Media Briefings

‘Willingness To Pay’: How To Get The Public To Put An Accurate Value On Healthcare Services

  • Published Date: January 2008

How can policy-makers get people to tell them how much they value things like
better health or an improved environment? New experimental research by Professor
Magnus Johannesson and colleagues has found a simple solution in the form of a
follow-up to hypothetical questions about people’s ‘willingness-to-pay’.
Their study, which is published in the January 2008 issue of The Economic Journal,
will be especially valuable for governments in determining which drugs or other
healthcare services to provide.
Governments need to be able to judge the value of certain goods in order to decide
whether they are worth spending taxpayers’ money on. These can range from
deciding whether to provide a drug through the NHS to deciding whether to invest in
improving air quality.
One way of estimating this value is to ask individuals how much they are willing to
pay for a certain outcome. The drawback of these hypothetical questions is that they
greatly overestimate the real willingness to pay. This new study demonstrates that
the willingness to pay can be accurately estimated by adding a simple follow-up
question about the certainty of hypothetical responses.
These results come from a field experiment in which two groups were asked how
much they would be prepared to pay for diabetes treatment:
• One group of patients (the ‘real’ group) could buy a diabetes management
programme delivered at their local pharmacy at a price of $15, $40, or $80.
• In the second (‘hypothetical’) group, patients answered a hypothetical
question about buying the diabetes management programme at those same
prices. In this second group, patients who answered ‘yes’ were asked if they
were ‘definitely sure’ or ‘probably sure’ of their response.
It turned out that about twice as many diabetes patients answered ‘yes’ in the
hypothetical group than in the real group. The value of diabetes treatment would
have been substantially overestimated inferred from the hypothetical group’s
first answer.
But the number of people in the hypothetical group who not only answered ‘yes’ but
also that they were ‘definitely sure’ corresponded closely to the number of buyers in
the real group.
By including the simple follow-up question about certainty, the overestimation in
hypothetical willingness to pay questions can therefore be adjusted for in a
simple way.
Notes for editors: ‘Eliciting Willingness to Pay without Bias: Evidence from a Field
Experiment’ by Karen Blumenschein, Glenn Blomquist, Magnus Johannesson,
Nancy Horn and Patricia Freeman is published in the January 2008 issue of The
Economic Journal
Magnus Johannesson is at the Stockholm School of Economics. Karen
Blumenschein, Glenn Blomquist and Patricia Freeman are at the University of
Kentucky. Nancy Horn and Patricia Freeman are of the American Pharmacy
Services Corporation.
For further information: contact Magnus Johannesson on +46 8 736 9443 (email:; Karen Blumenschein on +1 859 257 5778 (email:; Glenn Blomquist on +1 859 257 3924 (email:; Patricia Freeman on +1 859 323 1381 (email:; or Romesh Vaitilingam on 07768-661095 (email: