bounded rationality and the emergence of simplicity amidst complexity

Bounded rationality and the emergence of simplicityamidst complexity

by Narbadeshwar Mishra

Madras University School OF Economics(MUSE)Department of EconomicsUniversity of Madras

Chepauk, Chennai – 600005Tamilnadu , India

[email protected](91)7299844800

June2015

Bounded rationality and the emergence of simplicity amidst complexity

Abstract

The notion of rationality and the way in which it assumed and applied in economics is a much debated topic within the discipline itself and beyond. The notion of bounded rationality has recently gained considerable popularity in the behavioral and social sciences. The vast body of literature under the headings of 'behavioral economics' and 'economics and psychology' have attempted to make sense of the extent to which and the manner in which rationality in reality differsfrom rationality as it is assumed in economics. The term 'bounded rationality', which can be traced back to Herbert Simon's influential contributions in the 1950s, has been used by many when referring to departures from the conceptualization of rationality as consistency orrationality as maximization in mainstream economic theory. Today, the notion of bounded rationality has a permanent place in economics. Its impact has been profound in terms of our theoretical and empirical understanding of decision making and judgment, markets, organizations and institutions. Parallel to the current research in mapping the boundaries of bounded rationality (to paraphrase Kahneman's Nobel lecture title) is an increasingly influential line of research that attempts to transform economics, this time into a social science that embraces complexity theory. A core element within this research programme is its focus on the emergence of complex structures from micro-level interactions between relatively simple parts/elements/agents. An analysis of how bounded rationality and complexity is related should be of great interest to economists. Most economists are likely to agree on the bounded rational nature of the human species as well as the complex nature of the economy. Yet, thesetwo aspects have often been considered and researched separately with a few exceptions even though both are inextricably linked. This is a

key aspect of Herbert Simon's ideas. The purpose of this essay is to explore the relationship between the simple and the complex in economics by anchoring our analysis on bounded rationality. The point of view taken in this essay is that bounded rationality and the complexity of environment are both inextricably linked - that the emergence of complex social structures would not possible without interactions between bounded rational agents and vice-versa. Furthermore, the bounded rational nature of agents is in itself a consequence of a complex environment. What this implies in terms of a broader vision of economics is a topic worth exploring. : Rational choice theory rests on the assumption that decision makers have complete mental models of the events, consequences and acts that constitute their environment. They are considered to be individually rational if they hold preferences among acts that satisfy axioms such as ordering and independence, and they are collectively rational if they satisfy additional postulates of inter-agent consistency such as common knowledge and common prior beliefs. It follows that rational decision makers are expected-utility maximizers who dwell in conditions of equilibrium. But real decision makers are only boundedly rational, they must cope with disequilibrium and environmental change, and their decision models are incomplete. As such, they are often unable or unwilling to behave in accordance with the rationality axioms, they find the theory hard to apply to most personal and organizational decisions, and they regard the theory’s explanations of many economic and social phenomena to be unsatisfying.Models and experimental studies of bounded rationality, meanwhile, often focus on the behavior of unaided decision makers who employ strategies such as satisficing or adaptive learning that can be implemented with finite attention, memory, and computational ability.

Aim

The author looks at the subject matter from the point of view of economic theory. He is convinced of the necessity of reconstructing microeconomics on the basis of a more realistic picture of economic decision making. Moreover he thinks that there are strong reasons for modeling boundedly rational economic behavior as non-optimizing.

CONTENTS1. Introduction 1.1 What is Bounded Rationality? 1.2 How behavioral economics is different from traditional economics? 1.3 Bounded rationality - origins and development 1.3.1 Origin of bounded rationality

2. Forms of Rationality 2.1 Instrumental 2.2 Procedural 2.3 Expressive

3. Choice Axioms

4. Economist’s point of view on Bounded Rationality 4.1 Herbert’s Simon’s view 4.2 Kahneman’s & Tversky’s view 4.3 Reinhard Selten’s view 4.4 Gerd Gigerenzer’s view 4.5 Till Grüne-Yanoff’s view

5. Bounds of rationality

6. Some major concepts and theories of bounded rationality 6.1 Impossibility of unfamiliar optimization when decision time is scarce 6.2 Aspiration adaptation theory 6.2.1 Features of bounded rationality modelled by aspiration adaptation theory 6.3 Learning direction theory 6.4 Self control and Motivation 6.5 Hyperbolic discounting 6.6 Want generator and administrator 6.7 Rationality in the classical theory of the firm 6.7.1 The Limits of Rationality 6.7.2 Alternatives to the classical goals 6.8 Approaches to rational choice in chess 6.8.1 The game-theoretical definition of rationality in chess 6.8.2 Limits to rationality in chess 6.8.3 Satisficing and optimizing 6.9 Bounded Rationality in Design

7. What do we mean `choice under certainty?'

8. Heuristics and biases 8.1 Research on heuristics and biases 8.1.2 Critical assessment 8.2 Heuristics and ecological rationality 8.2.1 Fast and frugal heuristics 8.3 Bounded minds and ecological rationality 8.4 Heuristics for management decisions 8.5 Critique

9. Rationality in judgment and decision making - Dual process models as a unifying approach 9.1 Different concepts of rationality – are they mutually exclusive?

10. Bounded rationality, behavioral economics, and organization theory 11. The implications of 'bounded rationality' for the structure and conduct of public policy

11.1 Conclusion 11.2 Summary

12. Conclusion

13. Bibliography

1. Introduction

1.1 What is Bounded Rationality?

The notion of bounded rationality has recently gained considerable popularity in the behavioral and social sciences. Modern mainstream economic theory is largely based on an unrealistic picture of human decision making. Economic agents are portrayed as fully rational Bayesian maximizers of subjective utility. This view of economics is not based on empirical evidence, but rather on the simultaneous axiomization of utility and subjective probability. In the fundamentalbook of Savage the axioms are consistency requirements on actions withactions defined as mappings from states of the world to consequences (Savage 1954). One can only admire the imposing structure built by Savage. It has a strong intellectual appeal as a concept of ideal rationality. However, it is wrong to assume that human beings conform to this ideal.

1.2 How Behavioral Economics Differs from Traditional Economics

All of economics is meant to be about people’s behavior. So, what is behavioral economics, and how does it differ from the rest of economics?

Economics traditionally conceptualizes a world populated by calculating, unemotional maximizers that have been dubbed Homo economicus. The standard economic framework ignores or rules out virtually all the behavior studied by cognitive and social psychologists. This “unbehavioral” economic agent was once defended onnumerous grounds: some claimed that the model was “right”; most otherssimply argued that the standard model was easier to formalize and practically more relevant. Behavioral economics blossomed from the realization that neither point of view was correct.

The standard economic model of human behavior includes three unrealistic traits—unbounded rationality, unbounded willpower, and unbounded selfishness—all of which behavioral economics modifies.

Nobel Memorial Prize recipient HERBERT SIMON (1955) was an early critic of the idea that people have unlimited INFORMATION -processing capabilities. He suggested the term “bounded rationality” to describe a more realistic conception of human problem-solving ability. The failure to incorporate bounded rationality into economic models is just bad economics—the equivalent to presuming the existence of a freelunch. Since we have only so much brainpower and only so much time, wecannot be expected to solve difficult problems optimally. It is eminently rational for people to adopt rules of thumb as a way to economize on cognitive faculties. Yet the standard model ignores thesebounds.

Departures from rationality emerge both in judgments (beliefs) and in choices. The ways in which judgment diverges from rationality are extensive (see KAHNEMAN et al. 1982). Some illustrative examples includeoverconfidence, optimism, and extrapolation.

An example of suboptimal behavior involving two important behavioral concepts, loss aversion and mental accounting, is a mid-1990s study ofNew York City taxicab drivers (Camerer et al. 1997). These drivers paya fixed fee to rent their cabs for twelve hours and then keep all their revenues. They must decide how long to drive each day. The profit-maximizing strategy is to work longer hours on good days—rainy days or days with a big convention in town—and to quit early on bad days. Suppose, however, that cabbies set a target earnings level for each day and treat shortfalls relative to that target as a loss. Then they will end up quitting early on good days and working longer on baddays. The authors of the study found that this is precisely what they do.

Consider the second vulnerable tenet of standard economics, the assumption of complete self-control. Humans, even when we know what isbest, sometimes lack self-control. Most of us, at some point, have eaten, drunk, or spent too much, and exercised, saved, or worked too little. Though people have these self-control problems, they are at least somewhat aware of them: they join diet plans and buy cigarettes by the pack (because having an entire carton around is too tempting). They also pay more withholding taxes than they need to in order to

assure themselves a refund; in 1997, nearly ninety million tax returnspaid an average refund of around $1,300.

Finally, people are boundedly selfish. Although economic theory does not rule out altruism, as a practical matter economists stress self-interest as people’s primary motive. For example, the free-rider problems widely discussed in economics are predicted to occur because individuals cannot be expected to contribute to the public good unlesstheir private welfare is thus improved. But people do, in fact, often act selflessly. In 1998, for example, 70.1 percent of all households gave some money to CHARITY , the average dollar amount being 2.1 percent of household income.1 Likewise, 55.5 percent of the population age eighteen or more did volunteer work in 1998, with 3.5 hours per week being the average hours volunteered.2 Similar selfless behavior has been observed in controlled laboratory experiments. People often cooperate in PRISONERS’ DILEMMA games and turn down unfair offers in “ultimatum” games. (In an ultimatum game, the experimenter gives one player, the proposer, some money, say ten dollars. The proposer then makes an offer of x, equal or less than ten dollars, to the other player, the responder. If the responder accepts the offer, he gets x and the proposer gets 10 − x. If the responder rejects the offer, then both players get nothing. Standard economic theory predicts that proposers will offer a token amount (say twenty-five cents) and responders will accept, because twenty-five cents is better than nothing. But experiments have found that responders typically reject offers of less than 20 percent (two dollars in this example).

Although rational choice theory is the dominant paradigm of quantitative research in the social and economic sciences, it is not the only such paradigm—and of course not all social-scientific research is quantitative. The following is a brief survey of alternative paradigms that have useful lessons to offer.

1.3 Bounded rationality – Origins and development

1.3.1 Origin of bounded rationality

The idea of unbounded rationality assumes that people have unlimited computational power, time and knowledge (and from a management perspective you could also include unlimited money, if taken into account that most information in management settings, like benchmark studies, are not available for free). This form of omniscience leads to the final decision which maximizes the expected utility. Economic models, supposing unbounded rationality are drawn from homo economicus, which is the underlying principle of all neoclassical economic models (Morgan, 2006): maximizing expected utility and statistical Bayesian models. It seems obvious that the assumption of unbounded rationality is an unrealistic yardstick for human reasoning in the real world. Contrary to this objection, Bazerman and Moore, forinstance, these authors argue that a rational decision maker “will identify all relevant criteria in the decision making process” and that “an optimal search continues only until the cost of the search outweighs the value of the added information”. Some of the advice fromthese authors even sounds like a description of the concept of unbounded rationality, when they stress: “the rational decision maker carefully assesses the potential consequences on each of the identified criteria of selecting each of the alternative solutions”.

2. Forms of Rationality

2.1 Instrumental

This form of rationality is concerned with `means-ends' reasoning and is the form of rationality we will be dealing with this term. Some philosophers and many economists take this to be the only form of rationality. Instrumental rationality is a non-substantive account of rationality that takes preferences as given and primative facts about individuals that are not subject to further analysis.

2.2 Procedural

Procedural rationality, also called, `bounded rationality' is really another form of instrumental rationality that takes into account what it is instrumentally rational for an agent to do, given that she is a non-idealAgent.

2.3 Expressive

This form of rationality is also called `practical reasoning' and it is a substantive view of rationality that is much broader than the instrumental rationality of choice theory. Expressive rationality concerns itselfnot only with `means-ends' reasoning but also with the value of the ends themselves.

The following will help to illustrate the differences between expressive (practical) and instrumental rationality: When I found out that a famous philosopher smoked (many philosopher smoke actually) I commented to my friend Sebastian that his smoking was `irrational'. `How can such a smart person smoke?', I asked. Sebastian replied that since as I didn't know the philosopher's preferences, I was in no position to say he was acting inconsistently and violating rationality. I said, `no, the point is, he shouldn't have a desire to smoke, this desire (or preference for smoking over not smoking) itselfwas irrational'. What was the cause of our disagreement? I had a substantive, practical account of rationality in mind, while Sebastianwas speaking of purely instrumental rationality. But what would make smoking irrational, from an instrumental perspective? In order to answer this question we must examine what instrumental rationality is in more detail.

3. Choice Axioms

The standard approach to instrumental rationality is axiomatic. Colloquially, we take rational choices to be consistent and preferencemaximizing. This means that agents act rationally if their choices do not contradict one another and when their choices are the best elementin their choice set. More formally, these requirements are stated as

axioms. We'll begin with the basic three axioms necessary for establishing a preference ordering in cases of choice under certainty.

4. Economist’s point of view on Bounded Rationality

4.1 Herbert Simon’s view

Herb Simon pioneered the study of bounded models of rationality. Simonfamously argued that decision makers typically satisfice rather than optimize. According to Simon, a decision maker normally chooses an alternative that meets or exceeds specified criteria, even when this alternative is not guaranteed to be unique or in any sense optimal. For example, Simon argued that an organism—instead of scanning all thepossible alternatives, computing each probability of every outcome of each alternative, calculating the utility of each alternative, and thereupon selecting the optimal option with respect to expected utility typically chooses the first option that satisfies its ‘aspiration level.’ Simon insisted that real minds are adapted to real-world environments. Simon writes, “Human rational behavior is shaped by a scissors whose two blades are the structure of task environments and the computational capabilities of the actor”. Yet by paying undivided attention to the heuristics used by real agents, a very influential line of work on models of bounded rationality privileges one of these two blades over the other. This is the tradition initiated by the heuristics-and-biases program championed byKahneman and Tversky (see the classic and the more recent). According to followers of this tradition, laws of probability, statistics, and logic constitute normative laws of human reasoning; descriptively, nevertheless, human reasoners follow heuristics that are systematically biased and error-prone. But this is hardly Simon’s idea, who insisted on the accuracy of adapted and bounded methods.

The term bounded rationality is thought to have been coined by HerbertA. Simon. In Models of Man, Simon points out that most people are onlypartly rational, and are emotional/irrational in the remaining part oftheir actions. In another work, he states "boundedly rational agents experience limits in formulating and solving complex problems and in processing (receiving, storing, retrieving, transmitting) information". Simon describes a number of dimensions along which "classical" models of rationality can be made somewhat more realistic,

while sticking within the vein of fairly rigorous formalization. Theseinclude:

Limiting the types of utility functions Recognizing the costs of gathering and processing information Possibility of having a "vector" or "multi-valued" utility

function

Simon suggests that economic agents use heuristics to make decisions rather than a strict rigid rule of optimization. They do this because of the complexity of the situation, and their inability to process andcompute the expected utility of every alternative action. Deliberationcosts might be high and there are often other concurrent economic activities also requiring decisions.

Bounded rationality is not irrationality. A sharp distinction should be made here. Theories that incorporate constraints on the informationprocessing capacities of the actor may be called as theories of bounded rationality.

4.2 Kahneman’s & Tversky ’s view

The behavioural sciences, in particular economics, have for a long time relied on principles of rationality to model human behaviour. Rationality, however, is traditionally construed as a normative concept: it recommends certain actions, or even decrees how one ought to act. It may therefore not surprise that these principles of rationality are not universally obeyed in everyday choices. Observing such rationality-violating choices, increasing numbers of behavioural scientists have concluded that their models andtheories stand to gain from tinkering with the underlying rationality principles themselves. This line of research is today commonly known under the name ‘bounded rationality’. In all likelihood, the term ‘bounded rationality’ first appeared in print in Models of Man (Simon 198). It had various terminological precursors, notably Edgeworth’s ‘limited intelligence’ and ‘limited rationality’.

The term ‘bounded rationality’ is often employed to denote any evidence of the deficiencies of the standard models. Presumably, standard models are thought of as assuming ‘full’ rationality, so that

evidence against them is considered evidence for some not necessarily further specified ‘bound’on rationality.

4.3 Reinhard Selten’s view

H.A. Simon described decision making as a search process guided by aspiration levels. An aspiration level is a value of a goal variable which must be reached or surpassed by a satisfactory decision alternative. In the context of the theory of the firm one may think ofgoal variables like profit and market share.

Decision alternatives are not given but found one after the other in asearch process. In the simplest case the search process goes on until a satisfactory alternative is found which reaches or surpasses the aspiration levels on the goal variables and then this alternative is taken. Simon coined the word satisficing for this process.

Often satisficing is seen as the essence of Simon’s approach. However,there is more to it than just satisficing. Aspiration levels are not fixed once and for all, but dynamically adjusted to thesituation. They are raised, if it is easy to find satisfactory alternatives and lowered if satisfactoryalternatives are hard to come by. This adaptation of aspiration levelsis a central idea in Simon=searly writings on bounded rationality.

Three features characterize Simon’s original view of bounded rationality: Search for alternatives,satisficing, and aspiration adaptation.

4.4 Gerd Gigerenzer’s view

Gerd Gigerenzer opines that decision theorists have not really adheredto Simon's original ideas. Rather, they have considered how decisions may be crippled by limitations to rationality, or have modeled how people might cope with their inability to optimize. Gigerenzer proposes and shows that simple heuristics often lead to better decisions than theoretically optimal procedures.

On the one side, the unbounded rationality concept refers to concepts in which human judgment and decision making (JDM) incorporate a kind of a microprocessor which calculates the (optimal) outcome. A

caricature of this concept is a description of a super chip driven human agent that knows all the odds and is a perfect Bayesian calculator or as Thaler and Sunstein state: “Homo economicus can thinklike Albert Einstein, store as much memory as IBM‘s Big Blue, and exercise the willpower of Mahatma Gandhi” (Thaler & Sunstein, 2008). Some models use an optimization under constraints approach, however, that does not change the capabilities of homo economicus. The optimization under constraints model states that we search for furtherinformation until the costs exceeds the benefits (for an example see Stigler, 1961). This modification of the unbounded concept however leads to an infinite regress, because all the benefits and costs have to be computed somehow – and how shall a bounded mind conduct such sophisticated estimations? Rieskamp and Otto (2006) thus call this symptomatic issue “the recursive homunculi problem of deciding how to decide”.

The notion of bounded rationality, as introduced by Herbert Simon in the 1940s, is focused on how we deal with the limitations of social actors as human beings (Simon, 1945). First, heuristics and biases deploy norms from the neoclassical approach and benchmark humans with respect to these axioms (Kahneman & Tversky, 2000). Satisficing, a blend of sufficing and satisfying, is concentrated on finding solutions that are good enough and departs from the ideal of optimization (Simon, 1955). Finally, advocates of fast and frugal heuristics go a step beyond satisficing and point out that under certain circumstances these descriptive models can be seen as a normative standard (Gigerenzer et al., 1999). Gigerenzer and Gaissmaier (2011) define heuristic as “a strategy that ignores part ofthe information, with the goal of making decisions more quickly, frugally, and/or accurately than more complex methods”. This way they make a clear distinction between their view on mental shortcuts and the one provide by Kahneman and Tversky (2000).

4.5 Till Grüne-Yanoff’s view

The behavioural sciences, in particular economics, have for a long time relied on principles of rationality to model human behaviour. Rationality, however, is traditionally construed as a normative concept: it recommends certain actions, or even decrees how one ought to act. It may therefore not surprise that these principles of rationality are not universally obeyed in everyday choices. Observing

such rationality-violating choices, increasing numbers of behavioural scientists have concluded that their models andtheories stand to gain from tinkering with the underlying rationality principles themselves. This line of research is today commonly known under the name ‘bounded rationality’.

5. Bounds of rationality

Full rationality requires unlimited cognitive capabilities. Fully rational man is a mythical hero who knows the solutions of all mathematical problems and can immediately perform all computations, regardless of how difficult they are. Human beings are very different.Their cognitive capabilities are quite limited. For this reason alone the decision behavior of human beings cannot conform to the ideal of full rationality.It could be the case that in spite of obvious cognitive limitations the behavior of human beings isapproximately correctly described by the theory of full rationality. Confidence in this conjecture of approximate validity explains the tenacity with which many economists stick to the assumption of Bayesian maximization of subjectively expected utility. However, thereis overwhelming experimental evidence for substantial deviations from Bayesian rationality (Kahneman, D. P. Slovic and A. Tversky, 1982). People do not obey Bayes’rule, their probability judgements fail to satisfy basic requirements even in situations involving no risk and uncertainty.

The cognitive bounds of rationality are not the only ones. A decision maker may think that a choice is the only rational one, e.g. to stop smoking, but nevertheless not take it. Conclusions reached by rationaldeliberations may be overridden by strong emotional impulses. The lackof complete control over behavior is not due to motivational bounds ofbehavior rather than to cognitive ones.Boundedly rational decision making necessarily involves non-optimizingprocedures.

6. SOME MAJOR CONCEPTS AND THEORIES OF BOUNDED RATIONALITY

6.1 Impossibility of unfamiliar optimization when decision timeis scarce

Imagine a decision maker, who has to solve an optimization problem in order to maximize hisutility over a set of decision alternatives. Assume that decision timeis scarce in the sense thatthere is a deadline for choosing one of the alternatives. The decisionmaker has to do his bestwithin the available time.

