the distributions and size of companies end up being so large in the public markets that you have to think about what that looks like across your public equity portfolio. How have you thought about the exposures created by all these technology companies?
This is one of the things we have a tendency to argue about, because when we look at what we call our venture portfolio we have public companies in there, and the argument is that they’re really no longer venture but are public, and so should we be reclassing those positions? Then the question becomes: how do you allocate 10% of one fund into the public portfolio, so I never really paid a lot of attention to that. We just leave it all in venture.
For the liquidity aspect, it’s been interesting to watch the evolution from back in the 90s when everybody went public ASAP, to the 2000s and into the mid-teens where everybody was encouraging companies to stay private as long as possible, and now you’re seeing the flip where people are now going public again.
Venture is one of these things where you have to have a strong gut. A perfect example is that roughly 10 individual companies represent about 50% of our venture portfolio. That’s what venture is all about. You’re looking for that selection bias where you’re hoping to get those great winners; the Googles of this world where a client can invest 10 million dollars and end up with a billion dollars. And invariably there’s a lot of noise that comes along with a venture portfolio, a lot of stuff that’s going to generate flat to nothing, there’s a bunch of write-offs, and then you’re going to have some singles and doubles in there as well.
So when you’re left with 10 big names at the end of this exercise, do you think about hedging?
Back in the 90’s it was far easier to hedge, it was easier to borrow the stock to short the position. Now you have to look at the individual fund you’re in, because many of the funds have clauses in that prohibit you from hedging. So it’s something that you pay attention to; sometimes you can do something about it, and other times you’re just riding the position until you get the distribution.
You guys can’t seem to get the number 10 right in your portfolio in a one-in-one-out portfolio. Have you thought about picking the right ones compared to being in the asset class?
The dispersion of managers in venture capital is tremendous, and I have to admit I drive my team crazy because I sit there and say, ‘In its entirety, I view venture capital as a wealth-destroying asset class’. But it’s also an asset class where if you can invest in the best firms who get exposure to the Googles, the SpaceX, the Air BnBs, the DoorDashes and the Palantirs of the world you can make a lot of money. Not every fund out there is getting into those investments, and that’s why it really gets down to manager selection: it’s everything. And one of the things that we have developed is a belief in entrepreneurs who can sit there and empathise and sympathise with the people that are building the companies. They also have the ability to gain access to individual deals, that’s the thing really in venture — sourcing. If you can’t get access to what are going to be the 10 or 15 best deals of that year, it’s not worth being in the industry.
[Chris Culbertson, CFA, Managing Director, Bespoke Strategies team, Lowe, Brockenbrough & Co.]
Can you explain what was in your private market portfolio when you started, and how you’ve evolved it since?
Historically, the portfolio was allocated more towards later stage managers, and later stage companies. We actually have more of an agnostic stance when it comes to stage. But access and relationships are extremely important in venture. So my team was more focused on their relationships in the later stage.
My background and experience has been built at a number of organizations with a lot of seed and early stage exposure. So this knowledge, experience, and network has helped us build up the seed and early stage side of our venture capital portfolio. Importantly, it’s not just about allocating to the space solely to have exposure. We actually think that a lot of the managers in that space. Specifically, the next generation are truly different than their predecessors, both in mentality and the way that they’re communicating, interacting and building the relationships with the entrepreneurs as well.
Some of them are entrepreneurs so they’ve been in the trenches. They know how to build a company and know some of the areas challenges. That means they can perhaps identify and help these individuals as they build a company and avoid mistakes.
I would also say there are a number of instances where it’s not just about the money for them. Yes, they want to compound the capital and put up good performance numbers, but there’s something else that truly motivates them. It’s not just one thing you can identify, but they truly want to create a better world for everyone, or help create a product or service that is positively impacting society or a certain business or whatever it may be. There’s something else that’s non-monetary that is driving a lot of what the next generation has to offer.
How have you thought about access — because so many people refer to venture capital not as an asset class but an access class?
