The Effect of Employment Protection on Foreign Direct Investment in OECD Countries

This thesis examines the relation between employment protection and foreign direct investment (FDI) in OECD countries (Organisation for Economic Co-operation and Development). More specifically, it estimates the impact of employment protection legislation on the amount of Japanese foreign direct investment inflows into a set of fourteen OECD countries over the sixteen-year period 1998–2013. Foreign direct investment can have multiple beneficial effects on the host country, such as capital formation and spillover of knowledge and technology, which can improve productivity and lead to economic growth. Because of this, there is an extensive prior theoretical and empirical literature examining the determinants of FDI. This thesis introduces the most relevant theories and empirical results on the topic in the literature review sections, presents empirical analysis on the phenomenon and compares the results to the results of the previous literature. The theory section introduces the most relevant theories of employment protection and foreign direct investment. The theories on the topic are similar in their prediction of a negative relationship between employment protection and foreign direct investment. This suggests that increase in employment protection measures discourages foreign direct investment inflows. The negative relation arises from companies maximising profits by minimising production costs. The costs related to employees are a major expense to large companies, and they want to be able to adjust the number of employees according to current economic situation as easily as possible. The theory states that due to severance payments and restrictions on employment contracts, strict employment protection legislation hinders this adjustment. This makes a country with strict employment protection legislation less attractive investment target for the investor companies. The main econometric method used in this thesis is linear regression analysis. The estimated function is a function with natural logarithm of foreign direct investment as a dependent variable and natural logarithm form of employment protection variable as the independent variable with set of control variables. The data comes mainly from OECD database, apart from data for Japanese bilateral foreign direct investment data, which comes from Japan External Trade Organisation, and the average years of schooling, which comes from Barro and Lee (2010) Educational Attainment dataset, serving as a control for skill level. The baseline ordinary least squares (OLS) estimation results are then augmented with instrumental variable (IV) approach to account for endogeneity issues. The first instrumental variable used is power and political ideology of the ruling party in a country. The second instrumental variable used is unionisation rate. These variables capture variation in employment protection, while being unaffected by foreign direct investment inflows, the dependent variable, thus serving as adequate instruments. The main result of the thesis is that the employment protection seems to have a negligible impact on foreign direct investment. Baseline OLS estimation produces a negative significant coefficient for employment protection, suggesting that increase in employment protection decreases inflow of foreign direct investment. This is predicted by most of the theory. The IV results, however, state that the variable for power and political ideology does not serve as a good instrument, thus this model is discarded. The unionisation rate instrumental variable estimation yields non-significant positive coefficient for the employment protection. This suggests that employment protection is not an important determinant of Japanese foreign direct investment in the fourteen OECD countries selected for the analysis. The results of this thesis fit the prior literature fairly well. In general, the literature on the determinants of foreign direct investment is not unanimous in which country features are important for determining volumes of foreign investment received by a country. In most of the published empirical literature on the employment protection as a determinant of foreign direct investment, the results suggest that the relation is negative, but there are studies finding a complete opposite positive relation. Non-significant coefficients are also fairly common. The direction and magnitude of the impact seems to be sensitive to country and time period samples, as well as model specification. The results also vary depending on which one of the two available measures for employment protection has been used (survey of managers in the Global Competitiveness Report 1999 published by World Economic Forum and OECD employment protection indicators)