Workshop "Social welfare and climate policy", June 2021

CHIPS aims to improve welfare methodologies used to assess inequalities and risks arising from climate change. On 28 June 2021, the third CHIPS expert workshop brought together modelers and economic theorists working on issues of inequality and welfare assessment. The workshop consisted of two sessions: In the first session, the focus was on the role of inequality aversion, and disentangling the role of intra-region, inter-region and inter-generational inequality aversion. The second session was concerned with the impact of discounting on climate policy and how different welfare assessment may lead different discounting practices. Please find the full agenda here.

Beyond the presentation given by the CHIPS partners Marc Fleurbaey and Stéphane Zuber (from CNRS), we were particularly happy that three external experts accepted our invitations to present their research:

* Johannes Emmerling is Senior researcher at RFF-CMCC European Institute on Economics and the Environment. He is an expert in integrated assessment modelling, climate change and energy economics, risk and uncertainty, welfare economics and development.
* Francis Dennig is an Assistant Professor of Economics at Yale-NUS College in Singapore. His research focuses on policy questions relating to climate change and inequality, including how climate policy will affect poor people, population growth and health.
* Maddalena Ferranna is a research associate at the Harvard T.H. Chan School of Public Health. Her main research topics are welfare economics, global health and climate change economics. She is particularly interested in the development and application of methods to assess the equity implications of health and environmental interventions.

Report

Johannes Emmerling started the first session by presenting is on-going work with Pietro Andreoni, and Massimo Tavoni aiming at extending the RICE integrated assessment model to build a more regionally disaggregated model including more than 50 countries or regions. This works builds on historical inequality date and includes scenarios for future inequalities consistent with the SSPs. The presentation also included a discussion of different social welfare functions aggregating along the dimensions of income groups, space and time. The presentation concluded on the importance of inequality and redistribution for climate policy and mentioned some open questions (damage functions and the distribution of climate impacts).

Marc Fleurbaey and Stéphane Zuber then presented their work (which is part of the CHIPS project) on disentangling inequality aversion within and across regions. The paper investigates whether we could model a principle that would favor transfers within a country to transfers across countries. The theoretical response is negative: such a principle would bring inconsistencies, although we may be able to have weakened version of the principle when countries are all equal. The conclusion is that proximity preferences are hard to integrate into a social welfare function, but the paper proposes several new welfare criteria.

The first part of the workshop was concluded by a general discussion. It allowed to bring out additional questions that are relevant to the normative debate around inequalities: Can we use data to inform parameter choices, and if so what kind of data? Is the question of horizontal equity (for instance rural vs. urban) relevant, and how to take that into account? Are there other relevant kinds of inequalities (gender, race, etc.)?  

The discussion of discounting in the second half of the workshop was initiated by Francis Dennig’s talk on differentiated carbon prices (joint with D. Anthoff and J. Emmerling). The paper introduced trade in capital goods and the idea that in a second-best setting interest rates and hence discount rate cannot be equalized. As a consequence, carbon prices do not grow at the same rate in all countries. The only way out of this result is to impose time-varying Negishi weights that are arbitrary. 

In the final talk of the workshop, Maddalena Ferranna presented her work on discounting when inequality aversion and risk aversion are disentangled. Contrary to recent work using Epstein-Zin preferences to achieve this, Maddalena Ferranna studied ex post preferences consistent with the expected utility hypothesis and exhibiting a consistent role for increased inequality aversion. In particular, she showed that the social cost of carbon should be larger for more risk averse societies, and also that it increases when the difference between inequality aversion and risk aversion increases. The results were illustrated in a simple climate-economy model.

In the final open discussion round, the participants discussed to what extend complex social welfare function should be implemented in already complex integrated assessment models. Some of the work presented in the workshop show that we can include some additional welfare relevant dimensions and issues. But there are many difficulties on the empirical side to calibrate these phenomena and use meaningful welfare parameters. There was however a general agreement that these questions still offer interesting avenues for future research.  

We thank all of the presenters, discussants, and participants for a very fruitful seminar.