Title
Eliciting Laundered Preferences in the Context of Health and Wellbeing
Abstract
Most philosophers agree that only the satisfaction of laundered preferences, not the satisfaction of actual preferences, can be treated as a reliable indicator of wellbeing. Actual preferences often reflect cognitive error on the part of the agent, for example, and laundered preferences seek to correct for this. However, relatively little attention is paid to how laundering should proceed and to what desiderata laundering and its effects should satisfy. The proposed research intends to fill this gap. Three desiderata for laundering are identified in the proposal that the PhD will further develop, elaborate and refine together with a multidisciplinary supervisory team. Furthermore, practical implications of these refined desiderata will be explored. First, their consequences for experimental design will be investigated and, second, a few large scale experiments will be designed and conducted.
Key words
Laundered preferences, preference elicitation, wellbeing, health care
Description of the Proposed Research
Preferences and preference measurement are at the center of Action line 3, which deals with developing a (preference based) wellbeing measure for use in economic evaluations, and with patient preferences in the context of treatment choices. The underlying assumption behind the use of preferences is that preference satisfaction can be seen as an informative indicator of (partial or general) wellbeing. It has been suggested that such measurement can respect sovereignty (Bernheim 2016) and, more than other conceptions of wellbeing, such as the objectivist and hedonist conceptions (Parfit, 1984), respects the autonomy of agents.
For valuing health and wellbeing states, as well as (aspects of) health care interventions or distributional concerns, preference measurement is becoming increasingly important in the context of health care decision-making. These issues are not only central to Action line 3, but are also highly relevant from both a societal and a scientific perspective. The reliance on preference measurement stresses the importance of the normative underpinning of doing so, and of efforts to improve preference measurement for normative use in health care decision-making.
At the same time there are also good reasons to believe that the degree to which agents satisfy their preferences is not always a reliable indicator of their wellbeing (or social welfare). Empirical research strongly suggests that preferences exhibit various features that seem to compromise their central normative status in welfare assessment. There is massive evidence that human decision-making is prone to many behavioural heuristics and biases (Tversky and Kahneman 1974). If some preference of an agent is based on incorrect or incomplete information, or based on incorrect inferences on the basis of correct information, for example, it is hard to see how the satisfaction of that preference would contribute to the agent’s wellbeing. More generally, preferences do not seem to exhibit the consistency and stability that is required for welfare assessment. Agents that have particular preferences in the one context might have (and often actually do have) other (and sometimes reversed or opposite) preferences in other contexts. Preferences thus typically display a degree of context-sensitivity that undermines their usefulness as a normative benchmark in welfare assessment. Finally, agents might have antisocial preferences (such as malicious preferences and envy). Even though we might think that satisfying antisocial preferences contributes to the agent’s wellbeing, we think that health care policy (or, for that matter, any sort of policy) should not attend to such preferences.
Several philosophers have sought to overcome these weaknesses of a preference-based conception of wellbeing by proposing that laundered (also sometimes called ideal, purified) preferences, rather than actual preferences, should form the basis of welfare assessment (cf. Harsanyi, 1977, Goodin, 1986, Gauthier 1986, Griffin, 1986, Arneson 1990, Brandt, 1998). Indeed, it is no exaggeration to say that there is a consensus among philosophers that only laundered preferences could possibly serve as a legitimate basis for welfare assessment. Recently, Hausman (2011) argued that even though the preference-satisfaction conception of wellbeing is flawed, the degree to which the agents’ preferences are satisfied might nevertheless serve as a reliable indicator of the agents’ wellbeing. Hausman stresses that this is the case only if certain conditions are met. The conditions specified by Hausman are related primarily to the possibly “distorting” factors in actual preferences that laundering is supposed to dispense of, having mainly to do with cognitive limitations of agents. If these conditions are met, laundered preferences supposedly align better with personal wellbeing values (Haybron & Tiberius, 2015), the aspects of life that people believe to be constitutive to their overall quality of life.
Despite the broad agreement on the importance of laundering in philosophy, the issues of how laundering should proceed and what desiderata laundered preferences should ideally satisfy have received relatively little attention. Considering the closely related discussion in the field of economics around the ethics of nudging (Blumenthal-Barby & Burroughs 2012, Sunstein 2017, Sugden 2018), collaboration between philosophers and economists is essential (Angner, 2013). The research of the PhD student will focus on these issues in the context of health and wellbeing, and emphasize the issue of how we can check empirically whether the desiderata are satisfied to a reasonable extent. What follows below gives a first outline of the research conducted in this project. The PhD student, together with the multidisciplinary supervisory team, will further develop, elaborate and refine the points made.
