ECO-ECO-SYSTEM DYNAMICS
By JOHN D. SHILLING, PHD!
ABSTRACT
The United Nation’s promotion of Sustainable Development Goals (SDGs) requires
researchers and practitioners involved in planning and policy making to take fuller
account of dynamic interactions. Rio 92 and the Millennium Declaration supported
“integrated assessment models” for long-term policies. Most work still relies on
conventional economic relations (e.g. Stern Report)’ that monetize some environmental
and social variables, but are unclear about the dynamics of natural resources essential for
economies and societies. The global economic ecosystem’s complexity requires an
integrated, multidisciplinary, systemic approach to make sustainability policies. System
Dynamics provides an ideal basis for such models by incorporating real world causal
relations, cross-sector effects, change in resource stocks, and long-term effects of
continuing business as usual compared to the effects of assumptions about environmental
shifts and new policies More extensive use of it is needed to give politicians an
integrated long-term outlook by incorporating the interactions between the economy,
society, and ecosystem so they can generate sustainable policies. System dynamic
modeling must address these relations more completely; convince more people to use
these models; and convey the results to politicians. The Millennium Institute’s
Threshold21 model includes analysis of SDGs and sustainable development strategies --
a major step towards eco-eco system dynamics.
INTRODUCTION:
The United Nations called on member states in Rio +20 to reach an agreement on a new
set of Sustainable Development Goals (SDGs). This sends a strong message to the
development planning community -- researchers, modelers, and decision makers -- to
build models that take much better account of the full range of factors on which
sustainable development is based in order to provide better analysis for more sustainable
policy making. Following the UN Conference on Environment and Development in Rio
in 1992, the Millennium Declaration for the Millennium Development Goals (MDGs),
and subsequent programs to address sustainability, there has been discussion about the
development of the so-called “integrated assessment models” for long-term evaluation of
policy options. In practice however, the major modeling work on this (e.g. the Stern
Report)” still rely fundamentally on a largely economic framework and approach to
modeling. Beyond the normal economic relations, they include some variables outside
the economy, but they are based on conventional monetary valuation. They show that the
economic costs of not doing things like mitigating Green House Gas (GHG) emissions
can be high over time, and the benefits of shifting to things like renewable energy will
' Valuable assistance was provided by Matteo Pedercini and Ade Onasanya of the Millennium Institute
2 Stern, Sir Nicolas. 2006. Executive Summary, Stern Review on the Economics of Climate Change, New
? Stern, Sir Nicolas. 2006. Executive Summary, Stern Review on the E ics of Climate Change, New
Economics Foundation
- |
pay off over time compared to business as usual at a reasonable discount rate. However,
such monetization does not address many critical, ethical, and broader eco-system
sustainability issues, and it only takes account of direct effects, not indirect ones. It
thereby tends to obscure the actual dynamics of how the ecosystem and natural resources
are related to our economy and society and how they constitute the foundation of our
development. Taking account of these relations has become more critical for maintaining
the capacity of the ecosystem to support the increasing burden our expanding global
economy and population will place on these resources over the next decades and beyond.
The UN group working on the SDGs have stressed the necessity of taking account of the
relations among the economy, society, and environment, and need more tools to do so.
Environmental models have been developed in parallel with the economic ones. Many
are used to support the IFCC studies, which highlight the challenges faced in relation to
environmental issues such as climate change and resource depletion. They make
projections based on continued business as usual (BAU) economic activity and then
indicate the changes needed in certain economic outputs, like reduced GHG emissions, to
increase the chances of sustainability. This leads to recommendations to shift policies to
achieve some of their goals, like increasing renewable energy use or accepting crop shifts
due to climate change. But it does not take account of further relations and feedbacks
with more economic and social factors.
While both of these modeling approaches address many of the same sustainability issues,
like causes and changes needed to avoid serious problems such as excessive climate
change, they focus primarily on direct effects within their own economic or
environmental relations. One example was the early promotion of corn based ethanol as
a beneficial way of replacing fossil fuel use with renewables to reduce emissions. The
initial economic models considered only the cost of growing corn and extracting the
ethanol. They did not take full account of the emissions of all the related farm activity,
production of fertilizer, etc. When the full life cycle results of corn ethanol production
and the lower miles per gallon of ethanol compared to gasoline were included, the more
complete analysis showed that the overall effect would be a net increase in CO)
emissions per mile driven. The environmental models took account of the full production
elements related to corn based ethanol, but they did not consider the longer term effects
of the increased use of corn for ethanol would have on food availability, which would be
reduced and lead to higher food prices as more corn was shifted to ethanol production.
Fortunately, both of these issues were raised by people who had a more integrated view,
and this has reduced, but not totally eliminated, the production of corn-based ethanol.
While some models do take account of first step direct effects beyond their own sectors,
it has become increasingly clear that a truly integrated, multidisciplinary approach is
needed to represent the actual nature and dynamics of how our eco-systems, economies,
and societies interact over time, to provide a more coherent and consistent understanding
of challenges that need to be addressed, and to help policy makers cooperate to reach
agreement on more sustainable policies.
