Guetzkow, Harold, "Development and Utilization of Global/World Models in International Affairs", 1987

THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 209

DEVELOPMENT AND UTILIZATION OF GLOBAL/WORLD MODELS IN
INTERNATIONAL AFFAIRS

Harold Guetzkow
Northwestern University
Evanston, Illinois USA

ABSTRACT

The development and utilization of global/world models in the context
of political implications of world economic trends is overviewed in terms of
three streams of research, that of the multidisciplinary social scientists,
of the econometricians, and of the system engineers. Socio-political
processes impinge on world economic trends, just as such trends in turn
impact on politics, both national and international. The use of simulations
rich in system dynamics for the study of international affairs is found to

have much potential, as documented in findings obtained by researchers
throughout the world.

The creation and use of global models is not for the weak or the faint
of heart. Yet such simulations are exciting and important tools to
complement our understanding of some of the complex implications of world
trends. Their construction, however, taxes extant knowledge at our command
in the social sciences. Their demands for scholarly resources are
overwhelming, and they strain the patience of their users. Given
interdependence among nations, the potential of world simulations during the
next quarter-century for use in national and foreign policy-making is great,
Will the academies and institutes of planet Earth be able to accept the
challenge, thereby leading decision-makers throughout the world in the
development of global models worthy of the needs of mankind?

Research on simulations in international affairs by social scientists
stems from century-old war-gaming traditions, even though such work was
embodied only recently in person-computer format, as in our Inter-Nation
Simulation (Guetzkow 1959). In the 1960s econometricians began linking
their all-computer models of national economies into larger constructions,
as exemplified in LINK (Hickman and Klein 1985). Thanks to the initiatives
of the Club of Rome in the 1970s, systems engineers became interested in
world modeling, too, often using such computer languages as DYNAMO
(Forrester 1971) for their simulations. Today there are almost thirty
models of wide scope in existence (Siegmann 1986).

As Dina Zinnes perceptively commented, a simulation is but the
"construction of a dynamic model for purposes of: obtaining conclusions when
(mathematically) analytic results are not feasible" (Zinnes 1976: 222). In
coaching participants for their contributions to the panels on the THEME
"Global Modeling® at the 1982 meetings of the International Political
Science Association in Rio de Janeiro, coordinators Helio Jaguaribe and
Richard Merritt suggested the “central question is how scientists through
modeling techniques can develop representations of global processes and
their reciprocal interactions which are at once formally accurate, grounded
in data from the real world, and increasingly comprehensive" (Jaguaribe and
Merritt 1981). In surveying the eight models presented during the early
symposia of the International Institute for Applied Systems Analysis,
210 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

there was agreement that the "most important forces shaping the future are
social and political, and these forces are the least well represented in the
models so far" (Donella Meadows, Richardson, and Bruckmann (1982: p. xxii)
But some economists and systems engineers are joining hands with other
social scientists. As this paper demonstrates, given the advances in
sociology and political science, it is now possible to incorporate the
findings of those disciplines, too, in formal models of international
affairs.

I. THREE ILLUSTRATIVE FINDINGS ON POLITICAL IMPLICATIONS OF WORLD ECONOMIC
TRENDS (AND VICE-VERSA)

A. Tangential Political Implications from World 2/3, the World Integrated
Model (WIM), and from a Latin American World Model (Bariloche)

Researches of the industrial engineers in the United States and Western
Europe have had much visibility among intellectuals concerned with world
futures, even though their focus has been confined to the resource-economic
components of the international scene (Meads, Richardson, and Bruckman
1982). A China version of the Forrester/Meadows Model 3 was constructed
soon after the popularized The Limits of Growth appeared (Brinton,
Rosenberg, and Wolfe 1974; Wolfe 1975). Perhaps the ideological
controversies invoked by their models handicapped the engineers’ ability to
move more rigorously into political implications of the economic findings
about the ‘limits of growth® characteristic of the western capitalist world
models. Although they lament their neglect of formal inclusion of political
variables in their models, their speculation on the political consequences
indicates how even less-than-comprehensive formulations can be used as
vehicles in generating qualitative insights.

For example, in concern about the economic gap between North and South,
the World Integrated Model (WIM) was employed to check out the “Lima
Conference target that 25 percent of world industrial capacity be [located]
in the LDC’s by 2000." It was found that *in no case does WIM suggest that
the target can be realistically met." Through complementary runs of WIM, it
was revealed that were the target "to be met by international transfers
(aid, loans, investment abroad, tourism expenditures, and so on), it would
require approximately a tripling of those transfers as a percentage of
developed regions’ GNPs* (Hughes 1980: 200). Although there is no rigorous
discussion of political implications, the outcome of the study is rich in
policy significance. Given the lack of variables concerned with
international affairs, there was no opportunity for formal analysis of such
implications through the model.

