Moral responsibility and economic choice

Moral responsibility and economic choice Moral responsibility and economics Lászlo Zsolnai Budapest University of Economic Sciences, Budapest, Hungary 355 Moral responsibility emerges in choice situations where the choice has wideranging consequences and not only the decision maker but also other parties are affected by the outcome of the choice. In this paper I will refer to such situations as complex choice situations. The purpose of this paper is to present a normative but realistic model of responsible choice. The model is normative in the sense that it prescribes the way a morally responsible decision maker should behave in complex choice situations. However, the model is intended to be realistic in the sense that it does not require anything from the decision maker that would be contrary to the psychology of human choice behaviour. The basic fault of the rational choice model is its lack of psychological realism. Rational decision makers should maximize their utility functions under perfect information. This is a highly unrealistic requirement for human beings. By more than 50 years of research Herbert A. Simon has demonstrated that real-world decision makers are not capable of maximizing their utility functions, partly because of their strongly bounded computing capacities, partly because of the strictly limited information they usually have. Real-world decision makers are only capable of making satisficing choices (Simon, 1979). Daniel Kahneman’s recent research shows that decision makers are unable to foresee the real experienced utility of their chosen decision alternatives. For this reason their choices are not rational for most cases (Kahneman, 1994). The idea of moral responsibility Hans Jonas, the outstanding German-American philosopher, has injected the problem of moral responsibility into the modern moral discourse. For Jonas moral responsibility is essentially substantive in nature. The actor displays moral responsibility if he/she cares about the subject of his/her action. Power is inherent in responsibility relationship; there is an asymmetrical relation between the actor and the subject of his/her action. The actor can enter into a responsibility relationship either in a natural way or by contract. In the first case the subject becomes a subject of the actor’s action independently from his/her will. In the second case the subject becomes a subject of the actor’s action because he/she has established a contract with the actor. Jonas cites the parent and the statesman as ideal types of natural responsibility and contractual responsibility respectively. The parent is International Journal of Social Economics, Vol. 24 No. 4, 1997, pp. 355-363. © MCB University Press, 0306-8293 International Journal of Social Economics 24,4 356 responsible for his/her child not because of the child’s own will or even contrary to it. But the responsibility of the statesman comes from the political contract that he/she has established with his/her constituencies. The economic decision maker is somewhere in between the role of the parent and the role of the statesman. For example, while a corporate manager has a number of stakeholders (customers, employees, suppliers, creditors) with whom his/her company has contractual relationships, there are usually other stakeholders (the local community, the natural environment) who are affected by the company’s operations but towards whom the company has no contractual, only a natural responsibility. Consider the following famous case from business ethics. In 1978 three girls died in Winamac, Indiana because of the explosion of their Ford Pinto car. This was not the first case when this Ford model caused a serious accident by explosion. There were several law suits filed against Ford since it was proven that the top managers had been informed about the design problem of the model. Despite the warnings of their engineers, Ford managers decided to manufacture and sell the car (Hoffman, 1985). Our ethical intuition suggests that the Ford managers behaved in a nonresponsible way since they decided to put an unsafe model on the market. But how can this ethical intuition be defended on a more systematic ground? In the Ford Pinto case all the important components of complex choice situations can be identified. First, the decision maker has more than one decision alternative, that is, he/she can choose among different courses of action. Second, the decision maker or his/her organization has some goals that he/she wants to achieve. Third, in the choice situation there are ethical norms that represent the duties of the decision maker. Finally, other parties (stakeholders) are present who can be greatly affected by the outcome of the choice. In a complex choice situation the following components are the most important. (m ≥ 2) (1) A1, …, Ai, …, Am There are at least two decision alternatives for the decision maker. G1, …, Gj, …, Gn (n ≥ 1) (2) The decision maker or his/her organization has at least one goal to achieve. D1, …, Dk, …, Dp (p ≥ 1) (3) There exists at least one ethical norm that represents the duty of the decision maker. S1, …, Sq, …, Sr (r ≥ 1) (4) At least one