A new study by the Institute for Work & Health (IWH) replicates an earlier finding of a “union safety effect”—in which unionized companies have lower lost-time injury rates than non-unionized firms—in Ontario’s industrial, commercial and institutional (ICI) construction sector.
Using workers’ compensation claims data from 2012 to 2018, the study found unionized ICI construction companies had a 25-per-cent lower rate of lost-time injuries than non-unionized ones.
Sponsored by the Ontario Construction Secretariat (OCS), the study also found a 23-per-cent lower rate of musculoskeletal injuries and a 16-per-cent lower rate of critical or severe injuries in unionized companies, compared to non-unionized counterparts.
These findings are similar to those in the original study, also funded by OCS and published by IWH in 2015. OCS is a joint management-labour not-for-profit organization, created under the Ontario Labour Relations Act to represent the interests of the unionized ICI sector.
The earlier study, using claims data from 2006 to 2012, found lost-time claim rates were 14 per cent lower in unionized companies. However, unlike the 2015 finding that unionized companies had higher rates of no-lost-time claims, the latest study found no statistically significant difference between unionized and non-unionized firms in such claims.
Results were shared in a recent IWH Speaker Series webinar presented by IWH Scientist Dr. Lynda Robson, who co-led the study with IWH Senior Scientist and President Dr. Cam Mustard. A full report on the findings is also available on the IWH website.
This study was not designed to examine reasons for the observed union safety effect and, therefore, cannot be definitive about what those reasons are, says Robson. That said, the results are consistent with the focus of many unions on controlling occupational health and safety (OHS) risks at worksites, she adds.
We would expect this from what we know of the role of unions in workplaces. Their actions include putting OHS in collective agreements, delivering worker training, sharing OHS information, and participating in joint health and safety and worker trades committees, she says.
It may also be that the presence of a union empowers workers to report unsafe conditions and refuse unsafe work.
The follow-up study also indicates that the union effect on lost-time injury rates was greatest among the largest companies. Companies with 50 or more full-time equivalent employees (FTEs) had a 44-per-cent lower rate of lost-time claims, while companies with 20 to 49 FTEs and five to 19 FTEs had reductions of 24 per cent and 25 per cent, respectively. Notably, no union effect was found among the very small companies—i.e. those with fewer than five FTEs.
The absence of a union safety effect in companies with fewer than five workers needs exploring in future research, Robson says.
It appears that, in this sub-group of companies, factors related to small company size override any union safety effect. These factors could be a lack of OHS expertise, younger company age or lower average job tenure. As well, she notes, Ontario’s OHS regulatory requirements for these very small companies differ from those for larger companies.
The team also examined the union safety effect across seven different types of ICI construction work (the seven sub-sector categories with large enough samples to allow for analysis) and found that it varied. The largest union safety effects were found in industrial maintenance and repair contracting (where unionized firms had 72 per cent lower lost-time injury claims rates than non-union firms) and millwright and rigging work (where unionized firms had 67 per cent lower lost-time injury rates). The smaller union safety effects were found in electrical work (25 per cent lower) and excavating and grading (31 per cent lower).
How the study was done
The study sample drew on 24 lists of unionized contractors in 39 ICI construction sub-sectors. The lists, provided by unions and employer associations, represent a near-comprehensive picture of unionization in this sector, thus making comparison between unionized and non-unionized firms possible. For data on injury rates, the team used records obtained from the Workplace Safety and Insurance Board (WSIB).
Once the team completed the painstaking work of finding companies from the unionized lists in the WSIB dataset, it had a sample of 58,837 companies, operating 60,425 lines of business found in the ICI sector. Of the 60,425, 5,267 (8.7 per cent) were classified as unionized and 55,158 (91.3 per cent) as non-unionized. In terms of size of workforce, unionized companies in the sample employed 773,000 FTEs (44.6 per cent) and non-unionized companies employed 958,000 FTEs (55.4 per cent) during the period 2012-2018.
The research team’s findings of an association between unionization and lower injury rates take into account a number of factors that could also affect rates outside of union status. That is, the team conducted statistical analysis to disentangle the independent effect of unionization from potentially confounding factors such as firm size, firm complexity (e.g. number of lines of business), construction sub-sector type and location of firm.
The study did not examine work-related fatality rates. The number of fatalities was too small to allow the type of statistical analysis used in this study (e.g. to control for firm size and sector) to examine the effects of unionization.