The relationship of port efficiency and sector performance with foreign trade in the context of international maritime transportation: Turkey's foreign trade projection for 2035

The aim of this dissertation is to reveal the effects of maritime transport and ports on international trade, especially export volume, econometrically and statistically. With this scope, firstly, the changes in port efficiencies and total factor productivities in context of maritime transportation of important port countries evaluated by global indices in recent years were examined by Mamquist Data Envolopment Analysis (DEA). Then, in addition to economic variables such as Global Domestic Product per capita (GDPp), Purchasing Power Parity (PPP) and Official Exchange Rate (OE), the effect of the sector performance variables related to maritime transport and ports on Total Foreign Trade Volume (TDCH), Total Merchandise Trade Volume (TMCH), Merchandise Export Volume (ME), Merchandise Import Volume (MI) of countries were examined by short-term panel data analysis, respectively. As a result of these analyzes, it was determined that only the effect of container port traffic (CT) among the maritime transport and port performance variables was statistically significant on the dependent variables in all models, in addition to the economic variables. Westerlund ECM (Bootstrap) cointegration analysis and Pedroni's (2001) Mean Group Dynamic Least Squares (DOLSMG) estimator were used between the years 2000-2019, in order to reveal the long-term effect of the CT variable, which is thought to be directly related to the merchandise subject to export and import, on the TDCH, TMCH, ME and MI variables of these countries. As a result of these analyzes, it was determined that the long-term relationship between the CT variable and all dependent variables was statistically significant. In addition, as a result of the DOLSMG estimation model, the long-term effect of the change in the CT variable on the foreign trade variables of the countries was estimated for each of the 63 port countries constituting 94.2% of the world container port traffic. As in the result of the short-term panel data MG estimator as well as the results of DOLSMG estimator, it has been observed that a positive 1-unit change in the CT variable would have a positive effect on Turkey's total foreign trade, merchandise trade, export and import volume in the long term. Turkey, which has been on global trade routes for centuries due to its geographical location, is predicted to become an international logistics center for the global supply chain in the future. There are port infrastructure investments and mega container port projects in the 2010 National Coastal Master Plan, 2023 Turkey Export Strategy and Action Plan, 11th Development Plan (2019-2023) and 2021 Presidential Annual Program in order to use the advantages of Turkey due to its geopolitical position in context of the global trade routes more effectively, to increase its international competitiveness and finally to reach a higher export capacity. With these investment projects included in these action and development plans, it is aimed to increase the existing container port capacity in Turkey by 31,400,000 TEU in total. In this context, as the final analysis of this study, it was estimated whether these investment projects, which will be completed in 2035, are idle or effective, using the time-dependent change of the Turkey's CT variable. At this stage, stochastic process analyzes, GM (1,1) model and Markov chain analysis, were used. As a result of these processes, it has been determined that the total CT volume in Turkish ports estimated for the end of 2035 can be met with the load capacity of the planned ports and thus these ports are necessary and effective investments. Then, in order to determine how the estimated CT volume to be reached at the end of 2035 will affect the country's foreign trade, it combined respectively with the long-term possible growth rates in the CT variable obtained as a result of the GM (1,1) and Markov chain analysis, and the estimation results of the DOLSMG estimator for the variables obtained for Turkey. As a result of the integration of stochastic processes and econometric analysis, Turkey's merchadise export volume, merchandise import volume and total merchandise trade volume in 2035 were estimated. In the conclusion part of the thesis, the estimated data obtained were evaluated together with the Turkish maritime transport sector and export policies.