EcDev Journal

The System of Economic Development and its Complexity

Posted on Monday January 25, 2021
Figure 1
Figure 1
By: Kayleanna Giesinger, kgiesinger@uwaterloo.ca; Jeffrey Wilson, Jeffrey.wilson@uwaterloo.ca; Corresponding Author, kgiesing@uwaterloo.ca

Abstract

A local economy is a highly networked complex system made up of stakeholders, actors, and influencers. The very nature of a system is that its behaviour is anchored by its state, which is made up of a series of parameters and characteristics. If the mechanisms of a complex system are adequately understood, they can be manipulated to produce change within a community. The role of the economic developer and city planner comes in the form of targeted interventions to achieve positive autocatalytic changes in communities. In this article, commonly used complexity analysis tools, Systems Maps, and Causal Loop Diagrams are redeveloped for the application of economic development. Then, two case studies, Detroit, Michigan, and Kitchener, Ontario, are presented to assess the feasibility of applying complex systems techniques to inform economic development decision making. Also, to extract lessons to advance broader integration of systems analysis techniques into city planning efforts.

Key findings from the analysis include the importance of utilizing feedback loops and identifying direct connections that could prevent or catalyze progress. Overall, there is value in using complexity sciences for the pursuit of positive system change.

Keywords: complexity; economic development; planning; complex system; system map; causal loop diagram

Classification code: O12 (https://www.aeaweb.org/jel/guide/jel.php)

1.0 Introduction

Economic systems are complex (Schelling, 1978; Sinitskaya & Tesfatsion, 2015; Tesfatsion, 2002), but in our attempt to understand their behaviour, assumptions are used to simplify the systems for analysis. These simplifying assumptions result in the study of incomplete systems and often attempt to manipulate systems that are misguided (Brunner 2008). A local economy is a complex system that consists of a multitude of agents interconnected and influencing one another through cooperation and competition (Krugman, 1996; Schelling, 1978). At the same time, production and consumption activities make up the functions of the economic system (Matutinovic 2005). The economic outputs of a community can be the result of economic activities spanning several levels of functional interdependence.

Economic research in modeling complexity of marketplaces at a macro and micro level is extensive (Krugman, 1996; Schelling, 1978; Sinitskaya & Tesfatsion, 2015; Tesfatsion, 2002; Lo, 2012; Derman, 2009). However, these applications are focused on financial flows surrounding wealth creation, increasing the tax base, and creating jobs. Research has revealed that these indicators are insufficient for sustainable economic development and when exclusively considered, often increase economic and social inequalities (Leigh & Blakely, 2017; Malizia, 1994; Nijaki, 2013 ). Economic development is a specific type of system manipulation, which focuses on sustained community development and equitable wealth creation (Malizia, 1994). The challenge faced by an economic developer is to influence this highly complex system through deliberate system manipulations to attain the desired output. Development is the mechanism of shifting the stability of a system to a new state. Each state is made up of unique characteristics that are change-resistant (Brunner 2008). Therefore, for development to occur, economic developers' actions must be compatible with the existing system and reflect its limitations (Bar-Yam 2004). Overvalued activities can overwhelm the capacity of the institutions, while an undersized intervention does not stimulate enough to induce autocatalytic processes (Brunner 2008). The reality of this role, similar to many systems, is these changes are often unpredictable and require an iterative approach (Rihani 2002). Systems thinking and complexity theories offer a holistic system understanding to direct targeted interventions that more efficiently produce the desired output.

The objective of this research is to identify the complexity related to economic development and how this understanding can be used to inform economic development actions. The following questions are also explored:

  • What does it tell us about the social capital and welfare of the individuals when you apply these complexity theories?
  • What does it tell us about the local economy when we apply a lens of systems and complex theory to the community, its behaviour, and its influence?
  • Where are the optimal entry points to influence a system?
  • What characteristics and responses exist that can be utilized to influence a system positively?

