The analysis of human behaviour across different contexts is at the core of the studies of psychologists and sociologists. On the one hand, psychologists study in great detail how cognitive and emotional perceptions shape our decisions through thoughts and feelings in almost every aspect of our daily lives. Similarly, sociologists analyse how the influence of peer groups, friends and foes, family, and society at large influence individuals via different channels.
In the field of economics, on the other hand, the main focus of studies on individual behaviours have for long been focused mainly on economic decisions, such as. what and how much to buy or consume, one’s occupation and working choice, but also decisions about education, choice of living locations and many others. However, the prediction capability of standard economic theory – and in particular of the dominant neoclassical model – is uneven, to say the least. In fact, since the 1970s, many researchers such as Amos Tversky, Daniel Kahneman, George Ainslie, or Matthew Rabin tried to improve the predictions capability of economic models by developing frameworks allowing for more flexible and realistic considerations of individuals’ behaviour. Behavioural economics has since then picked up an increasing growing interest and enriched the discussions between researchers of many different fields: from economists to sociologists and psychologists, including very recently also neuroscientists.
Standard economic model
The traditional standard economic starting point of individual decision making could be roughly summarized with the welfare or “utility” value of individuals computed as the discounted, expected value of a utility function of an individual. This function depends only on one or very few characteristics, such as the amount of a good consumed or leisure time – while taking into account all available information. Assuming that the individual takes decisions by maximizing this utility allows inferring the predicted decision of individuals. Moreover, this assumption implies that the obtained utility can be used as some sort of measure of well-being of individuals. These two implications serve two central purposes in public economics: the prediction of reactions to policy interventions and the assessment of welfare losses and benefits of such policies.
The assumptions implied in this standard model are important and can be broadly classified in four categories.
- Firstly, considering an individual through a single utility function over a payoff (often in monetary terms) of only himself is actually restrictive. We can obviously think of taking into account more criteria (such as the distribution across individuals, environmental quality, social factors, etc) or assume that individuals’ utility might be influenced by others behaviour or characteristics (for instance, comparing income to others).
- Secondly, the fact that many decisions have to be taken under considerable uncertainty about the future could be considered in many different ways than by computing the simple expected value of this utility.
- Thirdly, estimating utility today and in the future (or in the past) also admits many ways of comparing e.g., pleasure from a vacation today or in ten years, whereas the classical model assumes standard exponential discounting over time using a constant time discount rate.
- Fourth, the classical model assumes that this utility is maximized by the individual taking into account all available information.
However, the individual might not be aware of all information available (limited information) or not be perfectly rational when making his choice (bounded rationality). Also, one might have two different utility concepts, and an individual can implicitly be described by: considering a certain utility or welfare maximization concept when making decisions (decision utility), but use a different utility specification when evaluating the decision’s consequences (experienced utility). Classical examples of this dichotomy are decisions that have costs and benefits at very different points in time. For instance, it is perfectly rational to smoke a (cheap and supposedly enjoyable) cigarette today facing an uncertain, but potentially very harmful health suffering in the future, so that in the long run individuals might prefer to not have smoked at all. Such seemingly irrational behaviour can be rationalized by considering two different utility concepts with different discount rates and/or ways how risks are taken into account. All these simplifications in the standard model make decisions of individuals tractable to be studied with mathematical methods, but they bring necessary and significant limitations to their applicability.
In order to achieve a more realistic model of human decision making and well-being, Behavioural economics starts from here, aiming to depart from the classical model to extend it in the four main categories described. Behavioural economic analyses developed in recent years can thereby serve several purposes: firstly, they allow us to get a better understanding and explaining of the actions of individuals – where the traditional simplistic utility maximization based on perfect rationalism fails. For instance, the domain of happiness research has focussed on finding the determinants of individual well-being or happiness. Examples of these findings include stronger than expected negative effects of unemployment, a lower happiness increase when income is above a certain level, or high costs for spending too much time commuting. Such results have important implications for optimal policies and therefore important consequences for what optimal policies should look like.Behavioural economics allows re-evaluating the effectiveness of existing policies when facing individuals who have different preferences than standard ones. Also, by empirically estimating preference parameters in certain domains, they allow to design new policy measures that take into account people’s behavioural reactions and thus allow a better targeting of policies. By going beyond the aforementioned classical standard utility model, such extensions can thus improve both the predictive power of economic models to improve public policies and avoid mispredictions and the measurement of individuals’ welfare changes resulting from such policies.
Recent works in economics has picked up established theories of psychology stating that individual’s utility not only depends on his or her own choice. Rather, this utility might also depend on how much other people consume, in the society at large or in one’s peer group, with respect to some forms of social norms or even on a created self-image. Researchers including Jean Tirole, Raj Chetty and Roland Benabou have in recent years aimed elaborating models of individual decision making that account for such potential impacts on individuals’ preferences, even more interlinking economic microeconomic modelling with behavioural traits.
Examples of how such mechanisms work are from Bénabou & Tirole, who argue that moral behaviour can be conceived as a result of individuals formation of their “identity”. That is, people are not fully aware of their own motivation or “morale” and their behaviour tells them something about their true self. In economic terms, morale or more specific behaviour can be seen as “investment in identity”. Such seemingly overly rationale has been indeed related to findings in the Psychology field.
Similarly, one can observe how societal rules and values impact individuals’ behaviour. In particular, “hard” societal norms such as laws or financial sanctions or rewards, but also “soft” norms like honor or social stigma affect how individuals behave. From a positive perspective, taking into account people’s’ reactions to social norms and formal laws or norms can explain the existence of certain legal norms and the evolvement of social norms. Rather than focusing on single and separated issues (donating to charity, waste recycling, trying to saving energy or water, organic food consumption etc.), the interlinkages between different domains of prosocial or “good” behaviour have been analysed from a conceptual perspective. Theories suggest that one could expect that being a good citizen in one domain (say, as a donor to charity) could be used as compensating for “bad” behaviour in other issues (Gneezy et al., 2014). Equally conceivable, individuals could perceive their image value from all behaviours or information about the “benefit” of good-doing, suggesting a more similar pattern across different domains.
Also in the context of decisions related to climate change, environmental concerns, or energy consumption, several aspects have in recent years been studied from a behavioural perspective. For instance, the unexpectedly low success of subsidizing energy saving technologies or appliances has puzzled researchers. However, the role of salience of information, unawareness of savings potentials, or implicit high discount rates of future energy savings have been suggested as explanations of individuals’ behaviour. Recycling habits and energy savings behaviour similarly have been analysed suggesting that social norms and image concerns are additional “arguments” for one’s utility function.
For policy makers, understanding better the psychological and economic reasoning of individuals when making decisions could therefore contribute to more successful policies, for instance in the case of taxes, rebates or subsidies for supporting more efficient energy devices, for switching from heavily polluting cars to more efficient ones or public transportation, or for increasing recycling shares.
It is noteworthy that the rationale for or against such policies in general stems from other considerations, notably health impacts of air pollution, or the greenhouse gas effect or other externalities. The main argument here is that if such policies are considered desirable, an analysis based on more realistic, behavioural economic approaches allows not only to design more effective policies, but typically also leads to increases in individual welfare, and finally allows for a better measurement of the social benefits of different policy options.