Relationship between human capital and economic growth in developing countries

II. After World War II, economic differences emerged more clearly with the increase in technological developments. Then, the problem of economic growth came to the fore. While physical capital was sufficient for economic growth in previous studies, the contribution of human capital to economic growth has begun to be investigated in today's studies. Due to the increasing importance of human capital, it has taken its place in the literature as one of the most important determinants of economic growth. The aim of this study is to define the concepts of human capital and economic growth. In addition, it is to determine whether there is a relationship between these variables by time series and econometric methods. For this purpose, first of all, the concepts of stationarity, long-term and short-term relationships and the theoretical structure of the tests to be used were examined. In the research, the human capital index and the 2011 base prices, which are used as indicators of economic growth, and the Real Gross Domestic Product (GDP) variables are discussed. The data used were obtained from Penn World Tables (PWT9). A time series was created for the period 1951-2017 for 26 developing countries with reliable data. In the analysis, firstly, the variables examined for each country were included by deciding on the most appropriate model. Then, the stationarity levels of the variables were determined by ADF and PP unit root tests. If the variables for each country are stationary at the same level, the Johansen Cointegration Test was used to understand whether there is a long-run relationship between them. Causality Analysis was applied in order to understand the direction of the relationship between the variables of the stable countries at the same level and whether there is a short-term relationship. If the variables examined for each country are stationary at the same level, the Vector Autoregression (UVAR) model is used if there is no long-term relationship between them, and the Vector Error Correction (VECM) model is used if there is a long-term relationship. It has been tested whether there is a short-term relationship between them. Toda-Yamamoto Causality Analysis was applied if the variables were integrated at the same/different level and regardless of whether they were cointegrated. In addition, Impulse-Response Analysis was used to determine the responses of the variables to a standard deviation shock on itself or on another variable, and the duration of these reactions in countries with a short-term relationship. Variance Decomposition was performed in order to determine what percentage of the change in these variables was caused by itself and what percentage was caused by the other variable. In the tests performed, the most frequently used α=0.05 significance level in the literature was chosen.