1.1 Literature Review

Academics and practitioners widely acknowledge that economies and economic development takes place in a complex system (Brunner, Application of complex systems research to efforts of international development, 2008a; Brunner, Tipping Points: Managing Complex Systems for Economic Development Success, 2008b; Rihani, 2002; Schelling, 1978; Sinitskaya & Tesfatsion, 2015; Tesfatsion, 2002). The perspective of complexity with regards to the economy is recognized broadly on a global (Brunner 2008a; Rihani 2002; Chistilin 2008), political (Chistilin 2008), and local level (Matutinovic 2005). Paul Krugman (1996) and Thomas Schelling (1978) were the first to acknowledge the need to understand economics through complexity using the theory of self-organizing behaviour and emergence. The value of their applications of complexity to economics awarded them the Nobel Prize in 2008 and 2005 respectively. In 1994, the discussion of economics split when Emil Malizia wrote that our current ideologies of economic development in local economies are increasing economic and social inequalities. Since then, economists and financial analysts have gone on to do extensive work to understand and model economies using a system’s lens and insights from complexity theory (Sinitskaya & Tesfatsion, 2015; Tesfatsion, 2002; Lo, 2012; Derman, 2009). However, the field of local economic development hasn’t completely made the connection to systems thinking and complexity. Rihani (2002), Brunner (2008a & b), Reddy & Minoiu (2006), and Chistilin (2008) agree that the current methods of understanding local economic development are limited to linear thinking which results in an ineffective understanding and planning. Studies (Rihani 2002; Brunner 2008a & b; Reddy & Minoiu 2006; Matutinovic 2005; Coviello & Islam 2006; Bar-Yam 2004; Thirlwall 2002; Chistilin 2008) dive into further understanding the non-linearity that exists through the system influences, tipping points, autocatalysis, resistance, management tools, and potential models. However, only Rihani (2002) and Brunner (2008a) apply their research to case studies, and none of the respective studies use field experimentation, empirical data, or models to advance understanding of the concepts (Rihani 2002; Thirlwall 2002; Bar-Yam 2004; Matutinovic 2005; Coviello & Islam 2006; Reddy & Minoiu 2006; Brunner 2008a; Brunner 2008 b; Chistilin 2008).

Influencing a complex system is a significant focus for Reddy & Minoiu (2006), Coviello & Islam (2006), Matutinovic (2005), Brunner (2008a), and Chistilin (2008). The appeal of understanding influence is due to the need to manipulate an economic system into a desirable state. Unfortunately, decisions and influence are constrained by the local economy's interaction networks, information, beliefs, and physical states  (Sinitskaya & Tesfatsion, 2015). Due to a lack of understanding, often the influence on a system has a surprising or opposite effect (Reddy & Minoiu 2006). To prevent this, researchers have looked into the most effective streams of impact and found that the system is most subjective to energy (Reddy & Minoiu 2006) (Coviello & Islam 2006), material (Chistilin 2008), and financial flows (Brunner 2008a). Also, the networks of economic systems highly influence development. Relationships between government, households, and institutions can all impact an economy (Matutinovic 2005) (Chistilin 2008). In an article by Brunner (2008a), he discusses additional challenges when it comes to influencing complex systems. His primary exploration being system resistance to change. In highly complex systems, affecting specific components of the system will have little to no actual impact. Additionally, economies with low-level effectiveness and resource use will experience significantly more resistance to change. Therefore, for interventions to be system compatible, they must be designed at an appropriate scale that flows with the pre-existing system dynamics (Bar-Yam 2004).

Autocatalysis is the process of any cyclical chain of processes that accelerate the activity of the following link (Ulanowicz 1999). This concept is often accompanied by the ideas of system resistance and self-organization. Brunner (2008a) describes autocatalysis as a phenomenon that can negate any system resistance and catalyze the desired system growth. Yaneer Bar-Yam (2007) explains that the use of autocatalysis can be achieved by the process of guidance, not strict planning or constraining. The socio-economic activities must not surround the intervention, so there is no dependency to sustain the change (Brunner 2008a). 

Several authors have focused on tools that work within complex economic systems (Brunner 2008a; Rihani, 2002). Since economic complex systems are filled with uncertainty, any intervention must be flexible, networked, iterative, and contain diverse approaches. Additionally, any program must integrate learning and feedback into the process to ensure community support and involvement (Rihani 2002; Brunner 2008a; Brunner 2008b). By having these characteristics, the interventions can accumulate knowledge for responding to opportunities, threats, and other events (Rihani 2002).

Brunner (2008b) suggests that institutions need to invest in management tools that specialize in complex economic system modeling with the following attributes:

  • Render the essential ongoing interactions of individual economic actors that stimulate structural dynamic change and innovation
  • Visually accessible simulation of decision parameters with a view of experimentation results in three-dimensional space
  • Allows experimentation with investment sizes

Positive and negative feedbacks have also been utilized for modeling system autocatalytic potential (Brunner 2008a). While studies call for the application of complex system tools to advance economic development understanding and techniques (Brunner 2008a&b; Rihani 2002; Chistilin 2008), no studies or institutions identified in this review have attempted to develop a complex system model.

1.2 Systems Theory and Complexity

Systems theories and complexity provide terminology and understanding for phenomena that linear systems cannot describe. These systems are described through non-linearity, which includes chaos and organized complexity (Kauffman 1993). Useful theories for understanding these systems include tipping points, bifurcations, emergence, and self-arranging behaviours (Krugman, 1996; Schelling, 1978).

1.2.1 Emergence and Self-Arranging Behaviours

Emergent or self-arranging behaviours is a large-scale or complex phenomenon that arises from simple small-scale interactions between agents (Fieguth 2017). This type of behaviour is difficult to anticipate due to the macro effects being a result of an enormous number of micro-interactions. Additionally, this occurrence is only witnessed within clusters of agents where the behaviour depends on what others are doing (Schelling, 1978). The evolution of this phenomenon can be linked to cyclical behaviour, where each member accelerates the activity of the succeeding link, also known as autocatalysis (Ulanowicz 1999). Also, emergence can be related to local economies through the microeconomic actions that lead to macroeconomic, aggregate behaviour of the system (Brunner 2008a  (Krugman, 1996)).

2.0 Analysis Tools (Methods)

For the identification of system characteristics within a local economy, two analysis tools were utilized. These tools are Systems Mapping and Causal Loop Diagrams. The purpose and uses of these tools are outlined in Table 1.

Table 1: Analysis Tools

Tool

Definition

Use

System Map

Visualization technique to

model system components and their interactions.

Describes and diagnoses a current state of a given system and identify gaps and opportunities for improvement.

Causal Loop Diagram

Visualization technique that aids in visualizing how different variables in a system are interrelated

Captures visualize and discuss relationships and feedback loops.

Both methods are similar in identifying system elements, interconnections, relationships, and boundaries. However, a System Map specializes in visualizing opportunities available within the system, while Causal Loop Diagrams aim to capture the actual causation and feedback loops within the system. The specific components of each tool and what they identify are presented in Table 2.

Table 2: Tool Components and Identifiers

Tool

System Map

Causal Loop Diagram

Component

Environment

Resources

Components

Elements

Interconnections

Boundaries (scope)

Relationships

Feedback

Resilience

Elements

Interconnections

Boundaries (scope)

Relationships

Identifies

Hierarchy

All Actors within a System

Emergence

Micro-Interactions

Feedback Loops

All Interactions within a System

Emergence

Micro-Interactions

Both Systems Mapping and Causal Loop diagrams are subjective tools often adapted to the purpose of the researcher. There are currently no specific maps or diagrams previously developed for economic development. Therefore, the tools used within this study draw inspiration from system analysis of non-economic systems but have been redeveloped to assess applicability and utility in supporting economic development decision making. These procedures, while grounded in systems theory and complexity science research, are unique and new methods for economic development research.

The System Mapping technique applied here was inspired by an energy access map created by Miller et al. (2008). The procedure for identifying system components was based on Johnson (2019). Additionally, the Causal Loop Diagram is inspired by the work of Riva et al. (2018). For both diagrams, the components and connections used are derived from Rihani (2002), McMonagle (2019), Brunner (2008), and Chistilin (2008).

3.0 Case Studies

The analysis tools compared the approaches and development outputs of two case studies. The locations of the case studies are in Kitchener, Ontario, and Detroit, Michigan. Both locations experienced economic hardships attributed to the dramatic decline of the manufacturing sector during the automotive industry recession. Detroit and Kitchener, however, responded with very different economic development approaches to address the economic challenges facing their respective cities. The locations of both cities are presented on the map in Figure 1.

Fig. 1 Canada-USA Border Map with Case Study Locations. The map is oriented with North in the upwards direction, the border is indicated with a black line and Detroit, Michigan, situated just below the line and Kitchener, Ontario, situated above further away from the Canada-USA border. The scale is one-third of the map is 200km of land

The city of Kitchener is located in southern Ontario, Canada, and 100km away from Toronto. It is home to a population of around 230000 people (Statistics Canada 2017). Starting in 1850, Kitchener established a large number of manufacturing firms, and by 1950 it was considered a leader in auto parts manufacturing (Spigel & Bathelt 2019).  The automotive industry experienced a recession in the years 1957 to 1961, significantly impacting the manufacturing job market. Figure 2 demonstrates the provincial decline of manufacturing jobs (Statistics Canada 2020).

Fig. 2 Ontario percentage of jobs in manufacturing over time (Statistics Canada 2020). The figure shows a downward trend in Manufacturing jobs in the province of Ontario from 1980 to 2019. The initial percentage of manufacturing jobs in 1980 is 19% and by 2020 the percentage has dropped to 9%. There is a 1:2 slope from 1980 to 1990, leveling out to almost horizontal from 1990 to 2000, then a steep decline of 3:2 from 2000 to 2010, leveling out at a gradual decline from 2010 to 2019

In the time between 1980 and 1990, the Kitchener downtown area became virtually abandoned with unfinished roadway projects and an empty shopping mall (English 2011). The economic history of Detroit, Michigan, follows a similar trajectory. As seen in Figure 1, Detroit, located in Northern Michigan, is the largest city on the United States–Canada border. The population of Detroit and the surrounding region is 670000 people. (United States Census Bureau, 2018). Similarly, Detroit was considered a leader in automotive parts manufacturing. When the recession hit around 1960, unemployment rose, and the population declined. Figure 3 tracks the jobs in manufacturing compared to the unemployment rate in Detroit, and Figure 4 demonstrates the population decline from 1950 to 2019 (YCharts 2020; FRED 2020; World Population Review 2020; Data USA 2020).