It is useful to distinguish between familiar and unfamiliar problems of this kind. A problem isfamiliar if the decision maker knows the optimal way to attack it. This means that he knows whatto do by prior training or mathematical investigation or that the problem is so simple that a suitable method immediately suggests itself.

In the case of an unfamiliar problem the decision maker must devise a method for finding thealternative to be chosen before it can be applied. This leads to two levels of decision makingactivities which both take time.

level 1: Finding the alternative to be chosen

level 2: Finding a method for level 1

What is the optimal approach to the problem of level 2 ? One can hardly imagine that this problem is familiar. Presumably a decision maker who does not immediately know what to do on level 1will also notbe familiar with the task of level 2. Therefore he has to spend some time finding an optimal method for solving the task of level 2. We arrive at level 3.It is clear that in this way we obtain an infinite sequence of levels k = 2, 3, ... provided that finding an optimal method for level k continues to be unfamiliar for every k.

level k: Finding a method for level k!

It is reasonable to assume that there is a positive minimum time whichis required for the decision making activities at each level k. Obviously this has the consequence that an optimization approach is not feasible when decision time is scarce. itstrongly suggests that a truly optimizing approach to unfamiliar decision problems with constrained decision time is not feasible.The impossibility of unfamiliar optimization, when decision time is scarce, would not be of greatimportance if most optimization problems faced by real people were familiar to them in the strictsense explained above. It is clear that the opposite is true.

6.2 Aspiration adaptation theory

As has been argued in the preceding section, there are reasons to believe that unfamiliaroptimization is impossible within the cognitive bounds of rationality,when decision time is scarce.This raises the following question: How can we model the non-optimizing behavior of boundedly rational economic agents? Only recently this aspiration adaptation theory has been made available in English (Selten 1998).Aspiration adaptation theory offers a coherent modelling approach to bounded rationality.

6.2.1 Features of bounded rationality modelled by aspiration adaptation theory

The decision behavior modelled by aspiration adaptation theory has some features which seem tobe of significance for the description of bounded rationality independently of modelling details.These features are listed below:1. Goal incomparability2. Local procedural preferences3. Integrated decisions on decision resources4. Decisions based on qualitative expectations5. Cautious optimism in the search for alternatives and the use of qualitative expectations

6. Risk related goal variables

Aspiration adaptation theory models decision making in a multi-goal framework with goalincomparability and local procedural preferences. These properties areembodied in the aspiration scheme. Integrated decisions on decision resources are modelled by aspiration adaptation involving decision resource stocks as goal variables. Aspiration adaptation theory also describes the use of qualitative expectations on the directions of change as the basis of a cautiously optimistic construction of expected feasible goal vectors for alternatives and aspiration adaptation among them. Search for alternatives is modelled as cautiously optimistic, too. Risk related goal variables explain how risks can be limited without probability judgements.

6.3 Learning direction theory

Learning direction theory (Selten and Stoecker 1986, Selten and Buchta1998) is anothersurprisingly sucessful approach to learning which is quite different from reinforcement theory. The basic idea can be illustrated by a simple example: Consider an archer who repeatedly tries to hit the trunk of a tree by bow and arrow. After a miss he will have the tendency to aim more to the left if the arrow passed the trunk at the right hand side and more to the right in the opposite case.The exampleis not as trivial as it may seem to be. The archer is guided by a qualitative causal model of his environment. This model relates changes of the angle at which the bow is held to changes of the direction in which the arrow flies. After a miss he sees on which sideof the trunk the arrow has passed. This feedback and the qualitative causal model enable the archer to draw a qualitative conclusion about what would have been better in the past period. He can determine in which direction of what he did a better alternative could have been found.The term ex post rationality refers to the analysis of what could havebeen done better, in contrast to ex ante rationality which reasons about future consequences of possible actions. Learning direction theory is based on ex post rationality. It requires reasoning, but only about the past, not about the future.

Learning direction theory can be applied to repeated decision tasks inwhich a parameter pt hasto be chosen in a sequence of periods t = 1, ..., T, provided that thedecision maker has aqualitative causal model and receives feedback which enables him to infer in which direction from what he did a better choice could have been found.The theory predicts that the parameter tends to be changed in the direction indicated by theinference, if it is changed at all. This is a weak prediction which, however, has proved to besuccessful in about a dozen experimental studies (see Selten 1998).Learning direction theory differs from reinforcement learning by a property referred to asimprovement orientation. It does not matter whether the choice of lastperiod resulted in a high or low payoff. What matters is the directionin which a better choice could have been found.

6.4 Self control and Motivation

The human control system determines the goal pursued by boundedly rational decisionmaking. Unfortunately we have no clear understanding of the interaction of different motivational forces. This is a serious difficulty for the development of a comprehensive theory of bounded rationality. Some decision problems are easy and others cause serious inner conflicts. What is an inner conflict? One approach to this question going back to Freudian psychoanalytic theory is the idea thatthe self is composed of several parts with different interests. Conflicts may arise among these components of the self.

6.5 Hyperbolic discounting

In economic theory the usual assumption about discounting streams of payoffs is that of a constant discount rate q with 0 < q < 1. The payoffs are utilities ut obtained in period t = 1, 2, ... This means that ut enters the discounted sum of the payoff stream with the weightqt. Experimental evidence shows that behavior is much better describedby hyperbolic discounting with weights 1 / (A + t) for ut, where A is a positive constant (Ainslie 1992 ). Thus a subject may prefer 95

money units now to 100 tomorrow, but 100 in a year and a day to 95 in a year. This involves a time inconsistency since after a year the second choice will look like the first one today. Ainslie models decision making as a game among multiple selves, one for each time period. The self of time t decides what is done at time t with the aimof maximizing its own hyperbolically discounted utility, taking into account what later selves are expected to do. This means that a subgame perfect equilibrium is played. However, the game can have morethan one such equilibrium, e.g. one in which the present self and all future selves continue to smoke and another one in which the present self stops to smoke and all future selves follow this example. The second equilibrium may be better for all of them. Assume that this is the case.The two equilibria have not yet been fully described. In the first one, the smoking equilibrium, all selves smoke independently of prior history. In the second one, the non-smoking equilibrium, the present self does not smoke and the later ones do not either, as long as none of them has deviated. If one of them smokes all the later ones will smoke under all circumstances.In order to make these equilibria work, one has to assume that the sequence of future selves isinfinite. Even if this is wrong one may argue that an analysis based on this idea nevertheless iscorrect, since it is known that boundedly rational game players can show behavior akin toequilibria of infinite supergames in finite supergames. (Selten and Stoecker 1986)Suppose that the person is in the smoking equilibrium. It would be better to switch to the nonsmoking equilibrium. However, there may be many other subgame perfect equilibria, among them a delayed non-smoking equilibrium in which the present self smokes, but all future selves don’t. Under plausible assumptions on payoffs this is the case and the delayed non-smoking equilibrium is better for the present selfthan the smoking equilibrium.In this way the inner conflict between stopping or continuing to smokecan be modelled as aproblem of equilibrium selection. This is a very interesting modellingapproach to the phenomenon of inner conflict, even if the game theoretic reasoning is not fully worked out by Ainslie. However, it isbased on strong rationality assumptions. The multiple selves are modelled as fully rational game players. A more plausible picture of inner conflicts faced by boundedly rational players requires another

kind of decision behavior . Maybe one should try to modify Ainslie`s theory in this direction.The split of the person into muliple selves with conflicting goals in itself is a bound of rationalityfor the person as a whole, even if it is not cognitive but motivational. Not only cognitive, but alsomotivational bounds of rationality must be taken into account by a comprehensive theory ofbounded ratonality.

6.6 Want generator and administrator

Otwin Becker (1967 ) has proposed a theory of household behavior whichextends aspirationadaptation theory to this context. The household divides its monthly income into a number of funds for different kinds of expenditures likea fund for food, a fund for clothing, a fund for entertainment, etc.. The goal variables are the fund sizes and upper price limits for wants, like the desire for a pair of shoes seen in the window of a shop, or an excursion advertised by a travel agency. Such wants are produced by a want generator, modelled as a random mechanism.When a want is generated by the want generator another instance, the administrator, checkswhether there is still enough money in the appropriate fund and whether the want remains underthe price limit for such desires. If the price limit is violated, the want is rejected. If the wantremains under the price limit but there is not enough money in the fund then the want will still begranted if transfer rules permit the transfer of the missing amount from another fund. The structure of these transfer rules will not be explained here. If such a transfer is not permissible, then the want is rejected.At the end of the spending period a new aspiration level for the next one is formed by aspirationadaptation in the light of recent experience. The details will not be explained here. If the household theory of Otwin Becker is applied to the spending behavior of a single person, then want generator and administrator are different personalty components. Conflicts between them are not modeled by the theory but it may be possible to extend itin this direction. Everyday experience suggests that sometimes wants are realized against the will of the administrator.

The split of a person into a mechanistically responding want generatorand a boundedly rationaladministrator seems to be a promising modelling approach not only to household theory but alsofor other areas of decision making.Rational decision making within the cognitive bounds of human beings must be non-optimizing.The discussion of motivation was restricted to only one aspect of it, the idea that aperson is subdidvided into several components which may be in conflictwith each other.

6.7 Rationality in the classical theory of the firm

The classical theory of the firm form provides a useful standard for comparing and differentiating theories of rationality, where profit isdefined as the difference between revenue and cost of production. The given conditions are two in number:

1. The demand function : The quantity demanded is a function of price:

(1)

Since revenue equals price times quantity, the demand function determines revenue :

(2)

2. The cost function : The cost of production is a function of the quantity produced :

(3)

If the quantity produced equals the quantity demanded,

(4)

Then the profit, to be maximized, is simply the difference between revenue and the cost of production :

(5)

And, under appropraite assumptions regarding diffrentiability, we willhave for the maximum profit :

(6)

The contraints in this theory, the demand function and cost function ,D and C, are both located in the actor’s environment. He is assumed tofind the solution of equation(6). To do this,he mmust have perfect knowledge of these constraints, and must be able to perform the necessary calculations.

6.7.1 The limits of rationality

Theories of bounded rationality can beconstructed here in a variety ofways. Risk and uncertainty can be introduced into demand functions, the costfunction, or both. For example ceratain parameters of one or bothe of these functions can assumed to be random variables with known distributions. Then the assumptions of acto’s perfect knowledge of these functions will be replaced by the assumptio of perfect knowledgeof the distributions of these functions. Usually, This change in assumptions becomes more difficult than in the corresponding case of certainty.

Another way in which rationality can be bounded is by assuming that the actor has only incomplete information about alternatives. Fewer models have been constructed to deal with this situation than in which he has incomplete information about consequences.

Finally, rationality can be bounded by assuming complexity in the cost function or other environmental constraints so great as to prevent theactor from calculating the best course of action.

6.7.2 Alternatives to the classical goals

The classical theory can be modified not only by the above mentioned ways but also by altering the nature of given goals. Some modern theories of the firm deolart from the classical theory, not along any of the dimensuins mentioned above, but by postulating different goals from the classical goal of maximaisation.

Baumol, for example, has developed a model in which the firm maximizessales subject to the constraint that should not be less than the specified “ satisfactory” level. According to this thoery of Baumol, equation (6) in classical model should be replaced by :

(6’)

Subject to the constrain that

(7)

It may be observed that the informational and computational requirements for applying Baumol’s theory to concrete situations arre not very different from the requirements of the classical model.

6.8 Approaches to rational choice in chess

A number of the persons who have engaged in research on rational decision-making have taken the game of chess as a microcosm that mirrors intresting properties of decision-making situations in the reak world. The research on rational choice in chess provides some useful illustrations of alternative approches to rationality.

The problem confronting a chess player whose turn is to move can be interpreted in either of two ways.

1. It can be interpreted as a problem of finding a good strategy – where “strategy” means a conditional sequence of moves, defining what move will be made at each successive stage after each possible response of the opponent.

2. The problem can be interpreted as one of the finding a set of accurate evaluations for the alternatives moved immediately before the player.From the classical point of view, these two problems are not distuinguishable.If the player has unlimited computaional power, it does not matter whether he selects a complte strategy for his future behavious in the game, or selects each of his moves, one at a time, when it is his turn to play. For the way in which he goes about evaluating the next move is by constructing alternative complete strategies for the entire future play of thegame, and selecting the one that promises the best return. This is the approach taken in the Von Neumann-Morgenstern theory of games.

6.8.1 The game-theoretical definition of rationality in chess

As Von Neumann and Morgenstern observed, chess is a trivial game. “…ifthe theory of chess were really fully known there would be nothing left to play”.

Now each player can specify an optimal strategy – a stratrgy that willguarantee him at least as good as outcome as any other – by specifyingwhich he would select at each branch in the tree whenever it is his move.

What “impracticality” meabs becomes more vivid when we calculate how much search would be involved in finding the game – theoretically correct strategy in chess. On the average, at any given position in a game of chess,there are about 30 legal moves - -in round numbers, for a move and its replies, an average of about 10x10x10 continuations.Forty moves would not be a unreasonable estimate of the average length of the game. Then there would be perhaps 10x10x10……120 times possible games of chesss. Obviously the exact number does not

matter: a number like 10x10x10…40 times would be less spectacular, butquite large enough to support the conclusions of the present argument.

Studies of the decision-making of chess players indicate strongly thatstrong players seldom look at as many as one hundred possibilities – that is one hundred continuation from the given position – in selecting a move or strategy. One hundred is a resonably large number , by some standards, butr somewhat smaller than 10x10x10…120 times ! Chess players do not consider all possible strategies and pickthe best , but generate and examine a rather small number,making a choice as soon as they discover one that they regard as satisfactory.

Before weconsider in detail how do they do it ,let us return to the classical model and ask whether there is any way in which we could make it relevant to the practical choice problem, taking account of the size of the problem space, ina agame like chess. One possible way would be to replace the actual problem space with a very much smaller space that approximates the actual onein some appropriate sense, and them apply the classical theory to the smaller approximate space.

Thus, the approximate scheme was not guaranteed to select the objectively best move, but only the move leading to the positions thatappeared best, in terms of these heuristic criteria.

What we refer to as “uncertainty” in chess or theorem proving, therefore, is uncertainty introduced into a perfectly certain environment by inability – computational inability-to ascertain the structureof that environment. But the result of the uncertainty, whatever its source, is the same: approximation must replace exactnessin reaching a decision. In particular, when the uncertainty takes the form of an unwidely problem – solving processmust incorporate mechanisms for determining when the search or evaluation will stop andan alternative will be chosen.

The optimal decision in the approximated world is not necessarily evena good decision in decision in the real world.

6.8.2 Limits to rationality in chess

Three limits on perfect rationality are there

1. Uncertainty about the consequences that would follow each alternative.

2. Incomplete information about the set of alternatives.

3. Complexity preventing the necessary computations from being carriedout.

Thus, chess illustrates how, in real world problem-solving situations,these three situations tend to merge.

If we describe the chess player as choosing a strategy, then his difficulty in behaving rationally - and the impossibility of his behaving as game theory says he should – resides in the fact that he has incomplete information as to what (strategies) are open to him. Hehas time to discover only a minute fraction of these strategies, and to specify the ones he discovers only incompletely.

Altenatively, if we describe the chess player as choosing a move,his difficulty in behaving rationally lies in the factthat he has only rough information about the consequences of adopting each of the alternatives (moves) that is open tohim. It would not be impossible for him to generate the whole set of his legal moves, for they seldom number more than about thirty. However, he can evaluatethem, even approximately, only by carrying out further analysis through the immense, branching, move tree. Since only a limited amount of processing time is available for the evaluation, he must allocate the time among the altenative moves. The practical facts of the matter arethat it is usually better to generate only a few of the entire set of legal moves evaluating these rather throughly,than it is to generate all of them, evaluating them superficially. Hence the good chess player does not examine all the moves open to him, but only a small fraction of them.

6.8.3 Satisficing and optimizing

The terms satisficing and optimizing, which we have already introduced, are labels for two broad approaches to rational behavior in situations where complexity and uncertainty make global rationalityimpossible. In these situations, optimization becomes approximate optimization – the description of the real-world situationis radicallysimplified untill reduced to a degree of complication that the decision maker can handle. Satisficing approaches seek this simplification in a somewhat different direction, retaining more of the detail of the real-world situation situation, but settling for a satisfactory, rather than an approximate-best, decision. One cannot predict in general which approach will lead to the better decisions asmeasured by their real-world consequences. In chess at least, good players have clearly found satisficing moreuseful than approximating-and-optimizing.

6.9 Bounded Rationality in Design

Complete designs,when they are finally arrived at, are not generally evaluated by comparing them with alternative designs, but by comparingthem with standards defined by aspiration levels. In the chess situation, as soon as the player discovers a startegy that guarantees a checkmate, he adopts it. He does not look for all possible checkmating strategies and adopt the best.

7. What do we mean `choice under certainty? '

Well choices under certainty are choices made when we know all of the options. Contrast this with choice under risk, where we don't know theoutcomes, but we know the probabilities with which they might occur. Finally, these both are different from uncertainty, where not even theprobabilities are known.

However, this is not the meaning of bounded rationality because, as already mentioned, an optimization under constraints approach does notrelieve social actors of the heavy burden of homo economicus. Herbert

Simon, who primarily studied behavior in organizations, is considered to be the father of the term bounded rationality (Selten, 2002; Simon,1945). According to Simon (for a summary see Simon, 1993), human agents have only limited information processing capabilities and are therefore not able to perform perfectly rational decisions (in the sense of unbounded rationality). In particular, their knowledge to evaluate consequences and alternatives of possible decisions, lacks adeeper understanding of all relevant factors. Simon in particular differentiates between an external uncertain world and our innate bounded cognitive capabilities: “Rationality is bounded when it fallsshort of omniscience. And the failures of omniscience are largely failures of knowing all the alternatives, uncertainty about relevant exogenous events, and inability to calculate consequences.” Furthermore Simon points out that our cognitive limitations have the consequence that we will not reach optimal decisions; instead we use satisficing decisions as our personal (and more realistic) yardstick. Simon states: “A decision maker who chooses the best available alternative according to some criterion is said to optimize; one who chooses an alternative that meets or exceeds specified criteria, but that is not guaranteed to be either unique or in any sense the best, is to satisfice.” Thus, Simon’s concept of bounded rationality and satisficing in a nutshell would be the following statement: a perfect and ideal solution might exist for our problems, but be-cause of our bounded mind we are not able to conduct the necessary cognitive steps to reach this goal. Paying respect to this fact, we use satisficing decisions and systematically deviate from ideals of unbounded rationality. With this strategy, we can reach satisfactory but not perfect outcomes. This approach is supported by the theory of the second best (Lipsey & Lancaster, 1956). According to this classic theory, simply getting closer to an optimization, does not necessarilylead to better overall decisions. Following this theory, it pays if a single condition cannot be reached (e.g. to high costs, lack of skills), to also depart from the other conditions and to reach this way a second-best outcome (i.e. a satisficing outcome). In Simon’s view it is important to recognize that social sciences are always bothnormative and descriptive, compared to natural sciences that are only descriptive. Thus, we cannot simply describe the decision making of humans without having a normative standard in mind. Simon sees social sciences in the same vein as engineering, where the goal is not only to describe how a bridge is build, but to construct the most efficientbridge possible with the available resources. This fundamentally distinguishes Simon’s view from classic economists like Savage (1954),who mainly use normative methods without considering actual human

behavior (Heukelom, 2007). One of the major credits of Simon’s work isthat he always insisted that these postulates (like the theory of subjective utility [SEUtheory]) might not fit empirical data. In one of his last publicationsSimon resumes his research program on bounded rationality: “Bounded rationality is simply the idea that the choices people make are determined not only by some consistent overall goal and the propertiesof the external world, but also by the knowledge that decision makers do and don’t have of the world, their ability or inability to evoke that knowledge when it is relevant, to work out the consequences of their actions, to conjure up possible courses of action, to cope with uncertainty (including uncertainty from the possible responses of other actors), and to adjucate among their many competing wants. Rationality is bounded becausethese abilities are severely limited. Consequently, rational behaviourin the real world is as much determined by the “inner environment” of the world people’s minds, both their memory contents and their processes, as by the “outer environment” of the world on which they act, and which acts on them.” The tricky part to describing human decision making properly, stems from Simon’s (1990) above mentioned scissor metaphor.

In line with this picture, the term bounded can be interpreted in two ways. On the one hand management scholars may consider limitations of the environment, like information costs (for example, a survey to findout employee satisfaction can be costly) or the difficulty of even gathering specificinformation (for example, which products are used by a competitor in order to conduct a benchmark). On the other hand, management scholars may consider cognitive limitations of the human agent, for example limitations in memory storage and imperfect evaluation of statistical information by the management.

However, Simon did not start a research program with empirical research on these issues. As a consequence the concept spread in different research areas with different meanings and different approaches to incorporate bounded rationality into other research programs (Klaes & Sent, 2005). In behavioral economics, for instance, bounded rationality is more in line with mainstream economics as Simonhad in mind. Looking at the current state of behavioral economics Sentstates: “Simon’s ideas are missing from the more recent development”.

According to Sent behavioral economics mainly relies on the heuristic and biases program from Kahneman and Tversky.

8. Heuristics and biases

8.1 Research on heuristics and biases

This program was to a large extent initiated in the late 1970s and early 1980s by the research of Daniel Kahneman and Amos Tversky. They showed that human cannot be accurately described in the terms of unbounded rationality, similar to the computer metaphor, as mentioned in the previous section. The central point of their position is that our judgments and decisions systematically depart from ideals of logicand probability theory and, thus, from rational behavior. Logic and probability theory laid the ground on which the homo economicus and its prototype of utility maximizer stand and which served as the dominant economic research program until the 1970s. The main thrust ofKahneman and Tversky’s research revealed EUT limitations in various scenarios within this approach, however, the axioms of unbounded rationality remained as yardsticks and as a normative frame .Kahneman and Tversky supported their bold hypothesis with a series of experiments, which showed that biases in decisions under risk occur when people estimate subjective values and probabilities inthe light of gains and losses, and with respect to a subjective point of reference. These results can be summarized in two classic graphs, which provide a descriptive picture of judgment under risk and the foundation of what is called prospect theory (see Fig. 1).

Fig.1. The value function of prospect theory.

For example, the subjective value of a five hundred Euro win is smaller than the subjective value of a five hundred Euro loss. This bias is not only valid for monetary incentives, consistent with Kahneman and Tversky (2000), it also explains why people fall victim to framing effects and are prone to loss aversive behavior. Gains and losses, as central reference points (and not final levels of wealth), are key issues when people regularly overestimate small probabilities of potentially positive (e.g. buying lottery tickets)and negative outcomes (e.g. signing an insurance). Compared to these unlikely events (see Thaler & Sunstein, 2008) medium to large probabilities are typically underestimated, as described in the weighting function. Other than these deviations from rationality axioms, people also fall victim to their limited cognitive capabilities. These limitations lead people in the view of Kahneman and Tversky to use heuristics which, due to their simplifying character, lead to a list of errors in our decision making. Heuristicsare, according to Kahneman and Frederick (2002), frequently used shortcuts in which a difficult question is answered by substituting itfor a shorter one. One of the first descriptions of heuristics appearsin 1972, when Kahneman and Tversky state: “People do not follow the principles of probabilitytheory in judging the likelihood of uncertain events. Apparently, people replace the laws of chance by heuristics.” In addition to the tendency to overestimate small risks and to underestimate high risks, the most mentioned and cited errors and biases by Kahneman and colleagues are: conjunction fallacy, overconfidence bias, base-rate fallacy (and base-rate neglect), representativeness, availability and anchoring.

NOTE:- (In decisions under risk the possible probabilities are known (e.g. the flip of a coin). In decisions

under uncertainty the probabilities, however, remain unknown. In the earlier versions prospect theory

only addressed decisions under risk, decisions under uncertainty were implemented later into the

theory (Kahneman & Tversky, 2000). This is relevant for settings of management decisions, since here

risks are

often unknown and thus managers face a high level of uncertainty)

In order to maximize certain criteria people do so based on probabilistic information. Thus, the mind represents information in a probabilistic way when making decisions according to this framework. This assumption is important because it allowed Kahneman and Tversky to provide both a normative and a descriptive theory of choice (and its differences); a distinction thatwas not well established particularly in economics, where no special attention was given to developing an explicitly descriptive model. This outline is summarized by Heukelom (2005): “If man is considered to be an intuitive statistician, and if the world is considered to present itself to the individual in terms of probabilistic informationof uncertain events, a decision that deviates from the theoretically optimal decision becomes a failure of the system, or an error of judgment.” The heuristics and biases program claims bounded rationality as well as its proper theoretical foundation (e.g. Camerer, 1998), since this program stresses errors that result from our limited cognitive capacity. Deviations from the ideal of optimization are viewed by proponents of this paradigm as failures that, at least up to a certain level, can becorrected and even must be corrected in order to gain good (and rational) decisions.

8.1.2 Critical assessment

The heuristics, resulting from our limited cognitive capabilities, as described by Kahneman and Tversky, are not considered as efficient solutions, rather (in the best case) as second best and quick-and-dirty techniques compared to a proper logical analysis. Many empiricalfindings corroborate the robustness of these deviations and show thatsocial actors systematically deviate from the standard model of economics (DellaVigna, 2009). However, there is a current debate as to how this program can be put in line with Simon’s view of an ecological rationality that includes both cognitive and environmental factors (Lopes, 1992; Sen, 2002).

The criticism stresses that heuristics and biases can be understood more as a vague label than as a proper and testable model (Gigerenzer,1996). Moreover, to classify behavior into categories of rationality orirrationality solely depends on the norms of rationality (Cohen, 1981). Advocates of the fast and frugal program highlight that using unrealistic yard sticks for human decision making will always lead to a picture of a highly biased and misguided agent (Todd, 2007).

With respect to the norms used by the heuristics and biases program McKenzie (2003, p. 405) thus notes: “When a rational model fails to describe behavior, a different rational model, not different behavior, might be called for.” By the same token Gigerenzer ́s earlier critique (1997) of the heuristics and biases program extends this statement by arguing that “it should be clear that the single trenchant conclusion reached by heuristics-and-biases program, namely that people are all too bad at reasoning, is itself, to a large degree, an illusion fostered by all-too-narrow norms of sound reasoning.” Moreover, some early critics point out that methodologicalissues play a crucial part, when demonstrating human irrationality (Cohen, 1981). From a philosophy of science point of view, some of thepropositions of this program are questioned for making predictions that explain every possible outcome. For example, the same bias (representativeness) accounts in the hot-hand fallacy and in the gambler’s fallacy for two completely different (even contrary) outcomes (Ayton & Fischer, 2004). The gambler's fallacy describes the illusion of a player that, for example, in game of roulette after a series of black the chances increase for red in the next round. The hot-hand fallacy is the other way round and is used as an explanation,for example, in basketball, arguing that a player who already scored several times is therefore more likely to score the next time. Resultslike this question the explanatory power of the heuristics and biases program, because using ex-post explanations is neither feasible (Popper, 1959), nor an adequate way for making possible prescriptive conclusions and raises the question of “how persuasive is a good fit”(Roberts & Pashler, 2000). Proponents of naturalistic decision making (NDM) like Gary Klein (Klein, 1993; Lipshitz, Klein, Orasnu, & Salas, 2001) criticize the heuristic and biases program. The goal of NDM researchers is to investigate the intuition of experts, using mostly recognition-primed decision techniques, which they also claim is in line with Simon’s (1993) suggestions regarding explanations of

human behavior. Like Gigerenzer and col-leagues they point out that intuition works as a fruitful way to make decisions. NDM advocates stress that their program is more in linewith an ecological view of human behavior, than with an emphasis of biases, as laid out by the heuristics and biases camp.

However, it should be noted that the heuristics and bias program revealed important insights in human JDM. We can now see more clearly that cognitive restrictions cause human errors and mistakes. The use of heuristics is a human strategy to reduce cognitive effort and needsto be studied more thoroughly. Moreover, this program stimulated research which tries to show the merits of heuristics in human JDM(judgment and decision making): the fast and frugal heuristics program.

8.2 Heuristics and ecological rationality

8.2.1 Fast and frugal heuristics

As an alternative to the heuristics and bias program, and to cope witha highly complex and uncertain environment and with limited time and knowledge, Gigerenzer and colleagues (Gigerenzer et al., 1999) proposed a number of heuristics as strategies in making inferences. These, by definition simple (Shah & Oppenheimer, 2008), heuristics useevolutionary based skills to adaptively make use of environmental structures. Since their working principles are rooted in the environment, their process follows an ecological rationality rather than an unbounded (economic) rationality. With this move heuristics are shifting away from the negative reputation they earned through much of the behavioral decision research. This fresh look at heuristics describes their quality depending upon their ability to interact with the environment and is in line with the original Greek meaning of the term heuristic, which can be translated as “serving to find out” (Gigerenzer & Gaissmaier, 2011).

Gigerenzer and Gaissmaier (2011) define heuristic as “a strategy that ignores part of the information, with the goal of making decisions more quickly, frugally, and/or accurately than more complex methods.”

This concept of judgment and decision making has some typical characteristics:

•It is a multiple strategy approach which is searching for different strategies according to the environment, task and cognitive capabilities (adaptive toolbox).

•Heuristics deliberately ignore information and this is operationalized by Shah and Oppenheimer: “1. Examining fewer cues. 2. Reducing the difficulty associated with retrieving and storing cue values. 3. Simplifying the weighting principles for cues. 4. Integrating less information. 5. Examining fewer alternatives” (Shah & Oppenheimer, 2008).” It is important to mention that not all five aspects are necessary components of a definition of heuristics, but as a minimum one of thefive has to be fulfilled.

•Gigerenzer and Gaissmaier attribute goal achievement to heuristics – more quickly, frugally and/or accurately – which is redundant as frugal hints again to information reduction. Moreover, it is problematic to enclose goals into the definition because it does not account for what happens if a strategy does not reach one goal or all three goals.Consequently, goals of heuristics should be looked at separately and the definition should rely on the criteria by Shah/Opppenheimer. It isalso important to highlight that heuristics in Gigerenzer ́s view are neither as-if models of optimization, nor solely descriptions of results – they are rules that describe the (problem-solving) process, in a fast and frugal fashion. They also fundamentally depart from concepts of rationality, like unbounded rationality and optimization under constraints. Furthermore, they go beyond the satisficing heuristics ofSimon (1979), because they are not necessarily second best choices – they can even outperform complex calculations under certain circumstances.

An example will illustrate how heuristics are seen in this program.

One of the heuristics in the adaptive toolbox is the take-the-best heuristic (TTB), coined by Gigerenzer and Goldstein (1996). We can describe this heuristic with five steps:-

1. The first step is pure recognition of the cue. For example, if a social actor is asked which company produces more cars out of Volvo orKhodro, she will choose Volvo, simply because she has heard of it.

2. The second step is the search for the values of the cues.

3. The third step uses the discrimination rule. A cue discriminates ifone has a positive value and the other has not. This can be illustrated again with the car example: is one type available at carscout24.com (positive cue value) and the other not (negative cue value)?

4. The fourth step is the cue-substitution principle and it states that if the cues do not discriminate go back to step two.

5. The last step is called maximizing rule for choice and it states that if no cue discriminates then choose randomly. The important pointin this lexicographic search strategy is that leaving out information can actually be helpful in making good inferences (compared to algorithms like multiple regression analysis or tallying for example).In order to make accurate predictions and to guide the decision maker,heuristics necessarily have three elements: a search rule, a stopping rule and a decision rule.The ideal environmental structure to make simple heuristics like TTB work in terms of efficiency and robustness,is mathematically described by Martignon and Hoffrage (1999). This environment is characterized by non compensatory information. This means that the cue weights (meaning how high the validity of the cue is) exponentially decrease, for example from 1.0 to 0.5 to 0.25 to 0.125 (...). In such an environment no other algorithm (like Dawes’s rule or Minimalist) can outperform the TTB heuristic. The nextcrucial feature of heuristics is the distinction between fitting and predicting. It is obvious that in order to explain the past ex post (fitting), it is helpful to use as many cues as possible. This may be the reason why we have got better at understanding the causes of the recent financial crisis; besides all the ad-hoc/post-hoc explanations that are not of any use. Despite all the good explanations of the past, there are serious problems to using these explanations to predict the next big crash ex-ante. One reason for this is called “noise” by cognitive scientists and refers to information that is

either redundant or that has no predictable value at all (neverthelessit can be expensive in terms of time and money). The danger of including too many informational cues of this type is called overfitting.

Other heuristics, which are supposed to be included in the adaptive toolbox and that are empirically supported are: recognition heuristic,default heuristics and earlier heuristics such as tallying, tit-for-tat, imitate the successful (for an overview of different heuristics including empirical findings see Gigerenzer & Brighton, 2009). The efficiency of heuristics is not a brand new subject for all fields of study and is, for instance, well documented in social contexts, where tit-for-tat as a fast and frugal strategy was shown to be highly efficient (Axelrod,1984), as well as simple rules (clearly evolutionary based) like imitate-the majority (Boyd & Richardson, 2005) or imitate-the-successful (Boyd & Richardson, 2005). These social heuristics use evolved and learned capabilities and are reducedto a minimalist set of building blocks.

8.3 Bounded minds and ecological rationality

Why do some heuristics work quite efficiently? According to Gigerenzerand Todd the working principle (that also accounts for the less-is-more effect) behind efficient heuristics is ecological rationality (Gigerenzer & Todd, 2008). In line with Berg (2010) a decision or judgment is ecological rational when itcan exploit structures of the environment. This exploitation works systematically and , thus, it is not a random process. In addition, the simplicity of the heuristic structure guarantees its robustness, which is shown in the ability to generalize dynamic and uncertain environments.This is of interest to management research because we have to keep in mind that the use of heuristics is not intended to fit a data set of the past. Contrary, heuristics are used to make predictions in a highly uncertain world. Nevertheless, is ecological rationality different from bounded rationality? Chase, Hertwig and Gigerenzer (1998, p. 212) describe ecological rationality and bounded rationalityas separate constructs: "We argue that to discover how the mind works,and how well, we need to understand how the mind functions under its own constraints – its bounded rationality – and how it exploits the

structure of the social and physical environments in which it must reach its goals – its ecological rationality.” Hoffrage and Reimer (2004) use the same distinction and add to this approach that “Models of ecological rationality describe the structure and the representation of information in actual environments and their match with mental strategies, such as bounded rational heuristics.” The interaction of adaptive minds with an uncertainenvironment (as the core element of ecological rationality) is also mentioned by Vernon Smith (2003), who points out in his Nobel MemorialLecture in 2002: “Ecological rationality uses reason to examine the behavior of individuals based on their experience and folk knowledge, who are ‘naïve’ in their ability to apply constructivist tools to the decisions they make; to understand the emergent order in human cultures; to discover the possible intelligence embodied in the rules,norms, and institutions of our cultural and biological heritage that are created from human interactions by not by deliberate human design.People follow rules without being able to articulate them, but they can be discovered.” The fruitful interplay of cognition and environment may also be illustrated by what is called situated cognition (Robbins & Aydede, 2009; and for a discussion of situated cognition in management research see Wrona, 2008). Followers of this program point out that the context (respectively environment) is not a passive entity which is perceived by social actors. Rather, situated cognition is embedded in the context and, especially in social contexts, it is described as a reciprocal interaction with individual cognition. Scholars of ecological rationality and those of situated cognition agree that context may influence cognitive processes and that these processes definitely use elements of the context. Another aspect of agreement is that situated cognition has a broad definition of what constitutes context, which isin line with ecological rationality. For example, both see social behavior as relevant context (Hertwig & Herzog, 2009). However, compared to ecological rationality, approaches of situated cognition do not tell us how individual minds exploit the structures of the context. The cognitive structures described by situated cognition-oriented scholars are seen to be directly influenced by the context. For ecological rationality this is not the case, because the cognitive structures (e.g. working memory) are rooted in an evolutionary development and thus do not object to a direct transformation.

8.4 Heuristics for management decisions

In the following we want to give some impressions of how fast and frugal heuristics might work in two areas of management research. We indicate possible research by reflections on bounded rationality, descriptions of situations and scenarios. They are intended as first impulses for researchers who wantto further investigate decision principles in their disciplines. Therefore, the empirical evidence is transferred from other studies and is not exclusive to this field. Management controllers – unboundedrational? German scholars developed different concepts of management control.

In one of these concepts the main function of management control is seen by Weber and Schäffer in securing the rationality of the decisionprocess and the final decisions made by the top management (J. Weber &Schäffer, 2008). After reviewing all the limitations people face due to their bounded rationality, it seems more than plausible to assume that these limitations also affect management controllers in their daily tasks. In the light of these findings, how are they able to secure rationality of the top management? One implication from the above statement may be that management controllers are able to secure rational decision making because they have a well learned capability to recognize and evaluate cues from their environment. Therefore, the normative stance of management control gives room for two interpretations.

1. In line with the heuristics and biases program, management controllers secure rationality because they know of the biases and, hence, they are able to make them disappear. This leads to the pictureof management controllers as experienced experts in human JDM who are devoid oferrors and biases.

2. Management controllers tend to be ecological rational since they systematically exploit structures of information in the environment (see Goldstein & Gigerenzer, 2002).

In our view, both interpretations have their merits. However, it is anempirical matter and it has yet to be shown in studies how management controllers act in reality. Despite these two opposing views it seems appropriate to assume that management controllers do not have privileged access towards rational decision making. Research should focus on descriptive models of how management controllers assist in decision making. When, for example, do they use heuristics? What kind of search strategies do management controllers use and how can they becompared to those of managers? Considering, for example, an investmentdecision scenario, where management controllers have to calculate the value of several performance indicators, empirical studies indicate that an information overload can lead to a poor decision quality (Volnhals & Hirsch, 2009) or at least to an inefficient use of the available information (Basel, 2010). It is very likely that this danger can be reduced usingfast and frugal heuristics like take-the-best in management accounting. In this way a management controller can avoid wasting precious time and money on evaluating too many cues and only base his decision on one single cue with the best validity. Of course requirements like non-compensatory cue weights (Martignon & Hoffrage, 1999) ideally have to be given in order that this kind of strategy canoutperform more complex calculations. Here as well we have to keep in mind that very little information in management settings is free: benchmarks or employee surveys usually cost both time and money. Even if more data of this kind is gathered it still can be found to be useless or misinterpreted for several reasons. This seems to be a promising field of research because management controllers often face a highly uncertain and dynamic environment. Imagine the following scenario :-

Social actors want to buy a new car and the only important criteria they take into account is life expectancy. Since Scandinavian cars have an excellent reputation regarding this kind of criteria they finally end up deciding. From the viewpoint of the information function, management controllers are labeled “chief information officer” (Link, 2002). This and the following story are slight adaptations of Gigerenzer (1991) between a Saab and a Volvo. First, they take a look at a magazine called “Cars and Technique” and see that in a big sample of over five hundred cars Volvo scored slightly higher than Saab. They are happy and think that they have made a good decision, having decided on the Volvo. Unfortunately, the day they planned to meet the car dealer they meet acar-loving neighbor on the s

treet. He tells them that his brand new Volvo just broke down. Now, dothey still buy the Volvo? According to heuristics and biases they should do so, since the car of the neighbor just counts as n =1 compared to the over five hundred cars in the test. If they now switchto Saab they are, according to this approach, a victim of base rate neglect bias. However it often seems that social actors make this baserate neglect with good reason. A short episode from our human history might illustrate why personal statements are sometimes preferred over large number statistics (for an overview in evolutionary based approaches see Cosmides & Tooby, 1997). Imagine humans living with their tribe in the middle of a rainforest. As their children grow up they teach them one survival skill (since they are busy hunting they only have time for a single skill): either climbing trees or swimming.The wise medicine man tells them that last year, of all the known tribes in the forest, twelve children died after falling from a tree, but only two had been killed by a crocodile during their swimming sessions. Following this advice, they are just walking down to the river when they meet a neighbor who tells them that he heard rumors about a big fat crocodile in their area. Again, social actors probablyfollow the single opinion. This might be explained with accessibility of the event, but coming back to “Simon’s scissors”we conclude, that the second opinion fitted better to their personal environment and they may make smarter decisions if they take into account ecological rationality (i.e. exploiting their social environment). In this case ecological rationality may reach its better fit and have the possibility to be more adaptive (they have had a dialogue with their neighbor) based on a form of competence trust that seems to be related to the closeness of the source and the non-commercial interests of it. This social heuristic which is of particular interest for marketing aspects could look like the following: trust the person most that has the most similarities with you – like living in the same area, having the same preferences etc. and who is considered as competent in this field (dueto their specific access to some media forms etc.). This kind of TTB heuristic could possible outwit many large scale statistics for the same reasons as described above. The working principle behind an opinion leader could be that he has privileged access to media information.This information is ecologically correlated with the cue value (for example, the opinion leader is the moderator in a chat group and frequently recognizes when new issues come up). This also might be the working principle behind effective word-of-mouth propaganda (Oetting, 2009) and the effectiveness of opinion leaders

(Weimann, Tustin, van Vuuren, & Joubert, 2007), especially when we know certain base rate rates butnevertheless seem to ignore them.

8.5 Critique

The fast and frugal heuristics program has been established as an alternative to the heuristics and bias program. However, despite the many studies for the use of fast and frugal heuristics in real life and laboratory settings, and a trend towards more evolutionary descriptions of human judgment and decision making, there are several doubts regarding this program. For example, Oppenheimer (2003) points to the fact that actual decision making involves usually far more calculations, than the simple mechanism described by Gigerenzer and colleagues. This finding is alsopickedup by Hilbig and Richter (2010), who argue that within the description of fast and frugal heuristics, it remains unclear how adaptive heuristics are selected (the so-called strategy selection problem). In addition, they present empirical evidence that the recognition heuristic is only one applied strategy among many. Thus, Gigerenzer and Brighton’s (2009) claim that “a majority of participants consistently followed the recognition heuristic” is not fully empirically supported.

Hilbig (2010) furthermore highlights that the adaptive toolbox needs to be more précised on the process level of reasoning and that even more complex mechanisms do not enforce severe information costs. Secchi and Bardone (2010) even suggest that the idea of bounded rationality in general needs an update, because new technologies and social resources were not integrated in this concept, when it was coined by Herbert Simon in the 1940s and 1950s. These authors propose that scholars should switch to a new approach called “extendable rationality”, which would allow for integrating technological advances into our concept of rationality. Moreover, since ecological rationality requires some implicit learning function, how exactly this learning (of heuristics) can be measured, or even improved, remains vague. Thisis true in particular for settings that are more complex like organizations and elaborate decision issues, for example mergers and acquisitions. Because most empirical studies of heuristics focus on

small scale problems and well defined issues, future research should bear this in mind and aim to increase ex-ternal validity by investigating real scenarios (Salas, Rosen, & Diaz Granados, 2010). To conclude the different aspects of this critique,the challenges for fast and frugal heuristics can be summarized into two categories:

1. The first addresses the nature of heuristics itself: how many typesexist? Are individual differences important (Stanovich & West, 2000)? For example, Scoot and Bruce (1995) state that “decision styleis not a personality trait but a habit based propensity to react in a certain way in a specific decision context”. Are some people more ecological rational and thus more successful than others? Is this, forexample, one of the reasons why Apple is so successful, because managers of the company constantly adapt their products to a complex changing environment, where simple and intuitive rules are preferred over complex calculations?

