A flood of stuff for you this week: This week’s longer column has two parts. First an exciting announcement: famed inventor Dean Kamen, famed entrepreneur and venture capitalist Bob Metcalfe and U.S. Senator John Kerry are amongst the lineup for this year’s Lux Executive Summit in October. Stay tuned! Second, I share an exclusive look at the new intensive findings from Lux Research on where the smart money (or dumb money) is putting theirs, with an exclusive inside look at the findings. Then third, an older column, I wrote one year ago on the eve of getting married and now our first anniversary. Time flies!
Here it is: Venture capital (VC) firms invested $702 million in nanotechnology start-ups last year across 61 deals, slightly down from $738 million across 73 deals in 2006. But this VC spending is sharply out of sync with investment returns. Although application-oriented life-sciences companies have delivered the majority of VC returns in nanotech, VC firms consistently devote most of their funding to companies in other areas, according to a new report from Lux Research entitled "How Venture Capitalists Are Misplaying Nanotech."
"Historical trends in nanotech exit returns are out of sync with VC spending," said Jacob Grose, Analyst at Lux Research and lead author of the report. "While alternative energy is hot right now, healthcare and life sciences companies have accounted for a staggering $1.68 billion of the $2.57 billion total valuation of nanotech start-ups at IPO. Correspondingly, revenue multiples at IPO have been an order of magnitude higher for the healthcare segment (206.2x on average) than in the four other segments we track, yet last year more nanotech VC deals closed in manufacturing and materials (16) than in healthcare and life sciences (15)."
To uncover the latest trends in nanotech VC, Lux Research took a snapshot of its ongoing, comprehensive database of the field, which contains every round of institutional VC funding in companies that are commercializing nanoscale structures with size-dependent properties. The report concludes that:
VC spending remains highly concentrated. The top 5% of venture-backed nanotech start-ups as measured by cumulative capital invested have received $1.24 billion since 1991, equal to 32% of cumulative VC funding through 2007. The top three: 1) optical equipment manufacturer, NeoPhotonics, at $205 million, 2) lithium-ion battery producer, A123Systems, at $133.8 million, and 3) drug developer, Acusphere, at $104.1 million.
In 2007, the energy and environment segment attracted the most nanotech venture capital, with 17 deals (up from 13 in 2006) worth a total of $227.2 million. Notable firms receiving funding included aforementioned battery specialist, A123Systems, with $70 million in two rounds, nanocrystalline solar ink developer, Innovalight, with a $28 million Series C, and organic photovoltaic developer, Konarka, with a $45 million Series F.
U.S. domination became even more pronounced in 2007, accounting for 90% of total VC activity by value. The tiny state of New Hampshire alone accounted for more funding in 2007 ($76.5 million in two companies, Finetex Technology and Nanocomp Technologies) than all countries outside the U.S. combined ($70 million).
Many nanotech start-ups are showing their age. While VC firms have told Lux that they expect their nanotech VC deals to deliver returns in six years (longer than they expect in other investment domains), of the 66 nanotech start-ups which received their first institutional VC funding in 2001 or earlier, 58% continue to operate - implying that most nanotech start-ups are taking longer to exit than VCs had expected.
"Nanotech start-ups with technologies that are tailored to target one or two specific applications tend to be much more successful than companies who develop a broad technology platform with no clear purpose. However, the latter platform technology companies still generate valuable intellectual property," noted Jurron Bradley, Senior Analyst and head of Lux Research's nanomaterials practice. "Large corporations should roll up materials and platform technology companies that have already had their R&D and pilot-stage manufacturing paid for by VC firms, buying into hard-built technologies on the cheap."
"How Venture Capitalists Are Misplaying Nanotech" is part of the LR Nanomaterials Intelligence service. Clients receive: 1) State of the Market reports every six months; 2) ongoing technology scouting reports and proprietary data points in the weekly Lux Research Nanomaterials Journal; and 3) on-demand inquiry with Lux Research analysts. For information on how to become a client, contact John Schwartz at john.schwartz@luxresearchinc.com or (646) 649-9582.
Now, my column from last year:
Like the periodic table, the lessons of Sherlock Holmes are “elementary”, my dear reader.
In one story Holmes asked Watson, “Do you recall a conversation with Inspector Gregory during our celebrated search for Silver Blaze? I drew his attention then to the curious incident of the dog in the night-time. 'The dog did nothing in the night-time,' the Inspector responded. That, I told him, was the curious incident.” We’ll return to Holmes in a few moments and I’ll tell you about what I call the ‘curious incident of the dog in the blur.’
So, I done got hitched. Before our wedding I was tasked by my wife to do something my cynical nature is particularly well suited for: to imagine all the abstract low-probability things that might happen that could wreck havoc on our wedding. Torrential downpours, grandparents passing, groomsmen with lost clothing, passport mishaps, pest infestation, guest with a persistent and disruptive cough during our exchange of vows, muddied wedding dress.
Predicting a circumstance can sometimes help prevent its occurrence--if it’s controllable. And most dramatic changes in life, markets and science often occur from circumstances that couldn’t be predicted. And even if they could be predicted, they might not be controllable (earthquakes, torrential downpours). But that doesn’t stop the circumstances’ actors, 24-hour cable pundits or policy makers from explaining pretty much everything away. This is natural, part of our biology. We seek causation. We seek confirming evidence.