Fig. 3 manufacturing jobs compared to unemployment in Detroit (YCharts 2020; FRED 2020). Two data trends are compared with scales on the primary and secondary axes. The primary axis is the number of total manufacturing jobs (in the thousands) in Detroit from 1990 to 2019, represented with a dotted black line and black triangle data points. The secondary axis is the unemployment rate from 1990 to 2019, represented with a solid black line and black circular data points. The number of manufacturing jobs begins with a gradual incline of 1:4 slope from 1990 to 2000, then a dramatic decrease of a 5:2 slope from 2000 to 2010, and then an increase with a slope of 3:2 from 2010 to 2015, and finally a gentler increase from 2010 to 2019 with a slope of 5:1. The trends of these data sets mirror each other

Fig. 4 population decline in Detroit from 1950 to 2019 (World Population Review 2020; Data USA 2020). The data trend is a decrease in population from around 1800000 people in 1950 to around 600000 people in 2019. The decrease is steepest from 1970 to 1990, and steep again from 2000 to 2010

As seen in these figures, unemployment rates are concerningly high from 1990 to 2015, and an additional source suggests that unemployment was non-existent before the recession (Thomas, 1990).

The cities took very different approaches to address the economic decline and high unemployment. The City of Kitchener focused its economic development on community-centered approaches. The City leveraged connections to post-secondary institutions, existing infrastructure, and community values to foster development. They did this through the encouragement of entrepreneurship and, more significantly, the establishment of Communitech in 1997 (Communitech 2020). Another significant economic development decision was establishing the University of Waterloo School of Pharmacy in downtown Kitchener in 2008 (University of Waterloo 2020). The university and Communitech ended up having significant impacts on Kitchener's economic trajectory, attracting talent seeking employers, and creating a community of entrepreneurship and cooperation (Spigel & Bathelt 2019). This influence has resulted in a successful transition from a traditional manufacturing economy to one that is increasingly led by the technology and science community of the Kitchener-Cambridge-Waterloo region. Supporting the successfully transitioned economy is Figure 5, comparing the decreasing unemployment compared to the increasing population in Kitchener, Cambridge, and Waterloo (Statistics Canada 2020).

Fig. 5 Kitchener-Cambridge-Waterloo unemployment compared to population growth (Statistics Canada 2020). Two data trends are compared with scales on the primary and secondary axes. The primary axis is the unemployment rate in Kitchener-Cambridge-Waterloo from 2001 to 2019, represented with a dotted black line and black triangle data points. The secondary axis is the population in the region from 2001 to 2019, represented with a solid black line and black circular data points. The unemployment rate begins at 6% in 2001, dips to 5.5% in 2005, increases to 8.2% in 2010, and then gradually decreases to 5% in 2019. The population linearly increases from 340000 people in 2001 to 450000 people in 2019

Today Kitchener is home to many technology start-ups, satellite offices, and headquarters of technology firms like google. The region has more than 1000 technology-related firms employing over 30,000 workers (Communitech 2018). The success of Kitchener indicates cooperation, innovation, and trust between firms and industries as an explanation for the region's success as a technology economy.

Detroit focused on corporate support strategies in an attempt to retain the existing automobile industry (Thomas 1990). Market forces controlled the design, location, and quality of commercial and residential development within the city. To combat unemployment, the local government made a large-scale investment to support the construction of a state-of-the-art auto assembly plant in 1983, Poletown (Digaetano & Lawless 1999). The decision faced substantial opposition from local community groups that rallied against the use of public subsidies to support the automotive companies. Still, Poletown was built and resulted in less employment than projected and public unrest (Digaetano & Lawless 1999). Despite the failure of the Poletown plant, during the 1980s and 1990s, Detroit continued to heavily subsidize the auto industry with projects such as investing 264 million into a new state-of-the-art Chrysler assembly plant (Digaetano & Lawless 1999). However, again the project was met with public opposition, cost overruns, and the mishandling of land acquisition (Digaetano & Lawless 1999). There were other plans similar to Poletown and the Chrysler assembly plant that was proposed and were either completely rejected by the public or met with controversy and lost momentum. Today the population of Detroit continues to decrease as the job market has not fully recovered and unemployment remains above the national average by 5% (YCharts 2020).

4.0 Results and Analysis

The resulting map and diagram are presented below. Figure 6 is a local economy Systems Map, including the relationships between the economic developer and other agents - policy makers, financers, community, researchers, influencers, and economy. It identifies the direct interactions between an economic developer and the surrounding agents. Additionally, the indirect influences of the economic developer are shown through the interrelationships of the agents. Figure 7 is an economic developers' Causal Loop Diagram, which highlights the place of the community, industry, and local economy in both case studies. The influences of the system and loops are exemplified with the causal arrows connecting the system components. From these two diagrams, lessons are identified related to feedback loops and direct connections.

Fig. 6 Local economy Systems Map, including the role of the economic developer. The system has been sectioned into six categories, Policy Makers, Financers, Community, Researchers, Influencers, and the Economy. In the center is the role of the Economic Developers. Each category has been subcategorized into local, global, public, private when relevant. Within these subcategories are the actual actors. Each of the actors is connected through activities such as services and sales, funding, grants, loans, information sharing, expertise, influence, and development actions. The activities between each agent are signified with a grey arrow from the actor to the influenced. The direct influence and approaches to influencing are represented with grey arrows from the Economic Developers

Fig. 7 Economic developer Causal Loop Diagram, highlighting the place of the community, industry, and local economy and their connections. System components that are highly influenced, identified in the system map, are connected with arrows to one another through influence on one another. The direction of influence is indicated by the direction of the arrows. The power of influence is identified with the number of potential influences from a component. Community, industry, and local economy are highlighted with an oval since they are the focus of the study.

4.1 Feedback Loops

Feedback loops are an integral concept in the pursuit of understanding economic development in complexity. Looking at Figure 7, and the highlighted areas of focus, essential observations are noted. Concerning the local economy, three feedback loops are identified, as presented in Figure 8.

Fig. 8 There are three local economy feedback loops identified from Figure 7. These loops are represented with arrows connecting components into an infinite circle. Positive feedbacks are signified with a plus sign in the center of the cycle and the negative loops are indicated with a subtraction sign. The first is the influence of Access to Financial Capital on Propensity to Invest and then influences the Local Economy, which cycles back into positive feedback. The second is the Job Market Demand influencing Average Income which impacts Income Inequality and then the Local Economy, which cycles back into a negative feedback loop. The third feedback loop is the influence of the Job Market Demand on Local Wealth, which influences Propensity to Invest again influencing the Local Economy. This is also a positive feedback loop

The positive feedbacks are identified with a plus sign in the center, while the negative feedback has a negative sign. Similarly, the feedback loops connected to the community are presented in Figure 9.

Fig. 9 There are four community feedback loops identified from Figure 7. These loops are represented with arrows connecting components into an infinite circle. All of these feedback loops could be positive or negative and is signified with an addition and subtraction sign in the center of the cycle. The first loop connects Local Government to Residents, then to Community. The second loop connects Local Businesses to Residents and Local Government, then to the Community. The third loop connects Residents to Local Businesses, then to Local Government. The last loop connects Infrastructure to Residents, then to Local Government

Unlike the loops in Figure 8, these feedbacks can act positively or negatively depending on the system influences and states. Table 3 compiles the observations of the feedback loops and their relevance to the case studies.  The most significant observations being the lack of feedback loops connected to industry and the high concentration of feedback loops connected to and surrounding the community. These contrasting results indicate the strengths of using a community-oriented economic development approach, which eases the resistance to change by using autocatalysis to stimulate development. For example, the City of Kitchener focused on attracting groups that built community connections and incubated talent, by funding entrepreneurship and start-ups through organizations such as Velocity and Communitech. These organizations attracted creative and talented clusters that settled in the area and feedback into the development of the community.

In comparison, if economic development focuses on a system component that doesn't feed back into itself, mechanisms like autocatalysis cannot be utilized which slows down or reverses the process of development and require large amounts of resources to sustain. Detroit, Michigan exemplifies this through its economic crash. The city funneled its resources into economic development through the industry. As a result, it dropped from the 5th largest city in America to the 23rd (Salins 2018), acting more like a sink for resources than a tool for development.

Table 3: Feedback Loop Observations

#

Observation

Meaning

Complexity

Case Study

1

Positive and negative local economy feedback loops

Money creates more money, but the distribution of the economy is easily bumped into a negative feedback loop with income inequality

Microeconomic actions from any point in the loop can lead to macroeconomic, aggregate behavior of the system

Detroit, Michigan started out by trying to promote the industry in the area. Eventually, the system spiraled, and planners lost control of the design of city infrastructure (Salins 2018).

2

The negative local economy feedback loop is closely connected to industry

Having an industry-focused economic development strategy has the risk of entering into the negative local economy feedback loop considering it is one step away from this loop

3

No feedback loops connected to industry

Considering the lack of feedback loops connected to industry suggests a sink of energy

  • It is useful to tap into system components that will increase development efficiency

If economic development is focusing on a system component that doesn't feed back into itself, mechanisms like autocatalysis cannot be utilized, which will slow down the process of development and require large amounts of resources.

 

Detroit, Michigan, observed the downfalls of putting your energy into a sink

  • When economic developers allowed for the market to drive development, the resources became a sink for singular growth rather than holistic development
  • Federal and State governments began building highways to the suburbs, while cities began neglecting their infrastructure (Salins 2018).

4

Many feedback loops connected to and surrounding the community

The community can be a powerful tool in influencing change, but it also can be a systems greatest challenge

  • There are so many close feedbacks directly related to or surrounding the community
  • If these loops act positively, the workload of an economic developer can be lightened, and autocatalysis can be utilized for the stimulation of development
  • If these loops act negatively, there can be extreme resistance to state change making it nearly impossible for development to take place

Feedback loops stimulate the emergence and self-arranging behaviours, for better or for worse

systems can self-arrange to criticality, thus reaching a tipping point sooner, for better or for worse

Kitchener observed positive results by targeting the community for development.

  • Their approach pursued attracting groups that built community connections and incubated talent
  • E.g. Velocity and Communitech
  • This allowed for strong connections between community, residents, and local government
  • These components built on and supported one another to the point where Kitchener has a vibrant entrepreneurial community and has attracted some of the top technology companies in the world.

5

The community has a feedback loop with the local economy, connecting through the local government and the job market

The community can influence cyclical local economic change (positive or negative) without direct interventions

Self-arranging behaviour at the community level can cause an independent economic change

Kitchener shows that positive outcomes can be achieved when growth and development goals are integrated and rooted in the community

4.2 Direct Connections

Direct connections represent the influencing capacity or power of an agent on the other components of the system. The connections to community and industry from Figure 7 are identified in Figure 10.

Fig. 10 Community and Industry are directly connected to a number of components. These direct connections are represented with a theoretical equation. The first equation connects Local Government, Residents, and Local Businesses as influencing the Community. Then, the Community influencing Local Government, Residents, and Local Businesses. The second equation connects Distributors, Manufacturers, and Crowding to influencing Industry, which then, influences Technology Development, Job Market Demand, and Manufacturers

Figure 10 emphasizes the secure connection the community has with local government, residents, and businesses. Table 4 compiles the observations of the direct connections and their relevance to the case studies. The implication of the community influence being strongly connected to this internal network creating a sense of ownership and responsibility between their actions. The community is personally invested in the outcomes of its actions, and therefore, its interactions with the other agents naturally work towards positive development. For example, the City of Kitchener's downtown core is a hub for low-income support, which has stimulated many community lead initiatives to support itself without any interventions. The Working Centre is an example of a community-led initiative that emerged in response to the economic decline triggered by the loss of manufacturing to support skill-building, retraining, and employment. The City of Kitchener's economic development strategy was supported by insights and knowledge sharing from the community connections in place and reciprocally targeted funds to expand strong community networks (City of Kitchener 2019).

In contrast, the industry influences the internal system but is influenced by external components like distributors, manufacturers, and crowding. Therefore, industry actions, concerning the system whole, are more self-serving due to the lack of connection it has with the complete system. Detroit experienced the challenges of this when after their development efforts focused on industry failure, the industry relocated without significant impacts. Both of these observations show the significance of the type of connecting behaviours between actors. These relationships dictate the direction of self-arranging behaviour that takes place within a system.

The second observation is that the direct interactions observed in the Causal Loop diagram are influenced by additional elements observed in the Systems Map. Meaning that the influences and relationships of the system (the state) have an impact as well. Therefore, it is essential to remember the limitations of the complexity tools and there is uncertainty in system manipulation.

Table 4: Direct Connections Observations

#

Observation

Meaning

Complexity

Case Study

1

Community influence is internally connected

The community is directly connected to internal system components which cause personally invest in their impact

The level of micro-investment will dictate the direction of positive/ negative self-arranging behaviour that takes place within a system

Kitchener's intervention was successful because the initiatives had a purpose for the community, local government, and residents, and each member was motivated to support one another towards a common goal

2

Industry influence is shaped by external factors

The industry is less personally connected to its impact due to its external influences

Detroit focusing solely on industry growth, so when the market drove the development, it wasn't considering any other component other than its own growth, which leads to segregated growth rather than holistic development

3

The direct interactions observed in the Causal Loop diagram are influenced by additional elements observed in the Systems Map

There is always a level of uncertainty when working with any part of the complex system

The micro outputs that will result from direct interactions observed in the causal loop diagram will have unpredictability due to the system state, which is formed through the systems map influences and relationships.

Kitchener utilized the various actors to influence a desirable output.

 

Detroit didn't consider the entire picture and underutilized system components.

  • Industrial plants were located along waterways, railway lines, and major arterial roads to minimize cost and maximize profit (Salins 2018).

5.0 Conclusions and Recommendations

A local economy is a highly networked system consisting of a multitude of agents interconnected through cooperation and competition while production and consumption activities make up the functions of the economic system (Matutinovic 2005). The economic outputs of a community can be the result of economic activities spanning several levels of functional interdependence. The challenge of the economic developer and city planner is to influence this system through deliberate system manipulations for the desired output. Economic development is a specific type of system manipulation where, if the mechanisms of a system are adequately understood, the stability of a system is shifted to a new state to produce positive change within a community. Systems thinking and complexity theories offer a holistic system understanding to direct targeted interventions that more efficiently produce the desired output. Using systems thinking and complexity science tools can be utilized for informed economic development actions.

The economic developer's role is explored through the analysis techniques of System Mapping and Causal Loop Diagrams. Both methods are similar in identifying system elements, interconnections, relationships, and boundaries. However, a System Map specializes in visualizing opportunities available within the system, while Causal Loop Diagrams aim to capture the actual causation and feedback loops within the system. Currently, no other study previously developed these maps as a tool for economic development. The tools used within this study draw inspiration from complex system analysis of non-economic systems but have been redeveloped for the purpose of supporting economic developers and community leaders. These procedures, while grounded in complexity science research, are unique and new methods for economic development research.

