Expanding the behavioral foundations of labor economics.
In this paper I focus on the first of these two components - the psychological model of the human agent - and the implications that alternative specifications of this model have for theory and research in modern labor economics. From the early 1930s through most of the 1980s, the dominant trend in the discipline of economics was to formalize and axiomatize the model of the human agent, minimize its reliance on or reference to assumptions and theories in psychology, and always look first to differences in environmental states (for example, prices and incomes) rather than differences in "tastes." This effort is most apparent in the development of general equilibrium theory and, in labor economics, in the highly influential work of Gary Becker.
These trends, reinforced by the widely cited methodological position of Friedman (1953) that realism of assumptions is not an appropriate basis for choosing among theories, resulted in a gradual withering of research that sought to explore the intellectual interface between economics, psychology, and other behavioral and social science disciplines (Solow 1997). In the past decade, however, a modest but discernible rebirth of interest has taken place in this type of scholarship, often under the rubric of "behavioral economics." Examples of contemporary economists writing in this vein on labor-related subjects are George Akerlof, Edward Lazear, Robert Solow, Roy Radner, Richard Thaler, Robert Frank, Oliver Williamson, Thomas Juster, and Jon Elster. Also highly influential, it should be noted, has been the work of psychologists Daniel Kahneman and Amos Tversky. Even Gary Becker, who formerly made it a cardinal principle (as stated by Mincer and quoted in Fuchs 1994:184) that "'changes in tastes' is the economist's admission of defeat," has in recent years argued for a more expansive consideration of psychological factors, as in remarks contained in his Nobel lecture (Becker 1993:401): "An important step in expanding the traditional analysis of individual rational choice is to incorporate into the theory a much richer class of attitudes, preferences, and calculations."(1)
These trends notwithstanding, the large bulk of economics research continues to eschew wider application and integration of psychology (and sociology) into economic theory. This is certainly the case in labor economics, which, I believe, has lagged behind other areas of economics in this task.(2) As an indication, of all the articles in the Journal of Labor Economics over a six-year period (1992-97), I found only two that substantively modified the rational choice model (Main, O'Reilly, and Wade 1993; Barron and Gjerde 1997).(3) Modestly more positive conclusions emerge if one canvasses the broader literature in labor economics, where works such as Akerlof (1990), Lazear (1991, 1995), Piore (1995), Lindbeck (1997), and Encinosa, Gaynor, and Rebitzer (1997) have in one way or another introduced greater psychological complexity into the model of the human agent.
Given this lacuna in the labor economics literature, and my desire to promote greater interdisciplinary research and collaboration, in this paper I undertake three tasks: (a) explication of the various psychological assumptions and processes embedded in the rational choice model, (b) discussion and critique of these assumptions based on modern research in psychology, and (c) illustration through numerous empirical examples of labor market outcomes of the gain in insight and explanatory power that may be achieved by incorporating a richer psychological model of the human agent into economic theory. The analysis focuses on three central psychological constructs - motivation, cognition, and emotion (affect)rain the context of a five-step model of the human behavioral process.
Four caveats must be noted. First, the relevant literature in psychology, as well as in the related fields of sociology and organizational behavior, is immense, and I make no pretense that this paper does more than scratch the surface. Second, the set of topics covered is selective and the emphasis given to each reflects my (possibly idiosyncratic) assessment of their relative importance and potential for illuminating substantively interesting aspects of theory and research in labor economics. Third, in places the tenor of the discussion is critical of mainstream, neoclassical labor economics research, and particularly the Chicago version popularized by Becker, Stigler, and others, for what I regard as an unduly rigid and narrow commitment to the rational choice model, but the intent is not to disparage neoclassical economists (or individual subgroups thereof) or their highly impressive and productive work in labor economics but, rather, to broaden the theoretical base in the field and promote a constructive, interdisciplinary dialogue. Finally, for every generalization made in this paper (for example, neoclassical theory does not take into account norms of fairness), I recognize and acknowledge that exceptions occur and counter-examples can be found. My claim, however, is that these counter-examples are indeed exceptions and not the rule.
The Rational Choice Model of the Human Agent
Labor economics contains various schools of thought, but without question the neoclassical school is the dominant theoretical paradigm in the field. A central component of the neoclassical paradigm is the rational choice model. A composite description of this model is as follows (Buchanan 1989; Mas-Colell, Whinston, and Green 1995):
- The appropriate level of analysis is the individual human being (methodological individualism);
- Human behavior is goal-oriented;
- The goal people strive for is maximization of individual utility, which is often equated with "satisfaction" or "happiness";
- Satisfaction or happiness is related to the fulfillment of individual "wants," which may be pecuniary or nonpecuniary;
- People have consistent, well-ordered preference relations among alternative items that satisfy wants, and these preference relations are independent of the actions and opinions of other people;
- In satisfying wants, people are confronted by a scarcity of resources, which forces them to choose among desired objects and activities;
- People exercise rational choice in that they consciously and consistently weigh the benefits and costs of alternative courses of action and choose in a logical manner the behavior that maximizes net satisfaction;
- Human behavior is externally motivated in that it is opportunities and costs, and differences or changes therein, that are the active agents driving the choice process and not changes in tastes;
- Internal human mental properties and processes (for example, the origin of wants, the structure of preferences, and the cognitive process of perception and information processing) are treated as a "black box" for purposes of predicting human behavior.
These assumptions collectively give rise to the standard model of the human agent in neoclassical economics research. Certainly a heterogeneity of opinion exists among labor economists about the degree to which one can modify and extend these assumptions and still remain "neoclassical," with some being more ecumenical in this regard than others, but I think the vast majority would assent to the general proposition that the assumptions of rational behavior, utility maximization, self-interest, and well-ordered preferences are core parts of whatever defines the essence of neoclassical economics. The validity of this statement, as well as that of the composite description of the rational choice model provided above, is attested to in these statements by well-known economists and scholars:
The economist's normal approach to analyzing consumption and leisure choices assumes that individuals maximize utility with preferences that depend at any moment only on the goods and services they consume at that time. These preferences are assumed to be independent of both past and future consumption, and of the behavior of everyone else. (Becker 1996:3-4)
Economics can be distinguished from other social sciences by the belief that most (all?) behavior can be explained by assuming that agents have stable, well-defined preferences and make rational choices consistent with those preferences in markets that (eventually) clear. (Kahneman, Knetsch, and Thaler 1991:193)
Mathematical modeling rose to preeminence in economics, and a sparse set of canonical hypotheses - Robert Solow has characterized them as greed, rationality, and equilibrium - became the maintained hypotheses in almost all branches of the subject. (David Kreps 1997:59)
In what follows, I shall for ease of exposition treat "rational choice model" and "neoclassical" as largely interchangeable terms, but with recognition that the latter is in fact broader and more inclusive than the former and that some (many?) labor economists who consider themselves neoclassical are not necessarily committed in their research to all features of the rational choice model.
Human Behavior: A Schematic Representation
To help organize the discussion and give it conceptual coherence, a useful exercise is to decompose the process of human behavior into five discrete steps. This is done in Figure 1. Not only is this five-step model of interest as a new conceptualization, it materially facilitates the transition from the economist's way of thinking to that of the psychologist, and vice-versa.
Starting on the left-hand side of the diagram, the first step in the behavioral chain is the assumed goal of the human agent. The goal is the objective or purpose the person seeks and is thus a central element of the motivational process (Locke and Latham 1990). In economics, as indicated earlier, the goal is assumed to be maximization of utility.
Since utility maximization is a global objective, the person must choose one course of action or behavior from among the available options (the opportunity set) in order to effectuate the goal. The choice process is shaped by the three factors represented in the second step of the diagram. These three factors are the self, wants, and the environment.
The self is the active agent in the decision-making process and is where the cognitive faculties necessary for decision-making are located, such as perception and reasoning. The self is also where emotions are experienced.
Wants are desires for specific items or activities that provide utility to the organism. Wants serve to both arouse and guide behavior in a specific direction in pursuit of the overall goal of well-being or happiness. Since wants are subjective feelings, they vary in intensity across items and activities. In making choices, therefore, people have to establish a preference ordering among the wants. This preference ordering is assumed in economics to satisfy two fundamental conditions, transitivity and completeness (Mas-Colell, Whinston, and Green 1995). Additional auxiliary conditions of continuity, convexity, and nonsatiation are also frequently assumed.(4)
The final factor that influences behavior in this part of the behavioral chain is the environment external to the individual. The environment defines both the opportunities available to the person and the limits or constraints on attainable outcomes.
The next (third) step in the chain of behavior is decision-making. Decision-making is a cognitive process that takes place within the person (self) and involves the twin processes of perception and reasoning. Perception and reasoning, as already noted, are most often assumed to be objective, omniscient, and effortless, although a number of studies weaken or in some way modify one or more of these assumptions.
The fourth step in the behavioral sequence is the initiation of overt action: motivated by the desire to maximize utility, the person acts to attain the outcome that mental deliberation has determined to be optimal. Within the limits imposed by environmental constraints and the probability distributions governing uncertain events, neoclassical models typically assume that the action undertaken by the agent will result in the desired outcome (that is, that the agent will reach the equilibrium solution to the constrained utility maximization problem, although possibly in terms of expected value or with a short-run lag in adjustment).
The final step in the behavioral chain is the change in the agent's utility resulting from the attainment of the optimal outcome. This change in utility may be either positive or negative, depending on whether the environment has become more accommodative or constraining, and thus represents a flow variable that alters the stock of total utility (satisfaction) existing at the beginning
of the behavioral sequence. Reflective of the equilibrium nature of economic models, the net increment or decrement of utility experienced at this stage is typically assumed to lead to no further adjustments in behavior.
Analysis and Critique
The next task is to critically examine each of the five steps in the behavioral chain illustrated in Figure 1 in light of wellknown and accepted psychological theories and empirical research. For each step, I attempt to identify one or more general principles from the psychological literature that can improve the ability of economists to explain and predict aspects of labor market behavior. In order to provide additional conceptual structure to the discussion, note that the various particularistic psychological concepts and theories discussed here pertain to one or more of three fundamental psychological processes: motivation (the forces that energize and guide behavior), cognition (the mental process of information gathering and processing), and emotion (subjective mental feelings, such as anger, love, and jealousy).
Utility maximization. The concept of utility was formalized by the eighteenth-century English philosopher Jeremy Bentham, although the basic idea goes back to the Greeks (see Georgescu-Roegen 1968; Baxter 1988). Bentham defined utility as "that property in any object, whereby it tends to produce benefit, advantage, pleasure, good, or happiness" (Mack 1969:86). Bentham often equated utility with "pleasure" and then contrasted it with its opposite sensation, "pain."
In order to blunt the criticism of psychologists and institutional economists about the unrealistic and primitive nature of the "economic man" model, as well as to remove any association between utility theory and the related, somewhat unsavory doctrine of hedonism (individual behavior is motivated by the pursuit of pleasure), neoclassical economists in the 1930s and 1940s attempted to reconstruct the utility concept to make it "psychology-free." The attempt was only partly successful (Sen 1973). Work by Pareto, Hicks, and Samuelson demonstrated that the major theorems of microeconomics can be deduced without explicit reference to utility, as long as the consumer (or worker) has a preference ordering that satisfies the conditions cited earlier for consistent choice.(5) In this version of utility theory, "utility" is nothing more than an ordinal ranking of preference over alternative items of choice, starting with the least preferred and extending to the most preferred (Harsanyi 1997).
Assuming that preferences are well ordered, the standard position of economists is that it is unnecessary for the purposes of deriving "if A then B" predictions to inquire any further into the nature of preferences, such as why one item is preferred over another or where this preference originates in the human psyche, and thus economics has little need or use for psychology.(6) This position, while having much to recommend it based on considerations of both parsimony and generality, has occasioned ongoing debate and controversy among economists, psychologists, and philosophers.
In this regard, it is first observed by critics that while neoclassical economists typically assert in their theoretical work, or when pressed on the matter, that "utility" is simply an ordinal ranking and has no substantive psychological content, they nonetheless proceed with considerable frequency, both at a textbook level (for example, Ehrenberg and Smith 1997; Borjas 1996) and in advanced theoretical work (for example, Becker 1976:92, 94, 145), to equate utility with psychologically loaded terms such as satisfaction, happiness, pleasure, and psychic income. Mixing these types of utility constructs (sometimes called "decision utility" versus "experienced utility") in economic theory is not merely an issue in semantics for, as Kahneman, Wakker, and Sarin (1997) show, they yield quite different predictions about human behavior.
A second line of attack focuses on Friedman's (1953) claim that prediction, not realism of assumptions, should be the basis for judging among theories, a position used by many economists to justify forgoing any deeper exploration of the utility concept. The counter-response is to ask whether the task of prediction would not be enhanced by a fuller understanding of the antecedents and determinants of human satisfaction or happiness. Thus, it is helpful to know that A is preferred to B, but would it not be even more helpful to know why and under what conditions A is preferred to B (Lagueux 1993; Sunstein 1996)? Likewise, even if the predictive ability of the rational choice model cannot be improved by inclusion of additional psychological considerations, these factors are crucial to the goal of understanding social phenomena.
Finally, by choosing to omit an explicit psychological base for the utility construct, economists are forced into a number of questionable methodological practices. That is, the variables that are the presumed source of utility in economic models are often introduced in a selective, ad hoc manner depending on the subject at hand. For example, a model of child care labor supply puts a "quality of child care" variable in the utility function (Blau and Hagy 1998), while racial prejudice is characterized as a "taste for discrimination," intrinsic sources of satisfaction from work as "psychic income," and the increase in preference for items such as music appreciation and addictive drugs that follows from repeated consumption as the effect of greater "consumption capital" (Becker 1996).(7)
One way out of these difficulties is to try to give greater psychological substance to the utility concept and see what new implications or predictions can be obtained. Since economists most often equate utility with satisfaction or happiness, the place to begin is with these two psychological concepts. Satisfaction and happiness are both emotional or "affective" states that are almost universally viewed as positive, often pleasurable, and hence desirable. The two are not equivalent concepts, however. Consumption of additional goods and services generally creates a sensation of increased satisfaction, but not necessarily increased happiness (per the homily that "money can't buy happiness"). Because happiness is a broader, longer-run construct than satisfaction, and has also been found in numerous "quality of life" studies to be the descriptor people most often use to assess and rank their subjective well-being (Lane 1991), it is the psychological construct adopted here as a potential measure of utility.(8)
Happiness, according to psychologists, is an emotional state or mood that reflects a subjective sense of well-being and satisfaction with life as a whole (Lazarus 1991; Averill and More 1993). More specifically, happiness arises from the interaction of an evaluative judgment that significant progress is being made toward the realization of important life goals, a perception that the self is largely responsible for the success in goal attainment, and the presence of a relatively benign existential background (that is, a positive event may have little power to induce happiness if it occurs in a negative life context, such as prolonged unemployment or a dysfunctional marriage).
This conceptualization of happiness, plus the findings from the extensive empirical research on the subject by psychologists, have several implications:
- Happiness has a large relational (or relative) component. The amount of utility an individual derives from, say, an annual salary of $V depends, in part, on how $V compares to the salary $X received by a referent other (for example, a co-worker). Another relational dimension is a comparison between goal and accomplishment. The utility derived from a salary increase of $Y, for example, is dependent on whether the person's goal or target increase for the year of $Z is above or below $Y.(9) A wage increase of 6% can make a person unhappy if he or she expected 10%.
The relational nature of utility has significant implications for economic analysis (Tversky and Kahneman 1991; McAdams 1992). If economics were truly a social science, the standard utility function in microeconomic analysis would have to be broadened to include relational variables, such as relative income. But this rarely happens (Killingsworth 1983, and Pencavel 1986, are good examples; exceptions include Frank 1985; Frank and Cook 1995; and Neubourg and Maarten 1994). Neoclassical economists typically justify this exclusion by maintaining that relational variables would significantly complicate the model without much gain in predictive power, or that academic research advances most rapidly through division of labor and the role of economists in this scheme is to focus on external, market-related variables, such as prices and incomes, leaving the more internal, mental set of variables to the attention of researchers in psychology, sociology, and so on (Solow 1997). For certain economic problems, this perspective is fruitful and appropriate. For a number of others, however, it seems much less so. Two examples are illustrative.
The first concerns the relationship between income and utility. In neoclassical labor supply models, the usual practice is to include in the utility function the human agent's level of real income and then postulate that the first derivative is positive (utility increases with income). But does it? The evidence, according to Easterlin (1995), is unambiguously "yes" only if the person's income increases relative to that of other people. If everyone's income level increases an equal amount, on the other hand, utility stays relatively unchanged. He adduces this conclusion from data for large samples of people from ten countries that show self-reported "life satisfaction" remained constant (and actually decreased in the United States) over the 1970s-1980s even as real income registered a significant gain. People who moved upward in the income distribution, on the other hand, reported increased happiness. The implication is that relative income, as well as absolute income, is likely to be an important determinant of labor supply. While the latter is included in nearly every economic model of labor supply, the former is included only very infrequently despite the fact that Duesenberry (1949) advanced a relative income theory of consumption a half-century ago. This appears to be a serious omission given that the consumption function and labor supply function are, in theory, duals of each other (Killingsworth 1983) - a concern strengthened by the frequent empirical discrepancy between the predictions of neoclassical labor supply theory and observed behavior (per the comment of Pencavel [1986:95], on male labor supply studies, that "the scientific procedure is surely to regard the theory as it has been formulated and tested to date as having been refuted by the evidence").(10)
As a second example, consider the issue of why workers join unions. The standard neoclassical approach is to posit a utility function for a representative worker that makes the utility of union services a function of the difference between the wage on a union and nonunion job - the larger the differential, the greater the inducement to join (Pencavel 1971; Abowd and Farber 1982). Field research reveals, however, that most workers who decide to seek union representation do so primarily in response to a "treatment" issue - that is, what they receive in wages (or promotions, respect, or some other desideratum) does not match what referent others receive or what the workers think they should receive relative to some subjectively determined standard or target (Brett 1980; Kaufman and Stephan 1995). Fifty years ago Ross (1948) coined the term "orbits of coercive comparison" to describe the relational nature of workers' wage expectations, but it and other interpersonal and inter-group relativities are typically given modest-to-negligible attention in microeconomic analyses of unions (for example, Hirsch and Addison 1986).
- Feelings of self-efficacy and self-esteem are fundamental to the experience of happiness. Not only is goal attainment important to happiness, so too is the perception that the self is the causal agent in the process. For this reason, people frequently report that it is not the goal itself that yields happiness but rather the positive feelings of self-worth that accompany successful achievement of the goal.
An application of this idea is to the subject of work. The standard microeconomic model of labor supply almost uniformly portrays work (at least on the margin) as a source of disutility and an instrumental activity people do only to earn an income (Killingsworth 1983; Burtless and Moffitt 1985). This may be called a "theory X" view of work (McGregor 1960). Early institutional labor economists, on the other hand, took a different, "theory Y" view. For example, Richard T. Ely (Ely and Bohn 1935:210) said that "work is the greatest of all blessings" because it contributes to people's sense of goal attainment. Recent empirical studies of life satisfaction also find that most people rate satisfying work as more important to personal happiness than the level of income, material possessions, and even most forms of leisure (Lane 1991; Juster 1991).(11)
The implication is that the neoclassical labor supply model may well misconstrue in certain fundamental respects the reason people supply labor. Yes, people do so in part to earn a living, but of equal or more importance for many is the psychological gratification work provides (or can provide if it is structured the right way). Why is this omission a problem? Consider the issue of work hours.
Managerial and professional employees have the longest work weeks among occupational groups. The standard interpretation of this pattern using neoclassical labor supply theory would be that these people choose to work these long hours as a response to the incentive of high wages and salaries.(12) An alternative explanation is that managerial and professional workers have a positive preference for work because of the nature of their jobs (for example, greater room for creativity). If such is the case, it is probable that predictions derived from neoclassical labor supply theory will be erroneous, while empirical estimates of income and substitution effects will most likely be biased by the strong correlation between tastes for work and rates of pay. Indicative is the statement of Robert Hall (1986:57),
Econometric work has almost universally found that both substitution and income responses of labor supply are substantially smaller in the experimental data than in survey data. That observation confirms the misgivings that veterans of labor supply estimation in survey data have always had - that wages and preferences favoring work are positively correlated in the population.
- Social norms and values are an important determinant of individual happiness. The concept of social norm is the most important theoretical construct in sociology (Coleman 1988). It is defined as (1988:140) "expectations about action - one's own action, that of others, or both - which express what action is right or what action is wrong.... Norms are accompanied by sanctions, either some sort of reward ... or some sort of punishment." Social norms, such as conceptions of equity, gender roles, and social status, are only infrequently included in labor economics models. Rather, the usual practice is to specify an individual's utility function in individualistic terms - an individual's utility is a function of his/her (say) wage but not the wage of other people or groups (but see Frank 1985; Elster 1989; Binmore and Samuelson 1994; Neubourg and Maarten 1994; Lindbeck 1997).(13)
The danger of omitting social norms from the utility function is easily seen (see Sunstein 1996 for many other examples). One of the central conclusions of the Hawthorne experiments of the 1930s, for example, is that employee work effort is significantly influenced by the norms of the workgroup as to what constitutes an appropriate level of output. Ignorance of this fact led early practitioners of scientific management to regard workers as irrational because they did not increase output in response to increased incentive rates (Whyte 1955). Although most labor economists study organizational pay practices without reference to social norms, an exception is a recent study by Encinosa, Gaynor, and Rebitzer (1997). They found evidence that social norms influence both levels of pay and work effort and, as a consequence, issues of efficiency in determining pay cannot be divorced from issues of distribution. They concluded (p. 40), "Our analysis suggests that the relationship between efficiency and distribution may be rooted in group psychology and sociology. If so, then the economics of organizations will need to carefully consider aspects of human behavior that have so far been largely overlooked."
Norms regarding procedural and distributive justice likewise heavily influence employee satisfaction with pay rates (as in the union joining decision), as is discussed more fully in a later section. Research also documents both that well-defined social norms concerning appropriate lines of work for men and women exist and that these norms play a significant role in occupational choice and patterns of occupational segregation (Blau and Ferber 1986; Glick, Wilk, and Perreault 1995).
The self. Next consider the triad of environment, self, and wants in Figure 1. Since the discussion in this paper concerns the model of the human agent, the environment is discussed no further.
Economic theory does not distinguish a specific psychological entity called the "self" (but see Thaler and Shefrin 1981). The human agent is assumed to have a goal of utility maximization, a set of wants and preferences, and a brain that calculates the optimal course of action for goal attainment. The "self" would be viewed by most economists as simply another name for the human agent and thus a descriptor that lacks independent theoretical content. Behavioral psychologists in the tradition of B. F. Skinner (1969) hold much the same view on the grounds that the "self" is a metaphysical concept whose existence and influence on stimulus-response relationships can be neither proven nor disproven (Burns 1979).
A large and growing body of theory and evidence developed by cognitive and "humanist" psychologists points in the opposite direction, however. This literature suggests that a concept of "self" is a meaningful construct, plays a significant role in shaping human behavior, does so through a wider range of "self-concepts than the self-interest motive emphasized in economics, and explains numerous actions that cannot easily be rationalized in terms of a theory of externally induced motivation (Deci 1975; Rosenberg 1979; Frank 1988; Lazarus 1991; Epstein 1993).
Psychologists define the self-concept as the composite view of oneself that is formed through direct experience and evaluations derived from interaction with significant others (Ban dura 1986:409). Part of the self-concept is based on objective characteristics (for example, gender and occupation), while another part represents subjectively formed impressions and opinions (for example, perceptions of worthiness and beauty).
The self-concept has several important ramifications for understanding work-related human behavior. For example:
- It provides a conceptual foundation for internally motivated behavior (Deci 1975; Deci and Ryan 1985; Kanfer 1992). Such behavior has little or no relationship to an external reward or sanction - it arises from a drive within the person and is performed for its own sake. In the language of economics, intrinsic motivation is endogenous while the extrinsic motivation assumed in the rational choice model is exogenous. The origin of this drive, according to Deci (1975), is the innate need of every person to feel competent and self-determining relative to his or her environment. This drive imparts both a persistence and dynamic growth to human behavior that equilibrium, stimulus-response models cannot easily explain. This defect of the economic man model was sharply attacked by institutional economists, such as Thorstein Veblen (1898) and John R. Commons (1934), and has resurfaced in contemporary institutional work, such as Lutz and Lux (1988).(14) The central idea in this literature is that humans not only respond to their environment, as in the rational choice model, but also actively try to control and shape the environment through the exercise of the human will, conscious purpose, and foresight and planning.
This idea can be applied to work effort. If work is a source of disutility (on the margin), as assumed in most economic models, workers have an incentive to call in sick, work at minimum speed, and loaf whenever possible. To prevent this, firms hire supervisors and managers to monitor employees and set up elaborate disciplinary procedures to provide deterrents and punishments - a mode of operation and set of calculations largely explicable in terms of the model of crime prevention developed by Becker (1976:39-85). It is widely argued in the management and organizational economics literature, on the other hand, that a cheaper and more effective method exists to elicit employee work effort and dedication (Lawler 1992; Levine 1995). That is to make the work interesting and self-managed, such as through autonomous work teams and job enlargement/job enrichment. When these practices are implemented, it is argued, work becomes much more intrinsically satisfying and workers become more internally motivated to work hard and attend regularly; as a result, firms no longer need as many supervisors and punitive disciplinary systems, and employee productivity reaches higher levels than can be obtained through manipulation of external rewards and punishments.
As with most of the examples cited in this paper, empirical evidence can be found that both supports and contradicts the superiority of a "behavioral" explanation (for supporting evidence, see Nicholson and Martocchio 1995). But one does not have to decide this issue to nonetheless affirm what is the main point being made here - that different models of the human agent give rise to markedly different predictions about labor market phenomena, and enough empirical evidence exists in support of a behavioral model that labor economists would be well-served to give it increased attention.
- The second ramification for explaining work-related behavior arises from the interaction of two other "self" concepts: self-esteem and self-efficacy.
Self-esteem is the normative evaluation of one's self and involves attributes such as personal worth, goodness, appearance, competence, and likability. It is formed from various types of evaluations, such as subjective self-appraisal, objective appraisal of one's record, subjective appraisal of what others think, and objective statements by others.
Self-efficacy is a belief that one is competent and exercises significant personal control over the direction of one's life and the events in it. Persons high in self-efficacy believe they can master most situations through personal initiative, while those low in self-efficacy believe they have little power to make a difference in various life events. Rotter (1966) has characterized the former group as having an "internal locus of control," meaning a strong sense that the self can influence the environment, while the latter group has an "external locus of control," meaning that they perceive the self has little influence on the environment.
An implication of self-esteem and self-efficacy is that research based on a neoclassical, rational choice model may seriously misrepresent the most effective solution to ending welfare dependency (Foster 1993). Typically, such a model leads one to focus on manipulating financial incentives associated with welfare programs, such as the implicit tax rate and benefit guarantee level, as the means to encourage greater work effort and job-taking. The negative income tax experiments revealed, however, that a change in financial incentives alone will lead to a disappointingly small movement of people from welfare to work (Munnell 1986). One explanation is that the people most likely to be on welfare are also those most likely to suffer from low self-esteem and low self-efficacy - feelings only likely to be reinforced by the welfare experience. Given this mind-set, a change in financial incentives elicits little positive response since welfare recipients feel ineffective and powerless to make their lives better even when given an incentive. Psychologists call this condition "learned helplessness" (Seligman 1976). The solution to welfare dependency, then, must include not only a change in financial incentives but also interventions that build self-esteem and self-efficacy, such as personal improvement and skill enhancement programs.
Wants. The third part of the bracketed triad in Figure 1 is "wants." Wants are desires for objects that, when consumed or experienced, yield that psychic form of pleasure economists often refer to as utility. Wants are important in the economist's model of human behavior because they energize behavior and direct it toward specific ends. Despite the importance of wants, neoclassical economists typically give scant attention to their origin and nature.
The crucial aspect of wants for economic theory is the assumption that they give rise to well-ordered preference relations among the items of choice in the opportunity set. As noted earlier, these aspects of preference include transitivity, completeness, continuity, convexity, and nonsatiation. Several of these assumptions are examined below.
Of these assumptions, the one that has attracted the most attention from both economists and psychologists is transitivity (Thaler 1991; Slovic 1995). Psychologists Kahneman and Tversky, in particular, have conducted numerous pioneering experiments and have found that the transitivity condition is often violated even in relatively simple choice situations (see the references in Kahneman 1994). One of their findings (Kahneman and Tversky 1979) that is particularly pertinent to labor economics is that people systematically overweight small-probability events and underweight large-probability events. As noted by Viscusi (1993), this implies that in a work setting employees will underestimate the increased probability of major injury or death due to a deterioration in safety and health conditions, and consequently will not demand a sufficiently high compensating differential. Contrary to the predictions of the standard competitive model, therefore, safety and health expenditures may be undersupplied by firms and injuries and fatalities may be above socially optimal levels.
A second cause of intransitivity is a change in preferences over time. In economic analyses of labor supply, for example, the standard practice is to assume that indifference curves in the income-leisure model are unchanged from one period to another and observed changes in labor force participation rates or work hours are due to changes in budget constraints. In a more complex time allocation model (Becker 1976; Killingsworth 1983), the change in labor supply may also arise from changes in the technology of household production, such as those due to time-saving appliances, and endogenous changes in female fertility. In both models, however, the maintained hypothesis is that tastes are stable (Becker 1976) for, if such is not the case, it becomes quite difficult to identify (in a statistical sense) the labor supply curve or the underlying income and substitution effects.
The assumption of stable tastes is not itself objectionable and, indeed, is a very productive device for deriving hypotheses about labor supply behavior. What is objectionable is that few labor supply studies ever investigate (and many do not even question) whether the assumption is in fact a valid one. That it may not be is suggested by the reversal in the downward, century-long trend in work hours in the United States and the research of sociologist Arlie Russell Hochschild (1997).
From 1900 to 1980 annual work hours declined substantially, a trend explained by neoclassical labor supply models (for example, Killingsworth 1983; Smith and Ward 1985) largely on the basis of rising real wages, a greater demand for leisure (a dominant income effect), and the induced decline in fertility. But real wages have continued to increase since 1980, albeit at a much slower rate, yet annual work hours in the last fifteen years have reversed course and started to noticeably increase. A possible explanation is that recent improvements in household technology have been so dramatic that they have allowed family members (particularly women) to have more of both leisure and market work. But Hochschild, based on extensive first-hand observation and interviews, argues that the phenomenon instead primarily reflects a profound, ongoing shift in preferences with respect to market work versus time spent at home in leisure or household production. In particular, she argues that the workplace and household are switching roles in the lives of many people. That is, whereas people once saw work as onerous and a source of disutility, and time at home as a refuge and source of satisfaction, the trend is increasingly the opposite, due both to improvements in the quality of worklife and to people's desire to escape from burdensome home responsibilities and emotionally taxing personal relationships. No doubt these developments, assuming they truly are occurring - admittedly a big "if" - can also be explained in a way consistent with the stable tastes assumption of neoclassical theory, but the question remains whether, on net, doing so illuminates or obfuscates the true dynamics of behavioral change.
Although intransitivity of preferences has received considerable attention in the economic psychology literature, the other assumptions of the rational choice model have not. Two of those examined here are continuity and nonsatiation.
The psychological concepts that imperil continuity and nonsatiation are needs and ethical values.
A need is a necessity - something the organism must have at some minimum level in order to continue to function effectively. Some needs are biological in nature, such as food, shelter, and sex; others are social in nature, such as the need to be loved and have the respect of others; and yet others are psychological in nature, such as the need to have fun and feel competent. Although the exact proportions remain a matter of debate, it is commonly held that most needs are in part innate to the human species and in part socially constructed (Kanfer 1992; Triandis 1995)
Needs and wants are distinct but closely related concepts (Georgescu-Roegen 1968; Baxter 1988; Elkins and Max-Neef 1992:181-214). Needs, by their nature, are fewer in number, of greater intensity, and motivationally antecedent to wants. A state of hunger, for example, gives rise to a need for food and this need, in turn, creates a desire for a variety of food items (for example, hamburgers, hot dogs) that represent the "wants" in standard economic theory. Although early neoclassical economists, such as Alfred Marshall (Haines 1982), distinguished needs from wants and assumed, furthermore, that needs can be ordered in a hierarchy of importance or immediacy, the concept of need mostly disappeared from economic discourse in the decades following World War II (but see Lutz and Lux 1988; Baxter and Moosa 1996). This stands in sharp contrast, it should be noted, to the field of organizational behavior and subfield of economic psychology, where need-based ("content") theories of motivation are widely cited and used in the study of work and effort (Kanfer 1992; Lea, Tarpy, and Webley 1987).
Can the concept of need illuminate important behaviors or issues in labor economics? Several examples suggest it can. One concerns the ethical or normative justification for a minimum wage. Many early twentieth-century economists, including Alfred Marshall, A. C. Pigou, and the American institutionalists, favored government minimum wage legislation as a means to redistribute income to low-wage workers (Cooter and Rappoport 1984). Their argument was based, in part, on a cardinal notion of utility (the notion that marginal utility is measurable and comparable across people), the law of diminishing returns, and a philosophical commitment to "utilitarianism" (a social welfare goal of "the greatest good for the greatest number"). They reasoned that the marginal utility of income must be lower for a rich person than for a poor person (due to the law of diminishing returns) and thus social welfare (as measured by the aggregate value of individual utilities) would be increased if income were redistributed via a minimum wage law (because higher income groups will receive lower returns on capital and pay higher prices on consumption goods while the wage bill received by low-wage workers will increase, assuming an inelastic demand curve for labor).
This position was undermined by the development of the ordinal utility concept in the 1930s and the parallel popularization of the Pareto criterion for evaluating changes in social welfare. According to the proponents of ordinal utility, there is no scientific basis for comparing marginal utilities across people, because (a) utility is impossible to measure and (b) the law of diminishing returns is inapplicable, since the intensity of a rich person's wants may well exceed that of a poor person's wants. Likewise, the doctrine of the "greatest good for the greatest number" (utilitarianism) was supplanted by the criterion of Pareto optimality, according to which a law such as the minimum wage can be credited with increasing social welfare only when at least one person's utility is increased and no one else's is decreased (or, somewhat less restrictively, the gainers are potentially able to compensate the losers) - a comparison that can be made using an ordinal measure of utility. Clearly a minimum wage law cannot satisfy the Pareto criterion as easily as it does the utilitarian criterion, a difficulty that, in conjunction with the position that it is impossible to compare the size of marginal utilities across individuals, led these neoclassical economists to claim that there exists no scientific basis for a normative position in favor of a minimum wage law.
But, say modern proponents of the minimum wage, such a normative position can indeed be established if the concept of want is replaced by that of need. This line of thought harks back to the concept of a "living wage" popularized by the early institutionalists (Glickman 1997). They defined a living wage as the income each individual must have to meet the minimum needs of food, shelter, health care, and so on, that are required for continued human existence. In this respect it corresponds closely to the poverty line concept developed in the early 1960s by the federal government (Ruggles 1990). According to the institutionalists, while the intensity of wants may be noncomparable across people, the satisfaction of basic needs is quite capable of comparison. The key difference between a want and a basic need is that if the latter is not satisfied the organism perishes. While it may be impossible to compare the marginal utility Person X gets from a new car to that obtained by Person Y from the same good, one can surely make a credible statement that Person X is better off than Person Y if the former gets the minimum necessary level of goods for sustained life and the latter does not (an infinitely large decrement in utility?). Thus, from an institutional perspective, the concept of need - when combined with a utilitarian criterion of social welfare - can in fact provide an ethical basis for a minimum wage law to the extent such a law redistributes income so as to give every worker the requisite living wage.
The need concept also helps illuminate a number of other issues of positive economics. Consider, for example, the conditions of continuity and nonsatiation in the theory of rational choice.
Continuity means that items of choice are divisible and substitutable, with the implication that indifference curves are smooth and continuous, thus making possible marginal adjustments in behavior. The concept of need, however, may preclude substitution and marginal adjustment. For example, when asked (or forced) in a work situation to choose among items satisfying different need categories (for example, physical safety versus job security), people feel conflicted, experience negative emotional responses such as frustration and anger, and frequently respond not by making consistent choices and trade-offs but by initiating stratagems aimed at postponing choice, eliminating the necessity of choosing, or opening up other, more preferred choices (Simon 1982:407). Examples include employee drug abuse (an escape from difficult choices), acts of violence against supervisors or managers, and participation in a union organizing drive.
The concept of need also calls into question the common assumption in rational choice theory of nonsatiation (that is, "more is better than less"). Many needs, particularly lower order needs associated with biological functions, exhibit homeostasis (Petri 1986): motivation is aroused when the organism experiences a need deficit (a person becomes hungry, for example); this deficit triggers behaviors aimed at satisfying the need (the person fixes a sandwich and eats it); and the behaviors cease once the need deficit is eliminated (the person is no longer hungry). Once one need is satisfied, the organism focuses on fulfilling the next highest-priority unsatisfied need.
An economic implication of needs and the homeostasis concept is that firms with super-normal profits may engage in rent sharing with employees by paying higher wages than the going rate (Kaufman 1988; Groshen 1992; Hildreth and Oswald 1997). Standard neoclassical theory assumes that the only goal of firms is maximum profit, a corollary of which is that they will never pay wages above the minimum rate dictated by market forces. Empirical research frequently finds, however, that firms in concentrated or regulated product markets pay higher wages for otherwise similar labor (Kwoka 1983). A reasonable hypothesis is that these firms earn super-normal profits due to barriers to entry in the product market and share a portion of these profits in the form of higher wages (or, for example, reduced demands for work effort, or laxer discipline; predictions contrary to most efficiency wage models). But economists have been notably resistant to exploring this possibility. Indeed, Richard Thaler remarked,
The suggestion that firms do not maximize profits was once considered heresy, equivalent to a belief in upward sloping demand curves. In recent years, however, the old fashioned notion of "managerial discretion" has been given the respectable term "agency theory." [Still], I know of no formal attempt to explain interindustry wage differentials with an agency model in which managers have a taste for both profits and highly paid employees. (1989:186)
Such a theory can be developed, however, by using alternative behavioral concepts from the psychology literature. From a psychological perspective, the key issue is to identify the motivational forces that determine the direction and intensity of action on the part of the firm's managers (the agents of the stockholders). One approach (Kaufman 1990) is to use the concept of need and, in particular, incorporate into the managerial utility function a "hierarchy of needs" theory of motivation, such as that developed by Maslow (1954), Herzberg (1966), or Alderfer (1972).(15) As long as profit is below the firm's "survival" level, managers make decisions solely focused on minimizing cost and maximizing profit. Once profit exceeds the survival level, this need is satiated and the next need, such as management security, becomes salient. The firm at this point "satisfices" with respect to profit and uses additional revenue to satisfy the security need, say by paying above-average wages to reduce the likelihood of a union organizing drive or to gain a public image as a benevolent employer. (Note that this theory of satisficing differs from that advanced by Simon .)
Should the firm become unionized or for some other reason incur higher costs, the potential or actual reduction of profit to below the survival level motivates management to "tighten up" in search for cost and efficiency gains (for an example, see Clark 1980). Although this type of behavioral response, now generally referred to in labor economics as a "shock effect" and as "X-inefficiency" in the organizational economics literature (Kaufman 1997; Leibenstein 1987), has long been dismissed as ad hoc or of little quantitative importance by Stigler (1947, 1976) and other neoclassical critics, institutional and behavioral-oriented labor economists (Commons and Andrews 1936; Lester 1946; Annable 1984) have long argued that it is an important phenomenon and provides an explanation for, among other things, the relatively small or nonexistent decline in employment that follows a minimum wage increase, per the recent findings of Card and Krueger (1995).(16) Existence of a shock effect-like phenomenon also provides one explanation (among several) for the contention of Freeman and Medoff (1984) that unionization promotes higher productivity in companies.
The second concept that imperils continuity and nonsatiation is values. Economists and psychologists typically think of values differently. "Value" in economics is the relative worth of an item, generally expressed as the price a person is willing to pay for it. Psychologists' concept of "value" sometimes resembles economists' (for example, the term "valence" in expectancy theory is typically defined as the subjective value of an item), but often "value" has a separate, quite distinct meaning related more to sociology, ethics, and theology (Etzioni 1986; Frank 1988; Hausman and McPherson 1996). In this sense a value is a social norm, an ethical principle, or a moral precept.
Analytically, the effect of moral values and ethical principles is to introduce various kinks, gaps, and instabilities into people's preference rankings and indifference maps. Tversky and Kahneman (1991) showed, for example, that preference rankings are "reference dependent" in that people evaluate prospective gains, losses, and trade-offs relative to a base-line or reference point. One such form of reference is the status or treatment afforded close associates. A specific application of this idea is equity theory (Adams 1963). This motivational theory posits that the utility value a person gains from a particular rate of pay is dependent on an evaluative judgment of his or her rewards-to-effort ratio in relation to that of similar co-workers. Should the employee discover that these other workers are paid more for the same job, a feeling of inequity arises and the utility of the wage drops sharply. The effect, as shown by Tversky and Kahneman, is to cause a shift in the person's indifference curves toward the origin, in direct contradiction to the "stable tastes" hypothesis of Becker (1976) and others.
On an empirical level, ethical values and principles help explain a variety of choices and actions observed in labor markets that cannot be fully accounted for by the rational choice model. Two examples regarding unions illustrate the point. The first concerns the large variation across countries in union density. Union joining is a form of collective action and, as Olson (1965) argued, is in part a form of public good in the workplace and thus subject to the free rider problem. But if the free rider problem is generic, what accounts for the diversity in union density across countries? The tendency in economic research (for example, Blanchflower and Freeman 1992) is to look for economic or legal factors that provide incentives or support for collective action, such as the size of the union wage premium, the extent of government market regulation, or the strength of legal prohibitions of anti-union employment practices. Not only do these explanations leave a considerable portion of the variation unexplained, but wages, market regulation, and legal rules are themselves endogenously determined (similar to right-to-work laws, as pointed out by Farber 1984).
One approach is to supplement economic and legal factors with those of a more behavioral nature. As one possibility, Triandis (1995) showed that attitudes and values regarding individualism versus collectivism differ markedly across countries and these differences, in turn, have their roots in perceptions about the extent to which individual goals are complementary or conflictive vis-a-vis goals of other people and social groups. These attitudes and values are measurable and could well be included in estimating equations.
The second example concerns the strength of management resistance in a union organizing drive. Freeman and Kleiner (1990) predicted, for example, that the higher the ex ante probability of a union win in an organizing drive, the less money a profit-maximizing firm will spend to try to defeat the union (it is irrational for a company to spend a great amount of money if the union is sure to win anyway). Field research reveals, however, that often it is those firms that are furthest behind in the campaign that "pull out all the stops" to defeat the union (Kaufman and Stephan 1995). Why? Because remaining nonunion is a matter of principle to the owners/managers, making the utility value of remaining nonunion extremely large (reflected in statements such as, "We'll close this plant before letting it go union"), and thus legitimating expenditures far larger than can be justified solely on the basis of profit and loss.
Normatively, ethical values impose constraints on choice that are often in the social interest. As shown by Leibenstein (1987), Frank (1988), Miller (1991), and Hausman and McPherson (1996), appropriate social norms, ethical principles, and moral precepts promote economic efficiency and social cohesion by fostering trust and cooperation - psychological dispositions crucial to overcoming production inefficiencies caused by free rider, prisoner dilemma, and envy problems in organizations. This result calls into question, in turn, one of the most fundamental and widely accepted conclusions in modern labor economics - that more competitive market conditions promote greater economic efficiency. The reason, as illustrated in the repeated-play prisoner dilemma model (Miller 1991), is that the degree of cooperative behavior between two people (for example, a manager and employee, or two employees in a production team) is a positive function of the number of times the game is played and opportunities for interpersonal communication. These conditions are least likely to occur in a competitive labor market because the low cost of labor :mobility and the plethora of alternative job opportunities promote high employee turnover and a transient employment relationship.
Decision-making. The fourth step in the behavioral chain represented in Figure 1 is decision-making. In the previous step behavior has been motivated by the existence of needs and wants; now the organism must use its mental capabilities to choose the course of action that best satisfies these needs and wants, subject to the constraints on choice imposed by the environment.
Conventional neoclassical economic theory, as noted previously, makes strong assumptions about the decision-making capacity of the human agent. In textbook models, the person is typically assumed to have complete and objective knowledge of the environment and the mental ability to effortlessly calculate the behavioral choice that maximizes utility. Rational behavior, then, is synonymous with choosing the optimal outcome.(17) More advanced models may relax certain of the assumptions by making outcomes probabilistic, such as in subjective expected utility theory, or by making the acquisition of information costly, as in search models, but in nearly all neoclassical models the core assumption of rationality is maintained - that the human agent has the mental ability and motivation to calculate optimal outcomes.
The rationality assumption in neoclassical theory has been the subject of vigorous and frequent attacks by critics. Best known is Herbert Simon (1982), who argued that the neoclassical model of "objective rationality" is untenable because human beings do not possess the brain power necessary to perform the optimization calculations. He proposed an alternative form of rationality called "bounded rationality," whereby, in essence, behavior is purposeful but less than optimal due to human cognitive limitations.
Simon's conception of bounded rationality has found fruitful application in several areas, such as game theory, organizational economics, and transaction cost economics (Conlisk 1996). The extent to which it has been applied to topics in labor economics, however, has remained limited. Probably the most interesting and important work in this regard is by Oliver Williamson (1985). He argued that bounded rationality, in conjunction with opportunism and asset specificity, makes it impossible to construct completely contingent labor contracts and, thus, firms and workers have an incentive to switch from a short-run, market-mediated form of employment relationship to a longer-term relationship in which wages and conditions of employment are determined administratively in the firm, typically within the context of an internal labor market. An assumption of perfect foresight, as in the standard neoclassical model, does not provide this implication, he maintains.
An alternative conceptualization of bounded rationality that I introduced recently (Kaufman 1999) provides additional implications for labor market behavior. I noted, first, that Simon drew part of his inspiration for the concept of bounded rationality from the work of John Commons (1934) and, second, that Commons described the typical human condition to be one of purposeful behavior mixed with significant elements of "stupidity, passion and ignorance" (p. 874). Simon's notion of bounded rationality thus corresponds to the two factors of "stupidity" (limited mental cognitive ability) and "ignorance" (limited knowledge), but omits the role of "passion" (human emotional states).
This is a serious omission, I argued, for while emotions are in many ways a functional adjunct to behavior (Frank 1988; Elster 1998), extremes in emotion due to feelings of intense anger, frustration, depression, love, jealousy, lust, and so on can seriously or even completely disrupt logical, rational thought processes and thus create a condition of bounded rationality. I adduced this result using the Yerkes-Dodson law from psychology, which hypothesizes that low pressure/arousal results in a low performance level (such as in test-taking), that performance increases with additional pressure/arousal up to some threshold, and that performance decreases with yet additional increments of pressure/arousal (Petri 1986).
This conceptualization of bounded rationality has a variety of implications for labor and organizational economics. In a quest for greater profits and efficiency, for example, many American corporations over the past decade have substantially reduced their work forces through downsizings and restructurings. Yet, paradoxically, evidence both from surveys of business executives and from quantitative academic research reveals that profitability in many of these companies is lower after the downsizing than before (Cappelli et al. 1997). Why? One explanation is that downsizings and restructurings create widespread conditions of bounded rationality in the workplace. Illustrative, for example, is the observation of management consultant Judith Bardwick that downsizings and restructurings create rampant fear among employees, sometimes approaching a state of panic. "Fear smothers productivity," she wrote.
The whole quality of decision making suffers. Employees are always looking over their shoulders. Manager's are afraid to take risks. (1991:34)
Bardwick attributed the decline in quality of decision-making to three psychological consequences of excessive emotional arousal: denial (to avoid fear and stress, people deny that the situation has changed), time spent worrying (people become immobilized by anxiety and spend their time engaged in introspective worrying), and inconsistent behavior patterns (excessive fear leads to behaviors that are often extreme and volatile).
This conceptualization of bounded rationality can also help explain a paradox regarding labor strikes that has long puzzled economists. From the perspective of standard microeconomic theory, labor strikes appear to be irrational acts. The reason is that both the company and workers would be better off if they agreed to the post-strike wage and skipped the strike and all its attendant costs. In attempting to square the occurrence of strikes with the economic model of rational behavior, economists have advanced a number of explanations involving asymmetric information, uncertainty, and divergent expectations. Strikes are typically portrayed in these models as bargaining accidents or mistakes (Kennan 1986; Kaufman 1992).
An entirely different approach is to incorporate emotions and bounded rationality into bargaining theory. An example is provided by Wheeler's (1985) integrative model of industrial conflict. He included in the model a "rational behavior" path to conflict, such as are represented by the economic models just described. He maintained, however, that most strikes arise from a very different process. The linchpin of this alternative route to conflict is frustration-aggression theory - a construct popular in sociology and political science.
Frustration is an emotion arising from blocked goal attainment due to the perceived obstruction or influence of a third party. Low to moderate levels of frustration can be functional in that the human agent is motivated to exert greater effort and thought to solving the problem. As frustration grows, however, emotional feelings increasingly cloud rational decision-making until the person finally "loses it" and resorts to acts of violence and aggression against the person or thing blocking goal attainment.
Wheeler hypothesizes that the frustration-aggression mechanism explains the occurrence of many strikes. For example, empirical research shows that strikes increase in periods of higher inflation (Kaufman 1992). As workers see inflation erode the real value of their wage, the gap between expected real wage and actual real wage grows, leading to greater frustration among the rank and file. If frustration becomes large enough, the rank and file's decision-making process becomes increasingly distorted by emotion until a strike becomes quite likely, not because it pays in terms of' dollars and cents but because it satisfies a psychological need for aggression and catharsis.
Action. The fourth step in the behavioral process pictured in Figure 1 is "action" - the initiation and completion of behavior to obtain a desired outcome. It is at this step that the psychological process of motivation becomes manifest in overt effort and performance.
The direction and intensity of effort and performance are key variables for many aspects of labor market behavior. This is particularly true as it relates to wage rates, since wage rates are the price of labor and the level and change in prices in a market economy are what motivate employers and workers, respectively, to efficiently organize production, work hard, invest in new skills, and move capital and labor resources out of areas of declining demand to those of growing demand. Clearly, the efficiency of the labor market as an allocative device depends to a significant degree on the extent to which wage rates effectively serve this motivational function. Two examples suggest that the predictions derived from standard rational choice models may well be inaccurate, if not flatly incorrect, in this regard.
As one example, consider the incentive effect of higher wages on work effort. As earlier alluded to, in the 1910s the proponents of scientific management assumed that higher incentive rates of pay would elicit greater employee work effort. Eighty years later this assumption is still widespread in labor economics. An example is efficiency wage theory. Most efficiency wage models (for example, Shapiro and Stiglitz 1984) are largely neoclassical in spirit in that they use the rational choice model and explain variation in employee work effort on the basis of changes in external rewards and punishment (for example, individual work effort is a positive function of the premium in wages received over the market level and the probability of unemployment if the individual is fired for shirking). Although this specification is intuitively plausible and finds support in empirical studies in the labor economics literature, a quite different prediction emerges when a more behavioral conception of the human agent is instead used.
In particular, a body of research in psychology and organizational behavior concludes that financial incentives may have much the opposite impact on work effort (for example, Deci and Ryan 1985). After an extensive review of this work, for example, Kohn (1993:133) stated, "We have no grounds for assuming that paying people more will lead them to do better quality work - or even, over the longer run, more work." The reasons, in a nutshell, are: first, workers perceive manipulation of external rewards and punishments as exactly that - attempts by management to manipulate and control them - and react by trying to evade or subvert the control mechanism; and, second, use of external rewards and punishments to elicit greater work effort backfires because it destroys intrinsic forms of motivation.
The position taken by Kohn and others on the role of financial incentives is controversial, and conflicting evidence exists on the matter. Also, economist David Kreps (1997a) has suggested an alternative explanation for why greater financial rewards may retard work effort that does not involve psychological concepts of extrinsic versus intrinsic motivation. Unless the behavioral explanation provided by Deci, Kohn, and others is completely without empirical support, however, the main point being stressed here retains its validity - that in studying the relationship between financial incentives and work effort labor economics research would benefit from giving greater consideration to an alternative, more psychologically enriched model of the human agent.
A second example concerns the impact of emotional states on motivation and action. As already noted, when a person fails to achieve a goal a typical emotional response is frustration. Another is depression. Depression typically follows frustration and stems from a personal attribution that the self is largely responsible for goal failure and lacks the means or ability to rectify the situation. The behavioral manifestations of depression are passiveness, withdrawal, and lack of self-confidence - behaviors that often only further inhibit goal attainment and intensify the state of depression. The net effect is to make the structure of preferences endogenous.
A practical application of this phenomenon to labor markets is with respect to the discouraged worker effect. A discouraged worker is an unemployed person who is unable to find a job and after some length of job search becomes discouraged, quits looking for work, and drops out of the labor force. The major theoretical tool used by economists to explain this phenomenon is search theory (Mortensen 1986; Devine and Kiefer 1991). It is, in effect, a theory of optimal job shopping, with respect both to the strategy used and to the time and resources devoted to the search. The basic idea is that the job seeker compares the marginal gain from additional search to the marginal cost and quits searching when the latter exceeds the former.
This problem can be represented in different ways, but according to the most popular interpretation, the worker obtains a sequence of job offers paying varying wage rates and stops searching when a job offer is equal to or exceeds some predetermined minimum acceptable level (the reservation wage). The usual outcome envisioned is that the person eventually receives such a job offer and transitions from "unemployed" to "employed." But more sophisticated search models endogenize the reservation wage and, in particular, show that under plausible conditions (for example, limited duration of unemployment insurance benefits) the optimal strategy for the job seeker is to gradually lower the reservation wage in direct proportion to the length of unemployment. It is possible that no acceptable job offer will have been received and at some point the reservation wage becomes less than the value of leisure and the person ceases the job search and exits the labor force as a discouraged worker.
The key difference between an economic explanation and a behavioral explanation of this event concerns the reasons for the decline in the reservation wage. The standard practice in microeconomic, rational choice models is to explain this decline based on objective, often pecuniary factors external to the person's psyche, such as liquidity constraints (limits on saving and borrowing), the exhaustion of unemployment benefits, or avoidance of a reputation among employers as a "lemon" (possibly a subjective assessment by employers but an objective fact for the job seeker).(18) A behavioral theory, on the other hand, looks to induced changes in the person's utility function as an important, complementary explanation.
A popular motivational theory in the behavioral sciences, for example, is expectancy theory (Vroom 1964; Kanfer 1992). It hypothesizes that the direction and intensity of action is a multiplicative function of three subjective estimates: "expectancy" (the person's estimated probability that he or she can accomplish the task), "instrumentality" (the estimated probability that accomplishing the task will be rewarded), and "valence" (the value of the reward). While standard economic theory assumes stable preferences, from a behavioral perspective preferences are likely to be endogenous because each of these three probabilities is influenced by external events and the emotions that accompany them, such as the depression and anxiety that arise from a longer than expected job search. Thus, from a behavioral perspective the decline in the reservation wage and the ultimate decision to drop out of the labor force, while stemming in part from rational economic considerations, also reflect a deterioration in "tastes" for work caused by a decline in assessed expectancy, instrumentality, and valence. This perspective, it should be noted, provides one explanation for unemployment hysteresis and path-dependent behavior. It is also consistent with the position taken by the early institutional economists (Commons and Andrews 1936) that one of the worst evils of unemployment is its corrosive effect on personal character (Kaufman 1997). This position has recently received fresh examination and corroboration in the work of Darity and Goldsmith (1996).
Utility change. The final step in the behavioral chain represented in Figure 1 is the change in utility that results from the action undertaken in the previous step. The change in utility may be either positive or negative, depending in rational actor economic models on whether the environment (for example, as represented by budget constraints) has become more accommodative or restrictive.
Three important amendments to this scenario are suggested by a psychological perspective. Each involves a feedback loop from the step "change in utility" to prior parts of the model. The principal conceptual effect with regard to microeconomic theory is that equilibrium is replaced by a dynamic adjustment process.
One feedback loop goes from "change in utility" to "decision-making" and represents the learning process (Hogarth 1987). Learning is often absent from neoclassical models since the human agent is assumed to have perfect information about the environment and unbounded powers of reasoning and calculation. These assumptions are, of course, relaxed in certain applications, such as search theory and various game theory models, and in these situations the human agent can no longer move to the optimal outcome in one step but, rather, must approach it incrementally through a process of observation, trial, error, and adjustment.
The crucial issue with respect to learning is not so much whether it is included in the model but the speed and accuracy of the learning process. At one extreme, for example, is the economic theory of rational expectations. Although future states of the world are unknown, it is assumed that economic agents accurately assess and incorporate all pertinent information and generate in each time period updated, unbiased estimates of the relevant variables. At the other extreme are various learning theories in the behavioral sciences, nearly all of which portray the learning process as slow and subject to numerous biases (Weis 1992). An example is Bandura's (1986) model of social learning, which distinguishes three steps in the learning process: attention, retention, and production (transference from memory to action). Given limited brain capacity and sensory ability, in combination with the filtering effect of values and attitudes, people perform these three functions with a great deal of selectivity, which, in turn, imparts bias into learning.
The different approaches taken by economic and behavioral research can be illustrated by considering wage bargaining and strikes. Most microeconomic bargaining models do not include an explicit learning mechanism for the negotiators. Instead, they specify profit and utility functions for the company and union and solve these through constrained optimization techniques to derive a predicted wage settlement point (Kennan 1986). More behaviorally oriented bargaining models (Walton and McKersie 1965; Cross 1969), on the other hand, typically include a learning process. Cross, for example, embeds the learning process in a set of union-firm reaction functions that, in the form of simultaneous differential equations, are solved for the predicted wage.
Explicit consideration of the learning process in the latter group of models leads to conclusions different from those gained from the former. For example, Cross maintains that one reason strikes occur is because both parties typically enter negotiations with over-optimistic expectations of how much the other side will give in. It is only through the process of offer and counter-offer during negotiations that learning takes place - the faster and more efficient the learning process, the less probability a strike will occur. Kennan (1986) claimed, however, that Cross's model rests on irrational behavior - that is, it is irrational for human agents to be consistently mistaken (over-optimistic) in their expectations. It follows, in turn, that if the negotiators' expectations are on average correct, consideration of the learning process has little of substance to offer bargaining theory.
Although most economists would probably agree with Kennan (given the popularity of the rational expectations hypothesis), psychological research on the learning process lends support to Cross's position. According to a psychological proposition known as the "fundamental attribution error" (Petri 1986:269), people systematically overrate their influence and control over events in order to heighten their sense of self-efficacy and self-esteem. Persistent overoptimism in bargaining is one manifestation of the fundamental attribution error, and a consequence is more strikes than would otherwise occur.(19) Kennan is correct, however, that learning should cause a reduction over time in the parties' degree of over-optimism. Research finds, for example, that strikes are more frequent in new bargaining relationships (where overoptimism is likely most pronounced) than in well-established ones (Kaufman 1992).
A second feedback loop exists between "change in utility" and the "wants" variable in Figure 1 (see Fisher and Locke 1992; Bowles 1998). This occurs through two reinforcing psychological channels. The first is described by aspiration level theory. An aspiration is an object of desire and thus closely resembles the "wants" of economic theory. Research shows that aspirations are not a "given" but, rather, rise with success and fall with failure (Katona 1975; Inglehart and Rabier 1986). The second channel is described by cognitive dissonance theory (Festinger 1957). Cognitive dissonance occurs when there is a disjunction between observed behavior (women earn less than men) and personal attitudes and beliefs (men and women are equally productive in the workplace). Because dissonance is aversive, people take action to eliminate it, either by changing their beliefs to fit observed behavior (for example, women deserve lower pay, due, say, to lower labor force attachment) or by undertaking action to change observed behavior (for example, promoting strengthened anti-discrimination legislation). Although a change in either values or observed behavior can reduce dissonance, the former is frequently the easier option.
Both theories suggest that the structure of preferences is not independent of behavior, as is usually assumed in microeconomic theory. Two examples that have aspiration theory contained within them have already been given: the discouraged worker effect and welfare dependency. Cognitive dissonance theory has relevance with respect to explaining gender and racial differences in occupational attainment and earnings.
Why, for example, were less than 15% of lawyers in 1980 female? A well-known neoclassical explanation is that women voluntarily chose other occupations that had a smaller earnings penalty for intermittent labor force attachment (Pollachek 1981). The typical explanation provided by law schools is that few qualified women applied at that time (Epstein 1981). A more behavioral explanation is that women (and other minorities) who desired to practice law experienced large cognitive dissonance - they placed a high utility value on becoming a lawyer but soon discovered the path was largely blocked by discrimination in law school admission policies and law firm hiring practices (for example, as in the case of current Supreme Court Justice Ruth Bader Ginzberg, who graduated from Harvard Law School only to find no major law firm would hire her). To reduce dissonance, some persevered and became lawyers but many others took the other route and, in order to protect self-concept, developed a variety of explanations and rationalizations to justify downgrading their desire to be a lawyer (Devine 1989). Because most neoclassical economists take preferences as a "given," the decision of these women to give up a career in law and pursue instead, say, elementary education appears to be a product of voluntary choice and a logical adaptation to their dual role of mother and wage earner. An alternative explanation, however, is that their preferences adapted to accommodate reality (see Haberfeld 1992) - a reality in which the seemingly free market forces of supply and demand are a veil hiding a discriminatory pattern of social norms, institutional access, and firm hiring behavior (Rebitzer 1993; Kim 1999).
Finally, a third feedback loop is from "change in utility" to "action." Once an action is completed in step four of the behavioral sequence, the change in utility experienced in the last step is significantly influenced by comparisons, particularly with respect to equity or fairness. Although equity is notoriously difficult to measure, conceptually it rests on two fundamental considerations: first, that an appropriate balance exists between "inputs" and "outputs" (for example, the pay is commensurate with the amount of output produced); and, second, that the decision process used to determine the outcome is procedurally correct (Adams 1963; Sheppard, Lewicki, and Minton 1992). These twin dimensions of equity are often called distributive justice and procedural justice. If the action violates either conception of justice, utility is markedly reduced. Furthermore, the feeling (emotion) of inequity or injustice is a catalyst for further action in order to resolve the injustice or retaliate in response to it. Injustice and inequity thus disrupt the equilibrium process and precipitate additional rounds of action and reaction.
Models by labor economists seldom include considerations of equity and fairness, and yet their influence permeates the workplace and labor market (but see Knetsch and Thaler 1986). In this regard, Albert Rees (1993:243) stated,
The neoclassical theory of wage determination, which I taught for 30 years and have tried to explain in my textbook [Rees 1973], has nothing to say about fairness.... Beginning in the mid-1970s, I began to find myself in a series of roles in which I participated in setting or controlling wages and salaries.... In none of those roles did I find the theory I had been teaching for so long to be of the slightest help.... The one factor that seemed to be of overwhelming importance in all these real-world situations was fairness.
Arthur Ross (1948) made much the same observation a half-century ago when he noted (p. 74), "Ideas of equity and justice ... are the strongest equalizing tendencies in wage determination." Ross's position has been widely ignored in modern labor economics, but both Rees's statement and recent empirical research (for example, Levine 1993) point to its continued validity.
The problem labor economics incurs by omitting considerations of fairness is illustrated by the oft-noted downward rigidity of money wages. Competitive theory predicts that money wages should decline in the presence of an excess supply of labor, but they only infrequently and slowly do (Kahn 1997). Neoclassical economists have developed a variety of explanations for this aberrant behavior, including wage rigidities introduced by minimum wage laws and unions, implicit contract theory, and an "unemployment is really disguised leisure" argument.
Research by institutional economists several decades ago, as well as more recent work (Blinder and Choi 1990; Bewley 1997), suggests that these "economic" explanations are of secondary importance to one that is more psychological in origin. Writing in 1929, for example, Sumner Slichter offered this explanation for downward wage rigidity:
The new personnel policies [of the 1920s] have increased the bargaining power of labor because they have made the efficiency of labor depend more than ever before upon the willingness of men to do their best.... There is abundant evidence that the reluctance of managers to reduce wages ... has been partly due to the fear that wage cuts would destroy the good will which has been built up at considerable trouble and expense. (Slichter 1929:431-32)
As Slichter explained in more detail, companies in the 1920s deliberately pursued a "high wage" policy by paying production employees wages greater than market rates. They did this as a motivational device, believing greater productivity, reduced turnover, and less union activity would more than offset the cost of higher wages. Having obtained these benefits, for firms to then cut wages back to market rates would be perceived by workers as unfair and illegitimate. They would react, in turn, by quitting, striking, shirking, and numerous other means not captured in standard "labor-as-a-commodity" models of labor markets. Recent behavioral-based efficiency wage models (Akerlof 1990) that incorporate social norms and considerations of equity, on the other hand, recapture Slichter's basic insight and offer predictions matched by empirical evidence (Fehr, Kirchler, and Weichbold, and Gachter 1998).
Although the rational choice model has provided the conceptual foundation for an immense and impressive body of scholarship in labor economics, it also has weaknesses that have heretofore received only modest scrutiny by most labor economists, weaknesses stemming primarily from its simplified psychological profiling of the individual human agent. These weaknesses are surprising given the greater importance of psychological (and sociological) factors in labor markets than in most other economic spheres, and also given the growing body of research in several other branches of economics (for example, game theory and financial economics) that makes use of behavioral economic concepts and methods.
The purpose of this paper is to move this research agenda in labor economics forward another step. To do so, I have taken a close look at the model of the human agent in labor economics in terms of a five-step conceptual framework of individual behavior and, for each step, suggested various ways this component of the economic model could be broadened with concepts and theories from the field of psychology. This expanded model of the human agent is largely preservative of the rational choice paradigm in that it seeks only to add more psychological and sociological structure (for example, social norms) against which individual calculation of benefits and costs takes place, but in some other respects (such as the hypothesis of bounded rationality due to emotional extremes) it directly challenges the basic tenets of the rational choice model. I have also attempted to address the "So what?" question by providing a number of examples of behaviors and outcomes in labor markets that are better understood or explained with a more psychologically informed model of man. The major psychological constructs that underlie this analysis are motivation, cognition, and emotion.
In closing, and to avoid possible confusion and misunderstanding, I would like to briefly state what I am not saying in this essay. I am not saying, for example, that the economist's model of the human agent should be rejected because it is an unrealistic representation of the nature of human beings (though it certainly is). Over the years numerous critics have had fun satirizing homo economicus, apparently with the thought that highlighting the absurdities of its assumptions would discredit it as a theoretical construct. Not only is this strategy non-scientific, it has also manifestly failed, as the rational choice model continues to dominate economics and is rapidly gaining ground in sociology, political science, and other social science disciplines (Baron and Hannan 1994). Thus, I have taken pains in this paper to avoid questioning or impugning the validity of assumptions per se and have instead focused on the consequences of assumptions for making predictions.
It is also not my position that the standard model of the human agent in labor economics should be entirely discarded and replaced by some other, more psychologically complex model, such as the one advanced here. For a wide range of issues (for example, the labor supply impact of an overtime law) the standard model is entirely appropriate, and anything more complex will only complicate the analysis with little gain in explanation or prediction. I have attempted to demonstrate, however, that for a large number of subjects the task of explanation and prediction is facilitated by consideration of a richer set of psychological concepts and processes. Although some economists will see this eclecticism as an open door for ad hoc theorizing (for example, one can manipulate the specification of the "tastes" variable so the theory fits any conceivable set of facts), this danger seems preferable to the current situation in which economists do exactly this when they selectively introduce a "taste for discrimination" or "taste for children" variable into the utility function but without benefit of any underlying theoretical justification.
Finally, it is not my position that disciplinary imperialism a la the Chicago school is an inappropriate research strategy. Science is promoted by extending a theory's reach to as wide a range of behaviors as possible, as amply illustrated by the pathbreaking applications of price theory to subjects such as labor supply, human capital, discrimination, information and job search, crime, and marriage and fertility.(20) What I am against is the narrowness, insularity, and defensiveness that this approach can impart to research and scholarly discourse if carried to excess.(21) It is certainly true that the advancement of knowledge benefits from specialization and comparative advantage like nearly all other forms of production, and hence economists have good reason to emphasize prices, incomes, and other environmental constraints on choice. But specialization can be and has been, in my opinion, carried too far and practiced too rigidly. The production of knowledge is inherently an interdependent, multi-disciplinary exercise (per team models of production, such as Alchian and Demsetz 1972). And, I should make clear, the interdisciplinary approach I advocate is a two-way street - economists need to be more open to research in the behavioral sciences but behavioral scientists in fields such as organizational behavior and sociology also have much to learn from economics.
To sum up, note that the closest real world representation of the neoclassical "economic man" is a child under the age of eight (Harris 1993:238-39). The behavior of children of this age group is accurately predicted by a "getting what you want makes you happy" model, which is to say the model of self-interested, rational choice. At about eight years of age, however, other considerations besides personal gain begin to influence behavior, such as the feelings of other people, the importance of observing conventions and norms, and the positive experience of self-control, and it becomes apparent to the child that single-minded pursuit of personal gain is neither appropriate nor the best way to attain happiness. In later years of life, it is still possible to predict a significant portion of the child-now-turned-adult's behavior based on the "maximization of personal gain" model, but that model also misses or misinterprets a great deal.(22) The contention of this paper, in a sentence, is that the social goal of advancing knowledge in labor economics would be materially advanced by taking some of the energy and creativity that currently goes into elaboration and defense of' the "child" model and using it instead to develop a richer, more complex "adult" model.
1 Two different research programs in economics have sought to broaden and endogenize psychological and sociological aspects of the rational choice model, but the approaches taken are quite different. The first is the "economic" approach advanced by Gary Becker, and the second is the "behavioral" approach pioneered by people such as Herbert Simon, Harvey Leibenstein, and George Katona. The key difference between the two is that the former stays within the neoclassical paradigm and seeks to accomplish the "broadening" goal by generalizing price theory with new concepts and methods drawn from or inspired by economic theory, while the latter looks to broaden the microeconomic model, or replace the rational choice construct altogether, with concepts and theories drawn directly from the behavioral sciences. Thus, Becker (1996) incorporated social norms of fairness into microeconomic theory by generalizing the human capital construct to form a new variable called "social capital," while Akerlof (1990) pursued the same end but by importing into economic analysis Adams's (1963) social psychological "equity theory" of motivation. The thesis of this paper is that both research programs advance knowledge and are in certain respects complementary, but that for a wide range of labor market outcomes the behavioral approach is indispensable for purposes of understanding and prediction. It is thus this approach that is focused on and advocated here.
2 The institutional paradigm of the 1910s-30s, associated with Commons and colleagues of the Wisconsin School, and the neo-institutional or "postwar" paradigm of the 1940s-50s, associated with Dunlop, Kerr, Lester, Reynolds, and Ross, significantly influenced research in labor economics during these periods and made extensive use of a behavioral model of the human agent (Kaufman 1988, 1994, 1997). But their influence greatly declined from the 1960s onward. A new and growing influence on labor research, by way of contrast, is organizational economics (for example, Alchian and Demsetz 1972; Holmstrom and Milgrom 1994) and the economics of personnel (for example, Lazear 1995), which may, at first glance, also seem to be behavioral. Both use concepts such as principal-agent theory, asymmetric information, moral hazard, and adverse selection that in some way introduce imperfections or biases in the information available to decision-makers. Most often, however, these imperfections and biases in information are introduced as a feature of the environment, not the human agent. Thus, the principal-agent problem arises because the principal cannot costlessly monitor the performance of the agent due to spatial distance, the complex or unobservable nature of job tasks, and so on. An exception is Williamson's (1985) transaction cost theory of economic organization, since the imperfection in information and decision-making power is modeled as bounded rationality on the part of the human agent.
3 The topics of the articles are executive pay and peer pressure in work groups. An article by Mullahy and Sindelar (1993) on alcoholism, work, and income contained citations to several psychiatric diagnostic manuals, but I did not count this article because these works are largely atheoretic in nature.
4 Transitivity implies that if A is preferred to B and B is preferred to C, then A is preferred to C; completeness implies that the decision-maker can establish a preference ranking for all combinations of goods in the opportunity set. Continuity, convexity, and nonsatiation give rise to well-ordered, continuous indifference curves. See Mas-Colell, Whinston, and Green (1995).
5 Pareto (1927:120) remarked, for example, that once a person's indifference curves are known, "the individual can disappear, provided he leaves us with this photograph of his tastes."
6 This predisposition has been strengthened by the conclusion of Becker (1976) that a number of basic economic relationships, such as downward sloping demand curves, can be derived even if human agents act irrationally.
7 Paul Samuelson echoed the point made here when he said of Becker's concept of a "taste for discrimination," "[it] wasn't so much wrong as it was empty; 'tastes for discrimination' are not an explanation of behavior but merely a ghost that gets blamed for observed events" (cited in Alhadeff 1986:185). A similar comment can be made about Becker's (1996) attempt to explain phenomena such as habit, addiction, and the size of intertemporal discount rates by using new theoretical constructs he called "personal capital," "social capital," and "imagination capital." There also appears to be inconsistency between the statement of Stigler and Becker (1977:89) that "no significant behavior has been illuminated by assumptions of differences in tastes" and Becker's earlier theory of discrimination (Becker 1957), in which the size of the majority-minority wage differential is explained, in part, by the distribution of tastes for discrimination among employers and employees.
8 That happiness and satisfaction are different concepts and the former has primacy over the latter is illustrated by a thought experiment proposed by Nozick (1989). Imagine an "experience machine" that can, by activating relevant brain circuits, produce any set of feelings or sensations you desire (for example, friendship, the taste of great food, orgasm). You may select them on the machine in any combination and for as long as desired. Assuming all your physiological needs are also met, so that you remain in perfect health, would you choose to spend the rest of your life attached to the machine? Most people, Nozick conjectures, would say "no."
9 Neoclassical theory specifies utility as a positive function of the level or stock of "goods" (for example, income). A behavioral perspective suggests, however, that it is the gap or difference between desired and actual levels of "goods" that is the stronger determinant of utility (see Simon 1982:209-38). According to Michalos (1986), empirical research in the behavioral sciences provides strong support for gap-theoretic models, although controversy continues over the appropriate specification. One gap model, for example, is the difference between a person's actual and expected amount of a good, while a second is the difference between the amount of a good a person has and that possessed by a referent group (as in a relative deprivation model in sociology). Gap models, because they are inherently relational, make it easier and more natural to introduce issues of fairness, social comparisons, and emotional states, such as frustration-aggression, than is true with neoclassical economic models. The gap-theoretic specification also implies that models of adaptive change, such as those that make use of difference and differential equations, are more appropriate than models that use calculus to solve constrained optimization problems (see Day and Groves 1975; Simon 1982:209-38).
10 In addition to relative income, labor supply is also influenced by target income behavior (Dunn 1978; Camerer, Babcock, Lowenstein, and Thaler 1997; Altman 1998) - a fact ignored in neoclassical labor supply research. Target income behavior not only is readily explicable in terms of well-accepted behavioral theories of motivation (e.g., goal setting) but also provides economists with additional information on the structure of indifference curves that, in turn, improves economists' ability to predict the size and direction of income and substitution effects. (With target income behavior, indifference curves have a small marginal rate of substitution, creating large income effects and small substitution effects.)
11 based on detailed data collected from a national study of time allocation, Juster (1991:95) concluded, "The chief surprise in these data is the next-to-last finding ... that the intrinsic rewards from work are, on average, at least as high as the intrinsic rewards from leisure. If that result is taken at face value, it suggests that economists need to do a major rethinking of the elements that go into individual utility functions." Also relevant is the finding that prime-age lottery winners who quit work report, on average, less satisfaction with life (Gould 1995).
12 Other factors could also account for this pattern that are consistent with the income-leisure model, such as systematic differences in nonlabor income or differential coverage of overtime laws.
13 Norms and interdependent preferences are also slowly being introduced into the economics literature in studies of altruism and conformity. See Rotemberg (1994) and Bernheim (1994) for examples.
Lazear (1991) briefly discussed the role of norms in pay determination, but concluded (p. 95) that the concept is "somewhat unsatisfying."
14 Veblen (pp. 389-90) first satirized the extrinsically motivated, equilibrium nature of economic man in these words: "[economic man is] a lightning calculator of pleasures and pains, who oscillates like a homogeneous globule of desire of happiness under the impulse of stimuli that shift him about the area, but leave him intact.... He is an isolated, definitive human datum, in stable equilibrium except for the buffets of the impinging forces that displace him in one direction or another." He then suggested an alternative conceptualization that emphasizes internal motivation: "It is the characteristic of man to do something, not simply to suffer the pleasures and pains through the impact of suitable forces. He is not simply a bundle of desires ... , but rather a coherent structure of propensities and habits which seek realization and expression through unfolding activity" (pp. 390-91).
15 Maslow's theory is perhaps best known but has not received much empirical support. Herzberg's and Alderfer's need-based theories have done better on this score, but remain controversial. Kanfer (1992) concluded that need-based theories cannot, by themselves, adequately explain human motivation but have a role to play as part of a larger, more integrative theory that also includes cognitive models, such as Vroom's (1964) expectancy theory. My advocacy of the needs concept for economic theory is in the same spirit.
16 Paradoxically, however, Card and Krueger do not use a shock effect argument, or any other behavioral-based theory, to explain the absence of a negative employment effect (see Chapter 11), despite their acknowledgment (Chapter 1) that it was exactly these kinds of noneconomic considerations that the "social labor economists" of the 1940s-1950s (e.g., Richard Lester) used as their central explanatory variables. In this vein, Osterman (1995) noted in his review of the book (p. 841), "There is no sense in any of the models of the group or social aspects of the labor market, even though several of the authors' findings - such as the idea of a 'just' wage and the importance of maintaining the relative wage structure - point in this direction."
17 Becker stated, for example, "Everyone more or less agrees that rational behavior simply implies consistent maximization of a well-ordered function, such as a utility or profit function" (1976:153).
18 Mortensen stated, for example, that the "more convincing explanation for a declining reservation wage is the likely possibility that most unemployed workers are liquidity constrained" (1986:859-60). Neither his comprehensive review of the search literature, nor that by Devine and Kiefer (1991), gives more than passing attention to the role of psychological factors in the search process in general and the discouraged worker effect in particular.
19 Two other examples provide further evidence of the systematic, quantitatively important bias that exists in human perception and learning. In the first, Gerhart and Milkovich reported that the responses received from employers in a 1990 survey of pay rates "indicated that none of the 124 organizations reported that they were below the median.... we know of no population where the laws of statistics permit everyone to be at or above the median" (1992:488). This finding, it should be noted, is entirely consistent with the finding of Lloyd Reynolds (1951) forty years earlier in his study of wage determination in the New Haven, Connecticut labor market. In the firms in the lower half of the wage distribution, 80% of the workers interviewed stated their wages compared favorably with wages in other plants in the area (p. 214). In the second example, Meyer (1975) surveyed a large number of employees in several firms and found that 95% rated themselves above average in job performance.
20 A further extension of price theory may be to the study of psychology and basic psychological processes such as motivation, per the suggestive observation of Kanfer (1992:81) that "almost all contemporary theories [in psychology] recognize that motivation refers to dynamic resource allocation processes."
21 An example of economic imperialism carried to excess is Stigler and Becker (1977). They claim there are two approaches to dealing with "tastes" in explaining economic behavior. The first is whenever economic behavior appears to rest on differences in tastes the inquiry is abandoned or relegated to psychologists or sociologists, given what they characterize (p. 76) as "the conventional view of inscrutable, often capricious tastes." They advocate a second "economic" approach in which tastes are assumed to be identical across people and the economist instead "continues to search for differences in prices or incomes to explain any differences or changes in behavior." The latter approach is certainly a legitimate research strategy, albeit one I predict will ultimately be unsuccessful; the objectionable part is the straw man of "inscrutable, often capricious tastes" they erect to support it. No citations are provided to support this characterization of tastes, and it ignores altogether the many credible, empirically supported theories of human motivation, cognition, and affect in the literature of psychology and dismisses out-of-hand the theoretical work of well-respected behavioral economists, such as Simon and Leibenstein.
The view that tastes are inscrutable and capricious, and that this rationalizes neglecting psychological factors in economic behavior, is long-standing and generic to mainstream neoclassical economics- per the observation of Commons (1924:3) 75 years ago: "Early [classical and neoclassical] theories attempted to get away from the human will, since that was conceived to be internal, capricious, not subject to law, and therefore, economics should be reduced to one of the nature sciences, analogous to chemistry, physics, or physiology. It would be a theory of commodities or mechanisms, not a theory of the will." A proposition of both behavioral and institutional economics is that the structure of tastes, and differences therein, are governed by their own laws and these laws can be used to improve the predictive ability of economic theory. This represents a middle way between the two approaches proposed by Stigler and Becker, and it is what I try to elucidate in this paper.
22 Economists (see Kagel, Battalio, and Green 1995) have also performed numerous laboratory experiments with animals, such as pigeons and rats, and have found that their behavior accords in a number of respects with the predictions of microeconomic theory (for example, quantity of grain pellets demanded declines with an increase in the price). Just as the behavior of adults has many similarities to that of children, so too does human behavior mimic animal behavior in a number of respects. But the differences are also profound, for, as Commons (1934) noted, humans have a purposeful will, are forward looking, and have unique needs related to personal growth and development. These aspects of human psychology are not well captured by microeconomic theory.
Abowd, John, and Henry Farber. 1982. "Job Queues and the Union Status of Workers." Industrial and Labor Relations Review, Vol. 35, No. 3 (April), pp. 354-67.
Adams, J. Stacy. 1963. "Toward an Understanding of Inequity." Journal of Abnormal Psychology, Vol. 67 (November), pp. 422-36.
Akerlof, George. 1990. "The Fair Wage-Effort Hypothesis and Unemployment." Quarterly Journal of Economics, Vol. 105 (May), pp. 255-84.
Alchian, Armen, and Harold Demsetz. 1972. "Production, Information Costs, and Economic Organization." American Economic Review, Vol. 62 (December), pp. 777-95.
Alderfer, Clayton. 1972. Existence, Relatedness, and Growth. New York: Free Press.
Alhadeff, D. 1982. Microeconomics and Human Behavior. Berkeley: University of California Press.
Altman, Morris. 1998. "Preferences and Labor Supply: Casting Some Light into the Black Box of Income-Leisure Choice." Unpublished paper, Department of Economics, University of Saskatchewan.
Annable, James. 1984. The Price of Industrial Labor. Lexington, Mass: Lexington Books.
Averill, James R., and Thomas A. More. 1993. "Happiness." In Michael Lewis and Jeannette M. Haviland, eds., Handbook of Emotions. New York: Guilford, pp. 617-29.
Bandura, Albert. 1986. Social Foundations of Thought and Action. Englewood Cliffs, N.J.: Prentice-Hall.
Bardwick, Judith M. 1991. Danger in the Comfort Zone. New York: AMACOM.
Baron, James. 1988. "The Employment Relation as a Social Relation." Journal of Japanese and International Economy, Vol. 2 (December), pp. 492-525.
Baron, James, and Michael Hannan. 1994. "The Impact of Economics on Contemporary Sociology." Journal of Economic Literature, Vol. 32 (September), pp. 1111-46.
Barron, John, and Kathy Paulson Gjerde. 1997. "Peer Pressure in an Agency Relationship." Journal of Labor Economics, Vol. 15 (April), pp. 234-54.
Baxter, J. L. 1988. Social and Psychological Foundations of Economic Analysis New York: Harvester-Wheatsheaf.
Baxter, J. L., and I. A. Moosa. 1996. "The Consumption Function: A Basic Needs Hypothesis." Journal of Economic Behavior and Organization, Vol. 31, No. 1, pp. 85-100.
Becker, Gary. 1957. The Economics of Discrimination. Chicago: University of Chicago Press.
-----. 1976. The Economic Approach to Human Behavior. Chicago: University of Chicago Press.
-----. 1993. "Nobel Lecture: The Economic Way of Looking at Behavior." Journal of Political Economy, Vol. 101 (June), pp. 385-409.
-----. 1996. Accounting for Tastes. Cambridge, Mass.: Harvard University Press.
Bernheim, B. Douglas. 1994. "A Theory of Conformity." Journal of Political Economy, Vol. 102, No. 5, pp. 841-77.
Bewly, Truman F. 1997. "A Depressed Labor Market as Explained by Participants." Unpublished paper. New Haven: Yale University.
Binmore, Ken, and Larry Samuelson. 1994. "An Economist's Perspective on the Evolution of Norms." Journal of Institutional and Theoretical Economics, Vol. 150, No. 1, pp. 45-63.
Blanchflower, David, and Richard Freeman. 1992. "Unionism in the United States and Other Advanced OECD Countries." Industrial Relations, Vol. 31 (Winter), pp. 56-79.
Blau, David, and Alison Hagy. 1998. "The Demand for Quality in Child Care." Journal of Political Economy, Vol. 106 (February), pp. 104-46.
Blau, Francine, and Marianne Ferber. The Economics of Women, Men, and Work. Englewood Cliffs, NJ.: Prentice-Hall.
Blinder, Alan, and Choi. 1990. "A Shred of Evidence on Theories of Wage Stickiness." Quarterly Journal of Economics, Vol. 105 (November), pp. 1003-15.
Borjas, George. 1996. Labor Economics. New York: McGraw-Hill.
Bowles, Samuel. 1998. "Endogenous Preferences: The Cultural Consequences of Markets and Other Economic: Institutions" Journal of Economic Literature, Vol. 36 (March), pp. 75-111.
Brett, Jean. 1980. "Why Employees Join Unions." Organizational Dynamics, Vol. 8 (Spring), pp. 47-59.
Buchanan, James. 1989. "Rational Choice Models in the Social Sciences." In James Buchanan, Explorations into Constitutional Economics. College Station: Texas A&M Press, pp. 37-50.
Burns, R. B. 1979. The Self-Concept. London: Longman.
Burtless, Gary, and Robert Moffitt. 1985. "The Joint Choice of Retirement Age and Postretirement Hours of Work." Journal of Labor Economics, Vol. 3 (April), pp. 209-36.
Camerer, Colin, Linda Babcock, George Lowenstein, and Richard Thaler. 1997. "Labor Supply of New York City Cabdrivers: One Day at a Time." Quarterly Journal of Economics, Vol. 112 (May), pp. 407-41.
Cappelli, Peter, et al. 1997. Change at Work. New York: Oxford University Press.
Card, David, and Alan Krueger. 1995. Myth and Measurement: The New Economics of the Minimum Wage. Princeton, N.J.: Princeton University Press.
Clark, Kim. 1980. "The Impact of Unionization on Productivity: A Case Study." Industrial and Labor Relations Review, Vol. 33, No. 4 (July), pp. 451-69.
Coleman, James S. 1988. "Norms as Social Capital." In Gerald Radnitzky and Peter Bernholz, eds., Economic Imperialism: The Economic Method Applied outside the Field of Economics. New York: Paragon, pp. 133-55.
-----. 1934. Institutional Economics. New York: MacMillan.
Commons, John R. 1924. Legal Foundations of Capitalism. New York: MacMillan.
Commons, John R., and John Andrews. 1936. Principles of Labor Legislation. New York: Harper & Bros.
Conlisk, John. 1996. "Why Bounded Rationality?" Journal of Economic Literature, Vol. 34 (June), pp. 669-700.
Cooter, Robert, and Peter Rappoport. 1984. "Were the Ordinalists Wrong about Welfare Economics?" Journal of Economic Literature, Vol. 22 (June), pp. 507-30.
Cross, John. 1969. The Economics of Bargaining. New York: Basic Books.
Darity, William, and Arthur Goldsmith. 1996. "Social Psychology, Unemployment, and Macroeconomics." Journal of Economic Perspectives, Vol. 10 (Winter), pp. 121-40.
Day, Richard, and Theodore Groves, eds. 1975. Adaptive Economic Models. New York: Academic Press.
Deci, Edward L. 1975. Internal Motivation. New York: Plenum.
Deci, Edward L., and Richard M. Ryan. 1985. Intrinsic Motivation and Self-Determination in Human Behavior. New York: Plenum.
Devine, Patricia. 1989. "Automatic and Controlled Procedures in Prejudice: The Role of Stereotypes and Personal Beliefs." In Anthony Pratkanis, Steven Breckler, and Anthony Greenwald, eds., Attitude Structure and Function. Hillsdale, N.J.: L. Erlbaum, pp. 181-212.
Devine, Theresa, and Nicholas Kiefer. 1991. Empirical Labor Economics: The Search Approach. New York: Oxford University Press.
Duesenberry, James. 1949. Income, Saving, and the Theory of Consumer Behavior. Cambridge, Mass.: Harvard University Press.
Dunn, L. F. 1978. "An Empirical Indifference Curve for Income and Leisure." Review of Economics and Statistics, Vol. 60 (November), pp. 275-82.
Easterlin, Richard. 1995. "Will Raising the Incomes of All Increase the Happiness of All?" Journal of Economic Behavior and Organization, Vol. 27, No. 1, pp. 35-47.
Ehrenberg, Ronald G., and Robert S. Smith. 1997. Modern Labor Economics, 6th ed. New York: Harper Collins.
Elkins, Paul, and Manfred Max-Neef, eds. 1992. Real-Life Economics: Understanding Wealth Creation. New York: Routledge.
Elster, Jon. 1989. "Social Norms and Economic Theory." Journal of Economic Perspectives, Vol. 3 (Fall), pp. 99-117.
-----. 1998. "Emotions and Economic Theory." Journal of Economic Literature, Vol. 36 (March), pp. 4774.
Ely, Richard, and Frank Bohn. 1935. The Great Change. New York: Thomas Nelson.
Encinosa, William III, Martin Gaynor, and James Rebitzer. 1997. "The Sociology of Groups and the Economics of Incentives: Theory and Evidence on Compensation Systems." Working Paper 5953. Cambridge, Mass.: National Bureau of Economic Research.
Epstein, Cynthia. 1981. Women in Law. New York: Basic Books.
Epstein, Seymour. 1993. "Emotion and Self-Theory." In Michael Lewis and Jeannette M. Haviland, eds., Handbook of Emotions, pp. 313-26. New York: Guilford.
Etzioni, Amitai. 1986. "The Case for a Multiple Utility Conception." Economics and Philosophy, Vol. 2 (October), pp. 159-83.
Farber, Henry. 1984. "Right to Work Laws and the Extent of Unionization." Journal of Labor Economics, Vol. 2 (July), pp. 319-52.
Fehr, Ernst, Erich Kirchler, Andreas Weichbold, and Simon Gachter. 1998. "When Social Norms Overpower Competition: Gift Exchange in Experimental Labor Markets." Journal of Labor Economics, Vol. 16 (April), pp. 324-51.
Festinger, Leon. 1957. A Theory of Cognitive Dissonance. Stanford, Calif.: Stanford University Press.
Fisher, Cynthia D., and Edwin A. Locke. 1992. "The New Look in Job Satisfaction Research and Theory." In C.J. Cranny, Patricia Cain Smith, and Eugene F. Stone, eds., Job Satisfaction. Lexington, Mass.: Lexington Books, pp. 165-94.
Foster, E. Michael. 1993. "Labor Economics and Public Policy: Dominance of Constraints or Preferences?" In William Darity, Jr., ed., Labor Economics: Problems in Analyzing Labor Markets. Boston, Mass.: Kluwer, pp. 269-94.
Frank, Robert H. 1985. Choosing the Right Pond. New York: Oxford University Press.
-----. 1988. Passions within Reason: The Strategic Role of the Emotions. New York: Norton.
Frank, Robert H., and Philip Cook. 1995. The Winner-Take-All Society. New York: Free Press.
Freeman, Richard, and Morris Kleiner. 1990. "Employer Behavior in the Face of Union Organizing Drives." Industrial and Labor Relations Review, Vol. 43, No. 3 (April), pp. 351-65.
Freeman, Richard, and James Medoff. 1984. What Do Unions Do? New York: Basic Books.
Friedman, Milton. 1953. Essays in Positive Economics. Chicago: University of Chicago Press.
Fuchs, Victor. 1994. "Nobel Laureate Gary S. Becker: Ideas about Facts." Journal of Economic Perspectives, Vol. 8 (Spring), pp. 183-92.
Georgescu-Roegen, Nicholas. 1968. "Utility." In International Encyclopedia of the Social Sciences, Vol. 16. New York: MacMillan, pp. 236-67.
Gerhart, Barry, and George Milkovich. 1990. "Compensation." In Marvin Dunnette and Leaetta Hough, eds., Handbook of Industrial and Organizational Psychology, 2nd ed., Vol. 3. Palo Alto, Calif.: Consulting Psychologists Press, pp. 481-570.
Glick, Peter, Korin Wilk, and Michele Perreault. 1995. "Images of Occupations: Components of Gender and Status in Occupational Stereotypes." Sex Roles, Vol. 32 (May), pp. 565-82.
Glickman, Lawrence. 1997. The Living Wage. Ithaca, N.Y.: Cornell University Press.
Gould, Lois. 1995. "Ticket to Trouble." New York Times Magazine (April 23), pp. 40-41, 54, 89.
Groshen, Erica. 1992. "Sources of Intra-Industry Wage Dispersion: How Much Do Employers Matter?" Quarterly Journal of Economics, Vol. 106 (May), pp. 869-84.
Haberfeld, Yitchak. 1992. "Pay, Valence of Pay, and Gender: A Simultaneous Equations Model." Journal of Economic Psychology, Vol. 13, No. 1, pp. 93-109.
Haines, Walter. 1982. "Psychoeconomics of Human Needs: Maslow's Hierarchy and Marshall's Organic Growth." Journal of Behavioral Economics, Vol. 11 (Winter), pp. 97-121.
Hall, Robert 1986. "Discussion." In Alicia Munnell, ed., Lessons from the Negative Income Tax Experiments. Washington, D.C.: Brookings Institution, pp. 56-59.
Harris, Paul L. 1993. "Understanding Emotion." In Michael Lewis and Jeannette Haviland, eds., Handbook of Emotions, pp. 237-46. New York: Guilford.
Harsanyi, John. 1997. "Utilities, Preferences, and Substantive Goods" Social Choice and Welfare, Vol. 14, No. 1, pp. 129-45.
Hausman, David, and Michael McPherson. 1996. Economic Analysis and Moral Philosophy New York: Cambridge University Press.
Herzberg, Frederick. 1966. Work and the Nature of Man. Cleveland: World.
Hildreth, Andrew K. G., and Andrew J. Oswald. 1997. "Rent-Sharing and Wages: Evidence from Company and Establishment Panels." Journal of Labor Economics, Vol. 15 (April), pp. 318-37.
Hirsch, Barry, and John Addison 1986. The Economic Analysis of Unions London: Allen & Unwin.
Hochschild, Arlie Russell. 1997. The Time Bind. New York: Henry Holt.
Hogarth, Robin. 1987. Judgement and Choice, 2nd ed. New York: John Wiley.
Holmstrom, Bengt, and Paul Milgrom. 1994. "The Firm as an Incentive System" American Economic Review, Vol. 84, No. 4, pp. 972-91.
Inglehart, Ronald, and Jacques-Rene Rabier. 1986. "Aspirations Adapt to Situations - But Why Are the Belgians So Much Happier Than the French?" In Frank M. Andrews, ed., Research on the Quality of Life. Ann Arbor, Mich.: Institute for Social Research, University of Michigan, pp. 1-56.
Isen, Alice M. 1993. "Positive Affect and Decision Making." In Michael Lewis and Jeannette Haviland, eds., Handbook of Emotions. New York: Guilford Press, pp. 261-77.
Juster, F. Thomas. 1991. "Rethinking Utility Theory: A Summary Statement." In Amitai Etzioni and Paul Lawrence, eds., Socio-Economics: Toward a New Synthesis. Armonk, N.Y.: M.E. Sharp, pp. 85-104.
Kagel, John, Raymond Battalio, and Leonard Green. 1995. Economic Choice Theory: An Experimental Analysis of Animal Behavior. New York: Cambridge University Press.
Kahn, Shulamit. 1997. "Evidence of Nominal Wage Stickiness from Microdata." American Economic Review, Vol. 87 (December), pp. 993-1008.
Kahneman, Daniel 1994. "New Challenges to the Rationality Assumption." Journal of Institutional and Theoretical Economics, Vol. 150, No. 1, pp. 18-36.
Kahneman, Daniel, Jack Knetsch, and Richard Thaler. 1986. "Fairness as a Constraint on Profit Seeking: Entitlements in the Market." American Economic Review, Vol. 76 (September), pp. 728-41.
-----. 1991. "Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias." Journal of Economic Perspectives, Vol. 5 (Winter), pp. 193-206.
Kahneman, Daniel, and Amos Tversky. 1979. "Prospect Theory: An Analysis of Decisions under Risk." Econometrica, Vol. 47 (March), pp. 263-91.
Kahneman, Daniel, Peter Wakker, and Rakesh Sarin. 1997. "Back to Bentham? Explorations of Experienced Utility." Quarterly Journal of Economics, Vol. 112 (May), pp. 375-405.
