TIME FAMINE IN THE AGE OF CONVENIENCE: A SYSTEMS PERSPECTIVE ON FOOD SECURITY IN DETROIT By Kyle R. Metta          A THESIS Submitted to Michigan State University in partial fulfillment of the requirements  for the degree of  Community Sustainability Ð Master of Science  2016   ABSTRACT TIME FAMINE IN THE AGE OF CONVENIENCE: A SYSTEMS PERSPECTIVE ON FOOD SECURITY IN DETROIT  By Kyle R. Metta  We conducted group model building workshops and semi-structured interviews with stakeholders concerned with food security in the City of Detroit. The group built a causal loop diagram to identify structural feedbacks that drive household food insecurity. A quantitative system dynamics model was then developed to integrate community knowledge with secondary academic data. The model is used to simulate a householdÕs consumption of fresh fruits and vegetables over time, serving as a proxy for the nutritional quality of all meals consumed. Our model demonstrates that singular interventions do marginally reduce general food insecurity and increase the consumption of nutritious foods, but the largest impacts result from utilizing behavioral and household resource interventions in combination. The model also shows that solutions may have unintended consequences, increasing household food spoilage, and increasing away from home consumption. The model provides a novel way to examine the experiential dimensions of food and nutritional security. We conclude by supporting research on  how time, as a resource stock, is a contributing factor to household food access (McKenzie, 2014) and how perceptions of time scarcity can lead to unhealthy food choices (Jabs & Devine, 2006). We demonstrate how a householdÕs stock of time is experienced in two dimensions, both temporal distance, the time it takes to procure food, and convenience pressure, the way perceptions of time scarcity affect food choices. We also point to the strength of household resource interventions when coupled with a behavioral component. Largely, we contribute and document how a participatory process can be utilized to better understand food security in an urban environment.    iii  ACKNOWLEDGMENTS   This thesis would not have been possible without the support and guidance from faculty and colleagues at Michigan State University. I would especially like to acknowledge the mentorship of Dr. Laura Schmitt Olabisi as her commitment to participatory research inspired and made this work possible. I would also like to thank my committee members Dr. Maria Claudia Lopez and Dr. Michael Hamm for their patience and support especially when things proved difficult. Their perspectives were invaluable throughout the process. I would also like to acknowledge Dr. Robert Richardson for his consistent and unyielding pursuit of interesting and meaningful science. His guidance and open door have shaped my academic and personal character immensely.  Thank you to my community partner, Ren”e Wallace from Food Plus Detroit. She spent countless hours planning and assisting us with this work and provided the experience and connection to community. I would also like to acknowledge the participants of our modeling workshops for their time and fearlessness when introduced to new things.  Additionally, I thank Sara Torrez, Candace Yost, Britta Yost, Jill Hardy, and Udita Sanga for their comments and edits throughout the writing and design process. Finally, thank you to my family for their unwavering support and love throughout my educational journey. Thank you to my mother Lynn, for teaching me empathy, kindness, and the importance of community.  Thank you to my sister Kimberly for her support and always being the loudest one cheering.   iv   TABLE OF CONTENTS   LIST OF TABLES ................................................................................................................... vi!LIST OF FIGURES ................................................................................................................ vii!INTRODUCTION ................................................................................................................... 1!CHAPTER 1   EXAMINING THE COMPLEXITY OF URBAN HOUSEHOLD FOOD INSECURITY WITH STAKEHOLDERS IN DETROIT, MI .............................................. 3!1.1 Introduction ............................................................................................................................................. 3!1.2. Literature Review of Drivers of Food Insecurity .............................................................................. 4!1.3.1 Research Design & Methods .............................................................................................................. 7!1.3.2 Methods ............................................................................................................................................. 8!1.3.2.1 Interviews .................................................................................................................................. 8!1.3.2.2 Participatory Model Building ................................................................................................. 8!1.3.2.3 Workshop 1: Causal Loop Diagrams ............................................................................... 9!1.3.2.4 Home Economics Causal Loop Diagram ................................................................. 11!1.3.2.5 Cultural Nutritional Casual Loop Diagram .............................................................. 12!1.3.2.6 Socio-Political Causal Loop Diagram ........................................................................ 14!1.3.2.7 Peer Networks of Influence Causal Loop Diagram ................................................ 16!1.3.3 Developing the Quantitative Model ...................................................................................... 17!1.3.3.1 Model Description ............................................................................................................ 18!1.3.3.2 Household Economics Module ................................................................................. 20!1.3.3.3 Time Cost Module ........................................................................................................ 21!1.3.3.4 Preference for Fresh Fruits & Vegetable Module ................................................... 21!1.3.3.4 Nutritional Security Module ....................................................................................... 22!1.3.3.5 Purchasing ...................................................................................................................... 22!1.3.4 Model Demonstration & Validation: Workshop 2 .............................................................. 23!1.4 Results ..................................................................................................................................................... 25!1.4.1 Reference Mode ............................................................................................................................. 25!1.4.2 Parameter Uncertainty ................................................................................................................... 26!1.4.2.1 Cost of Healthy Meals .......................................................................................................... 26!1.4.2.2 Alternative Food Economy Growth Rate ......................................................................... 27!1.4.3 Scenario Results ............................................................................................................................. 28!1.4.3.1 Scenario 1 ................................................................................................................................ 29!1.4.3.2 Scenario 1 Interventions ....................................................................................................... 29!1.4.3.3 Scenario 2 ................................................................................................................................ 31!1.4.3.4 Scenario 2 Interventions ....................................................................................................... 32!1.4.3.5 Scenario 3 ................................................................................................................................ 33!1.4.3.6 Scenario 3 Interventions ....................................................................................................... 34!1.5 Discussion ............................................................................................................................................... 35!1.5.1 Limitations ................................................................................................................................. 38!1.5.2 Potential Policy Implications .................................................................................................. 39!v   1.5 Conclusions ........................................................................................................................................ 40!CHAPTER 2   FACING THE PUBLIC: REPORT ON BARRIERS TO HOUSEHOLD FOOD SECURITY IN DETROIT, MI .................................................................................. 42!2.1 Introduction ........................................................................................................................................... 42!2.1.1 Why a Systems Approach? ........................................................................................................... 43!2.2 Process .................................................................................................................................................... 44!2.3 Model Narrative ..................................................................................................................................... 49!2.4 Simulation Workshop ........................................................................................................................... 50!2.5 Results ..................................................................................................................................................... 50!2.6 Food Policy Council Recommendations ........................................................................................... 55!APPENDIX ............................................................................................................................. 58!APPENDIX A: Model Equations ............................................................................................................ 59!APPENDIX B: Semi-Structured Interview Questions ......................................................................... 61!APPENDIX C: IntervieweesÕ Affiliations ............................................................................................... 62!BIBLIOGRAPHY ................................................................................................................... 63! vi   LIST OF TABLES   Table 1 - 1: Interface Intervention Definitions ............................................................................................................................... 24!Table 1 - 2: Model Equations ............................................................................................................................................................. 60!Table 1 - 3: Stakeholder Affiliations .................................................................................................................................................. 62!   vii  LIST OF FIGURES   Figure 1 - 1: Research Design .............................................................................................................................................................. 7!Figure 1 - 2: Aggregate Causal Loop Diagram ................................................................................................................................ 11!Figure 1 - 3: Home Economics CLD ............................................................................................................................................... 12!Figure 1 - 4: Cultural Nutritional CLD ............................................................................................................................................. 14!Figure 1 - 5: Socio - Political CLD .................................................................................................................................................... 16!Figure 1 - 6: Peer Networks of Influence CLD .............................................................................................................................. 17!Figure 1 - 7: Model Overview ............................................................................................................................................................ 20!Figure 1 - 8: Reference Mode & Sensitivity Analysis ..................................................................................................................... 28!Figure 1 - 9: Scenario 1 ........................................................................................................................................................................ 31!Figure 1 - 10: Scenario 2 ..................................................................................................................................................................... 33!Figure 1 - 11: Scenario 3 ..................................................................................................................................................................... 35! Figure 2 - 1: Example of Feedback Loop ....................................................................................................................................... 45!Figure 2 - 2: Aggregate CLD .............................................................................................................................................................. 46!Figure 2 - 3: Home Economics Segment ......................................................................................................................................... 47!Figure 2 - 4: Cultural Nutrition Segment ......................................................................................................................................... 48!Figure 2 - 5: Household 1 Results ..................................................................................................................................................... 51!Figure 2 - 6: Household 2 Results ..................................................................................................................................................... 54!Figure 2 - 7 Alternative Food Economy .......................................................................................................................................... 55!Figure 2 - 8 Graphical Equations ...................................................................................................................................................... 59! !   1  INTRODUCTION Food security exists in a complex system with notable tensions between social-ecological realities, urban development, and social justice. It is an issue that at its core is personal and intimate. Its prevalence is lauded as a reflection of the community, its character and assets. Its absence is deemed a product of larger systemic issues embedded in layers of intricacy. For communities, it is a problem rooted in questions of self-sufficiency but is dependent on the role of charitable organizations, and the trickling flow of benefits from federal programs. For activists it serves as an opportunity to assert justice and build an equitable food landscape. At the core, community food security is Òa situation in which all community residents obtain a safe, culturally acceptable, nutritionally adequate diet through a sustainable food system that maximizes community self-reliance and justiceÓ (Bellows & Hamm, 2002).   Nationally 17.4 million households in 2009 had difficulty providing enough food for family members due to insufficient resources (Pothukuchi, 2011). For Detroit, food insecurity issues are felt, to some degree, by over 550,000 community members who live in areas where the food system is Òout of balanceÓÑwhere it is far easier to access low quality, unhealthy foods than it is to find fresh fruits and vegetables (Gallagher, 2007). Food insecurity is experienced through malnutrition, obesity, heart disease, cognitive dysfunction, poor education outcomes and losses in workforce productivity (Pothukuchi 2011; Gallagher 2007; Walker, et al., 2010; Larson, et al., 2009; Drewnowski 2004). This thesis investigates food security issues in Detroit as a community engaged scholarship project. Michigan State University defines publicly engaged scholarship Òas a scholarly endeavor that cross-cuts instruction, research and creative activities, and service; fulfills unit and university missions; and is focused on collaboration with and benefits to communities external to the universityÓ. From the 2  outset, our research engaged with Detroit stakeholders working to identify and inform strategic planning around food systems and food security.  Our goal was the co-creation of new knowledge while developing an understanding of both the tacit and experiential dimensions of this community problem.  In Chapter 1 we document how our more focused research question emerged out of early discussions with community partners and stakeholders. Namely, how the complex interactions of household resource endowments generate patterns of food (in)security.  We investigate this question with a technique - Group Model Building (GMB) - and proceed to build a quantitative system dynamics model of household food security in Detroit (Van den Belt, 2004). We use an expanded understanding of household resources to include household time stocks, knowledge, and income, and document the complex interaction of these household resources with the built environment and food landscape.  We point to the emergence of a Òconvenience cultureÓ and how this phenomenon may be compounding socio-economic and geographic factors limiting food and nutritional opportunities for Detroit households (Jabs & Devine, 2006).   Chapter 2 is a policy report for the Detroit Food Policy Council, one of our community partners for this research. It frames the research findings from Chapter 1 in an applied context that is relevant to food policy practitioners and advocates. It identifies key ways that local food policy councils generally and the DFPC, can utilize this research to address food and nutritional security with local policy. As an appendix, we also provide a model user guide with the hope that it spurs systems thinking and broader understanding of the systems governing food security in Detroit.   3  CHAPTER 1   EXAMINING THE COMPLEXITY OF URBAN HOUSEHOLD FOOD INSECURITY WITH STAKEHOLDERS IN DETROIT, MI  1.1 Introduction A householdÕs food security status is shaped by the complex interactions between its endowed resources and the broader food environment (Campbell, 1991). Early literature on hunger documents the necessity of household income to access and participate in the food system (Webb Patrick;Coates & Ser, 2003). In more recent literature, the food environment and physical access to food retailers has become an emerging area of study, engendered by the concept of food deserts (Beaulac, Kristjansson, & Cummins, 2009; Guy, Clarke, & Eyre, 2004; McKenzie, 2014). Though the food desert literature has its critiques and limitations (Wrigley, Warm, Margetts, & Whelan, 2002), many empirical studies have concluded that there is a relationship between physical access to full-service grocery retailers and nutrition related health outcomes (Hendrickson, Smith, & Eikenberry, 2006). One area improving our understanding of urban food security is research on food environments, which include measures of market composition mix between healthier and less healthy options, and how households access food establishments (Shannon N. Zenk et al., 2012; Widener, Farber, Neutens, & Horner, 2013; Wrigley, Warm, & Margetts, 2003).  There is also a recent development in the behavioral health literature to better understand the effect perceptions of time scarcity have on food consumption choices. For this research, we partnered with FoodPlus Detroit and the Detroit Food Policy Council to first identify a meaningful problem statement that was both useful for our partners and aligned with our 4  research interests. There was shared interest in understanding how community food insecurity manifests as a result of system feedback.  A major goal of the research was to develop a shared understanding of the broader system and to test the impact of proposed management and policy strategies. A more specific research question emerged - how do complex interactions of household resource endowments generate patterns of food (in)security.  1.2. Literature Review of Drivers of Food Insecurity The food security literature identifies barriers to urban food security as being a result of the constraints of household resources and characteristics of the larger food environment (Beaulac et al., 2009; Campbell, 1991; Walker et al., 2010). C.C. Campbell (1991) describes the systems, private, public and informal, that impact resources at the household level and affect food security--namely money, time, information, and health status. Campbell distinguishes between the experiential dimensions of food security and the social context, where the experiential explains the food security outcomes as diet sufficiency and its effects on health and quality of life. Campbell writes, Òat the community level, the constructs most often included are the availability of food markets, the actual quantity and quality of food present in food marketsÓ and the Òability of people, both financially (considering price relative to individualsÕ abilities to command resources) and physically (in terms of transportation issues or physical disabilities), to acquire the food that is availableÓ (Campbell, 1991). The Campbell framework also provides a useful way to characterize the food security literature. In this conceptualization, a householdÕs resources are a product of, and often defined by, the dynamics of larger community systemsÑthe local economy, labor market, education, and nonfood expenditure prices of housing, taxes etc. The traditional policy action to combat hunger has been through government food assistance programs which focus on increasing household purchasing power in food markets (Vasquez et al., 2007; Walker et al., 2010; Wiig & Smith, 2009).  In reviewing 5  the literature, there is a heavy focus on the characterization and access to the Ònormal food systemÓ. This work often includes references to and descriptions of Òfood desertsÓ.  The term Ôfood desertÕ is first referenced in Scotland by public housing residents to describe living conditions and its meaning has evolved over time (Walker et al., 2010). The underlying principle of a food desert is the absence of supermarkets in a geographic location, but no firm definition exists and authors often expand on this definition (Guy et al., 2004; Hendrickson et al., 2006; Walker et al., 2010). Henderson et al describe it as ÔÔurban areas with 10 or fewer stores and no stores with more than 20 employeesÕÕ (Hendrickson et al., 2006). Cummins and Macintyre (2002) expand on this definition to include an emphasis on food quality,  stating that there are ÔÔpoor urban areas, where residents cannot buy affordable, healthy foodÕÕ (Cummins and Macintyre, 2002). Beaulac (2009), in a comprehensive meta study on food deserts finds that Òevidence is both abundant and robust enough to conclude that Americans living in low-income and minority areas tend to have poor access to healthy foodÓ (Beaulac et al., 2009). Hendrickson et al. (2006) studying the quality and availability of food in urban grocery stores find that prices are higher and food quality is poorer, in areas where poverty is the highest. Additionally, Hendrickson et al. find that there is less quantity and variety offered at stores in impoverished areas. They also found food prices in the urban food desert were higher than in suburban neighborhoods (Hendrickson et al., 2006). Moreover, lack of transportation is an access barrier described in the literature, as many low-income households lack access to cars and are unable to afford the costs of getting to larger supermarkets outside of their immediate neighborhoods (Guy et al., 2004; Hendrickson et al., 2006; Rose & Richards, 2004).  Hillier et al. finds that low income parents travel further to shop for food (Hillier et al., 2011). Clifton, in a case study examining mobility strategies for low income food shoppers found that the most common and useful strategy is for households to purchase a vehicle for transportation 6  (Clifton, 2004). The interaction of spatial proximity and how people access food through the transportation system is being addressed by some in the field to include spatial-temporal measurements in food environment studies (McKenzie, 2014; Rose & Richards, 2004; Widener et al., 2013).  Behavioral health researchers have investigated how perceptions of time scarcity affect food consumption choices. Furst (1996) developed a conceptual model of food choice making, documenting that time, as a resource stock, influences food choices (Furst et al., 1996). In a review of the literature on perceptions of time scarcity and food choices, Jabs and Devine (2006) document the growth in interest to further understand how time influences food decision making (Jabs & Devine, 2006). They pull from the research citing how changes to intra-family dynamics have influenced meal planning had how these changes are due added time pressure (Connors, Bisogni, Sobal, & Devine, 2001) and that time is an important barrier to healthy behaviors (Furst et al., 1996). They document that time scarcity has been linked to obesity (Cawley, 2004) and the rapid sale of convenience products (including food (Gofton, 1995). They point out that sale of convenience foods are on the rise (Jekanowski, 1999), that fast food sales have increased for low income households, and that convenience foods and foods eaten outside of the home have lower nutritional value (Guthrie, Lin, & Frazao, 2002).  Much of the reviewed literature has called for systems thinking around food and nutrition security. The community food security literature states that  to conquer food insecurity it is necessary to address the governing systems first (Bellows & Hamm, 2002; Hamm & Bellows C, 2003; Pothukuchi, 2011). CampbellÕs food security framework emphasizes the interconnectedness of systems and household resources and the systemic barriers to achieving security (Campbell, 1991). In a review of the literature on food security and health disparities, Walker et al. (2010) conclude by recommending Òan innovative method such as concept mapping, a participatory research method 7  that allows hypotheses to be generatedÉÓ and using the data to provide Òunderstanding of the complexity of food access and the food environment, while providing a basis for program planning and policy development aimed at addressing access to healthy and affordable foodsÓ(Walker et al., 2010).  1.3.1 Research Design & Methods This research was designed as a community engaged scholarship project. We partnered with FoodPlus Detroit" and the Detroit Food Policy Council# to first identify a meaningful problem statement that was both useful for our partners and aligned with our research interests. Engaged scholarship has reciprocal learning at its core.  Once the problem statement was established, we worked with our community partners to design the research process illustrated in Figure 1-1. The research design began by identifying key stakeholders in Detroit with experiential knowledge of the systems governing food insecurity. We then conducted semi-structured interviews with stakeholders around barriers to household food security in the city. The next step was to conduct a workshop in Detroit to construct a qualitative model of the system behavior. We then used the outcomes from this workshop and the interviews to develop and build a quantitative system dynamics model of urban food security. We then demonstrated this model to community stakeholders in Workshop 2 and received feedback on its assumptions and behavior.                                                 " www.Foodplusdetroit.org # www.detroitfoodpolicycouncil.net Figure 1 - 1: Research Design 8  1.3.2 Methods 1.3.2.1 Interviews Our community partners identified fifteen (15) key stakeholders to include in the semi structured interview sessions. Stakeholders were affiliated with or represented interests from urban agriculture, local government, food sales and distribution, economic development, emergency food services, small business owners, and entrepreneurs (see Appendix A for full list). The interviewees were prompted with questions focused on the patterns and drivers of food insecurity, over time (see Appendix B for interview structure). Our goal was to elicit comments that would inform the system structure and behavior, so we also asked about perceptions of proposed solutions, both those that have been and have not been implemented. We also asked about perceptions of the future and if the participants expected things in the food system to improve, worsen, or stay the same. This process was conducted to prime participants to think about the systemic issues governing food security in their communities. The data was used to inform the quantitative system dynamics model and the scenarios tested in Section 4.  1.3.2.2 Participatory Model Building Participatory modeling or Group Model Building (GMB) is a tool that has been used to mediate consensus and understanding of a problem statement (Hovmand, Ford, Kyriakakis, Brown, & Brown, 2009; Van den Belt, 2004). It is useful when multiple stakeholders hold competing mental models of how a system operates (Van den Belt 2004; Olabisi 2008; Hirsch et. al, 2007). Like traditional system dynamics modeling, it utilizes a simulation tool to examine the behavior of complex systems over time (Olabisi, 2008). Its main features are the ability to represent feedback (circular causal relationships) and stock-and-flow dynamics.  Through simulation and informal maps, the models assist with understanding the endogenous sources of system behavior (CITATION).  9   System dynamics has commonly been used to represent embedded complex problems that have time-delayed outcomes and nonlinear responses to interventions (Sterman, 2001; Forrester, 1994). It has been used to rigorously test the implications and effectiveness of policy interventions at the community, state, and national level (Stave, 2002). Furthermore, the development of easily understood models  can provide a context and application that can inform  both stakeholders and policy makers alike (Ghaffarzadegan, Lyneis, & Richardson, 2011). The process, though relatively new in the food systems literature, is well placed to capture stakeholder knowledge, facilitate discussion, and engender system level understanding.  Causal Loop Diagrams (CLD) can be used to illustrate and document the causal mechanisms and feedbacks governing a system  (Kirkwood, 2013; Van den Belt, 2004). Creating CLDs is a process that explicitly lays out assumptions of causal relationships and identifies any mutually causing variables, or feedback (Sterman 1992).  Much of the food systems and food security literature makes use of terminology that makes the topic well placed within a feedback approachÑi.e. vicious cycles, poverty traps etc. 1.3.2.3 Workshop 1: Causal Loop Diagrams Workshop 1 of our research design centered around mapping potential barriers of food security in Detroit. With guidance from our community partners, we invited participants and key stakeholders with unique and experiential knowledge of the food system to participate. Workshop 1 began with the focal question: What are the drivers of food insecurity in the City. This focal question was open to different units of analysis (community, household etc.). The workshop allowed stakeholders to work in small groups to diagram and map their perceptions of the system structure. The small groups worked independently, with assistance from facilitators who answered technical questions. The small groups then explained their diagrams to the larger group for input, critique, and to gain group consensus that the diagrams were accurate. 10  The modeling team then worked together to integrate and aggregate the diagrams into a qualitative model. This iterative process resulted in Fig 2. Fully assembled, the qualitative model documents fifteen feedback loops, thirteen of which are reinforcing and two of which are negative or balancing. The diagram represents the stakeholder views of the system and its causal mechanisms.  There are four segments of the aggregated CLD addressing the multiple broad areas that the group identified. These segments, which are interlinked in the aggregate diagram, have been identified as the Home Economic, Cultural-Nutritional, Socio-Political, and Peer Network segments.  Each segment has specific features and drivers and some are operating at different scales. Though some of the segments deal with macro level system behavior, all groups identified how the processes impact community and household food security.        11  Figure 1 - 2: Aggregate Causal Loop Diagram   1.3.2.4 Home Economics Causal Loop Diagram The Home Economics segment of the CLD captures the variables and conditions, at the household level, that lead to food and nutritional security outcomes.  The diagram segment is centered around household income. The first feedback loop illustrates that low income restricts the householdÕs ability to purchase and consume good food, creating negative health effects. Negative health effects are seen to reduce employment opportunities and lower worker performance, which reinforces lower household income.  The second feedback loop begins with access to transportation, affecting a householdÕs ability to buy healthy food, which affects health status negatively, which leads to 12  employment and income reductions as well as lowering access to transportation. Access to transportation also impacts a householdÕs employment opportunities, which affects household income, creating another reinforcing feedback loop. Figure 1 - 3: Home Economics CLD     1.3.2.5 Cultural Nutritional Casual Loop Diagram  The Cultural-Nutritional diagram region has many reinforcing feedback loops as illustrated in Figure 1- 4. The loops are clustered together because they deal with the attainment of nutritional information, the development of nutritional preferences, and the transfer of necessary skills needed to become nutritionally secure. The first loop deals with life skills, such as food preparation and budgeting, which are seen as not being taught by the under-performing educational system. The lack of these life skills is believed to perpetuate the educational systemÕs under-performance and is reinforced by intergenerational effects. The health of the local food industry, and its economic 13  potential, is seen to have a reinforcing relationship with the education system. In return, the health of the local food industry is seen to improve knowledge of the food system, creating more interest and activity in the local food industry, making it more viable and strong. Nutritional literacy, which the stakeholders defined as having the knowledge to acquire foods that are healthy, is dependent on the educational system, and has reinforcing relationships with cultural and family preferences and the level of exposure to other food ways and food traditions.  The group also identified a reinforcing cultural feedback loop with a phenomenon stakeholders called ÒFast-CultureÓ. Fast-culture was described as both the personal preferences and social norms that prioritize convenience living; immediate gratification, on demand results, and quick responses from others. This fast-culture was identified as being the source of time and convenience pressure, which has a reinforcing feedback loop with the nutritional value of food consumed. The logic is that more convenience pressure will reduce the quality of food consumed, by increasing the demand for convenience foodsÑhighly processed, prepared, or frozen meals. This consumption of convenience foods in turn, reinforces the fast-cultural norm and expectation. The group also identified a variable referred to as ÒconsciousnessÓ, which is also reinforcing the effects of fast culture and the nutritional value of food consumed. This variable was explained to be at a larger scale and stronger than fast culture alone, and represents the generational shift and acceptance of fast culture as a way of being.  14    1.3.2.6 Socio-Political Causal Loop Diagram  The Socio-Political segment of the diagram represents the broader social and political dimensions that limit (or empower) communities to seize opportunities to become food secure. There was a broad belief that people in power hold hostility towards low-income and minority communities and that this hostility was both historical and omnipresent. Though the literature on causal loop diagraming suggests that chosen variables should be neutral or the positive tense of the variable (Kirkwood, 2013), many of the variables in this segment were written in a way that captured the groupÕs true meaning and understanding of the role the variables play in the system.  Figure 1 - 4: Cultural Nutritional CLD 15  Cultural racism was described as a powerful force affecting many of the system variables and processes. The group identified a reinforcing feedback loop with cultural racism and images in the media of people of color (POC). These images are informed and shaped by cultural racism, and the dissemination of these images perpetuates the very racism that created them. The group also identified geographic racism as a variable affected by cultural racism and pressure for gentrification. Geographic racism was a term used in the group to describe the historical and systematic process of underfunding black neighborhoods, restrictive housing policies, segregation, and environmental injustice in Detroit (Glaeser, Resseger, & Tobio, 2008; Reardon et al., 2008; Schulz, Williams, Israel, & Lempert, 2002; Zenk, Schulz, Israel, et al., 2005). Geographic racism is shown in the diagram to influence polices that are restricting access to conventional food stores, lowering access to transportation for POC via routing of public transportation, and the placement of other environmental obstructions (highways etc.). Gentrification has a reinforcing feedback loop where neighborhood activities (like urban food production) are not valued by the local government. This dismissive valuing of neighborhood activities increases gentrification pressure, and leads to restrictive regulation on neighborhood activities.  Racism also influences both legislation and the legislators who make food assistance program decisions. In the Home Economics CLD segment, food assistance is in a balancing loop with household income (the more in need a household the more resources are allocated) but in this segment, it is shown that policy makers may use legislation of food assistance programs against POC, the poor, and the elderly by restricting access and publically shaming recipients as scape goats, or sources of larger budgetary problems. The use of these political weapons increases the stress of households, reduces confidence in the job market, and lowers a householdÕs ability to purchase food; all leading to negative health outcomes.  16   1.3.2.7 Peer Networks of Influence Causal Loop Diagram  The final diagram segment, Figure 1-6, contains elements from all of the other segments, but is isolated here as it deals with how personal networks shape and are shaped by the food system. At the center of this segment is a variable called Network, which describes the social network of individuals as well as the aggregate network within the community. The network serves as a resource vessel, a capacity tool for social and political change, and a source of information for cultural and family preferences.  As a resource, the individual social network can assist in finding employment (and educational) opportunities. These new employment opportunities then evolve the social network of the Figure 1 - 5: Socio - Political CLD 17  individual in a reinforcing feedback loop. The group also identified that an individual social network may also act as a resource by connecting individuals who may be able to pool resources to cope with food insecurity. This is represented in the diagram through transportation access. An individual may be able to access transportation through their network and increase access to healthy food options.  There is a very complicated feedback loop between the Network variable and a householdÕs ability to purchase good food. The process functions when households experience food insecurity and reach out to their networks. Through cooperation and collaboration, they organize to effect policy change through legislation. This balancing feedback loop works through many variables; increasing food assistance programs, increasing support for community activities and urban agriculture, and increasing access to transportation, all of which increase the access to, and consumption of Ògood foodÓ.  1.3.3 Developing the Quantitative Model  We utilized the qualitative CLD (Figure 1-2) to inform the creation of a quantitative system dynamics model (Kirkwood, 2013). Because the qualitative CLD documents subsystems operating at Figure 1 - 6: Peer Networks of Influence CLD 18  different temporal and geographic scales, we worked with our community partners to identify a model structure that would address and inform the discussions coming out of Workshop 1 and incorporate many of the feedback loops from the CLD. The contents of the CLD demonstrate how the system is operating at two scales: the larger community food system, and the dynamics influencing household food security. With our community partners, we chose to build the quantitative model at the household level as there was great interest in how policy interventions impact food security status at the household level. It was also believed that the modelÕs output at this scale would be more easily understood by community members and associated researchers, and inform immediate policy considerations. 1.3.3.1 Model Description  The system dynamics model depicts a single household in the City of Detroit. The household is programmed to make food purchasing and consumption decisions for different types of food products.  The household attempts to fill its food pantry stock by purchasing Òhealthy foodÓ or Òconvenience foodÓ. It is constrained, however, based on available income and time.  The time constraint is introduced by the physical distance the household is from produce vendors, and the type of transportation available to the household. The model utilizes the daily recommended consumption of fresh fruits and vegetables as a proxy for Òhealthy Ò food consumption (U.S. Department of Health and Human Services and U.S. Department of Agriculture, 2015). A list of all model equations can be found in Appendix C. The modeled household makes two decisions every time step:  the type of food to purchase and the type of food to consume. The purchasing decision functions by maximizing healthy preferences given the constraints of time and income. The consumption decision is based on current food stocks, current time stocks, and the householdÕs perception of time scarcity. The householdÕs time perception is a graphical function which depicts a sigmoid line for the relationship of available free 19  time and healthy food consumption.  