Skip to main content

An econometric approach to estimating support prices and measures of productivity change in public hospitals

Abstract

In industry sectors where market prices for goods and services are unavailable, it is common to use estimated output and input distance functions to estimate rates of productivity change. It is also possible, but less common, to use estimated distance functions to estimate the normalised support (or efficient) prices of individual inputs and outputs. A problem that arises in the econometric estimation of these functions is that more than one variable in the estimating equation may be endogenous. In such cases, maximum likelihood estimation can lead to biased and inconsistent parameter estimates. To solve the problem, we use linear programming to construct a quantity index. The distance function is then written in the form of a conventional stochastic frontier model where the explanatory variables are unambiguously exogenous. We use this approach to estimate productivity indexes, measures of environmental change, levels of efficiency, and support prices for a sample of Australian public hospitals.

This is a preview of subscription content, access via your institution.

Fig. 1

Notes

  1. A small portion (<7 % in 2007–2008) of public hospital funding comes from insurance funds and other non-government sources.

  2. In February 2011 the Commonwealth Government announced that it will fund a lower percentage of efficient prices.

  3. If input prices are available then an estimated cost function will give the minimum cost of producing (i.e., the efficient price of) a vector of outputs. The first derivatives of the cost function with respect to individual outputs will give the efficient prices of those outputs under marginal cost pricing.

  4. Let \(a=(a_1,\ldots,a_K)^{\prime}\). The notation a ≧ 0 K means that a k  ≥ 0 for \(k=1,\ldots,K\). The notation a ≥ 0 K means that a k  ≥ 0 for \(k=1,\ldots,K\) and there exists at least one value \(k\in \{1,\ldots,K\}\) where a k  > 0. The notation a > 0 K means that a k  > 0 for \(k=1,\ldots,K. \)

  5. The output and input distance functions are defined as \(D_O(x,q,z)=\inf\{\delta>0:(x,q/\delta)\in T(z)\}\) and \(D_I(x,q,z)=\sup \{\rho > 0:(x/\rho,q) \in T(z)\}\). If the production possibilities set is given by (1) then \(D_O(x,q,z)=\inf\{\delta>0:h(q/\delta)^r \le b(z)g(x)\}=\inf\{\delta>0:h(q)^r/\delta \le b(z)g(x)\}=\inf\{\delta>0:\delta \ge h(q)^r/[b(z)g(x)]\}\) = h(q)r/[b(z)g(x)] and \(D_I(x,q,z)=\sup \{\rho > 0:h(q)^r \le b(z)g(x/\rho)\}=\sup \{\rho > 0:h(q)^r \le b(z)\rho^{-r}g(x)\}=\sup\{\rho > 0:\rho \le [b(z)g(x)]^{1/r}/h(q)\}=[b(z)g(x)]^{1/r}/h(q)\).

  6. Coelli (2000, pp. 10–16) suggests that it is also fine in the case where firms minimise costs and there is no statistical noise (i.e., when the frontier is deterministic).

  7. For a discussion of Lowe indexes and their properties see, for example, Hill (2008).

  8. Most of these variants involve constraining the weights a it in problem (22) to be strictly positive, something that can be done easily in Excel Solver. However, using Excel to solve these weight-restricted problems for every observation in the dataset can be time-consuming if the number of observations is large.

  9. These axioms include the linear homogeneity, identity, homogeneity of degree zero, commensurability, proportionality and transitivity axioms listed in O’Donnell (2012c), plus a weaker version of the monotonicity axiom listed in that paper. This list of basic axioms differs slightly from lists of axioms and tests found elsewhere in the literature [e.g., Balk (2008)] where quantity indexes are required to be functions of prices.

  10. If there is no variation in the output vector then the minimisation problem (23) is equivalent to the standard data envelopment analysis (DEA) LP used to compute the reciprocal of the it-th technical efficiency score under the assumption of constant returns to scale (CRS)—see, for example, O’Donnell (2010, p. 543, eq. 6.6). Thus, aggregate inputs can be computed quickly by (1) constructing an artificial dataset in which there is only one observation-invariant output, then (2) using this artificial dataset to solve the standard DEA technical efficiency problem under the assumption of CRS. Importantly, the ‘technical efficiency’ estimates reported by DEA software packages should not be interpreted as measures of efficiency—they are the reciprocals of aggregate inputs. In this paper we implemented this fast-computing procedure using the \(\hbox{DPIN}^{\rm TM}\) software.

  11. An alternative to the half-normal is the truncated normal distribution. The half-normal was chosen because (1) the truncated normal involves the estimation of an additional parameter, and (2) if the idiosyncratic error is normally distributed (as it is here), the truncated normal can make it harder to distinguish inefficiency from noise [Coelli et al. (2005, p. 252)].

