SafetyInsights.org

This was really interesting. It studied CEO investment style on workplace safety.

Two styles were included: underconfident and overconfident CEOS. Underconfident CEOs are those that are less certain about future company value or performance and thus underinvest in order to maximise short-term value.

Underinvestment has been shown to contribute to workplace accidents. Underinvestment can involve too little investment in CAPEX which can result in old machinery, poor maintenance, and lack of safety features.

Conversely, overconfident CEOs – those CEOs that overinvest in CAPEX and R&D to maximise future performance, since they’re confident the investment will pay off, are hypothesised to inadvertently have a positive impact on workplace safety. This is because while trying to maximise future earnings, the newer equipment introduces new safety protocols and features.

Note that an interesting feature of this study isn’t about CEOs that prioritise safety or not but rather how CEOs inadvertently create safer environments by investing to enhance the financial performance of the company.

There’s way too much to cover in this paper both for the methodology and findings. But data was based on all CEO compensation data from 1992 – 2015 for that region and a host of other data.

Results

As expected, overconfidence (again, CEOs that overinvest in CAPEX & R&D, which creates asset growth and PP&E growth) is associated negatively and statistically significantly with accidents after controlling for a host of factors (negative association as in, as overconfidence investment goes up, accidents go down).

This effect was also economically significant. Quoting the paper, “a one standard deviation increase in CEO confidence is associated with a 2.1 percentage point reduction in the number of accidents per year” (p3). Further, they found that “overconfident CEOs [were] associated with a 14 percentage point reduction in the number of accidents for firms with internal capital constraints, and a 19 percentage point reduction for those with borrowing constraints” (p4).

 In all, overconfident CEOs experience around 2.4% fewer accidents compared to underconfident CEOs and 1.3% fewer accidents per employee.

For other findings:

·        R&D intensive firms had fewer accidents compared to non-R&D heavy firms

·        Longer tenured CEOs experienced fewer accidents, hypothesised due to CEOs becoming more familiar with their companies and cognisant of how to prevent those failures

·        The effect was stronger in more accident prone industries compared to less accident prone industries such that overconfident CEOs had 4.4% fewer accidents overall and 2.17% fewer accidents per employees

·        Overconfident CEOs had a greater impact on safety in states with relatively lax labour laws compared to states with more strict labour laws. According to the paper, this finding “implies that in states with strict labor laws, all CEOs are forced to invest more in worker safety. However, in states with weaker labor laws, where CEOs might otherwise underinvest, overconfident CEOs’ investment activities have the greatest impact” (p4).

·        On the above, firms operating in strict labour law states experienced around 2.1% fewer accidents compared to firms in more lax labour law states (which highlights the impact of legislation).

Additionally, when firms were divided into groupings of being cash constrained or not cash constrained, overconfidence only reduced accidents in the cash constrained sample. Overconfidence in non-cash constrained firms had no statistically significant impact, being unsurprising since those CEOS could engage in significant CAPEX.

It’s said that even with cash constraints, overconfident CEOs continue to invest in CAPEX; further indirectly improving safety.

Moreover, unionisation was found to improve safety – more unionised states had 4.8% lower accidents compared to lower unionised states. Unionisation was found to offset the impact of overconfidence, possibly suggesting that unionisation forces non-overconfident CEOs to invest in safety enhancing actions.

Next they looked at the relationship that reduced accident risk has on higher labour productivity and thus higher firm value. It’s said that “accidents are negatively related to firm value and operating performance. This supports the notion that reducing accidents benefits both workers and shareholders due to the significant costs that those accidents can impose on firms” (p24).

Finally, authors address the question about what happens when CEOs change? They focus specifically on a non-confident CEO changing to an overconfident CEO rather than the reverse (since an overconfident CEO will have a lasting impact for months to years after they leave which will mask the true effect). A statistically significant reduction on accidents was found when a firm replaced a non-overconfident CEO with an overconfident CEO – reducing accidents by 6.35%.

Authors: Banerjee, Suman, Mark Humphery-Jenner, Pawan Jain, and Vikram Nanda, 

Study link: https://acfr.aut.ac.nz/__data/assets/pdf_file/0007/578752/Mark-HumphreyJenner-Environmental,-Soc-and-Gov.pdf

Link to the LinkedIn article: https://www.linkedin.com/pulse/safety-first-overconfident-ceos-reduced-workplace-ben-hutchinson