SafetyInsights.org

This ran a content analysis of Annual Reports (AP) of publicly listed Australian non-domestic construction companies to study executive safety incentives.

They cover some known issues with safety performance metrics. For common injury measures like LTI:

• “These types of measures have been criticized because they often combine and treat all types of injuries (irrespective of severity) equally as recordable incidents” (p9)
• “LTI measures (which include fatalities) can be used as evidence that declines in injury rates reflects improvements in health and safety performance” (p9). That is, an organisation can kill people but still demonstrate “improved” injury performance
• Lagging injury measures tend not to be measures of safety (which they define here as the freedom of risk to injury) but instead tend to be measures of productivity
• On the above, injury measures don’t measure the above type of risk, nor do they “Objectively evaluate risk drivers (latent conditions or hazards) or the effectiveness of controls over those hazards” (p9).
• Further, while LTIs “can confirm, in hindsight, that a risk had been present at the time of an injury” (p9), the absence of an injury doesn’t indicate the absence of an uncontrolled risk. This is highlighted by a number of major accidents that were preceded by “impressive injury free periods” (p9)
• They note that accordingly, measures like LTIs, MTIs etc. “do not and cannot provide a valid measure of safety”
• For the interaction of injury measures into performance incentive schemes, like long-term incentive plans, even if LTIs etc are integrated into incentives there is “potential for manipulation in order to secure bonus payments is a significant issue”
• As the authors previously noted, “linking bonus payments to reportable injury rates can have unintended or “perverse” consequences” (p9)
• Expanding on the above, “Underreporting may stem directly from pressure not to report. It may also result from sophisticated classification strategies that ensure that injuries do not count as injuries for the purpose of bonus calculations” (p10) and these issues also likely apply to manipulation of process safety indicators.

Results
First, although most companies reported on injury metrics in their AP, there was instances where companies indicated improved safety performance in the current period compared to the last – despite workplace fatalities having occurred.

Overall, companies were found to express a strong safety commitment in their AP with safety being central to long-term planning. Despite this, only short-term incentive plans, prioritising lagging injury metric performance was found. Instead, LTIPs were exclusively related to financial measures associated with share value and organizational profitability” (p11).

None of the long-term incentive plans (LTIPs) were found to link safety performance to senior executive bonuses – with exclusively financial targets set. The excess reliance on these types of injury metrics problematically conflate incidents of different severity.

A consequence of LTIPs is that these schemes may “direct the attention of CEOs and KMP to initiatives that focus on share price and earning performance and, as a result, overlook safety performance” (p11).
Although all of the companies indicated having a Board-appointed Committee responsible for safety strategy and governance, little evidence was found linking safety governance committees to the remuneration committees that have oversight of LTIPs.

The authors suggest that, in this sample, executives are “not incentivized to align their long-term decision-making with long-term safety objectives” (p.1). Findings support the authors previous work in the energy sector, where LTIPs are only linked to financial performance while “safety performance is essentially irrelevant” (p.11).

Despite the recognised limitations of lagging indicators, the authors note that this is not evidence to abandon lagging indicators in favour of leading – as leading are likely as corruptible to manipulation in order for bonuses to be achieved, e.g. counting the frequency of something rather than its efficacy.

Executive safety incentives can lead to unintended consequences.

Authors: McDermott, V., Zhang, R. P., Hopkins, A., & Hayes, J. (2018). Construction Management and Economics, 36(5), 276-290.

Study link: https://doi.org/10.1080/01446193.2017.1381752

Link to the LinkedIn article: https://www.linkedin.com/feed/update/urn:li:ugcPost:6941876983119130624?updateEntityUrn=urn%3Ali%3Afs_updateV2%3A%28urn%3Ali%3AugcPost%3A6941876983119130624%2CFEED_DETAIL%2CEMPTY%2CDEFAULT%2Cfalse%29