3 Political interest groups supported by those who make money from the corn based ethanol also have
worked to maintain its use.
on Dine
System Dynamics is a highly innovative approach to modeling that offers an ideal
platform for the development of integrated and multidisciplinary models. It offers a solid
basis for the analysis of factors that underlie the proposed SDGs, for analyzing the full
cross sector effects of policies to achieve the SDGs, and for monitoring and evaluating
progress being made by the sustainable development strategies. It has been developed
over the past 50 years and has been used in a number of interesting areas. A broadly
promulgated example of applying system dynamics to address such sustainability issues
in an integrated manner is the Millennium Institute’s Threshold-21 (T21) model, which
has been implemented in more than 35 countries to help them create and implement
strategies to achieve the Millennium Development Goals (MDG), adapt to climate
change, and improve welfare in a sustainable manner. Several examples of its use are
described in Boxes below. The model has also been adapted globally in UNEP’s Green
Economy Report to show how shifts to ‘greener’ investment can improve economic
results over time. This Green Economy approach is now being applied nationally in
developing countries.
It is clear that System Dynamics has highly advanced features that will make a major
contribution to the extension of integrated modeling to help define and achieve the SDGs
and help countries create sustainable policies. The application of its real world causal
relation process contributes a lot to learning about how factors interact, directly and
indirectly, and highlights these factors to policy makers. It takes account of more cross-
sector relations and facilitates the incorporation of detailed sector studies and views of
participating sector experts, which contributes to building more cooperation. It can also
provide a longer term view than most other models. This contributes a great deal to
understanding the real world relations being modeled and illustrating where there are
positive and negative effects several steps through the causal process and into other
sectors. This goes beyond most orthodox economic modeling, which tries to adapt the
model to the theoretical views of the economic relationships being modeled and takes
little account of results that go through non-economic areas. Despite its use in a number
of areas and some useful application of system dynamics in policy making system,
dynamic modeling has not yet reached a level of use where it has a consistent impact on
public policy makers or their economic and other advisors. In view of the critical impact
economic development has had and will continue to have on the environment, it is
increasingly important to be able to take account of the interaction between the economy
and the ecosystem. System dynamics can increase people’s capacity to do this. More
effort is needed to do eco-eco-system dynamic modeling and to expand its use in both
academia and the political world. ISDC and other system dynamic groups can help
achieve this.
Way System Dynamics Is VITAL:
This modeling approach moves well beyond conventional models in a number of ways.
This paper will focus primarily on comparisons with socio-economic modeling, and to
some extent with environmental modeling. Much progress has been made in most
modeling techniques over the past decades, and vast advances in computer technology
3
have permitted much more complex models to be created and run quickly. However,
many shortfalls remain.
Briefly, established economic models are based on translating economic theory into a
quantitative framework. Simple ones tend to be linear (or exponentially linear) to show
the effects of different growth rates of the main variables on the major economic
balances, but they include little, if anything, about the interactions among the economic
sectors. These include the IMF and World Bank accounting and RMSM.. models. They
generate little useful information beyond the short term, and that is even more illustrative
of the effects of different assumed growth rates than descriptive of the real causal
relations in the economy. They take account of only a limited number of economic
relations focused on budget, investment, and trade balances, and they do not include
social or environmental factors in generating their projections. They are based to some
extent on econometric determined correlations.*
More advanced models, such as computable general equilibrium (CGE) models, take
much more account of the economic relations among the economic sectors included.
Extending the early Input-Output models, CGEs include exchanges among a number of
economic sectors, investment, consumption, the government, and trade, which cover a
large number of markets. This involves creating a large matrix showing the flows from
all the producers, consumers, etc. to and from each other. The flow in each cell in the
matrix is then converted into an equation that in the context of a simultaneous solution to
the whole model will determine the supply by the provider, the demand by the user, and
the price of the exchange that will balance the supply and demand in that market. Their
overall solutions generate balanced market equilibria in all cells of the matrix where
exchanges occur and are based on achieving an optimum equilibrium for the overall
economy. The optimum is typically defined as maximizing GDP, and may be subject to
constraints about minimum or maximum acceptable values of some variables. Typically,
the CGE model is solved to compare the optimum results of different policies or other
assumptions with the base case set of policies. This is a comparative static result that
provides useful information about how optimum GDP and all the other variables that will
shift as a result of the policy changes tested. However, the process does not show the
path of the changes that occurs for any variable, just the comparison of “instantaneously
calculated” equilibrium values between the two solution of the model. It also does not
show how long the transition will take, since it assumes that all markets will reach the
new equilibrium instantly.
Progress has been made to further reflect reality in CGE models, partly to better address
MDG goals. These CGEs are referred to as MAMS.° They are designed to include a
number of the MDG targets from the social sector and generate scenarios over time,
using one year intervals, They assume that markets don’t reach equilibrium instantly, but
*T helped develop the RMSM model of the World Bank many years ago and have analyzed many
applications. I also reviewed many of the IMF models in countries I was working on.
