Paper presented at the 32™ International Conference of the System Dynamics Society
July 20-24, 2014, Delft, The Netherlands
Dynamic decision making in coupled social-
ecological systems
Smallholder farmers’ goals, resources and constraints in
improving food security and adapting to climate change
in Zambia
Maria Saldarriaga’’, Progress H. Nyanga’, Birgit Kopainsky"”
system Dynamics Group, Department of Geography, University of Bergen, Postbox 7800, 5020
Bergen, Norway
+ e-mail: maria.saldarriaga@geog.uib.no
* e-mail: birgit.kopainsky@geog.uib.no; phone: +47 55 58 30 92; fax: +47 55 58 3099
2 Geography and Environmental Studies Department, School of Natural Sciences, University of
Zambia, P.O. Box 32379, Lusaka, Zambia, 10101
Abstract
Climate change will lead to significant yield reductions in maize dominated farming systems in
sub Saharan Africa. Combined with a growing and more demanding population, food systems in
this region thus face the challenge of undergoing a considerable transformation in order to
meet the challenges of achieving food security and adapting to climate change. Increasing food
security and adapting to climate change is a dynamic decision making task that involves a wide
range of stakeholders such as farmers, the private sector, consumers, civil society, and policy-
makers. In this paper, we focus on the particular stakeholder group of small-scale farmers in
Zambia and collect interview data on the multiple decisions they make in the course of a year.
Our data provides a rich description of farmers’ dynamic decision making and their adaptive
capacity to deal with existing and future challenges related to food security. As people also
need an enabling institutional and policy environment to successfully adapt in the longer term
and diversify livelihoods for positive wealth accumulation, we reflect on the use of a multi-
method approach that combines our qualitative interviews with quantitative system dynamics
modeling.
Introduction
Climate change will lead to significant yield reductions in maize dominated farming systems in sub
Saharan Africa (Lobell et al., 2008). Combined with a growing and more demanding population,
food systems in this region thus face the challenge of undergoing a considerable transformation in
order to meet the challenges of achieving food security and adapting to climate change. Adaptive
practices are manifold and they can occur at multiple levels and at different scales (e.g., Below et
al., 2010; Easterling et al., 2007; Vermeulen et al., 2010).
Food systems are social-ecological systems that consist of biophysical and social factors which are
linked through feedback mechanisms (Berkes et al., 2003; Ericksen, 2008b). These mechanisms
determine the development of food system outcomes such as food security, environmental
welfare and social welfare over time. The food security situation of a given unit of analysis such as
a country can be explained in terms of three components (Ericksen, Stewart, Dixon, et al., 2010).
Food availability is the amount, type and quality of food a unit has at its disposal to consume,
either through local production, distribution, or exchange for money, labor or other items of
value. Access to food can be analyzed in terms of the affordability of food that is available, how
well allocation mechanisms such as markets and government policies work, and whether
consumers can meet their food preferences. The utilization of food refers to the ability to
consume and benefit from food. Stability is an important dimension of all these components and
describes their behavior over time.
Inter- and transdisciplinary research is generally seen as key to overcoming fundamental problems
in the analysis of social-ecological systems (Carpenter et al., 2009; Hammond & Dubé, 2012;
Ostrom, 2009). Such problems include the multiplicity of disciplines involved when assessing a
broad range of possible outcomes; the multiplicity of stakeholders with potentially conflicting
interests and differing intervention pathways; and the difficulty of modeling complex dynamics
across the multiple scales and levels of a food system. Such research therefore requires hybrid
frameworks where multiple qualitative and quantitative methodologies are applied, making use of
a combination of existing quantitative sources, case studies, and stakeholder input (Engle et al.,
2013; Ericksen et al., 2009; Janssen & Anderies, 2013). This allows system complexity to be
captured in an efficient way, enabling the identification of synergies and trade-offs, and the design
of corresponding policy and management actions, even in contexts where detailed data collection
is not feasible.
In a different stream of work, we develop a system dynamics model that aims at supporting
decision makers at different levels to effectively manage adaptation to climate change for
improving and maintaining food security in Zambia. The simulation model studies the behavior
over time especially of the availability and access dimension of food security as well as some
dimensions of social and environmental welfare, and it analyzes their reactions to different
management and policy actions. This present paper contributes to an understanding of how
stakeholders in the maize dominated farming systems in Zambia make decisions to improve food
security outcomes and adapt to climate change. For this purpose, we use qualitative research
methods such as interviews. The results from the interviews contribute to the refinement of the
simulation model and its calibration. They are also used to derive implications for the
communication and use of the simulation model. This refers to guidelines on how seemingly
effective management and policy actions might have to be reformulated, adjusted and combined
so that they are consistent with stakeholders’ conceptions of the food system and thus facilitate
adoption and diffusion of such actions.
In this paper, we focus on the particular stakeholder group of small-scale farmers. This stakeholder
group interacts with food systems on a daily basis and over long time periods. It thus possess
crucial knowledge of food system dynamics, together with associated management practices
(Berkes et al., 2000; Janssen & Anderies, 2013). Small-scale farmers account for the vast majority
of farms, cropped area, maize production, and fertilizer use in Zambia. For example, as of the
2011/12 agricultural year, small-scale farmers accounted for 99% of the farms, 94% of total
cropped area, 98% of maize area planted, 95% of maize production, 75% of total fertilizer use, and
93% of the fertilizer used on maize (figures are based on the 2011/12 Crop Forecast Survey). At
the same time, little is understood of how small-scale farmers make decisions and manage their
complex resource systems (Janssen & Anderies, 2013). By investigating small-scale farmers’
dynamic decision making, we address the following research questions:
¢ What are the decisions that small-scale farmers make every year, including their coping and
adaptive practices to increase food security and adapt to climate change? The literature
distinguishes between coping and adaptation (e.g., Ericksen, Stewart, Eriksen, et al., 2010).
Coping with challenges allows survival and protection of short-term food security or income.
However, it often wears down assets that will be needed in the future. Adaptation, on the
other hand, can be considered as modifications in behavior or strategies that enable farmers
to continue to develop in the face of change over the long run.