2. The second challenge concerns the application of fast and frugal heuristics and is of special interest for management research. Are there normative/prescriptive consequences from working principles, like less-is-more, and how do management scholars have to re-think their assumptions regarding rationality and optimal decisions? Gigerenzer (2007) stresses that the fast and frugal program describes how real people solve real problems. Therefore, it seems important to collect as many empirical examples of applications in real management contexts. However, until now there have not been many studies of this kind. One of the few studies is that of Wübben and Wan genheim (2008), who showed that in retail marketing fast and frugal heuristics, predictingnon-active consumers, can be more efficientthan complex binomial distribution models. In a recent finance study by DeMiguel, Garlappi and Uppal (2009) they showed that a naïve 1/N rule of portfolio selection is not outperformed by more sophisticated methods. Nevertheless, the most important insight from the large existing body of research on bounded rationality is, to be successful, in a fundamentally uncertain world, managers sometimes have no other choicethan to rely on their adaptive capabilities in thinking and deciding. There is growing evidence that the human mind is equipped with efficient strategies for this search which stems from our evolutionary past .

9. Rationality in judgment and decision making - Dual process models as a unifying approach

9.1 Different concepts of rationality – are they mutually exclusive? In this concluding chapter we want to frame the different programs andshow their commonalities. Therefore, we do not stress their differences, but show how they might reasonably be linked. However, wewill focus on the concepts of bounded rationality and will not elaborate on the link to unbounded rationality, concepts which exist for instance, in economics.

Starting with the distinction made by Chase, Hertwig and Gigerenzer (1998) between bounded and ecological rationality. As already stressed, within ecological rationality it is of utmost importance to look at how the environment influences the tasks (c) and how the environment shapes and has shaped the cognitive capacities of social actors (b) (Elio, 2002). The key critique on the heuristics and bias program is its main reliance on the relationship between tasks and cognitive capacities of social actors (a). However, in our view both programs show us not only the pitfalls, but also the merits of human reasoning. Humans do not constantly err, because they use biased heuristics, nor do humans constantly engage in efficiently exploiting their environment – because they use fast and frugal heuristics. A more realistic picture looks like this: humans have an evolutionary past in which they constantly learned and adapted to their biological and social environment and this shaped their cognitive capacities. Forinstance, they learned to detect cheaters in social exchange situations which indicates a special heuristic for this important taskin a social environment (Cosmides & Tooby, 2005). In addition, humans are not error free and, even more importantly; they face a wide range of tasks in a modern technological environment. Research shows that ifhumans are not familiar with those tasks they make errors when trying to solve them but, if they were tutored a lot of those problems disappeared. Eventually, both programs could profit from a wider framework of reasoning.

10. Bounded rationality, behavioral economics, and organization theory

Over the same period in which rational choice theory has flowered, a parallel theory of bounded rationality in economics and organization theory has been developed by Herbert Simon, James March, Richard Cyert, Sidney Winter, Richard Nelson, Richard Thaler, and many others.Landmarks in this field include Simon’s Administrative Behavior (1947), March and Simon’s Organizations (1958), and Cyert and March’s Behavioral Theory of the Firm (1963), and Nelson and Winter’s An Evolutionary Theory of Economic Change (1982). (Recent perspectives are given by March 1994 and Simon 1997.) This stream of research emphasizes that individuals typically have limited powers of attention, memory, and calculation, and their decision-making behavior therefore departs from the model ofthe perfectly rational expected-utility maximizer. Boundedly rational agents satisfice rather than optimize—that is, they take the first available alternative that meets some threshold of acceptabilitydetermined by heredity or experience. They assume roles in organizations or social structures and then espouse values and employ decision-making rules appropriate to those roles—but the same agent may play different roles, depending on the decision context.1 They construct meaning and identity through their actions, not merely obtain adesired set of consequences. Organizational decision processes are often characterized by ambiguity rather than clarity of action, which isnot always a bad thing—it may be a source of innovation and adaptationto environmental change.

Although no one really disputes the premise that rationality is bounded, the membrane between rational choice theory and boundedly-rational choice theory has remained only semi-permeable. Rational choice theorists often add small amounts of noise to their models and/or use nonanalytic techniques such as simulation to investigate the behavior of less-than-perfectly-rational agents, but they still mainly rely on parametric models of optimizing or quasi-optimizing behavior by individual agents. Organization theorists occasionally

1

use mathematical models of optimization, especially when describing the “technological core” of organizations, but otherwise they tend to use a more interpretive vocabulary for describing organizational decision processes.

11. The implications of 'bounded rationality' for the

structure and conduct of public policy

“The concept of policy making drive the assumption of bounded rationality. Uncertainty, complexity, and weak selective pressure characterize those contexts”2. We as rational beings make sensible decisions and choices by understanding things logically. While making decisions, there are complex things which challenges and at times inhibits our capacity to understand everything there is to know beforemaking a decision. Limitation is not only of capacity but extends beyond human abilities and control. Limitations, may it be due to timeconstraints or inadequate information, and decisions are made to meet most but not all needs. A complete understanding of all the factors before making a decision is humanly impossible due to “human’s lack ofcognitive resources to optimize”3.One may analyze a decision taken by an individual or a local institution, to see it make sense and sound rational from an individual’s perspective, but from a broader point of view it may havedisastrous consequences. Decisions are made by people by choosing whatmakes most amount of sense to them, which is guided by their level of understanding, availability of time and preferences. Irrespective of his intelligence, a decision maker will be faced and must work under three unavoidable constraints: limited information, human brain capacity and time. This lack of cognitive resource to optimize, Herbert A Simon pointed out that “the relevant probabilities of an outcome will usually remain unknown with sufficient precision and our memories are weak and unreliable”4 This is called bounded rationality.The idea of ‘Bounded rationality’ was first proposed by Herbert A Simon, in his celebrated book Models of Man published in 1957 published by John Wiley and Sons, Inc., Unlike many theories, Simon proposed that “individual decision making is limited by many boundaries.” This bounded rationality as a concept, is opposed to unbounded rationality.2

3

4

Unbounded rationality assumes man as a reasonable being, knowing everything under the sun. Many models in social science are unbounded rationality variety. Bounded rationality as a decision making is a study of how people make decisions in an uncertain world. Bounded rationality as an idea has atleast three different meanings. And two of them do not associate or deal with the uncertain worlds. Those two want to study decision making under ‘risk’, where everything is known exactly including the probabilities. From an economist’s point of viewbounded rationality is the study of “optimization under constraints”. And then there is another school of thought, who deals with bounds by adding more and more parameters.Secondly, in psychology, it is allegedly an optimal answer to a simpleproblem, mostly by logical probability and people deviate from that because they pay attention to the content and not just the formalism and this deviation is interpreted as a bias or error and the explanation is because of our cognitive limitations.The first one emphasizes rationality and the second one irrationality.The third program which Herbert Simon represents is by asking the question ‘How do people make decisions when optimization is out of reach, in an uncertain world?’ What Simon presents is that decisions are made rationally only after evaluating all the choices and choosingthe satisfactory solution rather than an optimal one. This kind of decision makers, Simon called them‘satisficers’5. In decision making, satisficers would prefer and perhaps chooses that option, which goes on to address most needs rather than the perfect or ‘optimal’ solution. That is because every individual or organization is limited by his “schema”. Schema is a mental structure that dictates our conduct by simplifying gathered information and logically organizing it according to our cognitive ability. Schemas are interrelated with one another, which implicates our final decision. It is shaped based on one’s knowledge and capacity which assess and forecast the structure and conduct of a choice or decision. In public policy, schemas can be used to understand the reasoning behind policy solutions to policy issues. Schemas give us the insight to compare andunderstand what is best between the choices, facilitating the satisfier to reach a decision.For example, a group spends hours proposing and discussing policy measures for water management. The group discusses six choices of which the issue may have one obvious and rational answer as a solution, but final decision settle on solution C which may be totallyirrational to some who thinks solution A was the best and rational

5

answer to the issue. When we examine this carefully and understand theunderlying schema in the light of bounded rationality, the rationalityof choosing Solution C over A is that the decision maker was not able to rationally structure and associate solution A to the issue. But he by choosing solution C which met an acceptability threshold to providing a satisfying answer considering the cognitive limitation he is posed with, the decision maker demonstrates that his action is influenced by bounder rationality.Satisficing is not necessarily bad decision. Human tends to satisfice,”when there is an unlimited amount of information available and it is necessary to eliminate options, satisficing is beneficial because it helps the person making the decision effectively and efficiently reach a conclusion”.

Decisions therefore in policy process are influenced by many factors from individual’s ability to process ideas to his inability to cope with everything there to know about it. Factors like market regulations, personal and political philosophy, citizens with burning issues, media all contribute to the complexity of the structure and conduct of public policy.Bounded rationality, when applied to public policy decisions, policy makers, for reasons of above said constraints and as Simon points out the ‘cognitive factors’ can only make a best possible decision , rather than a perfect or optimal solution. Simon proposes that perfect decision to be made there is to exist, perfect circumstances including perfect information, which is practically impossible by human minds with its cognitive limitations to cope, and this has wide implications for the structure and conduct of public policy.The implication of this bounded rationality is that, if a policy makercomes up with a specific policy decision to address a particular problem, he would come up with altogether different policy solution tothe same problem if faced at an another time. Instead of applying the same optimal solution like the earlier they would for reasons of complexity of the situation and cognitive limitation, time constraintsand changed preferences policy maker will opt for heuristic approach, to come up with best possible solution which may or may not be an optimal solution. This implies that bounded rationality is a variable in decision making.When a policy is proposed its implications are understood with respectto the problem, the player and the policy itself. The public policy process when set into motion, at the foundation we have agenda settingfollowed by option formation, implementation and finally policy evaluation. The policy maker has to identify an issue as a policy

issue before he can define the problem. The Simon’s bounded rationality concept asserts that the problem definition is sometimes hindered by the cognitive limitations of the human brain, by preventing the decision maker to understand few crucial factors leading to that issue not being recognized as a policy issue, making it not a legitimate concern.Above agenda setting, next is policy formulation. During this step, the agenda that was set forth is looked at and evaluated by its effectiveness. In this pyramid of policy making, bounded rationality affects each level from identifying the problem to definition and evaluation of the policy, leading the decision maker to choose and settle on a decision which meets an acceptability threshold rather than the ‘optimal solution’.Decision making in public policy is about how policy maker act in their world. It is most importantly about the processes, how a policy maker studies, manages and influences the issue and the people. When a public policy maker arrives at an optimal decision to a specific problem considering and knowing all the probabilities, alternatives and consequences it is a small world decision, that can practically happen only in a laboratory. But we live in a real and what can be called as big world where problems are not straight forward and nor isthere any perfect solution. As satisficers who are optimal locally with limitation imposed by the cognitive ability there is no way for policy makers knowing all the probabilities, alternatives and consequences to a decision. Bounded rationality, as proposed by Simon is about this big world decision making. This boundary on the rationality influences the conduct of public policy by recognizing the“bottleneck” that exists in matters of information regarding possible alternatives, consequences and time constrains under which decisions must be made.The implications of bounded rationality for the structure and conduct of policy formulation regarding water management can be understood here. Water crisis in some places are a major and complex problem. Thecomplexity of water management is due to the combination of natural and human systems, which consists many interdependent elements, vast uncertainties and multiple stake holders. The issue is fairly unstructured and information scattered. Adding to this is uncertain knowledge, is disagreements on cognitive elements (Van de Graaf and Hoppe 19966 ; Hisschemoller and Hoppe 20017 ). The policy maker before arriving at a decision is faced with for consideration many factors