5 For example, Aid, Service Delivery, and the Millennium Devel Goals in an Economy-wide
Framework, by Frangois Bourguignon, Carolina Diaz-Bonilla and Hans Lofgren of the World Bank, July,
2008
-4--
provide an exogenous assumption about how close the market will move toward
equilibrium in each sector in each period. They also provide ‘rules’ about what changes
will occur in the base input set of variables in each period going forward. These would
include exogenous assumptions about such things as the population growth, and
endogenous changes resulting from the value of some variables in a period, such as net
saving, which is carried forward as new investment into the next period, thereby
increasing the capital stock and output capacity. These models also include factors like
education, health, access to water in what are called MDG production functions that can
be incorporated into the economic base of the MAMS model. These versions of the CGE
models produce better results and show year by year shifts in all variables as they
approach optimization, but still are quite limited in the links between the social and
environmental factors that affect and are affected by economic development.° They also
are based on the assumptions that economies will always tend toward a stable
equilibrium, have unlimited access to necessary resources (unless limits are specified in
the model) or generate substitutes efficiently, and can dispose of any waste. They also
typically ignore what economists call externalities, which are side effects that are not
included in the normal market actions. For example, a new chemical plant will generate
positive GDP effects which are taken into account, but if it produces pollution in nearby
streams, the negative effects of that pollution downstream on farmland, health, etc. are
not taken into account because they are not part of the economic market. As economic
activity becomes more pervasive and global, there are more externalities that need to be
taken into account. These models take a huge amount of time and effort to construct and
run, and only highly skilled modelers can really understand their relations, how they
generate scenarios, and properly interpret the results.
However, these economic models remain mostly within their own narrow boxes and take
little account of the relations of the economic activity with the rest of society and the
world, nor of the sustainability of the development beyond certain MDGs. Conventional
economic models that focus entirely on the standard economic relations underlie most
analyses of proposed economic policies. In fact, many politicians and their analysts are
using even more constrained models that only represent their ‘partisan’ views of what is
the right economic policy, despite considerable lack of supporting evidence to justify the
results. In addition, they typically ignore the social and environmental issues.’
© Some models do try to include economic-type equations of social factors like education and effects of
health expenditures, but these are often outside the simultaneous solution indicating change over time.
One approach to create an endogenous demand for education included relations based on the assumption
that consumers of education were able to estimate the discounted present value of the benefits of education
(even before they had been educated) in order to determine the amount of education that is worth having
compared to the value of working on the family farm.
7 Economic models based on the assumptions of pure free markets, supply side stimulus, etc. produce
policy recommendations that do not take account of the lack of validity in the real world of many of the
assumptions on which such models and their results are based. In addition, they tend to ignoring social and
environmental factors. So they are not justifiable to support sound policies, but are often used to support
political positions of special interest groups.
a5
Various environmental models used by IFCC and others do address many of the critical
environmental changes occurring, which are often the result of economic activity, such as
climate change. They estimate how things like warming are likely to affect agricultural
production, drought and flooding, and other factors. But they do not take account of the
feedback and interactions between the environment and the economic and social
activities. They do provide valuable input that can be used in system dynamic models to
further expand the eco-eco-dynamics that need to be addressed.
The rapid growth of the world’s population and extensive globalization of the world’s
economic activity mean that a much broader set of relations needs to be taken into
account. With the increased complexity and density of economic and social activity, the
multitude of critical direct and indirect interrelations among economy, society, and
environment factors have much more significant effects than was the case a century ago
when the population was one third its current level and economic activity much smaller.
It is increasingly clear than simply basing policies on the modeling of a sector or other
limited area will miss critical cross sector relations and thereby will not produce the best
results and may well cause serious problems for sustainability. This is why it is very
important to apply system dynamics more broadly for public policy making and get the
message out to the decision makers to clearly demonstrate more realistic and broader
effects of different policies and the kinds of dangers that will arise in the future.
STRENGTHS OF SYSTEM DYNAMICS:
System dynamics is a much more appropriate modeling system to address these issues of
achieving sustainable development of our society and economy, which depend on the
limited and variable foundation of the eco-system. There are several very important
aspects of system dynamics that make it appropriate for these applications. It has no
inherent limitations on the variables, sectors, or time frame to be considered. This
enables the modeler to adapted the structure to the key aspects of the situation being
addressed. The results can be readily and clearly demonstrated to policy makers, as can
the relations that lead to the results. Valid relations that the policy maker wants to
consider can also be added, which can improve the model and build the confidence of the
policy maker. This process is very important in that it contributes to learning more about
what needs to the done to design development policies that will help better achieve more
sustainable development over the long term in a highly populated world, demonstrate
how achieving SDGs will positively affect overall well being while protecting the
foundations in the ecosystems, and convince policy makers and the public that these
policies are necessary to assure more sustainable development, so as to gain their support.
The main strengths of System Dynamics are summarized below.
System dynamic models are based on real world causal relations. This information can
be derived from many sources, but needs to be validated as actual causal relations, not
just correlations. While these relations need to be quantitative, they do not need to be
expressed in economic value terms, which allows a much more extensive and
comprehensive model base. Causal loop diagrams also help understand the complexity
of many of these relations much more easily than normal economic model equations, and
they demonstrate both positive and negative feedback loops. Doing this causal analysis
-6—
in cooperation with experts in various related areas also helps learn more about the
relations that need to be taken into account and the critical steps that occur in the causal
relations. See Box 1.