¢ What are the determinants of these decisions, that is, the social, environmental and economic
factors that restrain or enable these decisions?
¢ What are the outcomes of these decisions in terms of food security and also in terms of
environmental and social welfare?
Methods
The study area consisted of four agro-ecological zones in Zambia (western, central, southern,
eastern) where we conducted in-depth qualitative interviews with 20 farmers. These farmers
constitute a sub-set of 470 farmers who had participated in a large-scale monitoring and
evaluation program on conservation agriculture (Aune et al. (2012)); Nyanga and Johnsen (2010)))
and for whom extensive survey and semi-structured interview data is available.
In a previous paper (Saldarriaga et al., 2013), we described the data collection method in detail.
Based on the experiences with the pilot interviews, we adapted the procedures to focus more on
the multiple decisions that small-scale farmers make in the course of a year. For this purpose, we
designed and worked with a food security wheel that showed the individual months of a year and
into which all information about farming decisions as well as coping and adaptive practices were
entered (Figure 1).
Interviews were held in local languages and subsequently translated and transcribed. Data analysis
involved the analysis of the interviews in fine-grained fashion to identify and track categories of
knowledge (Parnafes et al., 2008) and to develop theories of such knowledge (Cobb et al., 2003;
diSessa & Cobb, 2004).
Figure 1: Chart facilitating the elicitation of farming decisions, coping and adaptive practices (left
hand side of the figure: pre-interview; right hand side of the figure: example post-interview)
e0l3/le/08 11:33
Results
In the presentation of results, we start with the last research question, which was about the
outcomes of farmers’ decisions in terms of food security. We then explore the decisions that lead
to these outcomes and differentiate between decisions regarding farming activities and decisions
representing coping and adaptive practices. In the discussion section, we turn to the remaining
research question, which is about the determinants of the observed decisions.
Food system outcomes: Behavior patterns of food security throughout the year
In Zambia, there are three main seasons:
¢ The rain season (mainza), which in the analyzed farming year started in the middle of
November 2012 and ended in the middle of March 2013. Climate change causes rains to start
later in the year. In the last two decades, it has shifted the onset of rains from mid October to
end of November or even mid December (Neubert et al., 2011).
¢ The cold season (mupeyo), which in the analyzed farming year started in the middle of March
2013 and ended in August 2013.
¢ The rest of the year corresponds to the hot season (chillimo).
Food security of farming households’ varies considerably throughout the year (Figure 2). As soon
as the harvest starts in March and at the latest in April, farmers have enough to eat. This is the
period during which farmers are most food secure. The duration of this period, however, depends
on the rain patterns during the previous months. From August to October, some farmers still
affirm to have enough while others have to begin cutting down their consumption. As of October,
farmers need to cut down on their consumption while waiting for the harvest to be ready. For
some farmers, March is already a month where they have sufficient food. Others face a critical
period with barely enough food during February and March. Among other factors, this is explained
by a small harvest in the previous year and a late harvest in the current year. A delay in the onset
of rain aggravates and prolongs this period of scarcity.
Figure 2: Food security throughout the year
MUPEYO
cold S€250p) ~€
or
So far, we explored food security in terms of the amount consumed by a household. In system
dynamics terminology, we explained the pattern of food security throughout the year based on
the changes in the stock of available food, particularly maize. However, farmers also feel food
insecure if the stock of maize is sufficient for the household but if the variety of the available food
is poor. In a next step, we thus analyze the types of foods available to farmers throughout the
year. For this purpose, we split farmers’ diets into four categories. The order of the categories
does not represent the importance of each category in farmers’ diets.
A. Maize based foods: chibwantu (fermented maize drink), machebele (dried maize boiled),
magwaza (dried cooked maize), musozya (maize samp) and nshima (maize pulp).
B. Vegetables and legumes from farmers’ fields: e.g., beans, ground nuts, okra and sweet
potatoes.
C. Wild vegetables and fruits: e.g., blackjack leaves, ibonwe and lumanda (types of wild spinach),
lusala (traditional root), matobo (wild fruit).
D. Meet and dairy: beef, chicken, fish, eggs and milk.
Although our data does not allow us to evaluate the importance of these categories in farmers’
diets, there are important patterns that we can highlight. Figure 3 shows the food categories
available to farmers in each of the three seasons.
Figure 3: Food categories by season (upper half of the figure: detailed overview; lower half: food
categories by season in the food security wheel)
and legumes from fields es 7
a %s
sesate®
The highest variety of foods comes from farmers’ fields and, as expected, they are consumed more
during the cold season (mupeyo). This season is also the one where farmers are the most food
secure. A high variety of vegetables and legumes from the fields are also consumed towards the
end of the rain season. However, the availability of the foods found in this category depends on
the conditions of the rain season. Under appropriate rain conditions, cowpeas my be planted in
November and harvested in January, mushrooms may be harvested in November and groundnuts
may be planted between December and January and harvested in March. While the season for the
main maize harvest arrives (usually in May), these early harvest foods improve farmers’ food
security after the hot season where food consumption is cut down.
5
While maize based foods are less diverse than the other foods coming from the fields, they are all
eaten throughout the year. Some local varieties of maize can mature as early as January. This fresh
maize helps mitigate food insecurity while the main harvest season arrives.
Although most farmers keep chicken and a few of them heard cattle, meet and dairy foods are
rarely consumed. These animals represent mainly an insurance against family difficulties (e.g.,
illness, school fees) or special circumstances. Those that do not keep cattle sometimes exchange
maize for beef, and chicken for milk.
Wild fruit and vegetables complement farmers diet most year around. Although less diverse than
the foods from the fields, wild foods are regarded of great importance. Wild vegetables and fruit
can be found within the villages and children are sometimes in charge of collecting them.
Finally, we put dried foods into a separate category. Figure 3 shows that most of these foods are
consumed during the hot season (chillimo). Some of the foods that are eaten fresh during the cold
season (mupeyo) are dried and stored for later consumption. During the hot season (chillimo)
foods such as groundnuts, okra and pumpkin leaves are only available dried. Other foods such as
squash are difficult to keep for later consumption.
