Why Labor is not to put to work like Capital

Seminar Paper 2010 25 Pages

Business economics - General



1 Introduction
1.1 The Partial Gift Exchange Model
1.2 The Neoclassical Model of the Labor Market

2 Social Preferences
2.1 Sources of Social Preferences

3 A Gift-Exchange Experiment
3.1 “Does Fairness prevent Market Clearing? An Experimental In­vestigation” by Fehr et al. (1993)
3.2 Experimental Design
3.3 Experimental Predictions and Results

4 An amended Version of the latter Gift-Exchange Ex­periment
4.1 “Do incentive contracts undermine voluntary cooperation?” by Fehr et al. (2002)
4.2 Experimental Design
4.3 Experimental Predictions and Results

5 Conclusion
5.1 Literature
5.2 Positive Analysis and Personal View

1 Introduction

A market that needs to be treated differently than other good markets, is the labor market. The selling of labor is different, because working not only pro­vides income, but it is also seen as a status symbol and a source of pride (Solow, 1990). To explain behavior that contradicts with the neoclassical model of the labor market one has to deviate from the so called Homo Oe- conomicus. Hence, this is rather an empirical question[1]. Through the devel­opment of experimental methods it has become possible to compare actions based on rational assumption with actual behavior in a controlled laborato- rian environment. The results might be used to rule out general norms of behavior and therefore help to develope recommendations for policy makers.

As we will observe in the below presented experiments, the majority of the participants does not behave according to concept of being fully rational. That wages affect worker's effort through a partial gift exchange framework will be the focus of this paper. In several experimental investigations it has been found that fairness plays an important role in numerous economic transac­tions, for labor market transactions e.g. Fehr, E. et al. (1993).

The seminar paper is constructed as follows. In the opening section I intro­duce the model by Akerlof (1982) and the neoclassical model. Section 2, will describe the concept of social preferences. In the following two sections the experimental findings of Fehr et al. (1993) (section 3) and Fehr and Gächter (2002) (section 4) will be presented. Section 5 discusses the literature on ex­perimental economics concerning the gift-exchange hypothesis. Additionally I attempt a positive analysis and state my own opinion.

1.1 The Partial Gift Exchange Model

The partial gift-exchange model also known as the fair wage-effort hypothe­sis has been developed by Akerlof (1982). Akerlof (1982) explains the case of involuntary unemployment in the consequence of a wages above the market- clearing level. Reciprocally individuals respond with a higher effort level and a buyer is willing to pay more than the minimum in order to receive a higher quality. Reciprocal fair behavior in fact is only one part of the explanation for market outcome above the neoclassical predictions. Since fairness is hard to define one can at least distinguish between positive and negative reciprocity as features of fair behavior. If individual 1 is willing to take costly actions into account in order to pay back at least a part of the kindness he has received from individual 2, is denoted as positive reciprocity. In contrast, negative reci­procity leads to a costly action by person 1 with the aim of reducing individual 2’s payoff. One cannot deny that fairness considerations have an impact on la­bor market principal-agent relationships. Akerlof (1982) considered fairness as an important factor in the interaction of firms and workers. He constructed a formal relationship that represents the perceived fair wage of an individual i at time [illustration not visible in this excerpt]. The work rules are considered as e, unemployment benefit is represented by b and the unemployment stock by u.

Akerlof’s (1982) model is presented below.

illustration not visible in this excerpt


[illustration not visible in this excerpt] is the perceived fair wage of individual i at t+1

[illustration not visible in this excerpt] is the actual wage of individual i in previous period(s)

ω0 is the wage paid of others in the individual’s reference set in current and previous periods

bu is unemployment benefits of individuals in the reference set in current and previous periods

u is the number of unemployed in the reference set in current and previous periods

ei is the individual’s work rules in current and previous periods[2]

e0 the work rules of persons in the individual’s reference set in current and previous periods

These factors influence the concept of the worker of what is a fair wage given the current situation. As an important aspect of deriving a value of the fair wage, not only individual’s current and past experiences are taken into ac­count, but also past and current aspects of individual’s reference group. This leads us to the individual reference group theory, where it depends on how other workers in a similar situation are treated, whether a wage is consid­ered as fair to the worker. Various experiments about the gift exchange model are designed in a way that these group effects do not appear, as we will see later. Although these effects substantially influence the fairness concept. Be­sides the reference group consideration, also the actual unemployment stock u as well as the opportunity cost of employment b do have an impact on the fair wage concept. The model by Akerlof (1982) states that there is a benefi­cial exchange between the firm and the worker. This type of exchange is called gift exchange. As a responds to a higher wage of the firm the worker exceeds his effort above minimum required, which is an equilibrium outcome that is higher than the market clearing outcome. Akerlof (1982) defines a primary labor market and a secondary labor market. In the secondary market the market clears. The primary labor market results in a market outcome above the neoclassical one, where involuntary unemployment occurs and the market does not clear. According to Akerlof (1982, p.568), “the gift-exchange economy and the neoclassical economy differ in at least one fundamental respect.”

1.2 The Neoclassical Model of the Labor Market

In the neoclassical model of the labor market, individuals are assumed to be purely selfish and utility maximizer. In order that also firms are strict profit maximizer, they pay the minimum wage. Therefore the worker provides effort at the minimum effort level just as much as necessary to maintain employ­ment. This framework results in market clearing. Worker maximize their utility including the wage paid and the effort U = (ω,β)[3] where e stands for the effort of the worker and ω reflects the wage. When the firm sets a wage under the market-clearing level, the firm receives no labor at all. As far as the firm is profit maximizing the choice of the wage level must be equal to the market-clearing wage [illustration not visible in this excerpt], since paying more does not lead to a beneficial out­come for the firm. This equilibrium stands in contrast to the gift-exchange set up, where it is somehow beneficial to the firm to pay higher wages [illustration not visible in this excerpt] to the worker. And the firm takes into consideration that paying above the min­imum wage level leads to an increase in worker’s effort [illustration not visible in this excerpt]. Therefore the neoclassical objective functions need to be modified, according to match the empirical observations made by a number of experimental economists in the recent 20 years.

2 Social Preferences

2.1 Sources of Social Preferences

Social preferences indicate that an individual does not only care about its own payoff but also includes other relevant individual’s payoff in their fair­ness concept. In experimental studies it has been shown that the assumption of pure selfish individuals does not hold for all individuals[1]. The empirical results of studying gift-exchange experiments provide evidence that fairness considerations influences the behavior of individuals substantially. If one takes preferences for fair behavior in to account, it might explain a part of the deviation from the neoclassical model. In the labor market reciprocally fair behavior leads to an outcome which exceeds the competitive one. Firms are willing to pay higher wages in order to get back an effort that exceeds the minimum requirement. Likewise are firms not willing to reduce wages, since they anticipate that workers cut their level of effort and judge the firm’s action as not fair[2]. From the firm’s perspective only a motivated worker can be profitable, even more profitable if the effort level is raised. This can be achieved by forming the right incentives, with the wage as an important in­strument. Individuals also judge wage cuts differently, it highly depends on the reasons why the pay is cut[3]. Gift-exchange can be either seen as a durable action or as an one time bonus payment to the worker[4]. It is obviously, that fairness is hard to define, since every individual has its own definition of what is perceived as fair. Akerlof (1982) points out that fairness is due to an indi­vidual reference set. So, it seems to be crucial for the worker how others are treated in a comparable situation. Another source of developing social pref­erences is inequity aversion. Inequity aversion means, that individuals do not only care about their own payoff, but rather on how the payoff is finally distributed. Dufwenberg and Kirchsteiger (2000) however challenge the ex­perimental results on inequity aversion in their paper “Reciprocity and Wage Undercutting” and question that in principal-agent relationships the payoffs are actually comparable. In the end the judgment of generous or hostile ac­tions remains a private information, this holds for both market sides. Ex­perimental economics tries to isolate the different motivational sources for reciprocal fair behavior, as for instance the two following examples.


[1] See Falk 2001

[2] See Akerlof (1982)

[3] Akerlof (1982)

[1] See for instance Fehr et al. 1993, 1998

[2] See Falk et al. (2003) “On the Nature of Fair Behavior”

[3] See Kahneman et al. (1986)

[4] See Bellemare et al. (2007) “Gift Exchange within a Firm: Evidence from a Field Experi­ment”


ISBN (eBook)
ISBN (Book)
File size
525 KB
Catalog Number
Institution / College
Humboldt-University of Berlin – Wirtschaftstheorie II (Mikro)
Labor Capital



Title: Why Labor is not to put to work like Capital