System dynamic models readily take account of externalities beyond the economic
markets. They can include pollution and the effects it has on other factors downstream,
which may not be monetarily valued. For example, the effects of the pollution on
reduced soil fertility, higher illness rates, and lower worker productivity and/or higher
Box 1: Mozambique
In 2004, MI applied its T21 model in Mozambique with the support of the Carter Center to help
develop their Agenda 2025 Development Strategy. One of the main policies considered was
constructing roads in rural subsistence agricultural areas to give farmers more access to markets for
inputs and to sell their crops. Incorporating detailed information from the Agricultural and
Transportation Ministries, the model showed how over time, incomes would increase in the areas
with access to the newly build roads — reducing poverty. It also showed improvements in urban
incomes and agricultural exports due to the increased production and commercial activity. However,
when this was discussed with a broad based group of policy makers, the Health Ministry pointed out
that increasing the roads and commerce on them would be an increase in HIV/AIDs infections.
Based on their studies, these factors were included in the model, which then showed the negative
social side-effects of the roads, as well as the positive economic results, which were now lower due to
the labor force reductions and increased mortality in affected areas. As a result, policy makers got
together to incorporate HIV preventative measures and treatment capacity into the road building
program, protecting the benefits and mitigating the negative side effects.
mortality rates are likely to affect the economy over the longer term through such factors
as a reduced work force and higher medical expenses. These need to be taken into
account in making decisions, representing non-market based values of improving human
welfare, and attaining more sustainable development.
System dynamics readily takes a long term-view in its scenarios. While it cannot make
prefect predictions, it does illustrate likely longer term paths of all the variables, which
may shift from continuous trends typical of most economic models to tipping points and
major shifts. This is very important in dealing with sustainability and environmental
issues, since they are associated with long term effects by human standards, but which are
quite short term in environmental periods. It is also important to be able to take account
of structural changes that will occur over time and not just assume that current stable
structures will continue. There are many inherent structural changes that occur — shifts in
the structure of the population will affect the economy; depletion of resources (water,
fertile land, etc.) will affect the bases for production; climate change will affect many
things. The normal time lags for many of these results is much longer than what most
humans or models are used to dealing with, but the long-term effects certainly need to be
addressed if humanity is to survive as long in the future as it has in the past. It is
important to take account of structural changes due to both endogenous factors in the
model and exogenous ones that are likely to occur. See Box 2. And different rates of
7
exogenous change can be run to see how sensitive the rest of the model is to the rate of
8
change.
Box 2: USA Fuel Efficiency
In 2007, MI developed a T21 model to address energy issues in the USA. It was used by several
policy makers to demonstrate that increasing the CAFE standard of fuel efficiency would
actually help the economy over time. Applying the proposed increase in the CAFE standard
slowly increased average car fuel efficiency over about 30 years, which was how long it took for
the more fuel efficient cars to constitute all of the fleet. GDP grew more compared to continuing
the base level of fuel efficiency due to several factors, primarily less need to import oil and
lower expenditures on auto fuel that led to more expenditures on other consumer items based on
local production. This helped create more jobs and demand, which raised production. And CO2
emissions declined compared to the base case. These gains more than offset minor losses in the
automobile and gasoline industries. This convinced more policy makers to support the higher
CAFE standard, and it was passed. However, looking further into the future, CO2 emissions
started going up again. The model showed that the increased domestic production that resulted
from the reduction in gasoline consumption led to more use of energy by local industries, which
was provided by coal produced electricity at the margin. This highlighted the need for further
CO2 reduction policies and demonstrated the value of long term integrated scenarios.
These strengths illustrate the importance of using system dynamics to get beyond the
economic factors and produce a more complete view of the relations among economic,
social, and environmental factors. It gets out of the conventional boxes and takes account
of the cross sector relations that are, in fact, how the world really functions. This
approach promotes more thinking outside the box and encourages those working with the
model to take account of more cross sector relations and learn from specialists in the
sectors about the relations that are involved in their sectors and with other sectors. This
broader spectrum of modeling is very important.
Other important key factors include how system dynamic models can be readily updated
to add additional sectors and more information about existing sectors. These models
produce graphic scenarios to demonstrate and compare the results of different policies
and assumptions, as well as tables. This makes the results much more comprehensible to
non-modelers, which includes most policy makers and the public. Such comparisons of
different policies and assumptions with system dynamic models demonstrate the results
of policies across many variables. This, combined with the causal link charts, helps
convey the critical relations and their results to non-modelers, which can lead them to
have more comprehensive discussions and generate more integrated policies. This
transparency enables groups of policy makers to get together and discuss policy options
and combinations of policies across different sectors beyond their normal relatoins,
* As the system dynamic model becomes more extensive and complex, more of these exogenous factors can
be made endogenous, for example the rate of temperature increase due to GHG emis
--8--
sharing the ‘common policy language’ of the system dynamics model. It makes it easier
to reach compromises when the overall results and benefits to the whole constituency are
demonstrated. See Box 3. These factors help explain why system dynamics modeling
which integrates economic, social, and environmental factors needs to be more broadly
applied and promoted.