The Club of Rome models were immediately challenged by a group of neo-
Marxist mathematicians operating from the Bariloche Foundation in Argentina
(Herrera, Scolnik et al. 1976). Thee content with mere polemics, these
scholars constructed “A Latin American World Model" using optimization
techniques (Figure 1). ‘Instead of centering on industrial output as related
to non-renewable resources and persistent pollution, the Bariloche model
employed “life expectancy at birth"'as its dependent variable, in terms of
such basic needs as food, housing and education. According to their
simulations, economic trends permitted fulfillment‘of basic needs in the
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 21)

early 1990s for Latin America and by 2010 for Africa. But their model, even
though extended from 1960 through 2040, predicted continuing shortcomings
for Asia (p. 93).

Figure 1 Flowchart of the Bariloche Model

Overall the political implications of the alternative formulation of
the Latin American Model were profound. If there is egalitarian income
distribution instead of contemporary configurations, they found three to
five times less GNP per capita would be required to satisfy basic needs
(Herrera, Scolnik et al. 1976: 104). They then speculated, “The obstacles
that currently stand in the way of the harmonious development of humanity
are not physical or economic in the strict sense, but essentially
sociopolitical" (p. 107). Yet their model did not incorporate a means for
checking out the "radical modifications to the sociopolitical structure of
the world" (p. 108) which might create a "new society."

It is important to realize that significant modifications can be made
in extant models without undertaking an almost complete reconstruction. The
drastic alterations by the Bariloche group to the simulations commissioned
by the Club of Rome may not be necessary. For example, by altering
parameters concerned with technological innovation in the Forrester/Meadows
World 3 simulation, Bremer (1980) was able to create a model of "bloom" from
a model widely proclaimed as a model of "doom." More recently, Akashi
(1984) created his "F-Model" version of the Forrester world model (Forrester
1971) by inserting feedback loops embodying "human ingenuity" with respect
to the "production of natural resources and the control of pollution
generation." Without changing the fundamental structure of this initial
Club of Rome model, by making parameter changes in the first instance and by
adding feedback loops in the second, the outputs were radically transformed.

B. Generation of Political Implications from LINK and GIOM through the Use
of Scenarios

The quality of the modeling of the international economic system by the
econometricians, including both its trade and monetary components, is
certified in its leadership by Nobel laureates, namely Lawrence R. Klein and
Wassily Leontief. Although the work in the United Nations had begun already
in the early 1960s within its Department of International Economic and
Social Affairs, Leontief and his colleagues later reinforced the effort
(Leontief with Carter and Petri 1977). Using the Wharton School at the
University of Pennsylvania as a base of operations, Hickman and Klein (1979)
have been able to inspire indigenous national modeling efforts by
collaborators through the world. In a most creative fashion, they then
linked some thirty-one national models together, allowing exogenous
variables for each nation to be created endogenously by other nations in the
system, as represented schematically in Figure 2.

Figure 2 Schematic Diagram of LINK System

Although no explicit socio-political variables are incorporated into
the models of the econometricians, they have used comparisons among
alternative scenarios for seeking insight into political implications of
economic trends. For example, concerned with the politics of oil, Klein,
212 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

Figure |

Flowchart of the Bariloche Model (Herrera,
pp 40 & 41)

Skotnik et al.,

1 Torai population
2 Active population secondary sector
3 Protein & calories per person
4 Urbanization rate & houses per family
5 Initial school enrollment rate
6 Active population pamary sector
J Total capieal 1980
8 Birhrate
9 Lifeexpecrancy
10 Total labour 1981
At Oprimization
42 Total capial 1981
13 Producuon of protein & calories

22 Initial school enrollment rate
23 Quality of consumer goods & services
24 Total capwal 1982,

25 Birhrte

26 Lifeexpectancy

27 Active population primary sector .
28 Tota labour 1982

29 Optmuzation

30 Total capral 1982

31 Production protein & calories

32. Urbamzanion é¢ housing

33 Macroeducation: educational places
134 Consumer goods & services

14 Urbanization & housing 35 Capital goous
1§ Macroeducation: educanonal places 36 Total population 1982

16 Ce ‘goods & 37 Active population secondary sector
17 Capral goods 38 Protein & clones per person

18 Total population 1981
19 Acuve population secondary sector
20 Procem & calones per person,

21 Urbanization tate & houses per family

39 Urbanzanon race & houses per family
40 Initial school enrollment

41 Quality of consumer goods & services
42 Total eapual 1983

1982,
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 213

Figure 2 Schematic Diagram of LINK System (Klein, Pauly and Voisin 1982, p.
11)

