Pro-Environmental Behavior in High Cost Situations –Evidence from a Mixed-Mode Panel in Germany

Empirical studies reveal that even respondents with pro-environmental concerns frequently do not show pro-environmental behavior. According to the low-cost hypothesis on the one hand pro-environmental concerns only matter in situation where pro-environmental behavior is perceived as low-priced by the actors, pointing to additive effects of instrumental incentives (e.g. low costs) and concerns or a positive interaction between them. On the other hand, dual process theories claim a negative interaction between pro-environmental concerns and constraints, i.e. actors with strong concerns behave in an automatic-spontaneous way by neglecting the costs. Our study operationalizes both approaches to assess which one performs better in an empirical test. The data come from two waves of the Gesis Panel, where two different forced choice scenarios were presented, systematically changing the costs of pro-environmental decisions. Using a multilevel-logistic regression model, we are able to control for the perceived definition of the situation (high vs. low-cost) and will check for framing effects regarding the interpretation of differential wording of gains. If respondents have to choose between green energy and conventional energy, pro-environmental concerns have a robust positive effect on choosing green energy regardless of the definition of the situation as high or low cost. Further, strong pro-environmental concerns have still an effect on the choice if conventional energy is much cheaper than green energy. If respondents have to choose between sustainable and conventional investment opportunities the effects of environmental concerns depend on the situation´s definition as high or low cost: while we find positive effects of environmental concerns on the probability to choose the green investment in low cost situations, we do not find any significant effect for high cost situations. We conclude that the explanatory power of the low cost hypotheses depend on the framing of the choice problem.