Amartya Sense is making a reasoned choice of a piece of fruit from a basket passed round at a dinner party. Filling in background circumstances as required, catalogue a range of bases on which his selection might be made and examine what role, if any, self-interest could be held to play in each of them. Do you think that there is a clear relationship between choice, preference and self-interest?
Essay no. 4 for Paper 14, Philosophy of Economics
Submitted in part-fulfilment of the requirements for the MPhil in Development Studies at the University of Cambridge
2004 - 2005
Much of traditional economic theory centres on the idea of the rational economic agent acting according to his self-interest, aspiring to achieve the best set of preferences to maximise his welfare, which can be derived from his or her choices (Sen 1991:14). This essay will investigate whether such a clear relationship between self-interest, preference and choice actually exists, discussing some of the possible dilemmas by examining whether Sen would act in accordance to the rational economic man (or fool) at a dinner party.
Sen (1977) asserts that people can make choices that are not in one’s self-interest. He refers to this as ‘commitment’. Commitment refers to the situation in which an individual chooses an action for reasons of say, duty, even if that generates a lower anticipated welfare than an alternative action. The choice made can happen to maximise the individual’s welfare. The individual would however, still make this choice even if it ceased to maximise his or her personal welfare (Sen 1977:326-7). In this case, as Sen asserts “a wedge” is driven “between personal choice and personal welfare, and much of traditional economic theory relies on the identity of the two” (Sen 1977:329). In the case of commitment preference, choice and self-interest do not correspond. In this case an individual’s self-interest can not coincide with choice, as the choice made has not maximised the individual’s welfare. Additionally, if one assumes that choice always reflects preference, than self-interest would not be reflected by either of these since the individual’s choice/preference does not maximise his or her self-interest (Sen 1982:69). The term preference appears to be ambiguous, how this could possibly be resolved will be discussed later on.
Sen (1977) also, points out that other people’s welfare levels can directly (positively) affect an individual’s personal welfare. Hence ‘self-sacrificing’ behaviour is (partially) motivated by an individual’s self-interest. Sen refers to this as ‘sympathy’ (Sen 1977:326). Arguably this type of behaviour is not necessarily purely motivated by self-interest. The fact that an individual has taken others into consideration appears to allow for a degree of altruism and a lesser degree of pure self-interest (Collard 1978:7). These points can best be illustrated by an example.
EXAMPLE Imagine Sen making a reasoned choice of a piece of fruit from a basket passed round at a dinner party. A number of circumstances could affect his choice. The menu over which the choice is being made could sway Sen’s choice by, for example increasing the number of favoured fruits (Sen 1997:753). Sen’s relationship and role with regard to the other guests could also influence his choice (Sen 1997:761). He might be more willing to choose the favoured piece of fruit if it was his birthday or might not if he was visiting people for the first time and was on his ‘best behaviour’. The way the choice is presented could also influence Sen’s choice, especially in the case where the host (and other guests) insists that Sen has the favoured piece of fruit (Sen:1997:747). Additionally, his knowledge of the other guest’s preferences as well as his own preference or non-preference (indifference) could impact Sen’s reasoned choice (Sen 1997:753). In this particular case, imagine that most of the fruit in the basket are regarded as ordinary fruit, like apples and bananas. There is however, one exception, a ripe mango, which most people would probably like best, including Sen. Hence, if Sen only took his own interest into account he would choose the mango. It is however, highly probable that Sen will not choose the mango.
There are a number of reasons why Sen partially motivated by self-interest, would not choose the mango. He might feel that leaving the mango for someone else’s enjoyment would positively affect his own welfare. Thus he might choose to act out of sympathy for others (Sen 1999:270). Related to this choice could be Sen considering the fact that he might lose more from choosing the mango by feeling guilty about his choice (Sen 1997:748). Sen might also believe that choosing the mango would affect his reputation with regard to the other guests and/or he might be relying on his ‘self-sacrificing’ act to benefit him in the future, thus again acting out of a degree of self-interest (Sen 1997:747, Hahn 1991:8). In addition, Sen could purposefully pass the basket of fruit on without making a choice, hoping that the others would choose the apples and bananas out of the same considerations he can think of, which would leave him with the mango. Sen refers to this move as ‘strategic nobility’ (Sen 1997:761-2).
Sen might also not choose the mango out of a sense of duty, moral considerations and/or acknowledgement of common norms (Sen 1977:748). It is however, questionable whether Sen acting out of these considerations would be acting out of commitment. If Sen could choose the mango without anyone else noticing, but would still choose not to due to, for example, a sense of duty he would be acting out of commitment. The presence of the other guests however, questions whether Sen could actually act out of pure commitment as a result of reputation effects. This may depend on the nature of the dinner party. One can imagine a set up in which the guests might not pay any attention to Sen’s choice. They might be too caught up in conversation and the setting might not be very intimate. The fact however, that Sen is deliberating so carefully suggests that the set up is a ‘noticing’ one. In this situation, it appears impossible to divorce reputation affects from commitment. Sen’s deliberation reinforces his acknowledgement of the situation and the impact his choice could have on the other guests. This suggests that his motivation not to choose the mango even if motivated by a sense of duty or recognition of a common norm (‘to never choose the best fruit as a guest’) must also be motivated by his recognition that by not acting this way he would cause the other guests to directly frown upon his behaviour, negatively affecting his reputation.
