Using Ostrom’s DPs as Fuzzy Sets to Analyse How Water Policies Challenge Community-Based Water Governance in Colombia

Ostrom’s design principles have been broadly used to analyse the governance of common pool resources. However, they are normally assessed as present or absent. We challenge this assumption by considering them as fuzzy sets where membership scores range from 0 to 1, because, in real situations, the design principles can be present at a certain level. We define categories to assess the level of membership and apply it to a single case study analysing how changes in water policy can affect the community-based management of the Water User Association of Mondomo (Colombia). In rural areas of Colombia, most water and sanitation services are provided by water user associations, wherein civil society has developed governance systems based on active citizen involvement and community-based management. Some of these associations have been operating for decades and are essential pillars of the local social fabric. However, recent changes in the country’s policy and legal framework threaten these long-lasting governance systems. The results show that most of the design principles would suffer important changes that undermine the governance system. Essential principles for sustainable community-based governance, such as the congruence of the rules with local conditions, the local monitoring and sanctioning capacities, the internal conflict-resolution mechanisms and the recognition of the rights to organize, are dramatically reduced after the policy implementation.


Introduction
Water is at the core of sustainable development, as recognised by Sustainable Development Goal 6 [1].Water resources, and the range of services they provide, underpin poverty reduction, economic growth and environmental sustainability.However, in 2015, an estimated 663 million people were still using unimproved water sources or surface water, and eight out of 10 people who lack improved drinking water sources live in rural areas [2].Although water policies regulate the provision, rights of access and use, and distribution and disposal of water resources, developing country governments often lack the means to effectively provide water and sanitation, especially in rural areas.This situation has forced citizens to develop alternative governance systems to secure their access to water [3].
Governance has been defined as a way of collectively managing social relations to define or allocate possible resources [4,5] between public or private actors to obtain mutual gains while also

The Institutional Analysis and Development Framework (IAD)
Institutions are broadly defined as the prescriptions and procedures that humans use to organize all forms of interactions [27].The IAD framework was created and refined by Ostrom and her colleagues [28,29] as a systematic and consistent method of analysing the deeper structures that constitute specific action situations where human behaviour is central to understanding contextual changes [26].The framework is a conceptual map; it integrates different parts or sections that need to be approached to understand how institutions change based on potential and real changes in context (biophysical, economic, social, legal, etc.) (Figure 1).This WUA has more than 60 years of history and is considered a successful case of democratic water governance in Colombia.It is also considered an example of robust collective action following Ostrom's approach [16].To describe Mondomo's WUA, we used the general variables included in the Institutional Analysis and Development framework (IAD) proposed by Ostrom [26].To analyse how the new regulation may affect the robustness of the governance system of Mondomo's WUA, we assessed how the membership scores of the DPs change.
This study further contributes to the literature by: (1) considering the DPs as fuzzy sets and not as dichotomous categories, as well as proposing a transparent and reliable protocol to establish categories to assign membership scores that can be of high interest in the analysis of other common pool resources (CPR) governance systems; (2) analysing the influence of the new Colombian legal framework on the governance systems of traditional WUAs; and (3) using the DPs to analyse water provision and distribution systems (there is an extensive literature using them to analyse irrigation systems but not these water systems).By provision and distribution systems, we refer to the facilities used to supply water from its source to the point of usage.Given the relevance of WUAs in Colombia, this paper supports the advancement of knowledge.

The Institutional Analysis and Development Framework (IAD)
Institutions are broadly defined as the prescriptions and procedures that humans use to organize all forms of interactions [27].The IAD framework was created and refined by Ostrom and her colleagues [28,29] as a systematic and consistent method of analysing the deeper structures that constitute specific action situations where human behaviour is central to understanding contextual changes [26].The framework is a conceptual map; it integrates different parts or sections that need to be approached to understand how institutions change based on potential and real changes in context (biophysical, economic, social, legal, etc.) (Figure 1).A framework for institutional analysis [26].
The focal unit of analysis is the "action arena", which is the social space where participants interact.It is composed of an action situation and the participants in that situation.Action arena can be considered a set of dependent variables affected by three clusters of exogenous variables: (1) the attributes of the biophysical or material conditions that are acted upon in these arenas; (2) the structure and defining conditions of the community within which the action arena is placed; and (3) the rules used by participants to regulate their relationships [29].The IAD framework intends to explore the different conditions that play a role in institutional change as well as the possible changes that different institutional configurations might produce in the action arena and in the exogenous variables.Finally, actors' interactions with other actors and with the exogenous attributes produce different outcomes that can be analysed by different evaluation criteria, thus producing outcomes that reinforce complex and changing behaviours [29].Therefore, there is no single way to approach institutional change; such change can be defined based on research objectives or based on the disciplines according to which one is attempting to understand the conceptualization of actions [26].The focal unit of analysis is the "action arena", which is the social space where participants interact.It is composed of an action situation and the participants in that situation.Action arena can be considered a set of dependent variables affected by three clusters of exogenous variables: (1) the attributes of the biophysical or material conditions that are acted upon in these arenas; (2) the structure and defining conditions of the community within which the action arena is placed; and (3) the rules used by participants to regulate their relationships [29].The IAD framework intends to explore the different conditions that play a role in institutional change as well as the possible changes that different institutional configurations might produce in the action arena and in the exogenous variables.Finally, actors' interactions with other actors and with the exogenous attributes produce different outcomes that can be analysed by different evaluation criteria, thus producing outcomes that reinforce complex Water 2017, 9, 535 4 of 20 and changing behaviours [29].Therefore, there is no single way to approach institutional change; such change can be defined based on research objectives or based on the disciplines according to which one is attempting to understand the conceptualization of actions [26].

