A systems approach to exploring institutional absorptive capacity and adaptability

David Vaandrager
Erasmus University Rotterdam, NLD

Abstract

An important ability of institutions is their capacity to recognize new external knowledge so that they can revitalize outdated routines and stay clear of institutional lock in. However not much is known about what determines the absorptive capacity of institutions. By analyzing mental models of agents within an institution through Group Model Building. We contribute to literature on absorptive capacity and show how institutions could potentially increase absorptive capacity at the micro agent-level while circumventing problems of institutional lock-in. The experiment shows that Group Model Building provides an avenue for reflecting on the internal logic of the institution. It is able to show the limited applicability of the internal logic in producing a desired outcome for certain messy problems. Although the agents acknowledged the shortcomings of the internal logic, they also accepted them and found dealing with them too complex. The article also reflects on how the experiment can be made more effective for future endeavors.

Introduction

An important ability of institutions is their capacity to recognize new external knowledge (Nandakumar et al., 2014) so that they can revitalize outdated routines and stay clear of institutional lock in. When an institution recognizes new knowledge, it needs to internalize it through a process of acquisition, assimilation, transformation and exploitation (Cordes-Berszinn, 2013). Only when knowledge is truly internalized can we speak of knowledge absorption. Knowledge absorption in turn is necessary for an institution to adapt to its changing environment. However not much is known about what determines the absorptive capacity of institutions and why some seem to hold on to their routines even if those routines are outdated (Maskell & Malmberg, 2007). In fact, well established institutions can survive even if their outdated routines make them obsolete.

Understanding absorptive capacity and the way institutions change presents us with certain challenges. One of the most important challenges is the duality of structure and agency (Bathelt & Glückler, 2014). As institutions become ingrained into society, they form a macro level structure that guides our actions by steering our way of thinking into certain directions. Our actions might seem natural and self-evident when in fact they are being steered by institutions. Although the structure influences its agents the agents at the micro level also influence the structure (Maskell & Malmberg, 2007). In fact, this micro level is vital as organizations are only able to learn through individuals who learn (Senge, 2006).

The downward causation from the macro structure level towards the micro agent level and the upward causation from the micro agent level to the macro structure level leads to the conclusion that an understanding of institutions can only be obtained by analytically extracting it from social practice (Bathelt & Glückler, 2014). Although downward causation is relatively well-known few contributions in the social sciences exist about how institutions can be transformed at the micro level of individual action (Maskell & Malmberg, 2007). Senge (2006) presents us with a possible solution to this problem, he states that although macro structures hold us prisoner beyond our own awareness, once we are able to see those structures and name them, they no longer have the same hold on us. Our goal is to test this assumption. By uncovering the structure in a case-study we hope to see if it is in fact true that it will help agents to acknowledge and absorb more information from the changing environment than would have been the case without any awareness of that structure holding them prisoner. This would place the institution in a better position to help it adapt to its ever-changing environment.

In this article we study the case of the municipal authority of Rotterdam an institution confronted with new external knowledge due to its changing environment. We give an empirical account by using a social practice approach while we seek to determine the link between actions of agents within their organizational context. To start we will give a short overview of our case. Next, we will discuss institutional theory as it contains insights into the evolution of the institutional macro structure and its information processing mechanisms. We will also take a look at systems theory, where we will discuss the concept of mental models in order to get a clear understanding of how individuals at the micro level process information. Changing these mental models is vital to the learning process and creating the all-important awareness of the macro structure. The actual changing of those mental models will be achieved by applying the Group Model Build method, which will be discussed in a methodological paragraph. Once method and theory have been established, we present our results. By analyzing mental models, we contribute to literature on absorptive capacity at a micro level and how institutions can increase the potential for transformation at the micro level, circumventing problems of macro structure lock-in.

The case

In their quest for more effective and efficient management tools spurred on by budget cuts the strategy department of the Rotterdam municipal authority asked the researchers to develop a control panel in an experimental setting that could offer the possibility to 'steer' society. This control panel was to be based on an approach rooted in systems theory. The idea was based on an assumption that if the researchers managed to teach the strategists some basic skills in approaching a problem systemically, they themselves would become aware of the existing organizational macro structure and would recognize the sort of data that would be needed to rethink that structure or in this case the levers of the theoretical control panel. In order to accomplish this the data would have to go beyond the preexisting assumptions of the institute and in theory create new insights into the influence of the institute on its environment. To keep complexity manageable, we asked the strategist to pick one specific issue on which municipal authority wanted to exert its influence, the issue the strategists choose was that of maintaining cleanliness in public space. A workshop was held where the basic systemic skills were taught after which the strategists were given an assignment to venture out into the environment and gather information on their chosen issue from different perspectives. The exercise was meant as an experiment and no daunting accountability had to be given: free thinking was all that was required. At least this was true within the settings of the experiment. Our initial effort uncovered the boundaries of the institutional macro structure. A further effort was initiated to extend those boundaries. The results of which are presented here.

Institutions: processing information from the environment

The human ability to simplify situations is a very useful skill. As humans we try to deduce patterns and organize complexity to make it manageable and inform our actions. It streamlines our decision-making process and can generate quick and simple answers (Levinthal & March, 1993). Although this is a very useful skill, it runs the risk of creating what H.L. Mencken is misquoted as saying: For every complex problem there is an answer that is clear, simple, and wrong (Below, 2017). Actions that are informed by clear and simple answers tend to generate a host of surprises and paradoxes (Hood & Peters, 2004). So next to the ability of generating quick and often useful answers individuals are also prone to typical mistakes they have their typical limitations and their own blind spots (Drucker, 1980). This is no different for organizations where typical mistakes emerge from two mechanisms that facilitate learning from experience. The first is simplifying experience, which restricts complexity and the second is specialization which focusses on fine tuning certain adaptive responses to specific situations. Although this form of learning will on average lead to improvements in organizational performance at the same time it will limit improvements. As previous improvements become standard practice it causes a disincentive to look for any further improvements which in turn leads to different forms of myopia, particularly ignoring the long run, ignoring the bigger picture and a tendency to overlook failure (Levinthal & March, 1993).

