ABSTRACT
This study explores how four different types of environmental reporting regulations affect reporting practices. Accounting Act requirements, accounting standard requirements, accounting standard recommendations, and no regulation/voluntary disclosure are associated with different levels of reporting obligations. Disclosures made by enterprises subject to regulations are compared with those of enterprises that are not. There are separate regression models for each type of regulation. The sample consists of 235 enterprises from the private and public sectors. Content analysis is used to measure environmental disclosure. Enterprises subject to regulations report significantly more types of the information content required by law than other enterprises, which is in line with the higher regulatory legitimacy risk. There is no such difference in disclosure between the two groups of enterprises for the information required and recommended by the accounting standard. This may suggest that pragmatic, cognitive, and moral legitimacy issues outweigh the regulatory legitimacy risk for these types of information, or that legitimacy risks are generally low. Enforcement of regulations will increase the regulatory risk. For information that is voluntary for all enterprises to disclose, enterprises that are not subject to any regulations report significantly more types of information than those that are. This result is not in line with predictions made from any of the four types of legitimacy. Some alternative explanations are discussed. Since regulatory regimes may include several types of means, the main contribution is the comparison of four types of regulations within the same regime, as opposed to analysing only one type of regulation at a time such as in the extant literature. The study also explores different types of legitimacy, and addresses the lack of research on environmental reporting in the public sector.
Acknowledgements
The author gratefully acknowledges the comments from Professor Lars Fallan, Trondheim Business School; the journal editor, Professor Thomas Riise Johansen, Copenhagen Business School; and two anonymous reviewers, and the assistance of Helene Brekke, Inger Tone Hofsmo, Mette Johansen, Mari Ramstad, and Catrine Vang.
ORCID
Even Fallan http://orcid.org/0000-0002-7272-1742
Notes
1. Further, this study does not analyse compliance with regulations or the level of disclosure, only the difference in reporting between enterprises subject to regulations or not (and different other corporate characteristics).
2. The environmental reporting regulations apply even to companies that have to prepare financial (group) statements according to International Financial Reporting Standards (IFRS) because of the European Economic Area agreement.
3. Voluntary reporting standards (e.g. Global Reporting Initiative) might be included in . Such standards are not considered in this study.
4. As of fiscal year 2013 (which was reported for the first time in 2014), an additional section is operative (3–3c). However, this study's data do not cover the 2013 annual reports, so these requirements are not discussed here.
5. However, auditors should control whether environmental information content in line with the requirements of the Accounting Act and NRS 16 is provided in the board of directors’ report. This task is also included their checklists.
6. It might be of interest to stakeholders whether enterprises with whom they have exchanges/interdependencies comply with (reporting) regulations or not, but that pragmatic legitimacy risk is likely to evaporate more quickly than the accompanying normative regulatory legitimacy risk when non-compliance does not appear to have consequences for the enterprise.
7. These predictions are based solely on risks directly related to reporting and reporting regulation. Moral, cognitive, and pragmatic legitimacy may also affect environmental reporting because reporting is used as a legitimation tool concerning risks related to other issues (Lindblom Citation2010). A consideration of such risks is outside this study's scope.
8. Except in model 1, where environmental risk (p = 0.055) just misses the 5% boundary.
9. Of course, ideally, all enterprises subject to regulation should comply with the regulations in regression models 1–3, so that no variables (except possibly REGULATION) are significant. In reality, that is not the case.
10. The most basic issue here is whether the information is important for stakeholders.
11. Even the descriptive statistics in this study show that the actual level of reporting for the dependent variables are much lower than the maximum theoretical score.