Ostrom's Design Principles as Fuzzy Sets
After extensive study of long-lasting CPR systems, Ostrom enumerated eight design principles that characterized their governance systems.These principles enabled resource users to engage in collective actions that led to sustainable management [16].These DPs were present in long-lasting systems but were absent in those that collapsed; thus, she proposed using the DPs to characterize successful self-governance strategies.The DPs were defined as follows: (1) Clearly defined boundaries: the boundaries of the resource system (e.g., irrigation system or fishery) and the individuals or households with rights to harvest resource units are clearly defined; (2) Proportional equivalence between benefits and costs: the rules specifying the amount of resources that a user is allocated are related to local conditions and to rules requiring labour, materials, and/or monetary inputs; (3) Collective-choice arrangements: many of the individuals affected by harvesting and protection rules are included in the group that can modify these rules; (4) Monitoring: monitors, who actively audit biophysical conditions and user behaviour, are at least partially accountable to the users and/or are the users themselves; (5) Graduated sanctions: users who violate rules-in-use are likely to receive graduated sanctions, depending on the seriousness and context of the offense, from other users, from officials accountable to these users, or from both; (6) Conflict-resolution mechanisms: users and their officials have rapid access to low-cost, local action arenas to resolve conflicts among users or between users and officials; (7) Minimal recognition of rights to organize: the rights of users to devise their own institutions are not challenged by external governmental authorities, and users have long-term tenure rights to the resource; (8) Nested enterprises: appropriation, provision, monitoring, enforcement, conflict resolution, and governance activities are organized in multiple layers of nested enterprises.This last DP applies to resources that are parts of larger systems.
Cox et al. [30] with the analysis of approximately 100 case studies, provided evidence that the design principles hold up when challenged with data and suggested a revision of the DPs, splitting DP1, DP2 and DP4 into two principles each, which led to 11 DPs (Table 1).We analysed this 11 DPs.In this research, we consider the DPs as fuzzy sets, since we assume that they are not just present or absent, but can have a gradation of presence.Fuzzy sets can help social scientists conceptualize social and political phenomena as sets with imprecise boundaries between membership and non-membership.Crisp sets perceive cases only as members or non-members of a set.The problem is that many core concepts in social research are best understood as graded sets [31].Fuzzy sets are not ordinal scales or mere rankings of categories, but each degree represents a different empirical situation [25].The basic idea behind fuzzy sets is to permit the scaling of membership scores and thus allow partial or fuzzy membership.A membership score of 1 indicates full membership in a set, scores close to 1 (e.g., 0.8 or 0.9) indicate strong but not quite full membership in a set, scores less than 0.5 but greater than 0 (e.g., 0.2 and 0.3) indicate that objects are more "out" than "in" but still weak members of a set, and a score of 0 indicates full non-membership in a set.Thus, fuzzy sets combine qualitative and quantitative assessment: 1 and 0 are qualitative assignments ("fully in" and "fully out", respectively); values between 0 and 1 indicate partial membership.The 0.5 score is also qualitatively anchored, for it indicates the point of maximum ambiguity (fuzziness) in the assessment of whether a case is more "in" or "out" of a set [32].Fuzzy sets require that researchers establish a very close correspondence between fuzzy membership scores and the concepts they represent, as well as a careful calibration of the fuzzy scores that reflect the degree of membership [25].

The New Water Regulatory Framework in Colombia
Colombia's WUAs actually feel threatened by a regulation in process of implementation, proposed in 2014 by the National Council of Economic and Social Policy [19] and partially ratified by national law number 1898 [33].This framework has as its core objective, the promotion of access to drinking water and basic sanitation in Colombia's rural areas by improving the general conditions of access to good quality water for all rural inhabitants.Its objectives are: (1) strengthening the institutional capacity of the state to intervene in rural areas; (2) promoting the structuring of sustainable schemes for drinking water provision and sanitation in rural areas; (3) investing in infrastructures; and (4) promoting effective practices of sanitary and environmental management.The regulation will be implemented in two phases during 10 years, starting in 2014 with its presentation and discussion with WUA´s national representatives.
At the local level, the regulation is perceived as a water privatization process [24] that will affect the autonomy and governance of WUAs.The emphasis of the new regulation in the implementation of activities through national government agencies limit the role of local communities and organized WUA's in implementing it [34].Several outcomes of the policy are considered as threats by WUAs: (1) it opens options to increase the number and type of users with access to the WUAs' water facilities; (2) it introduces new priorities in water access displacing household needs in favor of other economic sectors; (3) it transfers the control and supervision of water quality to national agencies, undermining the management capacities of WUAs and probably increasing water prices; (4) it challenges community-based governance systems based on solidarity, reciprocity and direct democracy; (5) it reduces the legitimacy of social participation in water management systems; and (6) it only affects WUA's and aqueducts that serve rural villages and towns, but does not guarantee services to isolated areas.The Colombian Network of Community Pipelines has recently contested this regulation by proposing to the Government a draft regulation that preserve the right to the community-based management of water and maintain the WUA's autonomy, preventing what they called as the "privatization" of water supply to rural communities [35].

