Skip to content

The Effect of Collective Bargaining Legislation on Strikes and Wages

Literature › Works

Peter Cramton; Morley Gunderson; Joseph Tracy (1999). The Effect of Collective Bargaining Legislation on Strikes and Wages. The Review of Economics and Statistics, 81(3). MIT Press.
Type journal article
Scope Comparative and methods
Method econometric
Period studied 1967–1993
Regions Canada, Canadian provinces, Federal jurisdiction
Themes comparative theory, unions strikes, labour markets wages, methods data sources
Citation key cramton1999

Argument

The paper is not about South Africa. It tests how Canadian provincial and federal collective-bargaining laws affect strike incidence, strike duration and wage settlements. It uses a strategic bargaining model with private information about the firm's willingness to pay, in which unions choose between a strike threat and a holdout threat. It finds that compulsory conciliation and cooling-off periods do little to reduce strike costs, and that general reopener provisions and mandatory strike votes lower negotiation costs. Bans on replacement workers raise strike frequency, strike duration and real wages, which shifts quasi-rents from employers to unions. A welfare calculation puts the cost to employers of a replacement ban at about $4.8 million per contract.

Evidence and method

All private-sector Canadian contract negotiations covering 500 or more workers, January 1967 to March 1993 (4,340 contracts in the main sample). Sources are Labour Canada's Major Wage Settlements and Work Stoppage databases, Statistics Canada CANSIM macro series, and policy variables coded from statutes by month and province (Gunderson et al., Budd, CCH Canadian Labour Law Reporter). The 888 strikes in the wage file were linked to the work-stoppage file.

Findings

  • Headline finding on replacement bans: higher negotiation costs and a transfer of rents from employers to unions. (p. 2)

    Legislation banning the use of replacement workers appears to lead to significantly higher negotiation costs and redistribution of quasi-rents from employers to unions.

  • Conciliation policies are largely ineffective in reducing strike costs. (p. 2)

    The data indicate that conciliation policies have largely been ineffective in reducing strike costs.

  • Estimated average cost of a replacement ban to employers. (p. 2)

    The estimated average cost to employers from a replacement ban is around $4.8 million (1993 Canadian) per contract.

  • Mandatory strike votes lower strike incidence, which suggests some strikes start without majority support from members. (p. 5)

    Mandatory strike votes are associated with a reduction of over seven percentage points in the likelihood of a strike.

  • General reopener provisions lower strike incidence. (p. 6)

    General reopener provisions are associated with a statistically significant drop of six percentage points in strike incidence.

  • Replacement bans lengthen strikes substantially, with the sample of all strikes giving the more plausible estimate. (p. 7)

    In the sample of all strikes, a replacement ban is associated with a 50% increase in the average strike duration.

  • Conciliation is associated with longer strikes, the opposite of its intent. (p. 7)

    Conciliation in general is associated with longer strike durations, although the effects are imprecisely measured.

  • Replacement bans raise real wage settlements, with a larger long-run effect. (p. 9)

    the ban on replacement workers-is associated with slightly more than a 4% increase in real wages. The long-run effect of a replacement ban is around 14%.

  • When strike variables are instrumented, wage settlements fall with strike length. (p. 10)

    we find a positive average wage effect and a negative and significant slope to the concession function, which corresponds to roughly an 8% decline in wages per 100 days of strike.

  • Budd's null result for replacement bans on wages is sample-dependent. (p. 11)

    Budd's finding of no effect of replacement bans on wage settlements is sensitive to the shorter time period and narrower industry mix of bargaining units in his sample.

  • Unions gain and employers lose from replacement bans. (p. 13)

    Unions receive sufficiently higher wage settlements to offset their share of these higher negotiation costs, while employers are made significantly worse off.