Monday, October 21, 2013

Lecture on Identity, Motivation and Incentives

I am currently giving a set of lectures as part of a module "Behavioural Economic: Concepts and Theories" in Stirling. I am posting brief informal summaries of some of these lectures on the blog to generate discussion. Thanks to Mark Egan for a lot of help in putting these together online. 

The section on Identity, Motivation and Incentives contains a lot of interlocking aspects. Many of these topics are very heavily connected to the idea that emotions influence behaviour and peoples responses to outcomes and we will revisit some of these ideas in the next topic - also many of them are connected to other topics in the course such as rationality more broadly and well-being, which we will look at also. 

1. Introduction

The basic idea behind the lecture is that self-interest is generally conceived as the main motivation for different types of behaviour such as saving, investing, working and so on but that, increasingly, behavioural economics is examining how other motivations such as altruism and the desire to conform might influence economic behaviour and outcomes. The first point we make in the lecture is that self-interest is an "add-on" to rationality. Technically, it is quite possible to be rational, as outlined in the first few lectures, and also be motivated by concern for others and so on. However, there are a number of points during the lecture where wider influences on behaviour clash with the idea that people are rational, as defined by having stable preferences and making consistent choices. A way of thinking about this topic is to ask some questions like: do I care about other people outside my family so much that I would genuinely give up things to help them? Do I change my core preferences as those around me change theirs? Would I be independent in situations where I was asked to do something wrong by someone in a position of authority?


2. The influence of peers and groups
The first aspect of motivation that goes beyond self-interest is the idea of herding and peer effects. There is a high correlation between an individual's behaviour in any economic domain and the behaviour of their peer group. We looked at the very famous "Dartmouth paper" that showed that the pre-college characteristics of flatmates that students were randomly assigned to live with had big effects on their behaviour. If you are randomly assigned to someone who drank before coming to college, you are more likely to drink during college - similarly, you are more likely to study if you are assigned to someone who did well at school. These results raise questions about the idea of fully stable economic preferences. 

Fig 1. The set-up
Moving on from this, we examined the idea that "group processes" may influence behaviour. The most striking example of this is Milgram's 'Behavioural Study of Obedience'. During the most famous of these experiments, Stanley Milgrim had 40 male participants between the ages of 20 and 50 play the role of 'teacher' to the 'learner' in the adjacent room. In the room with the teacher was a stern looking experimenter wearing an official looking coat (Fig 1). The task of the teacher was to administer increasingly powerful electric shocks to the learner whenever he made a mistake on the ostensible memory task he was working on - in reality the learner was a confederate working with the experimenter. There were no real electric shocks being administered, although the learner was trained to react to them as if they were real.

Anticipating that many of the participants would become uncomfortable as they heard increasing pained screams from the next room, the experimenters were allowed to prod them. In the case of objections, the experimenter told the teacher "Please continue". If objections continued, they would reply in the following order: "The experiment requires that you continue", followed by "It is absolutely essential that you continue" and lastly "You have no other choice, you must go on".

Fig 2. The results
Before running the experiment, Milgram polled 40 psychiatrists who agreed that "only 0.1% of the subjects would administer the highest shock on the board" - essentially it was thought that only a psychopath would continue all the way to the end where the voltage level was marked XXX and clearly hazardous. In reality (Fig 2), almost 2/3rds of participants went all the way to the end, even when some of them were clearly uncomfortable with the process.

It seems, from a long line of psychological research, that people will do extreme things well beyond what they would predict they would if they are told to do so by someone in a position of authority. In terms of historical context this study came out in the same year as Eichmann in Jerusalem, which popularized the concept of the 'banality of evil'.

As an addition factor, conformity to norms and reaction to persuasion may also have complex effects on individual behaviour. The Zimbardo prison experiment is a classic example of how randomly assigned social categories can have strong effects on people's actions.


3. Motivation to Behave in Group Situations

Fig 3. The Ultimatum Game
We focus on complex social and economic situations, as are represented in the Prisoner's Dilemma and Ultimatum (Fig 3) bargaining games which are two of the most famous experiments in economics.

The key paper for this topic is the paper by Ernst Fehr on Trust. While this paper does not discuss every aspect of how people behave in group situations, it serves as a good example of how this works and is sufficient to use to explain these concepts. Fehr provides a very useful working definition of trust and explains how trust can help to solve social problems that mirror those of the prisoner dilemma. He argues that trust, in some sense, involves processing risk but that it involves more than just risk preferences. Specifically, trust contains elements of an emotional engagement with others and that "betrayal aversion" can lead people to feel a lot worse if they lose in a game involving trust than simply if they lose a gamble. This is a key insight for behaviour economics; namely that one solution to cooperative games is that people trust each other and reach the pareto-optimal solution.