Case Study Selection
Mondomo village locates in the Santander de Quilichao municipality (Cauca Department, Colombia).The village is located at 1350 m a.s.l. and is mainly composed of Afro-Colombian, indigenous and mestizo inhabitants.Mestizos and Afro-Colombians mainly locate in Mondomo's town, while indigenous communities locate in the upper areas of the Mondomo's river watershead.
Water 2017, 9, 535 6 of 20 Mondomo village has approximately 3500 inhabitants and is currently considered an agro-industrial centre because of its high provision of water and productive lands, as well as its proximity to roads and ports.Livelihoods are linked to agriculture and the transformation of cassava to extract starch (rallería).Starch extraction is the area's principal economic activity, resulting in high water demand and pollution.This processing creates a complex trade-off between using water for human consumption and for agroindustry activities [34].
Water provision in the region comes mainly from independent WUAs [36].The Mondomo's WUA initiated approximately 60 years ago, when a small group of water users defined their particular needs, built an aqueduct and developed a payment scheme and participation processes.They prioritized household needs, followed by water use for agricultural production.The first aqueduct was mainly dedicated to water distribution but lacked an effective sanitation system.This aqueduct was managed by a water administration committee plagued by management problems and economic deficit.
In 1994, a seismic shift damaged the water distribution system, acting as a trigger factor to configure a private-public alliance around the construction of a new aqueduct.Different public and private institutions [37] worked with the community in the design of the new aqueduct and in the definition of the management rules, where the beneficiaries were the system's administrators and could define the local water governance scheme [38].Design and execution activities for the new aqueduct lasted from 1994 to 1998 and approximately 718,000 USD were invested by national and international donors (World Bank, European Union and Inter-American Foundation).Initially, the community was sceptical due to the long tradition of corruption and uncompleted projects.People were also frustrated with the lack of transparency and control in the management of the first aqueduct.However, a local University fostered a legitimate participatory process that attracted the interest of most of the inhabitants, overcoming their traditional resistance and mistrust to work with publics agencies and authorities [39].
The principal achievements of this alliance were as follows: the design, construction and start-up of a complete clean drinking water supply, with almost 100% coverage; the creation of the Mondomo's WUA, as a public utility company with a community base; the land acquisition of plots on both sides of the principal watershed to guarantee water access; the creation of a new culture around participation processes, rationalization of water consumption and payment for the aqueduct services; the effectiveness and transparency in the management of economic resources and investments; the community's credibility in managing a public service; and a strong feeling of ownership among the users [40].
The Mondomo's WUA has been selected as case study due to several reasons: (1) it is considered as an example in the Colombian context of efficient combination of water national policies with local needs [39]; (2) it uses technologies and maintenance methods that reduce electricity and chemical products use [37]; (3) it currently fully covers increasing demand of water maintaining quality and preserving the autonomous water governance designed schemes [21]; and (4) it belongs to the founder group of one of the biggest second level WUA's platform called AQUACOL [41,42].Additionally, in 1998, it was distinguished by the World Bank as an example of an integral local initiative focused on poverty alleviation in Colombia [43].