  12. The percentage increase in maximum aggregate output (and maximum revenue) per period, holding inputs and the output mix constant, is \( -\partial {\ln} {D_O}(x_{it}, q_{it}, z_{it})/\partial t = \partial \ln b(z_{it})/\partial t = \gamma_1 = \phi_1 r. \)

  13. For information on the growing problem of antibiotic resistance, see Miller and Miller (2011). The failure of technological advances to keep pace with deteriorations in the physical environment is sometimes referred to as the Red Queen effect. This a reference to the Red Queen’s race in Lewis Carroll’s Through the Looking-Glass. The Red Queen said “it takes all the running you can do, to keep in the same place. If you want to get somewhere else, you must run at least twice as fast as that!”.

  14. If the technology is represented by (1) then the cost function is given by (4) \(\Rightarrow MC_{nit}\equiv {\partial C(w_{it},q_{it},z_{it})}/{\partial q_{nit}=C(w_{it},q_{it},z_{it})}\times {\partial \ln h(q_{it})/\partial q_{nit}} \Rightarrow p_{{nit}}^{*} = rMC_{{nit}} /D_{I} (x_{{it}} ,q_{{it}} ,z_{{it}} )^{r}.\) It follows that if D I (x it q it z it ) = 1 then \( p_{{nit}}^{*} = rMC_{{nit}} \).

References

  • Aigner D, Lovell C, Schmidt P (1977) Formulation and estimation of stochastic frontier production function models. J Econ 6:21–37

    Article  Google Scholar 

  • Balk B (2008) Price and quantity index numbers: models for measuring aggregate change and difference. Cambridge University Press, New York

    Book  Google Scholar 

  • Barbetta G, Turati G, Zago A (2001) On the impact of ownership structure and hospital efficiency in Italy. In European workshop on efficiency and productivity. University of Oviedo.

  • Battese G, Coelli T (1988) Prediction of firm-level technical efficiencies with a generalized frontier production function and panel data. J Econ 38:387–399

    Article  Google Scholar 

  • Biorn E, Hagen TP, Iversen T, Magnussen J (2002) The effect of activity-based financing on hospital efficiency: a panel data analysis of DEA efficiency scores 1992–2000. HERO Online Working Paper Series No. 8, Oslo University

  • Bradford D, Kleit A, Krousel-Wood M, Re R (2001) Stochastic frontier estimation of cost models within the hospital. Rev Econ Stat 83:302–309

    Article  Google Scholar 

  • Carey K (2003) Hospital cost efficiency and system membership. Inquiry 40:25–38

    Article  Google Scholar 

  • Cherchye L, Moesen W, Rogge N, van Puyenbroeck T, Saisana M, Saltelli A, Liska R, Tarantola S (2008) Creating composite indicators with DEA and robustness analysis: the case of the Technology Achievement Index. J Oper Res Soc 59:239–251

    Article  Google Scholar 

  • Chirikos TN, Sear AM (2000) Measuring hospital efficiency: a comparison of two approaches. Health Serv Res 34:1389–1408

    Google Scholar 

  • Coelli T (2000) On the econometric estimation of the distance function representation of a production technology. Centre for Operations Research and Econometrics (CORE) discussion papers, Universite Catholique de Louvain.

  • Coelli T, Rao D, O’Donnell C, Battese G (2005) An introduction to efficiency and productivity analysis, 2nd edn. Springer, New York

    Google Scholar 

  • Commonwealth of Australia (2010) A national health and hospitals network for Australia’s future. Canberra.

  • Deily ME, McKay NL, Dorner FH (2000) Exit and inefficiency—the effects of ownership type. J Hum Resour 35:734–747

    Article  Google Scholar 

  • Dervaux B, Ferrier G, Leleu H, Valdmanis V (2004) Comparing French and US hospital technologies: a directional input distance function approach. Appl Econ 36:1065–1081

    Article  Google Scholar 

  • Despotis D (2005) A reassessment of the Human Development Index via data envelopment analysis. J Oper Res Soc 56:969–980

    Article  Google Scholar 

  • Diaz M, Sanchez R (2008) Firm size and productivity in Spain: a stochastic frontier analysis. Small Bus Econ 30:315–323

    Article  Google Scholar 

  • Färe R, Primont D (1995) Multi-output production and duality: theory and applications. Kluwer Academic Publishers, Boston

    Book  Google Scholar 

  • Fernandez C, Koop G, Steel M (2000) A Bayesian analysis of multiple-output production frontiers. J Econ 98:47–79

    Article  Google Scholar 

  • Ferrari A (2006) The internal market and hospital efficiency: a stochastic distance function approach. Appl Econ 38:2121–2130

    Article  Google Scholar 

  • Folland ST, Hofler R (2001) How reliable are hospital efficiency estimates? Exploiting the dual to homothetic production. Health Econ 10:683–698