When time is more available, the household attempts to consume healthier food if healthy food stocks are available. Convenience foods are thought of as highly processed/prepared foods that are often consumed as a time saver (Brunner, van der Horst, & Siegrist, 2010). If the household is low on time, they will eat these foods (if they are available in the pantry). Alternatively, if the household does not have enough food to eat, they will consume emergency food, which represents any food security coping strategy (not eating, going to a soup kitchen, food bank, eating at a friend or relativeÕs etc.) (Maxwell, 1996). If the household is low on time, and does not have enough convenience food, they will consume a prepared meal, or an Ôaway from home mealÕ and in some cases Òfast-foodÓ (Stewart, Blisard, Bhuyan, & Nayga Jr, 2004). Household decisions are influenced by these constraints and by happenings in the broader food system in Detroit. The model takes into account a simplified version of the food system, including the proximity of retail grocery stores, the amount of available alternative food system options (community gardens, farmers markets, CSAÕs), and the effects of peer influence. These broader food system influences are also affected by the householdÕs preferences, as there is a reinforcing feedback loop between household preference for fresh fruits and vegetables and the growth rate of the alternative food system.  Figure 1-7 illustrates the model structure; the oblong shapes represent six sub-models or modules that are interconnected in function. The Home Economics module is where the household sells their labor-time on the market and receives a wage. This model structures the amount of money that can be allocated for food, housing, transportation and bills, and receives feedback by way of a Health Event from the Nutritional Security Module.  The Nutritional Security module is where the purchase and consumption decisions are made for the household. This module uses inputs generated from the other modules and follows simple rules for allocating resources. This model has a reinforcing feedback loop with the Home Economics model; 20  more income for food leads to a higher level of nutritional food security which leads to a more stable availability of labor (time) and income for food (minus health related expenses).  There is also a balancing feedback loop if Home Economics increase income by way of working overtime, reducing time stock, and then reducing nutritional security. Figure 1 - 7: Model Overview  1.3.3.2 Household Economics Module The Home Economics Module follows a stock flow diagram that tracks the HouseholdÕs monthly income, and income allocated for food. Income is generated by way of labor and the costs of this labor (commuting) are also included in the outflow expenses. Expenses have priority of outflow 1, signifying that this household first pays its housing and transportation costs before allocating money for food. Transportation expenses include gas ($3.35/gal) per mile driven, and a monthly payment 21  for car servicing, insurance and leasing. This function can be toggled off which defaults the model to use public transportation, which takes more time but is significantly cheaper.  This food money then flows into a stock called Income for Food, which also has an inflow of Food Assistance, which is calculated using the USDA methodology for Supplemental Nutrition Assistance Program benefits (Food Research and Action Center, 2012).  Money Spent is the outflow for the Income for Food stock. This amount spent is generated in the Nutritional Security Module and represents what the house is spending per month on food. There is also an expense labeled Health Event which adds $200 a month in expenses if the event happens, which is to signify being sick, which may cost a family both work days and medical expenses. The module is defaulted to use an hourly work week, which is highly variable, between 120 and 200 hours a month, representing that many hourly employees have inconsistent scheduling and income.  1.3.3.3 Time Cost Module The Time Cost module uses a simple stock flow structure to depict time in and time out of a householdÕs available time stock. Each month 30*24 hours are added to the time stock, which expire in a month and join the Primary Time out. Primary time out also includes work hours, commuting hours, and other time (where food decisions are made).  The model calculates commute time by dividing hours worked in the month by an 8hr shift for commutes and multiplying by the median distance traveled for work in Detroit. Speed is captured in the Car Speed and Public Transportation Speed converters, which are 45 mph and 15 mph respectively. !"##$%&'()#&*++#",%-./'-"$01'2"03&4.&,5%-'"6'1-)6%78#&4)9,'':"##$%),5'4)1%9,:&7%09,1;"0%9%)",'1;&&4 1.3.3.4 Preference for Fresh Fruits & Vegetable Module  The Preference for Fresh Fruits and Vegetables module is where the household preference for healthy food is modeled. This preference represents a goal that the Nutritional & Food Security 22  Module utilizes to calculate purchasing decisions. The dynamics of this module are influenced by healthy eating education programs, peer behavior, and the level of household exposure to healthy options. It is important to understand that this preference goal is not updated based on food security outcomes. This assumption is important to the model implementation and how we interoperate the results. By having the preferences immune to the influence of food security outcomes, we preserve the opportunity to analyze a householdÕs ability to access and consume foods that it desires.  In contrast, a preference goal that is sensitive to food security outcomes, would have a household that may prefer and purchase more expensive healthy food options, and identify that this spending behavior leads to food shortages later in the month. If the preference goal was then updated, on account of this shortage, the household would then appear more food secure, by diminishing the gap between what it would like to be consuming and what it is able to consume. 1.3.3.4 Nutritional Security Module  The Nutritional Security Module is where the household makes decisions about which food to purchase and which foods to consume. It is a bi-flow relationship between two pantry stocks called Healthy Food and Convenience Food. These stocks are calculated in meals. Meals are purchased (inflow) once a month and consumed (outflow) at a rate of 3 meals a day per household member. 1.3.3.5 Purchasing The inflows follow a simple set of rules for how the Household will purchase food. It assumes that the household is trying to maximize its healthy food preferences and purchase of healthy foods given the constraints of time and income. Purchase quantity is limited to 45 meals a trip if the household does not have access to a car. Convenience foods are purchased at a quantity that satisfies the need to replenish the total stock of meals per month. Convenience meals purchased are a function of healthy meals purchased in the same time period. To illustrate this relationship, if the 23  household is one (1) member; they require ninety (90) meals per month. If they purchase thirty (30) healthy meals in a month the model purchases up to sixty (60) convenient meals, if the income for food is available.  The household also tries to maximize healthy meals consumed, which is limited by time and Healthy Meal stocks. We used a graphical function that illustrates the perceived time one needs to prepare food, which we derived from the American Time Use Survey. Convenience meals consumed is also a function of the healthy meals consumed, much like the purchase function. Besides going hungry, the household follows two more rules to satisfy their food needs. If they have low time and healthy foods the household can consume food outside the home, Ôfast foodÕ. If they have time and inadequate meals in their pantries, they seek emergency food, which can be an array of different coping strategies that the home may utilize to eat. 1.3.4 Model Demonstration & Validation: Workshop 2  Workshop 2, held in East Lansing, Michigan, was designed to demonstrate the quantitative system dynamics model with community stakeholders and to elicit feedback on the model behavior and assumptions. A Graphical User Interface (GUI) was designed to allow the stakeholders to interact with and navigate the model first hand. The GUI connected model parameters to sliders and buttons making it simple to change assumptions and analyze the results.  Stakeholders were encouraged to make hypotheses about system behavior and to test these with the model.  From the GUI display, stakeholders could also select key interventions which the modeling team made accessible with a single click. These interventions were designed with information from the stakeholder interviews about possible solutions to food insecurity. Table 1 describes these 24  interventions and their operations. The stakeholders were also prompted to create their own Ôon the spotÕ system interventions and were guided by facilitators to incorporate them into the model.  Stakeholders were also prompted to give feedback on the model and its assumptions. This was an effort to both validate the model accuracy and improve the model structure. Stakeholders identified three (3) areas of concern with the model, the first being the nutritional composition of ÒAway from home/Prepared MealsÓ. In the model interface these meals are categorized as ÒFast FoodÓ which brought connotations of unhealthy behavior. Though the model tracks these meals separately, it does detract away from the overall health of the household and contributes to the probability of a health event$. While research supports the finding that away from home meals are of lower nutritional value (Guthrie et al., 2002; Jekanowski, 1999; Stewart et al., 2004) this is on average, and may not represent the preferences of all households as there are potentially healthier options.   The second concern was that of bundled food prices. In the model interface, stakeholders could select the relative price of a healthy meal compared to a convenience meal. This feature started                                                 3 Health Events are costly stochastic events whose likeliness of happening increase when adequate levels of ÒhealthyÓ food is not consumed. These events result in an increase in expenses for the month and a deduction from the time stock.  Table 1 - 1: Interface Intervention Definitions 25  discussion about this assumption. In the results section, we utilized sensitivity analysis with this relative price and found that it is important, but not a major driver of nutritional food security status.  The third concern with the model was that it lacked a feedback mechanism between changes in the Alternative Food Economy (AFE )and Preferences for FF&V. This feedback loop was added to reflect that as the AFE expands, it increases exposure to nutritious food and increases demand. The model described in the above section incorporates these additions and considerations. Overall the stakeholder group felt that the proposed model accurately captured their views of the complexity of household food and nutritional insecurity. The model was tested with other standard validity methods (Barlas, 1996). The equations were reviewed for consistency with the CLD and stakeholder interviews. The model was checked for consistency of units throughout. The model generated reasonable behavior for a wide range of parameter values including for extreme conditions. 1.4 Results 1.4.1 Reference Mode A system dynamics model reference mode is used to illustrate the problem statement that the modeling effort seeks to examine (Sterman, J,  1992). For this modelling effort, we are tracking a householdÕs consumption of fresh fruits and vegetables, which serves as a proxy for the nutritional quality of all meals consumed. The reason we are tracking food consumed, and not traditional food security measures like food access, is to examine the experiential and behavioral dimensions of food and nutritional security.  The reference mode is run with no interventions and is parameterized to represent a typical food and nutritionally insecure household in Detroit. The median household income for Detroit ($26,325) 26  and the median commuting time (26.6 minutes one-way) are used for parameters. Figure 1-8 Chart A documents the types of meals the household is consuming by percentage when the model is run for two years (24 monthly time steps). We see that the household is consuming far less than the recommended daily intake of nutritious foods and this tracks well with documented consumption habits for the residents of Detroit (Gunderston, C et al. 2016; Zenk,, et al. 2005). 1.4.2 Parameter Uncertainty  There are some model parameters that we have had to estimate because the secondary data was inconclusive or the value of an input variable was truly unknown. For these parameters we have tested the model with multiple runs, varying the parameter values incrementally.  This technique, referred to as sensitivity analysis, allows the team to understand how these parameter assumptions affect the model behavior (Hayward,  1961). Because system dynamics is a behavior-oriented simulation tool, the sensitivity of this system behavior (equilibrium points, oscillation amplitude) will be of most concern (Barlas, 1996). If the model behavior is highly sensitive to the set variable/assumption, this will help identify areas for future data collection and research.   1.4.2.1 Cost of Healthy Meals  There is some debate in the literature on the price difference between healthy and non healthy foods (Carlson and Fraz‰o 2012; Zenk, et al. 2005).  