Lubell et al., 2010. Ecology of games water management.

Ref: Lubell, M., Robins, G., & Wang, P. (2010). Policy coordination in an ecology of water management games. 2010 Annual Meeting of the Midwest Political Science Association, Chicago, IL.

Blends perspectives of Long (1958) and Scharpf (1997): ecology of games (EG) and actor-centered institutionalism.

Actor hypothesis: Politically powerful players coordinate policy activities by participating (and possibly creating) in many different types of policy games, and becoming “popular” with other actors, who want to participate in the same venues as the politically powerful.

Institutions hypothesis: Political interaction is driven by the interactive strategies of purposive actors operating within institutional settings, that at the same time enable and constrain these strategies.

Institutions consist of the formal and informal rules that structure human interaction by defining the set of actions that may be chosen, and the payoffs for those actions. A problem of cooperation: institutions evolve to reduce the transaction costs of economic exchange.

Actors must deal with two simultaneous problem:

1. Finding and implementing mutually beneficial policies (efficiency)
2. Bargaining over the distribution of mutual benefits (distribution)

Coordination in an EG involves both efficiency and political power, because actors will use political resources in an attempt to capture the greatest share of the gains over cooperation.

Political conflict and bargaining over efficiency gains is one source of transaction costs in the EG. Effective coordination mechanisms would minimize the transaction costs of searching for mutually beneficial solutions, bargaining over distribution, and monitoring & enforcing the resulting constellation of decisions and agreements.

Analytical approach: exponential random graph models (ERGM) for bipartite network structures (Wang, 2009).

Empirical system: SF Bay Water Management.

Collective action problems: Water quality, water supply, climate change, biodiversity

Actors: Federal, state, and local government agencies, special districts, environmental groups, economic interest groups, researchers.

Ongoing policy games: Collaborative partnerships, regulatory processes (e.g., Total Maximum Daily Load planning under Clean Water Act, advisory commissions to government organizations, associations of interest groups (e.g., collaborative institutions such as Integrated Regional Water Management)).

Adaptation of EG Framework

Six interrelated concepts:

• Policy issues
• Policy actors
• Policy institutions
• Policy games
• Policy arenas
• Time

Policy issues: Some type of collective action problem such as water pollution, air pollution, traffic congestion, loss of biodiversity.

Strategic structure is identical to traditional game theory: payoffs from using resources are interdependent, actors ignore the social costs of their decisions, and equilibrium outcomes are often inefficient. Added complication of EG framework: Issues may be interconnected, so decisions made in the context of one issue may directly affect payoffs in other issues.

Policy institutions: Collective-choice settings where actors jointly make decisions about the “operational” rules governing individual issues. They typically have jurisdiction over multiple issues at a given time, and thus policy issues are linked to multiple games. In the real world, these are often referred to as “planning processes” that shape implementation of specific resource management activities.

Policy actors: Have a stake in outcomes of collective-choice and the resulting rules governing specific issues. Can represent individual resource users (e.g., fisherman, farmers) or political actors (e.g., agency officials, interest groups, elected officials). Actors participate in policy games with jurisdiction over issues they care about, and also form networks with other actors in order to gain key political resources (information, credibility, and political influence).

Policy games: Defined by the constellation of actors, institutions, and issues that are at hand in a particular decision space. Games are not the same as institutions: linkages between behaviors and outcomes described by a game is a function of institutional rules.

Policy arenas: Territorially defined subsystems that encompass multiple issues, multiple institutions, and multiple actors. E.g., what are the most important policy games in a particular territory. This is really the ecology of games.

Time: Things can change over time. Changes can be endogenously driven by the actors as they participate in different games, try out different strategies, engage in policy learning, and even create new institutions. Change can also come exogenously, e.g. according to the dynamics of the underlying resources, or from higher-level institutions.

Key question: Does cooperation evolve, and how robust are cooperative interactions to exogenous changes?

Actors as coordinators: Actors are characterized by their capabilities, preferences, and perceptions. Capabilities are a function of the resources an actor commands to influence outcomes in ways that are consistent with preferences.

Institutions as coordinators: Institutions have rules that constrain the strategies of actors and define the link between strategies and payoffs. Different institutional arrangements have more or less ability to solve different types of collective-action problems. E.g., they may influence the transaction costs for searching for mutually beneficial solutions, bargaining, and monitoring/enforcing agreements.

Collaborative institutions emphasize specific types of rules: inclusive participation of multiple actors, consensus decision-making, integration of scientific information, voluntary implementation, and place-based activities.

No a priori expectation of whether actors or institutions are the primary coordinators within the EG.

Network Representation and Hypotheses

The EG is represented as a bipartite network where each policy actor participates in one or more games.

The assumption is the actors are choosing which games to participate in given the current set of available games, although the creation of new games is possible.

Network activity

Network activity (or popularity) is defined as the number of ties a node has, and is called the degree of a node.

Network centralization and degree dispersion

Networks that exhibit high centralization are likely to have more influence over decisions in the rest of the system. Network centralization and degree dispersion is represented by “two-star” configurations, where a node has connections to two other nodes. A node with degree d is involved in d(d-1)/2 distinct two-stars [note: I’m not sure this is true]. For a given level of network activity, the presence of more two-stars indicates a more centralized network structure based around a smaller number of high degree nodes.

Network closure and clustering

For unipartite networks, network closure consists of tight, circular network structures. For bipartite networks, the simplest closed structure is the four-cycle (C4), which occurs when actors of the same type are tied to the same multiple institutions and when institutions of the same type are tied to the same actors. The clustering coefficient for bipartite networks is defined (Wang et al., 2009) as 4 times the number of 4-cycles divided by the number of 3-paths (L3).

Some data from the study

They analyzed actors and institutions involved with water management in the SF Bay area. This was done with survey data collected by questioning a number of the involved actors. After the network analysis, here are some interesting findings.

• For actors, the mode and median number of connections is one, with an average degree of 3.09.
• For institutions, the mode degree is 5, with a median of 7 and an average of 10.33.
• Of the types analyzed, Federal and state agencies are the most central actors, and collaborative partnerships are the most central institutions.

Used exponential random graph models (ERGM) to analyze the data against models.

The most accurate was the one probabilistic model which controlled for attribute activity. The observed probability of ties was fixed and then probabilistically distributed across each graph in the distribution. There is also the addition of parameters which control for different types of actors and institutions. This generated an almost perfect match for the number of ties for each type, but also generated significantly fewer two-stars and 4-cycles relative to the observed data. The main explanation for this result is linked to the reputation of different types of actors and institutions for solving problems.

Two critical questions not answered:

1. It is unclear whether actors are participating in institutions to solve collective-action problems or to exert political power to achieve policy preferences. E.g., some issues may be zero-sum games, where actors are participating in institutions in order to shift policies in their favor at the expense of other actors.
2. A different perspective on policy effectiveness. There is no single coordinating entity. Also, the amount of redundancy may in fact be beneficial in such a complex adaptive system (and one involving uncertainty). Some redundancy may be beneficial as it allows actors and institutions to experiment and make mistakes without destroying the entire system. Network theorists have pointed out the adaptive capacity of other systems with “fat-tail” distributions – i.e. a few high-degree nodes and many low-degree nodes. E.g., the central nervous system can survive many small errors but is vulnerable to attacks on central, high-degree nodes. Similar processes may be at play in the ecology of games.