Can a Multi-Criteria Methodology Fit with Non-Profit Institutions’ Decision-Making? An Application in a Spanish Non-Profit Association

The 2008 crisis occurred in Spain while the Social Economy was undergoing a process of development and transformation to meet growing demand. The new stage uncovered organizational inefficiency in the Social Economy’s institutions that might put at riskh8 their survival and highlighted new challenges that these organizations had to face. It was necessary to move towards a strategic model that would allow them to respond more swiftly and effectively to external shocks without renouncing the values and principles of the Social Economy. This paper analyzes the effects of the 2008 crisis on the institutions of the Spanish non-profit sector and suggests a methodology to help them decide between different affordable future strategies. We use a multi-criteria decision-making methodology to carry out a case study focused on a non-profit association, whose main goal is to take care of disabled people. We show how to help the association make decisions in an efficient, rigorous and democratic way.


Introduction
In the years before the great recession of 2008, Spanish economic growth and the enlargement of the public budget created a large increase in the service sector and welfare state. This economic model was based on privatization and the hegemony of private companies (Banco de España 2017). However, institutions in the Social Economy, the main goal of which is not to maximize profits, were also increasing. The Social Economy gathers those initiatives that are more interested in community profit than economic profit: it is a third sector, located between the public and private sector, that is essential to achieve more balanced and fair development from the social and economic points of view (Castells 2017).
Various forms of associationism stirred by a number of visions, ideologies, theories, and philosophies had an influence in the formation of the concept of Social Economy. The history of Social Economy is linked with the oldest forms of human association, but it was at the end of the nineteenth century when the concept of a Social Economy was institutionalized (Defourny and Develtere 1997;Moulaert and Ailenei 2005). The first connections are found in French popular associations, and cooperatives emerged as a response of the most vulnerable social groups to new living conditions stemming from industrial development in the 18th and 19th centuries. These associations, with their values and their principles, opened the way to the emergence of organizations where all the economic agents have the power of ownership. The first time that the concept of Social Economy appeared was in 1830 when the economist Charles Dunoyer published a treatise on Social Economy (Monzón and Chaves 2012). In the nineteenth century the term "Social Economy" designated not only a type of organization created by workers, but also an approach that integrated social issues into the study of economy (Bastidas 2001).
The idea of the Social Economy increased in importance in Europe during the first third of the twentieth century; however, after the Second World War, Western Europe showed growth that was characterized by the private capital sector. The concept of the Social Economy stepped back and lost part of its role as a safeguard of social welfare because the state took the lead in this issue. It was not until the crisis of the welfare state in the last quarter of the twentieth century when an interest in the Social Economy was renovated (Monzón and Chaves 2012). The concept spread through a triple process of recognition: mutual recognition of organizations, recognition by the state, and recognition by the scientific community (Bastidas 2001).
The expansion of the Social Economy has enlarged both its complexity and its academic and scientific interest. The plurality of companies, institutions, and entities that have been established, as well as the plurality of needs, problems, and social demands considered, have led to the proliferation of a large number of terms related to the Social Economy: The Solidarity Economy, Collaborative Economy, Economy for the Common Good, Third Sector, and Circular Economy, to highlight the primary terms. While it is true that there are important similarities among these terms, it is also true that there are significant differences, so it is necessary that any study should clearly define where the focus of the analysis would lie. 1 1. The market or corporate subsector is integrated by companies under a democratic organizational structure, where the profit distribution does not link with the partners' capital investment. These organizations are created to satisfy their partners' needs and are market producers, which means that their output is mainly intended for sale on the market at economically significant prices. The surpluses can be distributed among their user members, although not in proportion to the capital or the fees provided by the members, but according to the member's transactions with the organization. 2. The non-market subsector is integrated by private formally organized non-profit institutions serving households. It also includes private entities-mainly associations and foundations-serving families and households and can trade on the market at economically non-significant prices. Such organizations seek to promote the recognition and exercise of social rights and to achieve cohesion and active social inclusion of people in all their dimensions. Particular support is given by these entities to those people and social groups that are in a more vulnerable situation or at risk of social exclusion. Their main sources of resources are donations, partners' dues and subventions. The surpluses, if there are any, cannot be appropriated by the institution members (Fundación Luís Vives 2012; Monzón and Chaves 2016).