Figure 1- 8, Chart B. and C. demonstrate the modelÕs sensitivity to relatively more expensive costs of healthy foods. As determined by the model structure, the Percent of Healthy Food Consumed Graph (Chart B) shows that for most variance in meal price, the behavior is mostly consistent. This is because the modeled household seeks to maximize its preference for healthy food, which is not 27  influenced by perceptions of affordability4. However, the more expensive runs of the model reveal that the Percent of Emergency Food Consumed is sensitive to this input, varying between 0.3% of total food consumed on the lowest end and 5.6% total food consumed on the high end. 1.4.2.2 Alternative Food Economy Growth Rate It is uncertain how the Alternative Food Economy (AFE) is evolving in the City of Detroit. Some stakeholders believed it to be growing at a rapid rate, others less so. Figure 1- 8 Chart D demonstrates the modelÕs sensitivity to changes in this growth rate on the percentage of healthy foods consumed. The growth rate is modulated incrementally starting with a 0.0% per year growth and moving up to 10% per year. Chart D demonstrates that this growth is an important variable in the model. Increasing the growth rate increases the demand for healthy meals and decreases the amount of healthy meals the household is required to travel long distances to procure. Increasing the growth rate also produces a small shift in the amount of Ôaway from home mealsÕ the household consumes, as the householdÕs preferences have changed, but it is still time constrained. This behavior is due to a time delay between how quickly the AFE responds to increases in demand. The growth rate is important because there is a feedback loop between FF&V Preferences and the AFE, where the more the household prefers healthy food the more the AFE grows, the more the household will be exposed to healthy foods, changing its preferences.                                                 4 Assumptions related to household preferences are discussed in the model description under Preferences for FF&V. The assumption is that preferences are not updated based on food security outcomes or perceptions of relative prices.  28  Figure 1 - 8: Reference Mode & Sensitivity Analysis   increasing  1.4.3 Scenario Results For the modeling workshop, the model parameters were flexible so stakeholders could create their own scenarios. We included many preprogramed interventions as outlined in Table 1. However, to interpret the effectiveness of interventions, we have created a scenario space that describes the initial household conditions and documents the model behavior when different interventions are applied. 29  The interventions modeled below fall into simplified categories because of how they act upon or influence model parameters. The focus is on household behavior, preferences, and access to resources. The influence of peer networks and the local food environment are modeled as endogenous to the household in these model runs.  1.4.3.1 Scenario 1 Scenario 1 represents a Detroit household that is quite vulnerable to food and nutritional insecurity. The scenario simulates a household of three, which has one income earner making the minimum wage, with a highly variable work schedule. The household is not participating in any federal or state supplemental nutrition programs during baseline, and does not have access to a vehicle. This variation in work schedule produces two constraints on the household, the first being the variation in income and the second the amount of time the household has to procure and consume meals. The variation in food consumption by type, shown in Figure 1-8 A, is largely driven by the ebbs and flows of this work schedule. Over the two-year period, this results in the household consuming 34% emergency meals, because it is lacking the financial resources to purchase food, 42% convenience meals, and 24% healthy meals. 1.4.3.2 Scenario 1 Interventions Graph C and D in Figure 1- 9 demonstrate the effects of the pre programmed interventions. Applying for and receiving SNAP does marginally increase the quantity of healthy meals the household is consuming by a 6 percentage point difference. SNAPÕs largest role in this scenario is reducing the amount of emergency meals the household is consuming by a difference of 36 percentage points. In the intervention, the emergency meals are largely replaced with convenience meals and not healthy meals due to access, time shortages, and preferences. Adding a healthy eating education program, to increase healthy food preferences, also increases consumption of healthy 30  meals by 5 percentage points, but increases the amount of emergency meals the household consumes as well. This counterintuitive outcome is driven by the increased time and financial resources a household is using to meet this healthy eating goal. Because the inflow of financial and time resources is variable, in time steps where these resources are scarce, the household no longer has the resources to purchase less expensive meals, both financially, or in terms of time costs.  The healthy eating education program, which focuses on shifting household preferences, also influences the growth rate of the local food economy, and this marginally increases access and exposure to healthier food options.  Graph D illustrates healthy eating consumption when the model is set to receive SNAP benefits and is allocated an additional hour for each day. This combination of household resources stabilizes the consumption of healthy foods.  The additional hour of time per day represents interventions that are targeting saving households time. This additional time could be improvements to public transportation speeds, development of organizational skills, or a change to family dynamics or other similar interventions. 31  Figure 1 - 9: Scenario 1  1.4.3.3 Scenario 2  Scenario 2 represents a Detroit household that is vulnerable to food and nutritional insecurity. The scenario simulates a household of three, which has one income earner making $12 an hour, and a variable work schedule. The household is participating in the federal SNAP program and has access to a vehicle. The baseline run for this scenario, illustrated in Figure 1-10 A, shows that the household is consuming 38% healthy meals, 56% convenience meals, 2% away from home meals, and 4% emergency meals. The variability of the diet is largely driven by the variable work schedule placing pressure on the time stock. Figure 1-10 B, shows how perceived time scarcity effects 32  consumption decisions. Each model run in Figure 1-10 B, increases the time stock incrementally. The final model run (6) increases the time stock by one hour per day and reduces the variability and increases quantity of healthy food consumed. 1.4.3.4 Scenario 2 Interventions Graph C in Figure 1-10 demonstrates the effects of various interventions on household healthy food consumption. The first intervention is a healthy eating education campaign that targets household food preferences. This intervention works to increase healthy meal consumption by 10 percentage points, decreases the consumption of convenience meals by 22 percentage points, and increases away from home meal consumption by 12 percentage points. The increase in consumption of away from home meals, which tend to be less healthy, is counterintuitive. It is caused by the increase in preferences for healthy meals and the household time stock remaining scarce. When the household perceives its time stock as scarce, the household tries to consume a convenience meal, with none available the household consumes a prepared meal, or away from home meal instead. The next intervention is a combination of additional time and the education component previously noted. This intervention has the effect of increasing healthy meal consumption by 34 percentage points, reducing convenience meals by 33 percentage points and reducing away from home meals by 2 percentage points. This combination intervention has the outcome of a fairly consistent diet with an average of 75% of meals being healthy.  The variability of the diet in this scenario is largely driven by the variable work schedule placing pressure on the time stock, and less so the variability in income. Figure 1-10 Chart B shows how perceived time scarcity effects consumption decisions. Each model run in Figure 1-10 Chart B increases the time stock incrementally. The final model run (6) increases the time stock by one hour per day and reduces the variability of and increases quantity of healthy food consumed. The variability of run six (6) in Chart B. is largely driven by the variability in income over the period. 33  Figure 1 - 10: Scenario 2  1.4.3.5 Scenario 3 Scenario 3 represents a Detroit household that is vulnerable to food and nutritional insecurity. The scenario simulates a household of three, which has one income earner making $18 an hour, and a variable work schedule. The household is participating in the federal SNAP program, though it is only periodically eligible for benefits%, and has access to a vehicle. The baseline run for this scenario, illustrated in Figure 1-11 A. shows that the household is consuming 26% healthy meals, 70% convenience meals, 5% away from home meals, and 0% emergency meals. The variability of the diet is largely driven by the variable work schedule placing pressure on the time stock. Figure 1-11 B.                                                 5 According to Feeding America, in Wayne County, where Detroit is, 18% of the food insecure population, is ineligible for federal and state food assistance programs. This scenario explores these dynamics.  34  shows how perceived time scarcity affects consumption decisions. Each model run in Figure 1- 11 B. increases the time stock incrementally. The final model run (6) increases the time stock by one hour per day and reduces the variability and increases quantity of healthy food consumed. 1.4.3.6 Scenario 3 Interventions Figure 1- 11, Chart C. demonstrates the model output for various interventions for this scenario. The first run represents the baseline with no interventions. The second run (and first intervention) is a healthy eating education campaign targeted at increasing the householdÕs awareness and preferences for healthy meals. For this scenario the intervention works to increase healthy meal consumption by 13 percentage points, decrease convenience meal consumption by 25 percentage points, and increase away from home consumption by 12 percentage points. There was no change to emergency food consumption. The increase in away from home meal consumption, as in Scenario 2, may be counter-intuitive, but is a result of the decrease in convenient meal purchasing, and perceived time scarcity. The household perceives that it is time poor, and then wishes to consume a convenience meal, but with limited meals in its pantry stock it then choses to eat a prepared meal, away from home. The third intervention combines the time intervention, adding an extra hour of perceived free time per day to the time stock, and the healthy eating education program. It results in an increase of healthy meals consumed by 38 percentage points, decreasing convenience meals by 36 percentage points, and decreasing away from home meal consumption by 2.5 percentage points. 35  Figure 1 - 11: Scenario 3  1.5 Discussion Our model results document how specific parameter limits govern the dynamics of household food and nutritional security. These limits operate by restricting a householdÕs ability to exploit the food security opportunities it is endowed with and those that are created by the interventions. As all three scenarios demonstrate, the effects of singular interventions are largely ineffective because the opportunity set they create is tempered by other limits. For instance, in Scenario 1 the variability of the householdÕs healthy eating behavior is being driven by the variability of the work schedule, both in terms of an income limit and time pressure limit. Applying a food income intervention is not fully effective as time pressure is then the dominant limit. There is also a complex interaction in how singular interventions fail to be effective solutions. Adding a vehicle to the household endowment 36  does reduce the pressure of the time stock, but comes at a cost, reducing available income for meals and potentially reducing savings for times of higher economic need. We also document how healthy eating education interventions can work to outpace a householdÕs healthy eating preferences from its physical access, economic realities, and time constraints. Largely the results suggest that interventions are much more effective if they are designed to target multiple limits or drivers of food insecurity.  Much of the literature around household food security deals with what Campbell describes as the Òsocial aspects of food security,Ó focusing on household resources and characteristics of the food environment (Campbell 1991). This focus on the social aspects is evident in a literature review by Walker et al.(Walker et al., 2010). Though useful for the creation of food security indicators and monitoring, this focus may lead to a limited understanding of the complexity and systemic factors that may lead to a household experiencing food insecurity. Research that has included the experiential dimension of food security has done so through the use of food diaries and survey methods (Walker et al.  2010; Wrigley, et al. 2003; Wrigley et al. 2002). This approach has revealed implications for households living in different food environments, but is limited in quantity of studies and the scope of dynamics that are able to be observed. An advantage of our modeling approach has been the ability to study the experiential dimensions of food security from stakeholder perspectives and to simulate these dynamics over time. In our results section, we illustrated counterintuitive behavior, in which some interventions lead to an increase in away from home meal consumption or more reliance on emergency meal coping mechanisms. This was driven by system feedback and delayed effects between food availability and household preferences.  This system behavior may reveal unintended consequences of interventions and programs that fail to include an experiential focus. Though the implications of this model are limited, we believe that it demonstrates 37  why more research is required to compare social and experiential food security indicators, and to accurately capture the consequences of living within different food environments.  