Main
Welcome To Venture Examiner

On Venture Examiner I share my thoughts on the venture capital industry, alternative ways of funding, supporting and fostering innovation, opportunities in the emerging markets and other topics relevant to my experiences....MORE.

DISCLAIMER: The content of this site reflects my thoughts only and is not affiliated with any other party...MORE.

I also share other things that are interesting or important to me on my personal site: aakarvachhani.com

 

You can also follow me on Twitter (see below)

Follow Me On Twitter
Can't display this module in this section.
Subscribe to E-mail Updates:
Can't display this module in this section.
Search Venture Examiner
Sunday
Jan312010

Venture Capital Overhang: Shrinking

With 2009 now behind us, full final year-end venture industry data is available. There’s plenty to glean from all the fundraising, investment and exit data. Much of it tells us what we already knew or expected: fundraising and investment are down, and exits have improved, but just slightly. There’s so much you can analyze, but I’ll focus on something I’ve done in the past, which is looking at the “venture capital overhang.” This is the difference between the aggregate capital raised by venture capitalists and the amount invested. It gives us a rough idea of how much capital VCs have available for investment, sometimes referred to as “dry powder.” The chart below shows venture fundraising, investment, the difference between fundraising and investment (as the overhang) and the cumulative overhang for the last ten years.

 

The cumulative overhang for the last decade for the U.S. venture capital industry totals close to $90 billion, using my methodology and data from PwC, Thompson Reuters and the NVCA. As with so much of the data on the venture capital industry, the calculation is not perfect. Things like management fees and recycled capital are unaccounted for. There’s also the issue of investments made outside of the U.S.  which are not captured in the PwC MoneyTree data.  Rather than focusing on exact numbers, its more important to focus on trends and to look at the big picture.

For one, there’s clearly capital out there for venture capitalists to invest. It’s probably becoming more concentrated across a fewer number of firms - as I mentioned in my last post, good firms will continue to be able to raise capital. The overhang number is down from my previous calculation earlier this year, which signalz to that capital will be a bit scarcer.  Going forward, we should see more years like 2009 and 2003 where the levels of investment and fundraising have less of a gap and less of an overhang is created. Now, you don’t want things going in the other direction, where we have more capital invested than raised because that would of course be unsustainable. But then some would also argue that the huge levels of overhang amassed in years past were also unsustainable, which is probably true.

There needs to be certain level of reasonability maintained in the industry and less overhang will force venture capital firms to be more prudent in deploying capital. This doesn’t mean, however, that great new ideas won’t get funded, because VCs clearly have plenty of dry powder. If anything we’ll see more early / seed stage deals which not only require less capital, but have more potential upside and also bring the industry back closer to its roots of more risk taking. 

Thursday
Jan142010

Venture Fundraising in 2010

2009 was clearly a difficult year for venture firms – continued turmoil in the public markets and the broader economy prolonged the dearth of venture-backed IPO and M&A activity, extending the liquidity drought for venture firms. Illiquidity negatively impacted fund performance, and more importantly the confidence limited partners (investors in venture capital funds) have in the asset class. The drop in confidence is most evident in their commitments to venture funds, which in 2009 fell significantly. According to Dow Jones, “overall VC fund-raising fell 54.6% to $13 billion across 120 funds from the $28.7 billion collected by 204 funds in 2008. It was the slowest year since 2003.” Here are a few things to watch for in 2010 in terms of fundraising:

Commitments to Top Tier Funds:

Fundraising totals for 2009 would have been worse had it not been for New Enterprise Associates (NEA) closing its thirteenth fund with $2.5 billion in commitments. While the fund took longer than expected to close, the fact that it was eventually able to do at a such a large size shows that institutional investors still have an appetite for firms like NEA that have a record of consistently delivering top quartile returns. This will be a theme going forward – we will see the most sold performers (firms such as Sequoia, Kliener Perkins, Matrix, Battery, etc.) continue to be able to raise capital, but fund sizes will still come down. If for some reason we see a top firm unable to get close to its fundraising target, it would be a sign that limited partner perception of the asset class is worse than feared. The shockwaves would be felt across the venture universe.

 The Numerator Effect

Over the past couple of years, the “denominator effect” has been a central issue for most large institutional investors / limited partners. Some quick background for the unfamiliar: If you think of an institutional investor’s allocation to venture capital as a fraction, the denominator is the total value of their total investment portfolio. The numerator is what is invested in venture capital. Stocks and bonds are traded daily, whereas venture capital is only valuated quarterly. When stock prices fell during the recession, it brought down the value of the overall portfolio, or the denominator, but at the same time, the percent actually invested in venture capital went up because the value of venture portfolios 1) are reported on a lag and therefore had yet to be written down in line with the public markets, and 2) didn’t declines as much relative to marketable securities.

