An Examination of Public Hospital Productivity and its Persistence: An Index Number Approach

• Author: Wenda Yan,Anthony Scott,Choon C. Cheng,Vijaya Sundararajan,Jongsay Yong
• DOI: http://doi.org/10.1111/1467-8462.12375
• Published date: 01 September 2020
• Date: 01 September 2020

The Australian Economic Review, vol. 53, no. 3, pp. 343–359 DOI: 10.1111/1467-8462.12375

An Examination of Public Hospital Productivity and its

Persistence: An Index Number Approach

Choon C. Cheng, Anthony Scott, Vijaya Sundararajan, Wenda Yan and

Jongsay Yong*

Abstract

This paper measures the level and growth of

total factor productivity (TFP) of public

hospitals in Victoria, Australia, using an

index number approach. We further examine

the persistence of productivity over time, and

the extent to which productivity varies with

hospital characteristics such as hospital size.

Hospital administrative data from Victoria

from 2007–08 to 2011–12 are used. We find

substantial variation in TFP across hospitals:

large hospitals perform significantly better

than small hospitals, both in TFP level and

growth. Productivity level is highly persistent

over time, but not productivity growth.

1. Introduction

The costs of health care have been rising in

most developed economies due primarily to

population ageing and advancement in med-

ical technologies. Public financing of hospitals

in many countries has struggled to keep pace

with demand, especially in the face of recent

tumultuous global economic situations. Given

that the rise in demand for care is outstripping

the increase in public finance, improving the

productivity and efficiency of hospitals is one

of few viable policy options. Consequently,

measuring and understanding the productivity

of hospitals has become a topic of interest in

recent years. In the United Kingdom, for

example, alleged negative productivity growth

of the UK medical care sector became such a

major concern that a special parliamentary

commission was formed to explore the issue

(Atkinson 2005). In the United States,

Medicare payments have been linked to

expected improvements in hospital perfor-

mance on a number of cost and quality

dimensions under the Affordable Care Act

(Blumenthal, Abrams and Nuzum 2015;

Cutler 2015; Nyweide et al. 2015). In

Australia, the efficiency of hospitals has also

been in the spotlight following the introduc-

tion of a national activity‐based funding

scheme for all public hospitals from 2012

onward (IHPA 2013).

The objectives of this paper are twofold.

First, we aim to document the level and changes

in the productivity of public hospitals over time

in Victoria, Australia, using an index number

approach. Second, we aim to uncover what

* Cheng: Department of Health and Human Services,

Victoria, Australia; Scott, Yan and Yong: Melbourne

Institute of Applied Economic and Social Research, The

University of Melbourne, Parkville Victoria, Australia;

Sundararajan: School of Psychology and Public Health,

La Trobe University, Bundoora, Victoria, Australia.

Corresponding author: Yong, email <jongsay@unimelb.

edu.au>. This research was funded by an NHMRC

Partnership Grant (Grant ID: 567217) and in collaboration

with the Victorian Department of Health and Human

Services (formerly the Victorian Department of Health).

The views expressed herein are those of the authors and

do not reflect the views of the Victorian Department of

Health and Human Services.

© 2020 The University of Melbourne, Melbourne Institute: Applied Economic & Social Research,

Faculty of Business and Economics

Published by John Wiley & Sons Australia, Ltd

drives the productivity variation across public

hospitals in Victoria. Productivity levels are

measured as the ratio of aggregate output to

aggregate input, both measured by a form of

quantity indices. The paper further investigates

the persistence of productivity and whether

productivity of public hospitals is related to

hospital characteristics, such as hospital size.

For the latter we use a standard methodology of

panel data regressions, while the investigation

of the persistence of productivity makes use of

models with lagged dependent variables.

The concept of productivity is designed to

capture the capability of production units (for

example, firms in an economy, public hospitals

in our context) in converting inputs into outputs.

In practice it has been used to measure

technological progress, efficiency and general

capability of production units in organising

production. There are many notions of produc-

tivity; common ones are labour productivity and

total factor productivity (TFP). The former

measures the level of aggregate output per unit

of labour inputs (for example, dollars of revenue

per full‐time equivalent (FTE) employee), while

TFP, also known as multifactor productivity,

broadens the measure of inputs to include all

factors of production. This paper will concen-

trate on the TFP measure.

It is important to distinguish between TFP

level and growth (Hulten 2001; Diewert and

Nakamura 2006, 2007). The latter is the rate of

change of the TFP level and its measurement

involves a standard for comparison in the form

of a reference period or reference production

unit. The choice of a reference point is thus an

important consideration in the computation of

TFP growth; estimates of productivity growth

will vary if different reference points are used.

Moreover, TFP growth provides little informa-

tion for understanding the relative performance

of different hospitals––hospitals with low TFP

level may grow at a fast rate simply because

they were starting at a low base.

With a population of about 5 million,

Victoria is the second largest state in

Australia by population size. The State has

about 120 public hospitals during the study

period. All citizens in Australia have uni-

versal access to care provided by public

hospitals,whicharejointlyfundedbythe

federal and state governments with approxi-

mately an equal share. State governments

aretaskedwithmanagingandregulating

public hospitals in their respective state.

Victoria is the first state in Australia to

adopt a prospective case mix funding system

for public hospitals (Duckett 1998). Under

this system, hospital admissions are classi-

fied into diagnostic related groups (DRGs)

and hospitals are funded based on the

number and type of patients treated ac-

cordingtoafixed payment schedule.

Besides in‐patient admissions, which con-

stitute the bulk of hospital services, other

major activities include outpatient services,

subacute and non‐acute services, teaching

and research and mental health services.

These latter services are funded on a cost‐

recovery or block grant basis in Victoria.

In 2011–2012, health care expenditures for

Victoria were approximately A$12 billion,

accounting for about a quarter of the State's

total operating expenses.

1

Of the State's total

health spending, public hospitals take the

largest share. Thus the productivity and

efficiency of public hospitals have important

budgetary implications. However, there have

been few studies on hospital productivity in

the published, as well as grey literature, and

little is known about public hospital produc-

tivity in Australia. This paper aims to fill that

gap. By relating hospital productivity to

observed characteristics of hospitals, it also

aims to provide a starting point for policy-

makers and analysts to explore strategies to

improve hospital productivity and efficiency.

We use Victorian hospital administrative data

for the five‐year period from 2007–08 to

2011–12 to construct quantity indices of output

and inputs for public hospitals. Well‐known

challenges in measuring outputs and inputs

of hospital care (for example, Triplett 2011;

Chansky, Garner and Raichoudhary 2013) are

discussed and addressed. We examine the trends

of both TFP level and growth of individual

hospitals, and find a high degree of variation in

TFP across hospitals. Moreover, TFP levels are

found to be highly persistent over time, but not

TFP growth. Large hospitals were found to

344 The Australian Economic Review September 2020

© 2020 The University of Melbourne, Melbourne Institute: Applied Economic & Social Research, Faculty of Business

and Economics