Causal Relationships in Coupled Human-Natural Systems

My early training was in ecology. After graduate training in economics and statistics, I became interested in bringing insights from those fields back to the natural sciences. Economics and biostatistics have developed a range of designs for drawing conclusions from non-experimental data. These designs can offer new ways to investigate causal questions that natural scientists have long been asking.

How do changes in species diversity affect ecosystem productivity in a world in which human impacts are ubiquitous? How do changes in roads and other impervious surfaces affect flood magnitudes? How do ecosystems such as forests moderate the effects of extreme weather events?

Answering questions like these requires more than documenting patterns: it requires estimating what would have happened under different conditions. I work with ecologists and hydrologists to apply causal-inference designs to these questions, with the aim of advancing scientific understanding and informing policy.

Related Publications

Clarifying the Effect of Biodiversity on Productivity in Natural Ecosystems with Longitudinal Data and Methods for Causal Inference. Dee, L, PJ Ferraro et al. 2023. Nature Communications

Causal Effect of Impervious Surface on Annual Flood Magnitude for the United States.
Blum, A., PJ Ferraro, S Archfield, & K Ryberg. 2020. Geophysical Research Letters.

Causal Inference in Coupled Human and Natural Systems
Ferraro, PJ, J Sanchirico, & M Smith. 2019. Proceedings of the National Academy of Sciences.
 

The Effects of Environmental Programs on Human Welfare and the Environment

Evidence-based environmental policy and programs require credible evidence about what works, what doesn’t, and why. To generate this evidence, my collaborators and I use research designs and statistical methods to estimate the effects of environmental interventions, both regulatory and voluntary, and to elucidate the mechanisms through which they affect human welfare and the environment.

Should we expect that listing a species under the Endangered Species Act will make the species better off than had it not been listed at all? If so, by how much will it be better off? Should we expect national parks and reserves to slow habitat conversion and benefit local communities? If so, through what mechanisms do those effects arise and can we redesign park systems to do more for both conservation and local welfare?

These questions are difficult to answer because programs are rarely implemented with evaluation in mind. Weak research designs can produce misleading conclusions. Alongside conducting research, I work to help researchers across disciplines use stronger designs. I also work to encourage environmental funders to create incentives for programs to use these designs to learn, including from programs that fail to meet their goals.

Related Publications

Through What Mechanisms Do Protected Areas Affect Environmental and Social Outcomes?
Ferraro, PJ, & M Hanauer. 2015. Philosophical Transactions of the Royal Society 370.

Conditions Associated with Protected Area Success in Conservation and Poverty Reduction.
Ferraro, PJ, M Hanauer, & K Sims. 2011. Proceedings of the National Academy of Sciences 108(34): 13913-13918. 

Designing Better Environmental Programs with Behavioral Science and Experimental Designs

This research is grounded in two beliefs:

1. Environmental problems are human behavior problems. Yet environmental programs and policies are typically designed and managed by natural scientists, engineers, and lawyers, not behavioral scientists. This insight lead me to switch from ecology to economics and, more broadly, to the behavioral sciences. To help environmental scientists and program managers design science-based solutions with human behavior in mind, I directed or co-directed two centers, CBEAR and EPIC.

2. Every promising idea is an opportunity to learn. When we implement an idea as something worth trying, rather than as a hypothesis to test, we miss a chance to build evidence. Scientific principles can help us understand not only the natural environment, but also how environmental programs perform and how people respond to them. For more than two decades, I have encouraged government and non-governmental institutions to embed science-based learning into program implementation. This “implementation science” was central to the missions of CBEAR and EPIC. It requires academics and program managers to collaborate and deliberately design programs, or changes to programs, to generate evidence.

I am especially interested in understudied questions about behavior change. How can we influence organizations, businesses and other groups where decisions are made by groups of individuals? For example, can peer comparisons encourage polluters to reduce their pollution beyond what they are legally required to do? How can we induce persistent behavioral change with short-term interventions? Can short-term incentives for using environmentally friendly technologies or protecting forests encourage long-term environmental and economic impacts?

Related Publications

Create a Culture of Experiments in Environmental Programs.
Ferraro, PJ et al. 2023. Science

Behavioral Biases among Producers: Experimental evidence of anchoring in procurement auctions.
Ferraro, PJ, K Messer, P Shukla, C Weigel. 2022. The Review of Economics & Statistics. Preprint. Data and Code

Synthesizing Evidence in Sustainability Science through Harmonized Experiments: Community monitoring in common pool resources. Ferraro, PJ, Agrawal, A. 2021. Proceedings of the National Academy of Sciences. Special Issue

The Effects of Anti-Poverty Programs on the Environment

Two of the great global challenges of the 21st century are to reduce poverty and slow global environmental change, yet solutions to these challenges are often framed as conflicting with each other. Much of my earlier research examined whether actions to slow global environmental change increase or reduce poverty. I am increasingly interested in the reverse quesion: Do actions to reduce poverty affect the environment and, if so, through what mechanisms.? Very little is known about these causal relationships.

For example, can targeted anti-poverty programs achieve their goals without exacerbating ecosystem degradation? We don’t have much evidence to answer that question. This gap in the evidence base is important because expenditures on anti-poverty programs, particularly in low- and middle-income countries, dwarf expenditures on environmental programs.

Related Publication

Conditional Cash Transfers to Alleviate Poverty also Reduced Deforestation in Indonesia. Ferraro, PJ & R. Simorangkir. 2020. Science Advances.