Greenhaven Road Capital letter to investors for the fourth quarter ended December 31, 2016.
Dear Fellow Investors,
In the fourth quarter of 2016, we generated positive returns of more than 5%, bringing the full-year 2016 gross return to 18.7%. Please check your individual statements for net returns, as they will vary by class. This compares to 12% for the S&P 500 and 21.3% for the Russell 2000 over the same period. We continue to believe that shorter-term (quarterly or single-year) relative performance measurements are largely irrelevant, whether favorable and unfavorable. While we are pleased to have outperformed the S&P 500 in each of the last four years, we consider it more meaningful that, over the last three years, five years, and since inception, we have outperformed the major indices.
As we go on this journey investing together, I try to keep you informed of how I am approaching investments. What are the frameworks that I am using? What am I paying attention to? Clearly, it would be less time consuming to just send statements, but my aspiration is to provide a higher level of understanding and transparency. Not only because this is what I would want as a limited partner, but also because I believe it increases the likelihood that you all will “stick around” when we hit inevitable rough patches. Our concentrated, value-focused, contrarian investment style will inevitably zig when the market zags and zag when it zigs, leading to both outperformance and, at times, underperformance. Knowing what we own and why we own it is your right as an LP, and should provide comfort in difficult times.
Where Are The Backhoes? Buy More Fiat
In the Q3 letter, I wrote about the attractiveness of “invisible companies” that lay outside the vortex of indexing. I have also written about the benefits of being a small fund, allowing us to pursue the broadest range of investments, and have paid homage to Marc Andreessen’s “Why Software is Eating the World” thesis. During the fourth quarter, I found myself thinking a lot about backhoes of all things.
Some people follow Kim Kardashian on Twitter; I follow investors and people who post anything and everything about Charlie Munger, Mohnish Pabrai lectures, and links to Howard Marks and Murray Stahl essays the moment they drop. It was through Twitter (I wish I could give better attribution) that I came across a 1996 column by venture capitalist Bill Gurley, “Backhoes Don’t Obey Moore’s Law: A Story of Convergence.”1 I had never read this 20-year-old article before, but it instantly resonated. In the age of dial-up internet, people could see the future of connected computers and the benefits of the broad availability of high speed internet, but Gurley effectively diagnosed that the bottleneck to realizing the utopia of computers connected with broadband would not be the computers, but rather the ability to lay the pipes which would enable the connections. Unfortunately, backhoes only improve at 12% per year, far more slowly than computers.
Obviously, the evolution of the internet and the passage of 20 years answered many of the questions posed in the article, but as I look at new and existing investments, I now try and understand what might be the metaphorical “backhoe” for the given situation that will delay convergence or the demise of an existing product. The common narrative of our times is that virtual reality, 3D printing, self-driving cars, blockchains, and drone delivery will transform industries. Fortunes will be made and fortunes will be lost. Often the mispricings we exploit exist because the market is over-emphasizing the speed at which change will happen, and underestimating the earnings power of the incumbent business.
I believe that one of our holdings – Fiat – is a situation where the market is/was forgetting about the backhoes. Over the course of the fourth quarter, we substantially added to our Fiat position (which we have owned for 3+ years) for two reasons. The first reason was “backhoe” related. Fiat has underinvested in self-driving cars. If that is not “bad” enough, there is a vision of the future in which the model of individually-owned cars will go away as Uber and self-driving cars converge. Why own a car when it is not used 96% of the time? Just hail a self-driving car on your smartphone and avoid the cost of ownership. In this draconian scenario for the legacy of auto manufacturers, fewer cars are sold, and the ones that are sold are made by technology companies, not car companies. For a futurist, Fiat is one of the most disadvantaged auto manufacturers. As a value investor, I don’t think it matters, there are “backhoes.”
The futuristic vision of car ownership being replaced by an Uber-like fleet of self-driving cars has several likely backhoes. One of the best examinations on the issues facing autonomous cars is this presentation by Frank Chen of Andreessen Horowitz (http://a16z.com/2017/01/06/selfdriving-cars-frank-chen/). There is a massive leap that has to be made from today’s “assisted driving” cars, which have advanced cruise control and can parallel park, to a “fully autonomous” car that does not even have a steering wheel and will never be driven by a human. Highway driving is relatively easy, but city driving is more complicated; sunny days are relatively easy, but snow-covered roads are more challenging. A self-driving car that works 99% of the time but still relies on occasional human intervention is not the real game changer; 100% reliability is necessary to change the paradigm. To get to 100%, outlier events such as navigating construction and accidents must be overcome. Other challenges/backhoes include contextual challenges of programming a car to drive alongside other humans. There are also cyber security challenges to prevent hackable cars, cost challenges to make them accessible to consumers… I could go on, but suffice it to say, there are many challenges.
Ultimately, I think the real backhoe for fully autonomous cars will be regulatory. Even after all of the technical challenges are overcome, I can hear the well-intentioned regulators and legislators acknowledging the potential of fully autonomous cars while still encouraging caution and taking a “wait and see” approach. Yes, there will be early adopter cities and even states, but a self-driving car that only works in San Jose and Michigan on sunny days is not that valuable, and hardly a death blow to Fiat. Chris Urmson, who ran the Google autonomous car project, has estimated it could be as long as 30 years before we have fully autonomous cars. When I synthesize all of the potential challenges to self-driving, I am certain that the impact on sales will not be material in the next five years, which I consider our investment horizon.
Another common argument against owning any of the auto manufacturers is that auto sales in the U.S. have reached “peak SAAR” (Seasonally Adjusted Annual Rate). Clearly, a low margin and high fixed cost business will see earnings decline as volumes decline. Current volumes are 18M per year and The Great Recession saw SAAR fall to as low as 9M units per year. Every monthly sales report is probed for weakness and evidence that peak SAAR has been reached. I think the most likely