노동조합의 고용효과

This study conducts empirical panel data analyses on whether there are negative impacts of labor unions on employment growth in Korea, and if so how much. The second purpose of this study is to examine and solve some methodological issues of endogeneity problem and limitations of within estimators in estimating the impacts of labor unions on employment growth. To solve the methodological problems, this study estimates random effects models of panel data with instrumental variables and dynamic panel data analyses, and also attempts to replace labor union dummy with the age of labor unions as the main explanatory variable in some dynamic panel models. The balanced panel data with 5 rounds(2005-2013, biannually) of Workplace Panel Survey is used.
The results of this study indicate that there are negative impacts of labor unions on employment growth in Korea. The negative impacts, significant at least at α=0.10 with one-tailed test, are found in all the models of fixed models, random effects model, random effects model with instrumental variables, and dynamic panel models in this study. The results indicate that the impacts of labor union on employment growth are distributed between 1.3~2.3% per annum. However, the impact of union age on employment growth is not statistically significant with p＝0.23, indicating that it is still possible to affect employment growth weakly. This study also finds that it is very important to control for the endogeneity problem appropriately, and that within estimators have limitation due to the small number of firms with union status changes. Finally the empirical results and the findings in methodological aspects are summarized, and their implications are discussed.