Methods
To consider the DPs as fuzzy sets, the first step was to define the categories that support varying degrees of membership in the set [44] and represent real situations.We found inspiration in the coding system proposed by Ratajczyk et al. [45], based on the codes established by Ostrom et al. [46] in the Common Pool Resources Coding Manual, but we proposed our own definition of the categories that represent potential situations in the governance systems of common pool resources.
The scores ranged from 0 ("fully out") to 1 ("fully in").We used 0.5 as the crossover point (neither in nor out) [25] and established 4 additional degrees of membership for each DP.The degrees of membership 0.2 and 0.4 are "more out than in", and those of 0.6 and 0.8 are "more in than out", and Water 2017, 9, 535 7 of 20 there are differences between them defining if they are "more or less in/out", and "mostly in/out" (see Table 2).The proposed degree of membership aimed to support the systematically coding of DPs and increase the replicability of this working method by other researchers.
The second step was to describe the actual situation of the water governance in Mondomo's WUA using the IAD framework [26].This descriptive single case study [47] is based on qualitative and quantitative information collected between 2012 and 2015.We initially collected secondary information from different technical, economic and legal sources to construct a narrative that describes the different spheres (environmental, social, economic, etc.) interacting in the water management.After, we analysed how the new legal framework could affect its governance system.
Finally, we scored membership for each of the DPs, both in the actual and in the most plausible situation after the change of policy.MacQueen et al. [48] recommend the use of two or more coders for qualitative analysis.We used four coders (two researchers and two members of NGOs working on water management issues), all of them with working experience in the area, in WUAs and in community-based management and governance.Both the number of coders and the clear definition of the conditions that defined each score were intended to increase the transparency, reliability and replicability of the research and to avoid the traditional bias of subjectivity in the social sciences.To assess intercoder reliability we calculated Krippendorff alpha [49] using the irr-package in R. The coders used the existing literature, most grey literature (Text S1), to characterize Mondomo's WUA, together with their knowledge of the situation.They independently scored each DP in both situations; after, the scores were compared.Often, the scores were similar; where differences existed, the coders assessed discrepancies and discussed the reasons behind their scores.The exercise of interpreting the scores was very useful for identifying possible biases and for reaching a final consensus about the score closest to reality.Finally, the differences between the actual and the new situations were analysed to obtain answers to the second research objective.Mondomo´s aqueduct is catalogued as the most advanced drinking water supply system in Colombia's rural areas.The technology used for water purification is multi-stage filtration with low consumption of chemicals and the distribution networks operate by gravity, minimizing the use of energy for distribution [39].Wastewater is collected by a system of sewage and discharged without treatment to the Zanjón Tiembla, a tributary of Mondomo River.The sewage service is provided by Mondomo´s aqueduct is catalogued as the most advanced drinking water supply system in Colombia's rural areas.The technology used for water purification is multi-stage filtration with low consumption of chemicals and the distribution networks operate by gravity, minimizing the use of energy for distribution [39].Wastewater is collected by a system of sewage and discharged without treatment to the Zanjón Tiembla, a tributary of Mondomo River.The sewage service is provided by the Municipality, and the cleaning service by a public service company that has an agreement with the WUA to include this charge in billing.As a management strategy, the WUA develops several reforestation programs with the participation of environmental agencies and the local community in their upper lands [40].

• Attributes of Community
Most of the users of the aqueduct are mestizo, but at the upper lands locate the indigenous councils of La Concepción and Las Delicias.Approximately 780 Mondomo's houses are currently connected to, the aqueduct [39].In the town centre, there are seven neighbourhoods, each with a Community Action Board.The Association of Community Action Boards has become the most representative community body.Another instance of strong representation is the guild of the rallanderos (processors of cassava starch), as this industry is one of the main employers in the area.Community members are in charge of the operation and maintenance of the water system.A police unit addresses minor conflicts and represents the municipal administration in the area.Nowadays, illegal mining at the upper part of the watershed is jeopardizing the water governance structure defined by the WUA, due to the difficulties to solve conflict with armed actors [20].

• Rules
The aqueduct has a water concession issued by the local environmental agency (Corporación Autónoma Regional del Cauca (CRC)), which, according to the Ministry of Environment and Sustainable Development, is responsible for administering and granting authorizations for the use of water (Law 99 of 1993).The maximum authority of the WUA is the Assembly of Users that defines the general management of the WUA, hosts annual meeting to analyse the performance and the users behaviour and implement changes in the general statutes, if needed.The Managing Board oversees the implementation of the Assembly's decisions [50].
The rules for the use of water are set forth in the statutes of the WUA, which establishes the rights and duties of the associates and the functions of both administrative bodies.The statutes and the procedures for the water provision service were structured per the guidelines of Law 142 of 1994 [34].The initial rule design focused on the autonomy of local water governance, and was a participatory process accompanied by research centres and universities.From the beginning, the community develop a monitoring system, more oriented to capacity building than to punishment or exclusion [41].
The WUA has defined constitutional rules, internal operation rules and users' regulation terms that clearly count it as a community enterprise.The two first sets define the rules for elections of the representative board, the duties and responsibilities of members and the aqueduct working method, based on participatory and democratic constitution and functioning.The second set defines the tariffs (entrance fee and water prices according to socioeconomic conditions), the mechanisms of conflict resolution, and the duties of users (reforestation activities, payment, collective work, participation in the WUA Board, etc.).The current WUA administrative board consider that the new water regulation will particularly affect their autonomy to define fees, entrance regulations and reduce the investment in conservation activities [34].Therefore, they have actively participated in the design of the mentioned draft submitted to the national government by the Colombian Network of Community Pipelines [35].
In terms of conflict management, the protocol to manage conflict internally is based on traditional values such as cooperation, reciprocity and the identity created around the aqueduct beneficiaries that help to maintain a low-cost conflict solving arena [50].However, most of the actual conflicts are with other communities that allow illegal mining in their territory.The attempts to establish agreements with members of the indigenous council La Concepción that permit productive (and polluting) activities in conservation areas have not worked [34].Campaigns have also been launched by the CRC to prohibit bauxite mining in the tributary streams of the community aqueduct.