Our focus on experiential outcomes, and participatory methods, allows our model to take an expanded view of household resources; incorporating household knowledge, time availability, preferences, and income. We believe that documenting the interactions of these resources is a novel and important outcome of this research. The model output shows that households face periods of food insecurity when income and time availability fluctuates with variable work schedules. We were also able to merge research findings on behavioral health and food environments to explore the importance of time as a resource stock (Daly, 1996; Jabs & Devine, 2006; Jabs et al., 2007; McKenzie, 2014). Time affects the model as both a stock and through the perception of time scarcity. As a resource stock, available time is a limit to the procurement of food items. This is an interaction with the food environment, physical distance to fully supported grocery stores, and the availability of and access to, transportation. This finding largely supports the incorporation of temporal distance, and time distance measures into the analysis of food deserts and food security (McKenzie, 2014; Rose & Richards, 2004). Secondly, a householdÕs perception of the necessary time to cook, clean, and consume food leads households to choose alternative options for consumption, even if their food stock is plentiful. In our model, this leads to the consumption of potentially less healthy options, and food spoilage. Though these dimensions of time have been utilized in the literature separately, we believe the interaction of available time at these two points in the decision process is important. Our model explicitly assumes that temporal distance drains the household of their time resource stock when they must travel for a long time procuring food; this then shapes how they perceive available time and their household food knowledge, when they make the consumption decision. Our model shows that coupling a time component with many of the other interventions has reinforcing effects, multiplying the effectiveness of interventions.  38  Further research is necessary to test the nature of these dynamics at different scales. It is also important to consider the more macro scale dimensions of the CLD, notably the socio-political segment, which potentially can be important to the long-term system behavior, in how household actions shape the food landscape and access issues. 1.5.1 Limitations The system dynamics model presented in this article is based on an integration of stakeholder mental models with academic theory and secondary empirical data. Our stakeholder group largely represented practitioner knowledge and expert testimony from years of experience working in the Detroit food system. A fair criticism of our process, is that we did not (to the best of our knowledge) receive first hand experiential knowledge of food insecurity. The household decision process in our model is based on theory and our assumption that households will attempt to maximize their healthy eating preferences. For our purposes this was useful in studying the system behavior and allowed us to test the impact of policy change on the opportunity set for households. At worst, this assumption would mean that our model is demonstrating the best case scenario for fulfilling preferences given the constraints in the system. A group model building process with food insecure households could prove very advantageous and yield more system discoveries, as well as providing another source of validation for the model findings.   Another limitation is that our model, though it includes a representation of temporal distance as a function of transportation speed and distance, is not a geographically explicit model. A geographically specific model could introduce other elements into the temporal distance calculation, including congestion, road conditions, public transportation schedules, walkability, and safety. We do not believe that these variables would drastically change the general system behavior, but could add great clarity into the heterogeneous landscape of household food security within the city.  39  Our model may be limited by the way we approached intra-household dynamics. In the model, all household activities that require time, including all aspects of procuring, preparing, and cleaning up of meals is attributed to the same time stock. Some research exists on how the shift in intra-household dynamics impacts food consumption and time allocation decisions, though we didnÕt find conclusive evidence to represent these effects in the model. It was also not represented in the group causal loop diagram, but we feel it could be important, especially in circumstances in which households are utilizing emergency food coping mechanisms. Finally, it also assumed that the household is homogenous in respect to eating preferences and dietary requirements. There could be an important delay in how a family adapts to shifts in preferences by the primary food decision maker. For instance, a parent could purchase healthier meal options and receive feedback or resistance from family members, which may result in the food going to waste. This could result in reshaping the preferences of the purchaser in a balancing feedback loop.  1.5.2 Potential Policy Implications  Interpreting the model behavior can be useful for informing policy considerations. It should be done with the cautious understanding that the model is not meant to be predictive, but used as a tool to better understand the interconnectedness of variables driving system behavior. Given the limitations outlined above, we believe there are policy and programmatic areas where the model can help inform discussion.  Our model demonstrates that coupling a time component with many interventions has reinforcing effects, multiplying the impact of interventions. Conceptualizing a time intervention may be difficult, and further research is needed, but here we will point to some hypothetical interventions that may be considered. For instance, at the national level, food assistance programs could make allowances for additional costs of semi-prepared healthy food options. We believe this could help reduce household time pressure, but more research is needed. Information and research on the marginal 40  time savings and price premiums for such a program change are out of scope for this project, but could reveal very important considerations. We also envision programs that assist people in understanding the true time it takes to prepare and consume healthy foods, and linking these programs to farmerÕs markets and grocery stores where people are purchasing their groceries. 1.5 Conclusions This modeling effort demonstrates the usefulness of utilizing a participatory process to unpack a complex social issue. The research design enabled the modelling to be iterative and allowed participants to see the benefits of collaborative research and systems thinking. The qualitative CLD documented and explored stakeholder understanding and knowledge of systemic structural issues facing residents of Detroit and how the combination of these forces interacting may limit opportunities. The quantitative model allowed us to explore the experiential dimensions of food and nutritional security and test stakeholder assumptions of how various interventions should be structured and implemented. The system dynamics model demonstrated the multiple drivers of food insecurity at the household level for residents of Detroit. Some of these drivers have been extensively documented in the literature including; the barriers of access, characteristics of the food environment, and the limits of household income (Campbell 1991; Hendrickson, et al., 2006; Walker, et al, 2010; Beaulac, et al., 2009; McKenzie 2014; Zenk, et al. 2005). We are also able to support findings that a householdÕs stock of available Ô free timeÕ and its perception of time are important factors in food related decision making (Furst et al., 1996; Jabs & Devine, 2006; McKenzie, 2014).  Our model adds to the understanding that these behavioral dimensions and access barriers interact to limit household food security opportunities. The modelÕs behavior also demonstrates the necessity of utilizing an expanded view of household resources, one that includes aspects of time management, knowledge, preferences, and peer behavior. We believe this research 41  has explanatory power in why these resources should be integrated into measurements of food security and is a novel and important outcome  42  CHAPTER 2   FACING THE PUBLIC: REPORT ON BARRIERS TO HOUSEHOLD FOOD SECURITY IN DETROIT, MI  2.1 Introduction This project report and policy recommendations were developed as the result of a two-year collaborative research process we embarked on with Food|Plus Detroit and members of the Detroit Food Policy CouncilÕs Research and Policy Committee. We have a shared interest in identifying how local food policy councils can address the drivers and barriers to food security. Our partnership was driven by a mutual understanding that community-university relationships can, when implemented correctly, result in collaborative and mutual creation of new knowledge and insight. We utilized a participatory process to gain understanding of the larger food landscape and food security challenges. We then built a demonstrative simulation tool to understand how households access the food system as a result of their personal resources interacting with the built environment, food landscape, and governance strategies.  Along with this report and recommendations, we have provided the Model Operation Guide which the DFPC can utilize to understand how policy interventions and program strategies affect household food security. The model is not predictive, but can assist in the development of strategies by demonstrating how food security manifests at the household level. We hope that this model will encourage and assist in applying a systems thinking perspective when designing policy strategies.  43   The DFPCÕs mission is to nurture Òthe development and maintenance of a sustainable, localized food system and a food-secure City of Detroit in which all of its residents are hunger-free, healthy, and benefit economically from the food system that impacts their lives.Ó To this end, the DFPC has been tasked with making policy recommendations to the Detroit City Council which is in the process of updating its Food Security Policy. The following report utilizes our model and research findings to analyze how DFPC proposed policy changes may impact the food security of households within the city.  2.1.1 Why a Systems Approach? As both researchers and citizens we believe that a systems thinking approach can empower communities to make complex and important decisions. Systems thinking is a way to view a problem and its causes as a whole system, recognizing that the patterns and cycles of behavior are a result of interrelated components and how they change overtime (Meadows 2008). Using systems thinking skills can add perspective to wicked problems and address the helplessness often associated with them. It can take this abstraction and provide a tangible understanding and an ability to find solutions to the root causes of problems (Meadows 2002; Stave 2002).   Food systems and the effect they have on Detroiters is inherently complex. The food system operates at many scales, both temporal and geographic (McKenzie, 2014). Food security is embedded in the food system and is influenced by forces in the economic, social, and transportation systems. It is resistant to singular solutions and dependent on context specific circumstances making it a Ôwicked problemÕ (Rittel & Webber, 1973). We believe this systems model and approach can add clarity, giving stakeholders the opportunity to envision new solutions and paths forward. 44  System dynamics (SD) is a quantitative simulation modeling tool used in a wide variety of fields to understand the behavior of complex systems over time. Its main features are the ability to represent feedback (circular causal relationships) and stock-and-flow dynamics. We chose SD for its intuitive understanding, and ability to display the interactions of system components operating at many different scales. SD is however not predictive, and its usefulness is to understand system behavior and to identify areas of importance and influence. 2.2 Process We identified our process collaboratively with our community partners, Food|Plus Detroit and members of the Detroit Food Policy Council. To build the model, we started by first interviewing key stakeholders with knowledge of the food and food security systems in Detroit. These interview questions focused on defining food security and why it is important in Detroit, and characterizing system behaviors. Our interviews also investigated proposed policy solutions and how these may or may not address the underling systemic drivers.  We then hosted a workshop in Detroit focused on diagraming the cause and effect relationships that may result in food insecurity. We used a technique called Causal Loop Diagramming to document the explicit paths and causal relationships that drive food insecurity (Kirkwood, 2013). One important feature of the method is that it focuses on variables that may have mutually causing relationships, or feedback. Examples of feedback loops are shown in Figure 2-1. The first image is a simple loop depicting population growth, where births per year are dependent on the current population and the current population depends on births per year. This type of feedback loop is reinforcing, meaning that the more population there is, the more births there will be. The second image illustrates a balancing feedback loop describing a social problem. In the loop, the severity of the social problem increases pressure to take action, which leads to 45  taking some action, and that action then reduces the severity of the problem. These types of loops are quite simple as they contain a few variables and only single loops. In complex systems we often see many variables and multiple feedback loops that combine to create behavior.   For our workshop we worked with four small groups of about five people per group to start mapping the causal structure of food insecurity in Detroit. We later combined these group diagrams in the full CLD. The full CLD, (shown below) has segments operating at different scales. These segments, which are interlinked in the full diagram, have been identified as the Home Economic, Cultural-Nutritional, Socio-Political, and Peer Network segments. Figure 2 - 1: Example of Feedback Loop 46  Figure 2 - 2: Aggregate CLD  Below, we explain in more detail two segments of the CLD that operate at the household level. These two are highlighted superficially because of this household focus and the important role they play in the quantitative system dynamics model. The Home Economics segment of the CLD deals with the variables and conditions, at the household level, that lead to food and nutritional (in)security.  The diagram segment is centered around household income. The first feedback loop illustrates that low income restricts the householdÕs ability to purchase and consume Ôgood foodÕ, creating negative health outcomes. 47  Negative health outcomes are seen to reduce employment opportunities and lower worker performance which reinforces and lowers household income.  