Abstract
Chapter 3 and 4 have shown that the response of wages to exogenous changes in the environment is crucial for the economy’s reaction. One of the most striking differences of the labour markets in various countries are the wage setting procedures. The preceding chapters presumed that wages are determined by negotiations between workers and firms. This assumption may be plausible for the United States. In most European countries, this assumption may be justified for the upper income classes. However, the influence of collective bargaining in most European countries is remarkable for middle and lower income classes.1 Consequently, the model presented in chapter 3 can only serve as a reference point for these countries.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
See e. g. Layard, Nickell, and Jackman (1992, p. 517-524) for the coverage of collectively negotiated wages in various countries.
There are two measures to estimate the unions’ influence: the density and the coverage. The union density counts the union members in total employment. However, the union density frequently underestimates the influence of unions as collectively negotiated wages are often paid to nonunion members. The coverage is the percentage of labour contracts rewarded by collectively bargained wages. Union density and coverage can differ extremely, especially in European countries. In France e. g. the density was only 28 percent in 1979 whereas the coverage exceeded 75 percent (Layard, Nickell, and Jackman, 1992, p. 88, 517).
See also Keuchel (1989), Berthold and Fehn (1996), Franz (1996), or Schürfeld (1998) for an overview of the German institutional setting.
Firm-level negotiations where unions, which bargain directly with firms over wages, can thus be subsumed under decentralised bargaining.
The reduction in employment is often inverse to seniority. In this situation the medium worker is protected against moderate dismissals. The unions’ objective function would then pay less or no attention to the utility of an unemployed person. See e.g. Pencavel (1991) for an overview.
See e. g. Lindbeck and Snower (1988) for the insider-outsider theory.
In Germany, the Federal Ministry of Labour has the possibility of declaring collective contracts to be compulsory for the whole industry (Franz, 1996). Berthold and Fehn (1996) mention that less than 10 percent of the collective contracts are actually declared compulsory, but that the Federal Ministry of Labour especially used this instrument in the service sector. In addition, a law has been enforced in 1996 saying that collectively negotiated wages which were declared compulsory in the building and constructing industry also apply to foreign employees of foreign firms currently working in Germany (Sachverständigenrat zur Begutachtung der gesamtwirtschaftlichen Entwicklung, 1996, p. 113). Both examples suggest that underbidding wages is difficult in Germany.
Examples are reported for a number of countries, e.g. the United States, the United Kingdom (Layard, Nickell, and Jackman, 1992, p.91), and Germany. A more recent case in Germany involves the firm-level contract between the union IG-Metall and the Volkswagen AG concluded in 1993 (Sachverständigenrat zur Begutachtung der gesamtwirtschaftlichen Entwicklung, 1994, p. 107). To prevent collective dismissals of about 30000 employees, both partners agreed on a contract specifying e.g. a reduction in weekly working time to 28.8 hours (4 days). At the same time, the wages were reduced by 20 percent. Operational dismissals were excluded for the duration of the contract.
See e. g. Espinosa and Rhee (1989) for this argument. See also Layard and Nickell (1989).
As organised workers normally have to pay some membership fees, the union can principally grant them a strike compensation. For the sake of simplicity the compensation is set equal to the unemployment benefits.
McDonald and Solow (1981) derive the same objective function. The authors interpret this function as the ”aggregated gain from employment, over and above the income” b every member starts with (McDonald and Solow, 1981, p. 898).
The union’s negotiations with the employers’ association in sector X is completely independent of the bargaining in sector Y. For this reason, there is no signaling character between the bargaining processes. In addition, the unions’ value function is not stochastic. Therefore, the unions’ discounted utility stream does not depend on an individual’s employment position. It implies that the individuals are fully substitutable from the unions’ point a view; it does not matter which particular person is employed or jobless.
See e. g. Davis (1993, p. 70, 92) or Rishel (1990) to derive the Bellman equation.
See e.g. Kamien and Schwartz (1991, p. 137f) or Sargent (1987) for a more formal derivation of this result.
The condition stems from equation (5.7). As in an equilibrium the time derivatives equal zero, the value function is positive if the profits are positive which directly leads to the desired condition.
See e.g. Appendix 5.6.1 for the derivation of the wage equation.
Binmore, Rubinstein, and Wolinsky (1986) show that in static problems the solution of a Nash bargaining process can also be derived from a sequential bargaining process. The validity of this result in dynamic problems is demonstrated by Coles and Wright (1998) under the additional restriction that the objective functions of both partners are linear and that the discount rates are identical.
If the wage equals the unemployment benefits an individual is indifferent to working or being unemployed in an equilibrium. Similar to the preceding chapters, it is assumed that persons prefer to work in these situations.
In contrast to a state of equilibrium, individuals are prepared to work for lower wages than the unemployment benefits during adjustment processes. Consider the following situation. The unemployment level is high so that the probability to quickly find another job is low. Suppose in addition, that the risk of becoming unemployed is low once the person is employed. However, the job seeker expects the wages to rise soon. Under these circumstances, waiting for a job offer that is rewarded by a wage higher than the unemployment benefits may yield the lower utility stream so that the individual accepts the job. Consequently, knowing this the union may in the very short run consent to lower wages than the unemployment benefits.
The firm’s problem is to choose the number of vacant jobs so that the discounted stream of profits is maximised. For the maximisation problem, the firms anticipate the wage agreement between the trade union and their own representatives. From the firm’s point of view, the difference lies in anticipating other wages, but the problem itself is unchanged.
In contrast to the model of chapter 4, no off-steady-state comparisons are possible here. The fact that the equation determining the evolution of the shadow price λ depends on LY is responsible for this characteristic. The dynamic system of equations can be modified so that the first derivatives with respect to time can be isolated on the left-hand side of the dynamic system of equations. However, the equation determining the evolution of labour will have a different form to the one in chapter 3 so that the criterion (4.14) cannot be used.
The export ratio is measured as the ratio of the total exports to the gross national product.
See Freytag, Meier, and Weiß (1998, p. 17, Figure B.2). The considered group of South East Asian countries consists of Indonesia, South Korea, Singapore and Thailand.
See Freytag, Meier, and Weiß (1998, p. 20, Table B.l).
The comparative advantage of developing countries may e. g. arise due to an abundance of low skilled workers relative to highly skilled labour compared to the industrialised countries. The differences in the relative factor endowments predict that the wages for low skilled employees are lower in the developing countries than in the industrial ones in a state of autarky. For this reason, the developing countries may produce goods using the factor unskilled labour relatively intensively at lower costs. See e.g. Markusen et al. (1995).
Similar to the models presented in chapter 3 and 4, the decline of one product price cannot lead to an outward shift of the Beveridge curve.
Equivalent figures as 3.1 and 3.2 can be drawn for the present model. Clearly, the curves corresponding to the condition for efficient production DY and the labour market equilibrium condition LMY under industry-level bargaining would have a different location. However the shape would be the same.
To derive the results presented in the Appendix 5.6.2, it was assumed that Fi(Li) is homogeneous of degree ai 0 < ai< 1. This is equivalent to presuming a Cobb-Douglas production function Fi(Ki,Li), where the capital K is fixed within the considered time horizon.
One may argue, that a bargaining power equal to the elasticity of labour ai would render a similar result: the unemployment rate reaches a maximum level, but it is independent of exogenous price changes. In the present model this possibility arises only under certain circumstances. Suppose, the firms in sector Y use more labour in the production than the ones in sector X, i.e. aY>ax.The the marginal profit stays indeed constant for small changes in the employment level of sector Y if α = aY. However, since aY >ax, the economy would have specialised on the production of good Y. Suppose next, that the sector X firms use more labour and production than the sector Y firms, i. e. ax >aY. Clearly, the marginal profit is invariant to changes in the sectoral employment if α = aY and the economy continues to produce both goods. If a decrease in the price pY is now considered the subscripts indicating the sector can be exchanged. It then becomes apparent that the unemployment rate will rise as in the new situation the marginal profit is not invariant to changes in the sectoral employment level.
See e. g. Glismann et al. (1986) for a description of the European Monetary System.
See e. g. Franz (1996) for Germany or Layard, Nickell, and Jackman (1992) for the other European countries.
Burda and Wyplosz (25) find a probability of 0.081 for Germany. Mortensen (91) used a value of 0.067 for the United States.
Layard, Nickell, and Jackman (75) use the ratio of the union’s (individual’s) and the firm’s time preference rate as the union’s bargaining power. In Binmore, Rubinstein, and Wolinsky (17), the bargaining power α is determined as λ2/λ1 +λ2, where λ2 (λ1) is the union’s (firm’s) belief that a breakdown of the negotiation occurs. Finally, Coles and Wright (28) define the unions bargaining power α as r2π2/(r1π1 + r2π2, where r2 (r1) is the union’s (firm’s) time preference rate and π2 (π2) is the probability that the union (firm) makes the next wage offer. Every definition for the union’s bargaining power ß relies on data which is not observable so that ß was chosen arbitrarily.
If the labour coefficients are chosen at ay = 0.6 and ax = 0.7 the unemployment rate turns out to be smaller in an collective bargaining than in an individual wage setting.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Weiß, P. (2001). An Open Economy with Industry-Level Bargaining. In: Unemployment in Open Economies. Lecture Notes in Economics and Mathematical Systems, vol 496. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-56569-4_5
Download citation
DOI: https://doi.org/10.1007/978-3-642-56569-4_5
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41161-1
Online ISBN: 978-3-642-56569-4
eBook Packages: Springer Book Archive