In terms of this established definition of the Social Economy and its two subsectors, this study deals with the non-market subsector (NMS) and, concretely, with the Spanish NMS.
The business environment has been seen to have increased uncertainty and instability due to globalization, increased economic and social technological changes, and intensifying competition within markets. Increasingly companies have to make decisions in a changing and competitive environment. Multi-criteria decision making is a decision methodology widely used by private companies that helps to increase the quality of their decisions. Under our point of view, even though there are other decision-making methodologies and the multi-criteria methodology is not necessarily the best, it can be useful for Social Economy organizations because it allows them to integrate both their democratic and participative principles with the necessity to make decisions in complex contexts.
The goal of this paper is to show how to implement a multi-criteria decision-making methodology in a non-profit association whose main goal is to take care of disabled people, in order to get the best alternative according to stakeholder preferences and the establishment of measurable performance criteria. We used the ELECTRE II (Elimination and Choice Expressing Reality) methodology, which is a widely used multicriteria methodology that evaluates multiple conflicting criteria in decision making.
We analyse the effects of the 2008 economic crisis on the institutions of the Spanish NMS and show, through a practical application, how ELECTRE II could be used to help them decide between different affordable future strategies in an efficient, rigorous, and democratic way. We have to bear in mind that the effects of the 2008 economic downturn hit at the core of the Social Economy because it had an impact on financial and human resources and on the volume and typology of social demands (Jaén 2017). Between 2010 and 2013 the public funding dropped roughly to 8%. The employment fell by 4,5%, while people at risk of social exclusion increased by 12% (Plataforma ONG de Acción Social;Plataforma de Tercer Sector and eeaGrant 2015;PricewaterhouseCoopers, ESADE, & Obra Social "La Caixa," 2014). As a result of the crisis and the changes in social demands that it brought about, the entities dedicated to insertion-integration and housing increased from 19% of total to 23% (Plataforma ONG de Acción Social; Plataforma de Tercer Sector and eeaGrant 2015). The new stage uncovered organizational inefficiency in the Social Economy's institutions that might put at risk their survival and highlighted new challenges that these organizations had to face (Marbán 2014).
The study is relevant because the future of institutions in the Social Economy requires making decisions to move towards a strategic model that would allow institutions to respond more swiftly and effectively to external shocks without renouncing the values and principles of the Social Economy. Our work provides an efficient methodology for assisting the decision-making process and furthermore shows how to implement this methodology in a real situation.
The paper is organized as follows. Section 2 presents a conceptual framework. Section 3 describes the effects of the 2008 crisis over the Spanish NMS. Section 4 presents the multi-criteria decision-making methodology. In Section 5 the ELECTRE II is applied. We conclude in Section 6.

Conceptual Framework
There is no doubt that in the current stage, the business environment has seen increased uncertainty and instability due to globalization, increasing economic and social technological changes, and intensified competition within the markets. In the past, in a less globalized environment, companies could survive with strategies that were not very dynamic, and with management models based on continuity. At present this is not possible, because companies have to carry out continuous adaptation in an extremely changing and competitive background. Their success depends to a large extent on their ability to make efficient and complex decisions in a dynamic global marketplace (Aydiner et al. 2019;Zopounidis and Doumpos 2002).
However, it is worth noting that to make decisions on the basis of the best possible information and to contemplate a wide range of perspectives relevant to the problem, through an structured process of analysis and choices of alternatives, outweighs individual capabilities (Braga et al. 2019). The decision-making process has to include experts, stakeholders, voters, and a company's staff, among the most important actors, who can provide evaluations of the alternatives under multiple criteria. Studies on the companies' organization and decision making highlight that multi-composition in the decision-making process normally implies disagreements. However, this is not a problem; quite the opposite, as disagreements can strengthen decision-making criteria. Heterogeneous teams, composed of members with different knowledge, skills, and/or professional experiences, tend to develop better decisions because they may have different knowledge of the problem (Braga et al. 2019;Kabak and Ervural 2017;Wu et al. 2015). Multiple dimensional decisions require multiple decision makers and also the use of appropriate decision practices, which help overcome both individual biases and the peculiar biases of the organization (Rousseau 2018).
