Intervention (Hidden)
We field a large-scale survey RCT. In particular, we assign consumers different climate change scenarios and elicit the impact of different scenarios on consumer expectations about macroeconomic variables.
Consumers’ expectations about the effects of climate change on the macroeconomy may be influenced by their socioeconomic characteristics, as well as by idiosyncratic, unobserved factors and beliefs that are likely time varying (e.g., concerns about the future in general, political beliefs, knowledge about the effects of climate change, etc.). Such unobservables are likely to also correlate with consumers’ awareness of different climate scenarios. As a result, estimating the causal effect of different climate change scenarios on consumers’ beliefs about the macroeconomy is empirically challenging and cannot be addressed with standard econometric methods.
To address this empirical challenge, we will implement a randomized controlled trial in the September 2024 (initial pilot) and June 2025 rounds of the CES as part of a special-purpose module fielded after the regular survey. Respondents will be randomly divided into four different groups at the country and recruitment type levels. Each group will receive a different hypothetical scenario for global average temperature change. First, we will explain that “According to historical data, the annual average global temperature in 2023 has increased significantly by about 1 degree Celsius compared to 50 years ago”. This information helps put the total change in temperature in perspective.
Thereafter, we will provide respondents with different temperature scenarios (ranging from -1.5 to +3 degrees Celsius) and subsequently elicit their qualitative expectations for important macroeconomic variables (e.g. consumer prices, growth, unemployment). Additionally, in June 2025, we will also collect quantitative forecasts for consumer prices and growth, as well as their willingness to pay, to mitigate an increase in average global temperatures.
In addition to their expectations about consumer prices, respondents indicate the expected consequences for the macroeconomy (economic growth and unemployment), asset prices (stock prices and house prices), public finances (government debt and taxes), as well as their household financial well-being. Apart from the economic variables, we ask for the consequences for biodiversity and immigration to verify the plausibility of consumer expectations for non-economic variables. Finally, we ask for the consequences of “the number of Oscar-winning movies,” which is included as a placebo outcome item, as there is no reason to expect a change related to global temperature changes. If consumers expect that global temperature systematically affects the number of Oscar-winning movies, this would be a reason for concern about the experiment design and the presence of survey demand effects on outcome variables. To avoid any order effects influencing our results, we block-randomized the order of the outcome variables, grouping those that logically belong together, such as the three items related to price expectations.