Food system activities: Decisions about farming activities as well as coping and adaptive
practices throughout the year
Figure 4 shows the farming activities that are performed throughout the year. To obtain this data,
we asked farmers to describe the activities realized during the previous farming season 2012 until
the day of the interview in December 2013. Farmers may do things differently from one year to
the other. However, the figure shows the concentration of particular activities during specific
periods of the year. By adding these farming activities as well as coping and adaptive practices to
the food security wheel, we can observe how the activities relate to the seasons and to farmers'
food security throughout the year.
Figure 4: Food security wheel with farmers’ decisions throughout the year
yeot potRtORS, PUMPKns,
ow CIN gs
rane ten
Crops are planted between November and January and harvested between January and July. The
main period for maize harvest starts in May, however, a local variety of maize called Tanda Nzala
matures over only one month and can be harvested as early as January. As shown earlier in the
analysis, this local maize variety together with other early harvest crops such as cow peas, help
mitigate food insecurity in the critical moths before the main harvest.
Farmers implement coping and adaptive practices to mitigate food insecurity. Coping and adaptive
practices often involve but are not limited to the utilization of food and production resources. An
example is drying vegetables and fish so that they can be stored until seasons of food scarcity.
From farmers’ descriptions of all food related activities that they perform throughout the year, we
found 11 coping and adaptive practices that farmers use to mitigate food insecurity. Table 1 lists
these practices and the respective seasons when they are mostly implemented. These practices
represent farmers' responses to past, present or potential future changes perceived in the food
available to the household. Therefore, the time of year when these practices are used is in most
cases associated to the level of food security in the household at that particular point. For
instance, farmers exchange food mostly during the cold season (mupeyo) when food is most
abundant, and during the hot season (chillimo) when milk, meet and fish are most available (see
Figure 3).
Table 1: Coping and adaptive practices
Season
Practice | chitimo | mupeyo | “3 | Description
a
(Hot) (cold) Rain)
Choosin
g seeds During the hot and the beginning of the rain season, farmers are preparing
and x * their land and acquiring all the inputs necessary for growing their crops.
crops
A critical decision during the farming season is to decide when to plant the
crops. If the seeds are planted before the rain, the rain will find the seeds in
the ground when it arrives, thus increasing the chances of germination.
: However, if the rains do not come on time, the crops may fail, increasing the
etigBsib chances of food insecurity. If, in contrast, the seeds are planted only after the
i x rain has arrived, the chances of crop failure may decrease but the rains may
planing be so strong that farmers may be unable to do the planting work.
time To predict the onset of the rain farmers use two heuristics:
1) assuming that the rains will begin at a similar time than in the previous year
2) using traditional knowledge of nature such as: the position of the moon in
the sky, the yield of wild fruit trees, and the direction of the wind
Borrowi Some farmers have access to credit, which they acquire to pay for farming
ng x inputs. This is done usually around August (end of cold season and beginning
money of hot season).
Piecework is a common source of income among small-scale farmers.
, Piecework consists mainly of working for other small-scale farmers doing
boing weeding (during the rain season) and harvesting (during the cold season).
piece * * * During the hot season, when the farmers begin preparing the land, it is also
work common to hire ox-carts. Children are sometimes in charge of doing this
work.
Some farmers take part time jobs to bring more income to the household.
Taking Women also play an important role in bringing extra income home. In some
other ¥ x %
: cases, they set up small shops in front of their houses where they re-sell
tabs products like soap and other utilities.
Two main categories can be distinguished in selling food:
1) farmers sell their crops after the main harvest in the cold season. Some
farmers, however, produce only enough for their own household
consumption; and
Selling
‘ood x x 2) farmers sell from their own reserves, i.e., from what has been stored for
household consumption. This is usually a critical practice used in cases of
illness or other emergencies. However, sometimes, even when maize stocks
are low, farmers exchange maize with the purpose of bringing variety to their
diets,
Season
Practice | chitimo | mupeyo | Mi? | Description
a
(Hot) (cold) ‘Rain
Exchanging food is perhaps the most common way of increasing food variety
Sechansl | x x in farmers' diets. After the main harvest, maize is usually exchanged for meat,
aefoed wild vegetables, fish, cabbage, tomatoes and onions and milk.
Food rationing is usually done after a sudden decrease in the food stored for
_ household consumption. For instance, after selling a part of their food stock
Rationin | x to cover a family emergency, farmers may need to reduce the per capita
asond consumption in the household. This helps stretch the food stock until the next,
harvest season.
Dried vegetables are commonly eaten during the hot season where they are
Drying not available fresh. Drying food seems to be the most common way of storing
food * * food for later consumption. Refrigeration was not available to any of the 20
households in our sample.
Farmers may buy food when money is available to acquire rarely consumed
; foods such as meat. However, in some cases, buying food is an emergency
euying x measure used when the food stock of the household has been completely
food depleted. This happens more commonly during the rain season where food
security is critical for some farmers.
Some farmers keep cattle, chicken and goat as insurance against family
ells x emergencies and for school fees payment. Sometimes however, animals are
animals sold to mitigate food insecurity.
The coping and adaptive practices listed in Table 1 can be categorized in three groups based on
how they act to mitigate farmers’ food insecurity:
¢ Food variety and duration: The first category involves practices through which farmers use the
food already available to the household to increase diet variety or extend the duration and
consumption life of the food stock. This includes exchanging, drying, rationing and buying
food. Practices that focus on mitigating food insecurity by increasing food variety and duration
are most common during the hot season (chillimo). This is indeed the time of year when food
stocks are getting low and farmers need to find ways to make this stock last until the next
harvest.
¢ Income available: The second category involves practices that increase the income available to
farmers, thus increasing the likelihood of food availability. This group includes selling animals
and food, taking other jobs and doing piecework, and borrowing money. Practices aimed at
increasing the income available to the household for food are most used during the rain
(mainza) and the cold (mupeyo) seasons. During these seasons, doing weeding and harvest
piecework can bring extra income to the household.
