Sub-theme 5. Modelling and multi-methods approaches in polycentric commons systems
Computational Social Science and the study of the commons
Which methods can we use beyond the standard surveys, interviews and case-studies to systematically capture mechanisms on the development and impact of diverse forms of institutions for collective action? The study of the commons and other institutions for collective action in itself is highly interdisciplinary and increasingly attracts the application of computational methodologies in the fields of amongst others history, sociology, economics and management studies. This panel demonstrates how the use of computational techniques can contribute to the analysis of institutions for collective action. The analysis of artificial societies through agent-based models, the creative application of experimental techniques and the analysis of digital trace data from online social media can bring the study of the commons to a next level. Can agent-based models go beyond simulation and tell us about the future of the commons? How can adaptations of experimental methods help us better understand how community enterprises impact their members and society? Next to impact in the physical world, institutions for collective action have found their way to the virtual world of digital platforms. However, can their positive societal influence reach beyond the offline world into online realities? The papers in this panel present interdisciplinary computational methods to analyse diverse institutions for collective action.
1. How to improve the management of transboundary commons? scenarios from an agent-based model simulating the effects of boundaries and communication in the case of crab fishery in the Amazon mangroves
1Centro de Desenvolvimento Sustentável. Universidade de Brasília (CDS – UnB), Brazil, 2Unité Mixte de Recherche Savoirs, Environnement et Sociétés (UMR SENS – CIRAD), France
To increase the chances of successful management of a common resource, communication among resource users is fundamental to align management strategies, as well as clear spatial boundaries around resource systems to avoid open access. Nonetheless, specific challenges arise in the case of transboundary commons which cannot be contained in areas delimited by human-defined boundaries. Crab fishery practiced in the mangrove forests of the Amazon coastal zone in Brazil deals with a typical transboundary resource; Marine Extractive Reserves have been created to organize the use of ecological resources among its beneficiaries. By analyzing this socioecological system, we observed that although setting boundaries is challenging, communication among fishers of the same community helps them decide where to fish. Based on our field results and on literature on the commons related to communication and boundaries, we built an agent-based model called SpaceCoord to simulate different forms of communication between users of a transboundary common resource, and different ways of delimiting resource systems, in order to test their effects on the resource. The results of the simulations show that the level of the resource increased in the scenario without boundaries and without communication, and decreased when the resource system was divided equally among resource users and when they communicated about the address of the fishing grounds they visited, revealing that communication and boundaries can increase the efficiency of resource user to catch more resources in the least amount of time. So, we underline the importance of encouraging communication about the resource conservation.
2. Flood-risk management in St.Maarten: building up insights from agent-based models and network analysis
Delft University of Technology, Netherlands1, 2Syracuse University, US
Governments of nations facing flood risk have dedicated substantial effort to investigating ways to mitigate, plan for, respond to, and recover from flood events. Institutions can be highly effective in regulating flood risk, but, as in other domains, their efficacy depends on the extent to which they are complied with. We formally assess compliance with flood risk management institutions through an in-depth case study in the Caribbean island of Sint Maarten. We aim to study how three different policies compare in the complex landscape of flood management: 1) distance from the beach, 2) 20cm-elevation for all buildings, 3) various elevations depending on the zone. To this aim, we use the IAD framework to build an agent-based model. This model has real-time interaction with flood models and primarily focuses on how institutions influence actor behaviour at the level of inhabitants. The results show that the existing 20cm-elevation policy is more effective than the other two policies if the rules are properly enforced. To cross-validate results, we complement the model with an independent network analysis to study whether the rules-in-form align with the rules-in-use. The network analysis shows that property owners largely follow the rules. However, those responsible for the enforcement mostly neglect the rules-in-form for various reasons, e.g., not seeing the actual impact or having a stake in policies not being implemented. This multi-method approach proved to be highly insightful in identifying the actual problem and the exact situation in which this problem arises.
3. Tools for Designing Robust Systems to Environmental Shock: the North Peruvian Coast Case
1Universidad Del Pacífico, Peru, 2Arizona State University, USA
The Peruvian north coast has been threatened by disastrous flood events during the past centuries, but they are expected to be more frequent and intense due to climate change. Extreme flood events damage physical infrastructure (e.g., reservoirs and canals) through which, for example, farmers are able to appropriate water to irrigate their crops. In arid regions like the north coast of Peru, agricultural activity depends entirely on physical infrastructure. Reservoirs capture runoff from the mountains, smoothing flow variations and saving surplus water for later use. Physical infrastructure is also affected by how it is maintained, and maintenance is affected by how common dilemmas (mainly appropriation and provisioning dilemmas) are solved. We developed a model for a basin located on the north Peruvian coast 1) to understand the core dynamics of the systems concerning the relationship between public infrastructure and collective action, 2) to understand the system’s robustness to extreme flood events, and 3) to explore potential interventions to increase its robustness. We calibrated the model to predict behaviors and results. The model revealed that collective action is needed. Given governmental limited actions, the hard and public infrastructure is damaged in El Niño episodes, and significant human suffering is experienced in the basin. An improvement in farmers’ collective action can change those results. The model shows how in the current situation, it is more effective to invest, if only investing one isolated time instead of regularly, in rules enforcement, than in the improvement of the physical infrastructure.
4. Using Player Types to Understand Cooperative Behaviour under Economic and Sociocultural Heterogeneity in Common-Pool Resources: Evidence from lab experiments and agent-based models
Rotterdam School of Management, Erasmus University Rotterdam, The Netherlands
This study aims to use player type classification to explain cooperative outcomes of two common-pool resource [CPR] experiments on the effect of economic and sociocultural heterogeneity on cooperation. The experiments contain 344 subjects from the United Kingdom and the Netherlands and 144 subjects from India. Multilevel regression, ordinal logistic regression, linear conditional-contribution profiles [LCPs] and agent-based models [ABMs] are used to analyse and replicate experimental outcomes on the micro- and macro-level. This paper concludes that the effect of a combination of economic and sociocultural heterogeneity on cooperation in CPRs depends on incompatibility between groups, which affects the percentage of conditional versus unconditional cooperators. Player type classification based on LCP scores shows that experimental data can be interpreted with player types, and that heterogeneity influences the player type distribution. Simple ABMs informed by experimental data are shown to be capable of replicating the main experimental outcomes, suggesting ABMs to be valuable in experimental research.
5. Investigating energy security in energy communities: Combining agent-based modelling and statistical analysis
Copernicus Institute of Sustainable Development, Utrecht University, Netherlands
Energy security is one of the most studied topics in energy-related literature, and various concepts and dimensions are introduced for energy security assessments. However, the literature is missing a study to prioritise these dimensions, as they cannot always be simultaneously addressed. Therefore, this study is focused on investigating the importance and prioritisation of energy security dimensions. The study narrowed down its focus on community energy systems as alternative, collective and decentralised energy systems, which are gaining momentum in the energy transition context. For the first time, a detailed agent-based model was developed to capture the energy security of energy communities, including the following dimensions; availability, infrastructure, prices, environment, governance, efficiency, and societal effects. The results articulate that energy communities are capable of contributing to the energy security of individual households, and they have substantial potential in CO2 emissions reduction (60% on average) while being affordable in the long run. In addition, the importance of project leadership (particularly regarding the municipality) concerning energy security performances was explored. In the next step, along with further institutional and behavioural insights, a statistical analysis was performed on the results to prioritise the importance of energy security dimensions. The statistical analysis revealed that infrastructure and governance are the most impactful dimensions for energy security assessments. Energy prices was among the least influential dimensions in determining energy security performances. Lastly, the study explored various existing energy security concepts (with their unique set of dimensions) to elaborate on their influence.