Updated: Jul 7, 2021
Despite improvements in construction processes, procedures, and safety equipment, fatalities continue to increase year over year. Risk compensation may be to blame; is it possible that by making sites safer we are in turn increasing risky behavior?
The next time you hold a major safety meeting on your construction project, I want you to add an additional topic to the agenda. Don't tell anyone you're going to do it; don't provide any prompting. I want you to ask everyone in attendance at the meeting if they can recall the topic of the previous day's toolbox talk. Odds are, most people in the room won't be able to.
A colleague of mine was working on a site in a remote area. It was January in Northern Alberta. If you aren't familiar with the climate, it was cold to say the least. Two supervisors weren't able to start their pickup truck. They gathered booster cables and walked out to boost the battery. After opening the hood, a lynx, who had curled up inside the engine to keep warm, jumped out at the pair. Don't worry, the lynx and the two men were uninjured.
This type of occurrence, while uncommon, does happen from time to time in cold climates. As such, it became the topic of the morning toolbox talk the next day. Now, if you were paying attention in the toolbox meeting, this would be a pretty memorable story. A lynx jumping out from under the hood of a pickup truck doesn't happen every day. Add a bit of theatrics to the explanation and it would certainly be a deviation from the standard topics of hand safety and situational awareness.
Interestingly, when this colleague stood up at the safety meeting and asked if anyone could recall the topic of yesterday's toolbox talk, no one raised their hand. Out of 120+ attendees, not one person thought the lynx story so memorable that they retained it in memory for longer than 24 hours. If teams can't remember the most memorable of toolbox talks, what's really going on at a practice level on our projects?
Construction Fatalities Aren't Decreasing
Despite our best efforts to improve health and safety performance on construction projects, the number of industry fatalities isn't decreasing proportionally. In the United States, 971 workers were killed on the job in 2017, a decrease of only 20 workers from 991 in 2016. These past years we watched as new technologies, better processes, and more comprehensive training were rolled out across the nation; the number of fatalities recorded proves contrary to predictions of broad safety improvements that should be realized.
If you analyze the number of fatalities by year, over the past decade, it's obvious that the numbers haven't been decreasing over time; in fact, they're increasing. The industry recorded the lowest number of fatalities, within the past decade, in 2011; the number of deaths has steadily increased each year after, with the exception of 2017 where the number dipped slightly. 253 more construction workers were killed in 2016 than in 2011. That's an astounding 34% increase in fatalities over 5 years of significant industry safety, process, and technological advancement.
The rate of construction fatalities per 100,000 full-time employees (FTEs) decreased slightly from 10.1 in 2016 to 9.5 in 2017. However, the construction fatality rate remains significantly higher than the industry average of 3.5. Construction is the third most dangerous industry to be employed in the United States, only behind transportation and warehousing (fatality rate of 15.1 per 100,000 FTEs) and agriculture, forestry, fishing and hunting (fatality rate of 23.0 per 100,000 FTEs). Conversely, the manufacturing industry is one of the safest industries with a fatality rate of 1.9 per 100,000 FTEs in 2017.
Despite our best efforts, construction workers are still being killed on the job. To be more specific, each Sunday as you are getting ready to head back to the office, 18 families are grieving the loss of loved ones killed on the job the previous week. That's a staggering number. Why aren't we better at protecting our workers now than we were in 2011?
What is Risk Compensation?
Risk compensation is a theory which postulates that individuals will adjust the way they interact with their environment based on changes in perceived level of risk. Essentially, individuals will modify their behavior as their perceived level of risk increases or decreases. The nature of the relationship between action and perceived risk is inverse; as the perceived risk increases, individuals exhibit less risk-taking behavior (exercise more caution). As perceived risk decreases, individuals exhibit greater risk-taking behavior.
While some dismiss this theory, there are several studies which illustrate its applicability in other industries. Risk compensation has been studied heavily in road traffic safety. A study by Gerald Wilde (1994) illustrated that taxicab drivers took more risks when their taxis were equipped with ABS brakes, thereby nullifying the effect of their potential for incident reduction. In Munich, taxicab drivers with ABS brakes and conventional brakes recorded the same number of incidents. Conversely, another study showed that ABS brakes reduce the risks of accidents but increase the risks of run-off-road incidents.
In 1997, Sweden made a monumental shift in their driving practices. Drivers stopped driving on the left side of the road and began driving on the right. For anyone who has driven in another country where you were required to drive on your uncomfortable side, you can empathize with Swedish drivers. It takes your utmost concentration and attention to follow the road and keep track of what to do at each intersection. Swedish drivers logged only 150 minor accidents that day; the number of traffic incidents decreased significantly over the next few months.
It is postulated that this decrease in incidents was the result of risk compensation. As drivers perceived the level of risk increasing, they decreased their risky behavior. Other studies have been conducted on driving law changes with similar findings. Risk compensation may not explain the behavior of each individual in each circumstance, but it certainly warrants analysis in relation to broad construction industry trends related to fatality rates.
If we apply the theory of risk compensation to construction fatality trends, it would suggest that the more safety measures we implement, and the more advanced the equipment and technology, the more risk accepting workers will be. Essentially, workers may be more accepting of risk in compensation for their perceived decrease in environmental risk. In turn, this increased risk taking would result in a stagnation, or even increases, in fatalities within the industry despite efforts and advancements to achieve the opposite outcome.
Risk compensation is a controversial theory. It implies that positive risk mitigation efforts may have minimal or even negative effects. Application of the theory also places the focus on the actions of the worker rather than on systems and practices. However, we can extrapolate the application of this theory beyond the worker. It is possible that employers, managers, or front-line supervisors become more accepting of risk-taking behavior as the perceived environmental risk is lessened.
For example, if all site workers are equipped with fall arrest equipment, supervisors may not place as much emphasis on ensuring that an aerial work platform is available to rescue a worker if they fall from heights. While the harness will save a worker from impact trauma, it will not save them from the dangers of suspension trauma if they can't be quickly rescued. In this example, perceived risk reduction through the use of fall arrest equipment may result in less focus on other risk management efforts, making project teams more complacent and less diligent in identifying and mitigating risk.
Is Risk Compensation Mitigating the Benefits of New Safety Measures?
Without conducting a detailed study, it is nearly impossible to determine if risk compensation is detracting from overall construction safety initiative benefits; however, the topic warrants greater analysis. 971 workers were killed on the job in the United States in 2017; that's 971 too many. As an industry, our efforts aren't driving the safety performance improvements that we expect to see in a decade of substantial safety, process, and technological advancement.
No matter how many systems we have in place, or how good our procedures are, any increase in risk tolerance can yield catastrophic outcomes. If any stakeholder in the project value chain is becoming more accepting of risk because of the increased deployment of tools, technology, and equipment, the benefits of each of these programs will be offset. We may not be able to tackle this problem tomorrow; however, if our teams can't even remember the topic of our toolbox talks, we may want to start with that.