یارانه حمل و نقل عمومی، اثر خروجی و بهره وری کل عوامل

This paper extends previous works on total factor productivity decomposition when firms receive both operating and capital subsidies. It shows that previous works considered either the lump-sum or substitution effects of these subsidies but not together. Using constrained cost minimization as the framework it offers formal proofs to show that cost increases are inevitable if the total effects of the subsidies are considered, and that total factor productivity growth results from increasing amounts of subsidies under economies of scale and in the absence of technical change. Applications of the decomposition equations derived to a sample of transit systems finds near constant returns to scale and negative contributions of these subsidies to total factor productivity growth. Technical change reverses this decline and results in total factor productivity growth. Further, it finds that the lump-sum effects of the subsidies reduce total factor productivity more than does the substitution effect.

More than two decades ago Kim and Spiegel (1987) examined the effects of lump-sum subsidies on the productivity of public transit systems that operated under rate of return regulation. Through their theoretical work, they showed that lump-sum subsidies affect factor proportions, output and cost. Following their theoretical work they estimated a neoclassical cost function that included lump-sum subsidy as a variable and used it to link these subsidies to decline in total factor productivity (TFP). Since then, few studies have followed the direction of Kim and Spiegel to theoretically derive models that link subsidies to TFP in public transit systems. They include Obeng and Sakano, 2000 and Obeng and Sakano, 2002 who studied operating and capital subsidies in a constrained output maximization framework and linked them to TFP. The major difference in these models is their treatment of subsidies. Kim and Spiegel (1987) treated subsidy as a variable in their model. Obeng and Sakano, 2000 and Obeng and Sakano, 2002 showed that because capital and operating subsidies are input oriented they make managers misperceive input prices. This misperception results in implied input prices that are lower than actual input prices and optimal input combinations that increase cost just as in Kim and Spiegel (1987). The empirical work of Obeng and Sakano (2000) reinforced the conclusions of Kim and Spiegel (1987) that subsidies contribute to TFP decline though in later work (Obeng & Sakano, 2002) they did not find this decline. In other studies, Kerstens, 1996 and Kerstens, 1999 found technical efficiency was negatively related to subsidies and in Matas and Raymond (1998) average efficiency was negatively related to subsidies. Nolan (1996) found negative relationships between state subsidies and technical efficiency and positive relationships between federal subsidies and technical efficiency. In the private sector, Bergstrom (2000) studied capital subsidies and found little evidence to show that they affected productivity. He argued just as Boame and Obeng (2005) did that subsidies advance the accumulation of technology in firms and could increase productivity. This is especially so because capital subsidies add to productive capacity and replacement investment such as modernizing existing capital stock and increasing efficiency (Harris & Trainor, 2005). In addition, Bergstrom (2000) argued that capital subsidies can help firms use economies of scale.

The objective of this paper is to examine the effects of operating and capital subsidies on TFP and extend previous work, particularly Obeng and Sakano, 2000 and Obeng and Sakano, 2002. This re-examination hinges on the fact that these subsidies have both lump-sum and input substitution effects. To the knowledge of the authors, these two effects have not been considered in the transport economics literature. In the theoretical sections of this paper two important proofs show that indeed these subsidies increase resource cost, and in the presence of diseconomies of scale reduces TFP when technical change is absent. The empirical section confirms these results showing that in transit systems characterized by slight diseconomies of scale, when both the lump-sum and substitution effects of the subsidies are considered together they decrease TFP by 0.02% per year. Comparatively, the substitution effects of the subsidies increase TFP by 0.03% per year when only those effects of subsidies on input use are considered. From these results, it is concluded that the inclusion of the total effect of the subsidies in TFP decomposition is important to fully understand the changes that have occurred in productivity.