Systems Thinking: A Tool for Municipalities

by Stephan E. Brown, M.A., and Daniel C. Lerch

Folks who do systems analysis have a great belief in ‘leverage points.’ These are places
within a complex system (a corporation, an economy, a living body, a city, an ecosys-
tem) where a small shift in one thing can produce big changes in everything....
...I don’t think there are cheap tickets to system change. You have to work at it,
whether that means rigorously analyzing a system or rigorously casting off paradigms.
In the end, it seems that leverage has less to do with pushing levers than it does with
disciplined thinking combined with strategically, profoundly, madly letting go.
—Donella Meadows, lead author, Limits to Growth (1972). From “Places to
Intervene in a System,” Whole Earth, Winter 1997.
Systems theory has been put to practical use in the business world for decades, helping orga-
nize global production processes and streamline multinational decision-making networks.
More recently, systems thinking concepts have been incorporated into a number of strategic
planning methods for local governments.a These and other tools can help municipalities bet-
ter understand the complex systems that are within them, and of which they are parts.
Systems thinking will also help municipalities to understand the role of key inputs like oil
and natural gas and to identify how municipalities are vulnerable to changes in the availabil-
ity and price of those inputs.

What is Systems Thinking?

“Don’t miss the forest for the trees.” This common figure of speech encapsulates the essence
of systems thinking: when we think in terms of systems, all we’re doing is looking at the
whole forest, and not just the trees.
What exactly do we see when we look at a forest? We see trees, certainly, but also animals,
brush, soil, water, and many other things. If we put a bunch of trees, bushes, a pile of dirt, a
tub of water and a family of squirrels together in a big room, however, we clearly wouldn’t
have a forest: we’d have a mess (or a bad art project). What makes a true forest are the rela-
tionships between all of its parts: the soil and water nourish the trees, the trees shelter the
animals, the animals eat the plants, and so on.
So, systems thinking is first and foremost about relationships. And when we think about
how the parts of a system relate to each other, we also notice changes: as the soil nourishes
the trees, the trees grow larger; when animals and plants die, they decompose and build up
more soil. By observing the relationships and changes in a system, we start to develop a com-
prehensive picture of how the system works. It also spurs us to ask important questions that
may help us understand the system better: What happens to the animals and soil as the trees
get bigger, and eventually die? What happens to the trees as the soil and animals change?
This way of thinking can be very helpful for understanding why things work (and change)
the way they do. It can be applied to anything that is a system—a collection of individual
parts working together—whether it’s a forest, a car, or a government program. It’s useful
because if we understand how complex things change, and why, we can make better deci-
sions to direct those changes as we see fit.
For example, 60 years ago we tried to stop forest fires as quickly as possible because we
thought that fires were simply destroying trees. Since then we’ve learned that fires are an
important part of forest systems, helping kill destructive insects, spread certain seeds, and
reduce deadwood that could fuel even more disastrous future fires. Today we contain some
fires and let others burn, and manage our healthier forests simultaneously for timber prod-
ucts, wildlife habitat, clean water and recreation. The better we understand the complex rela-
tionships in the forest system, the better we are at managing them.
When we do systems thinking, we are thinking about changes (or what systems thinkers call
“dynamics”) in terms of the relationships underlying them. Systems thinking thus views
problems as the products of some structure of relationships, in contrast to conventional lin-
ear thinking, which instead explains patterns in terms of simple causes and effects between
separate things.

The Municipality as a System

Just about everything that your municipality does or is responsible for can be thought of as a
set of relationships, and therefore as a system. For example, your town’s budgetary process
can be thought of as a system of relationships between incoming tax revenue, the expenses
of municipal departments, the priorities of elected officials, and the services the municipal-
ity provides. Similarly, your town’s street program can be thought of as a system of relation-

www.postcarboncities.net Appendix – Page 79

 ships between the needs of the local economy, the municipal Transportation Plan, the costs
Appendix of street construction and maintenance, and the city’s street maintenance activities.
We can visualize these relationships in a flowchart (Figure A-1). The flowchart below shows
that the Municipal Transportation Plan is derived from the needs of the local economy and
the costs of building and maintaining streets; the Transportation Plan then affects what will
happen with the street maintenance activities. Street construction costs can change quickly,
however, so these are also shown as influencing street maintenance activities. Compare all
the information this flowchart tells us with the simple list we might have produced had we
thought of the street program simply in terms of its components, and not its relationships
(Figure A-2).