Multi-criteria decision-making (MDM) is a decision methodology that can help to increase the quality of decisions by making the process more explicit, rigorous, rational, and efficient (Wang and Triantaphyllou 2008). Keeney and Raiffa (1976) provide the first complete exposition of the MDM methodology. They integrate uncertainty and multiple objectives by extending decision theory to incorporate multi-attributed consequences. The main assumption is that decision-makers wish to be coherent in taking decisions. MDM establishes preferences among an explicit set of options that the decision-making body has identified, and for which it has established measurable criteria. These different criteria are integrated to evaluate alternative decisions, scoring them according to stakeholder preferences and measurable performance data, helping to make the decision process reliable (Comunities and Local Govenment (CLG) 2009;Stoycheva et al. 2018). The main role of the techniques is to deal with the difficulties that decision-makers have in handling large amounts of complex information in a consistent way. The successful implementation of MDM depends on the correct choice of anyone who can make a useful contribution to the analysis. Furthermore, the stakeholders and those who guide the decision process have to have a clear vision of the values which guide every decision taken.
The advantages of MDM derive from the circumstance of confronting complex problems with an easy structure (Ferreira et al. 2011). It combines experts' judgement with the freedom of decision-makers in showing their preferences. Also, it can manage broad stakeholder participation, stimulating their discussion and, at the same time, a wide set of criteria (Achillas et al. 2013;Bhagtani 2008).
This methodology is very suitable in the environment of companies because the promotion of competitiveness, the search for efficiency, and the need to increase productivity are closely linked to decision scenarios involving multiple variables or selection criteria. For the business world it means choosing the most desirable option in the best case or the least unsatisfactory option in the worst (Berumen and Llamazares 2007) . Business fields where the methodology has been applied are very broad; among others they include the selection of portfolios, environmental impacts, organizational methods, investments in industrial equipment, selection of suppliers, design of consumer-oriented products, cost management, and engineering design. Also noteworthy are studies in the financial and health sector. In the public sector, the methodology has been used for public tendering or the evaluation and hierarchy of projects, among the most important areas. 2 If the methodology is useful in this broad spectrum of public and private organizations and companies, we believe it can also be useful for Social Economy organizations. Thus, as indicated by Hysa et al. (Hysa et al. 2018:1350: "The non-profit sector has been separated from the rest of global economy. Setting unreachable and unrealistic goals to the non-profit sector, and at the same time preventing it from using the tools of the free market, obstructs this important market player to solve social issues. Furthermore, the separation that has been created between the for-profit and non-profit sector, has relieved the for-profit sector from its social duties, and at the same time has increased the pressure and demand toward the non-profit sector to solve the social issues. The reason why social problems have not been solved yet, it is not only thanks to the failure of the neoliberal ideology, but also from denying the non-profit sector from using the free market strategies." Institutions in the Social Economy should make their democratic and participative principles compatible with those strategies of private companies that can improve their performance. Previously, studies focused on non-profit institutions have tended to explain them as institutional instruments for correcting market failures, particularly those related to public goods and information asymmetry problems . Their "social" side was emphasised. However, recent studies have shown that non-profit institutions need to go further. They can solve social, economic, and environmental problems by using the entrepreneurial engine and profit instruments. It would be a dangerously limiting approach to believe that they are institutions whose only aim is to reach 'social' outcomes (Hysa et al. 2018;Steiss 2003). The entities themselves have recognized that the sector has to adapt to the "real needs of society," "new profiles and demands," and the "new reality before the private company" (Fundación Luís Vives 2012).
These changes imply a more difficult management process, which demands, as in private companies, making efficient and complex decisions in a dynamic global marketplace. MDM methodology is appropriate for meeting these Social Economy challenges. MDM integrates multiple conflicting criteria to evaluate alternative decisions and allows a company to work using a participatory, democratic, and transparent decision-making process, which fits in well with the will to make decisions in a more complex stage without renouncing the values and principles of the Social Economy. In early 2010, the contraction of the Spanish GDP decreased as a result of the upturn in the global economy. However, the financial distress in European markets and the unsolved Spanish imbalances disrupted the framework of economic growth. In line with the rest of EU, Spanish fiscal policy changed from expansionary to contractionary to cut the deficit as a first step on the path to new, stable growth (Banco de España 2017).
The crisis hit the Spanish labor market strongly and exacerbated the structural problems that had hampered its performance. At the beginning of 2008, the unemployment rate started to increase, resulting in an unemployment rate of 23% at the end of 2011 and of 26% at the end of 2013. The most vulnerable groups were young people, immigrants, and the disabled and low skilled workers (Éltető 2011).