NATIONAL MODELS INTERFACE TRAOE MODEL

TRAOE-
SHARE
MATRIX

EXOGENOUS:
LAGS

WATIONAL,
SOLUTION

TRADE
é MODEL,

z lee ee

D

PX. PRICE CF EXPORTS,/ TO,

Tw WoRLO TRADE

Pw WORLD PRICE

3” CORKESPONDING VARIABLES IN TERMS
OF UNITED STATES DOLLARS,
214 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

Fardoust, and Filatov (1981) contrasted baseline outcomes for the 1980s with
those obtained from an alternative projection involving petroleum price
shocks analogous to those of the 1970s, as displayed in Table-1. Note how
differences in political systems (as revealed in the outcomes for the
"developed market economies" vs. the "centrally planned economies") are
reflected in the outputs for both gross domestic products and the trade
balances. The price shocks are contained by the centrally planned systems.
These same shocks produce deleterious effects within the western political
systems. Not having represented the socio-political processes within their
models, the econometricians are unable to explore specific ways in which
politics are implicated in the absorption of the shocks.

Table 1 Effect of Decade Oil Price Shocks (average annual percentage

changes in’ real terms for period 1980-1990) Klein, Fardoust and
Filatov 1982, p. 14)

Baseline Scenario
solution.

Gross domestic product

Developed market economies 34 3.0
Non-oil-exporting developing countries 31 40
Oil-exporting developing countries 58 58
Centrally planned economies 44 45

Consumer prices

Developed market economies 58
Non-oil-exporting developing countries 15.1 18.0
Oil-exporting developing countries 78

Centrally planned economies = -

Trade balance (absolute dollar values)

Developed market economies 123-2481
Non-oil-exporting developing countries -156.7  -154.7
Oil-exporting developing countries 1529 391.6
Centrally planned economies -34 13

Note: Oil-exporting developing countries are OPEC countries only.

Political implications of econometric processes, as represented in military
spending, have been touched by Leontief and Duchin (1983) in work completed
under contract with the UN Centre for Disarmament and Development and the U.
S. Arms Control and Disarmament Agency. Complementing their simulation of
economic trends from 1970 to the year 2000 with five scenarios involving
increases as well as decreases in military expenditures, along with
variations in aid, they found "it is the poorest of the less-developed
regions whose output and per capita consumption improved the most...."
Having but economic variables in their model, they concluded that ‘this
growth is explained by sizable additions to capital stocks made possible in
part by replacing military imports by shipments of machinery and other
capital goods" (p. 66). More recently, alternative simulations were
developed for the FUGI macroeconomic model by Onishi (1984). One of his
five alternative scenarios involved global disarmament, with a freeze of
defense expenditures at the 1982 level from 1983 through 1990. Although he
found that world economic trends would be accelerated generally, such
THE 1987 INTERNATIONAL: CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 215

improvement was not obtained for China and the

other Asian centraliy-
planned economies (Table 2).

Table 2 Annual Average Growth Rates of Real GDP (Onishi, 1984, per Table
5S)

ut rt

12831984 1985 BOBS 85490 BO~90 90-00
war Lo <e179 4292 312 08H 0H 19S.

VORLO ¢ EXCLUOING CPE) 2362 4353

319.207 .000

ane

- 2289 207
a€c! t3oe 1251
€c i700 Vite
OTHER ANE 2000 027
ome *S17 L763
OIL EXPORTING COUNTRIES T12t Sha.
‘OPEC IN HIOOLE EAST W156 1202
WON OIL EXPORTING COUNTRIES A7BY WAS
WI wits $830 1,064
vary 23 tynas
LATIN AMERICA MICS

Vor tlost
rays 9

ASTA
EAST ASIA e251 1.293

ASEAN

sN28 1286

OTHER ASTA ANG PACIFIC 463 lars

mtooLe EAST szos 1333
AFRICA

573 tate
e811 1,011 310

LATIN AMERICA ANO CARIBBEAN
we 1061 122.038.
‘USSR awo EAST cuRose 33

08s 1179 Lass
CHIMA AND OTHER ASIAN CPE wooo faat = faez = 001

In analog to the way in which changes were introduced by Bremer and Akashi
in the Club of Rome models, the econometricians develop their scenarios by
making changes in a constellation of parametric and minor structural
modifications, involving packages of some five to ten adjustments.

