Supporting Climate Adaptation Overseas by Providing Aid or Accepting Refugees

• In Event: Living with the Consequences of Climate Change

Abstract

What influences the willingness of Japanese people to assist developing countries to adapt to climate change? Japan is a global climate leader. In 1997, it hosted the Kyoto Conference, which led to the adoption of one of the most important climate agreements. Experts attribute climate change mainly to the burning of fossil fuels, which create carbon dioxide emissions. These emissions have been accumulating for the last two hundred years ever since coal and petroleum became major drivers of economic activities. From a historical context, developed countries have contributed most to these accumulated emissions. Because Japan is also an industrialized country, it bears substantial responsibility for climate change.

The negative impact of climate change is already visible worldwide. Japan is experiencing the increased frequency and severity of extreme weather events such as typhoons, floods, and heatwaves. But such damage tends to be more catastrophic in developing countries that lack the resources to adapt to sea-level rise and the increasing frequency of droughts, flooding, and hurricanes, all accentuated by climate change. As a responsible developed country, Japan is expected to assist developing countries in adapting to climate change. However, Japanese citizens might not be equally supportive of types of support or all developing countries. For example, their preferences may vary depending on which country gets the support, in what form, and for which type of extreme weather event.

To test the Japanese public’s willingness to provide climate adaptation assistance to developing countries, we will administer a survey-embedded experiment in Japan (N=3000). We designed a conjoint experiment, which enables the respondents to evaluate the effects of multiple policy attributes on their policy support. Participants are asked to compare two randomly assigned policy designs, shown side-by-side, and then express their preferences. (A sample set of policy pairs is presented at the bottom). We consider seven policy attributes concerning climate assistance to developing countries: (1) the continent in which the country is located to provide a sense of how geography and race might affect support (Bangladesh, Cambodia, Peru, and Tanzania); (2) the types of extreme weather event this country faces (wildfire, drought, sea level rise, flooding, and typhoon); (3) the volume of climate aid Japan will provide to this country ( 0, $1 billion, $2 billion, $3 billion and $4 billion); (4) the number of climate refugees Japan should accept from this country (0, 100, 250, 450, 700, 1000); (5) economic connections: value of Japanese exports to this country (0, $1 billion, $2 billion); (6) economic dependence: Japanese imports from this country (0, $1 billion, $2 billion); and (7) political solidarity: percentage of times this country voted with Japan in the United Nations over the last 10 years (0%, 40%, and
80%).

The second question we address is whether individuals benchmark policy proposals against
existing policies? One could benchmark in relation to past performance, or one could benchmark in relation to peers. To explore if benchmarking information creates a framing effect, we divide
participants into three groups (reference category and two treatment groups) and present benchmarking information and the conjoint table to participants in treatment groups. Participants in the reference (control) group are provided no benchmarking information, and they simply answer conjoint questions. Respondents in the first treatment group are provided with information on the annual level of climate aid that the Japanese government provides and the number of refugees the Japanese government accepts every year. Respondents in the second
treatment group are provided with the same information on average aid and refugees of peer
countries, the G7.

Our survey design enables us to evaluate how policy attributes influence the public’s willingness to provide climate assistance—be it aid or accepting climate refugees—to developing countries. By comparing conjoint results across experimental groups of benchmarks, we also analyze whether and how the public’s policy preferences are influenced by framing effects: comparison with current levels of climate assistance by Japan and by peers, i.e., G7 countries.

Author

• Azusa Uji, Kyoto University