Philosophers have not only highlighted potential blessings of laundering. Some of them also warned for possible problems and pitfalls. We think that in the present context two of them merit special attention. The first concerns the possibility that in the process of laundering preferences, those doing the laundering (the “launderers”) attribute to people (e.g., patients, decision makers or general public) or steer them in the direction of preferences that these people in their opinion ideally ought to have. Rather than reflecting people’s true, undistorted preferences, laundered preferences would then reflect the launderers’ preferred preferences. We call this the problem of paternalistic projection. Obviously, this problem can only be solved if things like manipulation and social pressure in the laundering process are kept to a minimum.
Second, several commentators have pointed out that even if we were able to find out what people would prefer if they had full information, reasoned correctly on the basis of that and the like, they do not necessarily identify with these laundered preferences also outside the context in which preferences are laundered. Outside the confines of the latter context, people might no longer recognize laundered preferences as their own preferences, or might no longer feel the moral pull of these preferences (so that they no longer feel committed to them.) Call this the problem of possible alienation. This problem is arguably overcome if people reflectively endorse their laundered preferences also outside the context in which their preferences have been laundered (cf. Anderson 1993).
In short, ideally the laundered preferences that we want to get at meet the following desiderata:
- They are free from cognitive error or lapses of instrumental rationality;
- They are affected as little as possible by manipulation by launderers, group pressure and the like;
- They are reflectively endorsed by people.
The next step will be to investigate how these desiderata (or some further refinements of them) can be implemented in protocols for empirical research. What procedures, in the process of eliciting laundered preferences, should be in place so that these desiderata are met to the fullest possible degree? In answering this research question, we build on earlier attempts to arrive at better-informed, better-reasoned and more consistent and stable preferences, such as the “Burgerforum: Keuzes in de Zorg” (2018) and “Time to Think” (Veldwijk, Johansson, Donkers, De Bekker-Grob 2018) initiatives in which members of the supervisory team for this PhD project were involved. These examples also show the relevance of these topics from both a societal and scientific angle. Improved ways of deriving preferences for health and wellbeing states, also considering distributions of health and health care, can have a profound impact on research as well as on health care decision-making.
In this project we will pay special attention to what additional procedures and control moments have to be in place in order to meet the second and third desiderata mentioned above to the largest possible degree. This will involve experimental designs in which we use different control groups and contexts. By providing different mechanisms to derive preferences (e.g., after a deliberative process or not) we can observe the impact on preferences and the desiderata. We will supplement measurement of behavioural effects with questionnaires and will use repeated measurements in order to investigate endorsement and stability of laundered preferences. Specifically with respect to the second desideratum, we think that some things can be learned from the extant literature on good practice with focus groups (Fern 2001; Onwuegbuzie et al. 2009).
In a final step, larger scale experiments will be designed and conducted. We are thinking here of implementing different forms of laundering (such as the use of educational tools in order to meet the first desideratum) in different groups and of observing what are their effects. This will first be done with student populations before turning to general population samples. In order to strengthen the relevance of this study for action line 3 and the overall Erasmus Initiative, we aim to align the preferences to be studied, and the context in which they will be studied, with choices made in action line 3.
Collaboration: Conrad Heilmann (Faculty of Philosophy, EIPE), Werner Brouwer (ESHPM), and Job van Exel and Kirsten Rohde (ESE) will all be involved in the supervisory board. Esther de Bekker-Grob (ESHPM) will advise on what preference-elicitation methods to use.
We will build on earlier successful collaborations, which combined philosophical and health economic insights. Examples include (successfully completed) joint supervision of PhD students Sofie Wouters (cf. Wouters et al. 2017) and Willem van der Deijl (van der Deijl 2017).
Knowledge utilization
The findings of this study are relevant for health and wellbeing state valuations in general, and the research that is conducted in action line 3 more in specific. After all, laundering preferences contributes to more appropriate measurement of the benefits of health care and therewith to smarter choices in health care. In addition, laundering preferences may have important implications in the context of eliciting societal preferences for treatments, in particular difficult choices about life extending or end-of-life treatments. As such, more insight into how laundering should proceed and to what desiderata laundering and its effects should satisfy, is crucial for health care (funding) decisions that wish to align with societal preferences, but also for initiatives that aim to promote citizen participation in policy development, choices and implementation. In other words, the findings of this project have relevance for health care research, policy and practice.
