In-group bias happens when people give preferential treatment to others they believe belong to their group. The group could be anything — gender-based, religious, a fellow fan of the same sports team. What matters is the perception of a shared identity that creates a sense of trust and disarms scrutiny.

“In-group bias is a prevalent and consistent psychological force,” says Sae-Seul Park, an assistant professor of strategic management at the University of Toronto’s Rotman School of Management.  “It bleeds over even into things that should be completely objective, such as judicial rulings.”

As she discovered, it also included marine vessel inspectors with shared nationality. Park developed a research relationship with a private marine inspection company in an unnamed jurisdiction and was given access to data on the regulatory compliance inspections done by its employees. The resulting study appeared in Strategic Management Journal.

Two years of data from 27,637 routine inspections for commercial vessels — from oil tankers to cargo ships — showed that inspections of domestic vessels were 11 per cent shorter on average than inspections for foreign ships, even though all inspections follow globally mandated standards and are conducted in English.

That changed after a catastrophic accident involving a severely overloaded domestic vessel. The incident, which resulted in hundreds of casualties, sparked national outrage when it was revealed that the ship had undergone only a brief inspection and did not meet basic safety standards. (It should be noted the vessel was not inpsected by the inspection firm analyzed by Park.) 

Following the accident and resulting outrage, inspections of domestic vessels became 18 per cent longer than those for foreign craft, equivalent to about one hour and 12 minutes.

That shift is an indication that the previous in-group trust with domestic clients was broken by the accident, prompting more rigorous inspections, Park says. This finding also provides evidence that shorter inspection times for domestic clients were indeed driven by in-group bias, rather than other factors.

If that were not the case, “we should see that inspectors exercise greater scrutiny for all inspections after that incident,” she says. “What we observe though is that everyone becomes more careful only for in-group vessels.”

The change was driven by inspectors with less additional training on top of the minimum requirements for the job. That group bumped up their inspection time for domestic vessels by an extra hour and 20 minutes, or 22 per cent. No such increase was seen for better trained inspectors, perhaps because there was less room for improvement, Park wrote in her study, co-authored with Sunkee Lee and Oliver Hahl, both of Carnegie Mellon University.

Additional training can be used as a marker of professionalism, the researchers wrote, and the data shows that professionalism can act as a counterweight to the influence of in-group bias on compliance monitoring.

The results show that “human capital is important — and individual human capital characteristics are important too,” says Park, meaning that companies concerned with enforcing standards need to pay attention not only to policies but to the people doing the work.

“No matter how hard a firm works to set optimal routines and manage operations, ultimately, it’s people who execute the actions that implement these strategies,” she says.