Cost of Street
Needs of the • Transportation Plan
Construction &
Local Economy
Maintenance • Traffic Signals
• Asphalt
• Concrete
• Fill
Municipal Street
• Work Crews
Transportation Maintenance
Plan Activities

Figure A-1: Flowchart of Some Key Relationships Figure A-2: List of Some Key Compo-
that Characterize the Municipal Street Program nents of the Municipal Street Program

Both of these representations are incomplete pictures, of course, but the flowchart contains
more information and suggests additional questions we can ask to gain a more complete pic-
ture. Looking at this diagram, we might next say: “Well, what influences the costs of street
construction? What happens when the streets are (or aren’t) maintained?” With systems
thinking, we keep asking questions and revising our picture until we have a model of a sys-
tem that makes sense to us and tells us what we want to know. For example, after a few

Cost of Oil

Local Economy Cost of Materials

Activity (asphalt, concrete, fill)

Cost of Labor

Cost of Street
Construction &
Maintenance

Municipal Street Transportation

Transportation Maintenance Network
Plan Activities Efficiency

rounds of additions and revisions we may end up with a revised flowchart like this:
Figure A-3. Flowchart of Some Key Relationships that Characterize the Municipal Street
Program (revised)

Asking about the costs of street construction got us thinking about the costs of materials and
labor. The cost of labor is largely determined by the local economy, and one of the main mate-
rial costs of street construction is asphalt, which is produced from oil, so we added these rela-

Page 80 – Appendix Post Carbon Cities: Planning for Energy and Climate Uncertainty
tionships to the picture. We also started to see additional relationships, such as the effect of
street paving on transportation network efficiency, which in turn affects the local economy.
This flowchart is rather limited compared to true systems modeling techniques, but it dem-
onstrates one of the main benefits of systems thinking: it gives us tools for identifying and
exploring complex relationships. For example, if we developed this flowchart such that we Systems thinking gives
could assign quantitative values to its elements, we could try changing certain variables to
see how those changes ripple throughout the entire system. Or, we could use different tech-
us tools for identifying
niques to identify weaknesses and trigger points, or test ways of improving the system’s resil- and exploring complex
ience to change. relationships.
Understanding Systems
Definition of a system
What is a system? All systems have two defining characteristicsb : First, a system is made up
of component elements, or subsystems, that are all related to each other in some way. Second,
a system has a structure, or metasystem, that determines how these elements relate to each
other.
Looking at Figure A-3, the street program flowchart, this definition tells us that (a) individ-
ual elements like “Cost of Labor” or “Transportation Network Efficiency” can be thought of
as systems in their own right, and (b) the relationships between these individual elements
are all part of a larger structure.
Boundaries
Whether we call something a “subsystem” or a “system” depends on our level of focus: a for-
est can be a system of trees, animals and streams, or it can be a subsystem of a larger ecolog-
ical region. The level of focus we choose also determines the boundary of our system.
When we draw a system boundary, we’re basically identifying which system elements are
interacting with each other to produce the pattern of behavior that we’re interested in
explaining. The elements that receive inputs sit inside the boundary, and the elements that
only provide inputs—but do not receive inputs themselves—are outside the boundary.c If we
drew a boundary around our flowchart in Figure A-3, it would include everything except If we don’t choose the right
“Cost of Oil”. boundary we may end up
Choosing the right level of focus and boundary is important because if we’re not clear about with bad analysis – and
these things, we might not include relevant elements (or mistakenly include irrelevant ones)
in our system and end up with a bad analysis—and then bad policy. then bad policy.