In order to overcome these limits an organization needs to shape its expectation to reflect the accuracy of results or else it runs the risk that it does not learn what it does well or poorly. As long as the results do not violate the expectations held, perceptual distortions may continue to exist beyond awareness. Without a developed sense of self-awareness an organization lacks the ability to compensate for its typical mistakes and blind spots (Drucker, 1980). Organizations with an inability to shape expectations beyond their internal criteria are associated with a low absorptive capacity. In contrast organizations able to form their expectations on the basis of opportunities present in the environment are associated with a high absorptive capacity (Van den Bosch et al., 1999). The mechanism responsible for low absorptive capacity can be illustrated with the way we develop our worldview. The worldview we construct or receive through our culture is often perceived to be unquestionable: it is the aggregate of group experience that has passed the test and can be accepted as truth without any further scrutiny (Boisot, 1995, Lowndes & Roberts, 2013). This stance can create an inability to recognize anomalous or conflicting stimuli occurring in our environment, as we tend to fit stimuli to existing expectations or categories. When this occurs perceptions are primarily driven by internal disposition more than by the stimulus itself. An internal disposition can be persistent which has to do with how it evolves over time. In the next paragraph we will take a closer look at how an internal disposition evolves and how it can create the macro structures that steer behavior by using the work of Boisot.

According to Boisot (1995) any physical system be it a man, organization or a quantum of matter that performs work (acts teleonomically) is subject to the principle of least action. The principle of least action can be explained through an economic production function. A production function expresses the relationship between quantities of production factors or inputs (like labor, capital and land) used and the amount of product or output obtained. The principle of least action means that a physical system will seek a mix of inputs that minimizes the cost of producing a given level of output, given that those inputs have differing prices. In other words, it economizes on its inputs. In traditional economics the inputs or production factors are land, capital and labor, but these factors of production are insufficient to describe production occurring in 'any physical system' towards that end Boisot uses the production factors of space-time and energy. An emerging property of a physical system consuming inputs of space-time and energy is data. Data can be thought of as originating in discernable differences in physical states of the world. A physical system will only register data if its sensing mechanism is able to discern it in terms of space time and energy. It takes energy to register discernable differences, the amount of which will depend on how sensitive the systems sensing mechanism is. The resulting data are stored in the system as memory. However, systems are limited by the amount of data they can handle which depends on their capacity to detect, store and process data. It is at this point that data becomes the subject of economization by converting it into information. The conversion of data into information is done through codification (reducing the number of data attributes) and abstraction (reducing the number of categories that codified data attributes can be assigned to in other words: generalization or abstract values). Data serves to function as an input for the production of information, raw data only becomes information when it modifies the knowledge that a system possesses, in other words when it is considered informative. As information starts to build up within the system the physical system becomes more complex, this enhances the adaptive capacities and helps to maintain its autonomy. Very much like Ashby's principle of requisite variety which states that a controller (e.g., an institution) must be at least as complex as its environment to survive (Kubik, 2015). The absorptive capacity or amount of knowledge that a system is able to internalize offers an indication for its abilities to act under certain circumstances.

This evolutionary path starts at the micro level with one physical system learning by experience how to utilize the available resources in its environment most effectively. Now if our physical system would be an individual surviving on a remote and isolated island and we would add an individual with whom the original inhabitant would constructively interact, a division of labor could emerge. In other words, the two inhabitants would form a social system and this social system is subject to the same evolutionary process as any other physical system. As this social system evolves it builds routines through coding and abstraction which enable it to economize on processing information and to simplify the decision making process (Maskell & Malmberg, 2007). The routines it builds form an internal logic which is an important step towards the formation of institutions. It is thus possible to conclude that institutions at the macro level are the result of upward causation from agents at the micro level.

Institutions are considered to be stores of social practices or structures through which we externalize collective memories, representations, values, rules etc. in order that they shall outlast us (Boisot, 1995). Some public organizations like the municipal authority of Rotterdam are able to take on institutional characteristics, the condition being that they are able to combine a strong internal logic with high levels of legitimacy (Boin & Christensen, 2008).

Institutions are structures designed to economize on social information processing: they offer large-scale economies of mental and physical effort. Through codification and abstraction of knowledge, which save on data processing, institutions seek to structure and store knowledge that is considered to have order preserving properties. In this process codification is used to select what information is relevant and abstraction acts as a vehicle for its diffusion: the spread of its acceptance. The production and exchange of information are both induced by social and economic transactions. Transactions occur when knowledge and information are passed between an agent and its environment. The transaction of information must overcome technical, semantic and pragmatic barriers to count as effective communication. Transacting information is thus costly in terms of time-space and energy and for this reason only transactions that prove to be salient in their consequences for social life, or are recurrent, are considered for the effort at institutionalization (Boisot, 1995). In other words recurrent information flows give rise to transaction structures like organizations and institutions. The existing stock of transaction structures acts as an attractor to uncommitted information transactions, because using existing transaction structures will drop marginal transaction costs. This is the same mechanism that creates a predisposition to be bound by inherited traditions of the past: relying on those traditions is considered a rational solution for a cognitive investment problem in terms of their marginal costs.

