Intra-industry benchmarking of discretionary corporate disclosure
5 July 2016
University of Antwerp, Stadscampus, Kapel Grauwzusters - Lange Sint-Annastraat 7 - 2000 Antwerp
Prof Walter Aerts
PhD defence Oveis Madadian - Faculty of Applied Economics
Relying on two theoretical perspectives from prior research on organisational behaviour, namely institutional theory and the behavioural theory of the firm (BTOF), this dissertation aims at exploring determinants and economic consequences of corporate disclosure practices that are characterised by means-ends uncertainty and lack sufficient normative guidance to lead managers (external observers) in their decision making processes (assessment of firms’ disclosures and underlying activities).
In this dissertation, we try to meet two key research objectives: (1) to investigate whether and how social comparison (i.e., comparison of performance against a socially-derived benchmark, such as performance of industry-based peer firms) and the related performance feedback (i.e., whether the firm’s disclosure behaviour pattern is better/worse than the benchmark) affect a firm’s disclosure behaviour; and (2) to examine implications of peer-based benchmarking for firms and for external observers, such as financial analysts and investors. To do so, we conduct four studies, of which the first three ones focus on selling, general and administrative expenses (SG&A) and the fourth one focuses on corporate social responsibility (CSR) disclosure.
Using a large sample of listed US firms, the first three studies document the following results. First, social comparison generally affects firms’ decisions to change the SG&A ratio (SG&A relative to sales). Moreover, the effect is documented to be stronger for firms (1) in the introduction and decline stages of life cycle, especially for those with a previous SG&A larger than the benchmark (average SG&A of their peers) and (2) in the peer groups characterised by higher similarity in terms of SG&A ratios. Second, higher similarity between the firm’s SG&A ratio and the benchmark is associated with the lower number of analysts following the firms. Also, increases in SG&A similarity are found to be negatively (positively) associated with the dispersion (accuracy) of financial analysts’ forecast of one year-ahead earnings, suggesting the uncertainty- (information asymmetry-) reducing effect of SG&A similarity. Third, SG&A similarity is found to be positively associated with future performance (net income and cash flow from operations) and stock returns only for firms with a previous SG&A ratio larger than the benchmark, confirming value relevance of SG&A similarity. In addition, change in SG&A similarity is found to be perceived as a credible signal by investors (i.e., positively associated with stock returns), again only for firms with a previous SG&A ratio larger than the benchmark.