Feedback Loops
In systems thinking we differentiate between “simple” systems and “complex” systems. In
simple systems, the chain of causes and effects between elements has a stopping point. Our
first simple flowchart (Figure A-1), for example, would have been a simple system because it
ended with “Street Maintenance Activities”.
In a complex system, however, the chain of causes and effects doesn’t have a stopping
point—it becomes a feedback loop. Technically speaking, a feedback loop is a circular con-
nection between two or more system elements in which a change in one element, or input,
causes other elements to generate a response, or output, that eventually feeds back to the
original element. In our more complex flowchart (Figure A-3), you can see a feedback loop
flowing through the following variables:
Local Economic Activity
—> Municipal Transportation Plan
—> Street Maintenance Activities
—> Transportation Network Efficiency
—> Local Economic Activity
There are actually four distinct feedback loops in this figure—all starting and ending with
“Local Economic Activity,” but taking different paths through “Cost of Labor,” “Cost of
Materials” and “Cost of Street Construction and Maintenance.”
Feedback loops can be either positive (“reinforcing”) or negative (“balancing”). In a positive
feedback loop, a change in one element will trigger changes that amplify, or reinforce, the
original change. In a negative feedback loop, a change in one element will trigger changes
that dampen, or balance, the original change. For example, if the feedback loop we described
above is a positive loop, we could then say that the effects of a decrease in local economic
activity would eventually result in further decreases in local economic activity.

www.postcarboncities.net Appendix – Page 81

 Parameters and Leverage Points
Appendix Systems are also composed of parameters. A parameter is a constant factor of a process, for
example, a fractional change rate such as “productivity,” “fertility,” or “depreciation.” In sys-
tems, time delays are important parameters. A time delay is the time it takes for a particular
element to respond to an input. Time delays can significantly impact how systems behave,
sometimes even making the difference between system success and failure.
By saying that parameters are “constant” factors, we mean that they are constant with
respect to a given level of system complexity. However, at a higher (more complex) level,
parameters can, themselves, be variable; that is, parameters can have parameters of their
own. For example, in our municipality’s street maintenance program, the productivity of
Leverage points are road paving may be constant up to a certain level of activity, and then suddenly increase
beyond that level by the introduction of efficiencies of scale.
opportunities for changing
When we build a model of a system, we ultimately want to identify the key parameters and
system behavior with the parts of its structure that seem to significantly influence the system’s overall behavior.
relatively little effort. These important parts are called leverage points because they represent opportunities for
changing system behavior with relatively little effort. When we adjust a leverage point we
generally make a small change to certain parameters or relationships impacting some rein-
forcing feedback loop(s); in doing so, we take advantage of that loop’s reinforcing cycle to
create a large effect in the larger system.
For example, if road paving productivity is a critical limiting factor on system performance
and is itself coupled to other factors in a positive (reinforcing) feedback loop, then a change
in productivity or one of its inputs could trigger a “virtuous cycle” of improvement in system
performance. In this case, productivity (or an input to it) is the leverage point we would want
to target.
Dynamic equilibrium, resilience and uncertainty
A particularly important system pattern is called dynamic equilibrium (or “steady state”
equilibrium). Think of the typical household heating system that regulates room temperature
using a thermostat and heater. Figure A-4 represents temperature change in a house.
Here, the vertical axis represents the house temperature, measured in degrees Fahrenheit,
while the horizontal axis represents time, measured in hours. The behavior pattern above is
called an oscillation pattern because the system fluctuates around a set point: in this case, 68
degrees F.
System thinkers call this pattern a “goal-seeking” behavior because it appears as if the sys-
tem is continuously comparing its actual
state to some “goal” state, and then adjust- 80o
ing itself to narrow the gap between the
Household Temperature (deg. F)

two. The entire system is governed by a

negative feedback loop that operates with 75o
some time delay; it’s this time delay that
generates the oscillating pattern. This
dynamic equilibrium pattern may seem 70o
inefficient, but it actually reflects system
resilience.
Healthy systems are resilient against 65o
uncertainty. In systems thinking, the con-
cepts of resilience and uncertainty start
with the idea that any system faces some 60o
risk of disruption whenever it is con- 0 12 24
fronted with a change it doesn’t expect Time (hours)
and for which it has no appropriate Figure A-4: Household temperature in dynamic
response. “Uncertainty” can be thought equilibrium
of as the amount of surprise the environ-
ment represents to a given system. It follows from this that the persistence over time of a pat-
tern of behavior represents a reduction in uncertainty between a system and its environment.d
“Resilience”, on the other hand, is the capacity or flexibility of a system to respond to chang-
ing environmental inputs in a consistent manner.e In the example above, the household
heating system demonstrates resilience against the uncertainty of outside temperature by
creating an “artificial” environment (or “bounded uncertainty”) it can in a sense predict and
therefore control.