In 2010, what we have already seen is that the denominator has rebounded – in line with the stock market (for example, the NASDAQ was up around 40% in 2009). However, the numerator, or value of institutional investors’ venture portfolios has remained suppressed – again, because venture capital valuations are reported on a lag. The real value of the numerator won’t be known until final year-end 2009 data is taken into account, which won’t be until April. Once that happens, institutional investors will really be able to get a true sense of where their allocations stand. This means that the second half of 2010 should see more commitments than the first half.  

 Attrition:

Early in 2009, PE Hub’s Dan Primack released a list of “The VC Walking Dead.” These were venture capital firms that were officially in business but which no longer had enough cash to add new portfolio companies. Presumably that meant they will no longer try or be able to raise subsequent funds.  Expect the list of firms that fall under this category to grow in 2010. The bar for venture firms will be much higher going forward. The amount of capital committed to the asset class will probably never (or not for a really long time) return to the levels of 1999-2000, or even 2007 for that matter. It’s the general consensus that there was too much capital in the venture industry and limited partners weary of the asset class have every reason to be extra judicious with their commitments. That spells bad news for undifferentiated firms, inexperienced firms, and firms with poor track records. 

Sunday
Dec272009

Drawing From Y-Combinator - A More Perfect Crowdsourced Venture Fund

I've written a few times now about the idea of a crowdsourced venture capital fund - where there would be a large number of small investors, each playing a role in the fund's investment decisions. It’s my belief that as the venture industry evolves, the disconnect that exists between investors, venture capitalists, entrepreneurs, and the tech community can be bridged well through such a fund. If you'd like some more background on what my ideas for a crowdsourced venture fund are you find it here, and here.

 I got thinking about a crowdsourced venture fund again after reading some more about the great stuff Y Combinator does. Y Combinator provides seed funding for startups, but money is just a small part of what they bring - typical investments are less than $20,000. Instead, where Y Combinator really provides value is in their work with the startups they fund. They provide hands-on guidance to help startups become successful, including forming the company, legal issues, developing the product(s), managing the company's growth, and even finding future funding. In an age where the cost of starting an internet company has gotten pretty low, Y Combinator, which has helped spawn great companies such as Disqus, Loopt, Scribd, Xobni, and Reddit, provides something more valuable through its expertise and connections. It got me thinking that a crowdsourced venture fund would need to be able to do something similar.

A crowdsourced venture fund would be best suited for making tech investments; particularly early-stage tech investments where the backing of a crowd (in this case the LP base as well) could help propel portfolio companies. You would also be able to draw from the wisdom of the crowd to help with any problems faced by the startups invested in. Here, you have an instant network, as long as the LP base remains on the tech-savvy side, which you would expect. But what about the nuts and bolts of a company and nurturing it properly early in its life? The truth is that most traditional venture capitalists don't do much there as you would think, which makes YCombinator special. In a crowdsourced fund you would ideally want Y Combinator-type VCs armed with their own connections which, along with input and backing from the crowd, would really create an ideal situation. You would be able to help entrepreneurs effectively through a variety of issues by drawing from the crowd, only having to make sure that the crowd is sufficiently engaged to want to lend support. Part of this is achieved through their investment into the fund itself. Part of it is also making them involved in the investment process.

What would a crowdsourced fund do with a very large pool of capital? It would be able to do what Y Combinator can't do: continue to fund the companies at later stages. Instead of having other venture firms come in for a series A or B round, ownership could be maintained in the companies as they grow. Of course you could always push for a larger ownership with the seed funding as well, but you have to be careful there as you want the entrepreneurs to be motivated with significant interest in their companies.

And what about the vetting investments? Y Combinator has an application process for companies, but for a crowdsourced fund, you would probably want a combination of companies applying for backing as well as the fund's VCs going out and sourcing investments in a traditional manner. Both sources of dealflow would be pooled and, as I've mentioned before, the crowd, or LP base, would be able to vote on the most promising companies, which the VCs would then use as input in making their final decisions. The reason you wouldn't leave it up purely to a vote is that you need to protect the confidentiality of potential investment and so voters would not have complete information when making decisions. You would also use the wisdom of the crowd by voting/collaborating on solutions to problems companies face that can't easily be solved by the VCs and would benefit from having input from the crowd. While the crowd, or LPs, wouldn't be compensated for their participation, they all have their investment in the funds at stake as a motivator.