C. Creation of Political Implications through Simulation in the Social
Science Models, SIMPEST and GLOBUS

Because the social scientists at the end of the 1950s already were
attempting to incorporate political variables into their hybrid person-
computer simulations of international affairs, as in the INTER-NATION
SIMULATION (INS) at Northwestern University (Guetzkow, Alger, Brody, Noel,
and Snyder, 1959), implications of world economic trends now can be explored
directly in a rigorous manner. The transformation of the INS into an all-
computer format allowed Bremer to study experimentally the political
outcomes of variation in world economic trends (Bremer 1977: Chapter 4).
The carefully validated work of Luterbacher and Allan (1982) in the SIMPEST
simulation of the superpower triad (PRC, USSR, and USA) integrally
incorporates such effects in its findings. More recently the twenty-five
nation GLOBUS being developed at the Wissenschaftszentrum in West Berlin has

demonstrated its emerging capacity to explore political implications of
global economic trends.

During the early 1970s, SIMPEST was unusual among extant models in
constructing its superpowers with a program structured to reflect its
nation’s attributes in a differential manner (Luterbacher and Allan, 1982).
WIM and SIPER distinguish among-nations only by using contrasting parameters
216 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

within a set of programs common to all nations.
the Chinese economic sub-model
centrally-planned agricultural

to the module for the USSR, but SIMPEST then further differ
production and imports. Figure 4 presents an example of

implications explicitly simulated through the SIMPEST prograi
impact of various components,

used for portraying military capabilities to the year 1990.

Figure 3 Simplified Representation of the
(Luterbacher and Allan, 1982, p. 418)

Military Influences of other Nations

As illustrate
within SIMPEST incorpora
along with non-agricultural se

din Figure 3,
tes not only
tors, sinilar
entiates grain
the political
min terns of

including economic trends upon two indices

Chinese Economic Submodel

f Grain laports
f Tora, ee a Inports of
Exports, Inports ———————Ingustrial
— Goods from
Gross Grain Private she Nese
Output ‘Consusption
see ——— NonwAgricultucad
Production Production |
Total Non-Agricfitural \imealetaiie
Population Donestic Capital Imported
tock Capital Stock
Agricultural
Capital Stock
Urban
Rurad
Population sepalscien
H Agricultural Won-Agricultural
[PD Thvestnance Donestie ——
Investments
a
Internal Political Sector:
Satisfaction of Maoists
action of the Hasses
Figure 4 Simulation Outputs for China for Military Capabilities
(Luterbacher and Allan, 1982, p. 422, as supplied by authors)
Ios : oe somasbanycee
N
D 8
Eo.
x
a
Vv
A 2
L
vo
E
°
s
T H T T a
2965 3970 3975 3800 ises isso
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 217

Although more comprehensive social science models permit one to check
out quite explicitly the military-political consequences of economic trends,
such simulations as SIMPEST may also be used in contrasting scenarios in
which the economic components are held constant but with variation in
packaging of political variables, per se. This is in contradistinction to
the work done by the systems engineers and econometricians, whose scenario
changes are limited to non-political variables. For example, Allan and
Luterbacher (1983) were able to check out the effects of a series of some
five scenarios against a base run covering the period 1965-2000 for the USA
vis-a-vis the USSR. They contrasted a scenario involving a new period of
detente with anew period of cold war, beginning in 1981 (pp. 297-303).
Later they developed outcomes from a specific policy orientation, as when
the USA was programmed to negotiate from a "position of strength" (pp. 313-
317). The huge build-ups in arms resulting from such an American
confrontation are represented in Figure 5. The authors note that such a
scenario, however, "stimulates the U.S. economy, adding 1 to 1.5 percent to
the growth rate." In working with more encompassing models, as with INS,
SIPER, and SIMPEST, it is possible not only to consider the impact of
econometric trends upon political components, but to study also the effects
of political variables upon economic outcomes, as just illustrated.

Figure 5 Strategic Indices for the Scenario (S) of Negotiations from a-
Position of Strength

In moving from but a few nations to an assembly of twenty-five nations,
covering some 74% of the world’s population and 82% of its GNP, Bremer and
his associates (Bremer, 1983, p. 1.4) found it expeditious to use prototype
program structures to represent differences among nation-types. Thus they
allowed for parametric differences within each prototype to handle
ywariations among the countries involved. GLOBUS consists of six modules, as
exhibited in Figure 6. Four are concerned with internal processes: a
"Domestic Economic Subsector" (ECOMOD), a “Domestic Political Subsector"
(POLMOD), a "Government Budget Subsector* (GOVMOD), and a “Demographic
Change Subsector" (DEMOD). The nations are linked together via a "Trade
Policy Subsector" (TRDMOD) and a ‘Foreign Policy Subsector" (FORMOD).
Qualitatively different modules are being built for developed, developing,
and centrally planned countries.

