Haavelmo hypothesis in saving spiral: The case of Turkey

This thesis aims to test the validity of the Haavelmo hypothesis, which argues that excessive foreign capital inflows may reduce domestic savings in developing countries, in the Turkish economy. For this purpose, two different models are constructed in the analysis of the study. In the first model, total savings deposits and in the second model, foreign currency deposit accounts are included in the model as dependent variables. Foreign direct investments, portfolio investments and trade openness variables constitute the independent variables. Covering the 1998:Q1-2023:Q1 period, the reason for constructing two different models is to determine whether foreign capital inflows increase foreign currency savings in Turkey if the Haavelmo hypothesis is valid. In other words, it is aimed to test the hypothesis "Do foreign capital inflows in Turkey increase foreign currency savings at the same time as they decrease domestic savings?". According to the results of the ARDL bounds test conducted in the analysis of the study, it is determined that there is a negative relationship between foreign direct investment inflows and domestic savings, while there is no statistically significant relationship between portfolio investments and domestic savings. According to the results of the analysis of the second model in which foreign currency savings is the dependent variable, there is no statistically significant relationship between foreign capital inflows and foreign currency deposit accounts. The results of the analysis suggest that Haavelmo's hypothesis is valid in the Turkish economy. In other words, foreign capital inflows lead to a decline in domestic savings in Turkey. In addition, there is no relationship between foreign capital inflows and foreign currency savings, and the decreasing domestic savings are not directed to foreign currency deposit accounts.