Stafford Beer also concluded that storing, modifying and transmitting information is vital for the effective operation of a system, information through codification and abstraction reduces variety which is one of the main techniques for a system to regulate itself. The reduction in variety not to simplify the system but to make outcomes more predictable (Flood, 1999). Institutions as systems are in fact expressions of the fight against entropy or chaos, they should provide a stable order. This order is determined by the dominant groups that have influence over information flows that are transmitted through institutions (Boisot, 1995). Structural integrity is achieved by the subtle coercion of social behavior or socialization of agents to the institutional order. If social reality were contested by different subjective realities and identities, a situation might occur where everything is relative and anything goes. The stability and predictability institutions provide are essential, because without them collective action within society would be impossible (Gupta et al., 2010). While uncertainty is reduced and conceptual stability is achieved it does come at the cost of perceptual texture and richness (Evans & Wuster, 2000, Boisot, 2006). Codification and abstraction enable information to increase its communicative efficiency and reach but require the compression of data which reduces communicative effectiveness. The byproduct of codification and abstraction is an information residue which is a source for blind spots (Brans & Rossbach, 1997).

In the end all the investments that have gone into coding and storing knowledge will not insulate an institution from external shocks. As circumstances shift the commitment to the preservation of its structures runs the risk of making an institution obsolete.

Institutions are embedded in societies that are faced with globalization, the rise of communication, the internet, the movement of migrant groups and higher levels of education. This puts pressure on shared values, identity and cohesiveness as cultures clash and political savvy increases. Typically, these institutions with routinized practices and rigid self-preserving structures are ill equipped to deal with or even acknowledge these challenges of rapid change (Innes & Booher, 2010). Through the storing and recording of knowledge institutions build up a commitment to the entropy reducing internal logic that is at their core. Those with the deepest investments will be the most resistant to change (Boisot, 1995, Jarvie & Stewart 2018). The narrative that comes with the internal logic creates a self-sustaining frame of reference. A feedback-loop that implicitly confirms the truthfulness of perceptions and that constitutes a barrier towards institutional absorptive capacity.

Without the ability to question the underlying assumptions, there will be no incentive to enact change, making an organization inert (Geiger & Antonacopoulou, 2009). Institutions built on a core schemata with key assumptions about the world that hold it together, base their legitimacy on the wide spread acceptance of those assumptions (Innes & Booher, 2010). An institutionally held norm that does no longer fit the changing values of society can destroy the consensus underlying that norm (Boin & Christensen, 2008). In order to stay relevant an institution needs to be able to identify, reconcile and accommodate different values in their transaction structure. This is harder when the rules embedded in institutions are made more concrete. Vagueness of rules makes it easier for an institution to adapt and to accommodate more variety and changes. Abstraction should have a sufficient measure of generality that it can also accommodate a wide range of contingencies (Boisot, 1995). Furthermore, it needs a method enabling it to reflect on its own assumptions, which requires a counter intuitive move to acknowledging the often chaotic and complex environment. The environment is complex because the circumstances that shape the environment are often interrelated and dynamic. Trying to understand emerging situations in the environment by reducing those situations into smaller easier to understand parts ignores the interactions between those parts which are crucial in creating the emerging situation. Effectively dealing with the new situation thus requires an understanding of the whole situation and its underlying dynamics which requires a good sensing mechanism and developed absorptive capacity. However the institutionalized frame of reference we often use to understand the environment is the result of codification and abstraction where our understanding of the environment has been reduced to simplified models of reality, which can be effective as long as the environment is relatively static or it doesn't change in vital parts of our models. The consequence is that governments that deny complexity, deny themselves opportunities to absorb new information and with it effective institutional responses to the environment that they find themselves in. Ultimately a tightly ordered command and control regime stops evolving with its environment (Boisot, 2006, Allen et al., 2010).

It is clear that if an organization wants to improve its absorptive capacity it needs a way to loosen the institutionalized frame, but how? An important consideration is that an organization can only learn through individuals who learn (Senge, 2006). It is at this level that we need to unravel the mental models if we want to improve or even understand absorptive capacity. However, it is not quite as easy as that: not unlike institutions individuals use data to actively construct reality; they build mental models of their environment which form the basis for their behavior (Gerrits, 2012; Vennix, 1996). Those mental models are highly subjective and influenced by many factors such as, culture, upbringing, schooling and the institutions in which they are embedded. Again referencing Senge (2006) once we are able to see those institutional structures and name them they no longer have the same hold on us. So can we improve institutional learning by a structure uncovering effort, is it possible to devise an approach that can increase absorptive capacity?

A systemic approach to charting the complex environment

This research needs a method that can basically do two things: the first is to understand an issue stemming from the complex environment of the institution, without viewing it through the institutional frame of reference or through the existing internal logic. Because the internal logic is a potential source for the information residue that creates blind spots. The second is some way to discern how an institution or in our case individuals at the micro level of that institution sense and absorb information. It is the friction between the structure of the issue and the information it provides and the amount of information that the institute and its individuals can actually discern that is of interest here.

In line with scholars like Von Bertalanffy (1968), Beer (1984), Checkland (2011) and Ackoff (1979), Taleb and Blythe (2011) suggest the best way to chart a complex issue is by looking at it through a systemic lens. Systems theory offers a relevant approach because it does the opposite of codification: where codification tries to view the world as efficiently as possible systems theory tries to understand the entirety of circumstances under which phenomena take shape. The circumstances include the contextual richness and dynamics (or fluctuations) that have been lost in the process of codification and abstraction.

