خبازان خودخواه، پرستاران مراقب ؟مدل انگیزش کار

Work contributes to people’s self-image in important ways. We propose a model in which effort is unobservable and where individuals have a preference for being important to others. This gives the following predictions: (1) if a worker’s effort is paid by his marginal productivity (bakers), effort is just like in the standard model. (2) If a worker’s wage is unaffected by his effort (nurses), more effort is provided than in the standard model. (3) To prevent that shirkers become nurses, nurses’ wages must be kept strictly lower than bakers’ income. At this wage level there will be too few nurses. (4) Overinvestment in nursing equipment can be justified as a means to attract motivated nurses. (5) Even with full income compensation, both nurses and bakers may experience a net utility loss when losing their job. (6) Similarly, both nurses and bakers may prefer work to welfare, even with full income compensation.

The Homo Oeconomicus model, which assumes that individuals care only about their own access to goods and services while keeping their own efforts low, has long been the benchmark for economic analyses of labor markets and worker behavior. In many contexts, empirical evidence has indeed proved to be nicely consistent with this model (e.g. Lazear, 2000). Nevertheless, in other cases the Homo Oeconomicus predictions seem too stark. Specifically, if a worker’s effort does not affect his pecuniary payoff, a Homo Oeconomicus employee will shirk as much as possible – regardless of the possible consequences for others. Anyone who has spent some time at a hospital, a school, or a university, knows that although this description might fit the behavior of some individuals, it definitely does not fit all (Heyes, 2004). Within health economics, the need to reconcile theoretical predictions with empirical evidence has in fact led numerous scholars to assume that physicians have altruistic preferences towards their patients.1 This approach, however, may be considered somewhat ad hoc, since it is not clear why physicians would generally be more altruistic than others. Similarly, the Homo Oeconomicus predictions make Scandinavian welfare states appear almost as logical inconsistencies. While studies indicate that social insurance generosity does increase the number of recipients and spell duration (e.g. Røed and Zhang, 2005 and Johansson and Palme, 2005), such effects are much less pronounced than one might expect. In Norway, for example, employees have the right, within certain limits, to claim sick leave with full wage compensation based only on self-declared sickness (no physician’s sickness certificate); still such self-declared sick leaves constitute only about 10 percent of the total level permitted by the rules (Eielsen, 2008; see also e.g. Aronsson et al., 2000). For Homo Oeconomicus, work affects utility only in terms of lost leisure (and, of course, through the income it generates). Thus, any utility loss associated with unemployment must be due to income loss. If unemployment benefits provided full income compensation, becoming unemployed would necessarily increase Homo Oeconomicus’ utility, due to increased leisure. Substantial empirical evidence, however, shows that while the relationship between income and happiness seems surprisingly weak, unemployment is a major cause of reduced happiness and life satisfaction (see, for example, Dolan et al., 2008, Theodossiou, 1998 and Oswald, 1997). All this may indicate that some aspect of work, wage payment aside, is valued by workers. In the present paper, we propose a model of work motivation. In our model, workers have a preference for a self-image as someone who is important to others. The strength of this preference, however, may vary between individuals. These assumptions are in line with evidence reported by, among others, Crewson (1997),2 and lead to predictions consistent with the phenomena discussed above. When effort is remunerated by its marginal productivity, workers’ choice of effort will be equivalent to the Homo Oeconomicus predictions. However, in jobs where neither individual effort nor individual productivity is verifiable, workers will exert more effort than Homo Oeconomicus. Thus, a given person’s effort choice may be quite consistent with the traditional model if she works as a baker, but not if she switches jobs and becomes a nurse. Turning next to the issue of self-selection into different occupations, we show that under plausible conditions, those with intermediate preferences for being important will seek employment in the sector where effort is rewarded by its marginal productivity, while individuals with the highest and lowest work motivation are attracted to jobs where effort cannot be verified: The highly motivated are attracted by the opportunity to be important; the poorly motivated are attracted by the opportunity to shirk. Since behavior in perfectly competitive market jobs is unaffected by the preference to be important, it is socially desirable that poorly motivated workers choose employment in jobs where economic incentives prevent shirking (bakers). This can be achieved by keeping wages low in jobs with unverifiable effort (nurses). Hence, while both the profit maximizing behavior of the baker and the other-regarding behavior of the nurse are consistent with the same underlying preferences, low wages can make strongly other-regarding individuals self-select into the nursing sector. If nurses’ wage is kept sufficiently low to keep shirkers out, the nursing sector will be strictly smaller than the first-best optimum. We show, however, that there may be other ways to recruit the highly motivated than increasing the wage: nursing jobs will become more attractive, but only to motivated workers, if one invests in capital equipment that increases nurses’ efficiency in helping others. This, in fact, provides an argument for overinvestment in certain types of public sector capital; for example better diagnostic equipment in hospitals, or library resources in universities and schools. Living off welfare instead of working is detrimental for the sense of being important to others: the welfare recipient produces nothing, and his consumption must be financed by others’ tax payments. The gain in terms of increased leisure must thus be weighted against the self-image loss. We show that workers with a sufficiently strong work motivation, including both nurses and bakers, will experience a strict utility loss if they lose their job, even in the case of full wage compensation. Such individuals will prefer work to living off welfare if they have the choice. Thus, they will not, for example, claim sick just in order to shirk. Several researchers have recently proposed models of altruistic or social preferences (e.g. Ellingsen and Johannesson, 2008, Benabou and Tirole, 2006, Konow, 2010, Rauscher, 2006, Brekke and Nyborg, 2008, Brekke et al., 2003, Andreoni, 1990 and Sugden, 1984). Our approach is closely related to this research. Moreover, the ideas put forward in the present paper have much in common with an interesting recent strand of literature on public service motivation (Delfgaauw and Dur, 2008, Francois, 2000, Francois, 2007, Dur and Glazer, 2008, Prendergast, 2007, Besley and Ghatak, 2005, Heyes, 2004, Frank, 2003 and Handy and Katz, 1998). A typical finding in the latter literature is that while the public sector (or an employer with a specific mission) may be attractive to highly motivated employees, a high wage will also attract “the wrong types” (Delfgaauw and Dur, 2008, Francois, 2007, Prendergast, 2007 and Heyes, 2004). Nevertheless, in spite of its intuitive appeal, the above-mentioned papers rely on particular model assumptions that limit the applicability of the conclusion. Delfgaauw and Dur (2008), for example, assume exogenously that workers receive a private benefit from effort, but only if employed by the public sector. One consequence is that, in contrast to the model presented below, Delfgaauw and Dur’s model would imply that a nurse’s work motivation vanishes completely if the publicly owned hospital she worked for were privatized. In Francois (2007), a strong work motivation originates from a strong private preference for the public good produced by the public sector. It has been shown by a number of authors that this kind of ‘pure altruism’ motivation is, typically, unable to explain substantial voluntary contributions to public goods (Warr, 1983, Bernheim, 1986 and Andreoni, 1988). In Francois (2007), voluntary contributions result from combining pure altruism with a rather particular production function: If the individual worker does not contribute, the public good might not be provided at all. With a production function implying, instead, that a worker’s shirking changes public good supply only marginally, it is hard to see how the motivation structure proposed by Francois (2007) could produce more than negligible voluntary contributions of unobservable effort. The model presented below, however, can explain voluntary effort even with more standard production functions. The general model of work motivation presented in the current paper yields similar conclusions concerning recruitment and self-selection of motivated workers as Delfgaauw and Dur (2008) and Francois (2007). In addition, it can explain a number of phenomena that cannot be analyzed within the framework of those models. Specifically, our model can explain the fact that many individuals prefer working to living off welfare, as well as the detrimental well-being effects of unemployment. Moreover, it provides new policy recommendations concerning alternative ways to attract motivated workers to jobs where performance pay cannot be used. Pure transfers do not represent real costs for an economy as a whole. Nevertheless, net transfers to an individual may well represent welfare changes for others. One distinguishing feature of our model – unlike any other model we know of – is that pure transfers, for example in terms of sickness benefits, unemployment insurance, or tax payments, will matter for a person’s perceived importance and thus his self-image and utility. This is important for several of our conclusions, for example the welfare effects of unemployment. In the model, workers may have a preference for regarding themselves as someone who is important to others. In social psychology, self-image is recognized as an important determinant of human behavior. Aronson et al. (2005, p. 166) summarize the findings as follows: “During the past half-century, social psychologists have discovered that one of the most powerful determinants of human behavior stems from our need to preserve a stable, positive self-image”. While self-image is still not a standard ingredient in economic analysis, a number of economic models involving the concept of self-image or identity have recently been developed (Akerlof and Kranton, 2000, Akerlof and Kranton, 2005, Benabou and Tirole, 2002, Benabou and Tirole, 2003, Benabou and Tirole, 2006, Brekke et al., 2003, Brekke and Nyborg, 2008 and Bruvoll and Nyborg, 2004, among others). While “being important” is hardly the only aspect of self-image that people care about, it is the only aspect we focus on here; note, in particular, that we abstract from any preferences for social approval.3 In the literature on voluntary contributions to public goods, it is common to model preferences either through “pure” or “impure” altruism (Andreoni, 1990). Pure altruism means that one cares about final outcomes in terms of public good supply, but not at all about one’s own effort to provide it (like in Francois, 2007). An impure altruist, on the other hand, derives a private benefit – the warm glow of giving – from his own contribution or effort. Several authors, like Delfgaauw and Dur (2008), assume that work motivation is caused by warm glow, while workers are not assumed to be concerned about the social consequences of their efforts.4 Each of these approaches are somewhat unsatisfactory: as mentioned above, pure altruism cannot generally explain substantial contributions; on the other hand, it does not seem reasonable that an altruist would only care about his own effort, independently of its consequences. The self-image approach presented below reconciles these concerns: A given effort produces the private good of a better self-image, but only to the extent that the effort is thought of as important to others’ welfare. Our aim has been to propose a model of work motivation which is applicable for almost any kind of job, not requiring ad hoc differentiation between types of jobs (physicians versus others, public versus private services, caregiving versus industrial production, etc.). In our model the crucial aspect of “importance” is not linked to specific characteristics of job tasks, enterprise organization, or ownership, as such, but to whether the employee’s efforts yield social welfare effects over and above the compensation she receives.

Above, we have formalized the idea that work is not only motivated by its monetary compensation, but also by a desire to be important to other people. We have shown that with lump-sum taxation, the desire to be important does not change the effort of workers receiving performance pay (bakers); it does, however, change the effort of those whose performance cannot be efficiently monitored and who thus are not paid according to their marginal productivity (nurses). Moreover, it affects behavior in both groups when it comes to the choice (whenever such a choice is available) between working and living off welfare: many individuals, including both bakers and nurses, will prefer to work even if welfare payments provide full income compensation. With the preference structure proposed above, nursing jobs – or rather, jobs where effort is not paid according to its marginal productivity – may be attractive for two very different reasons: they provide the opportunity to shirk, but also the opportunity to be important. Thus, unless nurses’ wage is kept strictly lower than wages in the baking sector (i.e. the sector where effort is paid by its marginal productivity), shirkers will be attracted to the nursing sector. If the wage is kept low enough to keep shirkers out, however, the public sector will be suboptimally small. One way to improve on this situation is to increase nurses’ social productivity through fixed capital investments. The easier it is for nurses to do a good job, the more attractive will the nursing profession be – but only for those who really want to do a good job. This argument provides a rationale for overinvestment in certain types of public good production capital, such as libraries in schools and universities, or diagnostic equipment in hospitals. Our model also provides one possible explanation for the detrimental effects of unemployment. Workers’ self-image loss when losing their job is not necessarily compensated by the decreased cost of effort. In fact, we show that in the case of full income compensation through unemployment benefits, every nurse would become weakly worse off if losing her job; the highly motivated bakers experience a net loss, while only the poorly motivated bakers can gain. The type of work motivation modelled here depends on the individuals’ perception of how important her effort is to others, not whether the firm she works in is publicly or privately owned, whether it is a profit or non-profit firm, or whether the worker herself has strong preferences for the good she produces. The crucial distinction between what we have called the “baking” and “nursing” sectors is that individual effort and/or productivity can be monitored in the former, but not in the latter sector. Interpretations of these two sectors as private and public, or profit and non-profit, are thus only valid to the extent that effort is more easily observable in private than in the public sector, or more so in profit than non-profit firms. Finally, our model of work motivation has been presented in terms of an exogenously fixed, internalized moral norm. Over time, such norms may be eroded or strengthened, depending, for example, on others’ behavior and the relative payoffs of available alternatives (see, for example, Lindbeck et al., 1999). While outside the scope of the present paper, integrating such dynamic aspects with our approach would be an interesting topic for future research.