The crisis affected the NMS in different ways. While the financial and credit perspective provided a clear long-run objective in the private sector, it should be considered a constraint rather than an objective in non-profit organizations (Grigoroudis et al. 2012;Kaplan 2001). We would like to point out that the NMS suffered greater financial instability during the recession than the private sector (Salamon et al. 2009). The credit crunch tightened the access to lines of credit and personal or mortgage loans (Fig. 2). These are important financial resources for solving liquidity problems or for financing programs that are executed before the payment of the public funds (Galindo et al. 2014). The low ratings also affected their guarantees and were basically linked to pre-approval of future collection of public funds.
Public funding withstood the first shock of the crisis thanks to budgetary inertia and the multi-year nature of many European fund programs. The inflection occurred in Fig. 1 Annual variation in GDP percentage. Source: Eurostat 2012, when budget cuts led to the cancellation of grants and programs; the merger of programs; late payments; the reduction, cancellation or absence of subsidies; and public calls without resolution and public calls with resolution but without charge (PricewaterhouseCoopers 2018). Public funds lost by 5 points between 2010 and 2013 (Fig. 3). The social work of savings banks, which are the main source of private financing, lost 7.6 points between 2010 and 2013, and the decline was motivated by bank restructuring in the first years of the crisis (Banco de España 2017). Own funds still showed a progressive increase (3.5 points), thanks to user fees and/or collaborating entities whose weight grew by 9.7 points between 2010 and 2013 (Plataforma ONG de Acción Social; Plataforma de Tercer Sector and eeaGrant 2015). With regard to the labor market, the crisis increased the discrimination of collectives more vulnerable or at risk of social exclusion. 4 The cut in social policies-especially in policies aimed at the social and labor welfare of vulnerable groups-had consequences for the ability to find employment and in working conditions (Caro 2017;Vidal 2013). The effects on employment in the NMS were not as devastating as those on the Spanish economy as a whole. During the period 2010-2013, data show an estimated loss higher than 27 thousand jobs. 5 However, the share of the NMS in Spanish total employment continued to increase (4.1% in 2010 and 4.6% in 2013) (Plataforma ONG de Acción Social; Plataforma de Tercer Sector and eeaGrant 2015). The percentage of entities 4 In 2009 disabilities people unemployment rate doubled that of the people without disabilities and had been increasing since mid-2007. As in other collectives, the crisis increased the labor supply of the disabled people due to the rise in the number of families with all their members unemployed (Fernández 2016;Huete et al. 2009). 5 We have not included employment data from the Spanish Red Cross, ONCE, and Caritas due to their relatively high weight. If these were included, it would appear that employment had increased by more than nine thousand people (Plataforma ONG  Overall, employment in the NMS withstood the crisis better than the Spanish economy as a whole. The crisis did, however, sharpen poverty and social exclusion in sectors that previously had not suffered from them, increasing basic social demand for basic goods such as food. This meant that in 2013, despite decreasing financial resources and cuts in social policies, direct care was almost 30% higher than in 2008 (Plataforma ONG de Acción Social 2017). The growth in volunteer figures during this period-by more than 18% between 2010 and 2013-and the fall in employment suggest that the formula chosen to meet this growing social demand was to replace paid work with volunteering (Plataforma ONG de Acción Social; Plataforma de Tercer Sector and eeaGrant 2015; Plataforma ONG de Acción Social 2017).
