Does it matter why power outages occur? An example of information-induced gender-heterogeneous preference on the intensive and extensive margins of choice

Abstract

People experience welfare losses in the event of power outages. A higher share of wind power in the power generation mix increases difficulties in forecasting power generation. This, in turn, challenges an effective utilization of wind power and increases the risk of power outages. Using a split sample approach, we test whether private households’ preferences for power outages are sensitive to information regarding the positive influence that controlled power outages can have on the utilization of wind energy in particular. Information increases the overall acceptance of contracts with low compensation, but influences females and males differently. Relative to female respondents, information increases male respondents’ preferences for accepting a contract, whilst information reduces preferences among females, relative to males, for contracts with higher levels of compensation.

Keywords:

• power outages
• information
• margins of choice
• gender

JEL Classification:

• C25
• C93
• D12

Acknowledgements

The authors would like to thank Astrid Kiil and Niels Coley from KORA, Danish Institute for Local and Regional Government Research, and Lars P. Geerdsen, Director at Krak Institute for Urban Economic Research, for their valuable comments. A thank also goes to the Danish Agency for Science, Technology and Innovation for financing the completion of this article. Any errors remaining are, of course, the sole responsibility of the authors.

Appendix A: Selection model

Logit model: Selection model (if received information =1, else =0)

Appendix B: Preference models

Logit model: Choice of contract in relation to the duration/power outage (marginal effects)

Logit model: Choice of contract in relation to the number of power outages/year (marginal effects)

Logit model: Choice of contract in relation to the compensation, DKK/year (marginal effects)

Notes

¹ The compensation levels are analysed in pairs in order to increase the statistical power of the analysis.