Figure 6 The Modular Components of a GLOBUS nation (Bremer, 1983, per
Figure 1)

In a recent exercise by Cusack (1984), scenarios were developed for
exploration in GLOBUS of the political impact of world economic trends upon
politics, ‘concomitant with interactive effects of the flow of hostility and
cooperation in the international arena. Low economic growth was combined
with stabilized hostile/cooperative relations in experiment 1--and with an
unfrozen, changing international political environment in experiment 2.
High economic growth was combined analogously with stabilized and then
changing hostility/cooperation in experiments 3 and 4. Simulating outcomes
with respect to military capability for the period from 1970 through 2000
yielded contrasting patterns for different groups of nations, as for those
in NATO vis-a-vis WTO. An interesting set of outputs is displayed in Figure
7 for the South. The figure tells that "economic growth...induces expansion
while decline in economic performance leads to stagnation in the
218 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

Figure 5 Strategic Indices for the Scenario (S) of Negotiations from a
Pogition of Strength (Allan and Luterpscher 1983, p. 315)

39900004
36c9000~+
33c0c0-4
3co00004
2700004
2400004
21oc004

18000¢-4

wumerp< xmoze

1s00004
1209004
0000-

600904

us -
s00004 a rs

1985 1970 1975 1930 19g5 1990 1935 2000
YEAR
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 219

Figure 6 The Modular Components of @ GLOBUS Nation (Bremer 1983, per Figure
ty

ECOMODx

GOVMODx

Determines changes in a
nation's aggregate output,
personal consumption,
savings, prices, capital
stock, interest rates,
money supply, etc...

Determines changes in a
government's taxing and
spending policies, including
defense, education, health,
administration, and foreign
aid.

POLMODx

DEMMODx

Determines changes in a
population's support and
opposition to the govern-
ment and the government's
reaction to opposition.

Determines changes in a
nation's demographic
structure, including labor
force, school and retire- |
ment age pupulation.

TROMODx

FORMODx

Determines changes in a
Nation's .import demand,
export prices, and
import biases.

Determines changes in a
government's reactivity to
hostility and cooperation
received from others.

220 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

accumulation of arms" with exacerbation of such effects due to the
interactions with the differences in the international political environment
(p. 63). Such effects were not obtained for the North.

Figure 7 Military Capability in South (Cusack, 1984, p. 62)

c 480 Exe. 3
A
Pp
a
B
I 4
L 320
I
T
Y
I
N 260
D
E
x
200

II. SOME METHODOLOGICAL CONSIDERATIONS

Many believe global simulations, regardless of their format (be they
@Banual, hybrid, or all-computer), are merely ways of mounting in a holistic
fashion something of the complexities in our theories about international
affairs. Often military groups prefer all-manual exercises, as befit their
traditions in war-gaming. Those rooted in systems engineering and
econometrics opt for all-computer simulations, given their proclivities for
“number-crunching.* Social scientists treasure the potential for creativity
exhibited by policy-makers in person-computer hybrids, recognizing
contemporary limitations of programs utilizing artificial intelligences for

decision-making. As illustrated above in Section I-C, however, they often
are constrained to employ all-computer configurations,. given the costs
incurred when involving experienced, adult humans in person-computer

experiments.

In global modeling, leverage in understanding derivations from one s
theories about international affairs is gained by virtue of the way in which
simulations permit exploration of concomitant relationships among variables.
These variables may be taken one at a time or assembled in packages as
modules. Note how in the illustrations given above in‘Section J there was a
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 221

progression from consideration of political implications of economic growth
to the reverse relationship, highlighting the impact of political facets
upon economic growth. Simulations permit rich feed-backs among variables
and between modules, so that one may experiment conceptually with a wide
variety of outcomes, per one’s research focus. Thus, a given global model
may interrelate such outcomes as resources, energy, pollution, population, .
economic growth, quality of life, and political stability.

Some researchers regard global modeling as an overwhelming task beyond
present competencies, believing it preferable at this time to develop more
sectoral approaches, involving concern, ‘for example, with food, as
illustrated in the MOIRA model, pioneered by the Dutch (Parikh and Rabar
1981). Such work, however, may impede fuller explanation by treating many
components as exogenous, ag dramatically illustrated in the modeling work of
the government of the United States (Barney 1980). In fact, when Robinson
and her colleagues experimented by de-linking modules within the Meadows’
World 3 and Mesarovic-Pestel-Hughes’ WIM, they found distortion in outcomes
(Barney 1980: Vol. II, pp. 663-681). Similarly, employing their nation-
centered economic models in supporting policy judgments, national banks in
Latin America and some planning offices in eastern Europe found themselves
encouraged in making overly optimistic assessments of their national futures
in the 1970s because of their neglect of impingements of model-exogenous
world economic developments.