• Action Situations
The WUA has a granted water concession and pays the CRC an annual fee for water use.It executes conservation actions with community members.Water purification treatment is performed according to the regulations of the Ministry of Health.The payment system and tariffs are defined in accordance with the guidelines of the water and basic sanitation regulation commission at the CRC and the water use fees follow the rules established by the superintendents of public services [21].

• Participants
The identified actors or participants that are-directly or indirectly-part of Mondomo´s WUA are as follows: water users, peasants and ranchers, "rallanderos", WUA Managing Board members, major municipalities, bauxite miners, the inspector of police, sanitation technicians, the CRC and the Universidad del Valle [37].The members of the WUA are organized in sub-commissions for health, education and water quality that work together with the representatives to improve the governance system through improving the social conditions of the members [34].

Interactions
The process of "formalization" of the WUA has entailed compliance with the sectoral regulations of the administration and with the operation of water and sanitation systems, but has rendered some agreements established by the community (i.e., free delivery of water to very poor families, the performance of some works by users) as illegal, undermining the WUA's operational capacity [50].A relationship of compliance between the association and the sectoral authorities is then established through the adoption of the normative package [21].Between users and the Managing Board, a collaborative relationship exists, evidenced by the presence of users in the assemblies and the active users' participation in conservation and maintenance activities.Contrarily, with the rallanderos and the WUA of Cascajal there is a competitive relationship to secure access to water resources.
Three factors related to social interactions are highlighted by users, representatives and international NGOs as drivers of WUA's success: (1) the strong and active participation of the local community, from the beginning; (2) the proven local capacity to solve local internal conflicts; and (3) the managing capacity of the WUA board, not only in solving the economic problems of the first water association but also in increasing the economic revenues to be invested in maintaining the aqueducts, using social water prices.

Evaluative Criteria
Mondomo's WUA indicators clearly differentiated it from other WUAs in Colombia: service coverage is 100%; water losses are less than 7%; micro-measurement coverage is 100%; and users' payment delays is 15% [40].The implementation of other criteria is pending, such as criteria to evaluate the co-responsibility for the pollution of the Mondomo River, as well as those to measure the vulnerability of the sub-basin, and the effectiveness of the adopted measures of protection.
The highly active participation of the WUA management board has also been recognized by different regional and national WUA second level platforms such as AQUACOL and the Colombian Network of Community Pipelines, strengthening the influence of this WUA´s in the recognition of alternative governance water models in Colombia [34].

Changes in the Desing Principles Due to the New Policy Regulations
Table 3 presents the score of each DP in the actual and future situations, and the descriptions that justify these scores.Table 3. DP scores according to the fuzzy membership categories.

1A User Boundaries
Currently: Access to water is regulated.New users must request admission and fulfil requirements (entrance fee, water fees and community work to maintain the aqueduct facilities), but Mondomo's WUA has no enforcement capacity (e.g., illegal users).
After policy changes: Water access can be determined by the capacity and willingness to pay and not necessarily by the collective action mechanisms that currently define access.Mondomo's WUA might lose capacity to exclude current nonusers.

1B Resource Boundaries
Currently: Access to water in Colombia is a right, and where water sources such as rivers or gullies exist, different actors can claim access.Thus, the limits cannot be easily defined, especially in rural areas.However, the constructed limits of the aqueducts are perfectly defined.Only users legally connected to the system can have access to water.
After policy changes: No major changes are expected because the conflicts do not depend on the resource boundaries but rather on the access to the water sources.

2A Congruence with Local Conditions
Currently: Appropriation and provision rules are congruent with local social and environmental conditions.A clearly established and collectively decided payment system based on the consumption level, the use of water and the economic status of households exists.Local rules not only cover provision and payment but also other activities such as the rationalization of use, improvement of water provision and fostering a water-saving culture.However, Mondomo's WUA has little capacity to modify external rules that do not consider its local conditions.
After policy changes: External rules might be imposed over the traditional schemes and local social structures, giving access to users according to their financial means.WUAs might function as private or mixed service provider companies, losing their local-social approach and their matching of local conditions.

2B Appropriation and Provision
Currently: The benefits obtained by users thanks to the building of the aqueduct and the actual performance of Mondomo's WUA greatly exceed the inputs.Most of the time, users have access to quality water in the quantities they need, and at a cheap price modulated as a function of the use and status of the users.
After policy changes: Costs might increase because WUAs will lose part of their decision-making and bargaining powers and water price will be decided by regional and national regulation systems.The potential expansion of users might surpass the resource provision levels, and benefits could be clearly reduced (e.g., many families could not afford the water price).

Collective-Choice Arrangements
Currently: Mondomo's WUA collective-choice arrangements are clearly defined.The managing board is appointed by the users and its work is based on transparency and accountability schemes.All individuals affected by the operational rules have the opportunity to participate in modifying them.The main decisions are taken in public assemblies open to all the users but are not always respected by non-users.
After policy changes: The local/traditional decision-making process might be jeopardized, and new and more bureaucratic processes can be imposed, complicating users' options to modify rules.