Access is certainly key, but it also becomes a network effect. The way that I portrayed it to others is: I’ve spent my entire career creating what is today a strong network in the space, but I can’t be everywhere. There’s way too many seed and early stage managers today, an exponential number compared to prior decades. So knowing who the best are comes through the work that we do and the networking that we continue to do to further build relationships. Our managers know what we’re looking for, so they are also a great source of information and introductions for us as well. A number of the more recent commitments that we’ve made have been through introductions from existing venture managers that we have, and I’d much rather take that call than someone who just randomly calls us up and says, ‘Hey, I’m raising a fund.’
It is about breaking down the barrier of access. It may be a situation where access is constrained, but our relationship with a manager and the institutions that we manage capital on behalf of, and all the missions that they are serving, help us get in the door. Then it’s about the people: it’s about how we convince them that we are the right partner and how they can convince us that they compound capital.
Different organisations have their own pitch to say you’re the one that they should take capital from. How do you deliver that pitch to a desirable venture manager?
I don’t think our pitch is any different. I think we all get better over time, by just knowing what we know and avoiding areas where we’re trying to reach too far.
We encourage managers to do references on us as well. Making a commitment is easy, and our managers raise capital from anyone. It’s about partnering with the best people and backing their process: those are much more important. Utilizing our network for references, both in regards to managers we are evaluating and offering them as references on us has become increasingly important.
[Dana Johns, Senior Portfolio Manager, Maryland State Retirement and Pension System]
You have this challenge of wanting to get access to this asset class but just having so much money to put to work. You have a small number of managers, and you mentioned a separate account partnership — what’s that all about?
We’re accounting around about a billion and a half of capital every vintage year. Generally, that’s going into buyout, so it’s a more buyout-centric type of portfolio, maybe allocating 50 to 70% of that to venture. Just before 2016, we had a CIO who was a big supporter of private markets and private equity and venture in particular. He was sharing his role with me and suggesting, ‘Hey, reach out to these managers, because clearly we understand you’.
I was knocking on some doors and introducing myself to some of these great managers we should be in but we were not in, and got a great reception but ultimately heard: ‘Thanks, nice to meet you, but we really don’t need your money,’ and ‘We don’t know you that well, and you’re a pension.’
We think of ourselves as good institutional investors and partners through our network. We prefer to invest with green managers rather than through third party or founder funds, and we really value those partnerships. We were introduced to a group called Tiger Iron, which had spun out of a founder fund. Our Tiger Iron journey began 2015/16 and we started due diligence, working out what this model might look like.
Basically, the Tiger Iron model has three institutional investors. We’re one of the three, and there are two other large institutional investors. We all have different goals in building a venture portfolio, and we, in particular, were focused on early stage because we did not have that exposure to early stage. There was some life science early stage, but not tech.
So the fund and the idea and the building of the relationship with Tiger Iron was that we would build a portfolio with roughly at about 100 million per year through a fund cycle. What Tiger Iron is driving at is to give access to the best managers. I can access those best managers through the relationships that Tiger Iron have built over the last 20 to 30 years.
That how we’re thinking about accessing venture. It’s the portfolios: we’ve done the first one, we’re not allocating a second one. So, generally, about 15 GP relationships, nearly 20 to 25 partnerships and some of these managers have an early stage and a late stage follow-on fund, We’re doing co-investment so our structure allows us to do direct investments into these companies, very selectively.
The commitment sizes are 5 to 30 million. We’re thinking about a portfolio of two to three or four of 30 million, and then we fill it in around the edges with our seed managers and Series A.
In this portfolio, and with capacity so constrained, how do you think about co-investments?
We hope to take advantage of co-investments. So from the buyout series we’re clearly very interested in co-investment because it’s no-fee, no-carry. We’d probably do a co-investment on the private equity side if there were any fees. But in our venture portfolio it’s no fee, generally there’s a little bit of carry — we’re looking at it as potentially shorter in duration, generally these are later stage.
When we’re thinking about a fund and putting this into the portfolio there are a number of different criteria. Is it a top quartile fund manager? Is the target multiple 3 to 3.5? Is there an access constraint? Does Tiger Iron have a relationship? There are a number of specific criteria to think about: where does the portfolio fit in terms of the VC being very sector focused or more generalist.
In terms of co-investment, I think it also helps to mitigate the very nominal fee that we’re being charged on invested capital. The first portfolio did about 25 million of co-investment out of the 300 million, so we’re kind of mitigating some of the fee that we’re charged through the co-investment.