The thought of a crowdsourced venture fund is definitely idyllic, and maybe even more so if you want to try to do some of the things Y Combinator does, but as capital starts to take a back seat to the other things venture funding should provide, it’s a model that seems to make more and more sense.

Previous posts on Crowdsourcing Venture:

Crowdsourcing Venture

Another Take on Crowdsourcing Venture

Thursday
Dec172009

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. 

Sunday
Nov222009

Observing Larger Trends in Healthcare VC

This year, venture investment into the healthcare sector surpassed investment in the traditionally heavyweight information technology sector for the first time in over a decade. This trend has been well covered and there are plenty of reasons as to why venture capitalists have shifted focus to healthcare (less cyclical, lower valuations, less fundamental change, etc).  What’s interesting about the trend is that there’s been a historical correlation between difficult economic environments and the relative level of healthcare venture capital investing.

Here’s a look at healthcare investment as a percentage of all venture capital investment over the past 10 years: 

 

What you notice, besides the clear increase in relative healthcare investment over the past decade, is two dramatic spikes: The first in the years following the tech bubble – which is partly the result of VCs shying away from the tech sector while those investing in healthcare maintained their investment pace. Plus, many tech VCs looked for safety in the relative less volatile healthcare sector. The second spike is more recent - starting last year as the recession took hold and what can be considered a small venture bubble burst.  The reasons for this spike are similar as the previous, and as with the previous spike, the absolute dollars invested in the sector have still declined.

What’s also interesting about the two spikes is that they are predicated by a short decline in relative healthcare investment, which makes you wonder - is the relative level of healthcare investment perhaps a leading indicator for venture capital bubbles? It appears so. During bubbles, VCs are more apt to pour cash into what can be more lucrative, but also more risky technology investments.

This recent spike, where healthcare investment has reached almost 40% of all venture investment, is probably just that - a spike. Unlike earlier this decade, the levels of healthcare investment should come back down again as clean technology and IT investment pick up. But the case for healthcare investment remains and we’ll probably see it remain around 30% of venture investment going forward. Demand will continue to be driven by an aging population, continued technological progress, declined pharma productivity and government programs such as those pushing healthcare IT and increased access to healthcare.

Data Source: PricewaterhouseCoopers/National Venture Capital Association MoneyTree™ Report, Data:  Thomson Reuters

Sunday
Nov152009

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.

Saturday
Nov072009

The 10-Year Venture Capital Return Doesn’t Mean Anything

Over the past year there’s been a lot of talk about how venture capital returns over a ten-year time horizon will no longer look so hot once this year is out. This is because after this year the ten-year timeframe will no longer include returns from 1999 – a year featuring huge exits through quick IPOs in what was the last year before the bursting of the tech bubble.  The 10-year venture capital return currently stands at 14.3% (as of 6/30/09 – VC returns are always on a quarter or quarter and a half lag), but has been in a steady decline. There’s been some speculation that the decline in this return will have a negative impact on the venture industry’s attractiveness to investors, but I don't think that is necessarily the case.

Over the last four quarters, the 10-year return has tumbled from over 40% to 14%. For a visual, here’s what the recent drop in the 10-year return looks like:

Before going further, here’s some background on how the return is calculated:

The 10-year return is an end-to-end venture capital fund-level return, meaning it looks at the fund-level (read: not company level) cash flows – to LPs only, meaning it is net of fees and carried interest ( a good thing because this is the true level of return an investor in venture capital funds would have received). The return is a standard IRR calculation except that it in the period in which it starts, in this case the second quarter of 1999, the cash flow pulls in the negative starting net asset value of all the funds comprising the index. Going forward, all contributions and distributions are netted to get the cash flows for each quarter (for timing purposes, the figure is assumed at the midpoint of each quarter). For the final quarter of the calculation the current net asset value of all the funds in the index is added in as a positive cash flow.

What’s the recipe for this drop in the 10-year return? For one, huge IPOs started tapering off after 1999 while at the same time, inflated valuations made the beginning negative cash flow much larger and tougher to overcome, dragging down returns.  Then add in the fact that there hasn’t been another period of fantastic exits since, and that the past year has been absolutely dismal for venture-backed exits and you get a free-falling 10-year venture capital return.