    Article  Google Scholar 

  • Frech HE, Mobley LR (2000) Efficiency, growth, and concentration: an empirical analysis of hospital markets. Econ Inq 38:369–384

    Google Scholar 

  • Herr A (2008) Cost and technical efficiency of German hospitals: does ownership matter? Health Econ 17(9):1057–1071

    Google Scholar 

  • Hill P (2008) Lowe Indices. In 2008 world congress on national accounts and economic performance measures for nations. Washington, DC

  • Horrace W, Schmidt P (1996) Confidence statements for efficiency estimates from stochastic frontier models. J Prod Anal 7:257–282

    Article  Google Scholar 

  • Jacobs R (2001) Alternative methods to examine hospital efficiency: data envelopment analysis and stochastic frontier analysis. Health Care Manag Sci 4:103–115

    Article  Google Scholar 

  • Linna M, Hakkinen U, Magnussen J (2006) Comparing hospital cost efficiency between Norway and Finland. Health Policy 77:268–278

    Article  Google Scholar 

  • Mahlberg B, Obersteiner M (2001) Remeasuring the HDI by data envelopment analysis. Interim report No. IR-01-069, International Institute for Applied Systems Analysis (IIASA), Laxenburg, Austria.

  • McKay N, Deily M, Dorner (2002) Ownership and changes in hospital inefficiency: 1986–1991. Inquiry 39(4):388–399

  • Miller A, Miller P (2011) Emerging trends in antibacterial discovery: answering the call to arms. Caister Academic Press, Norwich

    Google Scholar 

  • Morrison Paul C (2002) Productive structure and efficiency of public hospitals. In: Fox K (eds) Efficiency in the public sector. Kluwer Academic Publishers, Boston

    Google Scholar 

  • O’Donnell C (2010) Measuring and decomposing agricultural productivity and profitability change. Aust J Agric Resour Econ 54:527–560

    Article  Google Scholar 

  • O’Donnell C (2012a) An aggregate quantity framework for measuring and decomposing productivity change. J Prod Anal. doi:10.1007/s11123-012-0275-1

  • O’Donnell C (2012b) Econometric estimation of distance functions and associated measures of productivity and efficiency change. J Prod Anal. doi:10.1007/s11123-012-0311-1

  • O’Donnell C (2012c) Nonparametric estimates of the components of productivity and profitability change in US agriculture. Am J Agric Econ 94:873–890

    Article  Google Scholar 

  • Productivity Commission (2010) Public and private hospitals: multivariate analysis. Productivity Commission Supplement to Research Report, Canberra

  • Roibas D, Arias C (2004) Endogeneity problems in the estimation of multi-output technologies. Efficiency Series Papers No. 6/2004. University of Oviedo, Oviedo

    Google Scholar 

  • Rosko MD, Proenca J (2005) Impact of network and system use on hospital X-inefficiency. Health Care Manag Rev 30:69–79

    Article  Google Scholar 

  • Shephard R (1970) The theory of cost and production functions. Princeton University Press, Princeton

    Google Scholar 

  • Smet M (2007) Measuring performance in the presence of stochastic demand for hospital services: an analysis of Belgian general care hospitals. J Prod Anal 27:13–29

    Article  Google Scholar 

  • Street A (2003) How much confidence should we place in efficiency estimates. Health Econ 12:895–907

    Article  Google Scholar 

  • Street A, Jacobs R (2002) Relative performance evaluation of the English acute hospital sector. Appl Econ 34:1109–1119

    Article  Google Scholar 

  • Vitaliano D, Toren M (1996) Hospital cost and efficiency in a regime of stringent regulation. East Econ J 22:161–175

    Google Scholar 

  • Yong K, Harris A (1999) Efficiency of hospitals in Victoria under casemix funding: a stochastic frontier approach. Centre for Health Program Evaluation Working Papers No. 92. Monash University, Melbourne

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to C. J. O’Donnell.

Additional information

An earlier version of this paper entitled “An Econometric Approach to Estimating the Components of Productivity Change in Public Hospitals” was presented at the International Workshop on Efficiency and Productivity in Honour of Professor Knox Lovell, Elche, Spain, 4–5 October 2010.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

O’Donnell, C.J., Nguyen, K. An econometric approach to estimating support prices and measures of productivity change in public hospitals. J Prod Anal 40, 323–335 (2013). https://doi.org/10.1007/s11123-012-0312-0

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11123-012-0312-0

Keywords

  • Endogeneity
  • Distance functions
  • Shadow prices
  • Efficient prices
  • Total factor productivity
  • Environmental change
  • Technical efficiency
  • Scale efficiency

JEL Classification

  • C43
  • D24
  • I12
  • I18