In a recent blog, I contended that: (i) Better Place (BP) had an essential flaw in its approach because it underestimated Li-ion battery technological development; and (ii) better batteries would make costly battery-swap stations such as the ones envisioned by BP both meaningless and out of focus.
Subsequently, in a series of replies to comments on my piece both on EVWorld.com and LinkedIn.com, I argued that: (iii) Elon Musk never bought the swap idea betting on supercharging instead; and (iv) vertical integration gave Tesla (TSLA) a major advantage over its competitors, permitting it to learn by doing.
By the time I was writing my blog, I had no knowledge of Musk's swap idea. This gave rise to some misleading comments on my part. After Tesla's demonstration of its battery exchange scheme last week, I need to make the following clarifications and complementary remarks:
First, BP's approach was flawed not only because it underestimated Li-ion battery technological development but also because it originated as a commercial idea, not a production one. It has been known that Shai Agassi's swap system was based on a previous concept developed by Musk a few years ago. What appears to be astonishing is that both the Roadster and Model S may already have the swap capability incorporated.
This means that in its original production design Tesla included two key factors, both of which aimed at extending the range of its vehicles: One, a supercharging mechanism that implied development of better batteries with this kind of capability and two, a swap system that was indeed a sort of ace under Musk's sleeve.
By contrast, Agassi's approach rested essentially on the notion that EV owners shouldn't own the battery whose rental would constitute BP's main source of income and profit. The question then was whether owning an EV without a battery made any sense. And, as it turned out, it didn't, since it at most meant an eternal dependence on BP, something not too many car owners were willing to take. In this connection, it seemed much more reasonable to devise a complete leasing scheme (of EVs, including their corresponding batteries), which is in fact now being offered by a number of auto makers at the marketplace.
Another point worth mentioning in relation to BP's system is that the range of the electric car on which it relied (100 miles) was much more limited than that of Model S (250 miles). Here potential car owners got the impression that the cars they were partially acquiring were only a second-best solution to the oil problem.
Lastly, because supercharging was never part of Agassi's idea, BP couldn't offer it for free while charging a fee for swapping, like in the Tesla approach. The possibility of choice in Tesla's system is definitely an appealing way to attract customers, something missing in BP's. Nevertheless, supercharging will eventually go wireless and, to the extent that this revolutionary mechanism evolves and becomes widespread, it will most likely start displacing swapping as well. Note that wireless charging is already available for the Leaf and Volt cars.
Second, I maintain my point that better batteries would make $500,000 swap stations both meaningless and out of focus, but of course, we will have to wait a while to see real breakthroughs in the advanced energy storage sector. In the meantime, Tesla's swap mechanism is likely to succeed provided its business plan ensures an adequate return on investment in the next 8-10 years, the approximate time required for commercialization of different kinds of metal-air batteries that will also make supercharging, wireless charging and swapping (altogether) obsolete. In this context, Tesla should target its swap scheme toward potential geographical areas in the US, which have been seen resistant to change due to drivers' range anxiety.
Third, in my presentation for the Graphite Conference held in December 2011 in London, I argued that to understand the then difficult financial situation of a number of battery makers, we needed to look at the Li-ion battery value chain.
Following Te Velde (2006), the value chain approach not only involves the process-interlinked material and information flows as well as their spatial distribution, but also the relationships among chain actors insofar as cooperation, control and power. There are lead firms and lead agents that exert a leading and coordinating position within the chain and are able to set parameters in the entire chain. They are capable of controlling and coordinating other parts of the value chain.
Two types of lead firms can be identified: Those controlling production and processing processes (producer-driven); and (ii) those controlling knowledge -intensive marketing processes (buyer-driven). Electric vehicle makers were seen to exert both types of control over the rest of the value chain (See Figure 1). Finally, lead agents are an important driving force of methodological developments to influence production and consumption systems. I didn't refer to lead agents in my London presentation.
 See: Te Velde Dirk Willem et al (2006), "Entrepreneurship in value chains of non-timber forest products", Forest Policy and Economics, Volume 8, 725-741.
I then went on to indicate that many Li-ion battery producers were performing badly in the market because they didn't succeed in making important agreements with auto makers to produce Li-ion batteries for electric cars. In this connection, I foresaw strong trends toward vertical integration (i.e., a strategy whereby a firm gains control over its suppliers or distributors in order to increase the firm's power in the marketplace, reduce transaction costs and secure supplies or distribution channels) in the electric car industry and horizontal integration (i.e., a strategy whereby a firm usually acquires or mergers with a competitor in a same industry) in the Li-ion battery industry.
Tesla is one of the most vertically integrated companies in the electric car market today. This can be thought of as a major advantage of Tesla over its competitors. Instead of exerting control and power over other firms (e.g., Li-ion battery ones) of the rest of the value chain, it opted for integrating them to its major line of production. Regarding Li-ion batteries, even though it doesn't manufacture the cells, it has to do a great deal with their design not to mention their packaging and assembling, the last two tasks being performed by Tesla in their entirety. Among other things, vertical integration has allowed Tesla to learn by doing along the value chain. The relationship between vertical integration and learning by doing has been analyzed by Sorenson (2003). In that article, it is argued that in a volatile environment characterized by technological uncertainty and market instability, firms may benefit the most from learning by doing by vertically integrating.
The argument goes as follows. In general, in a highly volatile environment, companies cannot benefit effectively from learning by doing because by the time they acquire the knowledge, it no longer applies. Integrated companies are less affected by volatility because of the control the integrated firm has over its environment. In this connection, buffering can simplify the learning process by stabilizing the relationship between inputs and outputs. Hence integrated firms have the ability to direct the actions of internal suppliers. By increasing the predictability of future development paths and component supplies, vertical integration reduces exposure to volatility. Of course, vertical integration may not be the best way to go in a highly competitive world that requires continual innovation unless the company involved invests heavily in research and development.
And this happens to be the case with Tesla holding a number of patents in fields ranging from the drive train of electric cars to different kinds of batteries, including metal-air ones, and swapping. In fact, as another Seeking Alpha contributor has argued, the swap device may have been derived from the very line of production of Tesla's manufacturing plant, something not possible in BP's system due to its absolute lack of control over the production process of electric cars. So vertical integration, together with learning by doing and investment in research and development may be crucial for the success of Tesla's swap scheme. But again, this may be a short-lived approach, one that depends on how fast the "next best thing" in advanced energy storage arrives.
 See: Sorenson Olav (2003), "Independence and Adaptability: Organizational Learning and the Long-Term Effect of Integration", Management Science, Volume 49, No. 4, April.