¢ Certainty and resilience. The last category of practices increases the likelihood of food
availability by decreasing the uncertainty caused by the change in rainfall patterns and by
increasing farmers' resilience to change. Here we find choosing seeds, crops and choosing the
planting time. By the end of the cold season and the beginning of the rain season, before and
after the onset of the rains, farmers need to make important decisions about the crops and
maize seeds to be used and the appropriate time for planting these seeds. Furthermore,
although we focused on farmers' decisions about these three variables, many other choices
need to be made during the entire farming season (e.g., tillage (e.g., basins, ripping, plowing),
weeding (hand, herbicide), harvest (direct, stacking)).
Discussion and conclusions
Social-ecological systems such as maize dominated farming systems in Zambia are complex and
dynamic systems. They thus require stakeholders to continuously test, learn about, and develop
knowledge and understanding in order to cope with and adapt to change and uncertainty
(Carpenter & Gunderson, 2001; Ericksen, 2008a; Folke et al., 2005). Increasing food security and
adapting to climate change is a dynamic decision making task that involves a wide range of
stakeholders such as farmers, the private sector, consumers, civil society, and policy-makers. In
this paper, we focused on the particular stakeholder group of small-scale farmers and provided a
fine-grained description of the multiple decisions they need to make in the course of a year, the
determinants of these decisions and their outcomes that eventually feed back into subsequent
decisions.
Our results show that small-scale farmers perceive the dynamic complexity of the resource
systems they manage very well. They provide operational descriptions of stock management
problems such as the management of soil fertility or of food inventories and they try to make
decisions that rebuild and maintain safe stock levels. They also consider the main delays in the
system in their decisions. Persistently low levels of food security are mainly caused by farmers’
extremely low access to various assets and resources such as land, water, nutrients and seeds or
to alternative employment opportunities that would allow them to purchase food necessary for a
nutritious, balanced and culturally valuable nutrition.
The finding that persistent food security problems are caused only in lesser part by the
mismanagement of natural resources and more by demand-side factors (policy-led or induced by
demographic growth) are in line with results from other regions in sub-Saharan Africa (e.g.,
Mortimore & Harris, 2005). Our current data does not allow us to conclusively evaluate the role of
supply-side constraints such as scarcities of cultivable land, soil amendments, labor or capital. This
is illustrated in Table 2 that compares coping and adaptive practices suggested by the literature
(e.g., Ericksen, et al., 2010; Mendelsohn, 2008; Vermeulen et al., 2012) with the coping and
adaptive practices found in our interviews. The table shows that most practices that require
substantial investments such as capital- or water-related practices are unthinkable for the
interviewed farmers.
Table 2: Findings on small-scale farmers’ dynamic decision making
Decision Suggested by literature Found in interviews
Production input: land Expansion of agriculture land No
Reduced deforestation No
Reforestation No
Production input: capital Increasing mechanization (for adaptation) No
Decreasing mechanization (for mitigation) No
Production input: water Increasing irrigation No
Water storage and management No
Production input: seeds Use of improved seed varieties Yes
10
Production input: mineral fertilizer Increasing use of fertilizer (for adaptation) Yes
Decreasing use of fertilizer (for mitigation) No
Production management Match varieties to climate Yes
Experiment with alternative planting dates Yes
Experiment with alternative crops Yes
Move from growing crops to livestock and back Yes
Soil conservation Yes
Pest and disease management Marginal
Agroforestry Marginal
Production output Food storage No
Beyond production Exchange of food Yes
Distribution of food Yes
Laboring for food Yes
Laboring for cash Yes
Consumption of unusual food, e.g., wild plants Yes
Our data provides a rich description of the existing adaptive capacity of small-scale farmers to deal
with existing and future challenges related to increasing food security and adapting to climate
change. It is, however, increasingly recognized that people also need an enabling institutional and
policy environment to successfully adapt in the longer term and diversify livelihoods for positive
wealth accumulation (Adger et al., 2007; Ellis & Freeman, 2004; Ericksen, et al., 2010). To support
this process, we thus accompany our research in Zambia with quantitative system dynamics
modeling. The application of multiple methods to the rich range of challenges for maize
dominated farming systems in Zambia allows developing the basic principles that govern the
dynamics of social-ecological systems and their capacity to cope with and adapt to change
(Janssen & Anderies, 2013). The simulation model enables us to systematically explore the
consequences of management and policy actions on food system outcomes and, by doing so,
provides the basis for an enabling institutional and policy environment. Insights from case studies
and qualitative research, in turn, stimulate modeling. They also provide evidence of real
achievements and internal potentials and therefore point the way toward a foundation for
evidence-led policies for the management of social-ecological systems. Such evidence-led policies
aim to build on local experience, suggesting a more organic model for development rather than
aiming to transform seemingly inappropriate local practices (Mortimore & Harris, 2005).
Further qualitative research in this domain will thus expand the current focus on small-scale
farmers to other stakeholder groups and it will move from the analysis of current knowledge and
decision making to exchange and integration of knowledge held by different stakeholders. For
instance, the data described in this paper suggest that predictability and uncertainty regarding
economic and climatic context factors play an important role in small-scale farmers’ decision-
making. A next step thus needs to investigate how science-based knowledge held e.g., by
researchers on these subjects can integrate with farmers’ knowledge and how farmers will use it
for decision-making. This line of research will generate insights into the factors that need to be
considered in the implementation of seemingly effective management and policy actions and into
how these actions might have to be reformulated, adjusted and combined to be truly effective.
11
Acknowledgements
Work on this paper was supported by the Norwegian Research Council through the project
“Simulation based tools for linking knowledge with action to improve and maintain food security
in Africa” (contract number 217931/F10). The views and conclusions expressed in this paper are
those of the authors alone and do not necessarily reflect the views of the Norwegian Research
Council.
References
Adger, W. N., Agrawala, M. M. Q., Conde, C., O'Brien, K., Pulhin, J., Pulwarty, R., .. . Takahashi, K.
(2007). Assessment of adaptation practices, options, constraints and capacity. In M. L.
Parry, O. F. Canziani, J. P. Palutikof, P. J. van der Linden & C. E. Hanson (Eds.), Climate
Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to
the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 717-
743). Cambridge, UK: Cambridge University Press.
