As the purpose of our paper was to open up debate on the direction of travel of contingent valuation (CV) research, we obviously welcome the contributions of Professor Adler and Baker et al. However, although Professor Adler seems willing to join us on the boat and help determine which course to steer, it appears that Baker et al. would prefer to hijack it! Our response therefore naturally concerns the comments by Baker et al.

Although we could no doubt have been clearer in places, we are surprised at the suggestion that we had ‘settled on some rather alarming solutions’ or reached any decisions to ‘abandon’ CV, since an open debate concerning ‘whether there is a role for CV in health economics, and also how it might reach its desired potential’ cannot be started in this manner. Within the context of wishing to engage in such a debate, we would therefore like to reflect on two substantive points made by Baker et al.

First, the nature of the literature used as evidence. We drew upon a number of systematic literature reviews that we have been involved in, all of which are widely available and report their methodology, which focus on empirical CV surveys over a 20-year period. The references provided by Baker et al., largely covering Donaldson’s body of work with various colleagues, have a methodological focus. We agree that such work has had little impact on the conduct of empirical CV work not undertaken, as Baker et al. state, by an ‘expert’. The issue is the degree to which one should count studies conducted by ‘experts’ and those not (notwithstanding the need to define an ‘expert’ of course). If this ‘non-expert’ research is where the growth is taking place, then this raises some important questions: How to judge an ‘expert’? Who should be conducting/publishing such research and whether there is a need for training and/or expert manuals such as those in the environmental literature? This adds a new dimension to the debate which we would welcome discussion of, since, alternatively, one could argue that restricting CV research to the ‘expert’ could also stifle innovation (see below) as the research field becomes dominated by a few researchers which limits the perspectives and debates permissible.

Second, there are three comments concerning guidelines: (i) stifling innovation; (ii) guidelines for reporting or conduct of CV and (iii) the influence of National Oceanic and Atmospheric Administration (NOAA) guidelines. First, stifling innovation was an argument advanced against the cost-utility reference case produced by NICE (National Institute for Health and Clinical Excellence), but clearly methodological research in that area continues unabated and one might argue that the comparability of cost utility analysis (CUA) studies has improved for end-users. Contrary to what Baker et al. fear, there is no reason that a sensible use of such a reference case/guidelines would not result in a win-win situation. Second, Baker et al. state that they are not clear about whether we were concerned with guidelines for reporting or conducting CV studies. Our suggestion was a minimum reporting set to ensure the results are more comparable and useful to end users, but clearly a next – although more major and contentious – step would be guidance about conduct; even ‘experts’ have to start somewhere! Third, and as we stated, of course the NOAA guidelines are inappropriate for health; our point is that authors of CV studies in health still refer (on an ad hoc basis) to these guidelines for justification. Indeed, one of us pointed this out many years ago (Smith, Reference Smith2000). Clearly, one could either systematically confirm or discredit their use, but the important point is that more systematic research is needed to better understand and guide CV research in health, in order to help doers and users understand what is likely to constitute a good quality study. We certainly do not accept that papers simply ‘fall foul of space’; we accept that our list in Box 1 could be prioritised, but when 25% of published studies have failed to report the elicitation method used and 23% failed to report the price year, both of which take very little space, we would argue there is considerable room for improvement in the reporting of CV studies.

In conclusion, we would reiterate that our paper was merely meant to act as a vehicle for opening debate on what we consider to be a very important topic in health economics; it did not seek to be prescriptive but rather suggestive of the issues facing CV and of potential areas for future research. Clearly a debate about other alternatives which researchers or policy makers might choose to take is needed, but we would not agree with Baker et al. that the status quo is fine, because clearly the good methodological work that has been done (including by Donaldson and colleagues) has not (yet?) been adopted in empirical applications. Rather, we would argue that the debate to be had centres on how vaguely right we have to be to be in a better position than being precisely wrong.