The willingness to accept environmental fees in Poland

Discrete choice experiment - study participants (aged 18-64) will choose between hypothetical policy solutions aimed at climate mitigation and energy security. The choice tasks are random and differ in terms of attributes and their levels.

2022-08-08

2022-09-16

Willingness to pay (WTP) for four attributes:
1) mitigation of climate change impacts (2) decrease in the mortality rate due to poor air quality (3) limited imports of gas and oil from Russia (4) uninterrupted supply of energy.

Willingness to accept policies (e.g. carbon taxes, emission trading systems). Willingness to pay for particular attributes (1-4).

Heterogeneity of outcomes in terms of gender, education level, place of residence, the main source of heating, number and type of vehicles, occupation, incomes, energy expenditures, beliefs about climate change, beliefes about energy security, political preferences and trust in political institutions.

We set the experiment in the context of the residential energy sector in Poland. We study the willingness to pay for four attributes: (1) mitigation of climate change impacts (2) decrease in the mortality rate due to poor air quality (3) limited imports of gas and oil from Russia (4) uninterrupted supply of energy. We use the experiment to evaluate if achieving support for climate policy goals is more viable by environmental (attributes 1-2), or energy security (3-4) framing. We use the experiment to evaluate the acceptance of the introduction of environmental fee systems (e.g. a carbon tax or emission trading). We differentiate the policies between status quo (no policy), carbon tax + universal basic income and carbon tax + subsidy for home-based renewable energy solutions.
The study is based on an internet survey (CAWI). In the first part of the survey, we collect basic information about participants’ socio-demographic characteristics. Afterwards, we introduce a discrete choice experiment and ask all participants to state their preferences over hypothetical policy solutions comprised of four attributes with three levels each:
(1) mitigation of climate change impacts (a. Severe [large drop in agricultural production; substantial threat to human life due to heatwaves, floods and droughts]; b. Moderate [moderate drop in agricultural production; some threat to human life due to heatwaves; floods and droughts]; c. Mild [the most severe impacts of climate change are prevented; some effects of global warming present however not severe])
(2) decrease in the mortality rate due to poor air quality (a. No change [50,000 deaths per year]; b. Limited by half [25,000 deaths per year] c. Limited to a minimum [5,000 deaths per year]).
(3) limited imports of gas and oil from Russia (a. No change [imports of 20 billion m3 of gas and 30 million tons of oil annually]; b. Limited by half [imports of 10 billion m3 of gas and 15 million tons of oil annually]; c. Limited to zero [no imports of gas or oil]
(4) uninterrupted supply of energy (a. No change; b. Limited access [at least one outage between 7-10 a.m. per week, two Sundays per month without driving a car]; c. Minimum access [at least one outage between 7-10 a.m. and 6-9 p.m. per week, all Sundays without a car])
Each study participant will see five screens.
On each screen, there will be options for two public policies: a status quo (no policy implemented) and a new policy.
Each pair of policies will vary in terms of four attributes: the five mentioned above and the expected gain/loss from the introduction of the policy. The set of number attributes will be selected randomly (with equal probability).
The expected gains/losses from the introduction of the policies will be randomized (uniform distribution) in the range of {-24%,-20%,-16%,…,0,…,16%,20%,24%} deviations, taking into account the incomes and energy expenditures of the particular individual.
After the experimental part, we will collect additional information on political preferences and trust in institutions.

Randomization done by a computer system used to prepare the experiment. Each respondent will see five sets of two vignettes with attributes with a randomly selected level (1-3, with equal probabilities).
Additionally, in each vignette in a set of five, one vignette (randomly, with equal probabilities) will be a "status quo" option (with a cost/benefit of "0") and one (randomly, equal probability) will be either "tax + cash benefit" or "tax + subsidy" and a random value of cost/benefit drawn from a set of {-24%,-20%,-16%,…,0,…,16%,20%,24%} (with equal probabilities).

Individual

No

Choices clustered at the level of 10,000 indivudals

10,000 individuals responding to five choice tasks, 50,000 observations.

There is no division into treatment and control group. 10,000 individuals recieve five choice tasks each. The tasks are random as well as the attribute levels.