Global modeling may be viewed from an alternative perspective, in terms
of insights being generated through the use of artificial intelligence
programs. Instead of conceiving of researchers as objective agents
attempting to describe international affairs, perhaps it is fruitful to
realize their conceptions of the world may be matched with those of policy-
makers. Such are studied in research on cognitive maps of foreign policy
decision-makers (Bonham 1976). The work of Bennett with Alker (Bennett and
Alker 1977) is seminal; they employed a series of artificial intelligence
("AI") programs in tracing the decision-making of central actors for some
seventy years before and a little after the War of the Pacific (1879-1884),
It will not be long before the penetrating work of George and his colleagues
(George 1979) in the charting of “operational codes" will be ready for
mounting in computer formats, used in the simulation of cognitive maps of
policy-makers. Recently Schrodt (1986) has developed a "PWORLD" simulation,
richly endowed wth decision processes drawing extensively upon artificial
intelligence formulations.

Important in understanding the nature of global models is realization
that the outputs, as in all forms of deductive theorizing, are quite
dependent upon the substantive assumptions incorporated into the program.
At the same time, they are dependent upon the specification of parameters
used in weighting variables, as well as the assignment of initial
conditions. As mentioned above in Section I-B describing the modifications
made by Bremer (1980) and Akashi (1984) in the Forrester/Meadows Model, a
radical reversal of outputs may be obtained by changes (in these cases those
governing. innovation) in parameters and minor revision of program
assumptions, Thus, just as one can manipulate quantitative conclusions
through statistical techniques, so one can secure outcomes as desired in
simulation through selective handling of program assumptions, parameters,
and initializations. Users of global modeling, therefore, must be sensitive
222 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

to its fundamental yulnerability to political abuse, Recently Brewer (1983)
dramatized the significant costs and the unintended politico-social
consequences sometimes involved in building large-scale social systems,
drawing upon his extensive experiences in urban and military policy-related
activities. But what theory and what data in the social sciences are not
subject to such distortion when scholars lose professional integrity?

In reviewing Recent Developments in World Modeling for UNESCO’s program
entitled "Reflection. on World Problems and Future Oriented Studies,"

Heinrich Siegmann observed, "In recent years, modeling efforts have
increasingly sought to explicitly incorporate political features, the lack
of which had been a major criticism regarding earlier models.... New models

have been constructed, and equally notable, existing models have been
resurrected for modification, extension, or combination with other
models.... The data base is still considered inadequate but more, and more
easily usable data'sets have become available" (Siegmann 1986: personal
communication). An exhortatory description of the move toward the inclusion
of political components can be found in an essay, “Toward Integrated Global
Modeling" (Ward and Guetzkow 1979). Persons knowledgeable about
global/world modeling regard the GLOBUS model of the Wissenschaftszentrum in
West Berlin (Bremer 1985). as having incorporated more .domestic and

international political Processes in its formulation than any other
simulation to date.

III, IMPERATIVES FOR WISE SCHOLARS ~

The deliberation of the Sixth Symposium on Global Modeling convened by
the International Institute for Applied Systems Analysis in 1978 are replete
with sage observations for scholars ready to undertake global modeling
(Meadows, Richardson, and Bruckmann 1982: Chapter 5). Wise scholars

embarking on practical development of global modeling may wish to consider
the following imperatives:

--begin modestly, but in a holistic manner, allowing your model to
evolve early with feedback loops among its modules as its
complexity increases.

Exemplar: the evolutionary synthesis of . two macroeconometric
models articulated with a micromodel of mineral resources as
developed by Kaya and his associates (Kaya and Onishi 1980).

--collaborate early on with your potential users, so they may
thoroughly understand the limitations and potentials of your
construction as a complementary tool in their decision-making.

Exemplar: the close collaboration Gvishiani and his associates
maintain with government decision-makers, in that their Institute
for Systems Studies is sponsored by both the Academy of Sciences
and the State Committee on Science and Technology of the USSR
(Gelovani 1981). :
~-subject your modeling to competition, within your own team as
well as in confrontation with the models developed by others.
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY, CHINA 223

Exemplar: the comparative work among models being executed by the
Documentation Centre of Development-Policy Modelling of. the
Systems Studies Institute in Pune, India (Krishnayya 1984).

End Notes:. In the fall of 1983, Dr. Gerald 0. Barney and Patricia Maimon-
Music began editing the GLOBAL PERSPECTIVE QUARTERLY. In this newsletter
the editors "share a belief in the need for a global perspective to enable
us to identify and act on the many problems and possibilities before us.
Toward that end we will be sharing a wide range of information and ideas.
Each one of us needs to be able to organize diverse information into an
integrated, coherent whole, and have the capacity to view developments,
events and possibilities in their global context and their relative
importance" (Barney and Maimon-Music 1983: p. 2). One may subscribe to the
GLOBAL PERSPECTIVE QUARTERLY through the Global Studies Center, 1611 N. Kent
Street, Suite 600, Arlington, VA 22209 USA. The Center has also recently
issued its Managing a Nation: The Software Source Book.

