Bill Gurley schools NYU professor: Uber will be 25x bigger than you think
Predicting the future is hard. If it wasn’t, we’d all be venture capitalists and there’d be no money in it. For this reason, it’s not uncommon for two smart people to look at the same company or market and predict two very different outcomes.
About a month ago, I took NYU finance professor Aswath Damodaran to task for thinking small when it comes to Uber. My two main critiques were that Damodaran's analysis underestimated the size of the market opportunity the company is chasing – it’s more than just transporting people – and similarly it underestimated how much of that market the company can win.
Some people accused me of being harsh. If I was harsh, Bill Gurley was just completely and utterly brutal.
Gurley, who is one of the smartest and most articulate figures in all of venture capital (and who just happens to be an early Uber investor and board member) laid out a more than 4,500-word argument for why Damodaran underestimated the Uber opportunity by no less than a factor of 25. The Benchmark Capital partner was diplomatic, apologetic even, but that made his dismantling of the professor’s reasoning no less thorough.
You really should read Gurley’s entire piece because it is a master class in market sizing and the kind of thought exercise that is required to be a good VC – we're talking about handing over millions or occasionally billions in cash to a young company to create something that doesn’t exist yet. Below are a few of the high points of Gurley’s argument.
In choosing to use the historical size of the taxi and limousine market, Damodaran is making an implicit assumption that the future will look quite like the past. In other words, the arrival of a product or service like Uber will have zero impact on the overall market size of the car-for-hire transportation market. There are multiple reasons why this is a flawed assumption.For a similar historical example, he points to a 1980 McKinsey study commissioned by AT&T predicting that there would be 900,000 cell phone subscribers in the US by 2000, a figure that proved to be less than 1 percent of the actual 109 million in reality.
Gurley cites Uber’s falling pick-up times, dual-rating system that drives trust, comparatively high level of safety (recent high-speed chase notwithstanding), and overall convenience as factors that will lead to Uber being used in cases when taxis and black cars previously weren’t. Hence, modeling Uber off the historical car-for-hire market size is already thinking too small. The service, along with competitors like Lyft, are demonstrably increasing that market – Gurley cites an anecdotal statistic that his friends use Uber three-times more often today than they used taxis and black cars two years ago.
He further argues that Uber is replacing or reducing car ownership for many households. With the average annual cost of car ownership in America coming in a whopping $9,000 per year (and rising), the company’s potential addressable market suddenly becomes significantly larger from a dollar perspective. Gurley points out that with an assumption that the average ownership cost is just $6,000 globally, you quickly arrive at a $6 trillion (with a “T”) annual market opportunity for Uber by replacing a fraction of these cars. Damodaran capped the Uber opportunity at just $100 billion, a full 60-times smaller than the opportunity Gurley outlines.
As both Gurley and I pointed out, this even ignores the potential of Uber as a logistics service – or, frankly, addressing any incremental market opportunities beyond simply transporting people. To put it simply, there is no shortage of market opportunity for Uber to chase after. Of course, execution and winning those markets is an entirely different conversation.
Damodaran not only thinks Uber’s addressable market is small, but argues that the company can only win a fraction of that market – 10 percent to be precise. In my earlier post, I argued that Uber’s VCs likely have a number closer to 40 percent in mind. Gurley again makes my point for me, arguing that Uber’s network effects drive higher utility as the company grows, increasing customer utility and helping it capture more market share. He outlines a virtuous cycle in which pick-up times drop, geographic coverage and density increase, and drivers have less down time allowing the company to reduce prices. Combined this leads to happier customer and more demand for Uber.
Turning to more concrete numbers – without revealing any confidential company data – Gurley notes that by October 2012, Uber had already on-boarded more black cars in San Francisco (this was pre-UberX) than existed in the city when it launched two years prior. Moreover, the service was still growing at 20 percent per month while having already grown beyond 100 percent of the historical black-car market size, at least from a supply perspective.
By the same token, Uber CEO Travis Kalanick revealed in a June WSJ interview that the company’s business in San Francisco is currently “a very healthy multiple bigger than” the $120 million spent in total on taxis and limos in the city in 2009. This means that, at least in San Francisco, Uber has greater than 100 percent share of the “historical market demand” as well. Add in the $22 billion annual cost of car ownership in the city and you get a sense of where the ceiling truly lies.
As Gurley states, “Damodaran’s math simply does not hold up. This cannot be yesterday’s market.”
As I wrote a few weeks ago, when Uber raised at an eye-popping $17 billion valuation:
Uber’s latest round is a bet by its investors that the company will continue its rocketship growth, taking an ever-increasing share of the existing global taxi and livery market, while also growing that market well beyond its current size.In trying to offer a more realistic estimate of the Uber market opportunity, Gurley assumes that it expands the historical car-for-hire opportunity by 3-6X and wins between 2.5 and 12.5 percent of the “cost of car ownership market.” Combined, this gives Uber a total addressable market opportunity of between $450 billion and $1.35 trillion per year, or respectively 4.5 to 13.5 times larger that the $100 billion limit that Damodaran’s based on the historical car-for-hire market. The professor similarly limited Uber’s market share to just a 10 percent of that market. Gurley believes the company can own several time more.
Gurley acknowledges that he may be biased in his optimism toward Uber as an investor in the company. But, as he points out, Damodaran may be subject to his own biases. In an earlier version of his blog post the professor admits to never having used Uber:
I have to confess that I just downloaded the app and have not used it yet. I spend most of my of life either in the suburbs, where I can go for days without seeing a taxi, or in New York City, where I find that the subways are a vastly more time-efficient, cheaper and often safer mode of transportation than taxis.No one knows what the future will look like. The difference between Gurley (along with Kalanick) and Damodaran is that one is trying to create that future and expand its boundaries, while the other is busy arguing for its limitations.
Updated, 5:54pm PT at 7/12/14: Read professor Damodaran's thoughtful response.