Monopsony in the Japanese Labor Market

Minimum wage and wage bargaining as policy measures

One of the most important agenda for the Japanese economy is to raise wages which has remained stagnant in the recent years. To achieve the goal, policy measures have been proposed, which includes measures such as raising the minimum wage by the government, and encouraging wage bargaining between the employers and the labor unions in the spring wage round.

Minimum wage had seen a limited increase for a long time. However, since the government announce its intension in 2017 to achieve an minimum hourly wage of 1000 yen, an increase of more than 20 yen a year has been achieved with the exception of FY2020 when the COVID-19 hit the economy. In FY2022, minimum wage was raised by 31 yen, the largest increase on record.

As for the outcome of the spring wage round, increase in wages (which includes increase in wages due to higher seniority) had been below 2 percent since FY2002. However, since FY2014, when the government asked the employers and the labor union to raise wages, wages increased by more than 2 percent, with the exception of FY2021 due to the pandemic. In FY2022, wages increased by 2.20 percent (according to the Ministry of Health, Labour and Welfare).

Textbooks says these measures have negative consequences to the economy

These measures are now widely accepted as something natural for the government to implement. However, it is not so evident if we look at them through the eyes of labor economics textbooks.

If we assume that the labor market is under perfect competition, any minimum wage that is set equal to or below the equilibrium wage (determined at the intersection of the labor supply and labor demand curves) will not be effective. If the minimum wage is set above the minimum wage, labor supply will exceed labor demand, and unemployment will emerge.

It should be noted that, for such a conclusion to be true, the assumption that the labor market is under perfect competition need to hold: there should be sufficient number of both the employers and the workers. If the assumption does not hold, the textbook understanding of minimum wage will need to be altered.

Labor union is an example of a divergence from perfect competition. If a labor union is representing all the workers in the labor market (a monopoly union), it will aim to maximize the interest of the union members. As a result, labor unions will use the bargaining power to raise the wages to above equilibrium wage. At the same time, however, it will lead to lower employment than the equilibrium level. It could be understood as rewarding the “insiders” (members of the labor unions) at the cost of “outsiders” (unemployed).

If we could assume that workers can be represented by a single labor union, we should also be able to assume a situation where there is only one firm that demands labor (“the only employer in town”). This the situation which is called “monopsony”. If the labor market is under monopsony, textbook understanding of the labor market will need to be changed. That is because, a monopsony firm will make use of the bargaining power to affect wage and employment so that it could maximize its profits.

Labor market textbooks also pick up the topic of monopsony. However, it is picked up as an unusual situation where, for example, there is only one firm in some remote town so that whoever wants to work needs to find a job in the only employer in town. In such a situation, the employer will pay a wage that is lower than the equilibrium wage and will employ less workers that in the equilibrium.

When there is monopsony, minimum wage and wage bargaining by the labor unions will have a different implication than that in textbooks. In a monopsony, minimum wage and wage bargaining by the labor unions will be able to raise wages and increase employment at the same time. Ideally, it may even be able to realize a wage that is equivalent to equilibrium wage and an employment that is equal to the equilibrium level of employment. If that kind of a situation realizes, there will not be any unemployment either.

Studies of US and UK confirm the existence on monopoly in labor market

In the recent years, there is a growing interest in economist to find out whether monopsony exists, not only in some special circumstances, but also in wider areas. In doing so, the concept of monopsony is usually extended so that it applies to a situation where there is a limited number of firms (not necessarily only one) so that the concentration of labor demand is high. The purpose of the studies is to see whether the high concentration of labor demand is affecting wages and employment.

Theory of monopsony has been developed by Manning (2003). Empirical analysis has been done by Benmelech, Bergman, and Kim (2018) and Azar, Marinescu, and Steinbaum (2019) on US, and by Abel, Tenreyro, and Thwaites (2018) on UK. They tend to confirm the influence of monopsony on wage in wide range of areas and industries.

Can monopsony be confirmed in Japan as well?

Empirical analysis on Japan is still very limited. Izumi, Kodama, and Kwon (2020) , for example, confirm that when concentration of labor demand is high, wages tend to be low.

In the following, a simple analysis will be made to see whether there are signs of monopsony having any influence on wages in Japan.

a. HHIs of employment in prefectures

In order to see the degree of concentration of labor demand, Figure 1 shows the Herfindahl-Hirschman Index (HHI) of labor demand in each of the prefectures using the data from the 2014 Economic Census for Business Frame.