In systems theory all phenomena are a logical consequence of their circumstances. As we have demonstrated a phenomenon can be any kind of physical system like an institution or when looking at the environment it can be an emerging situation. As long as the phenomenon is made up of an interconnected set of elements that is coherently organized in a way that it achieves something (Meadows, 2009) we can also call it a system. In order for a system to generate a certain outcome the circumstances or its elements have to be organized in a specific way (Sterman, 2000). That specific organization gives the system its structure. It is the structure or the whole with its interdependencies and not the discrete parts that determine a systems outcome (Ryan, 2008). While a bureaucratic approach seeks to control outcomes often through narrow performance metrics (Hoggett, 1996, Dieffenbach, 2009) a systemic approach seeks to understand outcomes which are often no more than symptoms of the underlying structure (Sterman, 2000)

The understanding that outcomes or behavior are latent within the structure of a system, serves as a starting point for how the system works.

The systemic method used by the researchers with the strategy department of Rotterdam in constructing a control panel was that of Group model Building as developed by Vennix (1996). Group model building is a model building process where clients are involved in the construction of a model concerning a hard to grasp and often messy problem. The model is represented by causal loop diagrams that show the structure of the system responsible for producing a certain phenomenon. The causal loop diagram consists of elements or individual parts, which are connected through arrows that represent the relationships between those elements. Those relationships are flows of information between the elements, which keep the system together and determine how it behaves (Vaandrager et al., 2014). Relations between the elements are either positive or negative. A positive relation has an amplifying effect, a negative relation has an effect in the opposite direction or an opposing effect. An element can be connected to itself through relations with one or more other elements forming a loop of causal relations. These loops are also referred to as feedback loops. For example as humans we are constantly exposed to feedback loops: any action we take can change a situation and the resulting information we receive from that change (or lack thereof) serves as input for new decisions as Information is fed back to our senses (Sterman, 2000).

The method starts from the premise that the primary source of knowledge about the structure and its elements are the people involved in the system. People tend to interpret the same situation differently. Interpretations are mentally constructed, simplified and selective representations of reality (Vennix, 1996). These simplified representations or models are also known as an individual's mental model. Reality is actively filtered and what remains is determined by the short term, self-interest and values of an individual. This phenomenon is also called bounded rationality and denotes the limits of knowledge and computational capacity in decision makers (Simon, 1956). The mental models of the municipal strategists are subject to the internal logic of the municipal organization. The organization needs a dedicated staff around a shared mission. The shared mission in turn requires shared and aligned mental models of its employees (Boin & Christensen, 2008). It can thus be expected that the mental models of the strategist will show a strong congruence with that of the municipal internal logic.

To understand systemic structures and their outcomes it is necessary to go beyond individual assumptions. Challenging those assumptions and stretching the individual framework is called double loop learning (Herrfahrdt-Pähle & Pahl-Wostl, 2012). Awareness of the limited knowledge in the possession of an individual is best achieved by interacting with others who have another perspective. The need for differing perspectives often makes a Group Model building effort transdisciplinary. Sharing those perspectives does not lead to an objective truth, but it approaches it better than an individual could have done (Vennix, 1996).

Together with the strategists an attempt was made to build a model of the environment in focus group sessions. The sessions specifically targeted the messy problem of citizen responsibility with regards to cleanliness in public spaces. The systemic effort was to create an insight into the circumstances under which citizens would take up responsibility for the cleanliness of their own neighborhood. The strategists only managed to interview co-workers from the municipal authority and as a result we only got the mental models present within the institution and with it an insight into the workings of its internal logic.

Uncovering institutional barriers with the strategy department

The original effort at gathering data from the environment on the chosen issue of cleanliness in public space by the strategist had led to a disappointing result. When the researchers received the data resulting from the strategist's attempt, they were both surprised and disappointed. The strategist had failed to establish anything that could not have been established beforehand and they had shown no interest in perspectives that might have revealed otherwise. Something that was also evident from the fact that the strategists had almost exclusively interviewed people from within their own organization, while they were explicitly asked to venture out, interviewing different stakeholders including citizens. In short, an opportunity to learn and absorb new external knowledge from the surrounding environment was missed.

The information fell far short of what was needed to construct a control panel able to deal with a complex issue like maintaining cleanliness in public spaces. It did however present a first hint of an inability for these civil servants to absorb contextual richness. In itself an interesting conclusion that merited further exploration by the researchers. The original model of the presumed structure resulting from the strategist's efforts could be said to represent the way they perceived (or could perceive) the world filtered by their bureaucratic roles or the macro structure. A finding they themselves underscored by stating that it was not their premise to go beyond the barriers of their accountable institutional roles. The present bureaucratic system left no room for additional contextual richness. This triggered the researchers to study the model (fig. 1) and determine where the limits of their mental models lay and if these limits could be extended to include more contextual richness.

A first barrier occurred with regards to citizens' responsibility. The strategists held citizens in low esteem in the sense that they were viewed as being irresponsible. The consequences of citizen behavior were something that either needed to be countered or prevented and not something to be relied upon for anything like co-production. The dynamics leading towards certain behaviors by citizens were in general only understood superficially. They did make one interesting point: the strategists recognized the fact that the municipal authority had taken on a lot of responsibility in the past. A responsibility that had led them to a paternalistic role, due to this role citizens had started to view clean streets as a right as well as it having a negative impact on their sense of responsibility for a clean neighborhood. Ironically this in part created the lack of responsibility for which the citizens were scrutinized. But for the most part the lack of responsibility was seen as a result of the 'wrong' mentality with citizens.

This stance also resonated with the second barrier: If the government was to retreat alternative measures were deemed necessary to steer mentality in a direction that cleanliness standards as prior to the government retreat could be maintained. Those standards were set in concrete at cleanliness level 4, implying very little tolerance for litter. However, since most existing measures had self-admittedly proven ineffective, there was room for improvement. In addition, there was also a contradiction in expecting citizens' initiative (under the banner of self-organization) while at the same time the authorities seemed to be trying to achieve a very specific sort of behavior through correcting measures. Measures that should lead to a predictable and nonvolatile type of behavior leaving little room for creative initiatives.

The third barrier concerned knowledge about the dynamics of waste accumulation in public space. It was not clearly specified how it came into being and where the proverbial tap from which waste flows could be tightened. The only tap considered, was irresponsible citizenry and their behavior with regards to waste. But the main focus lay on the aspects that made it a concern for the municipal authority, squatting of the neighborhood, trash bags next to the container, a dirty city and how they could combat these symptoms.

The dynamics of citizen mentality were also ill understood. An understanding of citizen mentality is essential as it is the primary source for behavior regarding waste and it forms a central element for barrier one. In this case citizen mentality was only seen as a symptom to be controlled and not as an outcome to be understood. This was made clear by a remark that the municipal authority would have to establish a presence within the community so that it could set the standard for behavior regarding waste. It would be unacceptable for free riders to determine the measure of compliance to that standard. There was thus no room for volatility and only catalysts were named that could account for certain types of undesired behavior.

Figure1

Figure 1 The preliminary model and its blind spots

In order to get a sense of the contextual richness that was missing, researchers had a clear goal of seducing the strategists from their vested bureaucratic mindsets to see if we could get them to acknowledge information beyond the limits of their mental models. This second effort focused mainly on the first barrier of citizen responsibility and on the second barrier of the available means to steer citizen mentality. While exploring the complexity of these two barriers we also got a good sense of the internal logic of the municipal authority and how it shaped much of the mental models of the strategists. Figure 2 shows the model that resulted from the attempt at seducing the strategist to think beyond their mental models. Still nowhere near complete it reveals how much more relevant information from the environment can be acknowledged by critically reflecting on conventional concepts and their shortcomings. Acknowledgement however is still a big step removed from absorbing that information.

Figure2

Figure 2 The resulting model

The first barrier concerned the dynamics of responsibility: originally responsibility was mainly seen as the result of mentality. In order to create responsible citizens, influencing mentality was viewed as the main lever. Strategist had pointed out that there was something 'wrong' with civil mentality. An assumption that was made without considering the question: what motivates an individual to take responsibility? By addressing this question researchers tried to extend the limit of the mental model and trigger it to absorb more information from the environment. Initially strategists admitted there was no clear sense of what triggers people to behave responsibly. Some sought answers on an individual level, introducing financial stimuli, others thought that a sense of social cohesion accompanied by social control might lead to responsible citizens, still others made suggestions for a social engineering approach. Circumventing the question entirely an argument was made not to cut costs on cleanliness in public space as it can be considered a condition for responsible citizenship (see model via' quality physical context' and 'mentality' towards 'responsible citizens'). None of these responses satisfied the question because they were fixed on making citizens behave in a way that suited the ambitions of the municipal authority and not on citizens taking responsibility. There was a complete lack of sense what moved citizens, but a clear sense of what acceptable behavior should be. Up until this point it remained unclear why they thought responsible citizenry was so important for cleanliness in public space, once we asked this question it opened up a whole new realm of reasoning. The civil servants replied: "If it was just the cleanliness of public space we would be looking for, we might as well have gone with a technical solution: have the streets wiped two times a day by a crew of civil servants and be done with it. In fact, it might even be cheaper, but is that really the kind of society we want to live in? Politicians want a society where citizens have a level of commitment towards society, if you go with a technical solution you remove some of that responsibility". Responsible citizenry might have been the big underlying political ambition, the way it was executed could easily result in the complete opposite of responsible citizenry.

In research on citizens taking responsibility (see Arnstein, 1969; King et al., 1998) a crucial element is the real power of citizens to affect the outcome, participating without power often just frustrates citizens (Arnstein, 1969; King et al. 1998). In this case the outcome as to what constitutes an acceptable public space and how it should be maintained was based on a political decision. Even the amount of citizen initiative involved was stated to be a political choice (figure 2: via 'political motivation' and 'consensus' towards 'municipal responsibility' and from 'municipal responsibility' through 'political influence citizens' towards 'responsible citizens'). This effectively leaves citizens without true responsibility and ownership, which in turn makes them passive and less inclined to initiate communal projects, giving local government a further trigger to steer behavior in a way that reinforces lack of initiative, forming a reinforcing loop (Figure 2: 'civic initiative' and 'civil commitment' which through various causal relations trigger 'political motivation' and 'municipal responsibility' ultimately impacting on 'mentality' and 'responsible citizens') . In other words, this local government organized its own distrust regarding the capabilities of their citizens. Putnam (2002) describes a similar loop where a lack of engaging citizens in public affairs leads to a deterioration of their social capital which undermines institutional stability and effectiveness. The resulting political apathy only increases social distrust which acts to increase political apathy even further. Without the ability to reflect, the feedback confirms wrongful assumptions about distrust and can remain intact (Drucker, 1980; Geiger & Antonacopoulou, 2009). Ultimately the effort did provide an opportunity to reflect on the existing mental models and the way the institute organized its own distrust. The reinforcing loop showed the strategists they needed a counter intuitive move towards trusting citizens and letting go. The strategists identified these mechanics of the system themselves: "personally I am of the opinion that if government dares to let go and gives citizens more responsibility very nice things can happen and if government fails to muster the necessary courage citizens will remain passive and inert". We also managed to trace the question that lay behind the need for cleanliness in public space and responsible citizenry: "We want responsible citizens because we as citizens, voters and civil servants do not want to live in a society where anonymity and lack of commitment are rampant next to the fact that the political system is allergic to the idea of leaving people behind". Although we were starting to make some advances in identifying relevant mechanisms for creating responsible citizens the measures taken to make people responsible up to that point were steering the system away from the intended outcomes. A major shortcoming still in place was a lack of knowledge on what moved citizens, a problem that was not helped by the fact that no citizens were interviewed, these shortcomings also showed in the second barrier.