4A Monitoring Users
Currently: Appropriation and provision levels are monitored, recorded and delineate the payment system.Monitors are selected among the users and accountable to them (e.g., they can be asked to disclose their decisions regarding a user's violations of regulation codes).The information is used to make decisions.
After policy changes: The system monitoring can be taken over by external authorities.Mondomo's WUAs might face a decline in their capacity to monitor users' appropriation and provision levels because their options to decide who monitors users' actions might be reduced.

4B Monitoring the Resource
Currently: Monitors selected from the community continuously monitor water conditions to have information in real time about the conditions of the resource.The information collected is used to make group decisions.
After policy changes: The resource monitoring can be taken over by external authorities.The capability to keep systematic data to evaluate the state of the resource and the real capacity of users to access updated information might decrease.

Graduated Sanctions
Currently: Mondomo's WUA regulations define an internal gradual sanction scheme (known by all the users) that starts with a private caution, followed by a public caution; then, a penalty or fine is applied and, if none of these options works, the service is suspended.This sanction scheme is not only based on secure but includes social activities and collective work related to the maintenance of the aqueduct.There are also social sanctions.
After policy changes: Sanctions might be established by external officials not accountable to users and principally based on secure payment.

Conflict-Resolution Mechanisms
Currently: Mondomo's WUA has local well-established conciliation and bargaining protocols based on social values (e.g., high reciprocity) and accessible to all users, which helps to low-cost conflict solving.Quite rarely, conflicts among users are managed by national authorities (Superintendent of Public Services).
After policy changes: Conflicts might be negotiated and arbitrated by national-level institutions that are costly and difficult to access for users.

Minimal Recognition of Rights to Organize
Currently: Mondomo's WUA have de facto (users and non-users accept the existence and rules of WUA) and de jure (nationally recognised as public service provider association) rights to devise their own institutions and its organizational structure and functioning are clearly recognized by national and regional environmental authorities.
After policy changes: Traditional de jure water rights recognized as belonging to Mondomo's WUA might be transferred to new users and agencies.Internal decisions might not be recognised by the authorities.

Nested Enterprises
Currently: Mondomo's WUA is organized in multiple layers of nested enterprises with horizontal (its role exceeds water management and external authorities rely on it to implement different health, education and investment programs at local level) and vertical links (it belongs to AQUACOL, a second-tier organization, and to an aggrupation of regional WUAs).
After policy changes: They will continue participating in different vertical activities and organizations, but users fear they could lose independence and autonomy for horizontal links when other actors enter the scene.
The coding values and the Krippendorff alpha test for intercoder reliability, can be consulted in Supplementary Materials Tables S1 and S2.Results showed that the agreement between coders is higher for the situation after policy change (Krippendorff alfa = 0.490 and 0.829, respectively), but, in both cases, the percentage of agreement (0.88 and 0.96, respectively) is higher than the percentage of agreement by chance (0.77 and 0.79, respectively) at p-value < 0.05.
Figure 3 depicts how the scores change between the actual and the potential future situations.In the actual situation, all DPs except 2A score 0.8 or 1 on membership.However, the situation might dramatically change with the new legal and policy frameworks, leading to a general decrease in the membership scores of most DPs.
Water 2017, 9, 535 14 of 20 both cases, the percentage of agreement (0.88 and 0.96, respectively) is higher than the percentage of agreement by chance (0.77 and 0.79, respectively) at p-value < 0.05.
Figure 3 depicts how the scores change between the actual and the potential future situations.In the actual situation, all DPs except 2A score 0.8 or 1 on membership.However, the situation might dramatically change with the new legal and policy frameworks, leading to a general decrease in the membership scores of most DPs.

Considering DPs as Fuzzy Sets. Methodological Contributions
DPs are configurations of rules that support the emergence of effective governance [51] and one of the primary methodologies within the literature to analyse CPR governance.However, the method has received several critiques: (1) the problems to measure outcomes meaningfully and reliably [52,53]; (2) the measurement validity, especially with small samples [54]; (3) the difficulties to unpack causal complexity in qualitative work [55]; and (4) the lack of comparability between case studies [52,56].
Considering DPs as fuzzy sets can contribute to theory building by addressing some of them.The experience documented illustrates that approaching them as fuzzy sets allows a fine-grained definition [25] of these rules and avoids the simplification of considering the DP as simply present or absent.These rules can be present at a certain level or can be better developed in some systems than in others.The proposed categories establish a protocol to identify features that should be present in each of the DPs, support methodological transparency, help to systematically code the DPs in categories according to the extent to which these rules are present, and as such, impose consistency and reliability on the data collection, and facilitate comparisons and replicability of the findings [45].
Further, the categories proposed can help researchers to explore which combinations of rules are present for each membership score and create opportunities to identify the necessary steps toward sustainable governance models.
Finally, the visualization using a spider web diagram gives a quick insight of the situation (in our case of the initial and final ones).This graphical representation immediately communicates the DPs in need of attention.In the analysed case study, the tool and results can support Colombia WUAs

