Don’t Sweat Year-End Venture Performance Figures

Last week year-end 2009 US venture capital performance figures were made available by Cambridge Associates (venture capital returns are reported on a quarter lag – even longer when it’s year-end).  Despite improving in the fourth quarter, returns over all major time periods were pretty lackluster save for the 15 and 20-year returns which still include the tech bubble. The 10-year return, the period given the most attention because it the same length as the life of most venture funds, dropped into negative territory. What’s more is that the net to LP returns for every vintage year since 1998, except for 2003, is negative. This to me stood out as the most discerning stat –it means that average investor lost money in venture pretty much every year since 1998. Yes, top quartile managers have done better, posting positive returns with IRRs in the 2-6% range – but single-digit IRRs for venture capital still cannot be justified, especially when when you factor in the illiquidity premium.

There’s no reason to be shocked by the return figures however, since they were somewhat predictable. Plus, the lag in returns makes the year-end numbers look worse than they really are because we’ve seen the public markets improve significantly since the start of the year.  You can expect the venture capital index to rise in the neighborhood of 3-5% when first quarter returns are made available– improving as a result of the public markets as well as heightened liquidity. Still, the forthcoming return increases will not go far in improving the overall picture of venture over the past decade. This means venture investments will be tough to sell to limited partners, which will invariably lead to an attrition of funds and firms and eventually lower levels of capital being deployed.

Lower levels of capital deployment, however, is almost exactly what the industry needs. Remember I mentioned 2003 being the only positive vintage year median? Well it also happened the year with one of the lowest fundraising and investment levels. Also, even though it does not make perfect sense, looking back at the returns over the last decade, imagine if only the top 50% of funds/firms were in existence – the median figures I mentioned would represent the low, and the top quartile the median. Real top quartile returns then would most likely have been in the double digits – more than acceptable. While we should always be forward-looking, we can’t ignore the fact that the past has shown a smaller group of better mangers results in a healthier industry and better returns.

And if we want to be forward looking, there are plenty of reasons for investors to keep faith in venture and continue making commitments:

• You have a contrarian play at hand;
• Innovation, the lifeblood of venture, continues unobstructed;
• There’s the diversification case -venture proved to be an extremely valuable diversifier for institutional portfolios during the recession;
• Lastly we’re finally seeing a thawing of the exit markets and the return of liquidity through IPOs and M&A activity, especially by tech giants who have huge sums of cash on hand.

These changes should result in successful vintage year 2009 and 2010 funds, but when coupled with sustained lower levels of investment, should also bode well for the future of the industry. Gaps in funding from institutional investors will be filled by novel ways for entrepreneurs to raise funding. Examples include crowdfunding site like Profounder, and more basic funding platforms such as Kickstarter. These sources of capital fill niches that traditional venture had for the large part had gotten too big for, but fewer dollars flowing into startups means VCs will (and have already started to) move back down to earlier stage deals, which also historically happen to deliver the best returns. Looking back at overall venture performance isn't pretty, but keep in mind that past performance is rarely an indication of what’s to come. The venture industry is undergoing a transformation and performance for the next few years is primed to transform as well.

The Real Impact Of Overlooked Fund Return Considerations

The Private Equiteer recently brought up an aspect to private equity and venture capital returns that is often overlooked and unaccounted for: The fact that investors (limited partners) in funds have to set aside or plan around the capital they have committed to a fund. For those less familiar with private equity, investors in funds do not pay in the full amount they decide to invest in a fund right away. Instead, capital is called by the general partner as the fund makes new investments. Rarely do limited partners set aside their full commitment to a fund and hold cash to meet capital calls as they come. Most model around expectations provided by fund managers and hold only the amount of cash necessary to meet capital calls.

The Private Equiteer argues that opportunity cost of holding cash, or the risk of default associated with reserving inadequately should be factored into private equity returns. I would agree that there is some opportunity cost involved, but the simple fact is that virtually no limited partner holds the full amount of a commitment to a fund it has decided to invest in as cash – only for short periods to meet imminent capital calls, which in the grand scheme probably has a negligible effect on returns. There’s also a very limited chance that a limited partner defaults on a capital call. It’s extremely rare, and even if it does happen, there are remedies that would allow the limited partner to continue investing in the fund – rarely would all value be lost.