This paper is a revised and updated version of a contribution made to the
Seminar on "Political Implications of World Economic Trends" organized by
the Institute of World Economics and Politics in the Chinese Academy of
Social Sciences (CASS) of the People’s Republic of China with the U.s.
National Science Foundation’s (NSF) Delegation to the NSF-CASS Joint Program
Development Workshop in International Studies in Chongqing, October, 1984.
A Chinese translation of the original essay may be obtained from Professor
Harold Guetzkow, Northwestern University/Scott Hall, Evanston IL, 60201 USA.
224 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

REFERENCES

Akashi, H. (1984), “Prediction of the Future with Feedback by Man," in
Chestnut, H., R. Genser, P. Kopacek, and A. P. Wierzbicki, eds.
Supplemental Yays for Improving International Stability, Proceedings of
the IFAC Workshop, Laxenburg, Austria, 13-15 September 1983. Pergamon
Press, Oxford, pp 251-255.

Allan, Pierre and Urs Luterbacher (1983), "The Future of East-West
Relations: A Computer Simulation of Five Scenarios," Appendix AS in
Daniel Frei and Dieter Ruloff, East-West Relations, Vol. 2:
Methodology and Data. Oelgeschlager, Gunn and Hain, Cambridge, MA, pp
285-318

Barney, Gerald 0. (1980), The Global 2000 Report to the President of the
U.S.: Entering the 21st Century, a report in three volumes prepared by
the Council on Environmental Quality and the Department of State. Vol.
1, The Summary Report, 47 pp; Vol. 2, The Technical Report, 766 pp;
Vol. 3, The Government’s Global Model, 301 pp; U.S. Government
Printing Office, Washington,

Barney, Gerald 0., and Patricia Maimon-Music, eds. (1983), "Editors’
Column," Global Perspective Quarterly, Vol. 1 No. 1, pp 1-16

Bennett, James P., and Hayward R. Alker, Jr. (1977), *When National Security
Policies Bred Collective Insecurity: The War of the Pacific in a World
Politics Simulation," in Karl W. Deutsch, Bruno Fritsch, Helio
Jaguaribe, and Andrei S. Markovits, eds. Problems of World Modeling:
Political and Social Implications, Ballinger, Cambridge MA, pp 215-302

Bonham, G. Matthew (1976), "Cognitive Process Models and the Study of
Foreign Policy Decision-Making," in H. Bossel, S. Klaczko, and N.
Muller, eds. Systems Theory in the Social Sciences, Birkhauser Verlag,
Basel, 552 pp

Brener, Stuart A. (1977), Simulated Worlds: A Computer Model of National
Decision-Making. Princeton University Press, Princeton NJ, 249 pp

Bremer, Stuart A. (1980), "Technologischer Fortschritt und die Grenzen des
Wachstums,* in Stuart A. Bremer, Rolf Kappel, Peter Otto, Hannelore

Weck, and Ulrich Widmaier Weltmodellstudien: Wachstumsprobleme und
Losungsmoglichkeiten. Verlag Anton Hain, Konigstein/Ts., pp 21-92

Bremer, Stuart A. (1985), "The GLOBUS Model: History, Structure, and
Illustrative Results," in Michael Don Ward, ed. Theories, Models, and
Simulations in International Relations, Westview Press, Boulder, CO, pp
39-77

Brewer, Garry D. (1983), "Some Costs and Consequences of Large-Scale Social
Systems Modeling," Behavioral Science Vol. 28 No. 2, pp 166-185

Brinton, William, Gerald H. Rosenberg, and John Wolfe (1974), "CHINA1: A
Model of Growth for the People’s Republic of China," Simulation, Vol.
23 No. 5, pp 131-141
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 225

Costa, Antonio Maria and L. Peter Jennergren (1982), “Trade and Development
in the World Economy: Methodological Features of Project DYNAMICO,"
Journal of Policy Modeling, Vol. 4 No. 1, pp 3-22

Cusack, Thomas R. (1984), "The Evolution of Power, Threat, and Security:
Past and Potential Developments." Wissenschaftszentrum, Berlin, 69 pp

Forrester, Jay W. (1971), World Dynamics Wright-Allen, Cambridge MA, 142 pp

Gelovani, Viktor (1981), "An Integrative Modelling System as a Tool for
Analyzing Complex Socio-Economic Problems," Impact of Science on
Society, Vol. 31 No. 4, pp 413-422