Considering DPs as Fuzzy Sets Methodological Contributions
DPs are configurations of rules that support the emergence of effective governance [51] and one of the primary methodologies within the literature to analyse CPR governance.However, the method has received several critiques: (1) the problems to measure outcomes meaningfully and reliably [52,53]; (2) the measurement validity, especially with small samples [54]; (3) the difficulties to unpack causal complexity in qualitative work [55]; and (4) the lack of comparability between case studies [52,56].
Considering DPs as fuzzy sets can contribute to theory building by addressing some of them.The experience documented illustrates that approaching them as fuzzy sets allows a fine-grained definition [25] of these rules and avoids the simplification of considering the DP as simply present or absent.These rules can be present at a certain level or can be better developed in some systems than in others.The proposed categories establish a protocol to identify features that should be present in each of the DPs, support methodological transparency, help to systematically code the DPs in categories according to the extent to which these rules are present, and as such, impose consistency and reliability on the data collection, and facilitate comparisons and replicability of the findings [45].
Further, the categories proposed can help researchers to explore which combinations of rules are present for each membership score and create opportunities to identify the necessary steps toward sustainable governance models.
Finally, the visualization using a spider web diagram gives a quick insight of the situation (in our case of the initial and final ones).This graphical representation immediately communicates the DPs in need of attention.In the analysed case study, the tool and results can support Colombia WUAs in their discussions with policymakers of the draft regulation to preserve the rights to the Community-Based Management of Water [19,35].The results provide evidences on how the policy regulations and actions might (unintendedly) hamper highly empowered local governance systems (as represented by many WUAs in Colombia), and negatively affect long-lasting, effective collective-choice rules, and monitoring and sanctioning processes [57].Its use might assist communities, decision makers and politicians in creating awareness and setting priorities for action, and CPR users to identify the shortcomings of their governance system.

A Long-Lasting Governance System in Danger
DPs cannot be conceived as a protocol for successful management or as a panacea; rather, they are a first diagnostic tool to determine the "health" of the governance of a CPR [16]; further, not all DPs need to be present in successful governance systems, and the non-existence of some of them should not be considered a governance failure.From Figure 3 we deduce that the actual governance system of Mondomo's WUA can be considered healthy and robust [16], based on ecologically sustainable management of water [58] and lacking important conflicts [59], as most DPs scored close to 1. Mondomo's governance system, based on collective management and values such as trust, reciprocity and social cohesion, provides water for households and livelihoods at a social price that, in turn, covers the costs of the service and generates surplus.Further, it protects the sources of water and conforms to a social patrimony [20].
However, the new policy challenges this effective governance system.Some DPs (1A, 1B, 3, and 8) only change slightly, but others undergo moderate (2A and 2B) or intense changes (4A, 4B, 5, 6, and 7).Furthermore, the changes in the scores of six DPs (2A, 4A, 4B, 5, 6 and 7) place them below the crossover point for membership, which might introduce important pressures into the actual governance system.
Both DP2s decrease but DP-2A (congruence with local conditions) shifts to "more out than in" scores.Hence, the local rules based on solidarity (different water prices according to the household's socioeconomic status), self-organization to solve water problems and local cultural understanding of the value of water [60] lose relevance.External standardized rules not recognizing the social value of water nor its role in the community living standards can be imposed.Further, access and entitlements to water security are mediated by institutions and infrastructure [61].The potential shifts in institutions and new infrastructures for water-demanding sectors will decrease the water available to many families.
The actual situation of DP-2B (appropriation and provision) indicates the following: (1) users' involvement in management and collective work is considered fair and not excessively demanding; (2) the low tariffs derived from the technology used and the reduced distribution costs guarantee the sustainability of the system; and (3) the internal rules recognize the inputs (labour and involvement in decision making) of those users who worked on the construction of the aqueduct by establishing an entrance fee for those who did not participate.These values of genuine involvement, respect and appreciation for the governance system [62] will lose relevance if market principles regulate the appropriation and provision of the resource.
The modifications in monitoring (DP-4A and DP-4B) and sanctioning (DP5) challenge democratic water governance.The new policy reduces the WUAs power to monitor and sanction in favor of national regulation agencies, opening options to governance mismatches.These agencies are often considered inefficient, and users fear that this transfer of the monitoring process [63] might hamper water management, giving outsiders or powerful economic sectors access to water and decreasing the water allocation of current users.Further, the technology used and the community involvement provide high quality water at a low cost, with minimal use of chemicals and electricity, but the new situation will increase the costs of operation and consequently the price of water.Moreover, proximity and local relations allow the actual sanctioning system to consider the local conditions and the users' necessities before taking decisions, but the external agencies will not include these social considerations in their operational system.The power inequalities between current users and external agencies will decrease the capacity of the former to access the resource [15].Some rights, such as the rights to make transparent, equitable and evidence-based choices, can be lost by the local community as found by Pahl-Wostl et al. [64].
Conflicts (DP6) are actually managed by the WUA's board members.The democratic methods used for settling differences [60] solve conflicts effectively.Boards are composed of neighbors who are generally aware of each other's realities and problems, as well as those of the community.Although this does not preclude the existence of disagreement, it does enhance the Mondomo's WUA's ability to manage conflicts.The transfer of these powers to external agencies will dramatically erode this capacity, challenging the traditional values that characterized the relations among users and increasing the costs (time, money, accessibility) of solving them [65].
The dramatic change in the DP7 (recognition of rights to organize) score shows how the new policy regulations might challenge the current governance institutions.They are based on the periodic renewal of authority, the non-partisan nature of the WUA, and the awareness of community history.Community leaders have a strong sense of commitment and dedicate significant efforts, free of charge, to ensuring the WUA's functioning [60].The new regulation introduces the option to create new institutions based on market criteria and payment capacity, without considering the existing local social principles.Users fear that corruption practices could transfer power from the WUA to other actors and that operational management results will be more inefficient and expensive [63].
However, DPs do not just occur together, they interact to produce outcomes [59].Not only the number of DPs present, but the configurations of DPs determine the success of the collective action [59].These configurations differ according to the mobility of the resource and the intensity of the human-made infrastructure necessary for its management.In systems with high intensity of hard human-made infrastructure and high mobility, such as water provision and distribution systems, the combination of principles leading to successful governance are 1A (clearly defined social boundaries), 2A and 2B (congruence of rules and benefits), 5 (graduated sanctions) and 4A and 4B (monitoring of users and resources) [59].
All these principles suffer important reductions in membership in the new scenario; moreover, four of them (2A, 4A, 4B, and 5) shift to a "more out than in" score.The new regulation's failure to recognize the customary rights of Mondomo's users might lead to governance mismatches [66] and affect not only water security but also the social fabric articulated around Mondomo's WUA.This WUA faces the possibility that the actual local privatization of water by the aqueduct users might shift to what Araral [54] describes as an externally imposed privatization of local commons.Users might be frustrated by the difficulty of making decisions in mixed arenas subject to cognitive mismatch and power plays [67] and by the little recognition at national and regional scales of WUA's important contribution to water access and other basic services for the population [23].

