Blog Post

The shakeup of Kleiner Perkins exposes the short comings of venture capital

Valley venture capital firm Kleiner Perkins’ aggressive bet on cleantech has turned out to be a major driving force behind its struggles and recent shakeup. We’ve covered the topic many times (see the dangers of righteous investing) but Pandodaily nicely captured this week the essence of what’s at stake for Kleiner’s leader John Doerr — the guy who ushered Kleiner into its internet investing heights of the 90’s and led the decision to go whole hog into cleantech in recent years — if he can’t right the ship.

However, underlying Kleiner’s cleantech missteps is the reality of the short comings of traditional venture capital itself. Entrepreneurs, tech execs and media in the Valley look to the institution of venture capitalism as being the engine of innovation and the revolutionary spark needed to ignite disruptive change. But it’s actually just one way to fund innovation and it’s only proved to be particularly good at backing digital technologies based on the quick, large and predictable growth curves that Moore’s Law provides.

Workers inspecting panels in Solyndra's factory in April
Workers inspecting panels in Solyndra’s factory

The square peg in a cleantech hole

As we’ve written before, traditional VC hasn’t proven to be a good model for funding many of the sectors in cleantech, such as making solar panels, producing electric cars, and developing biofuels. It’s not good at funding basic scientific research, like driving breakthroughs in battery chemistry.

Many of these technologies have needed more money and taken longer to scale up from lab to commercial product than afforded by the traditional venture capital time frame. Just read the story of SunPower’s journey, and it’s enough to make you avoid the energy sector completely.

The energy startups also have failed to find as many exits, both IPOs and acquisitions, as desired, partly because the companies outside of tech don’t acquire startups at the pace that tech companies do (yet), and partly because commercial products and revenues have seemed to take longer for energy manufacturing companies to produce than web companies. As Bill Gates has put it, Moore’s Law has spoiled us.

Row of Fisker Karmas
Row of Fisker Karmas

For those of you who haven’t spent years following VC funds, this is how they work: traditional venture capital firms raise funds from groups of limited partners (LPs), which are made up of outside investors, like endowments, large corporates, wealthy individuals, and even peer investors. A VC firm can invest in whatever they want, but in order to continue to raise future funds from these LPs, the firm needs to make money for those LPs off of a few big exits from their investments. If the fund performs badly enough for long enough, they’ll have a hard time raising future money — and a venture firm’s lifeblood is its fund.

As many of us know, a lot of the venture capitalists that invested in young companies in these energy manufacturing sectors lost their shirts and, like Kleiner, had to restructure and change their investing strategies. Sure, investments in anything digital that still adheres to Moore’s Law, like Kleiner’s Opower and Nest, are a much better fit with the traditional venture capital. And of course there are outliers like Tesla that braved all odds and was led by a visionary.

The evolution of VC

But part of Kleiner’s problem has also been that traditional venture capital has been evolving. Accelerators now crowd the space trying to recreate the success of Y Combinator and Tech Stars. 500 Startups is out playing Moneyball and winning over young entrepreneurs. VC firms are experimenting with new things like design programs, and hiring “Communication Partners” in ways to stay relevant. VC is broken in other ways beyond its dabble with cleantech.

Every couple of months someone writes an article about how we were promised jet packs and we got Facebook. Well, Kleiner Perkins tried to create jetpacks and it got fracked. Disruptive innovation outside of digital in age-old industries like manufacturing, energy, chemicals, and transportation is expensive, difficult and not for the faint of heart.

A lot of that jetpack innovation will have to come from outside of the Valley. There are many other ways to fund innovation, like R&D from big companies, various methods of government support, academia, and investing from outside VC like private equity. Unfortunately these funding methods are much slower moving and their backers are less willing to take risks than VCs.

Corporations in particular have started to spend more on funding these types of energy innovations. They have the balance sheets to tackle these hard problems both through investments in startups as well as internally. I don’t think the corporates will be the answer to the cleantech VC bust, but they’re interesting nonetheless.

At our Roadmap conference in November, Om asked GE CMO Beth Comstock why a young person would want to work at GE when there are hot companies like Facebook out there. If there’s one compelling reason why anyone would want to, it’s because a company like GE can invest in R&D and tackle some of these massive, basic science, infrastructure and manufacturing problems. Because who’s more likely to deliver a jetpack, GE or Facebook?

19 Responses to “The shakeup of Kleiner Perkins exposes the short comings of venture capital”

  1. I am not so certain that this demonstrates the failure of the VC funding model rather than simply the relative difficulty of the problems. I think if you had asked 70 years ago, what will come first: a car powered by a battery the size of a small suitcase that can drive across the country or a computing machine that can (do whatever number of computations a computer can now do)? I think most scientists would have got the wrong answer. It simply turned out that the computing problem was much more easily solvable than the battery problem, but there was no way to know that a priori. There has been no profitable way to fund battery research over the past century — except that interminable Bunny ad.