Learning and “satisficing”

Systems can also adapt to deal with changing environments. This system adaptation process
is called “learning,” which can be thought of as any process that promotes resilience by
reducing uncertainty. For example, natural forest fires play an important role in some forest

Page 82 – Appendix Post Carbon Cities: Planning for Energy and Climate Uncertainty
ecosystems by regulating forest undergrowth to prevent cataclysmic fires. We could say that
a forest ecosystem like this has “learned” to deal with the potentially catastrophic threat of
fire by having structured itself in such a way that fires only clear the undergrowth that has
accumulated since the last fire, and therefore do not burn hot or long enough to kill mature
trees. The forest ecosystem is therefore more resilient because its “learned” structure reduces
the uncertainty posed by fire.
That structure, as you may remember from our original definition of a “system,” is called the
metasystem. A metasystem lends resilience by systematically changing its system in such a
way as to produce a controlled environmental disturbance. Each time the system “overshoots
the goal,” it produces an error (called a “steady state error”) that it then responds to. By pro-
ducing its own errors, the (meta)system can anticipate them, effectively reducing uncer-
tainty. This strategy by which a smaller problem is generated so as to head off a more serious
problem is called “error-controlled regulation” or “satisficing,” and is common in biological
and even social systems dealing with complex and turbulent environments.f
Learning is a form of adaptation, but it can also be addictive. Have you worked with some-
one who ritualistically creates their own crisis because they prefer a familiar problem to the
totally unfamiliar? Or consider the adage, “When the only tool you have is a hammer, every
problem begins to resemble a nail”; we similarly tend to perceive only those types of prob-
lems for which we have a solution. So long as a particular arrangement functions, this “addic-
tion” of creating problems to solve is not necessarily a problem. But we all know how diffi-
cult it can be to change a habit after the environment changes, even if it is plain to us that
the habit no longer serves us. In such cases, we may end up treating the symptoms instead of When we make municipal
the causeg. Systems thinking gives us a way of understanding both learning and the roots of plans and policies, we act as
resistance to learning: the two are deeply related.
“satisficing” metasystems
Because any change requires an expenditure of resources, learning follows an “economy of
learning about the urban
flexibility” in which the capacity to adapt is purchased at the price of habit formation.h
Alternatively, we can think of error—within an allowable range of tolerance, anyway—as the system we are attempting
price we pay for our (always limited) capacity to respond to multiple environmental to regulate.
demandsi. This trade-off applies to all kinds of learning, including the learning that ideally
occurs in planning and policymaking. When we make municipal plans and policies, we act
as “satisficing” metasystems learning about the urban system we are attempting to regulate.
Modeling
These basic concepts provide only a brief introduction to systems thinking, but they are also
useful tools in themselves for understanding complexity and change. The next step is to
learn about modeling, which gives us a way of deriving practical lessons from an otherwise
abstract picture of a system.
Systems thinking is fundamentally a decision-making process by which we create models of
system behaviors, and then test them to understand the system better. We can use models to
explain the causes of a system problem, and then develop solutions that will withstand the
range of expected conditions. For example, using a model of a municipal system, we could
propose various municipal policies by changing the structures or parameters that have been
identified as leverage points and then proceed to test these policy solutions under different
hypothetical conditions. Stakeholder-based or collaborative modeling can play a valuable role
in policy “learning,” starting with problem scoping or definition. Indeed, since much of
municipal complexity arises from the diversity of experiences and perspectives, modeling
should include stakeholders in the process as much as possible.j

Resources for Practical Applications

Systems thinking is still a relatively new tool for municipal policy and urban planning, but it
holds great promise for managing complex problems such as energy and climate uncertainty.
The resources listed below will help you further explore the practical applications of this
exciting field.
Pegasus Communications (www.thesystemsthinker.com)
Pegasus is a leading provider of practical resources on systems thinking, management inno-
vation, organizational change, and the next-generation workplace. Publisher of “The Systems
Thinker” newsletter, a highly-accessible print and online resource for learning about systems
thinking.
The Natural Step (www. naturalstep.com)
The Natural Step is a framework grounded in natural science that serves as a guide for busi-
nesses, communities, educators, government entities, and individuals on the path toward sus-
tainable development. Developed in Sweden in 1988, The Natural Step framework encour-
ages dialogue, consensus building, and systems thinking and creates the conditions for
profound change to occur.