The new methods are tested by applying them to two case studies, Detroit, Michigan (USA), and the City of Kitchener, Ontario (Canada). The first case study focuses on the economic downfall of Detroit, Michigan. Detroit Michigan focused their economic development on attracting industry to the area and missed out on valuable system feedbacks and connections. As a result, the community failed to stimulate a competitive industrial advantage, and the companies relocated. On the other hand, the second case study focuses on the community-centered economic development approach of the City of Kitchener. The City leveraged connections to post-secondary institutions, existing infrastructure, and community values to foster development. As a result, the stimulated growth within the community has catalyzed new connections, attracting more businesses and talent to the area.

The main lessons identified are the importance of utilizing feedback loops and identifying direct connections that could prevent or catalyze progress. The most significant observations being the lack of feedback loops connected to industry and the high concentration of feedback loops connected to and surrounding the community. These contrasting results indicate the strengths of using a community targeted economic development approach, like Kitchener, which lighten the effort required to stimulate change through the utilization of autocatalysis for stimulating development. Another important observation is that the industry is less connected to its impact due to its growth taking place as a result of external factors. These relationships dictate the direction of self-arranging behaviour that takes place within a system. Also, direct interactions observed in the Causal Loop diagram are influenced by additional elements observed in the Systems Map. Meaning that the influences and relationships of the system (the state) have an impact as well. Therefore, it is essential to remember the limitations of the tools, and there is uncertainty in system manipulation.

Overall, there is value in utilizing Systems Maps and Causal Loop Diagrams in the pursuit of system manipulation. However, the maps and diagrams built in this paper are not all-encompassing or universal. The fundamentals of successful economic development consist of knowledge building. Knowledge building can provide an institutional framework that leverages autocatalysis. Therefore, there is value in further exploration of industry systems and community systems to obtain further intricacies to understanding. Exploration can be done through building System Maps and Causal Loop Diagrams for the community and industry. Then, comparing the systems to one another and the local economy, would increase understanding of influence, which would help in targeted interventions with more predictable results. Additionally, applying the systems to more case studies would provide an opportunity for further lesson evaluation and reinforce the current lessons learned.

6.0 Acknowledgments: We would like to acknowledge the help of Hannah Murphy and Simon Lin in proofreading this article.

7.0 Author Contributions: Kayleanna Giesinger contributed to the study conception, design, material preparation, data collection, analysis, and writing of the first draft of the manuscript. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

References

Anna Johnson, D. P.-T. (2019). Student Guide to Mapping a System. Retrieved from Map the System: http://mapthesystem.sbs.ox.ac.uk/wp-content/uploads/2019/02/Student-Guide-to-Mapping-a-System-1.pdf

Bar-Yam, Y. (2004). Making Things Work. Cambridge, MA: NECSI Knowledge Press.

Bar-Yam, Y. (2007). Reducing Uncertainty to enable Better Investments and Clearer Justifications through Prediction of Impact by Multi-Scale Agent-Based Model, ADB in collaboration with the New England Complex Systems Institute. Paper at the Regional Aid for Trade Conference, Manila.

Brunner, H.-P. (2008a). Application of complex systems research to efforts of international development. Unifying Themes in Complex Systems, 580-587.

Brunner, H.-P. (2008b). Tipping Points: Managing Complex Systems for Economic Development Success. Evol. Inst. Econ. Rev., 5(1), 163-187.

Chistilin, D. (2008). Principles of self-organization and sustainable development of the world economy are the basis of global security. Unifying Themes in Complex Systems, 6, 382-389.

City of Kitchener. (2019). Economic development strategy. Retrieved from City of Kitchener: https://www.kitchener.ca/en/city-services/economic-development-strategy.aspx

Communitech. (2020). About Communitech. Retrieved from Communitech: https://www.communitech.ca/who-we-are/

Coviello, D., & Islam, R. (2006). Does Aid Help Improve Economic Institutions? World Bank Policy Research Working Paper, 3990.

Data USA. (2020). Detroit, MI. Retrieved from Data USA: https://datausa.io/profile/geo/detroit-mi/

Derman, E. (2009). Models. Financial Analysts Journal, 65(1), 28-33.

Detroit, MI Unemployment Rate. (2019, July). Retrieved from Y Charts: http://ycharts.com/indicators/detroit_mi_unemployment_rate

Digaetano, A., & Lawless, P. (1999). Urban Governance and Industrial Decline: Governing Structures and Policy Agendas in Birmingham and Sheffield, England, and Detroit, Michigan, 1980-1997. Urban Affairs Review, 34(4), 546-577.

English, J. (2011, August 8). Kitchener Meets Its Waterloo. Maclean's Magazine, pp. 66-69.

Fieguth, P. (2017). Introduction to Complex Systems. International: Springer Berlin Heidelberg.

FRED. (2020, April 17). All Employees: Manufacturing in Detroit-Warren-Dearborn, MI (MSA). Retrieved from FRED: https://fred.stlouisfed.org/series/DETR826MFG#0

Kauffman, S. (1993). The origins of order. Oxford: Oxford University Press.