There is no doubt that the future of the NMS institutions required confronting challenges, including internal reorganizations and greater optimization of resources. These institutions needed to develop specific and efficient strategies to avoid being overtaken by reality and to continue fulfilling their goals and objectives. The crisis showed that, in spite of the resilience of these entities, they had to develop a more flexible organizational strategic model to respond more swiftly and effectively to external shocks without renouncing the values and principles of the Social Economy. It was therefore necessary to make decisions that implied changes in management and the organizational and quality systems. For this reason, we believe that the application of the methodology proposed in the following section will help in these decisionmaking processes. There are many MDM models to analyze and rank the alternatives. 6 One such model, known as "outranking relations," includes the ELECTRE method. ELECTRE's origins go back in the mid-1960s and the European consultancy company (SEMA). At that time, Bernat Roy, who is widely recognized as the father of ELECTRE, and his colleagues worked on a concrete multi-criteria problem that dealt with the development of new activities in firms. Within the family of ELECTRE models, ELECTRE I was the one that was implemented first. 7 Scholars improved ELECTRE I with the development of ELECTRE II Bertier 1971, 1973), which is widely used when a final ranking of alternatives is needed by an analyst. Among the main factors for which we have chosen the ELECTRE method we wish to stress the following. First, it is an appropriate technique for the evaluation of projects because it allows a general ordering of the alternatives, even when there may be pairs of alternatives that remain incomparable due to the lack of sufficient references to distinguish between them. Second, it is a technique capable of dealing with both qualitative and quantitative information. Third, because it takes into account uncertainty and vagueness and it is less sensitive to changes in data in comparison to other methods, it is more stable and reliable (Rogers and Bruen 1999;Sabaei et al. 2015).
ELECTRE II fits with the goal of this study to analyze the best alternatives for implementing transformations and new strategies that would allow NMS entities to confront the future and, at the same time, to maintain the Social Economy philosophy. It is important to point out that the decision makers (DM) have to have access to full information and thoroughly understand the different available alternatives and criteria used to establish the outranking relationship, otherwise the ELECTRE evaluation method could produce results opposite to those desired (Wen-Chih 2005).
The ELECTRE II method is developed according to the following four steps: Step 1. Select the n alternatives (A), that the DM want to rank according to different criteria that may be in conflict.
Step 2. Select the m criteria (C), that the DM want to evaluate to carry out the ranking of the alternatives.
Step 3. Determine the relative weights (W) of the m criteria.
W ¼ fW j j ¼ 1; 2; …mg j and ∑ m j¼1 W j ¼ 1: Step 4. Apply the ELECTRE multi-criteria evaluation method. ELECTRE II is based on the evaluation of two indices-the concordance index and the discordance index-defined for each pair of alternatives-(A r ) and (A k ). The concordance index measures the strength of the hypothesis that alternative (A r ) is at least as good as alternative (A k ). The discordance index measures the strength of the possibility that this hypothesis is not true. To establish both indices, the following points are needed. We define the decision matrix, that is, the performance of the alternative (A i ) in terms of the criteria (C j ) as: The decision matrix is normalized following min-max normalization. The normalized value of feasible alternative (A i ) under criterion (C j ) is represented by g j (A i ).
If the criterion should be minimized: If the criterion should be maximized: Next, the criteria are classified into three categories: The sum of weights for which A r is better (more preferable) than A k is: The sum of weights for which (A r ) is indifferent to (A k ) is: The sum of weights for which (A r ) is worse than (A k ) is: That is: We can then establish the concordance and discordance indices. Concordance index At this point we need to define the ranking procedure. We work with TOPSIS index developed by Hwang and Yoon (1981), which is based on the concept that the chosen alternative should have the shortest distance from the ideal solution and the farthest from the anti-ideal solution. We first choose the largest value both in the concordance matrix (c*) and in the discordance matrix (d*), then we define the concordance dominance matrix and the discordance dominance matrix, the elements of which are, respectively: c' rk = c *c rk and d' rk = d *d rk We determine the aggregate dominance matrix, the elements of which are: From the aggregate dominance matrix, we calculate the mix evaluation value of each alternative, as follows: ;k≠r a rk r ¼ 1; 2; …; n The alternatives are ranked according to the increasing order of a r . The best alternative is A * ¼ maxa r We also apply an alternative ranking procedure to analyze the sensitivity of the ranking to the methodology used. In this sensitivity analysis, we define the average values of the concordance and discordance matrices (c ; d) as the acceptable values for the concordance and discordance threshold. We define the concordance/discordance dominance matrix according to the following rules: & Concordance: if the concordance matrix element c rk ≥ c the concordance dominance matrix element c′ rk = 1, otherwise c′ rk = 0 & Discordance: if the discordance matrix element d rk ≤ d the discordance dominance matrix element d′ rk = 1, otherwise d′ rk = 0 The aggregate dominance matrix is obtained by multiplying every element of the concordance dominance matrix by the discordance dominance matrix: We then build the new ranking from the "best" alternative, defined as that which is not outranked by others (sum of the column in the aggregate dominance matrix = 0), to the "worst," defined as that which shows the greatest number of alternatives that outrank it, (the maximum value of the sum of the columns in the aggregate dominance matrix). The best alternative of this ranking should coincide with that of the TOPSIS test to guarantee the robustness of our results.