Aune, J. B., Nyanga, P. H., & Johnsen, F. H. (2012). A Monitoring and Evaluation Report of the
Conservation Agriculture Project 1 (CAP1) in Zambia: Noragric.
Below, T., Artner, A., Siebert, R., & Sieber, S. (2010). Micro-level Practices to Adapt to Climate
Change for African Small-scale Farmers. A Review of Selected Literature /FPR/ Discussion
Paper (pp. 21). Washington, DC: International Food Policy Research Institute.
Berkes, F., Colding, J., & Folke, C. (2000). Rediscovery of Traditional Ecological Knowledge as
Adaptive Management. Ecological Applications, 10(5), 1251-1262. doi: 10.2307/2641280
Berkes, F., Colding, J., & Folke, C. (2003). Navigating Social-Ecological Systems. Building Resilience
for Complexity and Change. New York, NY: Cambridge University Press.
Carpenter, S. R., & Gunderson, L. H. (2001). Coping with collapse: Ecological and social dynamics in
ecosystem management. BioScience, 51(6), 451-457. doi: 10.1641/0006-
3568(2001)051[0451:cwceas]2.0.co;2
Carpenter, S. R., Mooney, H. A., Agard, J., Capistrano, D., DeFries, R. S., Diaz, S., .. . Whyte, A.
(2009). Science for managing ecosystem services: Beyond the Millennium Ecosystem
Assessment. Proceedings of the National Academy of Sciences of the United States of
America, 106(5), 1305-1312. doi: 10.1073/pnas.0808772106
Cobb, P., Confrey, J., diSessa, A., Lehrer, R., & Schauble, L. (2003). Design Experiments in
Educational Research. Educational Researcher, 32(1), 9-13.
diSessa, A. A., & Cobb, P. (2004). Ontological Innovation and the Role of Theory in Design
Experiments. Journal of the Learning Sciences, 13(1), 77 - 103.
Easterling, W. E., Aggarwal, P. K., Batima, P., Brander, K. M., Erda, L., Howden, S. M., .. . Tubiello,
F. N. (2007). Food, fibre and forest products. In M. L. Parry, O. F. Canziani, J. P. Palutikof, P.
J. van der Linden & C. E. Hanson (Eds.), Climate Change 2007: Impacts, Adaptation and
Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change (pp. 273-313). Cambridge, UK: Cambridge
University Press.
12
Ellis, F., & Freeman, H. A. (2004). Rural livelihoods and poverty reduction strategies in four African
countries. The Journal of Development Studies, 40(4), 1-30. doi:
10.1080/00220380410001673175
Engle, N., Bremond, A., Malone, E., & Moss, R. (2013). Towards a resilience indicator framework
for making climate-change adaptation decisions. Mitigation and Adaptation Strategies for
Global Change, 1-18. doi: 10.1007/s11027-013-9475-x
Ericksen, P. J. (2008a). Conceptualizing food systems for global environmental change research.
Global Environmental Change, 18(1), 234-245.
Ericksen, P. J. (2008b). What is the vulnerability of a food system to global environmental change?
Ecology and Society, 13(2 C7 - 14).
Ericksen, P. J., Ingram, J. S. I., & Liverman, D. M. (2009). Food security and global environmental
change: emerging challenges. Environmental Science & Policy, 12(4), 373-377. doi:
http://dx.doi.org/10.1016/j.envsci.2009.04.007
Ericksen, P. J., Stewart, B., Dixon, J., Barling, D., Loring, P., Anderson, M., & Ingram, J. S. I. (2010).
The Value of a Food System Approach. In J. S. I. Ingram, P. J. Ericksen & D. Liverman (Eds.),
Food Security and Global Environmental Change. London, Washington DC: Earthscan.
Ericksen, P. J., Stewart, B., Eriksen, S., Tschakert, P., Sabates-Wheeler, R., Hansen, J., & Thornton,
P. K. (2010). Adapting Food Systems. In J. S. 1. Ingram, P. J. Ericksen & D. Liverman (Eds.),
Food Security and Global Environmental Change (pp. 115-143). London & Washington DC:
Earthscan.
Folke, C., Hahn, T., Olsson, P., & Norberg, J. (2005). Adaptive governance of social-ecological
systems. Annual Review of Environment and Resources, 30(1), 441-473. doi:
doi:10.1146/annurev.energy.30.050504.144511
Hammond, R. A., & Dubé, L. (2012). A systems science perspective and transdisciplinary models for
food and nutrition security. Proceedings of the National Academy of Sciences, 109(31),
12356-12363. doi: 10.1073/pnas.0913003109
Janssen, M. A., & Anderies, J. M. (2013). A multi-method approach to study robustness of social—
ecological systems: the case of small-scale irrigation systems. Journal of Institutional
Economics, 9(04), 427-447. doi: doi:10.1017/S1744137413000180
Lobell, D. B., Burke, M. B., Tebaldi, C., Mastrandrea, M. D., Falcon, W. P., & Naylor, R. L. (2008).
Prioritizing climate change adaptation needs for food security in 2030. Science, 319(1), 607-
610.
Mendelsohn, R. (2008). The impact of climate change on agriculture in developing countries.
Journal of Natural Resources Policy Research, 1(1), 5-19. doi: 10.1080/19390450802495882
Mortimore, M., & Harris, F. (2005). Do small farmers’ achievements contradict the nutrient
depletion scenarios for Africa? Land Use Policy, 22(1), 43-56. doi:
http://dx.doi.org/10.1016/j.landusepol.2003.06.003
Neubert, S., Komm, M., Krumsiek, A., Schulte, A., Tatge, N., & Zeppenfeld, L. (2011). Agricultural
development in a changing climate in Zambia. Increasing resilience to climate change and
economic shocks in crop production (Vol. 57). Bonn: Deutsches Institut fur
Entwicklungspolitik.
Nyanga, P. H., & Johnsen, F. H. (2010). 2009/2010 Monitoring and Evaluation Draft Report:
Noragric.
13
Ostrom, E. (2009). A general framework for analyzing sustainability of social-ecological systems.