I’ve learned there are three words we don’t say enough and you almost surely never hear from newscasters, CEOs or most institutional investors. As Stevie Wonder sang, “These three words/ Sweet and simple/ These three words/ Short and kind/These three words/ Always kindles”.
No, not “I Love You”…but “I Don’t Know”
As Nassim Taleb has said, “We favor the visible, the embedded, the personal, the narrated, the tangible. We scorn the abstract.”
But it is the abstract that gets us. As I’ve written before and as Bill Miller of Legg Mason tells his analysts, 100% of the information we have (data and all the patterns we see in it) comes from the past. But 100% of the value of any decision (around an investment or life choice) comes from the future. And even the best scenario analyses can miss the very low-probability, low-frequency, high-impact events.
Here’s what I wrote quoting Taleb in this column nearly three years ago: "I have just completed a thorough statistical examination of the life of President Bush. For 55 years, close to 16,000 observations, he did not die once. I can hence pronounce him as immortal, with a high degree of statistical significance."
When we assume models are linear when they're not, we also increase the odds that we will pick the wrong model from which to make predictions. And as Keynes said, “It is better to be roughly right than precisely wrong”.
This cognitive bias and kind of reasoning made sense in our primitive ancestral past. If you don’t see a lion or an elephant, you’re pretty safe. But in our modern non-primitive environment, there are far more random, unpredictable low-probability events.
The point Taleb was making was our inability to deal with probabilities accurately and our tendency to reason inductively and incorrectly while presuming the future will continue linearly as the past (side note: this was something Ray Kurzweil drilled upon at MIT yesterday as he illustrated in example after example how technology improves at exponential—not linear--rates). People are horrible with probability and horrible with predicting risk. We just didn't evolve to weigh all the factors involved. Why? The logic is simple. In our ancestral environments, the people that acted quickly on emotions, ran away and survived. The people that sat around and over-analyzed situations got eaten by tigers. We don't have to run from tigers today, but our brains haven't changed much
since then.
And many people get confused over the difference between risk and uncertainty. Risk is when there's a *known* but *undesired* outcome. Uncertainty is having no clue and getting struck completely by surprise. Uncertainty is not knowing the underlying distribution. And as many find out the hard way, markets and much of life are uncertain.
This is when you're most vulnerable. These very, very low probability but high magnitude events. September 11th was such an event, so were the major historic stock crashes and bursting of tech bubbles. We call these: "Black Swans." Why
black swans? It comes from a philosophical thought experiment demonstrating the problem we have with inductive reasoning by asking this simple question: "How many white swans do you need to see to conclude that all swans are white?" The answer is infinite.
Because all you need to observe is just one black swan and the prior conclusion is false. Or try this admittedly silly thought experiment. Pretend you're a turkey born on January 1st, 2005. For the next 11 months, up through Thanksgiving, life is good. Actually, life is great. You (the turkey) wake up, get fed and walk around all day. Just in time to go to sleep and do it all over again the next day. And with each new day you become conditioned to assume that the next day will be just like the one before it. And this logic works all the way up till Thanksgiving. When suddenly--boom! Off with your head! The irony is that the closer the turkey is to death, the more confident it becomes. And so it is with many investors. We're not good at preparing for these low probability events--even though their magnitude could spell devastating consequences.
Now let me take you to two make-believe countries invented by Taleb. Mediocristan and Extremistan. First, we’re in Mediocristan and we take a random sample of 1,000 people. And we weigh them all. Next we take the fattest person we can find and we throw him or her in the mix. As a percentage, even that fattest of fat people won’t skew the groupvery much and they will represent a very miniscule percentage of the total. In Mediocristan, no single instance can influence the total: the exceptional in inconsequential
But in Extremistan, things are different. Instead of weight, think: income. Unlike weight, income is not normally distributed on a bell curve. If Warren Buffett walks into a bar, the average wealth just skyrocketed. The exceptional is consequential.
You can scorn the abstract in Mediocristan, but not in Extremistan. Harry Potter, Google, September 11, Long-Term Capital Management all live in Extremistan.
Back to Sherlock Holmes. So what is ‘the curious incident of the dog in the blur’? It has to do with confirmation bias and also illustrates the danger of watching too much CNBC, sucking up too much information (and why Buffett prospers in the quiet poise of Omaha).
Imagine this: You’re looking at a very blurry picture with a hidden dog and you can’t make it out. Meanwhile, my hand is on a dial that can increase the resolution.
If I turn the dial gradually and increase the resolution slowly, you don’t see the dog. But if I turn the dial more quickly, thus increasing the resolution quickly, you will see the dog more easily. Why? In the first case you are receiving more information (let’s say I turn the dial 10 notches each one increasing the resolution gradually; ie. once every 6 seconds for 60 seconds) and forming theory after theory of what you are seeing. In this case, you produce 9 theories.
If I instead turn the dial in 5 steps instead of 10, you are more likely to see the dog. In this case you produce only 4 theories of what you’re seeing. Instead of confirming theory after theory with consistency and confirmation bias, you have better odds of seeing something new.
This invokes the boiling frog syndrome. Throw a frog in boiling water it will jump out. But put it in room temperature water and gradually turn up the heat and it’ll get cooked. The danger of gradualism and information creep is ever present in markets with sell-side equity analysts who incrementally adjust their earnings figures and cluster like a herd.
Don’t be the frog. Be ware of blurry dogs. Listen for the ones that don’t bark. And watch for black swans—especially the kind that might derail a wedding!
Labels: alternative power, bias, power, Weekly Insider