Application of ELECTRE II
The institution for which we implemented the multi-criteria measurement system is a non-profit association called AdP. The main goal of AdP is to take care of disabled people. The organizational DNA is defined by its willingness to accompany disabled people throughout their lives. It is important highlight that the institution carries out entrepreneurial activities and services. The areas where AdP provides services to people and families and develops activities are shown in Fig. 4  The two areas of Food and Leisure and spare time are the primary focal points of entrepreneurial services and activities, while the remaining areas are mainly focused on Social Economy services and activities.
The 2008 crisis hit AdP action areas as described in section two. To achieve longterm viability and sustainable development, the institution sought for an equilibrium between effective performance and the institution's vision. To maintain this equilibrium, the management team decided to establish a new, more flexible development model, with a greater adaptability and that would allow more agility in the decisionmaking process.
The management team considered that the future development model should focus on a strategy reached after a participative process. The different actors involvedincluding caregivers, workers, users, families, and the management team-would have to have the opportunity to engage in the process.
Prior to the application of ELECTRE II, important and laborious preparatory work was necessary. To carry this out, three work groups (WG)-in addition to the DMwere created. The DM was integrated with eight individuals: two representatives of the board of trustees, two from the board of directors, two from the management team, one representative of the service areas, and the head of the economic and financing area. With respect to the WG, a WG was created for each of the following three stakeholder groups: workers, families and users, and common and service areas. Each WG was composed of ten representatives. The working process followed by the DM and the WG is synthetized in Fig. 5 and is explained in more detail below.

The first step-the definition of the strategies (alternatives in ELECTRE language)
and criteria-was completed by the DM, using the discussion group technique because it allowed cooperative work in an open and flexible scenario. They sought to generate a process of feedback among the participants that would lead them to assume responsibility as a group. The technique was appropriate because it gave prominence to the group. A moderator directed the conversation by opening dialogue and agreement spaces, but was not the engine of the debate.
This phase was essential because the information generated was absolutely necessary for carrying out the subsequent stages. It is worth noting that behind each alternative there was a wide strategic action plan that included, among other item, work lines, timing diagrams, environmental implications, and financial, economic and human resources. During this phase, the different criteria that should be maximized or minimized, as well as how to quantify them, were also established. The criteria laid down by the DM were the result of its qualitative evaluation of the future of AdP, so a previously established scale was necessary to quantify them. In this situation, a group could be compared by evaluating the level of its members' similarity-dissimilarity (Rogers et al. 2000) or by establishing an ordinal scale (Maystre and Bollinger 1999).
In the case of AdP, an ordinal scale was established, with a quantification range from zero to five, where zero indicated no link between criterion and strategy, while five indicated that the strategy in question accentuated the criterion.
The extent of AdP reference territory-where the institution may wish to spread its activities and services-was also resolved during this phase. The territory decided upon included the counties of the Catalonia Autonomous Community within a maximum distance of 100 km from AdP headquarters.
The DM also agreed upon the criteria-specific weightings. Weight allocation of criteria is essential in the ELECTRE II process. Among several possible techniques (Simos 1990), the "cards method" procedure revised by Figueira and Roy (2002) was chosen. This method is well adapted to ELECTRE II and helps stakeholders to think rigorously about how they wish to rank the different criteria in a given context.

2.
Once the DM had established the alternatives, criteria, and weights, the WGs stepped in. All of the information associated with the work by the DM was made  available to the WG members so they could prepare for the work sessions. It was hoped that each WG would use the information to construct its own decision matrix. This was done using the focus group technique, because the work to be developed required a more active role from the moderator in conducting the session in a more directive way, stimulating the group, and leading the group to achieve a decision matrix. 3. The DM assessed each WG's decision matrix and each criterion. If the scores awarded by the WGs showed fairly low dispersion, the final criterion score was the average. If the scores showed fairly great dispersion, the DM discussed which score would be the most suitable. Considering all of the information and the DM's knowledge of the association, if any of the final scores were thought unsuitable, the DM could discuss those scores and agree upon a different score. 4. Then all of the previous information was gathered to apply the ELECTRE II methodology and to reach the final decision.