Box 3: Getting Policy Makers Together: Namibia and Bhutan
In 2011-2, MI, supported by the Japanese CCA program administered by UNDP, worked with
Namibia to develop better Climate Change Adaptation policies. A meeting of major ministries
and other key stakeholders was held to elicit their main concerns about CCA and possible impacts
on other sectors. In discussing resources, the Mining Ministry pointed out that they would
expand production of uranium mining and provide necessary foreign exchange resources. When
asked what they needed to do so, they replied that they only needed more water and energy,
which they did not see as a problem. The ministries dealing with water and power immediately
pointed out that those were two of the most scarce resources available in Namibia and most
threatened by climate change. Based on this revelation, those involved in mining, energy, and
water agreed to meet and work our a mutually agreeable plan to manage water and energy for
mining while not aggravating other sectors covered in their CCA program.
Earlier, MI had worked with the government of Bhutan to help them meet their objectives of
improving Gross National Happiness (GNH). The government was committed to these goals
more than economic growth and felt it had adequate income supported by its exports of
hydropower to India. Having been involved in the development of the model to improve the
government’s policies to increase GNH with much more spending on social benefits — health,
education, water access, nutrition, etc. primarily in urban areas, the Strategy Directors of the
major ministries met to discuss the results. One thing the model demonstrated was that since 70%
of the population lived in rural areas and depended on agriculture, which would only grow at 1%
pa under the proposed policies, there would be massive migration into urban areas. It was
determined that this was not feasible, so the social ministries decided to shift some of their
investment resources to improve the rural economy incomes in order to reduce immigration. This
was done and proved successful in improving overall welfare.
MucHu Dong, MucH More To Do:
The most comprehensive indicator of the serious sustainability challenges we face is the
estimate of the current level of the global foot print. This represents the amount of
biologically productive land and sea area necessary to supply the resources a human
population consumes and to assimilate associated waste, all in a sustainable manner
consistent with the earth’s capacity to renew these resources. The current level of this
footprint at 1.5 earths means that the use of global resources by our economies and
societies amounts to one and a half times the sustainable amount available on the earth.
We are exhausting necessary resources — effectively draining the illustrative water stock
tank of system dynamic courses. But the large amounts of most of these natural resource
stocks keeps most users from understanding their depletion, since they seem to be fully
9
available in the short term. However, many parts of the world already face serious
problems of water availability and are encountering more soil depletion which will
reduce agricultural production and weaken their food security. So some areas are
beginning to understand the risks of the expanding footprint.
Probably the best known risk factor of the current economy’s impact on the environment
is climate change, due to emissions of GHG. This includes many indirect cycles beyond
fossil fuel emissions of CO2. As warming occurs, more methane is emitted from melting
arctic tundra. And as the amount of ice is reduced in the arctic, the exposed Arctic Ocean
absorbs more solar heat, which also leads to more warming and shifts in the Jet Streams
that affect weather volatility in the lower latitudes. Warming also affects water
availability in the rest of the world, agricultural productivity, and habitability of much
land. Some areas benefit, but most don’t.
Expansions of societies and increased use of agricultural land, various chemical
pollutants emitted, and other behavior factors of the modern economy are threatening the
habitats of many species, which increases their chances of their extinction. While not all
species can or need to be protected from extinction, as a number have become extinct
through natural processes over paleontological time, it is vitally important to assure many
do survive because they are vital to the survival of our productive and social processes.
For example, loss of bees would greatly hurt agriculture because they are vital for
pollination of many crops.
The structure of the environment and mix of species will continue to change over time,
but it is very important to prevent or mitigate changes in the environment and species
populations that will be detrimental to continued human development and welfare. Given
these increasingly vital interconnections between the economy, social expansion, and the
environmental foundation, it is very important to take them into account in designing
policies and in making the necessary changes in behavior patterns that will assure the
sustainability of the foundations of our human systems for our children and
grandchildren, not just increasing next year’s income as much as possible.
The SDGs need to be designed to provide important information about changes in critical
sustainability variables and to keep the public and policy makers informed about how
policies and activities are affecting sustainability. The specifics of the SDGs are still
being discussed. It is important for those working on them to include the key factors that
affect sustainable development and preserving the environmental foundation of economic
and social progress. Doing so will require taking fuller account of the relations among
these three dimensions of our global structure and to take a long term view, since many
factors that affect sustainability only appear after a long term in human terms, but it is
essential that appropriate actions be early enough to avoid and tipping point to assure that
we do not run into a sustainability abyss. There are critical tipping points that if passed
will mean that certain goals cannot be achieved. Defining SDGs, developing the policies
to achieve them, and monitoring how well progress is being made will require much
° See Global Footprint Network at http://www. footprintnetwork.org for more information about this.
-- 10 --
more comprehensive, real world based, and long term tools than have been typically used
in the past. Fortunately such tools exist in system dynamics.
System dynamic models have been used to address many of these important issues.
Limits to Growth’? raised a number of questions about the effects of continued traditional
development on the ecosystem. Other models have addressed sustainability issues,
including ones done by John Sterman and others at the Sloan School. '' The University
of Bergen and others teach courses on system dynamics in economics. The Millennium
Institute developed its system dynamic model, Threshold 21, when its founder, Dr. Jerry
Barney, learned in writing the USA study, Global 2000’? in 1980 that different US
government departments used different models with different basic assumptions about the
same exogenous variables. He recognized the need for a more consistent integrated
model and searched for one. After much effort, he discovered system dynamics at MIT,
met with Jay Forrester and others, and was able to use system dynamics to develop the
Threshold 21 model for the Millennium Institute, with much assistance from experts at
MIT. Since its creation in the 1990’s, T21 models have been used in a number of
countries to support more coherent strategic planning, as has been discussed in the Boxes.