    Further, the businesses created by Moore’s Law have had economic moats (esp for small firms) in a way other innovations do not have. This has significantly distorted our economy. Consider the advances in things like metallurgy. If a scientist at U.S. Steel creates a slightly better steel that is 1% stronger; 1% lighter; and 1% less brittle, he gets a pat on the back and a $5k Christmas bonus. Further, US Steel has a difficult time protecting the IP. So smart people go work at Google and Goldman Sachs, not US Steel.

    Think about the gene sequencing problem — we still do not know the answer. Is cheap gene sequencing going to open up a world of profitable drug development opportunities? Or is the relationship between genes and phenotypes much, much more complex than we suspected and so not useful?

  2. Kevin Strong

    “Why would anyone need a camera in a cell phone?”, a John Doerr question to me when I was presenting to a full Kleiner Perkins partners meeting in early 2002. (I was presenting the spinout of Conexant’s CMOS image sensor business, which they did invest in as a merger with one of their portfolio companies, and we already had camera module design wins in early 3G phones for the U.S. market).

  3. Felix Hoenikker

    I don’t believe its the VC model thats the issue. I believe its primarily an issue of VCs not understanding what they’re investing in. Furthermore the antiquated spray and prey, 1 in 10 will be a home run, simply does not apply when there are deep rooted incumbents. Lastly I don’t believe cleantech is nearly as large as anyone cared to realize. Money was cheap and cleantech was the new buzz word, of course the MBAs are going to rush to raise funds and LPs don’t know any better.

  4. Great article! I am working in East Africa right now and continually examining the local ecosystem of innovation and access to funds. There is a growing tech sector, which funds like the Savannah Fund have taken the initiative to address, but there is a significant need to elaborate models which will support the future of local technology innovation here, both in hardware and software. Google is definitely leading the way on that front from what I can see.

  5. Tesla WAS actually backed by VCs to start with, but they almost starved the company of capital at a critical early stage and Musk turned to 1) his own money 1) pre-selling cars and 3) government loans to keep the company going when VCs bailed.

    In that respect, VCs demonstrated a lack of the intestinal fortitude required to see a mega project through!

  6. Great, great article.

    The article is really 2 in 1: the VC story and the clean tech story. Regarding the latter – we need to tackle the definition of “clean tech” – the current approach often is too broad.

    I think Nest and Opower are terrific examples of what i call “energy + IT” companies. In my view, this is a subset of clean tech.

    These types of companies lend themselves more readily to the traditional venture capital model than, say a solar PV innovation business.

    Actually, i see the glass half full on Kleiner Perkins – how many clean tech VCs have had winners like Opower, Nest, Tesla, and Clean Power Finance.

    Pretty impressive.


  7. Daniel Kahn

    “Sure, investments in anything digital that still adheres to Moore’s Law, like Kleiner’s Opower and Nest, are a much better fit with the traditional venture capital.”

    How exactly do Nest and Opower adhere to Moore’s Law?

  8. The missed points in this article are that many cleantech. companies (solar, wind, storage, fuel cells, autos) depended on two other financing components both of which have dramatically declined worldside due to the 2009 recession.

    1. Unpredictable and decreasing state & federal tax subsidies (e.g. solar, wind, auto) to make them economical.
    2. Project finance (debt) for the installations. Project finance dried up in 2009 and with declining (and unpredictable) government subsidies many projects were no longer financeable.

    Many companies failed or merged as a result of these interrelated issues. (for the world) is these technologies are now close to parity with conventional power sources.

  9. Great article. And, yes, VC “fund” model is not well suited to the challenge for green or material science in general. We need more funds like Lux.

    I do have some challenges:
    1) whose more likely to fund an all-electric car, GM or a ex-paypal exec?

    2) There’s a deeper root causes for green tech and physical / hardware investment in US, regardless of source of funds: We’ve lost our tool and die skilled worker. The result of decades of off-shoring.

    3) We have to tease apart science/invention from innovation. For example, AT&T invented the transistor. Other innovators made all the money. The counter example, Genetech shows where science was a good bet.

    Keep up your great work. Steve Jurvetson has some great slides about how Moore’s law applies to everything – lighting, transportation, but the doubling time is years, or decades.

  10. Michael Sinsheimer

    Time horizons to exit and resources required there are key to determine which VCs will invest in “innovation” and longer term plays. In Medtech, we typically need 7-10 years of patient money to be involved given the regulatory pathway and other pre-commercial activities. We are working on technologies that can result in improvements in quality of life, extend life and/or help enhance our healthcare system. The more inpatient VC “healthcare” money is targeting ehealth given the seemingly lower regulatory burdens although 23andme may change that perspective. Remember with lower burdens, the entry barriers are low too leading to more competition and smaller returns.

  11. Trinadh Yerra

    Starting a hardware company is costly, let alone disrupting it. It needs consistent investment and takes long time, which is missing with these VCs. It just shows the motto of “change the world” of these VCs is a hollow fad.