The second feedback loop begins with access to transportation, which affects a householdÕs ability to purchase healthy food, which affects health status negatively, which leads to employment and income reductions and lowers access to transportation. Access to transportation also impacts a householdÕs employment opportunities, which affects household income, creating another reinforcing feedback loop.  Figure 2 - 3: Home Economics Segment   The Cultural-Nutritional segment of the CLD is highlighted here because it deals with the variables and processes that influence the flows of information and the formation of household preferences and skills that are necessary to become nutritionally secure.  The first loop demonstrates the reciprocal relationship between life skills, (thought of as food preparation and budgeting) and the performance of the educational system. It shows how, through intergenerational effects, the 48  educational system performance is reinforced by students not getting the skills necessary to be nutritionally successful.  Our process also identified a reinforcing loop with a phenomenon stakeholders called ÒFast CultureÓ.  This was described as the personal preferences and social expectations for immediate gratification, convenience living, and quick responses from others. This ÒFast CultureÓ is shown to lead to time and convenience pressure which is in a reinforcing feedback loop with the nutritional quality of food being consumed. When households are faced with time and convenience pressure, they prefer more convenience foods (highly processed, prepared, or frozen meals) and the consumption of these foods reinforces the expectation for convenience living.   Figure 2 - 4: Cultural Nutrition Segment  49  2.3 Model Narrative  A full description of the model can be found in the Model Guide6 which illustrates the stocks, flows, feedback loops, parameters and equations. Generally, thinking about the model in narrative form can enrich understanding.  The model is of a household in the city of Detroit that has two pantries, one for healthy foods and one for convenience foods (foods that are designed to be time saving, and tend to be less healthy). The household also has a stock of time. Some of this time is spent working (and receiving income), commuting, and completing other household tasks. The household then makes two decisions: what types of food to purchase, and later, what type of food to consume. This first decision, of what to shop for and purchase, is determined by the householdÕs preference for healthy foods, but is constrained by the amount of income they have to spend on food and the time they have to procure it. The food environment and physical distance from different food sources, is represented in the model as a time calculation; as it matters how the household will travel to procure food. Traveling by car is faster than traveling by bus or walking. The second decision the household makes is concerned with consumption. For this the household identifies how much time they have, and how they perceive this time. If the household perceives itself to be in a state of time scarcity, or not having enough time to take on the task of a healthy meal preparation, they reach for convenience food. If the household has no supply of convenience food, the household consumes an Ôaway from home mealÕ at a restaurant or other food service establishment. If the household perceives itself in a state of time surplus, or neutral time, it consumes the healthier option in their pantry. If the household pantry is low on food, the household will try to cope. This ÔcopingÕ is represented in the model as Ôemergency foodÕ but is referring to a number of different coping mechanisms, including borrowing money/food from a neighbor, sending children to eat at a friendÕs                                                 6 http://tinyurl.com/gp5ry5s   50  house, portion reduction, etc. (Maxwell, 1996). The household is also receiving input from the larger food system, which is simplified in the model but also is dynamic. For instance, we represent the Alternative Food Economy (AFE) as a growing force in the community bringing a portion of healthy meal products closer to households and exposing households to healthy option Ôfood waysÕ. 2.4 Simulation Workshop  We demonstrated this model at a second workshop, held in East Lansing, MI. We invited stakeholders from the first workshop to participate in the demonstration and to interact with the model. Stakeholders were encouraged to make hypotheses about system behavior and to test these with the model. Stakeholders were prompted to create their own Ôon the spotÕ system interventions and were guided by facilitators to represent these interventions with the model. We then received and incorporated their feedback into the model, making it more a more realistic and powerful representation of the experience in Detroit. 2.5 Results  To further explore the model, we designed scenarios that depict how different households, endowed with different resources, may experience food and nutritional security. We are also able to demonstrate how policy strategies may work to provide food insecurity relief. The following sections document the model output at baseline for different households represented in the City of Detroit. We then demonstrate and explain the effects of selected policy interventions. Though not all interventions are showcased, the interventions demonstrated below were chosen as they impact certain aspects of model behavior. For instance, a time intervention can represent any intervention that works to reduce the amount of time being withdrawn from the time stock. This could be decreasing commuting time through more efficient public transportation; reducing the time it takes to procure food by having more options closer, or adding time through household time saving 51  workshops. These interventions are all different, and their implementation would be important in practice, but generally affect model output in similar ways. Likewise, healthy eating education programs can be quite unique in practice. In the model, healthy eating education programs are a behavioral campaign that is targeted at the householdÕs eating behavior. The intervention is assumed to influence household preferences towards healthy foods, and then we test if the household is able to access and consume these foods. In practice, many interventions may have a combination of goals, targeting the physical environment, household behaviors, and household resources.   Figure 2 - 5: Household 1 Results  52  Figure 2-5 A above, is the output from the model and tracks the percentage composition of meals consumed over a two-year period. This represents a household in Detroit with poor physical access to supermarket stores, no private transportation and a varied hourly job, and which is not receiving Supplemental Nutrition Assistance Program (SNAP) benefits. The output shows that with these conditions the household is consuming about 24% healthy meals over the model run, 42% convenience foods, <1% prepared/away from home meals, and 34% emergency meals. What appears as peaks and valleys in the graph is due to the household responding to both food and time shortages as a result of income and hours worked being highly variable. Chart C shows the effects of selected interventions. Applying for and receiving SNAP does marginally increase the quantity of healthy meals the household is consuming by a 6 percentage point difference. SNAPÕs largest role in this scenario is reducing the amount of emergency meals the household is consuming by a difference of 36 percentage points. In the intervention, the emergency meals are largely replaced with convenience meals and not healthy meals due to access, time shortages, and preferences. Adding a healthy eating education program, to increase healthy food preferences, also increases consumption of healthy meals by 5 percentage points. When the education and SNAP interventions are run in unison, we see that overall the household is consuming more healthy meals. However, this does very little to address the general variability of the diet, as the driver of this behavior is linked to time constraints.  Chart B and D explore how the household responds to interventions that include a time component. This time intervention can represent any time saving element, whether it is more efficient transportation, time management training, or improved cooking/cleaning skills. In Chart B we ran the model with incrementally more Ôfree timeÕ, starting with an extra ten minutes a day and moving up to an additional hour of time per day. As the additional time comes closer to the full extra hour, the healthy consumption becomes much more consistent, as many of the smaller valleys 53  become smooth. In Chart D we illustrate what happens when the time intervention is combined with the SNAP benefits. This combined intervention largely stabilizes the consumption behavior, and the household is far closer to reaching its healthy eating preference and is consuming upwards of 46% healthy meals. This documents the importance of the household time stock.  When the model is set to represent a different household we see similar dynamics and behavior. The major differences are the hurdles the house must overcome to become food and nutritionally secure. Below, we document the model set to a household that is making more income, with a more demanding but less variable work schedule, which has access to a vehicle.  With no interventions, the household is consuming 26% healthy meals, 70% convenience meals and 5% away from home meals. The variability in diet is largely caused by the variations in the time stock. Though this household is modeled to have a more consistent work schedule, it is on average working more hours than Household 1, making it more sensitive to time in months where time becomes more scarce. Figure 2-6 Chart C shows healthy eating habits for Household 2 given the interventions of time and a healthy eating education program. As with Household 1, a time intervention of one additional hour of Ôfree timeÕ each day largely stabilizes the householdÕs consumption of healthy meals. Combining the time intervention with the healthy eating program increases healthy eating by shifting preferences, and the additional time allows the household to act on these preferences.  54  Figure 2 - 6: Household 2 Results  There is also interest in how the AFE is impacting food insecure households in Detroit. In our model the AFE represents all alternative food system components, including community gardens, farmers markets, and Community Supported Agriculture (CSAs). These components are all grouped together and impact the households similarly: by increasing exposure to different Ôfood waysÕ, working to shift preferences towards healthy, and by allowing a portion of the healthy food a household procures to be accessed more locally. A question arose in our community discussions regarding how quickly the AFE is growing. In Figure 2-7 we demonstrate how different growth rates of the AFE effect Household 1Õs healthy meal consumption by moderating the AFE growth rate incrementally from 0.0% to 10.0% per year. This demonstrates that this growth rate is an important variable in the model, as in later months the household responds to the growth by consuming a 55  healthier diet. Increasing the growth rate also creates a small shift in the amount of Ôaway from home mealsÕ the household consumes. This is because the AFE works to shift household preferences at a faster rate than the AFE can provide these additional healthy meals.  Figure 2 - 7 Alternative Food Economy  2.6 Food Policy Council Recommendations  Our model illuminates how barriers to food security may operate to limit a householdÕs ability to consume healthy food. These limits, broadly, are; time, purchasing power, and behavior. It is important to realize that many features of the system dictate how these household resources are or are not utilized. For instance, time stocks can be drawn down by an imbalanced food system where it is difficult to access healthy options, or they can be drawn down by not having adequate access to childcare, or by being subject to long commuting times. Similarly, a householdÕs purchasing power is relative to the prices in the marketplace and how system features impact the household money supply. If transportation costs are drawing down the money stock this works in a way similar to how expensive healthy options may be out of reach of the householdÕs purchasing power. Our model demonstrates how household behavior and preferences are dynamic and being shaped by the larger system. We utilized very simple rules to demonstrate how eating behavior may not be 56  representative of household preferences given the constraints of resources and physical features in the larger food landscape. However, it is important to do further research on the development and effectiveness of healthy eating programs. There also needs to be more research on the relationship between the AFE, household preferences and physical food access.  Our model starts to unravel the role that community based food systems and urban agriculture (AFE) may play in addressing food security and nutrition issues. Our stakeholders observe that the AFE may address access issues by bringing access to fresh foods closer to where it is needed. It may also have the effect of exposing residents to different food and food ways. This can help shape the preferences for healthy eating and living. It is important for food policy councils to consider current capacity and projected growth rates when reviewing AFE related policy.  Many of the federal and state policy programs address food insecurity by increasing the purchasing power of households. Though these programs have evolved to include nutrition related components, our model supports local food policy councils advocating for a review of the needs calculation and benefit allotments being used. A significant barrier to food and nutritional security in our model was the stock of time necessary to access and prepare healthy foods. Our findings support the idea that these programs could benefit by reviewing the time constraints that households face. Programmatically, this may be achieved through additional programs that provide resources for child care. Additionally, an adjustment to current benefit allotments, to include provisions for the price premium on semi-prepared healthy food products, may work to close the time gap for lower income families. The Thrifty Food Plan, utilized in all federally administered programs, does not account for time scarcity or semi-prepared healthy food premiums in the benefit allotment calculation, and assumes that preparation time has no cost to households. However, it remains unknown what the marginal cost of these products is, or what the Òtipping pointÓ in time allocation would be for individual households.  