When Too Much Is Too Much: Media and Stigma in the Nuclear Industry

In a study on nuclear power generation, Fabrizio Perretti (Department of Management and Technology) and former Bocconi PhD student Alessandro Piazza (now at Columbia Business School) find that firms are much more likely to abandon nuclear practices in periods when they are more intensely stigmatized. This effect is weaker (and even becomes negative) when media exposure is high or when the stigmatized practices are a large part of the firm's total business. The authors report these conclusions in a recently published article in Organization Science (Volume 26, Issue 3): Categorical Stigma and Firm Disengagement: Nuclear Power Generation in the United States, 1970 – 2000 (doi: 10.1287/orsc.2014.0964).

Since disengagement from a particular industry is costly, firms are unlikely to disengage easily. Therefore, the authors propose that firms will be more likely to disengage the more intense the stigmatization they face. More surprisingly perhaps, the second hypothesis predicts that media exposure reduces the likelihood that a firm disengages. This is because audiences do not respond as strongly to repeated information, as they get used to the information or overloaded and left unable to process it. Furthermore, firms that have survived an initial wave of stigmatization are less likely to be exposed to public scrutiny. The last hypothesis states that firms that have a higher extent of involvement in the stigmatized activities are less likely to disengage. These firms have too many vested interests, and the activity is considered "core" and integral to the purpose and identity of the organization and are thus unwilling to cease their activities.

The authors test these hypotheses in the context of US nuclear power generation plants in the period from 1970 to 2010, when the practice of nuclear energy generation was stigmatized due to manifold protests, two nuclear accidents, and the efforts of several environmentalist groups. However, the intensity of this stigma varies from year to year, which the authors measure as the proportion of articles on nuclear power generation appearing in the New York Times that highlight negative, as opposed to neutral or positive features of this practice. Furthermore, media exposure, measured as the total amount of articles published independent of tone, also varies over time. Lastly, some firms produce power using only nuclear power plants, whereas others also produce energy using coal, oil, or gas-fired plants. The extent of membership in the stigmatized category is thus the percentage of the nuclear plants the firm owns over the total number of plants.

In order to test these predictions, the authors use logit regressions as well as several robustness tests to test for alternative specifications and explanations. Indeed, these tests confirm all three hypotheses.

Overall, these results seem to indicate three important contributions to theory. First, a counterintuitive finding that firms that are more active in an industry, and thus more likely to be stigmatized, are less likely to disengage. This means that firms with an apparently strong identity are more willing to bear a stigma rather than to change their identity. Second, firms are not only willing to take action when attacks are directed to the firm itself, but even stigmatization by association with an activity can prompt firms to take drastic action. Last, an abundance of media-attention regarding the industry under stigma can actually reduce the amount of stigma people are able to perceive.