Assessing the Impact of Multimodal Transportation on Economic Growth: A Machine Learning and Cointegration Approach in 28 Countries

Abstract

In this study, the effect of freight and passenger transport in different modes on economic growth is determined for 28 selected countries. The Westerlund cointegration test is used to reveal the long-term relationship between freight and passenger transportation and growth. According to the cointegration analysis, all transportation modes (road, rail, and air) are cointegrated with growth. Additionally, machine learning models were used to predict growth based on each transportation mode for each country for the upcoming four years and to determine the importance of the input parameters. According to the importance of the parameter analysis, for the entire panel, rail transport is the most effective transport mode for economic growth. On a country-by-country basis, the findings differ. Rail transport is the strongest transport mode for growth in high-income countries. However, although it is not the dominant mode, the relative impact of air passenger transport is strong. In upper middle-income countries, there generally is not a dominant mode of transport, but in general, freight transport is important to economic growth. In passenger transportation, air passenger transport is the most prominent mode in these countries. In lower middle-income countries, rail freight is the strongest transport mode for economic growth.