ENS, CNRS, CMH, TU-Delft
Post-REPONSE analysis
Project's goal:
> Explore the position & contours of the company in collective bargaining
> Analyse configurations of collective bargaining in French companies
Using mixed-methods: modelling & monographies
Results:
Didry C. et al. (2021), Entreprises en Négociations, report for the French Ministry of Labour https://dares.travail-emploi.gouv.fr/publication/entreprises-en-negociations
- Distinction local/central in the literature
Livernash, E. R. (1963). “The relation of power to the structure and process of collective bargaining”. The Journal of Law and Economics, 6, 10-40.
"Management has its strongest power position when it has a reasonably large number of plants, geographically dispersed, producing on a nonintegrated basis, with plant bargaining by different unions with different expiration dates. But any form of plant bargaining, so long as production in one plant is not dependent on production in other plants and provided contract expiration dates do not coincide, gives management a strong bargaining position.” p. 24-25
- Distinction local/central in the literature
Kinnie, N. (1982), "Bargaining Structures in Multi‐Plant Companies." Employee Relations, 4(1), 23-26.
> Local bargaining gives more flexibility
Block, R., & Berg, P. (2009). “Joint Responsibility Unionism: A Multi-Plant Model of Collective Bargaining under Employment Security”. ILR Review, 63(1), 60-81
> Local cooperation between unions and management can help the firm increase profitability rather than wages, to maximize employment
Zagelmeyer, S. (2005). The employer's perspective on collective bargaining centralization: an analytical framework. Int. J of HR Managmnt, 16(9), 1623-39.
> Geographical concentration of plants favors centralized bargaining because of high comparability of working conditions in a homogenous labor market
- Distinction local/central in the literature
- Lacking data to validate systematic claims
Swidinsky, R. (1981). “The Effect of Bargaining Structure on Negotiated Wage Settlements”. Relations industrielles / Industrial Relations, 36 (2), 371–385.
2300 agreements in the Canadian private sector
Compares single-plant and multi-plant firms on wages settlements
> does not consider the level of bargaining within the firm
- Productive
- Geographical
Within the firm = locational choices for plants & HQs
- with CLAP/FLORES (France), BSD (UK), AES (US) etc.
Aarland, K., Davis, J. C., Henderson, J. V., & Ono, Y. (2007). “Spatial organization of firms: The decision to split production and administration”. The RAND Journal of Economics, 38(2), 480-494.
Henderson, J. V., & Ono, Y. (2008). Where do manufacturing firms locate their headquarters?. Journal of Urban Economics, 63(2), 431-450.
> depends on industry, size, geographical dispersion
- Productive
- Geographical
Within the group = locational choices for firms (FDI, mergers and acquisition)
- ORBIS (world)
Bohan, C., & Gautier, B. (2013). Multilevel analysis of Corporations Networks: A comparison between agro-food and automobile strategies for Urban Development. In Methods for Multilevel Analysis and Visualisation of Geographical Networks (pp. 155-176). Springer, Dordrecht.
- Productive
- Geographical
Within the firm & group:
- ORBIS (world)
- LIFI (France)
- CLAP (France)
Finance, O. (2016). Les villes françaises investies par les firmes transnationales étrangères: des réseaux d'entreprises aux établissements localisés (Doctoral dissertation, Paris 1)
> Size and regional effect of cities receiving FDI through multinationals
Sunk cost of bargaining at the establishment level c>0. At the firm level, the employer enjoys increasing returns in her bargaining technology: the cost c(N) is increasing but concave in the number of covered establishments, and 0 < c (2) < 2c.
Our model suggests three testable hypotheses:
H1: The geographic dispersion of establishments favours negotiation at central/company level.
H2: The heterogeneity of size/activity of establishments favours negotiation at the establishment level.
H3: The number of establishments belonging to the firm is positively related to a firm level bargaining.
There can also me mixed strategies for the employer:
- conducting bargaining at both central and local level
- pooling some establishments into a single negotiation but not all, especially the ones that are far from HQs, to make it hard for employee representatives to coordinate their actions and mobilize workers.
This leads to:
H4: The distance of a particular establishment from its head office may favour its inclusion in a multi-establishment bargaining.
Despite reputation for industrial dispute,
of companies with 10+ employees on given year in the 2010s:
<2% experienced a strike action
15% conducted collective bargaining
- of which 80% reached an agreement
> Only recognised unions can bargain with the employer
In establishments with 50+ employees: a union is recognised when it attracted >10% votes at last professional election
> When unions are present, employers have to open, each year, collective bargaining on legal list of topics:
...except if agreement on given topic in past 4 years still valid. NB: reaching an agreement is not mandatory! >30% votes
- wages
- profit sharing schemes
- working time
- gender equality
- quality of work life
(Until Summer 2016)
Level of bargaining: local vs company
= discretionary choice of the employer in multi-establishment firms.
New reforms during summer 2016,
...do not affect our data:
REPONSE 2017 based on 2014-2016 rounds of bargaining
REPONSE 2017 (DARES)
Ministry of Labor’s French Workplace Employment Relations Survey, equivalent of WERS (UK).
Every 6 years, mandatory questionnaire + face-to-face interviews
Sample stratified by employment size and industries, with sampling probability proportional to the size.
= 4 364 establishments with 10+ employees (non-agricultural)
Accessible through secure remove environment (CASD)
REPONSE 2017 (DARES)
Variables of interest: Wage bargaining (WB).
"Has there been bargaining on Salaries, bonuses and other allowances between 2014 and 2016 ?
- “No”
- “Yes only at the establishment level”
- “Yes at the company (or UES level)”
- “Yes both at the establishment and company (or UES levels)”
We also compare bargaining on gender equality (GB) and on working time (TB)
FLORES 2017 (INSEE)
File of local salaried employment and rewards
= core source for local business stats for France (non-military).
Key variables at establishment level:
> Almost complete map of companies and establishments operating in France
- parent company
- creation date
- industry code
- number of employees
- wages
- municipality of location
LIFI 2017 (INSEE)
Financial Links dataset
= covers French companies and financial links of subsidiarity
Key variables at company level:
- ultimate group owner
- municipality of ultimate group owner
REPONSE - FLORES - LIFI 2017
Merging data within secure remote environment
REPONSE - FLORES
Establishment level SIRET
LIFI - FLORES
Company level SIREN
FLORES - GEOFLA
Municipality level DCCOM
Geometries + Delineation of urban areas
To test hypotheses H1 to H4
Using structural characteristics of the establishment, company level characteristics, and co-location info with HQs.
4 364 establishments from REPONSE
- single-establishment companies
- establishments part of UES
- complete data (no NA)
= 1997 Observations in Core sample
Multinomial model
Multinomial model
Multinomial model
Multinomial model
Multinomial model
Multinomial model
WAGE BARGAINING
in 61% establishments (REPONSE)
in 73% establishments (Core sample)
of these:
- 11% at establishment level
- 67% at company level
- 22% at both levels simultaneously
Industry, size & age = significant
R2 = 40.5%
Presence of union representative & share of full-time workers significant
Industry, size & age = significant
R2 = 24.5%
H1
H2
H3
H4
GENDER EQUALITY BARGAINING
in 54% establishments (REPONSE)
in 65% establishments (Core sample)
of these:
- 10% at establishment level
- 70% at company level
- 20% at both levels simultaneously
Industry, size & age = significant
R2 = 48.9%
H1
H2
H3
H4
WORKING TIME BARGAINING
in 36% establishments (REPONSE)
in 44% establishments (Core sample)
of these:
- 17% at establishment level
- 61% at company level
- 22% at both levels simultaneously
Industry, size & age = significant
R2 = 29.2%
H1
H2
H3
H4
Europe converges towards the US scheme:
despite wide variety of structures, most collective bargaining occurs at either the company or workplace level.
Trade-off between bargaining close to workplace characteristics and limiting cost of bargaining (increases with n) by negotiating at the central level.
Empirically: geographical dispersion, size and productive organisation affect level of collective bargaining on major topics.
philippe.askenazy@ens.psl.eu
c.cottineau@tudelft.nl