This thesis studies the automotive market and the competition between the gasoline-fueled internal combustion engine car and a fuel cell-powered hydrogen car. The automotive market exhibits network effects, it gets more and more attractive to drive a certain type of car, the more consumers are driving the same type, for instance a gasoline car. The reason is that when more consumers are driving the same type of car, it gets more attractive for fuel and service stations to become compatible with that type of car. I.e. the more hydrogen cars there are on the road, the more fuel stations will invest to serve hydrogen cars. This increased number of fuel and service stations feeds back in to further increase in demand.
The fact that demand depends on the network size means that the incumbent gasoline car has an advantage over the emerging hydrogen car through its installed base of consumers. Even if it would be the case that all consumers prefer to drive a hydrogen car, and that the increase in consumer surplus would more than compensate the loss to gasoline stations who would go out of business, the transition might not occur. Each consumer would not be willing to switch before he knows that there will be fuel stations to support him, and fuel station owners would not invest to support the hydrogen car before they know that there are hydrogen cars to service. We might be locked in to the gasoline car.
The thesis first surveys the literature on network effects and defines the central concepts needed to conduct the analysis. The project is an extension of a previous work undertaken at Statistics Norway (Greaker and Heggedal, 2010), and their model is summarized. Then I construct my baseline model in which half of the consumers buy a new car each round of the game. Each car lasts for two rounds, therefore the game has an overlapping generations-structure. The game lasts for three periods, and hydrogen stations may enter the model in any of the first two periods.
A first finding is that for a broad range of parameter values there will only be hydrogen station entry in the first period. Period two-entry can only take place if the market grows substantially between periods while the fuel stations place a heavy discount factor on future earnings. Relying on this result, hydrogen station entry is therefore assumed to only take place in period one.
With entry in only one period, I find that sequential adoption decisions by consumers does not affect the likelihood of lock-in. The set of equilibria is identical to a one-shot game. What will affect the likelihood of lock-in, however, is the exit of gasoline stations if hydrogen proves popular, learning by doing that improves the hydrogen car technology relative to the gasoline technology, and more impatient consumers. All these factors will increase the likelihood of hydrogen cars making it past a critical mass of market share beyond which the market tips in favor of hydrogen.
I also consider a sponsor on the hydrogen side of the market. A sponsor is a single license holder who therefore stands to gain if the hydrogen technology diffuses in the market. A sponsor who can monopolize the hydrogen fuel stations will do so, and greatly limit the number of stations. A sponsor who cannot monopolize the sector may construct some stations himself and from there he can rely on the market to work in his advantage, private entrepreneurs will build the remaining stations and the hydrogen technology will be widely adopted.
The broad adoption of the hydrogen car will impose substantial losses on the oil producers. According to the U.S. Energy Information Administration, transportation uses 2/3 of all oil consumption in the U.S. (source: U.S. Energy Information Administration). It is unclear how this sector will respond to the emerging hydrogen car, but the literature on the green paradox suggests that they will not sit idle, we can at least expect oil prices to decline.