Fehr argues that countries with better social institutions arguably grow better and have better all-round outcomes, basically because in such countries it is easier to do business and interact in economic and social contexts because there is a basic degree of confidence in other people. We have spoken a lot about the difference between libertarianism and paternalism. This is another concept that we will talk a lot about - namely that markets are not perfect and a pure libertarian solution has many flaws but the state is not the only solution. The basic idea is that many economic problems are solved not by contracts but by social norms and implicit cooperation that is regulated not by laws or by fines but rather by complex social emotions such as trust. Trust is the example you should focus on, but in the next section of the lecture we will look at other examples of complex emotions and motivations that regulate economic behaviour in different ways. The basic idea is still the same.

4. Other Emotions & Economic Behaviour

We will look at this topic in more depth in the Emotion lecture. One consequence of relaxing the assumption of pure self-interest as a driver and looking at a broader range of emotions is that we open up a number of facets of human economic behaviour and attitudes that may have seemed outside of the realm of economics beforehand. As discussed above, trust and the emotions surrounding it are involved in some of the most important non-financial motivations of behaviour - but there are many other different types of motivations and emotions that arguably play a role in regulating complex economic situations involving groups. A few of them are discussed below:

(i) Discrimination and Hate: One consequence of being in different groups is that we may form a preference for our group over other groups. I referred in the lecture to a series of experiments that show that women and ethnic minorities are less likely to get called back to job interviews compared to whites even when the characteristics of each group have been randomly assigned on the CVs. Furthermore, we know that many people dislike people not of their own ethnicity and that many people favour restrictions in trade and migration. The real question (and one we will speak about in the Emotion lecture also) is whether such preferences are actually just irrational hangovers from the fact that we are basically animals with faulty cognitive equipment or whether they are rational preferences (albeit selfish preferences). For example, I may oppose globalization because of an irrational fear of foreigners but I may also oppose it because my industry has lots of nice protections from competition that would be eroded if restrictions were lifted. The Ku Klux Klan may have outwardly behaved in very silly and deplorable ways but as well as spreading hate it is arguable that their members may have been using the situation to improve their economic position.

(ii) Abhorrence: We discussed the idea that we may have motivations beyond just self-interest. For example, we may have strong beliefs that some markets simply should not exist. Al Roth, who won the Nobel Prize partly for his work on market design in organ donation, has a paper called "Repugnance as a constraint on markets" that addresses this in the context of whether it should be legal for a person to sell their own organs. 

(iii) Reference Effects: Another consequence of being in groups is that we evaluate ourselves relative to others. We will look at this in more depth in the well-being lecture.

(iv) Intrinsic Motivation: As discussed by Fehr and Falk and others, many people engage in tasks because they are intrinsically interested. Furthermore, people may have a desire to keep control over their own behaviour. 


5. Identity & Economics
The key paper for this is the paper on Economics and Identity by Akerlof and Kranton. This paper takes the view that looking at identity is vital to understand a wide range of economic phenomenon such as welfare dependency, ghettos, integration into the labour market, globalisation and economic growth. Identity emerges from the social categories we identify with or are members of by default. They outline a very simple model, which we will cover in the lecture, where membership of social categories enters directly into utility functions and use this to explain a range of economic phenomena such as gender discrimination 


Recommended Readings:
2. Akerlof (1998), Men without Children, The Economic Journal.
4. Fehr (2008), On the economics and biology of trust, IZA Discussion Paper.
3. Fehr & Falk (2001), Psychological Foundations of Incentives, Schumpeter Lecture at the European Economic Association Meeting.
4. Falk, Fehr & Fischbacher (2005), Driving Forces Behind Informal Sanctions, Econometrica.
5. Falk & Kosfeld (2006), The Hidden Costs of Control, American Economic Review.
6. Andreoni (1995), Cooperation in Public-Goods Experiments: Kindness or Confusion?, American Economic Review.
7. Milgram (1963), Behavioral Study of Obedience, Journal of Abnormal and Social Psychology.
8. Sacerdote (2001), Peer effects with random assignment: results for Dartmouth roommates, Quarterly Journal of Economics.

Supplementary Material:

No comments: