THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 633
ECONOMIC GROWTH AND DEVELOPMENT POLICY
IN OIL DEPENDENT INDONESIA
4 2
M. Tasrif Arif and Khalid Saeed ©
ABSTRACT
This study uses a system dynamics model to understand the
Process of economic growth in the oil dependent economy of
Indonesia. Many long-run growth patterns resulting from the
various intuitively appealing development policies are analysed
and an attempt is made to identify the best policy set for
attaining 2 sustainable growth pattern. The study shows that
influencing factor prices in a way to facilitate adoption of
capital intensive technologies increases acceleration of growth
and is a key policy to sustain growth in the long run
4, INTRODUCTION
This paper examines the oil dependent economic growth in
Indonesia over the past and explores development policy options
to sustain growth over the next 50 years over which oil
reserves are expected to be consumed. A generic system dynamics
model is constructed to analyze the problems of transition
from an oil-dependent economy to an independent one. The model
is based on the macroeconomic growth principles, while it also
incorporates the microeconomic market clearing mechanism, A
set of simulations with this model is used to examine
alternative policies for achieving a smooth transition from an
oil-dependent economy to an oil-independent one.
A policy history analysis precedes policy analysis in this
study. Such analysis improves confidence in the model and
lends itself to acquiring a good understanding of the system
under study, The policy design emphasis is also on influencing
the day to day decisions of the participants in the system
instead of intervening to fight internal trends. . It is
therefore, necessary to concentrate on elucidating decision
making process itself instead of simply using behavioral
equations prescribed by the economic theory as is done in case
of econometrics (Tomkins, 1981). Hence the model used is
somewhat large. The general structure of this model is
discussed in Section 4 of this paper. Further technical
details and a machine readable listing of .the model are
available from the authors on request
Development planning has traditionally required large
scale interventions by the government. Many of such
interventionist policies have been implemented in Indonesia
1, Assistant Lecturer, Electrical Engineering Department,
Bandung Institute of Technology, Bandung, Indonesia
2. Associate Professor, Industrial Engineering & Management
Division, Asian Institute of Technology, Bangkok, Thailand.
634 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY, SEVILLA, OCTOBER, 1986
(Lewis, 1984), These include (1) increasing government
spending, (2) encouraging investment, (3) controlling foreign
trade, (4) subsidizing domestic consumption of oil, and (5) to
some extent, controlling factor prices
This study, however, suggests that, in some sense, non-
intervention is the best policy. The tendency of the economy
to maintain a sustained growth pattern appears to be
facilitated by the flexibility of production technology and an
unencumbered working of the market clearing mechanisms.
Influencing factor prices ina way as to increase capital
intensity further helps since it strengthens the accelerator
mechanism and consequently also creates, larger multiplier
effects.
2. THE INDONESIAN ECONOMY: AN OVERVIEW
Under the so called New Order (which signifies the
assumption of political power by the new regime after a
military coup d'etat in 1965), the average rate of economic
growth in Indonesia has been almost 8 percent over the period
1969-1981. This translates into a real per capita growth rate
of over 5 percent (McCawley, 1983). As compared with per
capita growth rate of -0.2 percent during the period 1960-1965
(the period of so called guided democracy), such a rate of
economic growth is high, even by contemporary Asian standards,
and much higher than that achieved in Western Europe during the
nineteenth century or in Japan between 1867 and 1914 (Arndt,
1975).
This rapid growth has led many observers of the Indonesian
economy to study the mechanisms of its growth during 1970s. It
is generally believed that such rapid growth occurred largely
because of revenues from export of oil. Indeed, petroleum and
gas currently account for 75 percent of total exports and 60
Percent of the government's budgetary outlays (Pitt, 1985).
Although energy consumption in Indonesia is low, even by
developing country standards, it has grown very rapidly over
the past decade. Commercial energy consumption grew at a rate
of 14.7 percent per annum during 1972, which is among the
fastest rates of growth in energy consumption in the world and
is almost twice as fast as the GNP growth rate in Indonesia
(Pitt, 1985). In absolute terms, domestic oil consumption
accounted for about 124 million boe* in 1976 which was about 67
percent of the total commercial energy consumption (Energy
Planning For Development in Indonesia, 1981).
Undoubtedly, a major factor behind the rapid growth in
domestic oil consumption has been the subsidization policy of
the government. Under the New Order, subsidies on the domestic
sale of refined oil products have become a sizeable item in the
State Budget since late 1970s. In the early 1970s, the price
of domestic oil products was higher than what it cost to
produce these (Dick, 1980).
*
barrel oil equivalent
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986. 635
The increasing dependence of the economy on oil export
revenues, however, must be viewed with some alarm since oil
resources are finite and non-renewable. When oil runs out the
economy might experience a sudden and deep recession before it
can make a transition into an oil-independent state.
3. PAST BASIS OF DEVELOPMENT POLICY
It is widely recognized that if population is growing
real income must also grow. Its failure to do so, means a
reduction in living standards to the point where the population
ceases to grow (Hamberg, 1956). Thus, models of economic
growth have become indispensable to designing strategies for
the less developed countries. These models are further
supplemented by the humanitarian concerns and moralization as
bases for development policy. Unfortunately, the policies
issued under such bases often require large scale government
intervention into fighting the internal tendencies of complex
social systems, which is often economically wasteful and
politically dysfunctional (Saeed, 1985),
The relatively simpie Harrod-Domar model which is best
known among the various economic growth models, has been the
main instrument of policymaking in the past. According to this
model the basic remedy to the development problem is simply to
increase resources to be invested. Two policy implications of
this popular model have been to increase domestic savings and
foreign aid (Bhagwati, 1984). However, a particularly crucial
assumption. of the Harrod-Domar growth model is that production
takes place under fixed factor proportions. There is no
possibility ‘f, substituting labor for capital in production
(Solow, 1956). This assumption means that any excess of one or
the other is wasted. This assumption is in conflict with the
idea of clearing markets which is based on the notion that
private markets function efficiently
The idea of market clearing is also closely related to the
optimizing behavior of individuals, although, within a bounded
information set. People determine their individual choices of
work, consumption, and so on, in order to make themselves as
well off as possible. Market clearing also reflects the idea
that the people who participate in and organize markets - and
who are guided by the pursuit of their own interest - do not
waste resources, and thereby end up achieving efficient
outcomes (Barro, 1984). Last but not least, if policy
intervention is to be indirect, it must aim at affecting the
day to day decisions of the people, not the final outcome of
those decisions. Thus, a policy model concerning economic
growth must incorporate these day to day decisions which
determine how markets are cleared.
4. A SYSTEM DYNAMICS MODEL OF ECONOMIC GROWTH IN
AN OIL DEPENDENT ECONOMY
The main model of this study is a single sector two
factor system which incorporates national income accounting at
636 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
an aggregate level and mechanisms to determine price level
wage rate, interest rate, and technological mix. To this is
added an oil production and consumption system representing the
oil sector of the economy.
Following are the key assumptions of the model:
4. There. is only one non-oil commodity, - whose rate of
Production depends on the potential production rate
and the capacity utilization factor. Potential
production rate is modeled as a function of capital
and workers, whereas capacity utilization factor
depends on the short run aggregate demand and
inventory condition.
2. Inventory is introduced into the economic growth
submodel as a level containing the physical
accumulation of goods in the economy. It is increased
through production and imports, and depleted through
capital investment, government purchases, consumer
Purchases, and exports.
3. Excesses in inventory can be exported, while shortages
in inventory can be met by imports. Volumes of both
exports and imports can be controlled through
specified policy instruments, There is no limitation
outside of the country for exports and imports.
4, Financial markets are cleared through changes in real
interest rate. Fractional gross profits, tax,
propensity to consume, population growth rate, export
price of oil, electricity price, coal price, and
subsidy on domestic oil consumption are all
exogenously determined depending on government policy.
Government spending is also exogenously determined.
5. Desired oil exports are endogenously determined on
basis of oil reserves and the need for oil revenues,
but can be modulated by oil export policy.
6. Two main components of the oil submodel are the
potential oil production and the desired oil
production, Potential oil production is determined by
capital stock in this sector and its productivity.
The model considers capital as the sole production
factor input for producing oil, since oil sector is
highly capital intensive. Oil production is used for
domestic consumption and exports. Revenues from oil
exports are assumed to be used for importing goods
into the economy,
The following subsections describe briéfly the causal
relationships incorporated in the model and growth patterns
generated by simulating it.
4.1. Growth Mechanisms of the Model
The following well-known identity represents income
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 637
accounting in the model.
Y=sC+tI+G@+¢ X- mM) a)
where :
¥ income;
c consumption
I investment;
G government spending;
x value of exports; and
M value of imports
The growth mechanisms are embodied in the feedback
loops representing the multiplier-accelerator principle first
propounded by Samuelson (Forrester, Nathan B. ,1982). However.
instead of assuming presence of a market equilibrium all along
the growth path, the model also incorporates the micro-level
responses of the producers and consumers to changes in market
conditions since market clearing is the natural macrocomplement
to the microfoundations that underlie the model (Barro, 1984)
Figure 4.1 illustrates the main causal-loops of the
economic growth submodel.
Income represents Gross National Product (GNP) at constant
prices and is given by multiplying inventory outflow with
equilibrium price. Inventory outflow constitutes consumer
purchases, capital investment, government purchases, and net
exports in terms of goods, When oil sector is also connected
to this system, income is equal to inventory outflow times the
equilibrium price plus domestic oil sales (in real terms).
638 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986.
wanant cartacissort wat
carta + Ls asm
aetiet er
sntarveraent
nan
ae tt
nate
.
ie ae
a se 1
ab
ner rxroRy
F RATE FOR
Seseeo
a t.
covernment Gunvexroer TE mv enrony,
wea"
cara
aa
moicarto
roncnasts
+ +
tweome SALES
wa
smoormes
wmcomt
vt
savines 4 consumption
savesrineot 5. anes
+t,
pesiato omnvestto
Savincs
Fig. 4.1 Main Causal Loops in the Economic Growth Submodel
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 639
The multiplier is represented through a positive feedback
loop which is coupled with a negative feedback loop, as seen in
Figure 4.2.
+ +
Income €——— Inventory ~~ _ Inventory
outflow usage
+ +
Smoothed (+) Se)
income
+ + - +
Indicated Consumer Desired
consumer purchases inventory
purchases
“Fig. 4.2 The Multiplier Loop
The negative loop represents the effect of inventory
availability on consumption. The effect of wage rate on
consumption (as a budget constraint) is not considered.
The accelerator mechanism is represented through one main
positive feedback loop which is coupled with three
supplementary positive feedback loops, as seen in Figure 4.3
However, these positive feedback loops are coupled with several
negative feedback loops which are created by ‘the market
clearing mechanisms
Figure 4.4 shows the behavior of the model when it is
disturbed from equilibrium by stepping up government purchases.
In this experiment market is assumed to remain in equilibrium
and population is kept fixed. Also, oil sector is kept out of
the picture.
As seen in Figure 4.4, due to a step increase in
government purchases, income rises, which increases consumption
expenditure. Rising government purchases deplete the level of
inventory while they increase the shipments. Together, these
effects increase desired production, which fuels investment.
After some period of time as seen in Figure 4.4, income
levels off at its new equilibrium which is higher than its
initial value. Moreover, due to an increase in desired
production in the face of a fixed population, the unemployment
rate decreases to a lower value in the new equilibrium.
4.2 MARKET CLEARING MECHANISMS
Four market clearing mechanisms are built into the model
These are interest rate, general price level, wage rate, and
technological mix indicated by the capital labor ratio
Interest rate balances investment and saving rates which are
decoupled by a pool of uninvested savings. Uninvested savings
in excess of a level necessary to support desired investment at
640 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
Capical +
deoreciacion
Desired
site Production
(——+Capica.
° + +
4
LTavencory
+) oa
capical + Invencory 4h Inveacory
— ae —_—
inves cmenc oucélow usage.
ns
+
(oo) Desired.
invencory fos)
ee
Capical Sales
on orders = | e
+ Consumer Indicated
purchases production
race
CH) caoi
de te orde:
+ Capical Savings +———— Consumo cion
—— orders
+ L +
Cniavesced =, Incéresc =, Desired
savings race capical
Fig. 4.3 The Accelerator Loop
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 641
* eee
4 AMAA
pasa
ea
Wage réfe
4
an ee Hi i or
84 ; General pfice level
'
3
Pad
oe linterést tate ! 1 ,
Ann yens
Oe RE cere rere jenny evden (peerenenl oman
3°3'333
2 3 233 [Oe en a ne res te DDS oe
| covernned purchases ! !
f] ‘
1 t
| Unemployngat rate
senusz20 bob
ass ; z ; ; ; ; :
33 2 3 3 3 3 3 3
22 2 2 2 3 2 2 3
2 Ss as 5 a 2 5 2
2 2 8 5 a 2 8
Fig. 4.4 Behavior of Economic Growth Submodel with Assumptions
of Market Equilibrium and Fixed Population
current cost of capital, exerts a downward pressure on interest
rate, which encourages further investment. Interest rate is.
also influenced by price. An increase in price raises money
demand to support existing exchanges of goods which raises the
interest rate.
General price level adjusts towards a desired level at
which inventory is adequate to support shipments out of it
while it affects sales, consumption, and investment.
Wage rate clears the labor market. An increase in the
unemployment rate depresses wage rate. A depressed wage rate,
in turn, will increase the demand for workers. Subsequent.
hiring of workers restores balance between workers and the
unemployed.
Finally, the study also assumes that in the long run, the
optimal technology mix can be adjusted by using relatively
flexible production methods. This assumption allows for
substituting workers for capital in production without
compromising on production efficiency. The study assumes that
the optimal capital-labor mix will rise when real wage rate is
higher than the average value of the marginal productivity of
642 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
workers. Presence of such a technological flexibility.
however, depends a lot on how the government deals with the
import of technology. Thus, the process of clearing of the
technology market can be considered as a policy variable
Figure 4.5 shows the behavior of the model repeating the
“first experiment while the assumptions of fixity of price
interest rate, wage rate, and the optimal capital-labor ratio
are relaxed. As seen in Figure 4.5, income rises, and levels
off as in the first experiment. Price rises at first due to
the inventory shortage, and then declines converging to its
equilibrium value. This means that the price mechanism can
balance supply and demand of goods. Wage rate also increases
and levels off at a higher level as compared with its initial
value This increased wage rate is coterminus with a slight
decline in the unemployment rate (not evident in the figure due
to compression of scale).
Wage tate “Kincome | Gaptthi Actor ratio
eR ER ee
we WU General price level '
T
q
a) ‘ ' '
1 Tvinterest, rate
t
120.2
ee ie ee ee
2
2D
: 4 1 1 f '
Ioy 1 Governmeat purchases '
Beone H il fl : re! t
fas sas eT i y fons 7
Roowomrey |g ' : i
N . Unemployhent rate
! 1 ' ‘ f
j 1 1 1
L
2040.
2us0.
2070.
2u9u.
Fig. 4.5 Behavior of Economic Growth Submodel with Fixed
Population but with Market Clearing Mechanisms
Working
TRE 1900 INTERNA TIUNAL CUNPFERENCE UF TRE OTOTEM UINANILS OUUIELT. SEVILLA, UUIVBEN, 1900 OM
The increased wage rate raises the optimal capital-labor
ratio, as seen in Figure 4.5. It means that the constant
fraction of output accruing to capital is higher than its
initial value. In other words, the economy becomes more
capital intensive compared with the first experiment.
Consequently, at the new equilibrium, the interest rate is
higher than its normal value. It should be noted that the
unemployment rate is lower than in the first experiment since
increased capital intensity creates bigger acceleration and
also raises subsequent multiplier effects.
4.3 POPULATION GROWTH
When population is allowed to grow exponentially and the
increases are added to the pool of the unemployed, wage rate is
depressed This raises the demand for workers When
technological adjustments are possible, the reduced wage rate
decreases the optimal capital-labor ratio, which further fuels
worker hiring.
Figure 4.6 illustrates the model behavior repeating the
*
4 a aa
3 go ae
: I Leer
woveDdee G8 8 aoee nen pimps = Se
S. G & gaan ;
perth 555335555535
3323 °333
25° 2 222
c 2
3
® aie
343
238
3A 4
?