57  As local food policy councils look to address physical food access, we suggest that they also look at the temporal distance, or the time it takes households to access food markets.  This may be through public transportation updates, increased neighborhood walkability, or non-traditional vehicle ownership.  It is important for food policy councils to also incorporate this temporal distance calculation into their own assessments of the food landscape, as our model demonstrates.  There is growing understanding in the behavioral health community that perceptions of time scarcity impact the food choices people make. Though this is not directly related to food access, we believe this is important to food policy councilsÕ mission of equitable healthy living. We recommend that local food policy councils reach out to programmatic partners to identify ways to address these time perception barriersÑwhether it is through demonstrating time saving techniques in food preparation, or through time management skill development. Shifting how people perceive their current time allotment or how much time they perceive is required for healthy eating was very important to our model outcomes, and is an important area for intervention.   Many food insecure households (notably food system workers) are also hourly paid employees who experience inconsistent and varied work schedules.  Our model demonstrates how this inconsistency can impact income expectations, time stocks, and the ability of households to meaningfully plan meal schedules. We recommend that local food policy councils work with employers to identify ways to provide more consistent schedules.      58         APPENDIX     59  APPENDIX A: Model Equations  Figure 2 - 8 Graphical Equations     60  Table 1 - 2: Model Equations   61  APPENDIX B: Semi-Structured Interview Questions  Pre-Workshop Interview Guide: Food Security 1.!Why do you /your organization feel that food insecurity is an important issue in the city of Detroit? a.!What is your/your organizationÕs role in improving food security in Detroit? 2.!What are the major causes/drivers of food insecurity in Detroit?  3.!How have these drivers changed over time (if at all)?  4.!How are these drivers likely to change in the future? 5.!What strategies do you know of that have been used to improve food security in Detroit? a.!What resources do these strategies require? Who is responsible for implementing them?  b.!Have these strategies been effective? Why or why not? 6.!What strategies do you think should be used in the future to improve food security in Detroit? a.!What resources would these strategies require? Who would be responsible for implementing them?  7.!Some have advocated for urban livestock as a means of improving food security in Detroit. Do you think this is likely to be an effective strategy? Why or why not? What potential concerns should be addressed before urban livestock is permitted in Detroit? 8.!Finally, what other problems for Detroiters does food insecurity cause? (e.g. health, economic impacts).     62  APPENDIX C: IntervieweesÕ Affiliations  Table 1 - 3: Stakeholder Affiliations City Planning Commission/Detroit Food Policy Council GenesisHOPE CDC Detroit Future City Eastern Market Corp/Detroit Food Policy Council Michigan State University The Greening of Detroit Third Eye Group G Bailey Winston Enterprise Detroit Food Policy Council (3) Strategic Financial Strategies Neighborhood BUG Kearney Development Strategies Detroit Food and Fitness Collaborative/Detroit Food Policy Council   63            BIBLIOGRAPHY   64  BIBLIOGRAPHY   Article, S. (2004). Special Article Poverty and obesity&: the role of energy density and energy costs 1 , 2, 6Ð16.  Barlas, Y. (1996). Formal aspects of model validity and validation in system dynamics. System Dynamics Review, 12(3), 183Ð210. http://doi.org/10.1002/(SICI)1099-1727(199623)12:3<183::AID-SDR103>3.0.CO;2-4  Beaulac, J., Kristjansson, E., & Cummins, S. (2009). A systematic review of food deserts, 1966-2007. Preventing Chronic Disease, 6(3), A105. http://doi.org/A105 [pii]  Bellows, A. C., & Hamm, M. W. (2002). U.S.-Based Community Food Security: Influences, Practice, Debate. Journal for the Study of Food and Society, 6.1(Winter), 31Ð44. http://doi.org/10.2752/152897902786732725  Brunner, T. A., van der Horst, K., & Siegrist, M. (2010). Convenience food products. Drivers for consumption. Appetite, 55(3), 498Ð506. http://doi.org/10.1016/j.appet.2010.08.017  Campbell, C. C. (1991). Food insecurity: a nutritional outcome or a predictor variable? The Journal of Nutrition, 121(3), 408Ð415.  Carlson, A., & Fraz‰o, E. (2012). Are Healthy Foods Really More Expensive&? It Depends on How You Measure the Price, (96), 44. http://doi.org/10.2139/ssrn.2199553  Cawley, J. (2004). An economic framework for understanding physical activity and eating behaviors. American Journal of Preventive Medicine, 27(3 SUPPL.), 117Ð125. http://doi.org/10.1016/j.amepre.2004.06.012  Clifton, K. J. (2004). Mobility Strategies and Food Shopping for Low-Income Families: A Case Study. Journal of Planning Education and Research, 23(4), 402Ð413. http://doi.org/10.1177/0739456X04264919  Connors, M., Bisogni, C. A., Sobal, J., & Devine, C. M. (2001). Managing values in personal food systems. Appetite, 36(3), 189Ð200. http://doi.org/10.1006/appe.2001.0400  Daly, K. (1996). Families & time: Keeping pace in a hurried culture (7th ed.). Sagr Publications.  Food Research and Action Center. (2012). Supllemental Nutrition Assistance Program - Eligibility.  Forrester, J. W. (1994). System Dynamics, Systems Thinking, and Soft OR. System Dynamics Review, 10(2), 1Ð14. http://doi.org/10.1002/sdr.4260100211  65  Furst, T., Connors, M., Bisogni, C. A., Sobal, J., & Falk, L. W. (1996). Food Choice: A Conceptual Model of the Process. Appetite, 26(3), 247Ð265. http://doi.org/10.1006/appe.1996.0019  Gallagher, M. (2007). on Public Health in Detroit. Public Health.  Ghaffarzadegan, N., Lyneis, J., & Richardson, G. P. (2011). How small system dynamics models can help the public policy process. System Dynamics Review, 27(1), 22Ð44. http://doi.org/10.1002/sdr.442  Glaeser, E. L., Resseger, M. G., & Tobio, K. (2008). Urban Inequality. NBER Working Paper, 14419(14419), 1Ð63. http://doi.org/10.2139/ssrn.1299100  Gofton, L. (1995). Convenience and the moral status of consumer practices. In D. W. Marshall (Ed.), Food choice and the consumer (pp. 152Ð181).  Gunderston, C, A., Dewey, A., Crumbaugh, M., Kato, M., & Englehard, E. (2016). Map The Meal Gap 2016: Food Insecurity and Child Food Insecurity Estimates at the County Level.  Guthrie, J., Lin, B., & Frazao, E. (2002). Role of Food Prepared Away from Home in the American Diet, 1977-78 versus 1994-96: Changes and Consequences. Journal of Nutrition and Education and Behavoir, 34(3), 140.  Guy, C., Clarke, G., & Eyre, H. (2004). Food retail change and the growth of food deserts: a case study of Cardiff. International Journal of Retail & Distribution Management, 32(2), 72Ð88. http://doi.org/10.1108/09590550410521752  Hamm, M., & Bellows C, A. (2003). Community Food Security and Nutrition Educators. Journal of Nutrition and Education and Behavoir, 35(1).  Hendrickson, D., Smith, C., & Eikenberry, N. (2006). Fruit and vegetable access in four low-income food deserts communities in Minnesota. Agriculture and Human Values, 23(3), 371Ð383. http://doi.org/10.1007/s10460-006-9002-8  Hillier, A., Cannuscio, C. C., Karpyn, A., McLaughlin, J., Chilton, M., & Glanz, K. (2011). How Far Do Low-Income Parents Travel to Shop for Food? Empirical Evidence from Two Urban Neighborhoods. Urban Geography, 32(5), 712Ð729. http://doi.org/10.2747/0272-3638.32.5.712  Hirsch, G. B., Levine, R., & Miller, R. L. (2007). Using system dynamics modeling to understand the impact of social change initiatives. American Journal of Community Psychology, 39(3-4), 239Ð53. http://doi.org/10.1007/s10464-007-9114-3  Hovmand, P. S., Ford, D. N., Kyriakakis, S., Brown, G. W., & Brown, W. (2009). Victims arrested for domestic violence&: unintended consequences of arrest policies, 25(3), 161Ð181. http://doi.org/10.1002/sdr  Jabs, J., & Devine, C. M. (2006). Time scarcity and food choices: An overview. Appetite, 47(2), 196Ð204. http://doi.org/10.1016/j.appet.2006.02.014 66   Jabs, J., Devine, C. M., Bisogni, C. A., Farrell, T. J., Jastran, M., & Wethington, E. (2007). Trying to Find the Quickest Way: Employed MothersÕ Constructions of Time for Food. Journal of Nutrition Education and Behavior, 39(1), 18Ð25. http://doi.org/10.1016/j.jneb.2006.08.011  Jekanowski, M. D. (1999). Causes and consequences of fast food sales growth. FoodReview, 22(1), 11. Retrieved from http://dcu.summon.serialssolutions.com/link/0/eLvHCXMwY2BQSEm1tDBJMkwEZrRUC9NEk1QzS8tk05Q0Q8ski8Rk8E0NUa5mLiHmQV6WkUiluZsog4yba4izhy6sVIxPycmJNwFdbGliamAoxsCbCFr2nVcC3h6WAgBdRRt4  John Hayward (University of Glamorgan). (1961). Introduction to System Dynamics. System Dynamics, 1Ð18. http://doi.org/10.1049/ep.1968.0042  Kirkwood, C. C. (2013). System Behavior and Causal Loop Diagrams. System Dynamics Methods, (Forrester 1961), 1Ð14.  Larson, N. I., Story, M. T., & Nelson, M. C. (2009). Neighborhood Environments. Disparities in Access to Healthy Foods in the U.S. American Journal of Preventive Medicine, 36(1), 74Ð81.e10. http://doi.org/10.1016/j.amepre.2008.09.025  Maxwell, D. G. (1996). Measuring food insecurity: The frequency and severity of Òcoping strategies.Ó Food Policy, 21(3), 291Ð303. http://doi.org/10.1016/0306-9192(96)00005-X  McKenzie, B. S. (2014). Access to supermarkets among poorer neighborhoods: a comparison of time and distance measures. Urban Geography, 35(1), 133Ð151. http://doi.org/10.1080/02723638.2013.856195  Meadows, D. (2002). Dancing with Systems. The Systems Thinker, 13(2), 2Ð6.  Meadows, D. H. (2008). Thinking in Systems: A Primer. (D. Wright, Ed.) (Vol. 89). Susainability Institute. http://doi.org/10.1080/09644016.2011.589585  Olabisi, L. S. (2008). Participatory Modeling in Environmental Systems, (517).  Pothukuchi, K. (2011). The Detroit Food System Report 2009-2010, 1Ð76.  Reardon, S. F., Matthews, S., OÕSullivan, D., Lee, B. A., Firebaugh, G., Farrell, C. R., & Bischoff, K. (2008). The Geographic Scale of Metropolitan Racial Segregation. Demography, 45(3), 489Ð514. http://doi.org/10.1353/dem.0.0019  Rose, D., & Richards, R. (2004). Food store access and household fruit and vegetable use among participants in the US Food Stamp Program. Public Health Nutrition, 7(8), 1081Ð1088. http://doi.org/10.1079/PHN2004648  Schulz, A. J., Williams, D. R., Israel, B. A., & Lempert, L. B. (2002). Racial and Spatial Relations as Fundamental Determinants of Health in Detroit. Milbank Quarterly, 80(4), 677Ð707. http://doi.org/10.1111/1468-0009.00028 67   Shannon N. Zenk, Amy J. Schulz, Srimathi Kannan, Laurie L. Lachance, Graciela Mentz, & William Ridella. (2012). Neighborhood Retail Food Environment and Fruit and Vegetable Intake in a Multiethnic Urban Population. Am J Health Promot, 23(4), 255Ð264. http://doi.org/10.4278/ajhp.071204127.Neighborhood  Stave, K. A. (2002). Using system dynamics to improve public participation in environmental decisions. System Dynamics Review, 18(2), 139Ð167. http://doi.org/10.1002/sdr.237  Sterman, J. D. (1992). System Dynamics Modeling for Project Management. System, 1951, 286Ð294. http://doi.org/10.1109/SOCA.2007.45  Sterman, J. D. (2001). System Dynamics Modeling: Tools for Learning in a Complex World. California Management Review, 43(4), 8Ð25. http://doi.org/10.1111/j.1526-4637.2011.01127.x  Stewart, H., Blisard, N., Bhuyan, S., & Nayga Jr, R. M. (2004). The demand for food away from home: Full-service or fast food? Agricultural Economics Reports, 1Ð23. Retrieved from http://ideas.repec.org/p/ags/uerser/33953.html  U.S. Department of Health and Human Services and U.S. Department of Agriculture. (2015). 2015Ð2020 Dietary Guidelines for Americans. Retrieved from http://health.gov/dietaryguidelines/2015/guidelines/  Van den Belt, M. (2004). Mediated Modeling: A System Dynamcs Approach to Environmental Consensus Building. Island press.  Vasquez, V. B., Lanza, D., Hennessey-Lavery, S., Facente, S., Halpin, H. a, & Minkler, M. (2007). Addressing food security through public policy action in a community-based participatory research partnership. Health Promot Pract, 8(4), 342Ð349. http://doi.org/10.1177/1524839906298501  Walker, R. E., Keane, C. R., & Burke, J. G. (2010). Disparities and access to healthy food in the United States: A review of food deserts literature. Health and Place, 16(5), 876Ð884. http://doi.org/10.1016/j.healthplace.2010.04.013  Webb Patrick;Coates, J.&; H., & Ser, R. (2003). Measuring Food Insecuirty: Going Beyound Indicators of Income and Anthropometry, (November).  Widener, M. J., Farber, S., Neutens, T., & Horner, M. W. (2013). Using urban commuting data to calculate a spatiotemporal accessibility measure for food environment studies. Health and Place, 21, 1Ð9. http://doi.org/10.1016/j.healthplace.2013.01.004  Wiig, K., & Smith, C. (2009). The art of grocery shopping on a food stamp budget: factors influencing the food choices of low-income women as they try to make ends meet. Public Health Nutrition, 12(10), 1726Ð1734. http://doi.org/10.1017/S1368980008004102  Wrigley, N., Warm, D., & Margetts, B. (2003). Deprivation, diet, and food-retail access: Findings 68  from the Leeds Òfood desertsÓ study. Environment and Planning A, 35(1), 151Ð188. http://doi.org/10.1068/a35150  Wrigley, N., Warm, D., Margetts, B., & Whelan, A. (2002). Assessing the Impact of Improved Retail Access on Diet in a ÒFood DesertÓ: A Preliminary Report. Urban Studies, 39(11), 2061Ð2082. http://doi.org/10.1080/004209802200001136 Zenk, S. N., Schulz, A. J., Hollis-Neely, T., Campbell, R. T., Holmes, N., Watkins, G., É Odoms-Young, A. (2005). Fruit and vegetable intake in African Americans: Income and store characteristics. American Journal of Preventive Medicine, 29(1), 1Ð9. http://doi.org/10.1016/j.amepre.2005.03.002  Zenk, S. N., Schulz, A. J., Israel, B. a., James, S. a., Bao, S., & Wilson, M. L. (2005). Neighborhood racial composition, neighborhood poverty, and the spatial accessibility of supermarkets in metropolitan Detroit. American Journal of Public Health, 95(4), 660Ð667. http://doi.org/10.2105/AJPH.2004.042150