George, Alexander L. (1979), "The Causal Nexus Between Cognitive Belief and
Decision-Making Behavior: ‘'The Operational Code’ Belief System," in
L.S. Falkowski, ed. Psychologic: lodels ternational Politics
Westview Press, Boulder CO, pp 95-124 i

Guetzkow, Harold (1959), "A Use of Simulation in the Study of Inter-Nation
Relations," Behavioral Science, Vol. 4 No. 3, pp 183-191

Guetzkow, Harold, Chadwick F. Alger, Richard A. Brody, Robert C. Noel and
Richard C, Snyder (1963), Simul: nm International Relations:
Developments for Research and Teaching, Prentice-Hall, Englewood Cliffs
NJ, 248 pp

Herrera, A., H.D. Scolnik, G. Chichilnisky, G.C. Gallopin, J.E. Hardoy, D.
Mosrovich, E. Oteiza, G.L. deRomero Brest, C.E. Suarez, and L. Talavera
(1976), Catastrophe or New Society? A Latin American Worl. jel,
International Development Research Centre, Ottawa, Canada.

Hickman, Bert G., and Lawrence R. Klein (1985), "Recent Developments in
Project LINK," Social Science Research Council ITEMS, Vol. 39 Nos. 1-2,
pp. 7-11

Hughes, Barry B. (1980), World Modeling : The Mesarovic-Pestel World Mo¢
in_the Context of Its Contemporaries, Heath, Lexington, MA, 228 pp

Jaguaribe de Mattos, Helio, and Richard L. Merritt, (1981), "Search for a
New International Order," Participation, Vol. 5 No. 2, pp 28-29

Kaya, Y., and A. Onishi (1980), "FUGI--Future of Global Interdependence," in
Gerhart Bruckmann, ed. Input-Output Approaches in Global Modeling,
Proceedings of the Fifth IIASA Symposium on Global Modeling, IIASA
Proceedings Series 9, Pergamon, Oxford, pp 91-357

Klein, L-R., S. Fardoust, and V, Filatov (1981), "Oil Price Simulation with
Fixed and Flexible Exchange Rates in the Decade of the 1980s" in Klein,
Lawrence R., Peter Pauly and Pascal Voisin (1982) "The World Economy--A
Global Model, Perspectives in Computing, Vol. 2 No. 2, pp 13-i4
226 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA

Krishnayya, J.G. (1984), *Regional Information and Documentation Centre for
Development-Policy Modelling," Systems Research Institute, Pune, India,
one page announcement.

Leontief, Wassily with A.P. Carter and P.E. Petri (1977), The Future of the
World Economy, Oxford University Press, New York, 110 pp

Leontief, Wassily and Faye Duchin (1983), Military Spen Facts and
Figures, Worldwide Implications and Future Outlook, Oxford University
Press, New York, 124 pp

Luterbacher, Urs, and Pierre Allan (1982), *Modeling Politico-Economic
Interactions Within and Between Nations," International Political
Science Review, Vol. 3 No. 4, pp 404-433

Meadows, Donella, John Richardson, and Gerhart Bruckmann (1982), Groping in
the Dark: 's' jade _of Global lelling. John Wiley and Sons,
New York, 311 pp

Onishi, A. (1984), “Alternative Simulations of the FUGI Global Macroeconomic
Model," in Chestnut, H., R. Genser, P. Kopacek, and A. P. Wierzbicki,

eds. Supplemental Ways for Improving International Stability,
Proceedings of the IFAC Workshop, Laxenburg, Austria, 13-15 September

1983. Pergamon Press, Oxford, pp 67-76

Parikh, Kirit and Ferenc Rabar, eds. (1981), Food for Al, Sustai
World: The ITASA Food and Agriculture Progran international Institute

for Applied Systems Analysis, Laxenburg, Austria, 250 pp

Schrodt, Philip A. (1986), "PWORLD: A Precedent-Based Global Simulation,"
Northwestern University, Evanston, IL, 18 pp

Siegmann, Heinrich (1986), WZB: Wissenschaftszentrum Berlin (1986),
"Modelling: An Overview," Chapter 6 in United Nations Educational,
Scientific, and Cultural Organization, Pere: s_and Analys
World Problems (Major Programme I (Reflection on World Problems and
Future-Oriented Studies) Synoptic report 1984-1985], UNESCO:
Workshops, Paris, France, 123 pp

Wolfe, John R. (1975), "A Technical Report of CHINA1: A Model of Growth
for the People’s Republic of China." Dartmouth College Research
Program on Technology and Public Policy, Hanover NH, 36 pp

Zinnes, Dina A. (1976), Contemporary Research in International Relations; A
Perspective and a Critical Appraisal, Free Press, New York, 477 pp