There has been increasing interest in industrial policies in many countries around the world in the last couple of years (Note 1). Though various countries have introduced industrial policies for different periods of time, Japan's postwar industrial policy in particular has attracted a fair amount of world attention. The reason behind this is the view that the government's policy interventions contributed to Japan's economic growth during the twenty years after the war. Against this backdrop, many countries including the United States have followed the examples of Japan's industrial policy by conducting policy interventions that focus on particular industries (Note 2).
Along with the increasing interest in industrial policies, a number of empirical studies have examined the effects of Japan's industrial policy. In this column, I particularly want to focus on the influence of Japan's industrial policy on productivity and discuss the quantitative effects of this based on the latest joint study results with Professor Tetsuji Okazaki of the Graduate School of Economics, the University of Tokyo (Kiyota and Okazaki, 2010) (Note 3).
Industrial Policy and Productivity
Before moving on to the discussion, I would like to explain the meanings of two keywords used in this column. The first keyword is "industrial policy" and the second keyword is "productivity." The content of different industrial policies varies greatly. In this column, I will refer to "policy interventions focusing on particular industries," whether they are growth industries or declining industries, as "industrial policies."
Meanwhile, with regard to productivity, I will focus on productivity indicators such as labor productivity and total factor productivity (TFP). Labor productivity is widely used as a productivity indicator that measures how much output is produced per unit of labor input. Although labor productivity has the advantages that it is easy to use and easy to understand, it also has the disadvantage that it cannot take into account the influence of inputs other than the labor input. For example, output can be increased even without any change in the labor input and labor efficiency if capital input increased. As a result, labor productivity can be increased. However, this increase in output is simply the result of an increase in capital input and it does not necessarily mean that the production efficiency has been increased. TFP is an indicator that overcomes this kind of problem.
TFP is an indicator that measures how much output is produced per unit of combined production inputs including labor input (Note 4). I would like to take into account in my discussion not only labor productivity but also TFP.
Effects of Industrial Policy
Examples of quantitative analyses of Japan's industrial policy include Beason and Weinstein (1996), Okazaki and Korenaga (1999), Kiyota and Okazaki (2005), and others. Key results identified through these studies can be summarized in the following two points. First, industrial policy contributed to an increase in labor productivity. Second, industrial policy did not necessarily contribute to an increase in TFP. These results reflect the following mechanism: industrial policy encouraged capital accumulation through an expansion in investment, contributing to an increase in output which led to an increase in labor productivity. However, the increase in output was simply accompanied by an increased capital input and therefore did not lead to an increase in TFP.
If Japan's industrial policy did not contribute to an increase in TFP, the next important question would be: "is the reason that TFP did not grow because the policy did not work?" or "is the reason that TFP did not grow because the policy did work?" If TFP did not grow because the policy did not work, the reason should be found in the policy implementation. However, if TFP did not grow because the policy did work, the problem must lie in the design of the policy itself.
To answer this question, Kiyota and Okazaki (2010) focused their attention on Japan's cotton spinning industry and analyzed the effects of industrial policy quantitatively. In the 1950s, the top 10 leading cotton spinning firms accounted for almost half of the entire market share in the cotton spinning industry. In the cotton spinning industry in the 1950s, both imports from other countries and domestic competition were restricted. However, the industry faced international competition with trade liberalization in the 1960s. Prior to this trade liberalization, there was policy intervention in the cotton spinning industry. The primary purpose of the intervention was stabilization of the supply of cotton yarn. Kiyota and Okazaki (2010) conducted an analysis on how this policy intervention in the cotton spinning industry affected the TFP of cotton spinning firms.
The analysis showed that industrial policy worked effectively in controlling the output of spinning companies. The market share of the top 10 cotton spinning firms remained steady over nearly 20 years. However, control of the output led to a decrease in the overall TFP of the cotton spinning industry. This was attributable to a decrease in the TFP of the top 10 cotton spinning companies. The overall TFP of the spinning industry decreased because the market share of the top 10 cotton spinning firms was maintained despite the decrease in their TFP. As a result, the cotton spinning industry, which had continued to support the Japanese economy since the beginning of the 20th century, fell into decline in the mid-1960s.
The results of this study suggest that a policy focused on particular industries could produce accompanying side effects. The industrial policy for the cotton spinning industry contributed to market stabilization by maintaining the market share of large companies. However, the policy restrained the reallocation of resources from firms with low production efficiencies to companies with high production efficiencies, which deprived small and medium-sized firms with high production efficiencies of growth opportunities. As a result, the entire industry was forced to pay the price of a sluggish TFP growth rate.
Policy Implications
A lesson from the study by Kiyota and Okazaki (2010) would be that industrial policy cuts two ways. Policy interventions focusing on particular industries could be effective in achieving a specific objective. However, achieving that objective can potentially interfere with the reallocation of resources within the industry, and as a result interfere with the medium to long-term growth of productivity for the whole industry.
In addition, this lesson can be applied not only to the reallocation of resources within the industry but also to the reallocation of resources among industries. Though policy interventions contribute to an increase in TFP in a particular industry, they may interfere with the reallocation of resources from inefficient industries to efficient industries. In such a case, if the positive effects of TFP growth do not exceed the negative effects resulting from the misallocation of resources, we have to conclude that policy interventions have negative effects on the economy as a whole (Note 5). If we consider economic growth as the foundation for policy targets, we will need to design systems after taking a panoramic view of the entire economy to ensure that policy interventions will neither interfere with the reallocation of resources within the industry nor among industries.