Conclusions
Considering DPs as fuzzy sets open new avenues for CPR governance research.It addresses some of the problems of homogenizing different real situations and work with imprecise categories, and introduce more possibilities to compare situations.We are cautious to affirm that the proposed categories have to be the final ones and we cannot conclude on external validity [68], since we only tested the method in one case study.We propose a starting point to the scholarly community; the future testing of the method in a bigger number of case studies will permit to refine them and to be more categorical about their validity.
The new water regulation intends to increase the quality and access to water in rural areas, but, policymakers should not implement one-size-fits-all approaches that might led to the fading of long-lasting place-based governance approaches like those represented by many Colombia WUAs.The shift in the DPs highlights common dilemmas in CPR management and how the multi-level context and power asymmetries affect community-based management.As repeatedly noted by Ostrom, institutional arrangements are costly [16,29,69].Mondomo's users have made long-term investments in creating their institutional arrangements.As a result, they have a locally managed water governance system that is sustainable, efficient, democratic, adapted to local needs and provides water at a low cost.Further, the system is part of a social patrimony that strengthens the social fabric, local empowerment and the sustainable management not only of water but also of most local needs.The new policy should adapt to ensure the safeguarding of these democratic governance systems.
As has been discussed throughout this manuscript, the above-mentioned threats might or might not materialize.It will depend on how the policy is implemented in each place.We hopefully expect

7
Minimal Recognition of Rights to Organize 0.0 Users have no right to devise their own institutions 0.2 Users have de facto right to devise their own institutions, but their decisions are not considered by external governmental authorities 0.4 Users have de facto right to devise their own institutions, and their decisions are partially considered by external governmental authorities 0.6 Users have de facto right to devise their own institutions, and their decisions are considered by external governmental authorities 0.8 Users have de jure right to devise their own institutions, and their decisions are considered by external governmental authorities 1.0 Users have de facto and de jure rights to devise their own institutions, and their decisions are considered and respected by external governmental authorities 8 Nested Enterprises

Figure 3 .
Figure 3. DPs scores in the actual situation and after policy changes.

Figure 3 .
Figure 3. DPs scores in the actual situation and after policy changes.

Table 2
present the results for the first research objective, depicting the six categories of membership score established for each DP.

Table 2 .
Categories of membership score for each DP.
Appropriation and provision levels are poorly monitored by monitors who are not accountable to users 0.4 Appropriation and provision levels are systematically monitored by monitors who are not accountable to users 0.6 Appropriation and provision levels are occasionally monitored by monitors who are accountable to users; the information collected is non-necessarily used to inform group decisions 0.8 Appropriation and provision levels are often monitored by monitors who are accountable to users; the information collected is used to inform group decisions 1.0 Appropriation and provision levels are systematically monitored and recorded by monitors who are accountable to users; the information collected is used to inform group decisions