A BALANCE BETWEEN SELF-INTEREST AND ALTRUISM It appears that in this example Sen’s behaviour is to a degree always motivated by self-interest or arguably by different types of self-interest. Self-interest tends to invoke a negative connotation, implying something similar to selfishness. This example however, indicates that it was more often than not, in Sen’s self-interest to take others’ welfare into account because of the affect theirs have on his own. Acknowledging this indicates how self-interest does not necessarily exclude altruism (Collard 1978:4-5). I would assert that this type of ‘altruistic self-interest’ is inherent to human behaviour, because we are social beings and that it is exactly this type of behaviour that sustains society. One wonders whether it might even be possible to argue that altruism necessarily assumes a (small) degree of self-interest. This would reject the suggestion that commitment entails a non-sympathetic altruistic action, asserting that altruism inevitably implies a degree of sympathy.
A balance between ones narrow self-interest and altruistic self-interest appears important in social situations. If everyone would always act according to the interests of others, this would in the case of the fruit mean that no one would choose the mango. If people have acted out of sympathy, everyone would be left worse off since no one would gain from the anticipated benefits of allowing someone else to enjoy the mango. Where everyone shares the same subjective preference and acts out of sympathy it is essential that someone makes the purely self-interested choice. In a new social environment someone is therefore likely to intervene ‘forcing’ someone to choose the favoured choice. In Sen’s case the host (and other guests) might insist he has the mango or it is imaginable that the last person may be left with only the mango and (apologetically) will eat it.
If however, everyone has acted out of a sense of duty or common norm (‘never pick the favoured fruit as a guest’) enforced by reputation effects or if everyone feels they would feel guilty picking the mango, it would not matter if no one ended up eating the mango. If anyone one of them did, they would be worried what the others would think of them for not following the same ‘unwritten rule’ or would feel guilty. This however changes if the last person choosing was left with the mango. The last person in the group acting out of a sense of duty or following some common norm would be best off since the others would know he was forced to make this choice (which they essentially forced upon him by making their choices) and would therefore seem to exclude the possible negative reputation effects. The last person in a group of people acting out of guilt could arguably be worse off, if he or she continues to feel guilty being the only one in the presence of others sharing the same subjective preference, eating the favoured mango.
The previous discussion does not sound very realistic or familiar based on personal experiences at dinner parties or other similar situations that occur on a daily basis. It seems unlikely that a group of people would be motivated by the exact same considerations and even more unlikely that they would be motivated by a single consideration. It therefore seems credible to conclude that in general people make choices in social situations on the basis of a wide variety of considerations. And that a reasoned choice process necessarily entails a process of weighing these considerations, eventually resulting in a choice.
DIFFERENT TYPES OF PREFERENCE By breaking the choice process down, one can identify how preference is sensitive to the choice process, allowing one to identify different types of preferences linked to different types of considerations (Sen 1997:745). In social situations, the majority of people generally do not choose their “subjective preference” or choose out of “narrow self-interest”.1 Subjective preference refers to what an individual would choose without taking other people into account, acting purely out of “narrow self-interest”. Hence Sen’s subjective preference in the example above was choosing the mango. In a social situation it is in an individual’s self-interest to take a variety of considerations into account due to the fact that people are affected by other people’s reactions, welfare and/or future behaviour (Sen 1997:747). Sen would therefore be likely not to choose the mango himself out of reputation, sympathy and/or guilt considerations. These could be referred to as an individual’s “social preferences.” Additionally, a person could make “moral preferences” by acting out of commitment. This however, only seems to be the case in situations in which an individual can ‘free ride,’ free from any negative reputation affects. This implies that these cases can only entail public goods in impersonal situations (Sen 1977:330-1).
This leads one to conclude that whereas people are bound to one subjective preference and most likely one moral preference per individual choice preference, people are likely to have multiple social preferences in social situations. All of these social preferences involve a degree of self-interest and altruism. In the example set, all of Sen’s social considerations swayed in favour of not choosing the mango. It is possible to think of more difficult, ambiguous situations in which different social considerations might lead to different preferences. These would then have to be weighed against each other to reach a final choice.
In conclusion, it is clear that there is no clear relationship between self-interest, preference and choice. I have argued that where an individual can act on one choice from a set of alternatives in a social situation, he or she will have varying preferences according to different stages of the choice process linked to different considerations. These preferences vary on different levels of self-interest and most often include a degree of altruism (except in the case of subjective preferences). Generally, apart from in the exceptional case of commitment, people tend to choose an option that is at least partially motivated by self-interest as well as altruism.
Collard, D. 1978. Altruism and Economy: A study in non-selfish economics. Oxford: Martin Robertson.
Hahn, F. 1991. “Benevolence,” in J.G.T. Meeks (ed.). Thoughtful economic man: Essays on rationality, moral rules and benevolence. Cambridge: Cambridge University Press.
Sen, A.K. 1977. “Rational Fools: A Critique of the Behavioral Foundations of Economic Theory,” in Philosophy and Public Affairs. Vol. 6, no. 4, pages 317-344.
Sen, A.K. 1982. Choice, Welfare and Measurement. Oxford: Blackwell.
Sen, A.K. 1997. “Maximization of the Act of Choice,” in Econometrica. Vol. 65, No. 4, pages 745-779.
Sen, A.K. 1991. “Beneconfusion,” in J.G.T. Meeks (ed.). Thoughtful economic man: Essays on rationality, moral rules and benevolence. Cambridge: Cambridge University Press.
Sen, A.K. 1999. Development is Freedom. Oxford: Oxford University Press.
1 The following preference ranking was inspired by Sen’s discussion in Sen 1977:336-7. Subjective preferences is however, not defined the same way.