Understanding the regulator–regulatee relationship for developing safety culture

TL;DR

This qualitative nuclear-industry study finds that regulators influence regulatee safety culture through three perceived roles—enforcer, partner, and advisor/consultant—supported by relationships marked by professionalism, transparency, and balance between formal enforcement and informal exchange. It also identifies self-development facilitation and implicit influence as mechanisms, with regulator roles serving content, credibility, and will.

Key findings

• The regulator was perceived to hold three complementary roles for safety culture development: enforcer, partner/collaborator, and consultant/advisor. Dutyholders expected and accepted a strong enforcement role, while regulator inspectors noted that influence powers were used more often than formal enforcement.
• Effective regulator–regulatee relationships were characterized by three themes: balance between formal enforcement and informal exchange, professionalism, and transparency. Dutyholders reported never losing sight of the regulator’s enforcement powers, and regulators emphasized independence and appropriate, proportionate action.
• The roles were linked to three higher-order purposes: providing content/knowledge, stimulating will/motivation, and awarding credibility. These mapped onto Vogus et al.’s safety culture development processes of enablement, enactment, and elaboration.
• Mechanisms facilitating safety culture development were facilitating dutyholder self-development and implicit regulation, where safety culture was often addressed indirectly through other areas such as leadership. Some dutyholders wanted a more formal, explicit regulatory approach, such as a license condition.

Methods

• Qualitative iterative design in the nuclear industry. Data included 12 semi-structured interviews with employees from five nuclear dutyholder organizations, three follow-up interviews with senior nuclear inspectors, 12 supplementary interviews with nuclear regulator inspectors, and analysis of two regulator documents. Interviews were conducted July 2020–January 2021, averaged 60 min, and were transcribed verbatim.
• Thematic analysis followed Braun and Clarke’s six steps. Phase 1 was predominantly inductive; phases 2 and 3 used a more deductive approach with some inductive review. Semantic and latent coding were used, with triangulation, cross-coding, member checking, and NVivo for an audit trail.

Limitations

• Stated: the study is primarily interview-based and reflects individuals’ experiences at a given point in time; it did not include more objective data on regulator influence; question wording such as “collaboration” may have primed participants toward informal interactions; participants knew the regulator would receive a summary, which may have influenced responses. Future research is suggested in other industries and with quantitative or longitudinal designs.
• Evident from the text: the findings come from a small qualitative sample in one heavily regulated nuclear context and are based on perceptions, limiting generalizability and causal inference.