The knowledge generated in this project will be utilized along the same lines. The methods and findings will be reported in academic papers, and communicated to a broader non-academic audience through publications in field and policy journals. The methods and findings will also be discussed with academia at relevant conferences and with policy and practice (with a focus on national policy bodies such as ZiNL) during at least one workshop.
This project will further strengthen the collaboration between the faculty of philosophy, ESHPM and ESE, and also link the (germinating) Erasmus Values and Economics Network (EVEN), a joint initiative of these faculties/schools, with the EI Smarter Choices for Better Health.
Finally, the people involved in the project can Integrate the findings in their teaching, for example in the ESE minor ‘quality of life and happiness’ and the ESHPM minor (in development) ‘smarter choices for better health’.
Description of the Hosting Department
The Faculty of Philosophy of Erasmus University Rotterdam provides leading international research and education, with theoretical philosophy, continental philosophy, practical philosophy and the history of philosophy as central themes. The Faculty offers a three year bachelor programme in philosophy, a one year master programme in philosophy and a two year research master programme in philosophy and economics. Moreover, the Faculty offers a double degree (honours) undergraduate programme for students from other faculties at the Erasmus University. In total the Faculty has about 700 students and 25 academic staff members. Within the Faculty of Philosophy, EIPE (Erasmus Institute for Philosophy and Economics) stands out as an interdisciplinary Institute at the interface of philosophy and economics. EIPE does not only bring the two disciplines of philosophy and economics together, it also actively seeks to connect philosophers with economists. Cases in point are regular interactions and collaboration especially with researchers from ESE and ESHPM in EHERO, behavioural economics and health economics.
Appendix
References
Angner, E. (2013) Happiness and Well-Being. Res Philosophica 90(3): 365-385
Anderson, E. (1993) Value in Ethics and Economics, Harvard University Press.
Arneson, R. (1990) Liberalism, distributive subjectivism, and equal opportunity for welfare, Philosophy & Public Affairs 19: 158-194.
Bernheim, B.D. (2016) The good, the bad, and the ugly: A unified approach to Behavioral Welfare Economics, Journal of Cost-Benefit Analysis 7(1):12–68.
Blumenthal-Barby, J.S. & H. Burroughs (2012) Seeking better health care outcomes: The ethics of using the “nudge”, The American Journal of Bioethics 12(2): 1-10.
Brandt, R. (1998) A Theory of the Good and the Right, Prometheus Books
Fern, E.F. (2001) Advanced Focus Group Research, SAGE Publications.
Gauthier, D. (1986) Morals by Agreement, OUP.
Goodin, R.E. (1986) Laundering Preferences, in Elster, J. (ed.) Foundations of
Social Choice Theory, Cambridge, Cambridge University Press: 75-102
Griffin, J. (1986) Well-being: Its meaning, measurement and moral importance, Clarendon Press.
Harsanyi, J. (1977) Morality and the theory of rational behaviour, Social Research, 44(4).
Hausman, D. (2011) Preferences, Value, Choice, and Welfare, CUP.
Haybron DM, V. Tiberius (2015) Well-Being Policy: What Standard of Well-Being? Journal of the American Philosophical Association 1(4): 712-733.
Onwuegbuzie, A.J., W.B. Dickinson and N.L. Leech (2009) A Qualitative Framework for Collecting and Analyzing Data in Focus Group Research, International Journal of Qualitative Methods 8(3), 1-21.
Parfit, D. (1984) Reasons and Persons, OUP.
Sugden, R. (2018) ‘Better off, as judged by themselves’: a reply to Cass Sunstein, International Review of Economics 65(1): 9-13.
Sunstein, C. (2017) ‘Better off, as judged by themselves’: a comment on evaluating nudges. International Review of Economics 65(1): 1-8.
Tversky, A. and D. Kahneman (1974) Judgment under uncertainty: heuristics and biases, Science 185(4157): 1124-1131.
Van der Deijl, W. The Measurement of Wellbeing in Economics: Philosophical explorations. PhD thesis. Rotterdam: Erasmus University Rotterdam (retrieved from: https://repub.eur.nl/pub/102422/thesis-whole-september-27.pdf).
Veldwijk, J, J.V. Johansson, B. Donkers, E. de Bekker-Grob (2018) Mimicking real life decision-making in health: allowing respondents time-to-think in a discrete choice experiment (unpublished first draft)
Wouters, S., N.J.A. van Exel, K.I.M. Rohde, J.J. Vromen & W.B.F. Brouwer (2017) Acceptable health and priority weighting: discussing a reference-level approach using sufficientarian reasoning. Social Science & Medicine, 181 (May): 158-167.