Media Briefings

Competition Can Be Fatal: Evidence From The Nhs Internal Market

  • Published Date: January 2008

NHS reforms introduced in the early 1990s by the last Conservative government to
increase competition – the so-called ‘internal market’ – resulted in higher death rates
for heart attack victims in hospitals subject to competition. But it also led to faster
falls in waiting lists in those same hospitals. These are the findings of new research
by Professor Carol Propper and colleagues, published in the January 2008 issue of
The Economic Journal.
The study finds that the effect of competition on quality was to decrease hospitals’
performance in areas where it was not tracked – fatalities following emergency
admission for heart attacks – but improved their relative performance on a welltracked
aspect of performance – waiting times.
Competition in the NHS internal market varied over time and space. Competition is
simply not possible in some parts of the country since population density means
there is only one hospital. Elsewhere, the potential for competition is high with a large
proportion of the population enjoying relatively easy access to a relatively large
number of hospitals.
The study derives the possibility of competition from distances between hospitals.
Around one third of hospitals are located in areas where no competition is possible.
The extent of competition also varied with policy diktat from the Department of
Health: competition was introduced in 1991, actively promoted until around 1995,
then downplayed and, finally, more actively discouraged from 1997 onwards.
Using these spatial and time differences, the study measures the impact of
competition by comparing death rates following emergency heart attack admissions
at hospitals that could experience competition with death rates at hospitals that could
not from 1991 to 1999. Deaths from heart attack admissions have been argued in the
US context to be the ‘canary in the mine’ – an overall indicator of poor clinical quality.
This study finds that deaths were higher where competition was greater. In addition,
the magnitude of the negative effect of competition on quality was not trivial: the
researchers estimate that the impact of competition on death rates more than
cancelled out the 7% per year fall in death rates that occurred over this period due to
technological progress.
But the research also finds that competition had positive effects as well. Waiting
times fell more in hospitals located in competitive areas. In addition, there is softer
evidence that exchange of information between GPs and hospitals improved
considerably as a consequence of its introduction.
The bottom line is that competition in the NHS during the 1990s appears to have
damaged quality where quality was not monitored but it simultaneously improved
performance in one area – waiting times – that was easy for patients and GPs
to observe.
The Labour government has re-introduced competition into the NHS. The message
from this research is that measures of quality need to be widely available in order for
competition to improve the quality of healthcare. The authors hope that increased
provision of information since the period covered by their study will mean that the
new competitive framework will help to improve health outcomes.
Notes for editors: ‘Competition and Quality: Evidence from the NHS Internal Market
1991-9’ by Carol Propper, Simon Burgess and Denise Gossage is published in the
January 2008 issue of The Economic Journal.
Carol Propper and Simon Burgess are at the Centre of Markets and Public
Organisation at the University of Bristol. Denise Gossage was at CMPO when the
paper was written.
For further information: contact Carol Propper on 0117 331 0706 or 0207 594
9291 (email:; Simon Burgess on 0117 331 0764 (email:; or Romesh Vaitilingam on 07768 661095 (email: