Tony Hockley, Department of Social Policy, LSE
The 2018 LSE Annual Behavioural Public Policy Seminar demonstrated the breadth of ways in which behavioural science is reshaping approaches to public policy; whether in developing a better understanding what matters in individual decisions, or measuring and assessing the individual and social outcomes of policies. The interdisciplinary nature of the journal to which this blog is attached is reflected in this longstanding annual LSE seminar, organised and hosted by journal editor Adam Oliver. Even within a relatively small faculty like the LSE such opportunities to come together across disciplines for discussion and debate around new research are still surprisingly rare.
Paul Frijters from the LSE Centre for Economic Performance (CEP) revealed the state of the art in the use of the “WellBe” as a useful unit for the measurement of life satisfaction. He set out an ambition that government would one day have a common costing approach for policy decisions across all departments and activities. Frijters described the development of the first micro-macro model, being built “one intervention at a time”; starting with an assessment of the value of cognitive behavioural therapy as a treatment for depression. The “Incredible Years” programme on child behaviour will the next target for WellBe analysis. Unlike the QALY, often used in healthcare decisions, the WellBe captures externalities, offering a much broader assessment of impacts.
By focusing on a single aspect of wellbeing, a sense of self-determination, Joan Costa-i-Font drew the seminar’s attention to recent “locus of control” research with colleagues in Australia and France. He argued that this framework can be used to explain positive selection in health insurance; why private health insurance (on top of social health insurance) in Australia and Germany is bought by those who least need it.
Locus of control is one of four dimensions of core self-evaluation, alongside neuroticism, self-efficacy, and self-esteem. The research finds that a strong internal locus of control is associated with a heightened sensitivity to future losses, high valuation of care quality, and action on prevention. According to the research it offers a clear answer to the paradox of positive selection in health insurance. It also suggests that financial incentives or compulsion may be ineffective in tackling the problems associated with positive selection, which might be better addressed within the choice architecture of insurance.
At the 2017 seminar Julian Le Grand had described how, within fixed-price public services populated by both “knights” and “knaves” (motivated respectively by intrinsic values or extrinsic rewards) competition and choice will foster service improvement. This raised questions around the risk of external rewards crowding out the knightly motivations, dependent upon the knightly-knavish composition of individual provider units. Matthew Skellern from the LSE Department of Health Policy and colleagues in the Department of Economics, have been using the English National Health Service Staff Survey to investigate hospital-level altruism amongst managers and workers, and the relationship to reported performance on Acute Myocardial Infarction (AMI) mortality rates. Responses from the NHS Staff Survey do show the complexity of motivations: Hospitals in London, which previous research suggested saw the highest gains in quality from rising competition, revealed a combination of high levels of manager altruism and low levels of worker altruism. The research does find, as Le Grand predicted, that non-altruistic hospitals respond to competition by increasing quality, but that the response in altruistic hospitals is attenuated (or negative if the level of altruism recorded is well above the mean). It certainly does not suggest that treating everyone as a utilitarian knave would be a wise policy for public service motivation.
The seminar concluded with philosopher Alex Voorhoeve making the case for “Egalitarianism under Ambiguity”. Voorhoeve argued that both equality of the value of people’s prospects and equality of outcomes should feature in public policy decisions involving ambiguity, and explored what would follow if this form of egalitarianism were combined with ambiguity aversion. On this assumption, just as individual-level ambiguity aversion depresses the value of individual prospects (revealed in the Ellsberg Paradox), so social-level ambiguity aversion depresses the value of social prospects. Voorhoeve analysed when the aims of reducing inequality and limiting ambiguity are congruent and when they conflict. In essence, they are congruent whenever, in ambiguous prospects, one person’s gain is another’s loss; they are in conflict whenever, in order to prevent inequality, one would have to give everyone an ambiguous prospect in which all “sink or swim” together. He also demonstrated that ambiguity aversion renders a wide range of distributive views incompatible with the Pareto principle applied to ambiguous prospects and argued that this gives us reason to dispense with the Pareto principle applied to prospects.
Voorhoeve illustrated the argument’s practical implications by considering decision-making in the UK in relation to Swine Flu in 2009. Referring back to the UK decision to invest in swine flu vaccination Adam Oliver suggested that this particular decision may have been taken by politicians to minimise the potential for regret; they were acting as individual decisionmakers, whereas analysis within the seminar revealed a role for action as a “social decisionmaker”, with much broader consideration than individual utility.