But how important is the 10-year return really? The whole notion that the 10-year return’s fall will have a negative impact on the venture industry is actually hugely flawed. The pure numerical drop in the 10-year return should not be what deters investors. Why? Because sophisticated investors (limited partners) are smart enough to not have been looking at the 10-year return anyway. If anyone looked at the 10-year return as their reason to invest in venture, they clearly do not understand the asset class enough to be making decisions on investing in it. 10-years, despite being a round number, is just as arbitrary as any other number. Nothing makes it different than looking at 9 or 11 years. Even if you are comparing to public markets. The only thing that makes this current case unique is that we are on the brink of excluding a great period for venture. But by now, most investors have come to accept the tech bubble as an aberration, and new expectations for venture returns are much more tempered, albeit still relatively high because of the illiquidity and risk.

Plus., if you want to get technical, the 10-year return isn’t realistic for investors because to have gotten those 10-year returns they would have had to have gotten into funds in the prior one to four vintage years – the ones which were actively investing prior to the bubble and able to exit during the bubble.

Finally, no legit VC I can think of is selling their fund based on 10-year industry returns. In fact they probably look to avoid having to talk about the bubble years because the business is so different now. What happened 10 years ago is pretty much irrelevant. VCs are judged on their performance over the last five years, how their strategies can take advantage of the current environment, and their ability to exit deals over the past year and going forward.

What the drop in 10-year return does more than anything is highlight a major shift in the venture industry - a shift that investors and VCs were aware of years ago and probably ever since the bubble burst. This drop in the 10-year return is not new information and could have esily been predicited a few years ago. We’ve gone from some 250 venture-backed IPOs in 1999 to just 10 so far this year. We know there will be attrition in the industry, but investors that stick with venture capital should be better off because of it. Innovation continues unabated and top mangers continue to provide quality returns through building and growing companies - the way it should be.

Tuesday
Oct272009

VC Shift To Seed Stage Investing Is For Real

Third quarter venture capital investment data was made available last week which prompted me to re-examine the data I looked at in my previous post on the recent spike in seed stage investing by venture capitalists. More specifically, I wanted to see if the data trend held true to what seems to be going on anecdotally - are venture capitalists really dramatically shifting their focus to early and seed stage deals? The answer still seems to be a resounding yes. The chart below is a more detailed look at the percentage of all initial investments allocated to seed stage deals by venture capitalists by quarter since the post-bubble period (2001-2003).

You’ll notice the spike in the second quarter of this year, but the third quarter still represents the highest level of relative seed investment since the second quarter of 2005. Furthermore the data trend still clearly shows that venture capitalists have indeed continued to shift more of their focus to seed stage investing. Why? Well, as I’ve covered before, it’s a reflection of a few factors:

  • Lack of syndicate partners for later stage deals
  • Lack of capital or adequate reserves for later stage deals
  • Skepticism around the medium term prospect for exit (IPO and M&A markets)
  • The realization that more risk needs to be taken to achieve desired returns
  • A rise in the number of quality “venture-backable” start-ups and entrepreneurs (partially a product of the state of the US economy)

Data Source: NVCA PricewaterhouseCoopers/National Venture Capital Association.

Monday
Oct052009

Why Have Venture Capitalists Shifted To Seed Stage Investing?

On the heels of my last post on venture capital’s role in innovation, I decided to take a look at how active venture capitalists were in funding companies that are early in their innovation lifecycle. The proxy for this is the earliest stage at which venture capital can come in – seed capital or start-up investment. What I found was surprising. The chart below shows the percentage of new venture capital investment (in terms of dollar value) that went to seed /start-up stage companies each year over the last 15 years. This is the best way to look at this type of data - absolute figures tell you more about what the venture market is doing overall - to understand real deal trends you have to examine changes in proportions of investment over time.

What’s clearly surprising about this data is the recent spike in seed investment (relative to other stages) by venture capitalists. 32% of all new venture investment this year has gone to seed stage deals. What’s driving this?

I know for a fact that pure seed-stage venture funds have had trouble raising funds over the past few years (relative to balanced and later stage funds), which means that traditional venture funds are now leaning more towards seed stage deals. I can think of a few reasons why:

  • Venture capital firms are being forced to engage in cheaper, earlier stage deals because syndicate partners are increasingly tougher to find for later stage deals.
  • Venture capitalists are still worried about the future of the exit markets (the IPO market and M&A activity) and therefore are hesitant to engage in later stage deals. These are deals in which they would have to reserve adequate capital for if subsequent venture rounds are needed to sustain the companies. We've seen a lot of firms face reserve shortfalls over the past year as the exit markets have essentially been closed. Venture capitalists with the expectation that exit markets will remain tight would clearly be detered from making later stage investments and would perfer less capital intensive earlier stage deals.
  • Perhaps venture capitalists are simply going back to their roots and finally taking more risk again. There’s probably the realization that outsized returns can only be attained by generating higher return multiples off of earlier stage deals. Some of this might be pressure from limited partners - low multiple later stage deals just are not attractive, particularly when you consider the fees and illiquidity that come with commitments to venture funds.

