한국생산성학회

This study empirically analyzes the effects of the use of non-regular workers and indirect workers on the financial performance of companies in South Korea. The study uses data from the Workplace Panel Survey which consists of representative workplaces in South Korea. More specifically, this study extends the time series data to the most recent 2017 WPS wave, and uses function estimation methodology of Olley and Pakes(1996) to control endogeniety which usually occurs in the analysis of business productivity. Through the estimation process, this study is expected to contribute to providing the theoretical and empirical foundations and policy implications for setting the policy basis for the rigidity of the labor market and the level of employment protection. We found that the corporate performance from using non-regular workers differs according to the size of the employees, and the corporate productivity measured by added value per person has a linear negative relationship with the proportion of non-regular workers and indirect employment, but also a non-linear inverse U-shaped relationship in large companies. The result shows its significance in both analysis targeting all industries and analysis limited to manufacturing industries. It implies that the added value per person increases up to a certain proportion but decreases after the critical point when companies utilize non-regular and indirect employment. This leads to inference that the effect of corporate productivity of the use of non-regular workers can be expressed differently according to the distribution of the proportion of the use of non-regular workers. However, the result did not appear to be significant in terms of per capita operating income and sales, and it did not have a significant effect on businesses with less than 300 employees. It is likely to be due to the difference in investment methods for productivity improvement between small and large-scale businesses. Other than the use of non-regular and indirect employment, education and training expenditure per person, research expenditure, overseas market ratio also seem to affect corporate productivity measured by revenue, operating profit, and added value per employee. We found the education/training expenditure per person and the size of the company increased sales and operating profit per person throughout the analysis in general, but it did not have a significant relationship with added value per person. This study focuses on the corporate performance of non-regular and indirect employment, but as shown in the analysis, the financial performance of a company is influenced by various factors in addition to the difference in the proportion of employment types. The sales per person appears to be affected not only by the ratio of the use of non-regular workers, but also total number of employees, union status, training costs per person, and subcontracting transactions. In addition, the proportion of non-regular and indirect employment utilization appears to have different effects depending on the size of the employees, suggesting that the proportion of non-regular workers in companies should not be considered as a single factor for estimating or forecasting corporate performance. In addition, considering that variables such as entrepreneurship and organizational culture that cannot be observed in the survey data can also have a significant impact on corporate performance, it needs to be careful to put excessive weight only on employment decisions in improving corporate financial performance and productivity. In determining the level of employment protection and legislation, the government would be better not to focus on the use of non-regular workers, i.e., at what level to determine the level of quantitative use. Efforts, however, should be made to bring more balanced labor flexibility in terms of employment quality, such as the working conditions and environment of workers. The limitations of this study are as follows. The Workplace Panel Survey used in this study consists of companies with 30 or more employees as a population, and companies with less than 30 employees are excluded from the sample. In future studies, it is necessary to conduct an analysis through data reflecting more small businesses. Therefore, it is possible that the proportion of non-regular workers shown in the analysis data was observed to be lower than in reality, and this may cause bias in estimating the effect of using non-regular and indirect workers on corporate productivity estimated in this study. In addition, the Workplace Panel Survey includes indefinite contract workers in the composition of non-regular workers, and the items of indefinite contract workers are included relatively recent from the 2015 survey, resulting in a time series break in the survey. Considering the high proportion of indefinite contract workers among non-regular workers, it is possible that estimation bias may appear in the period prior to 2015 when indefinite contract workers were not included in the calculation of the proportion of non-regular workers. However, this is a problem that arises from the limitation of survey data, thus it is expected that more meaningful analysis results can be derived if research data that can supplement the limitations are secured in the following studies.