The T21 process has also been used by UNEP in its Green Economy Report to
demonstrate how shifting to green investment will improve both sustainability and human
welfare.'? But unfortunately, all of this work has only had limited effects on moving
decision makers toward more sustainable policies.
This is why it is very important to become more active in promoting the systemic
approach for critical decision making. Action is required at several levels. First, more
basic modeling needs to be done that relates the economy to the ecosystem in adequate
detail. It needs to analyze how major economic activities are likely to develop over time,
how they will effect the ecosystem, and what feedbacks there will be from the ecosystem
that will affect the economy and social wellbeing. These relations will be both positive
and negative, and it is critical to understand which will be most important. Such
modeling is not just extending existing relations; it involves studying the important cross
sector causal effects that would become more clear from integrating the economic and
ecosystem relations so that the modelers would learn more about the depth and direction
of such relations. This improved understanding of real world relations would provide the
basis for developing policies to improve the positive benefits and mitigate the negative
ones over the longer term. The relations that are analyzed and indicators developed need
to extend well beyond normal economic indicators of GDP and profits of special interest
groups. They need to take account of expanding global sustainability indicators,
including SDGs, preservation of essential environmental resources, and assuring human
10 Donella H. Meadows, Gary. Meadows, Jorgen Randers, and William W. Behrens III. (1972). The Limits
to Growth, New York: Universe Books
"MIT Economics doesn’t seem to have any interest, unfortunately.
'? ‘The Global 2000 Report to the President: Entering the 21st Century. 1980. A report by the Council on
Environmental Quality and the Department of State. By Gerald O. Bamey, Study Director
'S UNEP, 2011, Towards a Green Economy: Pathways to De and Poverty
Eradication, www.unep.org/greeneconomy
--11--
welfare. This does not mean trying to protect and preserve the environment and society
as it existed in the past, but assuring that the inevitable changes in both the economy and
ecosystem going forward avoid major cliffs and other factors that would undermine the
sustainability of the ecosystem foundation that supports our economies and societies.
Second, it is very important to get the message out that more comprehensive
methodologies need to be used at design and implement policies that will assure the
sustainability of the ecosystem that supports our economies and societies. Too many
people suppose that these problems are overstated, based on their hopes and beliefs that
major changes are not necessary and on the publicity supported by special interest groups
who want to protect their short term profits from continuing BAU. Decision makers and
the public need to understand the broader and longer term risks posed by continuing
business as usual, which many partisan groups support. While the scientific analysis of
the threats to the ecosystem and climate change are well established, if not precise, the
potential impacts on the economy and society over time have not been publicized clearly
enough.
System dynamic models designed to address these issues can more realistically and
convincingly demonstrate to decision makers and the public what is going to happen as a
result of continuing business as usual in terms of more comprehensive and meaningful
indicators and factors. The models produce easily understood graphic outputs that
compare results of different scenarios over time and demonstrate the causal relations that
lead to the results shown. They can reveal to non-modeler audiences what is likely to
happen, and they can be very effective in changing people’s minds -- especially where
the results of past decisions on the current problems being faced can be clearly
demonstrated. The boxes above give good examples of this. Based in the eco-eco-
system dynamic modeling that is available and can be strengthened, it would be possible
to get out the message explaining how to attain sustainable development by numerous
means, including by hard copy articles beyond academic journals, by website
information, and by widely circulated blogs.
Third, it is important to convince decision makers and their advisors to use this systemic
approach in their analysis and development of viable policies. Too many of even the
highly competent economic advisors still rely on conventional economic models and
theory to develop policies. They rarely look out of their sector boxes to see the broader
issues that need to be addressed and the impacts of the sector policies that they propose
on other sectors related to the environment and social welfare. It is important that they
learn to take a broader view of the effects of policies across sectors, which can be
achieved if they understand how to take a more integrated view of the issues being
addressed. Significant progress can be achieves by spreading the use of the systemic
approach more widely in academia and managing more sharing of information and
cooperation among the departmental silos that have grown up. While this greater
concentration on individual sectors is partly due to the increased amount of information
and details that must be mastered in each department, it needs to be recognized that this
increased knowledge also means that we have the basis to understand better the relations
among the sectors, which is essential for understanding how the world really functions
since all sectors interact in reality. This more integrated approach is increasingly
--12--
important due to the extent that global activities involving the economy, society, and
environment have become more complex and interactive in today’s world — and will
become much more so in tomorrow’s.
Reaching decision-makers can be difficult, but there have been cases where some success
has been achieved, such as the Montreal Protocol and the use of Threshold 21 to convince
a number of Congressmen to pass the CAFE increase in 2007. (Box 2) More effort is
needed to inform decision-makers and their advisors about the need for more coherent
develop policies, to demonstrate what policies will work best overall, and to identify the
negative side effects that need to be mitigated. This will require more publicity and more
means of making direct contacts with these policy making individuals to better inform
them about the systemic challenges being faced by means of non-partisan models. This
will help convince them of the real relations and impacts of strategies so they can build
their policies on a more integrated basis, which will certainly benefit their legacy as well
as the sustainability of the country.
