Firms are under increasing pressure to be transparent along a host of dimensions. Mandatory information disclosure programs abound, forcing firms to disclose the caloric content of foods they sell, the chemicals they use and emit, and of course, a host of financial information. Meanwhile, ‘voluntary’ information disclosure efforts are growing as well, as a multitude of ratings agencies and stakeholder representatives pressure firms to reveal various aspects of their environmental and social performances.
Of course, transparency is a means, not an end. The point of transparency is to reduce information asymmetry, and thereby allow investors, customers, community residents, and other stakeholders to make better decisions. In an ideal world, the increase in transparency would cause laggard firms to improve as they either were shamed by their poor performance or felt competitive pressures to improve because better performing firms were rewarded by the groups that use the information.
The evidence is starting to accumulate that the effect of transparency is highly contingent on a number of factors.
But, we don’t live in that ideal world, and the evidence is starting to accumulate that the effect of transparency is highly contingent on a number of factors. I’m going to cover three broad factors here, but I’m not pretending that this is a collectively exhaustive list. Instead, it’s kind of a convenience sample of factors that I’ve seen in research I’ve read recently or have been involved in myself. These factors seem to matter in determining whether disclosure programs actually lead to improvement. Or, perhaps more accurately, these factors affect both how much improvement the programs engender overall and where the improvements are felt most and least strongly.
1. Program Construction Matters: Fung et al. (2006) show that not all programs are created equally. They discuss the degree to which a program is embedded in both users and producers’ decision making as a crucial determinant of a program’s effectiveness. Embeddedness comes about when the information provided is considered important to its users and producers, is relatively comprehensible to users, and when the managers can act on the feedback they receive. For those of us using the Toxics Release Inventory for our sustainability research, it is noteworthy that this form of information disclosure does not score well in the Fung et al. (2006) analysis – the information is disaggregated and complex and is not usually available readily at the time when the users might best make use of it (say, during a house purchase decision).
2. Attributes of Disclosers Matter: Even holding the disclosure program constant, there is evidence that not all information producers are equally affected by disclosure. When disclosure shames an entity sufficiently (because, for example, it makes a ‘’worst of” list), it is more likely to improve (Chatterji and Toffel 2010). But shame does not seem to be felt equally by all information disclosure. There is some evidence that those organizations that are most likely to be sensitive to embarrassment are more likely to improve following disclosure. For example, privately held firms, where owners are presumably more closely identified with the firm than the more dispersed shareholders of publicly traded firms, seem to respond more to disclosure pressures (Doshi et al. 2013).
3. Attributes of the users who receive the information matter. Not all populations are equally sensitive to the information disclosed, or capable of acting upon it. Bollinger et al. (2011) find that caloric disclosure in Starbucks had a bigger impact in higher income and more educated areas of New York. In a current working paper, Arturs Kalnins and I have found that in the first two decades of the Toxics Release Inventory, emissions improved far slower in the poorer counties across the United States, raising the possibility that those people who had the least capacity to put pressures on polluting firms benefitted the least from the transparency that the firms were forced to provide.
Overall, the three issues I’ve identified (as well as some that I’ve no doubt omitted) point to the need for policy makers or pressure groups that promote disclosure to think about what can catalyze improvement following disclosure. It is an important, but incomplete step for disclosure to reduce the information asymmetry between companies and stakeholders. The next step requires better design of disclosure systems and attention to how to ensure that the sunlight that is shed on organizations does, in fact, disinfect them.
Citations:
Bollinger, Bryan, Phillip Leslie, and Alan Sorensen. “Calorie Posting in Chain Restaurants.” American Economic Journal: Economic Policy 3.1 (2011): 91-128.
Doshi, Anil R., Glen WS Dowell, and Michael W. Toffel. “How firms respond to mandatory information disclosure.” Strategic Management Journal 34.10 (2013): 1209-1231.
Weil, David, et al. “The effectiveness of regulatory disclosure policies.” Journal of Policy Analysis and Management 25.1 (2006): 155-181.
Glen Dowell is Assistant Professor of Management and Organizations at The Johnson School, Cornell University