Kanfer, Ruth. 1992. "Motivation Theory and Industrial Organizational Psychology." In Marvin Dunnette and Leatta Hough, eds., Handbook of Industrial Organizational Psychology, Vol. 1. Palo Alto, Calif.: Consulting Psychologists Press, pp. 75-170.
Katona, George. 1975. Psychological Economics. New York: Elsevier.
Kaufman, Bruce. 1988. "The Postwar View of Labor Markets and Wage Determination." In Bruce E. Kaufman, ed., How Labor Markets Work. Lexington, Mass.: Lexington Books, pp. 145-203.
-----. 1989. "Models of Man in Industrial Relations Research" Industrial and Labor Relations Review, Vol. 43, No. 1 (October), pp. 72-88.
-----. 1990. "A New Theory of Satisficing." Journal of Behavioral Economics, Vol. 19 (Spring), pp. 35-51.
-----. 1992. "Research on Strike Models and Outcomes in the 1980s: Accomplishments and Shortcomings." In David Lewin, Olivia Mitchell, and Peter Sherer, eds., Research Frontiers in Industrial Relations and Human Resources. Madison, Wis.: Industrial Relations Research Association, pp. 77-129.
-----. 1994. "The Evolution of Thought on the Competitive Nature of Labor Markets." In Clark Kerr and Paul Staudohar, eds., Labor Economics and Industrial Relations: Markets and Institutions. Cambridge, Mass: Harvard University Press, pp. 145-88.
-----. 1997. "Labor Markets and Employment Regulation: The View of the 'Old Institutionalists.'" In Bruce E. Kaufman, ed., Government Regulation of the Employment Relationship. Madison, Wis.: Industrial Relations Research Association, pp. 11-55.
-----. 1999 "Emotional Arousal as a Source of Bounded Rationality." Journal of Economic Behavior and Organization, forthcoming.
Kaufman, Bruce, and Paula E. Stephan. 1995. "The Role of Management Attorneys in Union Organizing Campaigns." Journal of Labor Research, Vol. 16 (Fall), pp. 439-54.
Kennan, John. 1986. "The Economics of Strikes." In Orley Ashenfelter and Richard Layard, eds., Handbook of Labor Economics, Vol. 2. New York: North-Holland, pp. 1091-1137.
Killingsworth, Mark R. 1983. Labor Supply. New York: Cambridge University Press.
Kim, Marlene. 1999. "Inertia and Discrimination in the California Civil Service." Industrial Relations, Vol. 38 (January), pp. 46-68.
Kohn, Alfie. 1993. Punished by Rewards. New York: Houghton Mifflin.
Kreps, David. 1997a. "Intrinsic Motivation and Extrinsic Incentives." American Economic Review, Vol. 87 (May), pp. 359-64.
-----. 1997b. "Economics: The Current Position." Daedalus, Vol. 126 (Winter), pp. 59-86.
Kwoka, John. 1983. "Monopoly, Plant, and Union Effects on Worker Wages." Industrial and Labor Relations Review, Vol. 36, No. 2 (January), pp. 251-57.
Lagueux, Maurice. 1993. "Kirzner vs. Becker: Rationality and Mechanisms in Economics" In Robert Hebert, ed., Perspectives on the History of Economic Thought, Vol. 9. New York: Edward Elgar, pp. 37-50.
Lane, Robert E. 1991. The Market Experience. New York: Cambridge University Press.
Lawler, Edward. 1992. The Ultimate Advantage. San Francisco: Jossey-Bass.
Lazarus, Richard S. 1991. Emotion and Adaptation New York: Oxford University Press.
Lazear, Edward. 1991. "Labor Economics and the Psychology of Organizations" Journal of Economic Perspectives, Vol. 5 (Spring), pp. 89-110.
-----. 1995. Personnel Economics. Cambridge, Mass.: MIT Press.
Lea, Stephen, Roger Tarpy, and Paul Webley. 1987. The Individual in the Economy: A Textbook of Economic Psychology. New York: Cambridge University Press.
Leibenstein, Harvey. 1987. Inside the Firm. Cambridge, Mass: Harvard University Press.
Levine, David. 1993. "Fairness, Markets, and Ability to Pay." American Economic Review, Vol. 83 (December), pp. 1241-59.
-----. 1995. Reinventing the Workplace. Washington, D.C.: Brookings Institution.
Lindbeck, Assar. 1997. "Incentives and Social Norms in Household Behavior." American Economic Review, Vol. 87 (May), pp. 369-77.
Locke, Edwin, and Gary Latham. 1990. A Theory of Goal-Setting. Englewood Cliffs, N.J.: Prentice Hall.
Lutz, Mark A., and Kenneth Lux. 1988. Humanistic Economics: The New Challenge New York: Bootstrap.
Mack, Mary. 1969. A Bentham Reader. New York: Pegasus.
Main, Brian G. M., Charles O'Reilly, and James Wade. 1993. "Top Executive Pay: Tournament or Team-work?" Journal of Labor Economics, Vol. 11 (October), pp. 606-28.
Mas-Colell, Andreu, Michael Whinston, and Jerry Green. 1995. Microeconomic Theory. New York: Oxford University Press.
Maslow, Abraham. 1954. Motivation and Personality. New York: Harper & Row.
McAdams, Richard. 1992. "Relative Preferences" Yale Law Journal, Vol. 102 (October), pp. 1-104.
McGregor, Douglas. 1960. The Human Side of Enterprise. New York: McGraw-Hill.
Meyer, Herbert. 1975. "The Pay for Performance Dilemma." Organizational Dynamics, Vol. 3 (Winter), pp. 22-38.
Michalos, Alex C. 1986. "Job Satisfaction, Marital Satisfaction, and the Quality of Life: A Review and a Preview" In Frank M. Andrews, ed., Research on the Quality of Life. Ann Arbor, Mich.: Institute for Social Research, University of Michigan, pp. 57-84.
Miller, Gary J. 1991. Managerial Dilemmas. New York: Cambridge University Press.
Mortensen, Dale. 1986. "Job Search and Labor Market Analysis." In Orley Ashenfelter and Richard Layerd, eds., Handbook of Labor Economics, Vol. 2. Amsterdam: North-Holland, pp. 849-919.
Mullahy, John, and Jody Sindelar. 1993. "Alcoholism, Work, and Income" Journal of Labor Economics, Vol. 11 (July), pp. 494-520.
Munnell, Alicia H., ed. 1986. Lessons from the Income Maintenance Experiments Washington, D.C.: Brookings Institution.
Neubourg, Chris de, and Maarten Vendrik. 1994. "An Extended Rationality Model of Social Norms in Labour Supply." Journal of Economic Psychology, Vol. 15, pp. 93-126.
Nicholson, Nigel, and Joseph Martocchio. 1995. "The Management of Absence: What Do We Know? What Can We Do?" In Gerald Ferris, Sherman Rosen, and Darold Barnum, eds., Handbook of Human Resource Management. Cambridge, Mass: Blackwell, pp. 597-629.
Nozick, R. 1989. The Examined Life. New York: Simon & Schuster.
Olson, Mancur. 1965. The Logic of Collective Action. Cambridge: Harvard University Press.
Osterman, Paul. 1995. "Comment" Industrial and Labor Relations Review, Vol. 48 (July), pp. 839-42.
Pareto, Vilfredo. 1927. Manual of Political Economy (1971 English translation). New York: Augustus Kelley.
Pencavel, John H. 1971. "The Demand for Union Services: An Exercise." Industrial and Labor Relations Review, Vol. 24, No. 2 (January), pp. 180-90.
-----. 1986. "Labor Supply of Men: A Survey." In Orley Ashenfelter and Richard Layard, eds., Handbook of Labor Economics, Vol. 1. Amsterdam: North-Holland, pp. 3-102.
Petri, Herbert. 1986. Motivation: Theory and Research, 2nd ed. Belmont, Calif.: Wadsworth.
Piore, Michael. 1995. Beyond Individualism. Cambridge, Mass.: Harvard University Press.
Pollachek, Solomon. 1981. "Occupational Self-Selection: A Human Capital Approach to Sex Differentials in Occupational Structure." Review of Economics and Statistics, Vol. 63 (February), pp. 60-69.
Rebitzer, James. 1993. "Radical Political Economy and the Economics of Labor Markets." Journal of Economic Literature, Vol. 31 (September), pp. 1394-1434.
Rees, Albert. 1993. "The Role of Fairness in Wage Determination." Journal of Labor Economics, Vol. 11 (January, pt. 1), pp. 243-52.
Reynolds, Lloyd 1951. The Structure of Labor Markets. New York: Harper.
Rosen, Sherwin. 1986. "The Theory of Compensating Differentials." In Orley Ashenfelter and Richard Layard, eds., Handbook of Labor Economics, Vol. 1. Amsterdam: North-Holland, pp. 641-92.
Rosenberg, Morris. 1979. Conceiving the Self. New York: Basic Books.
Ross, Arthur M. 1948. Trade Union Wage Policy. Berkeley: University of California Press.
Rottemberg, Julio. 1994. "Human Relations in the Workplace." Journal of Political Economy, Vol. 102, No. 4, pp. 684-718.
Rotter, J. B. 1966. "Generalized Expectancies for Internal versus External Control of Reinforcement." Psychological Monographs, Vol. 80, pp. 1-28.
Ruggles, Patricia. 1990. Drawing the Line: Alternative Poverty Measures and Their Implications for Public Policy. Washington, D.C.: Urban Institute.
Russell, Thomas, and Richard Thaler. 1985. "The Relevance of Quasi-Rationality in Competitive Markets." American Economic Review, Vol. 75 (December), pp. 1071-82.
Seager, Henry. 1913. "The Theory of the Minimum Wage." American Labor Legislation Review, Vol. 3, No. 3, pp. 81-91.
Seligman, M. E. P. 1976. Learned Helplessness and Depression in Animals and Man. Morristown, N.J.: General Learning Press.
Sen, Amartya. 1973. "Behavior and the Concept of Preference." Economica, Vol. 40 (August), pp. 241-59.
-----. 1977. "Rational Fools: A Critique of the Behavioral Foundations of Economic Theory." Philosophy and Public Affairs, Vol. 6 (Summer), pp. 317-44.
Shapiro, Carl, and Joseph Stiglitz. 1984. "Equilibrium Unemployment as a Worker Discipline Device." American Economic Review, Vol. 74 (June), pp. 433-44.
Sheppard, Blair, Roy Lewicki, and John Minton. 1992. Organizational Justice: The Search for Fairness in the Workplace. Lexington, Mass.: Lexington Books.
Simon, Herbert A. 1982. Models of Bounded Rationality, Vol. 2. Cambridge, Mass.: MIT Press.
Skinner, B. F. 1969. Contingencies of Reinforcement. New York: Appelton-Century-Crofts.
Slichter, Sumner. 1929. "The Current Labor Policies of American Industries." Quarterly Journal of Economics, Vol. 43 (May), pp. 393-435.
Slovic, Paul. 1995. "The Construction of Preference." American Psychologist, Vol. 50 (May), pp. 364-71.
Smith, James, and Michael Ward. 1985. "Time-Series Growth in the Female Labor Force" Journal of Labor Economics, Vol. 3 (January, supplement), pp. S59-S90.
Solow, Robert. 1997. "How Did Economics Get That Way and What Way Did It Get?" Daedalus, Vol. 126 (Winter), pp. 39-58.
Stigler, George. 1946. "The Economics of Minimum Wage Legislation." American Economic Review, Vol. 36 (June), pp. 358-65.
-----. 1976. "The Xistence of X-efficiency." American Economic Review, Vol. 66 (March), pp. 213-16.
Stigler, George, and Gary Becker. 1977. "De Gustibus Non Est Disputandum." American Economic Review, Vol. 67 (March), pp. 76-90.
Sunstein, Cass. 1996. "Social Norms and Social Roles." Columbia Law Review, Vol. 96 (May), pp. 903-68.
Thaler, Richard. 1989. "Anomalies: Interindustry Wage Differentials." Journal of Economic Perspectives, Vol. 3 (Spring), pp. 181-93.
-----. 1991. Quasi-Rational Economics. New York: Russell Sage Foundation.
Thaler, Richard, and H. Shefrin. 1981. "An Economic Theory of Self-Control." Journal of Political Economy, Vol. 89 (April), pp. 392-406.
Triandis, Harry. 1995. Individualism and Collectivism. Boulder, Col.: Westview.
Tversky, Amos, and Daniel Kahneman. 1991. "Loss Aversion in Riskless Choice: A Reference-Dependence Model." Quarterly Journal of Economics, Vol. 106 (November), pp. 1038-61.
Veblen, Thorstein. 1898. "Why Is Economics Not an Evolutionary Science?" Quarterly Journal of Economics, Vol. 12 (July), pp. 373-97.
Viscusi, W. Kip. 1993. "The Value of Risks to Life and Health." Journal of Economic Literature, Vol. 31 (December), pp. 1219-46.
Vroom, Victor H. 1964. Work and Motivation. New York: Wiley.
Walton, Richard, and Robert McKersie. 1965. A Behavioral Theory of Labor Negotiations. New York: McGraw-Hill.
Weis, Howard. 1992. "Learning Theory." In Marvin Dunnette and Leatta Hough, eds., Handbook of Industrial Organizational Psychology, Vol. 1. Palo Alto, Calif.: Consulting Psychologists Press, pp. 171-221.
Wheeler, Hoyt N. 1985. An Integrative Theory of Industrial Conflict. Columbia: University of South Carolina Press.
Whyte, William Foote. 1955. Money and Motivation: An Analysis of Incentives in Industry. New York: Harper.
Williamson, Oliver. 1985. The Economic Institutions of Capitalism. New York: Free Press.
Bruce E. Kaufman is Professor of Economics, Research Director of the W. J. Usery Center for the Workplace, and Senior Associate of the W. T. Beebe Institute of Personnel and Employment Relations at Georgia State University.
|Printer friendly Cite/link Email Feedback|
|Author:||Kaufman, Bruce E.|
|Date:||Apr 1, 1999|
|Previous Article:||Are mandated health and safety committees substitutes for or supplements to labor unions?|
|Next Article:||Minimum wages and poverty.|