Investigation of price bubbles in Borsa Istanbul indices: An ARDL approach

This study examines the existence of bubbles in asset prices in two stages. In the first stage, the presence of price bubble in Istanbul Stock Exchange (ISE) indices is examined. For this purpose, monthly data of the six indices between January 2005 and December 2018 were used. In the second stage, the relationship between the bubble series and macroeconomic variables was investigated by using ARDL boundary test. The study consists of three parts, in the first chapter the definition of bubble and the theoretical framework for the bubble. In addition, types of bubbles, including rational bubbles and irrational bubbles, and types of investors that cause bubble types have been described. In the second part of the study, the Efficient Markets Hypothesis, which contradicts the definition of bubble and has many advocates in the financial literature, is explained. In contrast to the bubbles that can be defined as the difference between the underlying value of the asset and the market value, the Efficient Markets Hypothesis states that the underlying value of an asset and its market value will be the same. Accordingly, the price of the assets accurately reflects all the information in the market. In case there is a new information entry into the markets, the prices will react quickly and there will be no difference between the basic value of the asset and the market value. According to the hypothesis, if there is a difference between the actual price of the asset and the market price, this difference will be eliminated by arbitrager. Therefore, there is no possibility of bubble formation in an efficient market. According to the Efficient Markets Hypothesis, the phenomenon that is perceived as bubble is the price movements that occur depending on the supply and demand of the asset. In the last part of the study, the presence of bubbles was investigated in the period between 2005-2018 and bubble formation was observed in the period between 2005-2008 according to the obtained bubble series. Similar bubble graphs were obtained in all indices. In addition to the ARDL test, the relationship between the bubble series and macroeconomic variables was analyzed by Granger causality test to determine the causality. According to the results of the analysis, it was determined that there is a bidirectional causality between ISE National 100 and ISE National 30 indices and money supply and consumer price index. In addition, the causality relationship between the independent variables was examined and it was seen that there was a bidirectional causality between the consumer price index and overnight interest rate, industrial production index and money supply.