Research: Union safety effect

By Erik Henningsen 

A few weeks ago, a friend in the labour movement pointed me to this new article about the impact of unions on workplace injuries. “Does ‘right to work’ imperil the right to health? The effect of labour unions on workplace fatalities” examines how legislative changes that impede unionization affects levels of injury and fatality in those workplaces.

The research on the union safety effect is mixed. Some of it supports the notion that unions results in lower levels of injury and fatalities. Some of it comes to the opposite conclusion.

There are lots of methodological challenges in doing these kinds of studies that help explain the mixed results. Comparing unionized and non-unionized workplaces, for example, sometimes suggests that unionized workplaces have higher levels of injury and fatalities. Critics of unions often use such data to conclude that there is no union safety effect. There are, of course, good potential explanations for these results.

For example, workers may be more likely to unionize in dangerous workplaces (spurred on by the danger) so there may be an apples-to-oranges issue with the comparison. The presence of a union may also embolden workers to report injuries, thereby creating the appearance of more injuries when, in fact, it is just a matter of more reported injuries.

This new study examines fatality levels in US states that have implemented so-called right to work laws. These laws weaken unions by allowing workers to opt out of paying union dues (thereby creating a free rider problem). The question is essentially, what effect do right-to-work laws have on workplace health.

The upshot is that a 1% decline in unionization (attributable to right-to-work laws) means about a 5% increase in occupational fatalities. Overall, RTW laws have resulted in a 14.2% increase in workplace mortality (this is bad). It also suggests that unionization reduces occupational fatalities (these being the most serious forms of injury and also the most likely injuries to be reported). The author discusses his study a bit more on his blog, which is an interesting read.

-- Bob Barnetson

-- Bob Barnetson