The above working process allowed AdP to decide the future strategy of the organization in a democratic way. We present the results of the working process by following the four steps enumerated in section three.
Step 1. Select the strategies (alternatives) and rank them according to different criteria that may be in conflict (A i ; i = 1,2…n). There was an important number of assumable strategies. The discussion process (Fig. 5), implied that the eight   Table 1 were the ones that were finally taken into consideration.
Step 2. Select the criteria for ranking the alternatives (C j j = 1,2,…,m). Following the Balanced Scorecard (BSC) system, as a performance measurement system, strategy evaluation system, and communication tool (Grigoroudis et al. 2012), four criteria perspectives were defined: 1. Financial: The financial perspective was designed to guarantee that AdP would be able to efficiently operate in the future, so the criteria were focused on the longterm viability of AdP. 2. Users: The user perspective mainly refers to the quality of the association services and are linked to the main role of AdP: caring for disabled people. 3. Internal institution: This perspective wants to reflect the extent to which users, families, and workers feel like an integral and fundamental part of the association and feel that they are valued and their opinions considered. 4. Learning and growth: The learning and growth is oriented towards reaching the best option for generating sustainable growth.
Each criterion was clustered into the perspective in which it fit best. Table 2 presents the perspectives and the criteria while indicating whether these would be maximized or minimized in the multi-criteria model.
Step 3. Determine the relative weights of the criteria (W j ; j = 1,2….m). The relative weights agreed upon by the DM are shown in Table 3.
Step 4. Apply the ELECTRE II multi-criteria evaluation method to select the most suitable strategy according to the scores and weights given to the different criteria. The normalized decision matrix is shown in Table 8 in the Appendix. The decision matrix and the weights drive the concordance and discordance matrices (Table 4 and Table 5, respectively).
Given these matrices, the aggregate dominance matrix (Table 6) and the mix evaluation value (Table 7) were deduced.
The mix evaluation value indicates that the best alternative is A7, and the sensitivity analysis shows the same result, providing robustness to this finding. This means that the best possible strategy for AdP in the future lies in maintaining the present institutional common services, empowering the philosophy of the Social Economy philosophy, and supporting territorial growth.
Finally, we want to highlight some of the main actions derived from the whole process that we have explained and which have enabled a new and more flexible development model.
& A people-centred organization model (PCO) was implemented in AdP's areas. The services offered and the groups assisted were expanded following the PCO model. & In the area of leisure and spare time, the supply of inclusive leisure services was increased and extended in the territory.
& An "insertion service" was implemented to give answers regarding the new employment needs of groups at risk of exclusion. & AdP spread its services along its territory of reference by opening two centres to attend to disabled people closer to their homes. & The figure of "facilitator" was created: a professional who accompanies and contributes to providing a more individualized care plan to disabled people. & The different areas work together to create sub-areas that undertake cross-cutting projects. & The empowerment of the stakeholders has been strengthened. Their participation in AdP bodies was increased; the participation of the different voices of the organization was encouraged through the creation of meeting spaces; and selfmanagement groups were created within the AdP areas. & AdP took measures to increase its economic sustainability through creation of new services and business lines, both in the area of attention to people and in the special employment centre. & Labour well-being was improved with actions regarding comfort in the working spaces, training, tools, and machinery. & The social and solidarity economy was promoted by strategic alliances and networking in the territory, through the organization of business conferences, promotion of actions to involve local companies in Social Economy projects, and participation in seminars to raise awareness of corporate social responsibility.