Science, 325(5939), 419-422. doi: 10.1126/science.1172133
Parnafes, O., Hammer, D., Louca, L., Sherin, B., Lee, V., Krakowski, M., .. . Edelson, D. (2008). How
to study learning processes? reflection on methods for fine-grain data analysis:
International Society of the Learning Sciences.
Saldarriaga, M., Kopainsky, B., & Alessi, S. M. (2013). Knowledge analysis in coupled social-
ecological systems. What do stakeholders in sub Saharan Africa know about the dynamic
complexity of climate change, agriculture and food security? . Paper presented at the 31st
International Conference of the System Dynamics Society, Cambridge, MA.
Vermeulen, S. J., Aggarwal, P. K., Ainslie, A., Angelone, C., Campbell, B. M., Challinor, A. J.,.. .
Wollenberg, E. (2010). Agriculture, food security and climate change: Outlook for
knowledge, tools and action CCAFS Report (pp. 16). Copenhagen: CGIAR-ESSP Program on
Climate Change, Agriculture and Food Security.
Vermeulen, S. J., Campbell, B. M., & Ingram, J. S. I. (2012). Climate change and food systems.
Annual Review of Environment and Resources, 37(1), 195-222. doi: doi:10.1146/annurev-
environ-020411-130608
14
VY. Output Behavior Analysis
In simulation experiments, time horizon is taken as 1250 days, which is approximately
3.4 years. Time period is one day. For all of these experiments, in first one year or two
years model adjust itself due to its initial values, then shows steady-state behavior.
However, it is not likely for an economy to last for three years in such crisis. Some
political change will eventually occur during such catastrophic crisis.
First part of the output behavior analysis is deterministic. The reference buying and
selling values of individuals and big investors are constant. In the second part,
experiments have a stochastic behavior. Buying/selling references and price of dollars
have normal distribution with mean is equal to reference values and standard deviation
is equal to 30% of the mean value.
When first scenario analyses were done, there were significant transient behaviors at the
beginning of each run. To avoid this, the initial values of money stocks (Dollars at
Individuals, Dollars in the Market, Dollars at Big Investors) are set at some point near
their equilibrium values for base run.
a. Scenario Analysis of Deterministic Models
In this part six different scenario analyses are conducted. The first scenario can be
considered as the base run of the model. In the first three experiments, there are not any
manipulators in the market. The second three experiments involve manipulators in the
market. For both cases (manipulators do not exist and do exist), first market behavior is
regular (non-speculative), then it is speculative, and finally an interest adjustment policy
is applied to speculative market conditions. In goal-seeking case, big investors and
individuals have the same buying function. In manipulative case, big investors have a
steeper buying function with respect to exchange rate changes. They can buy up to ten
times of their reference buying value.
i. Sensible Individuals and Sensible Big Investors
In this case, manipulation and speculation behaviors are excluded, as it is the base run
of the model. Market behavior is assumed to be regular. Individuals and big investors
buy US dollars when the price is low, and sell when the price is high. There is
borderline equilibrium in the market. As it is shown in Figure 17, there are only minor
oscillations in the market, which is consistent with real life.
bad z “Dalasi
4
ff
P\ J
i]
F oy
ie of
72500
Pace Days 1005 PH Wee Mer 72,2017
Pesca ones Socks
Figure 17: Model dj ics with Sensible Individuals and Sensible Big Ii
ii. Speculative Individuals and Sensible Big Investors
Even though, there are any not manipulators in the market, individuals speculate when
the rate of change of exchange rate is high, then they start to buy more dollars from the
market and with reinforcing loop, the exchange rate increases even more. Individuals
continue to buy despite the increasing exchange rate as they anticipate that price of a
dollar will rise even more in the future. On the other hand, big investors are not
assumed to show manipulative behavior here — they buy only when the exchange rate is
low.
Fo camtreroro T oeiacaSigmesars a Gaiam Dearth
rey 1
poe 402600 vied Mar 22 2017
aaa So Tara
ae
“ :
-
es
os : . ‘om a i
igs vere
Figure 18. & 18.b: Model dynamics with Speculative Individuals and Sensible Big I
with different time horizons
In Figure 18.a, it seems that an economic crisis occurs during the first half-year and
speculative cycles are observed to continue in the following years. However, after an
economic crisis in such magnitude, the economy of a country might collapse and all the
dynamics might change in real life. In Figure 18.b shows crisis dynamics in more detail.
iii. Speculative Individuals and Sensible Big Investors + Control of Interest Rate
by Central Bank
The preceding model is extended by adding an interest rate adjustment policy. As it can
be seen in Figure 19, at the beginning there is a sharp increase in exchange rate,
however interest rate adjustment brings economy to an equilibrium, and crisis is
prevented before it starts.
PP + Dotaexchanoenate 2 DabisABionestos Dorsal 1 Dols
A 2
FE oerat0
a =
\ fom a
F yy (pix: y
\ / WANT ANNO AW
1 F Sy x
Easel eZ ~~
i Saeco | UK
Mn
i
\ :
[|
E ae 7 | S|
0 rey e200 esr50
pase Days 1029PM Wea Mar22, 2017
= Noney Sits
Figure 19: Model dj ics with Speculative Individuals and Sensible Big Investors with
Central Bank intervention
Pr comecarerrae Zwieestae Sop
| er
f ones
: a0
i
oe
LX
b i w wy o
00 3250 00 ra
002 ays 10.29M Wes, war 22,2017
erst Rate
Figure 20: US Dollar Exch Rate dj ics depending on Interest Rate adjustment
The interest rate adjustment is so sharp in this scenario analysis, which is hard to apply
in real life. On the other hand, same interest rate policy can be applied in a smoother
form in real life. For example, increasing interest rate to 9.06% at the beginning, then
decreasing it gradually to its base level in one year is a possible scenario application.
iv. Sensible Individuals and Manipulative Big Investors
In this setting, even though big investors buy US Dollars in large quantities, since
individuals are assumed to be sensible, crisis does not occur in this market. Individuals
respond big investors move by buying less dollars.
[F DotarEschanaaRate 7 DolaABarNesos 7 Dolev Tolan
[> 2elttt
is abestt
q 89
le ¢
F 4
|:
00 3de0 wea es0 “2500
pase + Days 1032 PM Wied Ma 22,2017
base 2 Money Stocks
Figure 21: Model dj ics with Sensible Individuals and Manip ive Big Investors
y. Speculative Individuals and Manipulative Big Investors
In this case, both speculation and manipulation effects are included to the model. Big
investors manipulate the market by playing with huge number of US Dollars with
intention of making profit, meanwhile individuals panic and respond to market by
keeping US Dollars even tough price of a dollar is high. This run reflects the real life
economic market dynamics of crisis situations.
a ere “Dolrsinareet
\/ V :
V4
r
2
rs
1034 PM Wed, Mar22, 2017
Figure 22: Model dj ics with Sp lative Individuals and
Big
vi. Speculative Individuals and Manipulative Big Investors + Control of Interest
Rate by Central Bank
An interest rate adjustment policy is added to the preceding model. There is an
increasing exchange rate at the beginning, but interest rate adjustment prevents the crisis
before it happens. As it can be seen in Figure 23.b in detail, there is a sharp increase in
exchange rate, however interest rate adjustment policy brings exchange rate to
equilibrium.
Pa a “Dears
Re it J ' A
= a 5 a
a 1s Tage ee
: peers
pase £ i
ae a ea
; se
‘a
Bee AX
00 350 7500 1250
fase 1 Days 1025 PM Wed Mar 22,2017
fe Nene Stooks
Figure 23.a&23.b: Sp lative Individuals and Manip ive Big Investors with Central
Bank intervention (with different time horizons)
20
P+: ootarzxcnangerate 2 meresrate
| 3
I ‘a0
350
| 7
00 31280 a0 oaT50 "25001
foe 2 ays 1035 8 Wed Mar22 201
terest Rate
Figure 24: US Dollar Exch Rate dj ics depending on Interest Rate adjustment
As it is seen in Figures 19, 20, 23.a, 23.b and 24, by adjusting interest rate in
accordance with people’s expectations, keeping the exchange rate at a low and stable
level is possible.
b. Scenario Analysis of the Stochastic Versions of Models
In stochastic models, the buying/selling reference values of individuals and big
investors and dollar price have normal distribution. These references represent if there is
not any effect of exchange rate, inflation and interest rate on buying/selling behaviors.
The players in the market will buy/sell these US dollar quantities in daily basis. In
stochastic setting, noise is observed. Although the main pattern is conserved in the first
two settings, noise affects model behavior considerably in the third setting.
vii. | Sensible Individuals and Manipulative Big Investors
i Fi 7 7 Dotan
age t
Figure 25: Model dynamics with Se and i ive Big stors in
Stochastic Setting
21
viii. | Speculative Individuals and Manipulative Big Investors
Fiemniees soaagean sauamnaar vasa
FBR
: SoH ‘ i \
/
\ ‘
E pee
: i
| !
00 ooo 200 Fa
faoe ars 4042PM Wed, Mar 22,2017
Used None Socks
Figure 26: Model dj ics with Sp lative Individuals and Manip ive Big Investors in
Stochastic Setting
ix. Speculative Individuals and Manipulative Big Investors + Adjustment of
Interest Rate
Dalat
Figure 27: Speculative Individuals and Mt
ip Big Investors with Central Bank
intervention in Stochastic Setting
22
7 DatarExchargefale eae
| 3
sooo
+
x
=
= es
=
se
===
‘000 ‘1250 2500 ost 50 12600
Figure 28: US Dollar Exchange Rate dynamics depending on Interest Rate adjustment in
Stochastic Setting
Although interest rate policy lowers the exchange rate, unstable interest rate adjustment
dynamics do not seem to be realistic. Real life implementation of this policy can be
done by setting constant interest rates equal to moving averages of these rate
fluctuations.
VI. Conclusion and Future Work
Motivated by the volatile US Dollar exchange rate issue in Turkey, the impact of
speculation and manipulation on the dynamics of foreign exchange rate is examined in
this paper. In this context, this study attempts to analyze the relations among inflation,
interest rate, exchange rate and the monetary market supply-demand by using system
dynamics approach. This approach is adopted in order to better address the real-life
feedback interactions, and causal mechanisms behind the unstable foreign exchange rate
dynamics in Turkey.
A case study of Turkish exchange rate market is chosen; however the proposed model is
applicable/adaptable to other developing country economies, once the relevant
parameters are estimated.
As a base run, a goal-seeking market is modeled. The structural validity of the model is
thoroughly tested. Then, the effects of speculation and panic behavior among people,
the existence of manipulative investors and interest rate adjustment intervention by
Central Bank are incorporated by extending the base model for each scenario.
The most important conclusion from simulations is that the existence of
speculation/panic among individuals in the market is even more harmful than the
existence of big manipulators in causing possible exchange rate crises. Economic
institutions must be aware of this fact and lead the public perception and economic
behaviors with this principle. A policy that focuses on preventing speculation among
individuals is required in order to have a stable foreign exchange rate. As it is shown in
the interest rate adjustment scenarios, once the people’s expectations are satisfied, the
exchange rate variables come to equilibrium eventually.
23
This study focuses on the interacting effects of foreign exchange rate, interest rate and
inflation on the foreign exchange rate market, and circulation of foreign currency
among the different players in an economy. Even though the exchange rate and interest
rate are part of feedback loops, the inflation is not modeled as an endogenous model
structure. In future research, the monetary policies of central bank and the interactions
between these monetary policies and dollar market and inflation can be modeled
endogenously and examined.
VII. References
Al-Monitor. 2016, November 30. How the Turkish lira entered free fall. Retrieved
March 18, 2017, from: http://www.al-monitor.com/pulse/originals/2016/11/turkey-how-
government-contain-the-economic-crisis-blaze.html#ixzz4UGGBd3XN
Arslan, R. January 2014. Dolarin yiikselmesi hangi riskleri barindiriyor? BBC Turkce.
Retrieved from: http://www.bbc.com/turkce/ekonomi/2014/01/140124 dolar turk lirasi
Benmaran, M.L., Saaedi, A. 2014. Identification of speculative bubbles in Tehran stock
exchange, a system dynamics approach. Indian J.Sci.Res. 5 (1): 271-283, 2014
Bigpara.com.(n.d.). Dolar Kuru Bu yilki Dolar Lira Ahs Satis Kurlari igin Bigpara.
