The paper investigates four different factors that might contribute to the regional industry agglomeration, which is a situation where firms locate near each other in a specific region. The four factors are: intra-industry spillovers(positive intra-industrial externality), related industry spillovers, spinout effects, and location fixed effects.
Positive intra-industrial externality is about similar firms gather together in order to learn from each other’s spillovers. In class we also discussed about the economies of scales, which is a concept that if output(internal or external) is large enough, the cost can be reduced. In this case, we can apprehend as companies might gather together and share capitals(machines to produce specific parts) in order to reduce their cost.
Related industry spillover is a different factor which is mainly about how related industry plays an important role in the regional agglomeration. The authors used an example of C&W(carriage and wagon) industry for it predates the automobile industry and a lot of automobile companies were founded by C&W employees such as William Durant.
In order to access these two factors and their relationship with regional agglomeration, the authors conducted several sets of regression. Specifically they look at how the two factors affect automobile companies’ entry in the market. Through a series of variable collecting and data analysis, they conclude that intra-industry spillovers hardly affect the agglomeration while related industry spillovers play a significant role in it.
A spinout(spinoff) is a company created through an existing company. It might contribute to regional agglomeration because, as we discussed in class, founders tend to stay at where they created the companies due to several reasons.The authors believe that this factor is only significant in the later stage of industry development. The authors approach this factor through logit models of spinout entry(Table 5) and conclude from the regression that older firms tend to create spinouts, bigger firms tend to create spinouts, and spinouts tend to stay at their parents’ location.
Location fixed factor is the most obvious one out of the four. Firms gather in a specific location for its geographical advantages, which include variables such as resources and transportation. The authors recognize there exists some location fixed effects in explaining the agglomeration. Unfortunately this factor is not discussed in length in the paper.
Although these factors certainly applied to agglomeration of auto firms in the early days of the industry, I wonder if these factors are all still applicable. Japanese car companies, most of which originally established their North American headquarters in California, are now leaving California and headquartering in locations such as Plano, Texas (Toyota) and Nashville, TN (Nissan). If the positive intra-industrial externality factor were still in effect, wouldn’t these companies have considered relocating to Detroit?
Moody’s followup is good: Toyota’s engineering operation is in Detroit (or rather Ann Arbor) while Honda is in northern Ohio, both in easy driving distance from Detroit. Nissan is also in the lower midwest for its engineering. VW and the others all have Detroit offices. So for engineering, yes, intra-industry effects remain. For production logistics matter – as we heard at Metalsa, even if wages are lower, moving material around is costly so being near can offset such putative cost advantages.
At the company level, what functions benefit from proximity? How about suppliers? [see work by Thomas Klier on that, with Jim Rubenstein and others].