Conference Agenda

In this paper we study an under-explored implication of population aging, i.e., its effect on environmental quality and environment-related individual attitudes. In doing so we propose a novel classification of environmental outcomes: action-requiring environmental outcomes and nature-concerning environmental outcomes. The borderline difference between those two types of outcomes lies in the level of individual engagement and effort required to fulfill them. Using data on a panel of countries the 1995-2018 period, our findings reveal that whilst ageing has a pro-ecological impact on nature-concerning environmental outcomes, it has an anti- ecological effect on action-requiring ones. Using survey data analysis, we also demonstrated that population ageing has a negative and statistically significant on respondents’ engagement for the environment.

Past studies on the nexus between equity and environmental throughput suggests a concave relationship between household income and the emissions embodied in household consumption. If this is correct, any redistribution from richer households to poorer households would raise the aggregate emissions of a society, ceteris paribus. The empirical evidence for this concave relationship is based on data from household surveys, which suffer from undercoverage and underreporting of high-income households (“missing rich”). The income share of the missing rich, that is not accounted for in the household surveys, can be derived from national accounts, as well as data from tax registries. Existing studies either rely on those imprecise data or correct for the income of rich households through an income-emission elasticity. The shortcoming of this method is that the elasticity is derived from the uncorrected household data which leads to biased results.

The study at hand addresses this problem through an integrated approach which consists of three steps: First, we correct the income distribution of household surveys for the missing rich by combining them with tax data. Secondly, we predict the consumption of the high-income households for which no survey data are available with machine learning techniques. Finally, we associate carbon emissions with the consumption of all households through input-output tables.

We apply this approach to the case of Ecuador and find that measures of income inequality are higher when correcting for the missing rich households with exhaustive microdata. The aggregate income-emission elasticity is lower than in previous studies. Finally, our findings challenge the prevailing idea of a concave income-emission relationship, and therefore the existence of a social dilemma between inequality and emission reduction policies.

Migration is often perceived as a key tool to adapt to climate change. However, in developing countries, many households face liquidity constraints that prevent them from migrating when climate shocks occur. This generates a spatial misallocation of labor that impedes economic development. The goal of this paper is to quantify the cost of these misallocations. To do this, we focus on Sub-Saharan Africa, a region heavily concerned by the impacts of global warming and that displays large fertility rates. Using reduced-form estimations, we start by documenting that droughts cause out-migration in African districts except in the poorest ones, where we find no effect. We build a quantitative spatial model of migration and trade combined with satellite and census data to analyze this result's aggregate and future implications. We find that by 2050, 30 million potential migrants will be blocked by climate shocks, representing a strong welfare loss for the African economy. Our results further highlight important heterogeneity in the impacts of climate change across and within countries of the region.

Recent agricultural policy research draws attention to the importance of continuous conservation tillage, as many environmental benefits of conservation tillage are fully realized only when conservation tillage is used continuously over the years. However, little is known about the dynamics of farmer's tillage choices. Panel tillage data are scarce and incomplete. This study presents a method that relies entirely on publicly available tillage shares data to infer the probabilities of rotational and continuous conservation tillage. Using the framework of first-order Markov chains, we model tillage dynamics by estimating the probabilities of transition from one tillage-crop combination to another tillage-crop combination with spatially aggregated data. We use the combination of Quadratic Programming and Generalized Maximum and Cross-Entropy to infer the transition probabilities for the period of 1992-2008 for the state of Iowa, U.S. The results show that approximately one million acres (4000 km squared) of corn and soybeans moved away from continuous conservation tillage to greater tillage intensity practices in 2001-2008 period when corn price increased sharply. Spatially, more acreage in the southern and eastern Iowa, where soils are of lower productivity and more likely to be classified as highly erodible land, were taken out of continuous conservation tillage practice between 2001 and 2008 when compared with the rest of the state. Our findings imply that the shift toward corn monoculture in high crop prices periods might have increased soil erosion, nutrient runoff, and CO2 emissions from the soils.