stakeholder is present in the situation. Responsible choice involves finding and implementing the decision alternative which corresponds to the idea of moral responsibility in the given context. Ai* = Ω [A1, …, Ai, …, Am] (5) We need to define the decision rule Ω that selects the responsible course of action among the feasible ones. Deontology, rationality and respect In the context of economics Kenneth E. Goodpaster has presented the most developed model of responsible choice (Goodpaster, 1983, 1990; Goodpaster and Matthews, 1982). Using the conception of William K. Frankena (1980) Goodpaster distinguished two basic components of moral responsibility, namely rationality and respect. Rationality involves the following attributes: (1) lack of impulsiveness; (2) care in mapping out alternatives and consequences; (3) clarity about goals and purposes; and (4) attention to details of implementation. Rationality described by the attributes (1)-(4) differs strikingly from the rationality postulate of standard economics. The rationality concept used here is procedural in nature and does not require to maximize anything. Respect means a special awareness of and concern for the effects of one's decisions and policies on others, special in the sense that it goes beyond the kind of awareness and concern that would ordinarily be part of rationality, that is, beyond seeing others merely as instrumental to accomplishing one’s own purposes. Respect for others involves taking their needs and interests seriously. It is what Kant meant by the “categorical imperative” to treat others as valuable in and for themselves. In Goodpaster’s model responsible choice is to combine rationality and respect for others in decision making. In this model respect is basically, if not exclusively, a consequentialist account. This means that the decision maker considers the effects of his/her choice on the stakeholders. However, there are complex choice situations where a pure consequentialist account is not enough. In complex choice situations there might be marginal contributions, unforeseeable consequences, and distant effects. These phenomena create decision traps if the choice is based solely on consequentialist considerations. It is possible that the choice of the decision maker produces marginal contributions to the wellbeing of stakeholders but cumulative or aggregate effects of these marginal contributions might be detrimental for the stakeholders. Garret Hardin’s famous “tragedy of the commons” model describes such situations (Hardin, 1968). If the consequences of a certain course of action are partly or completely unforeseeable then the decision maker necessarily neglects them. Here lies another decision trap. If the consequences are foreseeable but distant in space and/or time then the decision maker Moral responsibility and economics 357 International Journal of Social Economics 24,4 358 discounts them. Effects beyond the normal (usually very narrow) space and time reference-frame of the decision maker are highly overdiscounted. Phenomena of marginal contributions, unforeseeable consequences, and distant effects show that non-consequentialist considerations are also required in complex choice situations. Deontological, that is non-consequentialist, considerations are based on ethical norms. Ethical norms can be understood as heuristic devices which help to avoid decision traps in complex choice situations. These devices capture the essence of evolutionary stable strategies of human communities. In a deontological account the value of decision alternatives does not depend on the real-world outcomes. It depends only on the correspondence of decision alternatives to the applying ethical norms. A number of well-known philosophers, sociologists and economists have emphasized the importance of deontological considerations in economic and political decision making (Elster, 1989; Epstein, 1987; Etzioni, 1988; Frank, 1988; Mansbridge, 1990; Nozick, 1974; Rawls, 1971; Sen, 1982). Goodpaster’s responsibility model can be enlarged to include the deontological aspect of choice. Hence responsible choice can be defined as a synthesis of deontology, rationality as goal achievement, and respect for stakeholders. Responsible decision makers try to make choices that involve applying ethical norms, promoting the achievement of their goals, and showing respect for affected parties. In complex choice situations three kinds of decision variables play a role. Decision alternatives have deontological value (D(A i)), instrumental (goal achievement) value (G(Ai)), and external value, that is value for the stakeholders (S(Ai)). The question is how can the value functions D(.), G(.), and S(.) be defined? The psychology of choice In the last few decades psychologists have discovered some basic regularities of human choice behaviour that should be taken seriously in developing a normative but realistic model of responsible choice. Based on a series of well-designed experimental studies Amos Tversky and Daniel Kahneman have stated the general features of the value functions of decision makers (Kahneman and Tversky, 1979). Decision makers display important differences in valuing positive and