Identity Decisions and Welfare

Experimental and behavioural studies have shown that people consistently deviate from the standard model of economic rationality in various ways that range from altruistic behaviour to preference reversal and proneness to cognitive failures (see e.g. Kahneman 2003). This obviously has theoretical implications and economists are trying to respond to those results by developing models on the basis of behavioural game theory, evolutionary game theory, and social interaction models or other adaptations of the standard utility maximising model (e.g. Akerlof and Kranton 2000, Benabou and Tirole 2002). These results and models are claimed to represent a more realistic version of the economic agent, taking account of the fact that people are social beings and that they are endowed with cognitive capacities that make them adapt to specific social circumstances. In particular, cognitive capacities such as self-signalling, self-confidence, but also socially inspired self-images have been shown to motivate and influence the individuals` behaviour. However, while this research improves the realisticness of the characterisation of the economic agent, it had so far not much to say about welfare evaluation. Indeed, standard economic thinking assumes unchanging preferences such that the satisfaction of people`s preferences can be compared at two different moments of time. These newly introduced cognitive features of the individual put a question mark not only on the standardly assumed "consumer sovereignty" of the economic agent, but also on the stability of preferences. Welfare considerations therefore need a substantial re-think. My research will continue in the vain of recent research of behavioural economists and proposes to study similar questions but from a different point of view. Indeed, while I am inspired by results of behavioural and experimental economics I do also think that these approaches leave too much leeway to uncontrolled cognitive features even though it has also been shown that people can develop a substantial amount of self-reflection and self-knowledge that give rise to more sophisticated decision process than currently assumed. I will thus follow a route that is reminiscent of what Tapas Majumdar (1980) and Prasanta Pattanaik (1980) called "The Rationality of Changing Choice" and propose to do two things. First I will continue developing the "personal identity" approach I have been elaborating during and after my doctoral studies and will analyse in particular how different and changing self-concepts will influence people`s behaviour. Second, I will study the consequences of this personal identity approach on welfare evaluations. The "identity approach" is the result of an interdisciplinary approach, considering economic rational choice, philosophical accounts of personal identity and philosophy of mind, as well as psychological studies on self and identity and my future research will continue to explore their common grounds. The identity conception that I started to develop assumes that people have a particular self-image of themselves. This self-image can be thought of as mental images of themselves in particular future situations (represented, for example in terms of a preference ordering) that motivate people to engage in those actions that bring them (or their "actual self", which is also a preference ordering) closer to the realisation of such self-image. However, we cannot assume that these self- images will be stable over time. They will necessarily change depending on people`s achievements, experiences or external influences. Hence, self-images will change - and so will people`s choices. The economic agent can thus be represented on the basis of changing preferences, but with change that is rationally motivated. This process of change will need further clarification and my aim is to develop an axiomatic of change that helps formalising "identity decisions". As for the second part of my research, my claim is that if people`s changes in preferences can (also) be understood from a rational point of view then we do not (always) face problems of welfare evaluation as in those situations where the underlying basis of comparison changes in an ad hoc manner. If an individual`s decision to change can be followed through time, then it will also be possible to develop an value-index indicating under what conditions the person`s welfare has increased and when it hasn`t. In particular, to see people`s choices as identity decisions based on the attempt to realise themselves, two factors that influence welfare can be distinguished: first the opportunity to become "who" one would like to be, and second the extent to which one has achieved to be the person one wished to be. As such this is related to research on freedom of choice, which obviously constitutes a crucial element in people`s identity decisions.

One of the key methodological tools in economics is the model. In a model, the economist starts with a few assumptions about how the world works and then changes some parameters in order to see what the consequences would be of such changes. This allows making predictions, and will help to make informed policy recommendations. One of these assumptions concerns peoples behaviour. In most cases, the economist assumes that people are rational, and this usually means in the economic context that they maximise their utility. In other words people, (named economic agents in a model) are endowed with preferences over different goods, services or outcomes, which are supposed to be stable over time, and try to choose what they prefer most, given their income or other constraints. Those stable preferences are often taken to be the identity of an economic agent, i.e. people are what they prefer. This also implies that economic identity is unchanging over time. Preferences are also the basis for welfare evaluations. A persons welfare is higher the more of their preferences are satisfied. Assumptions about the world are necessarily simplistic, it is argued, because if not, a model would become intractable. However, such argument is untenable if simplistic assumptions are false or hardly justifiable. This is where my research comes in. In particular, I was considering the assumption of unchanging preferences over time. My main hypothesis was that peoples preferences change over time. This hypothesis has wide reaching consequences, not only for the standard model of decision making, but also for welfare evaluations. The questions of my past research was then how best to integrate this assumption into economic theory, how to justify it, and what consequences this would have for the behaviour of economic agents and for the evaluation of welfare. To do so, I studied among others the psychological literature of internal conflict, which may arise when people experience several, at times even competing motivations or reasons to do certain things. On the basis of this, I was able to develop together with my co-author Ritxar Arlegi a model of choice that explains the formation of preferences, rather than assuming the unchanging existence of them from the outset. Preferences are then stable whenever the person has solved a conflict between different motivations that may push them to do or to want certain things, but otherwise people would make choices, which are not representative of their preferences, or which may be, as economists also say, inconsistent. This has of course consequences on what can be considered an economic agents identity, which cannot be any longer equalised with a given set of preferences (or consistent choices). I now rather propose to see identity as the set of frames within which people make choices, even inconsistent ones. But choices cannot violate the minimal rationality conditions these frames would imply to maintain a persons identity. Welfare evaluations have to be adapted accordingly to this new model and can for example take account of the impact inner conflict has on wellbeing.