What shall we talk about?: How differences in interest group type and language matter for gaining influence in regulatory rulemaking

Interest group literature has often aimed at explaining differences in influence on public policymaking. The empirical focus has been on stakeholder type, levels of insiderness and levels of resources as potential reasons for differing levels of influence. With the increasing prevalence of online consultations used by regulators, particularly in regulatory rulemaking, written language has gained in importance for interest group politics. And as the abilities and of generative AI language models such as ChatGPT advance, the question arises what the impact of the language used by interest groups is on their success in gaining influence. This paper zooms in on the importance of written language on interest group influence. It combines interest group-, bureaucratic reputation- and regulatory communication literature to establish expectations on what subjects in written text are most useful to what particular type of interest group in gaining influence. Using a dataset of 26,468 consultation comments left by stakeholders in EASA rulemaking consultations, I test what type of stakeholder using what type of language is most successful. I find that most stakeholder types, whether they represent public or private interests, are more influential when using technical language as opposed to moral language. Those representing government organizations and business associations excel when using legalistic and procedural language. The findings indicate that it matters what interest says what in regulatory rulemaking. This matters as concerns for undue interest group influence can partially be put to rest if it is clear that they gain influence on terms that are ethically defensible, such as providing expertise. While not testing the actual transfer of expertise or other resources, the findings do provide implications in this direction.