One way researchers analyze real-world, sustainability questions is by utilizing computational models.  These models use data to derive the relationships between economic, social, and natural systems and quantify the potential outcomes of public policies and behavioral changes.  Sustainability models have provided useful insights on a wide range of societal challenges,  such as energy transition, climate mitigation, and environmental protection.  

These models have played an important role in supporting environmental decision making and problem solving, but it can be difficult to codify the complexities of human systems and institutional rules  that influence sustainability outcomes.  Collaborating with researchers at the University of California San Diego and Delft University, C-PREE faculty member Wei Peng co-authored a study that uses modeling experiments to demonstrate the importance of institutional factors in computational sustainability models.  

Defined by the authors as the rules that constrain human behavior, institutions can both formally and informally effect sustainability outcomes, whether it is through officially-recognized policy mechanisms or unwritten rules that are dictated by social norms and beliefs.  To date, models are typically limited to simplified representations of institutions that are easier to codify, such as emission target policies or changes to technology costs.  But as Peng and her colleagues argue, these overly simplified assumptions about institutions  can limit the usefulness of models and can lead to biased conclusions that do not properly reflect socio-political realities .  

“Most energy models have been built around the physical rules governing the technologies,” explains Peng. “Modeling institutions is much harder. Some institutional factors can be represented in models through technology cost and deployment, such as electricity market rules. Others are deeply uncertain and complex, such as social norms, which makes quantitative modeling quite challenging.  In essence, the right model choice depends on what institutional factor one hopes to capture, and that's exactly the motivation of this study.”

In their study, the researchers experiment with codifying institutional factors with three types of models commonly used in sustainability research: Integrated assessment models (IAM), engineering-economic optimization (EEO) models, and agent-based models (ABM).  IAMs are typically used to analyze global climate policies and regional emissions, EEOs are used to analyze the design and planning of physical infrastructure, and ABMs focus on the interactions and behaviors of “rational actors” with the environment.  

Using U.S. climate and energy data, the researchers tested the effect of formal institutions by incorporating heterogeneous carbon prices in the IAM and EEO models.  The effect of informal institutions was tested by incorporating social norms and attitudes toward sustainable behavior in an ABM looking at solar power installation in the Netherlands.  The authors found that even the inclusion of one type of institution in these computational models had significant effects on sustainability outcomes.  

As quoted from UC San Diego Today, lead author Michael Davidson discusses the results in more detail.  

“Our report examines three common modeling approaches and shows that omitting human-related institutions can alter the cost and outcomes of sustainability models,” explains Michael Davidson, an assistant professor at the School of Global Policy and Strategy and the Jacobs School of Engineering. “Our findings reveal that incorporating only one of many institutions in computational models delivers measurable effects on sustainability outcomes, such as emission reductions of 8-11% and costs nearly 6% higher nationwide.”

Given the results of the study, the incorporation of institutional changes might be the next frontier for researchers looking to improve the accuracy of sustainability models.  Peng talks about how this type of research could shape the future of sustainability model specification and selection.  

“Each model has strengths and limitations,” says Peng. “This is an important context when choosing which model to use in the first place. I think the next frontier is to start exploring ways to endogenize institutional changes. Adding relevant processes within each model can be useful. Coupling different models is also a promising strategy to explore.”

The paper, “Simulating institutional heterogeneity in sustainability science,” was co-authored by Michael R. Davidson (School of Global Policy and Strategy, University of California San Diego), Tatiana Filatova (Department of Multi Actor Systems, Faculty of Technology, Policy and Management, Delft University of Technology), Wei Peng (School of Public and International Affairs, Andlinger Center for Energy and the Environment, Princeton University), Liz Verbeek (Department of Multi Actor Systems, Faculty of Technology, Policy and Management, Delft University of Technology), and Fikri Kucuksayacigil (School of Global Policy and Strategy, University of California San Diego). The paper appeared in Proceedings of the National Academy of Sciences of the United States on February 15th, 2024.