TL;DR
This paper argues that optimal energy policy must account for both external costs (like pollution) and internal costs (consumers undervaluing future energy expenses). They introduce the "Internality Dividend," showing that taxes on energy use can significantly boost welfare by correcting both issues simultaneously. The authors propose an "Internality Targeting Principle" for designing multi-instrument policies, suggesting a combination of energy taxes and product subsidies tailored to consumer biases.
What Problem Does This Paper Address?
Policymakers often design energy policies, like carbon taxes, to address negative externalities, such as pollution from burning fossil fuels. However, this paper highlights a crucial overlooked factor: consumers often undervalue the long-term energy costs associated with durable goods (like cars or appliances). This "internality" leads to underinvestment in energy-efficient options, even when they would be financially beneficial for the consumer over time. The challenge is to design policies that effectively address both external environmental costs and these internal consumer biases to maximize overall social welfare.
Methods
The authors develop a theoretical model of consumer choice for energy-using durables, incorporating both externalities and internalities. They use discrete choice simulations, specifically modeling the U.S. automobile market, to quantify the welfare effects of different policy interventions. A key methodological contribution is the development of "sufficient statistics" – reduced-form parameters that can be estimated from market data (like price elasticities) to approximate optimal policy without needing to fully estimate the underlying structural model of consumer behavior.
Datasets
The paper utilizes data from the U.S. automobile market for its simulations. Specific data sources mentioned include:
- ID Power and Associates "Power Information Network": For vehicle transaction data.
- National Vehicle Population Profile: For market shares.
- U.S. Environmental Protection Agency (EPA): For fuel economy (MPG) data.
- National Household Travel Survey: For vehicle utilization data.
Key Insights and Ideas
- The Internality Dividend: This is a core concept. It shows that policies designed to address externalities (like a carbon tax) can also generate significant welfare gains by correcting consumers' underestimation of future energy costs. This dividend can potentially double the welfare gains of a carbon tax.
- Internality Targeting Principle: When using multiple policy instruments (e.g., taxes and subsidies), the optimal design depends on the average "internality" of consumers most responsive to each instrument. This suggests a nuanced approach rather than a one-size-fits-all solution.
- Optimal Policy Mix: The paper suggests that optimal policy often involves an energy tax set below marginal external damages, combined with a subsidy for energy-efficient products. This combination is more effective at correcting both externalities and internalities than a single instrument alone.
- Sufficient Statistics for Policy: The development of reduced-form formulas allows policymakers to estimate optimal policies and welfare effects using observable market data (like price elasticities and demand responses) without needing to fully specify and estimate complex structural models of consumer behavior. This makes the findings more practically applicable.
Key Results
- In the auto market simulations, a carbon tax at marginal damages, when accounting for internalities, can yield more than double the social welfare gains compared to a scenario without internalities.
- The "second-best" optimal policy often involves a combination of an energy tax and a product subsidy. For example, a carbon tax of $0.15/gallon combined with a product subsidy of $665 per vehicle is shown to be highly effective.
- The paper demonstrates that policies designed to target biased consumers (those who undervalue energy costs) can be significantly more efficient than policies that treat all consumers as rational.
Limitations and Future Work
- Estimation Challenges: The authors acknowledge that precisely estimating the joint distributions of unobservables that determine optimal policy can be difficult. This is why they focus on developing "sufficient statistics" that are easier to estimate.
- Model Simplifications: The theoretical model makes simplifying assumptions, such as consumers having single unit demand and specific functional forms for utility. While the authors explore extensions (e.g., in the appendix), these simplifications are inherent to the model.
- Future Work: The paper suggests further research into the application of these principles to other energy-consuming durables and the empirical estimation of the derived sufficient statistics. They also highlight the importance of understanding how different types of behavioral biases (e.g., present bias, inattention) interact and how policies can be tailored to address them.
Summary updated