www.postcarboncities.net Appendix – Page 83

 Eco-municipalities (www.sjamesassociates.com; www.sekom.nu)
Appendix A quarter of all municipalities in Sweden have adopted the Natural Step framework as
guiding policy. Known as eco-municipalities (“ekokommuner”), these jurisdictions have
started a movement that has recently spread to the United States. Sarah James Associates
(www.sjamesassociates.com) provides consultation and resources on pursuing the eco-
municipality idea in the United States.
Resilience Alliance (www.resalliance.org)
The Resilience Alliance is a research organization comprised of scientists and practitioners
from many disciplines who collaborate to explore the dynamics of social-ecological systems.
The Resilience Alliance website includes an active discussion area and resource databases
that occasionally touch on urban issues.
Works Cited and Suggested Readings
Ashby, W.R. (1956). An introduction to cybernetics. London: Methuen & Co. Ltd.
Bateson, G. (1979). Mind and nature: A necessary unity. New York: Bantam.
Checkland, P. and Scholes, J. (1990). Soft systems methodology. New York: Wiley.
Innes, J.E. & Booher, D.E. (1999). Consensus building and complex adaptive systems: a
framework for evaluating collaborative planning. Journal of the American Planning
Association, 65(4), 412-23.
Khisty, C.J. (1995). Soft-systems methodology as learning and management tool. Journal of
Urban Planning and Development, 121(3), 91-107.
Lendaris, G.G. (1986). On systemness and the problem solver: tutorial comments, IEEE,
SMC-16(4).
Linstone, H.A. (1999). Decision making for technology executives: Using multiple perspectives to
improve performance. Boston: Artech House.
Mostashari, A. & Sussman, J. (2005). Stakeholder-assisted modeling and policy design pro-
cess for environmental decision-making. Journal of Environmental Assessment Policy &
Management, 7( 3) 355-386.
Presley, A. & Meade, L. (2002). The role of soft systems methodology in planning for sustain-
able production. GMI, 37, 101-110.
Purnomo, H., Mendoza, G.A., & Prabhu, R. (2004). Model for collaborative planning of com-
munity-managed resources based on qualitative soft systems approach. Journal of Tropical
Forest Science, 16(1), 106-31.
Richmond, B. (2001). An introduction to systems thinking, STELLA. High Performance
Systems, Inc.
Senge, P.M. (1990). The fifth discipline: The art and practice of the learning organization. 1st ed.
New York: Doubleday.
Seymoar, N.-K. (2004) Planning for Long-term Urban Sustainability: A Guide to Frameworks and
Tools. Vancouver, BC: +30 Network.
Shannon, C.E. & Weaver, W. (1975). The mathematical theory of information. Urbana, Ill:
University of Illinois Press (first published 1949).
Simon, H.A. (1996). The sciences of the artificial. Cambridge, Mass: MIT Press.
Sterman, J.D. (2000). Business dynamics: Systems thinking and modeling for a complex world.
Boston: Irwin McGraw-Hill.
Van den Belt, M. (2004). Mediated modeling: A system dynamics approach to environmental con-
sensus building. Washington, D.C.: Island Press.
Walker, B., Gunderson, L., Kinzig, A., Folke, C., Carpenter, S., & Schultz, L. (2006). A
handful of heuristics and some propositions for understanding resilience in socio-ecological
systems. Ecology and Society, 11(1), 13.
Online: http://www.ecologyandsociety.org/vol11/iss1/art13/.
Endnotes
a Seymoar, N.-K. 2004
b cf. Lendaris, 1986
c Sterman, 2000
d Shannon & Weaver, 1975
e Walker et al., 2006
f Ashby, 1956; Simon, 1996
g Senge, 1990
h Bateson, 1979
i Simon, 1996
j Innes & Booher, 1999; Linstone, 1999; Mendoza & Sussman, 2005; Purnomo et al., 2004

Page 84 – Appendix Post Carbon Cities: Planning for Energy and Climate Uncertainty