EMPIRICAL ANALYSIS OF TRANSPORT DEMAND AND EFFICIENCY OF DIFFERENT TRANSPORT MODES IN PAKISTAN

Abstract

It is well known that empirical analysis of transport demand is important in terms of making effective decisions regarding transport demand management and planning. Therefore, this thesis uses two objectives, in which the overall objective is to increase the understanding of how aggregate freight and passenger (travel) demand for rail and road in Pakistan is determined by various factors. Therefore, first objective investigates the demand for rail (freight and passenger) transport in Pakistan within Johansen’s multivariate co-integration framework using annual time series data over 1978-2018. Similarly, second part analyzes the determinants of the demand for both road freight and passenger demand. For that, annual time series data is used from 1980 to 2016, and ARDL bounds test of co-integration is employed because variables have different order of integration (I(0) and I(1)). The second objective estimates the (technical) efficiency of rail, road, and rail vs. road transport in Pakistan with a non-parametric approach of data envelopment analysis (DEA), and efficiency patterns are traced over time from 1980 to 2018. The empirical results show that rail freight (passenger) demand is relatively more inelastic in the long-run, offering important policy implications for Pakistan-railways to manage its operations through changes in freight-rate and fare. Similarly, the long-run passenger demand for rail and road is positively related to real per-capita GDP, which means that the demand for road and rail travel can be treated as the normal goods. Moreover, income elasticities of travel demand and output elasticities of freight demand across both rail and road are positive and relatively more elastic in the long-run, which implies that a significant increase in transport demand is expected in future as economic growth in Pakistan increases.Cross-price elasticities of rail and road transport demand are found positive and significant, indicating that both road and rail transport are substitutes. The demand for road transport is inelastic to its own-price, particularly passenger demand, which means that the role of market instruments is relatively less effective in terms of controlling the future growth of road transport demand. Although it is found that population and the passenger demand for rail and road are positively related in long-run, the effect of (urban) population on road travel demand is far more dominant than the impact of population on rail travel demand. Similarly,rail (road) densities are positively related to passenger demand for respective transport modes in the long-run. However, rail passenger demand is far more sensitive to rail density than road travel demand to road-density. In the short-run, a comparison of error-correction terms of passenger demand for rail and road indicate thatroad passenger demand adjusts relatively quickly than rail travel demand to restore xv the long-run equilibrium, while error in rail freight demand is corrected at a relatively faster rate than road freight demand to re-establish the long-run relationship.Finally, the results based on efficiency scores indicate that, in general, the technical efficiency of rail transport has increased over time, as Pakistan-railway has gradually shifted from inefficient coal-traction to a more efficient diesel-traction. A similar pattern is observed for road transport due to a relatively faster growth in its outputs than inputs. However, efficiency of road transport is higher than rail transport when relative efficiency comparisons are made.