The second barrier concerned the question in which way the 'right' mentality could be achieved. Mentality was seen as the result of upbringing, individualism, knowledge, social economic status, social control, and a sense that civilians regarded municipal services as a right. Some of which could be influenced with the municipal toolkit according to strategists. Individualism and social control were seen as two aspects for which the municipal authority lacked the tools at least in an instrumental sense. The only thing at the disposal of the municipal authority with a possible indirect effect was the housing supply. The reasoning was that if you build homes meant for permanent residence the flow of residents through the district could decrease. With more permanent residents in place a homogenous community could develop which would ultimately increase social cohesion and social control. Although this was self admittedly seen as a drastic measure without a guarantee for success. A well-established community would also make it easier to increase awareness on the need for cleanliness by addressing community leaders on the issue. Informing citizens on the necessity of cleanliness and responsibility could in general be achieved by launching promotional campaigns or educational programs. However increasing knowledge on certain topics does not automatically lead to the desired behavior, while at the same time it is one of the few measures that the municipal authority actually had at its disposal.

The prevalent mentality that many citizens regarded public services as a right and not as their duty was viewed as problematic. A situation that had historically grown from the fact that public hygiene had been deemed a lawful task of the municipal authority. The municipal authority had to guarantee a minimum level of cleanliness of the public space and this in turn had led to 'spoiled citizens' according to the strategists. New rules and legislation could be enacted to convince citizens to take responsibility for public space themselves. Two forms of rules and legislation were considered: those with a deterring function in the form of fines and those with a stimulating function in the form of rewards for producing less trash. New rules and legislation were disregarded as they could cause some perverse effects like citizens simply dumping their trash elsewhere.

Convenience was another important aspect of influencing mentality. Convenience was related to the amount and size of the trash disposal units available in a certain area, but the optimal size and amount was highly debated. Make them too big and numerous and you might create an incentive to increase waste production, make them to small and scarce and you might create more waste in the streets.

The net-effects of these measures were ill understood, there was not a clear sense of what triggers citizens and as a result making decisions was often not knowledge driven. All in all the strategists admitted that the municipal authority often operated from a gut feeling, looking for means they actually had at their disposal. No surprise than that the municipal toolkit in the past twenty to thirty years had proven it could not always deliver the envisioned results. Even so strategists were satisfied with using the means available, because at least they could show they were actively involved in combatting the problem and accountability within the organization could be given. In the end they did not feel responsible because ultimately, it's a political choice on which they have no influence. With regards to influencing mentality in obtaining responsible citizenry a strategist said: "it's an ambition that is heard every now and then: something we have to live with, although the goal is unattainable". An ambition which is reinforced by the fact that a politician would never be seen admitting that a problem is unsolvable. Although the strategists recognized the need for trust and letting go, as soon as we discussed the measures at their disposal, they were perfectly comfortable with accepting that nothing could be done. The routines of the system at the macro level dictated the kind of ambitions that could be realized and not the dynamics and information from the issues in the environment. Even if a systemic effort is able to show the limits of the mental models in use, it does not mean it automatically changes the reality of the macro structure in which the civil servants operate. At this point the force of the macro structure pulling the civil servants back in is stronger than the realization that they need to extend the limits of their mental models. Something more is needed.

The picture that we are left with is reminiscent of Boisot's (1995) institution trying to enforce its rules and practices by using tools of coercion. The strategists accept their roles as executors of policies in this system. The coercion is meant to push behavior within society in a direction that aligns it with the core values of the institution. A possible problem it could encounter is that the norm embedded in the institution and the values that citizens hold are different and the institution has few means of getting to grips with those values. Individualism, higher education and migrant groups create a fragmented society that is less inclined to obey centrally set rules. The ability to accommodate values is not only stifled by socially engineering the ideal citizen but impaired even further by the formulation of very concrete and stable rules, like the desire for cleanliness level 4. The limited scope in which the problems were viewed led the researchers to a third barrier.

The bureaucratic organization consists of several departments all with their own responsibilities, a problem emerging from its environment has to lie within the scope or internal logic of one of these departments. A problem outside of that scope theoretically does not exist. Even if a problem is identified it has to adhere to the bureaucracy, it has to fit political priorities, departmental vision statements, and budgetary labels. The problem is thus defined by the bureaucratic scope and not by its dynamics. If the bureaucracy due to its own structure is not able to target the underlying dynamics it risks a loss of control. This observation can be explained by the mechanism of institutionalization set out in our theoretic chapter. As systems, institutions code and abstract information. The mere act of codification is always a partial truth as some information is discarded in the quest for transactional efficiency. As the order that an institution exerts grows it will increase control over the diffusion of information. In this scenario alternative possibilities will have been eliminated, phenomena can appear to be unproblematic as interpretation and the course of action to be taken are clear. This can create a narrow scope through which it is no longer possible to identify and respond to phenomena outside that scope. This mechanism which Boisot (2006) calls the fossilization of the regime is responsible for creating blind spots.