Krugman, P. (1996). Chapter 1: Self-Organization in Space. In P. Krugman, Self-Organizing Economy (pp. 9-29). Hoboken, NJ: Wiley - Blackwell Publishers.

Leigh, N. G., & Blakely, E. J. (2017). Planning Local Economic Development: Theory and Practice. New York, NY: SAGE Publications Inc.

Lo, A. W. (2012). Adaptive Markets and the New World Order (corrected May 2012). Financial Analysts Journal, 68(2), 18-29.

Malizia, E. E. (1994). Redefinition of Economic Development. Economic Development Review, 12(2), 83.

Matutinovic, I. (2005). The Microeconomic Foundations of Business Cycles: From Institutions to Autocatalytic Networks. 1. Of Economic Issues, 39(4), 867-898.

McMonagle, R. (2019). Developing Growth Roadmaps for Toronto’s Green Industries. Sector Development Office, Economic Development & Culture City of Toronto.

Miller, C. M.-A. (2008). Poverty Eradication Through Energy Innovation: a Multi-Layer Design Framework for Social Value Creation. Retrieved from Affordable Energy for Humanity: https://ae4h.org/projects/energy4change

Nijaki, L. K. (2013 ). Going Beyond Growth: The Green Economy as a Sustainable Economic Development Strategy. In L. K. Nijaki, The Next Economics: Global Cases in Energy, Environment, 251 and Climate Change (pp. 251-274). Berlin/Heidelberg, Germany: Springer Science & Business Media.

Reddy, S., & Minoiu, C. (2006). Development Aid and Economic Growth: A Positive Lon-Run Relation. Columbia University working paper.

Rihani, S. (2002). Complex systems theory and development practice: understanding non-linear realities. London; New York: Zed Books; New York: Distributed in the USA exclusively by Palgrave.

Riva, F. A. (2018). Electricity access and rural development: Review of complex socio-economic dynamics and causal diagrams for more appropriate energy modeling. Energy for Sustainable Development, 203-223.

Salins, P. (2018, October 3). Rethinking Urban Planning in Detroit—and Beyond. Retrieved from Manhattan Institute: https://www.manhattan-institute.org/html/urban-policy-2018-rethinking-urban-planning-detroit-and-beyond-11520.html

Schelling, T. C. (1978). Chapter 1: Micromotives and Macrobehaviour. In T. C. Schelling, Micromotives and Macrobehaviour (pp. 1-26). New York, NY: W. W. Norton & Company Inc.

Sinitskaya, E., & Tesfatsion, L. (2015). Macroeconomies as constructively rational games. Journal of Economic Dynamics & Control, 61, 152-182.

Spigel, B., & Bathelt, H. (2019). Questioning cultural narratives of economic development: An investigation of Kitchener-Waterloo. The Canadian Geographer, 267–283.

Statistics Canada. (2017, February 13). Kitchener (city) census profile. Retrieved from 2016 Census data: https://www12.statcan.gc.ca/census-recensement/2016/dp-pd/prof/details/page.cfm?Lang=E&Geo1=CSD&Code1=3530013&Geo2=CD&Code2=3530&Data=Count&SearchText=kitchener&SearchType=Begins&SearchPR=01&B1=All&TABID=1

Statistics Canada. (2020, April 21). Labour force characteristics by industry, annual (x 1,000). Retrieved from Statistics Canada: https://www150.statcan.gc.ca/t1/tbl1/en/cv.action?pid=1410002301

Tesfatsion, L. (2002). Agent-Based Computational Economics: Growing Economies From the Bottom Up. Agent-Based Computational Economics, 8, 55-82.

The Social Compact Inc. (2006, October). Downtown Detroit In Focus: A Profile of Market Opportunity. Retrieved from The Urban Markets Initiative, Brookings Institution Metropolitan Policy Program: https://web.archive.org/web/20110812084749/http://www.downtowndetroit.org/ddp/newsroom/Downtown_Detroit_in_Focus.pdf

Thirlwall, A. P. (2002). The Nature of Economic Growth—An Alternative Framework for Understanding the Performance of Nations. Edward Elgar, Cheltenham UK and Northampton MA.

Thomas, J. M. (1990). Planning and Industrial Decline: Lessons from Postwar Detroit. American Planning Association. Journal of the American Planning Association; 56(3), 297-311.

Ulanowicz, R. (1999). Life after Newton: An Ecological Metaphysic. BioSystems, 40, 127-142.

United States Census Bureau. (2018, October 3). U.S. Census website. Retrieved from United States Census Bureau: https://www.census.gov

University of Waterloo. (2020). About the School of Pharmacy. Retrieved from University of Waterloo: https://uwaterloo.ca/pharmacy/about-school-pharmacy

World Population Review. (2020). Detroit, Michigan Population 2020. Retrieved from World Population Review: https://worldpopulationreview.com/us-cities/detroit-population/

YCharts. (2020, February). Detroit, ME Unemployment. Retrieved from YCharts: https://ycharts.com/indicators/detroit_me_unemployment