negative decision prospects. The value function is much steeper for negative outcomes than it is for positive outcomes. Tversky and Kahneman found that people are more sensitive to losses than to gains. |V(x)|< |V(–x)| (6) The magnitude of the value of a positive outcome x is smaller than the magnitude of the value of the corresponding negative outcome –x. Other experimental studies show that the loss aversion coefficient of decision makers is about 2, that is, a negative outcome can be compensated for them by a twice greater positive outcome (Kahneman, 1994). ~|V(–x)| |V(2x)|~ (7) Harvard psychologist Richard J. Herrnstein has discovered the so-called “matching law” after decades-long experimental research. The main finding is that it is the average reinforcement value of the decision alternatives that really counts for decision makers (Herrnstein and Prelec, 1991). Finally, the same experimental results of Reinhard Selten should be mentioned. He found that decision makers try to avoid trade-offs among different value dimensions (Selten, 1994). To be realistic, the normative model of responsible choice should be consistent with the above listed regularities of human choice behaviour. Modelling responsible choice Let Dk(.) be deontological value function as follows: 1  0 Dk ( Ai ) =   –2  if decision alternative Ai corresponds to ethical norm Dk ; if decision alternative Ai is neutral regarding ethical norm Dk ; (8) if decision alternative Ai violates ethical norm Dk ; Dk Ai characteristically shows the deontological value of decision alternative Ai regarding ethical norm Dk. Let Gj (.) be instrumental value function as follows: 1    0 G j ( Ai ) =   –2   if decision alternative Ai is positive for the achievement of goal G j ; if decision alternative Ai is neutral for the achievement of goal G j ; (9) if decision alternative Ai is negative for the achievement of goal G j ; Gj(Ai ) characteristically shows the deontological value of decision alternative Ai regarding the achievement of goal Gj. Finally, let Sq(Ai ) be external value function as follows: 1 if decision alternative A is good for stakeholder S ; i q  S q ( Ai ) = 0 if decision alternative Ai is neutral for stakeholder S q ;  –2 if decision alternative Ai is bad for stakeholder S q ; (10) Moral responsibility and economics 359 Sq(Ai ) characteristically shows the external value of decision alternative Ai International Journal of Social regarding stakeholder Sq. The above-defined value functions are Tversky and Kahneman-type value Economics functions since (8), (9), and (10) satisfy both (6) and (7). 24,4 How can the aggregate value functions D (.), G (.) and S (.) be construed? 360 Let w1, …, wk, …, wp be weights that represent the relative importance of ethical norms D1, …, Dk ,…, Dp for the given society. It is required that ∑ wk =1 (11) k Then D ( Ai ) = ∑ [ wk Dk ( Ai )] (12) k D(Ai ) shows the average deontological value of decision alternative Ai. Let u1, …, uj, …, un be weights that represent the relative importance of goals G1, …, Gj, …, Gn for the decision maker. It is required that ∑uj =1 (13) j Then G ( Ai ) = ∑ [u jG j ( Ai )] (14 ) j G(Ai) shows the average instrumental value of decision alternative Ai. Finally, let v1, …, vq, …, vr be weights that represent the relative stakes of stakeholders S1, …, Sq, …, Sr in the choice situation. It is required that ∑ vq =1 (15) q Then S ( Ai ) = ∑ [ vq Sq ( Ai )] (16 ) q S(Ai ) shows the average external value of decision alternative Ai. The following vector gives a multiple evaluation of decision alternative Ai. V– (Ai ) = [D(Ai), G(Ai), S(Ai)] (17) In contemporary moral philosophy three distinct viewpoints are identified from which an act can be evaluated, namely the viewpoint of the actor (doer relativity), the viewpoint of society (observer relativity), and the viewpoint of the stakeholders (receiver relativity) (Sen, 1982). In (17) the first component of the vector represents the evaluation of the decision alternative from the viewpoint of society. The second component of the vector represents the decision maker’s evaluation while the third component represents the evaluation of the stakeholders. The rule of responsible choice is stated as follows: (18) A*i = maximin [D(Ai), G(Ai), S(Ai)] Responsible choice demands the choice of the least worst alternative in the multidimensional space of deontological, instrumental, and external values. If there are two decision alternatives, A and B, then the responsible choice is A if and only if min [D(A), G(A), S(A)] > min [D(B), G(B), S(B)]. The underlying principle of responsible choice is that decision makers should make optimal compromises among the applying ethical norms, their own goals, and the interests of the stakeholders. Let us consider again the Ford Pinto case. The following simplified analysis serves as illustration only. The Ford managers’ main goal was to introduce the model without time delay and at no extra cost (G1). The applying ethical norm is to supply a safe product for the customers, to avoid jeopardizing their lives. (D1) Stakeholders were the actual and potential buyers and users of the car (S1). Essentially two alternatives were available for the managers. A 1 = manufacturing and selling the car without re-engineering; A2 = re-design the model and manufacturing and selling the safer version. Table I shows the multiple evaluation of the decision alternatives. It is easy to discover that A2 is the responsible course of action as it is suggested by our ethical intuition. Namely V– (A1) = [–2, 1, –2] while V– (A2) = [1, –2, 0]. A1 and A2 are equal in their worst components (–2) but the second worst component of A2 is better than that of A1 (0 versus –2). Using the maximin rule A2 is preferable to A1. A1 alternative A2 alternative Deontological value (D) Instrumental value (G) External value (S) –2 1 1 –2 –2 0 It is important to demonstrate that the rational choice model is a degenerate form of the responsible choice model. The rule of rational choice is as follows: (19) max G(Ai) (19) is equivalent to (18) if there are no applying ethical norms as well as no stakeholders in the choice situation, that is, D1, …, Dk, …, Dp and S1, …, Sq, …, Sr simply do not exist in the given context. From this it follows that responsible choice is the general case while rational choice is a special case only. The responsible decision maker The procedural model of responsible choice can be summarized as follows: Moral responsibility and economics 361 Table I. Multiple evaluation of the decision alternatives in the Ford Pinto case International Journal of Social Economics 24,4 362 (1) (2) (3) (4) (5) identifying the ethical norms which apply in the given choice situation; mapping out the stakeholders; defining the goals to be achieved; generation of decision alternatives; multiple evaluation of each alternative regarding the ethical norms, the decision maker’s goals and the stakeholders; (6) finding the least worst alternative in the multidimensional space of deontological, instrumental, and external values. There is always conflict or at least some tension among the ethical norms of society, the decision maker’s own goals, and the interests of the stakeholders. Making responsible choice is not an easy business at all. Responsible decision makers have some common character traits. On the one hand, they are rightly socialized persons, that is, they have interiorized the ethical norms of society (Etzioni, 1988). On the other hand, they have developed responsiveness towards their stakeholders (Freeman, 1984). In short, responsible decision makers display social commitment and empathy. These character traits remind us of the moral make up that Adam Smith presupposed about his fellow-businessmen (Sen, 1987). However, social commitment and empathy are such scarce resources in our economies that ethical awards and other prizes have been established for decision makers who have them. Rational choice does not produce satisficing outcomes in complex choice situations. Such situations require multiple considerations and optimal compromises across diverse value dimensions. References Elster, J. (1989), The Cement of Society, Cambridge University Press, Cambridge. Epstein, E.M. (1987), “The corporate social policy process: beyond business ethics, corporate social responsibility, and corporate social responsiveness”, California Management Review, No. 3. Etzioni, A. (1988), The Moral Dimension, The Free Press. New York, NY. Frank, R. (1988), Passions within Reason, W.W. Norton, New York, NY. Frankena, W.K. (1980), Thinking about Morality, University of Michigan Press, Ann Arbor, MI. Freeman, E.R. (1984), Strategic Management: A Stakeholder Approach, Pitman, Boston, MA; London. Goodpaster, K.E. (1983), “The concept of corporate conscience”, Journal of Business Ethics, No. 1. Goodpaster, K.E. (1990), “PASCAL and corporate conscience”, Strategic Direction, November. Goodpaster, K.E. and Matthews, J.B. (1982), “Can a corporation have a conscience?”, Harvard Business Review, January-February. Hardin, G. (1968), “The tragedy of the commons”, Science, Vol. 162, pp. 1243-8. Herrnstein, R.J. and Prelec, D. (1991), “Melioration: a theory of distributed choice”, Journal of Economic Perspectives, No. 3. Hoffman, M.W. (1995), “The Ford Pinto”, in Hoffman, M.W. and Frederick, R.E., Business Ethics, McGraw-Hill, Inc., New York, NY, pp. 552-8. Jonas, H. (1984), The Imperative of Responsibility, University of Chicago Press, Chicago, IL. Kahneman, D. (1994), “New challenges to the rationality assumption”, Journal of Institutional and Theoretical Economics, No. 1. Kahneman, D. and Tversky, A. (1979), “Prospect theory: an analyses of decision under risk”, Econometrica, No. 2. Mansbridge, J. (Ed.) (1990), Beyond Self-Interest, University of Chicago Press, Chicago, IL. Nozick, R. (1974), Anarchy, State, and Utopia, Basic Books, New York, NY. Rawls, J. (1971), A Theory of Justice, Harvard University Press, Cambridge and Boston, MA. Selten, R. (1994), “New challenges to the rationality assumption”, Journal of Institutional and Theoretical Economics, No. 1. Sen, A. (1982), “Rights and agency”, Philosophy and Public Affairs, No. 1. Sen, A. (1987), On Ethics and Economics, Blackwell, Oxford. Simon, H.A. (1979), “Rational decision making in business organizations”, American Economic Review, September. View publication stats Moral responsibility and economics 363