Discussion

This exercise has taught us a few valuable lessons on the crucial elements that were missing for upward causation and breaking through the macro structure. Our goal was to challenge the key assumptions within the organizational internal logic. This could have broadened the institutional scope making it more receptive to information emerging from its environment, increasing its adaptive capacity to that environment. There were circumstances that made it very hard to do so (Gerrits & Vaandrager, 2018):

Conclusion

In theory exposing the systemic structure of issues emerging in the environment of institutions should have helped the strategists to acknowledge and absorb new information. The recognition of this information could have been the beginning of a new evolutionary path for the organization through upward causation. It could have made the organization adapt to its changing environment. The effort fell short in the sense that the researchers never managed to enter an evolutionary path that would lead to the adaptive control panel. This does not mean the strategist were completely dissatisfied with the method. According to an internal evaluation conducted by the municipal authority afterwards the strategists were positive about several aspects (Gemeente Rotterdam, 2013):

The conclusion is that although you can make the strategists look at emerging issues in a different light even making them reflect on how things have been done up to that point and creating awareness for the fact that existing policies haven't been effective it is not enough to make them escape the forces exerted by the macro structure. To start to deal with this issue we will first take a look at how this force works.

By economizing on information transactions, institutions emerge as expressions of the fight against entropy (Boisot, 1995). The order they transmit is a perquisite for collective action and a way to sustain modern complex society (Gupta et al., 2010). The municipal authority of Rotterdam functions similarly: it wants to sustain and stabilize its environment. In this case it transmitted a norm of cleanliness that it wanted to enforce with subtle acts of coercion. It asked: what incentives does our structure provide to steer behavior in a way that it complies with the social order. In fact, it was very explicit about which cleanliness level it wanted: level 4.

The structure from which the municipal authority tried to formulate a solution is built like any other institution with the codification and abstraction of information. In this way institutions erect core schemata with key assumptions about the world that hold it together (Innes & Booher, 2010). Those key assumptions however are very rarely questioned and might cause a self-referential feedback loop (Geiger & Antonacopoulou, 2009). Furthermore, institutions might discard relevant information as the information does not fit the institutional scope or they may be prone to formulating very concrete codes making it more difficult to accommodate variety (Boisot, 1995). All these pitfalls were witnessed in our analysis of the strategist working for the municipal authority.

Although the systemic effort failed to produce a control panel, it did put the researchers in a position where they were able to reflect on the internal logic of the institution and how it poses barriers to understanding phenomena that emerge in its environment. Because in contrast to making information more efficient the systemic effort provides a way to understand the entirety of circumstances under which phenomena take shape.

An institution can develop a structure that has a narrow scope on reality, partially blinding it from the dynamics in its environment. The systemic effort that was conducted with the strategist unearthed some of the boundaries that are part of this structure. While cleanliness was clearly within the scope of the institution a lot of the circumstances that brought it about were not, simply because they were deemed beyond the municipal premise, or considered to be unsolvable. Examples range from the motivations of citizens, to the influence of politicians and the inadequacies of the existing municipal toolset. The mechanisms of the self-referential feedback loops, the concreteness of rules, and the narrow structural scope keep these boundaries firmly in place.

The systemic effort provided an avenue for reflecting on the internal logic of the municipal authority of Rotterdam. It was able to show the limited applicability of the internal logic in producing a desired outcome. Confronted with this limited applicability the strategists acknowledged the shortcomings of the internal logic, however they also accepted these shortcomings and found dealing with them too complex. So unlike Senge (2006) suggests even though we were able to uncover and even discuss the macro structure it did not decrease its hold over the strategists. The simple conclusion is much more is needed. At the same time this effort has shown that even a slight systemic effort can uncover information on emerging issues leading to insights on why policies fail and what could be relevant areas of intervention previously left unobserved. Though the municipal authority eventually would not continue their systemic endeavor, in a closing report on the findings they recognized the method could have its added value.

References

Ackoff, R.L. (1979). "Resurrecting the future of operational research," Journal of the Operational Research Society, ISSN 0160-5682, 30(3): 189-199.

Allen, P.M., Strathern, M., and Varga, L. (2010). "Complexity: The evolution of identity and diversity," Complexity, Difference and Identity, ISBN 9789048191864, pp. 41-60.

Arnstein, S.R. (1969). "A ladder of citizen participation," Journal of the American Institute of Planners, ISSN 0002-8991, 35(4): 216-224.

Bathelt, H. and Glückler, J. (2014). "Institutional change in economic geography," Progress in Human Geography, ISSN 0309-1325, 38(3): 340-363.

Beer, S. (1984). "The viable system model: Its provenance, development, methodology and pathology," Journal of the Operational Research Society, ISSN 0160-5682, 35(1): 7-25.

Below, B. (2017). "Out of complexity a third way?" in P. Love and J. Stockdale-Otárola (eds.), Debate the Issues: Complexity and Policy Making, ISBN 9789264271531, pp. 31-33.

Boin, A. and Christensen, T. (2008). "The development of public institutions reconsidering the role of leadership," Administration & Society, ISSN 0095-3997, 40(3): 271-297.

Boisot, M.H. (1995). Information Space. A Framework for Learning in Organizations, Institutions and Culture, ISBN 9780415114905.

Boisot, M.H. (2006). "Moving to the edge of chaos: Bureaucracy, IT and the challenge of complexity," Journal of Information Technology, ISSN 0268-3962, 21: 239-248.

Brans, M. and Rossbach, S. (1997). "The autopoiesis of administrative systems: Niklas Luhmann on public administration and public policy," Public Administration, ISSN 0033-3298, 75(3): 417-439.

