서강대학교

초록/요약

The evolution of the labor share, or the wage share, as part of Korea’s national income has become an important issue over recent years. The aggregate labor share in Korea has steadily risen since the mid-1950s, while some authors argue that the labor share has actually declined if self-employed earnings are taken into account. Above all, this controversy suggests that the movements of labor share in Korea since the 1950s do not mirror the long-term stability of the labor share that has appeared in other economies at least until the 1980s as a “stylized fact” of economic growth. Studying the dynamics of labor share at the micro level helps to learn about the relationship between various forces in the economy. Furthermore, estimating the labor share at a firm or establishment or at the plant level mitigates measurement issues that come about when using aggregate data. Meanwhile, studies based on a methodology that assumes perfect competition in product markets are somewhat limited, in that they do not take into consideration the dominating market position of the large firms that dominate East Asian economies. I explore recent issues concerning labor share in the Korean economy, looking at both the theory and the evidence, and then conduct a review of the literature on the relationship between labor share and market power. This paper contributes in three ways to studying the relationship between labor share as measured by payroll over value added and market power as measured by market concentration in the manufacturing sector in Korea. First, I find a negative relationship between labor share and market power in the manufacturing sector with the empirical evidence using the establishment-level data, and this allows us to better understand its macroeconomic implications. A rise in market power since the 1990s may have contributed to the fall in labor share in Korea. Moreover, as an indicator of market power, the HHI measure using sales shows more meaningful results in empirical tests than either concentration ratio measure (CR4, CR20) that uses sales or employment. This is different from Autor et al. (2017) who show that the concentration ratio measure is a more proper measure than the HHI measure when exploring the connectedness between labor share and market concentration in the U.S. Second, for the first time in Korea I construct a long time-series dataset that links several revisions of the Korean Standard Industrial Classification (KSIC) for five-digit KSIC industries and use micro data at the company level to analyze the relationship between labor share and market concentration. Finally, this paper contributes to the macroeconomic literature about labor share by showing measured labor shares within the corporate sector. Focusing on the labor share within corporations helps to address a potential measurement issue related to self-employed, because measures of labor share at the establishment-level are produced by using micro data from surveys of companies that have 10 or more workers Meanwhile, the analysis in this paper has some limitations because it does not cover the self-employed sector that plays a considerable role in the Korean economy, especially in the service sector. The reason is that the Mining and Manufacturing Survey, which is used as the basic dataset for this paper, does not include companies with fewer than 10 employees. Next, the change in labor share not only influences unskilled and skilled labor, but also shifts labor forces from the manufacturing sector to the service sector, which can lead to a change in composition between the manufacturing sector and the service sector, thereby affecting the overall labor share at the national level. Finally, in the empirical analysis to investigate the relationship between the labor share and market power, I considered only the market concentration as an explanatory variable affecting the labor share, and didn’t include several factors, such as the TFP, labor productivity, company size, imports, and so on. One of the main purposes of this paper is to highlight the differences between the Korean economy and the U.S. economy, compared with results of Auto et al. (2017), such as some factors affecting market concentration. Thus, dealing with the application of integrated methods goes beyond the scope of this paper and may be undertaken in future research.