A1
To change the present institutional common services by giving more economic and financial autonomy to each institutional area, but preserving the common services, empowering entrepreneurial activities, and strengthening territorial growth

A2
To change the current institutional common services by given more economic and financial autonomy to each institutional area, but preserving common services, empowering entrepreneurial activities, and taking an institutional policy of moderate territorial growth

A3
To change the current institutional common services by given more economic and financial autonomy to each institutional area, unlinking the entrepreneurial activities of the institution, empowering the philosophy of the Social Economy, and strengthening territorial growth

A4
To change the current institutional common services by given more economic and financial autonomy to each institutional area, unlinking the entrepreneurial activities of the institution, empowering the philosophy of Social Economy, and taking an institutional policy of moderate territorial growth

A5
To maintain the present institutional common services, empower entrepreneurial activities, and enhance territorial growth

A6
To maintain the institutional common services, empower entrepreneurial activities, and take an institutional policy of moderate territorial growth A7 To maintain the institutional common services, empower the philosophy of Social Economy, and enhance territorial growth

A8
To maintain the present institutional common services, empower the philosophy of Social Economy, and take an institutional policy of moderate territorial growth  In short, and schematically, the new model has implied changes in management and in the organizational and quality systems in topics such as the following: & Providing spaces and empowering the different AdP stakeholders to define new priorities and services, guaranteeing economic viability and growth in the AdP territory. & Increasing the channels of decision, contribution, and bidirectional participation in the definition of AdP's organizational issues. & Enhancing the strengths and skills of AdP's people to guarantee their capacities and well-being. & Having the resources and management support to carry out strategic and crosscutting projects.

Conclusions
Our study provides evidence that the decision-making tools used in private companies can be useful in the Social Economy. This does not mean copying objectives, but rather about borrowing the tools that private companies employ to improve their competitiveness and maximize benefits. The Social Economy, like capitalist companies, operates in constantly shifting economic environment, which requires a continuous review of their actions to define a viable future. This was particularly true of the 2008 crisis, because entities within the Social Economy did not show as much countercyclical behavior as in other downturns (Jaén 2017;Sala Rios et al. 2014;Sala-Ríos et al. 2018). The figures indicate that the subsector suffered a credit crunch, as public funds fell, and a loss of employment at a time when there was an increasing need to meet most basic social demands.
This scenario demonstrates that many of NMS entities need to refocus their organizational structures and redefine their strategies to become more flexible and resilient to economic shocks and the austerity policies that often follow such shocks. It is on this point that we propose the ELECTRE II methodology, widely used by private companies in decision-making, as a rigorous and efficient option that, furthermore, fits with the democratic participative process inherent to the philosophy of institutions within the Social Economy. Our proposal was about providing entities in the Social Economy  with a tool that would enable them to move forward towards new organizational formulas, which would be more flexible and adjustable to recessions, and do so with the active involvement of the different agents within the entity's project. The goal of this study has been achieved. Following a working process previously designed, and applying the ELECTRE II, it has been possible to select the most suitable strategy according to stakeholder preferences for a non-profit association named AdP, whose main goal is taking care of disabled people. Using ELECTRE II, AdP was able to decide how to align its business activities, territorial growth, and Social Economy philosophy.
We would like to highlight some limitations of this study. First, we have not gone into great detail about the wide range of strategic action plans, work lines, run times, resources, or environmental implications that hide in each alternative, and especially in the alternative chosen, because this would be beyond the scope of our research. Second, we have pointed out the advantages of both MDM and ELECTRE; however, there are also some disadvantages. With respect to MDM, the main weaknesses arise from the difficulty of choosing the stakeholders and timing their participation. Also, producing consistent scores can be difficult: some options could be better against one criterion but worse against another, and weighting becomes necessary to handle such cases; therefore, the subjectivity can be high. If we focus on ELECTRE, the outranking relation is not suitable when alternatives are very different because setting the preferences becomes very difficult. It requires an additional threshold to be introduced, and the ranking of the alternatives depends on the size of this threshold (Sabaei et al. 2015).
We have become aware of some future research lines that may follow from this practical application. In our opinion, MDM must first be applied to other internal areas of the Social Economy, such as, for example, the selection of environmentally sustainable processes, suppliers, new investments, or new markets. Secondly, once we have a sufficient background in this field, we will be able to go into an analysis of results and compare them with the past and with those obtained by capitalist companies to detect whether Social Economy organizations have increased their efficiency without losing their values and principles.
Finally, in our opinion, another field of future research should be oriented to the study and analysis of public measures to support the Social Economy. The economic growth model must be geared towards a more sustainable development in which the Social Economy has to play a more leading role. This paradigm shift demands a greater deployment of public measures that defend the Social Economy, not only to prevent turmoil from future crises such as the 2008 crisis, but also because it is necessary to defend it from capitalism that resists letting in new actors with values other than profit (Chaves and Savall 2013). That is why future research lines should focus on different public policy alternatives to bolster the Social Economy. The decision-making process could be supported by the multi-criteria decision-making methodology, and in this development Social Economy agents should have a key and active role.