Cost characteristics of hospitals
Introduction
In this paper we try to gain an insight into the literature on the cost structure of hospitals. The aim is not to give a complete overview of the existing empirical literature, the focus is rather on the theoretical and methodological aspects involved. Modern hospitals are complex multi-product organisations that should be analysed using the appropriate techniques. The more recent studies derive flexible hospital cost functions which are consistent with neo-classical production theory. Using multi-product cost functions and regressing total costs on various output levels and input prices allows the degree of economies of scale and scope to be calculated. However, before estimating a cost function, it is necessary to keep in mind some important considerations and complications. Can hospitals be considered as truly cost-minimising firms? Do they operate in their short- or long-run equilibrium? What are the local and the global properties of flexible functional forms?
The paper is structured as follows: first some methodological aspects of cost functions are discussed. More specifically, the two main categories of hospital cost function studies will be introduced, concentrating however on the more recent technique of using flexible functional forms derived from the microeconomic neo-classical theory of production. Complications resulting from the use of these flexible functional forms are also dealt with in this section. In the second part of the paper some specific hospital cost function studies are discussed to give the reader an example of the issues tackled in the methodological part. An alternative approach as well as some relevant related topics are also shortly discussed in the last section.
Section snippets
Hospital cost function estimation
The theory on economies of scale was originally developed for and applied to single-product outputs in industry. Later on the concepts of ray average cost, multi-product scale economies and economies of scope were introduced to relax the restrictive assumption of producing a single output. Originating from the theory of firms producing multiple-product outputs, this method can also be used to derive economies of scale and scope for (non-profit) organisations such as hospitals.
Breyer (1987)
Cost minimisation
The neo-classical cost function assumes that hospitals minimise (variable) costs, given that input prices and outputs are exogenous. Therefore, applying flexible functional forms such as the quadratic, translog or generalised translog cost function implies cost-minimising behaviour of hospitals. Cost minimisation has been questioned: given the not-for-profit status of most hospitals and the cost-based reimbursement system in many countries, this assumption may be violated. However, Pauly (1987)
Specific hospital cost function studies
This section does not intend to give a complete overview of the existing empirical research (e.g. NHS Centre for Reviews and Dissemination (1996) is a comprehensive survey). Instead, some specific (often cited) hospital cost function studies will be discussed to give the reader an example of the issues tackled in the previous methodological part. For every article we will specify which of these methodological aspects were actually adopted or discussed (e.g. the specification of the model, the
Concluding comments
We have tried to gain an insight into the literature on the cost structure of hospitals, which nowadays are complex multi-product organisations. The most appropriate technique to analyse hospital cost structure is to estimate a flexible functional form of a cost function which is derived from the neo-classical theory of the firm. This approach assumes that hospitals minimise (variable) costs, given that input prices and outputs are exogenous. Therefore, applying flexible functional forms such
Acknowledgements
The author would like to thank the referees of this journal for their useful comments.
References (26)
Maximum likelihood estimation of complete systems of demand equations
European Economic Review
(1969)- et al.
On the estimation of returns to scale using variable cost functions
Economics Letters
(1983) The specification of a hospital cost function. A comment on the recent literature
Journal of Health Economics
(1987)- et al.
Payment source and the cost of hospital careEvidence from a multiproduct cost function with multiple payers
Journal of Health Economics
(1996) Economies of scale in non-revenue producing cost centersImplications for hospital mergers
Journal of Health Economics
(1998)- et al.
Uncertain demand, the structure of hospital costs, and the cost of empty hospital beds
Journal of Health Economics
(1995) - et al.
Estimating hospital costs. A multiple-output analysis
Journal of Health Economics
(1986) The logged dependent variable, heterscedasticity, and the retransformation problem
Journal of Health Economics
(1998)Much ado about twoReconsidering retransformation and the two-part model in health econometrics
Journal of Health Economics
(1998)Competition among hospitals
Journal of Health Economics
(1988)
Exploring hospital production relationships with flexible functional forms
Journal of Health Economics
Uncertainty and the welfare economics of medical care
American Economic Review
Contestable markets and the theory of industry structure
Cited by (38)
Economies of scale and scope in the Danish hospital sector prior to radical restructuring plans
2012, Health PolicyCitation Excerpt :Consequently, we used the geographically distinct production unit, not the management entity, as the unit of analysis in our study. Most studies argue that no evidence exists that hospitals operate at their long-run equilibrium, i.e., hospitals do not adjust all their inputs to their cost-minimizing levels [9]. From a theoretical point of view, it is therefore appropriate to estimate short-run cost functions.
Restructuring hospital industry to control public health care expenditure: The role of input substitutability
2010, Economic ModellingCitation Excerpt :The results are quite similar to the ones presented in Section 4, with a cost elasticity with respect to Y (∂lnVC/∂lnY) equal to 0.602 and a cost elasticity with respect to K (∂lnVC/∂lnK) positive and equal to 0.082. This latter result is quite common in the literature (see, for example, Smet, 2002 and Aletras, 1999), and suggests that hospitals are over-investing in capital. The estimates of short-run scale economies–1/(∂lnVC/∂lnY)–are equal to 1.66, while long-run scale economies, computed using the formula introduced by Caves et al. (1981)–[1 −(∂lnVC/∂lnK)]/(∂lnVC/∂lnY)–are equal to 1.52.
Do waiting times reduce hospital costs?
2009, Journal of Health EconomicsCitation Excerpt :Since the coefficient is significant we can also exclude that hospitals are operating in the long-run equilibrium. Moreover, a positive coefficient implies overinvestment in capital (Cowing and Holtmann, 1983; Smet, 2002; Vita, 1990; Grannemann et al., 1986). The introduction of emergency admissions or day cases does not affect the results significantly (see columns 3 and 4).
Cost minimization models: Applications in a teaching hospital
2008, European Journal of Operational ResearchUNDERSTANDING AND ANTICIPATING LAG-TIME BIAS IN COST-EFFECTIVENESS STUDIES: THE ROLE OF TIME IN COST-EFFECTIVENESS ANALYSIS
2014, International Journal of Technology Assessment in Health Care