Figure 1 offers following observations.

The first observation is that HHI is relatively low. It ranges from 5 to 173 and an average of 27. They are much lower than 1500 which is the criteria used by the Fair Trade Commission when judging whether to allow merger to take place or not.

One of the reasons for the low HHI is the difference in the aggregate level of regions and industries. In calculating HHIs for Figure 1, data for all workers (excluding temporary workers) of all firms in all industries in a prefecture have been used. As a result, number of firms tends to be large and the share of regular workers for each of the firms tends to be small. In the previous studies, HHI has been calculated for firms in a certain industry in a limited area of land, such as commuting zones. Therefore, their HHI have been much higher that what is shown in Figure 1.

The second observation is that there is a large dispersion of HHIs among the prefectures. There is a tendency for HHIs in metropolitan areas such as Tokyo and Osaka to be low (low concentration) whereas those for prefectures such as Yamaguchi or Shiga tend to be high (high concentration). It seems to be a reasonable result judged from the differences in the agglomeration of firms and workers in these prefectures.

However, there are exceptions such as Hokkaido which has a low HHI. It should be a result of summing up high HHI regions with low HHI regions which leads to relatively lower HHI for the prefecture as a whole.

b. HHIs and wages

In Figure 1, average compensation of employees per employee is plotted with HHIs for each of the prefectures. There does not seem to be any clear relationship between the two. However, it may be because of the Japanese employment system which has a feature of a lifetime employments system. Under such an employment system, once a worker joins a firm, mobility is low so that the workers is expected to stay with the firm until retirement. It means that once the worker joins a firm, there is essentially no competition from other firms to hire the worker. If the objective is to see whether there is an influence of monopsony in wages, therefore, we should look at something different from an average wage.

From that point of view, what seems to be more appropriate is the wage you receive in the first year of employment (starting wage). Even in Japan, there is a fierce competition for the new graduates that seek employment every spring.

However, it should be noted that the starting wage which a freshman receives is also affected by the difference in productivity of the firms (which in turn may reflect the different industries they are operating in), the difference in price level, and the difference in business conditions. Therefore, in order to look for the influence of the difference in concentration of labor demand on wages, these factors need to be controlled.

c. HHI and staring wages

In order to control for these factors, using the cross section data for prefectures in 2014, starting wage of male and female high-school graduates was regressed by log HHI, log CPI, share of the manufacturing industries in gross prefectural products, share of non-regular workers, and unemployment rate. The coefficient for log HHI is negative and significant at a significance level of 5 percent. It suggests that monopsony is affecting staring wages negatively in the case of male and female high school graduates.

In order to see it visually, Figure 2 shows the relationship between starting wages and log HHI after controlling for other factors. A negative correlation can be seen between the teo variables.

The importance of monopsony in understanding wages

What we have seen above suggest that there seems to be a monopsony in the Japanese labor market and that it is influencing wages. There have been a number of empirical analysis that indicate that minimum wages do not seem to have affected employment in Japan as well as in other countries. Such a finding could be explained by the influence of monopsony.

Whether there is monopsony or not has an important implication for designing of economic policies. Minimum wages and wage bargaining by labor unions would have an important role to play if there is a monopsony. They would be important not only for its macroeconomic aspect but also for their role in fighting inequality.

The empirical analysis on monopsony in Japanese labor market is still very limited. Much more work needs to be done.

【References】

Abel, Will, Silvana Tenreryo, and Gregory Thwaites (2018), “Monopsony in the UK”, CFM Discussion Paper Series 2018-27, Center for Macroeconomics, London School of Economics and Political Science.
Azar, Jose, Ioana Marinescu, and Marshall I. Steinbaum (2019), “Labor Market Concentration”, NBER Working Paper 24147, National Bureau of Economic Research.
Benmelech, Efraim, Nittai Bergman, and Hyunseob Kim (2018), “Strong Employer and Weak Employees: How Does Employer Concentration Affect Wages?” NBER Working Paper 24307, National Bureau of Economic Research.
Izumi, Atsuko, Naomi Kodama, and Hyeog Ug Kwon (2020), “Labor Market Concentration on Wage, Employment, and Exit of Plants: Empirical Evidence with Minimum Wage Hike”, CPRC Discussion Paper Series, Competition Policy Research Center, Japan Fair Trade Commission.
Manning, Alan (2003), Monopsony in Motion, Princeton University Press.