اثر ساختار بازار بر استفاده از فن آوری در حال تحول

We study the speed at which technologies are adopted depending on how the market power is shared between the firms that sell technologies and the firms that buy them. Our results suggest that, because of a double marginalization problem, adoption is fastest when either sellers or buyers hold all the market power. Thus, competition between sides of the market may delay the adoption of technologies.

The adoption of new technologies is regarded as one of the main contributors to economic growth (see, for instance, Lucas, 1993 and Barro and Sala-i-Martin, 1995). The time at which a new technology is adopted has a significant impact on the growth rate of a country or a firm. Adopting new technologies too quickly may be disadvantageous given the sunk cost the establishment of a new technology carries. On the other hand, delaying the adoption of a new technology can lead to high opportunity costs or to a disadvantageous position with respect to competitors. This trade-off has been widely studied in the literature. Most of the literature so far has focused on the pace of adoption of new technologies from the perspective of a firm which by adopting a new technology incurs in a fixed cost that is compensated over time by the increase in productivity. In this respect, the problem of adopting new technologies was reduced to two basic settings: an optimal stopping problem (Farzin et al., 1998 and Jovanovic and Nyarko, 1996) and a competition game between firms buying technologies (Götz, 1999 and Chamley and Gale, 1994). Two articles, Stoneman and Ireland (1983) and Ireland and Stoneman (1986), modify this approach and take into account the interaction between firms adopting new technologies, buyers, and the firms that sell these technologies, sellers. Stoneman and Ireland (1983) try to replicate the fact that the penetration of a new technology usually follows a sigmoid (S-shaped) path. On the other hand, Ireland and Stoneman (1986) focus on the role of expectations (rational and bounded rational) in the adoption of new technologies. We follow Stoneman and Ireland (1983) and Ireland and Stoneman (1986) in that we consider the interaction between sellers and buyers but focus instead on how the different market structures, i.e. how the market power is shared between supply and demand, affect the timing of the adoption of new technologies. In the model we present sellers have a trade-off between price and time: setting up a higher price means that the income from selling the technology is higher. However, it also means that buyers wait more before adopting a new technology as technologies are more expensive. On the other hand, buyers have a trade-off between early adoption, which implies an earlier increase in productivity, and late adoption, which implies a greater (but later) increase in productivity since a newer technology is adopted. We explain how these trade-offs are solved when we consider three different market structures that are distinct in how the market power is shared among the buyers and the sellers. In the first market structure we consider there is only one firm selling technologies and many firms willing to buy technologies. Hence, in this setting, the supply side holds all the market power and buyers act as price takers.In the second market setting considered in this paper, there are many firms supplying technologies and only one firm interested in buying. In this setting the demand side holds all the market power and, therefore, sellers compete in prices and make profits equal to their outside option of not participating in the market. In the last market setting we consider there is one firm on each side of the market. In this last setting sellers and buyers compete for the surplus in the economy.1 We find that if there is competition between sellers and buyers then, because of a double marginalization problem, the adoption of new technologies occurs at a slower pace than when either sellers or buyers hold all the market power. This suggests that competition between both sides of the market, instead of competition within each side, can delay the adoption of new technologies. We round off our results by introducing some comparative statics. The present paper tries to shed light on the speed of adoption of new technologies with respect to different market structures. Rather than focusing on the nature of the technology itself or other factors, we choose to study how market power can explain the differences in speed of adoption. We do not claim market power is the only reason why we might observe different behaviors. However, as we shall show, it is a factor that can explain these differences by itself. From the theoretical point of view, many models study the optimal timing of technology adoption. Jovanovic and Nyarko (1996) present a model where the decision maker increases productivity by either learning by doing or by switching to a better technology. Karp and Lee (2001) extend this model by introducing discount factors. Farzin et al. (1998) introduce a setting where the increase in productivity caused by the adoption of the newest technology is known only in expected terms. Adopting a new technology has a sunk cost that is independent of the productivity level of the new technology. In an article by Doraszelski (2004) a distinction between technological breakthroughs and engineering refinements is introduced. Other papers that consider the adoption of new technologies from the point of view of the agents adopting the technologies (buyers) are Chamley and Gale (1994) and Götz (1999). From the empirical perspective, there is no doubt that the timing of technology adoption has been a concern. Hoppe (2002) presents a literature review on this topic. To cite some, Karshenas and Stoneman (1993) present a study on the diffusion of CNC (computer numerically controlled machine tools) in the UK engineering industry. Factors determining the delaying in the adoption of the new technology were found to be, among others, the learning effects and the cost of the new technology. Weiss (1994) studied the adoption of a new process technology called the surface-mount technology by printed circuit board manufacturers. The rest of the paper is organized as follows. In Section 2 we introduce the model. Section 3 presents our findings for the three different market structures considered. In Section 4 we present a comparative statics analysis. Finally, Section 5 concludes.

In this paper we investigated how different market structures affect the speed at which new technologies are adopted. A game between the demand side, firms buying technology, and the supply side, firms selling technology, was presented. Three different market scenarios were considered, one in which the supply holds all the market power, another in which the demand holds all the market power, and a third setting where there is competition between both supply and demand. In our results, we explained how these three different market structures affect the adoption of technologies. The speed of adoption when one side of the market holds all the market power is the same independently of which side holds the power. However, when no side of the economy has all the market power, i.e. competition between supply and demand, then the adoption occurs at a slower pace. This suggests that competition between the two sides of the market might decrease the speed of adoption and that competition within each side might increase the speed of adoption. The literature so far has focused mostly on the optimal timing of adoption of new technologies from the perspective of a firm that faces an exogenous process of technological change where the price of new technologies is also exogenous. To our knowledge, this is the first paper to consider how the differences in market structure can affect the speed of adoption of new technologies. Empirical studies often attributed the differences in the speed of adoption of technologies to the nature of the technology itself. We show that the market structure could also account for these differences in speeds of adoption.