Given the global nature of the ecosystem and the globalization of the economy, more
international consensus and cooperation will need to promoted. The recent work by
UNEP with the Green Economy Report is a first step in this direction. It uses a system
dynamic model" to demonstrate how shifting 2% of global investment to green activities
that reduce GHG emissions and help protect the environment will lead to an economic
structure that is more sustainable in 2050 than what will happen by continuing business
as usual. This model demonstrates that while there will be some slower growth in the
early years of the shift to greener investments, well before 2050, GDP and job growth
will pass BAU and more sustainability will be assured. The model demonstrates that
there will be more poverty reduction and that in key areas like agriculture, food security,
and water availability, the green investment program will assure sustainability, in
contrast to the business as usual path, which will cease being sustainable soon after 2050,
leading to serious problems. This is the kind of information that can be used to convince
policy makers to make better policies.
Fourth, continued support for this work is vital to its success. The development of the
world’s economies, societies, and their relations to the ecosystem are ongoing processes
that continually change. There will be no stable equilibrium or standard environmental
situation. The changes are due both to human based transformations and nature based
shifts in the ecosystem. It is important be able to incorporate these changes in the past
into the models being used and to take account, to the extent possible, how they will
continue to shift in the future. A significant part of this future change can be included in
sophisticated eco-eco-system dynamic models, and they can be regularly updated as new
data and causal relations become available. They can also include shifts in political and
social priorities, which also change over time.
These models should remain works in progress. This will require continued support and
development in academic institutions, further education of students in a broad range of
'* The global model used is based on Millennium Institute’s Threshold 21 and was developed by MI.
UNEP is now promoting the use of the model in many countries to help achieve green local policies.
-- 13 --
areas about the systemic approach, and continued presentations and dialogues with the
policy makers and their advisors. This relationship needs to be institutionalized over
time. One important element is to assure and emphasize that the eco-eco-system dynamic
models are based on the best possible representation of reality. They need to be assuredly
non-partisan, and should involve all major political interests to assure that the dynamic
Eco-Eco-System approach is broadly understood, accepted, and used. The more that
different groups become familiar with the models and are able to propose realistic
relations that can be incorporated, the more likely they will be to accept the results of the
models and reach agreements on the better policies, unless they are totally ideologically
driven. See Box 4. It is important to build and continue the support base for the eco-eco-
system dynamic models, incorporate a broad set of SDGs, and assure their continued
expansion and improvement. And this information needs to be continually disseminated
as broadly as possible.
Box 4: Bringing Together the Opposition; Jamaica
MI had worked with the Planning Institute of Jamaica (PIOJ) for several years in the early
2000’s to train their staff and adapt the T21 model to their issues. They used it to develop their
strategy to become a high income country. However, with an upcoming election, the Director
of the PIOJ was worried that the opposition party would win, and was likely to throw out the
tools used by its predecessor, including the T21 that PIOJ depended upon. So he decided to try
to convince them to keep the model. He arranged a meeting with the leaders of the opposition,
MI, and PIOJ to discuss the model. T21 was presented in some detail to the leaders of the
opposition, and their first response was that it was just another instrument that the governing
party used to prove its own points, so they had no reason to keep it. The director of PIOJ then
asked them to describe what were the key policies they wanted to implement. They described
five, and it was possible to run three of them in the model as it was structured. They were
quite interested in the results. MI also explained how the additional policies they were
interested in could easily be incorporated into the model to demonstrate the results broadly.
The opposition then expressed further interest in the model and asked more questions about
how they could use it. They decided it was non-partisan and agreed to continue its use in PIOJ
after they were elected. And they kept the Director in place. So T21 can form a viable and
enduring long term policy tool.
CONCLUSIONS:
Based on the above discussion, it is quite clear that we face serious challenges in order to
achieve sustainable development and protect the earth from a major ecosystem shift that
will seriously threaten the survival of our societies and economies. Current policies are
not addressing most of the major challenges to achieve sustainability, and the basic
modeling tools used for policy making tend to ignore most of the important non-
economic issues and cross sector causal effects. So policy makers and the public are not
well informed and not yet willing to make the necessary broad based changes in policies
and related behavior patterns. We have seen that to the extent the public has become
--14--
better informed about some specific issues and understands the risks and need for change,
more people have been willing to make changes. This highlights both that more
understanding of the cross sector sustainability issues and the effects of policies over time
is needed and that this information needs to be much more broadly disseminated in a non-
partisan manner. This fully supports the proposal to work harder to get the message out
systemically, along with the need to design stronger models and build support for better
policies to achieve more sustainability.
The System Dynamic approach has been demonstrated to be the most effective tool for
taking account of the interactions among the society, economy, and ecosystem; for
generating scenarios to illustrate what is likely to happen if no policies are changed,
particularly the negative effects on the supporting environment and society that need to
be addressed; and to demonstrate how more appropriate policies will assure more
sustainability and protect human welfare. It is quite effective in generating useful
indicators of progress toward sustainability (including SDGs), demonstrating how
cooperation across sectors can address the important challenges, and indicating how to
reduce or prevent the negative effects while promoting the positive ones. These results
can be transparently and convincingly presented to non-modelers and non-technicians,
which will help build support among the public and decision-makers, as has been
demonstrated in a number of cases described in the Boxes. Thus it is very important to
expand system dynamic modeling to take account of more sustainability issues; to get
more academics, students, and political advisors involved in the application of these
models; to get the message out to the public; and to actively promote more sustainable
policies. We need to strongly encourage more development and use of Eco-Eco-System
Dynamics.