The reason these two issues aren’t talked about too much is probably because they’re not really major  issues to begin with. Putting aside the risk of default (which is incredibly small), let’s take a look at the effect holding committed capital as cash would have on a fund’s return. If you remember, in my model for a crowdsourced venture capital fund, I suggested that all committed capital would have to be called at the onset of the fund to make things logistically simpler – perhaps as the private equity and venture capital industries evolve, we’ll see more of this. Below I’ve modeled out a hypothetical private equity or venture capital fund’s cash flows under a normal model (which assumes that cash comes in right at the time of a capital call) and also for a model where cash is held/called at the onset of a fund (same impact on returns). I’m using 5% as an interest rate for the cash and the rest of the cash flows for both models are the same. Here’s what we get:

As you can see, there is clearly an impact on the fund’s IRR - a difference of around 1.3% in this case, but with a return multiple of 1.6x under both scenarios. Is this a significant difference?  I would say it’s definitely material, but it depends on the investor. The difference is probably significant enough to impact investment decisions and overall portfolio performance, and its why funds do not call capital upfront (negative impact on IRR, even though all other performance is the same) and why limited partners don’t hold cash. They assume they can earn even more than the 5% I modeled in on their cash. The only benefit derived from calling capital upfront or holding a commitment as cash is eliminating the risk of default, but as I mentioned before, it’s such a small risk in the first place that it does not make sense to protect against in such a way.  That said, I do stand by the idea that for different models such as a crowdsourced fund, you would still want to call all capital upfront, even if you sacrifice a bit of your IRR. 

VC’s: How To Benchmark Yourselves Properly

Plenty of VCs are guilty of deceitfully presenting fund performance – especially when it comes to benchmarking themselves against their peers in marketing materials. There’s a reason why so many LPs complain about almost every VC claiming to be “top quartile.” When raising new funds or providing updates, you too frequently find venture capitalists benchmark their performance improperly, often in an attempt to make themselves look better. Instead of these attempts, existing and prospective LPs would probably find it much more amicable if VCs candidly presented their relative performance - in fact, being candid in presenting performance and benchmarking goes a lot further than you would think with most LPs.  Here’s a guide on how to go about benchmarking properly and in a way that LPs will surely appreciate:

Selecting a Benchmark:

Use the Cambridge Associates Benchmark Statistics - don’t even consider any alternatives. Their benchmarks are far and away the most comprehensive. A common alternative you see is the ThomsonReuters (Venture Economics) benchmarks but they fall so short of Cambridge in terms of sample size that their statistical validity is questionable - sophisticated LPs are aware of this issue. The Cambridge benchmarks exhibit slightly higher returns because of their selection bias (all funds that Cambridge clients are in get pulled in so some consider it more of an “institutional quality benchmark”), but they are the industry standard and using anything else immediately raises doubts for LPs.

Don’t have access to their quarterly benchmark statistics? Simply contact them and participate in their quarterly survey and you’re receive the benchmark statistics free of charge. You’ll have to provide your quarterly financials to them but don’t worry, all data is kept confidential and your performance remains anonymous. They’ll even sign an NDA if you ask them to.

What to Present:

Determining the Proper Vintage Year: Cambridge Associates determines a fund’s vintage year based on the partnership’s date of legal formation - not by when a fund holds its final close and not by when a fund started investing. This means that under most circumstances you should do the same, unless there’s a special case where a fund took over a year to raise or if the fund was legally formed so late in a year and didn’t start investing until late in the next year that its most logical to use the next vintage year. If either case applies, make sure you footnote the situation properly.

 Determining the Proper Asset Class: Usually it’s not difficult deciding whether the venture capital or private equity benchmark is most appropriate. But in some cases where a fund has a later stage or growth equity strategy, LPs like to see it benchmarked using private equity benchmarks. If you have a strategy that straddles venture and private equity, consider using both benchmarks.

 Performance: You must of course show IRR, but make sure it is the IRR NET TO LPs (net of fees and carry) not the gross fund IRR. Nothing is worse than a VC that either purposely or inadvertently shows just gross IRR and then even worse, benchmarks it against the net to LP benchmark. It’s ok to show a fund’s gross IRR but if you decide to do so, you must also show the net IRR as the next line item. Generally, when benchmarking you want to show just the Net IRR because if there’s a large discrepancy between the net and gross, it draws attention to the effect management fees and carry is having on the return to LPs.

 In addition to the IRR also show and benchmark the distributed/paid-in and total value/paid-in multiples. Showing fund level cash flows is a plus too. The reason for showing the multiple is to help iron out some of the effects timing has on the IRR and give a truer sense of the fund’s performance. This goes back to being candid – you don’t want to omit multiples if IRR has been boosted by a quick exit, and conversely, for older funds you might actually be doing yourself a disservice by not including the multiples if IRR has been dragged down because of timing.

 Not Meaningful Performance: A general rule you can follow is that a fund’s performance is not meaningful and thus okay not reporting on and benchmarking if there has been less than three years of activity. Just make sure you footnote why you have decided that a fund’s performance is not meaningful and why it hasn’t been benchmarked. But also be sure to have the performance handy and expect to provide it if an LP asks for it.

Here’s an example of how fund benchmarking should look (note: figures are fictional):

 

Note that these guidelines apply to sharing fund level performance only and that the guidelines for sharing company level returns are much different – something I’ll eventually post about later. In the meantime, feel free to provide feedback or ask questions about these guidelines.