6

7

like below average rainfall, recurring drought conditions, different stake holders like households, agriculture and industry. In most realistic situation there is no way the policy maker can identify exact and full information and preference of water use to come up withan optimal solution. As Simon presented, “these are practical limitations to human rationality and these limits are not static”8 . Different actors choose different policy measures. For some privatization may be the best choice, for others it may be restrictionon water use, and others choose information technology measure, water infrastructure, water metering etc. A policy decision maker in this scenario where the available data is limited comes up with a decision from simplification heuristics such as elimination by criteria. This may sound irrational, but under the light of bounded rationality, irrational choices are not necessarily made in an unpredictable or unreasonable manner.Here the question that begs the answer is, “How does one ascertain theobserved choices are consistent with bounded rationality if the data is incomplete?” The answer is, when we observe the choice procedure from all possible problems, the information about the underlying preferences can be identified.Another example of implications of bounded rationality for the structure and conduct of public policy is United State’s President Barak Obama’s “Patient Protection and Affordable Care Act 2010”. Thosewho frame the issues to be addressed by policy often exert influence guided by their rationality, choosing as a satisficer which reflects theoretical or experimental assumptions. So when recently the Supreme Court of United States ruled ‘Obama Care’ to be constitutionally valid,we can see how even when data available is completely accurate or not, in spite of the quality of the information at hand, a decision was made using the resources currently accessible to the decision maker.Many rational actors contended that Obama Care policy to be a bad one and irrational. The ‘optimizers’ who oppose the Obama Care argue that the rationality calls for “Good public policy that would encourage health care consumers to shop around and purchase insurance policies that are best suited to their own needs and spending priorities”9. Whenmany States sued the federal government saying the government has no power to force them to buy something, the Obama administration does have the power to buy something, because they have the power to regulate commerce.10 This power to force is not legal because 8

9

10

Constitution allows it, but in other words Congress cannot make peoplebuy but rather it has the power to tax people. This new policy of mandatory health insurance can be thought as tax, where citizens are taxed for not being insured. The point I am making is that “cognitive limitations” are not specific to decision maker or decision making butthis cognitive limitation applies and influence the non decision makers, here the beneficiary and the Republicans. Not passing this bill may be the rational alternative or optimal solution to some, but “decision makers do not simply choose among alternatives, but they have to generate the alternatives in the first place”.11 In this Obama Care policy, the policy issue was to make the previously expensive health care services affordable and it was about achieving wealth redistribution. The rational optimal solution is to bring in ‘British single payer’ system. But, due to the complexity of the health servicebusiness in the United States and factors like any change easily impacting the economy of State, Obama Care was a decision, weighing all the options and alternatives and chose the option which addresses most needs of the goal rather than the ‘optimal’ solution. The flaws or irrational decisions in the Obama Care, which others claim like this policy weakens economic recovery or the ‘penaltax’ (individual mandate) which are insufficient to the task, are all generated as the product of ‘bounded rationality’. Because, the goal of the satisficer President Obama is not economic recovery or free health insurance but primarily a health policy which is affordable, prioritizing the needs of old and the sick and wealth redistribution. When many States sued the Obama administration for Obama Care being unconstitutional, it wasthe unreliable information and limited capacity to understand the wider interpretation the Constitution and their cognitive limitation to understand the goal Obama Care is trying to achieve against the odds, all generated due to the bounded rationality. Obama’s Obama Careis a goal oriented policy where achieving maximum goals and not perfect goal is the rational choice.As a boundedly rational being, Obama Care has taken into considerationthe cognitive limitation of the decision maker in attempting to achieve these goals. The long term risk of this policy like weakening the economy of the State, not addressing the issue of burdens faced byemployers with more than fifty workers, which are the tendency of satisficer to operate on goals sequentially rather than simultaneously, and satisficing rather than optimizing, clearly demonstrates the implications of bounded rationality for the structureand conduct of public policy.

11

11.1 Conclusion

Every decision maker is confronted with the cognitive limitations of human brain and other factors. There is no way a decision maker can have a complete understanding of a situation before having to make a decision. And the like is also true to the person who is not a decision maker as illustrated in Obama Care policy issue. Information,more or less, tends to change people’s decision. Every decision maker is limited by the schemas we have and other decisional limitation. Thecomplex the issue, the harder the problem requires more thinking, thereby increasing the cognitive load. And where there is such cognitive overload, we bring in our coping mechanism to arrive at a decision, choosing the one that most meets our goal than going for theoptimal solution. And this implies that, public policy is mostly an “attempt” to resolve an issue. The meeting of the required goal in public policy is greatly limited by the resources that allow it.In conducting public polices, it can be argued that there can be no optimal agents, but agents who are in some sense locally optimal at best. In the large world Simon’s bounded rationality implies that, public policy is conducted based on uncertainties rather than on certainties. The implications of our actions as locally optimal are based on bounded rationality, which is senseless to a rational agent, but completely sensible from that person making such decisions.The bounded rationality in public policy is closely related to time and space, exaggerating our present and paying far less attention to the past and the future.”Based on our understanding of our situation we endeavor to satisfy our own aims as best as we can before we move on to the next situation” ( Herbert A Simon).Without policy makers voicing their opinion, problems would never get solved. The policy maker in the United States cannot possibly be cautious of all the policy problems the world faces today before making a decision, which implies that there can never be a ‘optimal’ or ‘perfect solutions’ to policies. It is safe to conclude that every decision maker is boundedly rational, and “people always do exactly what makes the most amount of sense to them, in the context of the moment, with their current understanding”12. Hence, bounded rationalityalways imposes specific restraints on the structure and conduct of public policies.

12

11.2 Summary

Bounded rationality is the idea that in decision-making, rationality of individuals is limited by the information they have, the cognitive limitations of their minds, and the finite amount of time they have tomake a decision. It was proposed by Herbert A. Simon as an alternativebasis for the mathematical modeling of decision making, as used in economics and related disciplines; it complements rationality as optimization,which views decision-making as a fully rational process of finding an optimal choice given the information available. Another way to look atbounded rationality is that, because decision-makers lack the ability and resources to arrive at the optimal solution, they instead apply their rationality only after having greatly simplified the choices available. Thus the decision-maker is a satisficer, one seeking a satisfactory solution rather than the optimal one. Simon used the analogy of a pair of scissors, where one blade is the "cognitive limitations" of actual humans and the other the "structures of the environment"; minds with limited cognitive resources can thus be successful by exploiting pre-existing structure and regularity in the environment.

Some models of human behavior in the social sciences assume that humans can be reasonably approximated or described as "rational" entities (see for example rational choice theory). Many economics models assume that people are on average rational, and can in large enough quantities be approximated to act according to their preferences. The concept of bounded rationality revises this assumption to account for the fact that perfectly rational decisions are often not feasible in practice because of the finite computationalresources available for making them.

12. Conclusion

The theory of rational decision has undergone extremely rapid development in the past thirty years. A considerable part of the impetus for this development came, during and since World War II, fromthe attempt to use formal decision procedures in actual real-world situations of considerable complexity. To deal with this complexity

the formal models have grown in power and sophistication. But complexity has also stimulated the development of new kinds of models of rational decision that take special account of the very limited information-gathering and computing capacity of human beings and theirassociated computers.

One response to the concern with uncertainty , with the difficulties of discovering or designing alternatives, and with computational complexity has been to introduce search and information transmission processes explicitly into the models. Another response has been to replace optimization criteria with criteria of satisfactory performance. The satisficing approach has been most often employed in models where “heuristic” or trial-and-error methods are used to aid the search of plausible alternatives.

As a result of all these developments, the decision maker today , in business , government, universities, has available to him an unprecedented collection of models and computational tools to aid him in his decision-making process. Whatever the compromises he must make with reality in order to comprehend and cope with it, these tools makesubstantially more tractable the task of matching man’s bounded capabilities with the difficulties of his problems.

13. References

Arrow, K. J. (2004). Is bounded rationality unboundedly rational? Lee, C. (2012). Bounded rationality and the emergence of simplicity amidst complexity Herbert A. Simon , 1972 : Theories of Bounded Rationality Honorata Sosnowska , Warsaw School of Economics : A choice of pension fund as an

example of Bounded Rationality Till Grüne-Yanoff : Bounded Rationality Royal Institute of Technology, Stockholm,

Sweden Selten, R, 1999,What is bounded rationality? Perdersen A. P, Bounded rationality : models for some fast and frugal heuristics Albers, W. and G. Albers, 1983, On the prominence structure in the decimal system, in:

Decision making under uncertainty Aumann, R. and S. Sorin, 1989, Cooperation and bounded recall, Games and Economic

Behavior Becker, O. and U. Leopold, 1996, The bounds & likelihood procedure Cyert, R. and J.G. March, 1963, 1992 (second ed.), A behavioral theory of the firm, New

Jersey, Prentice Hall. Gigerenzer, P.M. Todd, and the ABC Research Group, Simple Heuristic that make us

smart, New York, Oxford UniversityPress. Kahnemann, D., P. Slovic and A. Tversky (eds.), 1982, Judgement under Uncertainty,

Heuristics and Biases, Cambridge University Press. Newell, A. and H. Simon, 1972, Human Problem Solving, Prentice Hall Roth, A. and I. Erev, 1995, Learning in extensive form games, Games and Economic

Behavior. Sargent, Th., 1993, Bounded rationality in macroeconomics, Oxford, Clarendon Press. Munro, Alistai : The Economics of Non-Market Goods and Resources Bounded Rationality

and Public Policy