The Impact of Unionization Structures with Heterogeneous Firms and Rent‐Sharing Motives

• DOI: http://doi.org/10.1111/sjoe.12268
• Author: Marco Pinto
• Published date: 01 January 2019
• Date: 01 January 2019

Scand. J. of Economics 121(1), 298–325, 2019

DOI: 10.1111/sjoe.12268

The Impact of Unionization Structures with

Heterogeneous Firms and Rent-Sharing

Motives*

Marco de Pinto

IAAEU,Trier University, DE-54296 Trier, Germany

depinto@iaaeu.de

Abstract

How are unemployment and output affected if wagesare set on the sector level rather than firm

level? We take a new look at this question, allowing for heterogeneous firms and rent-sharing

motives.Without these motives, employmentand output are lower under sector-level wage-setting

due to higher wage markups. With rent-sharing motives,however, firm selection is higher under

sector-levelwage-setting, which tends to increase employment and output, thus counteracting the

markup effect. Simulations show that the firm-selection effect decreases the difference between

the two unionization structures substantially but it does not change the signs of the effects on

output and employment.

Keywords: Aggregate output; heterogeneous firms; rent-sharing motives; unemployment;

unionization structures

JEL classification:J3; J6; J51

I. Introduction

Despite decreasing union membership, trade unions are still one of the

predominant labor market institutions in Europe, and they have a significant

influence on labor market outcomes.1However, there are structural

differences with respect to the level at which bargaining takes place. As

illustrated in Figure 1, bargaining at the sectoral (or central) level is

mostly observed in central/north/south Europe, while firm-level bargaining

is predominant in west/east Europe. These institutional differences are

featured in the public (and academic) discussion about how to reduce

unemployment. One prevalent suggestion is that unemployment in countries

*I gratefully acknowledge helpful comments from Laszlo Goerke, J¨org Lingens, Mario Mechtel,

Jochen Michaelis, and Ulrich Zierahn, as well as from participants at research seminars in

M¨unster and Mannheim. I also thank two anonymous referees; their input has substantially

improved my paper.

1Visser (2015) reports (adjusted) union coverage rates for European countries. In 2013, the

average coveragerate in Europe was still about 61 percent.

C

The editors of The Scandinavian Journal of Economics 2017.

M. de Pinto 299

Fig. 1. Bargaining level in selected countries, 2014

Notes: Bargaining levels 1 and 2 indicate that bargaining takes place at the local/firm level or that bargaining is

alternating between sector and firm levels, respectively.Bargaining levels 3 and 4 indicate that bargaining takes place

at the sector level or that bargaining is alternating between central and sector levels,respectively. Bargaining level 5

means that bargaining takes place at the central (or cross-sector) level.European countries are categorized according

to van Klaveren and Schulten (2015), with abbreviationsdenoting: AU, Austria; BE, Belgium; CZ, Czech Republic;

DE, Germany; DK, Denmark; EL, Greece; ES, Spain; FI, Finland; FR, France; IE, Ireland; IT, Italy; JAP, Japan;

NL, the Netherlands; NO, Norway;PL, Poland; PT, Portugal; SE, Sweden; SK, Slovakia;UK, United Kingdom; US,

United States.

Source:Visser (2015), ownillustration.

with a sector-level wage-setting structure could be reduced if bargaining

were to be decentralized.

This argument is based on the highly influential paper by Calmfors and

Driffill (1988). In their model, sector-level trade unions set higher wages

compared to firm-level trade unions by implementing a larger markup over

the alternative income. They can do so because competition across sectors is

less intense than competition within sectors, such that the demand reduction

in response to a wage increase is relatively low. This markup effect causes

labor demand to decline, implying that aggregate employment and output

are lower under sector-level trade unions. However, this result is derived

under the assumptions that firms are homogeneous and that their number

per sector is fixed. In addition, an individual’s utility is only based on

absolute income terms, neglecting any other determinants of utility.

C

The editors of The Scandinavian Journal of Economics 2017.