-- 15 --
Additional References:
Books and Reports:
Barney, G., Blewett, J., and Barney, K. (1993) ‘Global 2000 Revisited: What Shall We Do?’,
Millennium Institute.
Bassi, A. (2008) ‘Modeling U.S. Energy Policy with Threshold 21’. VDM Verlag Dr. Mueller
E.K. ISBN-10: 3639048229; ISBN-13: 978-3639048223
Yudken, J. and Bassi, A. (2009) ‘Climate Policy Impacts on the Competitiveness of Energy-Intensive
Manufacturing Sectors’. National Commission on Energy Policy, Bipartisan Policy Center, Washington
DC, USA.
Bassi, A., Shilling, J., and Herren, R. (2007) ‘Informing the US Energy Policy Debate with
Threshold 21’, Millennium Institute, Arlington, VA.
Meadows Donella H., Jorgen Randers and Dennis L. Meadows, Limits to
Growth-The 30 year Update, 2004
Pedercini, M. and Barney, G. (2004) ‘Dynamic Analysis of MDG Interventions: The Ghana
Pilot’, Technical report prepared for Capacity Development Group, UNDP New York, NY.
Pedercini, M. (2004) ‘T21-Cape Verde: Report of the Poverty Reduction Strategy Paper
Exercise’, Technical report prepared for the Government of Cape Verde.
Pedercini, M., Bogdonoff, P., and Qu, W. (1999) ‘Chinese and Global Food Security to 2030:
Reducing the Uncertainties’, Final Report of the Strategy and Action Project for Chinese and
Global Food Security, Millennium Institute.
Articles and Papers:
Bassi, A., (2007) ‘The T21 model customized to the U.S.: General Overview’, Conference on
System Science, Management Science and System Dynamics, 19-21, October, Tongji University,
Shanghai, China.
Bassi, A. (2007) ‘Threshold 21-USA: Overview of the Energy Sectors’, Conference on System
Science, Management Science and System Dynamics, 19-21, October, Tongji University,
Shanghai, China.
Bassi, A. (2007) ‘Threshold 21 (T21) USA: Behavior Description’, in Proceedings of the 25"
International System Dynamics Conference, 29 July — 2 Aug., Boston, MA. 978-0-9745329-8-1
Bassi, A., Schoenberg, W., and Powers, R. ‘An Integrated Approach to Energy Prospects for North
America and the Rest of the World’, submitted to Energy Economics.
Bassi, A. (2006) ‘Modeling U.S. Energy with Threshold 21’, in Proceedings of the 24th
International Conference of the System Dynamics Society, 23-27 July 2006, Nijmegen.
Bassi, A. and Lorenz, T. (2005) ‘Comprehensibility as a discrimination criterion for Agent-Based
Modeling and System Dynamics: An empirical approach’, In Proceedings of the 23rd
International System Dynamics Conference, 17-21 July, Boston, MA.
-- 16 --
Bassi, A. (2005) ‘Strategic Analysis Evolution: Scenario planning and simulation based on the
methodology of System Dynamics’, Korean System Dynamics Review, Summer 2005.
Pedercini, M. (2005) ‘Potential Contributions of Existing Computer-Based Models to
Comparative Assessment of Development Options’ CCG Report, Conservation International,
Washington, DC.
Pedercini, M., Sanogo, S., and Camara, K. (2007) ‘Threshold21 Mali: System Dynamics-based National
Development Planning in Mali’, in Proceedings of the 25" International System Dynamics Conference,
29 July — 2 August, Boston, MA. 978-0-9745329-8-1
Pedercini, M. (2004) ‘Evaluation of Alternative Development Strategies for Papua, Indonesia: A
Regional Application of T21’, in Proceedings of the 22nd International Systems Dynamics
Conference, 25-29 July, 2004, Oxford, England.
Pedercini, M. and Barney, G. (2003) ‘Models for National Planning’, In Proceedings of the 21st
International System Dynamics Conference 2003, 20-24 July, New York, NY.
Pedercini, M. (2003) ‘Potential Contribution of Existing Computer-Based Models to
Comparative Assessment of Development Options’, Working Papers in System Dynamics,
University of Bergen, 1(2). ISSN: 1503-4860
Shilling, John D. (2003) ‘Can system Dynamics Flows Reach an Economic Equilibrium’,
Proceedings of the 21st International System Dynamics Conference 2003, 20-24 July, New York,
NY.
Qu, W., Barney, G., Shilling, J. and Chu, T. (2005) ‘Challenges Facing China in the Next Fifteen
Years’, in Proceedings of the International Systems Dynamics Conference, November 2005,
Shanghai, China.
Chu, T., Xu, C. and Qu, W. (2004) ‘Feedback Loops and Policy Scenarios in the Chinese Private
Vehicle Demand Model’, in Proceedings of the 22nd International Systems Dynamics
Conference, 25-29 July, 2004, Oxford, England.
Additional information is available on the Millennium Institute website: http://www.millennium-
institute.org/ in the resources section.
--17--