Checkland, P. (2011). "Autobiographical retrospectives: Learning your way to 'action to improve'—The development of soft systems thinking and soft systems methodology," International Journal of General Systems, ISSN 0308-1079, 40(05): 487-512.

Cordes-Berszinn, P. (2013). Dynamic Capabilities: How Organizational Structures Affect Knowledge Processes, ISBN 9781137351272.

Dieffenbach, T. (2009). "New public management in public sector organizations: the dark sides of managerialistic 'enlightenment'," Public Administration, ISSN 0033-3298, 87(4): 892-909.

Drucker, P.F. (1980). "The deadly sins in public administration," Public Administration Review, ISSN 0033-3352, 40(2): 103-106.

Evans, P. and Wuster, T.S. (2000). Blown to Bits: How the New Economics of Information Transforms Strategy, ISBN 9780875848778.

Flood, R.L. (1999). Rethinking the Fifth Discipline: Learning Within the Unknowable, ISBN 9780415185301.

Geiger, D. and Antonacopoulou, E. (2009). "Narratives and organizational dynamics: Exploring blind spots and organizational inertia," The Journal of Applied Behavioral Science, ISSN 0021-8863, 45(3): 411-436.

Gemeente Rotterdam (2013). Notitie: Evaluatie en vervolg traject System Dynamics.

Gerrits, L. (2012). Punching Clouds: An Introduction to the Complexity of Public Decision-Making, ISBN 9781938158001.

Gerrits, L. and Vaandrager, D. (2018). "Group model building in a pressure cooker: a field experiment with public policy," Systems Research and Behavioral Science, ISSN 1099-1743, 35(1): 139-151.

Gupta, J., Termeer, K., Klostermann, J., Meijerink, S., Van den Brink, M., Jonge, P., Nooteboom, S., and Bergsma, E. (2010). "The adaptive capacity wheel: A method to assess the inherent characteristics of institutions to enable the adaptive capacity of society," Environmental Science & Policy, ISSN 1462-9011, 13(6): 459-471.

Herrfahrdt-Pähle, E. and Pahl-Wostl, C. (2012). "Continuity and change in social-ecological systems: The role of institutional resilience," Ecology and Society, ISSN 1708-3087, 17(2).

Hoggett, P. (1996). "New modes of control in the public service," Public Administration, ISSN 0033-3298, 74(1): 9-32.

Hood, C and Peters, G. (2004). "The middle aging of new public management: Into the age of paradox?" Journal of Public Administration Research and Theory, ISSN 1053-1858, 14(3): 267-282.

Innes, J.E. and Booher, D.E. (2010). Planning with Complexity: An Introduction to Collaborative Rationality for Public Policy, ISBN 9780415779326.

Jarvie, W. and Stewart, J. (2018). "Conceptualizing learning in the public sector: the importance of context," International Journal of Public Sector Management, ISSN 0951-3558, 31(1): 14-30.

King, C.S., Feltey, K.M. and Susel, B.O.N. (1998). "The question of participation: Toward authentic public participation in public administration," Public Administration Review, ISSN 0033-3352, 58(4): 317-326.

Kubik, G.H. (2015). "Alpha societies and requisite variety: A projected framework for governance, education, and work in the 21st century," On the Horizon, ISSN 1074-8121.

Levinthal, D.A and March, J.G. (1993). "The myopia of learning," Strategic Management Journal, ISSN 1097-0266, 14(S2): 95-112.

Lowndes, V. and Roberts, M. (2013). Why Institutions Matter: The New Institutionalism in Political Science, ISBN 9780333929544.

Maskell, P., and Malmberg, A. (2007). "Myopia, knowledge development and cluster evolution," Journal of Economic Geography, ISSN 1468-2702, 7(5): 603-618.

Meadows, D.H. (2009). Thinking in Systems: A Primer, ISBN 9781603580557.

Nandakumar, M.K., Jharkharia, S., and Nair, A.S. (eds.) (2014). Organizational Flexibility and Competitiveness, ISBN 9788132216674.

Putnam, R.D. (2002). Democracies in Flux, ISBN 9780195171600.

Rouwette, E.A. (2012). "Does Group Model Building Work? Evidence from and Comments on the Paper by Videira et al.," Systems Research and Behavioral Science, ISSN 1099-1743, 29(6): 620-623.

Ryan, A. (2008). What Is a Systems Approach?

Senge P.M. (2006). The Fifth Discipline: The Art and Practice of the Learning Organization, ISBN 9780385517256.

Simon, H.A. (1956). "Rational choice and the structure of the environment," Psychological Review, ISSN 0033-295X, 63: 129-138.

Sterman, J.D. (2000). Business Dynamics: Systems Thinking and Modelling for a Complex World, ISBN 9780072389159.

Taleb, N.N. and Blyth, M. (2011). "The black swan of Cairo: How suppressing volatility makes the world less predictable and more dangerous," Foreign Affairs, ISSN 0015-7120, 90(3): 33-39.

Vaandrager, D., Gerrits, L., and Bressers, N. (2014). "Reconstructing the Blue Connection: a systemic inquiry into the improbable project of building a new river through a (sub)urban area," Systems Research and Behavioral Science, ISSN 1099-1743, 18: 689-706.

Van Den Bosch, F.A., Volberda, H.W., and De Boer, M. (1999). "Coevolution of firm absorptive capacity and knowledge environment: Organizational forms and combinative capabilities," Organization Science, ISSN 1047-7039, 10(5), 551-568.

Vennix, J.A.M. (1996). Group Model Building, Facilitating Team Learning Using System Dynamics, ISBN 9780471953555.

Von Bertalanffy, L. (1968). General System Theory, ISBN 9780807600153.