Retrieved March 18, 2017, from: http://www.bigpara.com/doviz/dolar/buyil/
Bloomberg HT. 2016, October 20". Merkez Bankast Rezervleri Arth. Retrieved from:
http://www.bloomberght.com/haberler/haber/1933 175-merkez-bankasi-rezervleri-artti
Capital.com. 2001, August 1° .Yasuk Alt Ekonomisi. Retrieved from:
http://www.capital.com.tr/ekonomi/makro-ekonomi/yastik-alti-ekonomisi-haberdetay-
2654
Dwenger N., Pavlov O. 2008. Feedback analysis of speculation in a foreign currency
market. Proceedings of International Conference of System Dynamics Society,
Retrieved from:
http://www.systemdynamics.org/conferences/2008/proceed/papers/PAVLO441.pdf
Egilmez, M. 2017, Jan 31". Faiz ve Kur Iliskisi. Retrieved from:
http:/Awww.mahfiegilmez.com/2017/01/faiz-ve-kur-iliskisi.html?m=1
European Central Bank. (n.d.). What is inflation? Retrieved March 18, 2017, from:
https://www.ecb.europa.eu/ecb/educational/hicp/html/index.en.html
Finansgiindem.com. 2016, December 3". Vatandasin 150 Milyar Dolart Var. Retrieved
from:http://www.finansgundem.com/haber/vatandasin-150-milyar-dolari-var/ 1143654
Furman, J. and Stiglitz J. E. 1998. “Economic Crises: Evidence and Insights From East
Asia” Brooking Papers on Economic Activity, No. 2, Brooking Institution, Washington
D.C.
Fxcm.com. 2016, March 23", Impact Of Inflation And Interest Rates On Exchange Rate
Trends. Retrieved from: https://www.fxcm.com/insights/exchange-rate-trends/
Internethaber.com. (n.d.). Dolar kuru 3.1140 lira olan tarihi zirvesini zorluyor!
Retrieved from: http://www.internethaber.com/dolar-kuru-3 1 140-lira-olan-tarihi-
zirvesini-zorluyor-1726420h.htm
24
Investopedia. 2016. How are international exchange rates set? Retrieved from:
http://www. investopedia.com/ask/answers/forex/how-forex-exchange-rates-set.as|
Klitgaard, T., Weir, L. 2004. Exchange Rate Changes and Net Positions of Speculators
in the Futures Market. FRBNY Economic Policy Review / May 2004
Mill, J.S. 1848. Principles of Political Economy. The Project Gutenberg E-book.
Release Date: September 27, 2009. Retrieved August 17, 2017 from:
http://library.umac.mo/ebooks/b3 1139632.pdf
Mohammadi, H., Kazemi, R., Maghsoudloo, H., Mehregan, E., Mashayekhi, A.N. 2010.
System Dynamic Approach for Analyzing Cyclic Mechanism in Land Market and Their
Effect on House Market Fluctuations. Proceedings of the 29th International Conference
of the System Dynamics Society, July 25- 29, Seoul (2010).
Pettinger, T. 2012. Inflation and Exchange Rates. Retrieved March 18, 2017 from:
http://www.economicshelp.org/blog/1605/economics/higher-inflation-and-exchange-
rates/
Salvatore, D. 1995. International Economics. Prentice Hall International Economics.
Fifth Edition
Sterman, J. D. 2000. Business dynamics: systems thinking and modeling for a complex
world. Boston :. Irwin/McGraw-Hill
TCMB Expectation Survey. 2016, December 15". Retrieved from:
http://www.tcmb.gov.tr/wps/wem/connect/d1ee7b1d-41 fc-40e6-907 1-
61685e7babba/BA-Rapor-Int.pdf?MOD=AJPERES&CACHEID=d1ee7b 1d-4 1 fe-40e6-
9071-61685e7babba
TCMB Foreign Currency Transaction Volume. 2016, January. Retrieved from:
http://www.temb.gov.tr/wps/wem/connect/fdbca103-e878-42c2-al5b-
d99e134b115f/OCAK2016.pdf7MOD=AJPERES&CACHEID=ROOTWORKSPACEf
dbcal03-e878-42c2-a15b-d99e134b115f
TCMB Inflation Expectation. 2015. Retrieved from:
http://www.tcmb.gov.tr/wps/wem/connect/temb+tr/temb+tr/main+menu/paratpolitikasi
/fiyattistikrari/enflasyonthedefleri
TCMB Interest Rates. 2016. Retrieved from:
http://www.tcmb.gov.tr/wps/wem/connect/tembten/temb+en/main+pagetsitetarea/cbrt
+policy+trates/cbrt+interest+rates
TCMB Monthly Foreign Exchange Buying/Selling Amounts. 2016. Retrieved from:
http://temb.gov.tr/wps/wem/connect/4d0d01 le-94ac-4027-b8d9-
a2a94b20cd1c/Ayl%C4%B 1k+D%C3 %Bb6viz+Al%C4%B 1m+Sat%C4%Blm+Tutarlar
%C4%B | .pdf?MOD=AJPERES
Wikipedia. (n.d.) Economy of Turkey. Retrieved from:
https://en.wikipedia.org/wiki/Economy of Turkey
World Bank. (n.d.). Turkey Overview. Retrieved from:
25
http://www.worldbank.org/en/country/turkey/overview
Yildirim A. 2016, July 29". Yabanci Déviz ald: ama Tiirkiye’den ¢ikmadi. Retrieved
from: http://www.businessht.com.tr/yorum/haber/1273628-yabanci-doviz-aldi-ama-
turkiyeden-cikmadi
Yildirim, A. 2016, December 12". Kampanyaya Yabancidan daha Fazla Destek Geldi.
http://www.bloomberght.com/ht-yazarlar/abdurrahman-yildirim/1965375-kampanyaya-
yabancidan-daha-fazla-destek-geldi
VIII. Appendix
Figure 29: Complete Stock Flow Diagram
26