Regardless of the reason, this shift to earlier stage investing can only be a good thing for the venture industry. The firms that are truly good at building companies and working with entrepreneurs will stand out and perhaps help repair the image of the venture industry.

Data Source: NVCA PricewaterhouseCoopers/National Venture Capital Association.

PricewaterhouseCoopers/National Venture Capital Association. MoneyTree™ Report, Data: Thomson Reuters.
Sunday
Sep272009

Venture's Role in Innovation

A big debate was spurred by Vivek Wadhwa last week when he lashed out at the venture community through a post on TechCruch.  Wadhwa contends venture capitalists do little, if anything, for innovation and even detract from innovation. His post comes on the heels of the National Venture Capital Association (NVCA)’s most recent “Venture Impact” report which highlights the importance and impact of venture-backed companies in the macro U.S. economy.  The report does make some outlandish claims, such as 12 million jobs, and 21% of GDP can be attributed to venture-backed companies. Clearly venture capitalists cannot realistically lay claim to those statistics. The report, meant to be a lobbying tool more than anything, includes any company that has received any type of venture funding at any point in its life under its definition of “venture-backed,” which is pretty misleading. Not to say that venture capital does not have a positive impact on the economy - it does - but it is definitely not responsible for the numbers the NVCA presents.

Wadhwa took things a bit further in his post though, and cites research (itself a bit questionable) that shows not only do the vast majority of successful entrepreneurs not need venture capital, but that those who do take it see their companies become less innovative. This leads Wadhwa to conclude that “gold digger” venture capitalists with MBAs have increasingly been simply funding “me-too” companies resulting in high failure rates and declining returns. Looking past the VC-bashing, his main argument really is that venture capital does not facilitate innovation and therefore does not play a meaningful role.

The argument that venture capital does not facilitate innovation is really not much of an argument at all; aside from the rare cases when VCs start their own companies, entrepreneurs are clearly the innovators - no VC would argue with that. So then the next question is around the role venture capital plays. VCs are not simply “middlemen” as Wadhwa states. It would be ignorant to group VCs into one bucket - some are better than others, but almost all VCs play a role in their portfolio companies’ management and strategy. Wadhwa points to research that shows a company’s innovation decreases after it receives venture capital. This is actually a product of maturation. Most companies far enough along to receive venture capital funding would see a decline in innovation regardless of whether or not they received that venture funding.  Venture investment is meant more to take an innovation to the next level, raising its impact and growing its reach, not to prompt innovation.

The role venture capital plays varies by industry as well. In the more visible internet sector, there is limited need for venture capital investment, particularly because the cost associated with starting and scaling web startups has become so low. Venture capitalists won’t always admit it, but it’s a space most do not understand well, and because of its high visibility they catch the more flack for failures. Other sectors such as biotech and cleantech need venture investment to scale. Innovation can be halted if not for investment by venture capitalist. Wadhwa does not account for this at all.

Looking at the big picture, there’s not necessarily causality between venture investment and innovation, rather the two go hand in hand. Innovation can only have limited impact without scaling which is often made possible by venture investment. At the macro level, as the US is faced with increase competition from abroad from countries like India and China, innovation will be paramount in our global competitiveness. The only way the US can continue to compete and maintain its current standard of living is by creating new jobs through innovation. Entrepreneurs will be at the forefront and VCs will continue to be there to back them. The innovation ecosystem is fragile and the last thing it needs is people like Wadhwa causing an unnecessary break in trust.

As an aside, Wadhwa in his piece also mentions that “VCs are looking for bailout money and tax-breaks.” I’m not sure what the basis for this claim is. Surely anecdotal evidence is not what bears the proof. There’s not a single VC I know of that would want to touch bailout money, given the caveats it would come with. Nor are VCs aggressively looking for tax breaks. The only major activity on that front is resistance against a change in capital gains tax, which is reasonable. The one exception where tax breaks have been asked for by VCs is in the cleantech industry where government subsidies have led to increased investment by not only venture capitalists, but by a wide range of investors. And finally, I’ve written about this earlier, but venture capitalists are sitting on approximately $120 billion of “dry powder,” not a figure that indicates the VC community is looking for handouts.