Four essays on efficiency and productivity of cultural institutions : empirical analyses of orchestras, theatres and museums
This thesis builds on the research tradition of Cultural Economics, particularly of its empirical strand. The thesis includes four articles that each analyse empirically a different facet of production of cultural services. Together they contribute to the two main themes of the thesis: features of production technology and applicability of different empirical methods in analysing production of cultural services. The four articles proceed in "chronological" order. The first article deals with problems shared by the subsequent studies - measurement of inputs, output(s) and quality of production - and employs neo-classical single-output cost functions and Structural Equation Modeling approach together with a cross-sectional data set of museums. The second article follows the earliest strand of cultural economics and examines, in a context of Baumol's cost disease, the assumption of stagnant productivity growth. The analysis utilises two index number approaches, Törngvist Approximated and Generalised Divisia Indices, and a panel data set of orchestras. The third article looks at the scale properties of production as well as tests the assumption of allocative efficiency by using non-linear single-output cost functions with a panel data set of theatres. The fourth article focuses on cost-efficiency of museums and utilises non-parametric Free Disposal Hull method and a cross-sectional data set of museums. The main findings of the thesis indicate, first, that production technologies of museums, orchestras and theatres vary substantially. The production technology of museums is shown, in a single-output setting, to exhibit homotheticity and homogeneity with respect to output as well as substantial increasing/decreasing scale economies, depending on the output measure. Analysis in a multiple output context reveals that on average a quarter of museums are cost-inefficient and private museums are more likely to be inefficient than the publicly owned ones. The production technologies of orchestras and theatres are alike with respect to single-output production as well as non-homotheticity and nonhomogeneity of production technology. They differ as to the scale properties - orchestras exhibit diseconomies while theatres are characterised by scale economies - as well as relative usage of labour input. Production in orchestras is characterised by stagnant productivity that is a result of scale effect cancelling out technical change. Theatres are shown to be moderately allocatively inefficient. As to the methodology, the thesis suggests that particularities of cultural institutions have to be taken into account. These particularities involve problems related to measurement of inputs, output(s) and quality of production, definition of production technology as well as problems arising form data. Methods employed in price-quantity space, allowing in-efficiencies of production and not imposing a priori assumptions on production technology are argued to be most applicable ones.