Specification of a long run transit cost function to estimate input bias from subsidies

Abstract

The common econometric approach to the study of subsidy impacts on costs does not consider the changes in the ratios of the marginal products of the inputs brought about by the subsidies and which affect the cost function. Instead, this approach includes subsidies as an explanatory variable in the cost function. This paper presents an alternative approach that explicitly considers the changes in the marginal products of the inputs when subsidies are offered and incorporates them directly into a cost function derived by assuming a Cobb‐Douglas technology. This cost function is estimated from a sample of 96 U.S. bus transit systems and the sizes of the inefficiencies from subsidies calculated. It is found that operating and capital subsidies increased costs by a factor of 2.2 and that these subsidies lead to mostly non‐capital bias when a composite measure of noncapital inputs is used. However, there is capital bias when capital is compared to fuel and labour separately. An additional finding of this paper is that subsidies increased the own price elasticities of demand and made fuel and capital demand price elastic.

Key words:

• Bus transit
• operating subsidies
• capital subsidies
• inefficiency
• cost