. ‘ j
4 ALS General |
33°3°3 Prec! 7
fn 7% level Y
7
He
Ao ‘ ~---
a9 3
at a Interest
‘ rate,
\
nepeuaseeibs soe §
33°38 q * é ‘ ; : ‘
33 2 3 3 3 3 3
RF 2 R = cs 3 = B
2 > z 3 3 a 5 3
5 2 g Fd g 8 rf
Fig. 4.6 Behavior of Economic Growth Submodel with Growing
Population and with Market Clearing Mechanisms
Working
644 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
second experiment, but without an exogenous step in government
spending, while an exogenous population growth is also
introduced, As described above, the rise in population raises
the pool of workers, thus increasing production capacity.
Increased production raises the level of inventory which
depresses price level. This reduced price is also followed by
a decline in interest rate since the investment rate is less
than the saving rate. The excess supply and the lower level of
capital-labor ratio do not encourage further investment
Therefore, more and more capital is substituted by labor. This
also limits acceleration while multiplier effects and aggregate
income decline. Thus, unemployment rate becomes high in spite
of a low wage rate
4.4 The 0i1 Submodel
The oil submodel determines potential oil production and
desired oil production. The potential oil production is given
by multiplying capital in the oil sector by the productivity of
this capital. The oil capital increases through capital
investment in this sector which is modeled as a function of
desired oil production and the forecast oil productivity of
capital. A decline in oil productivity raises the capital
investment needed to produce the same amount of oil. This
productivity, in turn, depends on the level of oil reserves
Figure 4.7 illustrates. the main causal-loops of the oil
submodel
Desired oil production is the sum of desired oil
production for domestic consumption and desired oil production
for exports. Desired oil production for domestic consumption
is determined through its forecast value. Desired oil
production for exports is a function of the ratio of oil export
revenues to income. When oil export revenues decline, there is
a tendency to export more oil in order to offset the declining
revenues and vice versa.
The domestic demand for oil is modelled as a function of
income, population, and the relative domestic oil price which
is oil price to the consumers divided by average energy price
The average energy price is an average of electricity, coal
and oil prices to consumer. The electricity and coal prices
are determined exogenously.
Oil price to consumer is domestic oil price or oil export
price (when oil import occurs) modulated by subsidy or tax.
The domestic oil price adjusts to its indicated value which is
a function of oil productivity of capital, life of oil capital
interest rate, and general price level.
Figure 4.8 shows the behavior of the model when the oil
submodel is coupled with the economic growth submodel. This
experiment also assumes government spending to be a fixed
function of average income and allows population to grow, In
this case, all market clearing mechanisms are assumed to be
working.
DE SOU RN Ler we aee e ERENCE UE TS STO LEAN S OVE LT SEVIELA, Uv Eben, [SOS CMR,
Subsidy on domestic
oil price Tacome
- |.
Oil price =, Domestic demand +
—
to consumer for oil Populacion
+
Interest
rate
+, Domestic oil Oil export
price + price
General * * Domestic oil
price | consumption
level - .
+
Oil exports
+
+
Desired oil
produccion
for domestic
consumption
Oil productivity:
of capital
+
a
Capital oil
sector
at
Capital investment
oil sector
+
+, Desired oil +
Oil reserves +——-0il production
Oil potential
production
+
+
Oil export
revenues
Desired oil production
for oil exports ———J
T+
Income
Fig. 4.7
production
Main Causal Loops in the 0il Submodel
646 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
aa » og 2-4 Pa
aS = } 3a ct
se NM cert . < .
HO FOHH 24M SHSM NM MAE ARR
BDHOeeSDH*# 505595. SBDN SHN NHN MH RHA
ENS TNGHHa
tule ee eee
yea p222
as ¥535
N c3en
Q
jwuu.t
General Brice revel
' 1 f
=a ast == NU our exper ~~~ |
ue
ener gh Naw. ts
3°38 34 recount:
1 L 1 1 ye ita |
‘ Interest rate 7nceue ; Ancemecner (capil Wage
1 i
DeHDIDIHDH Tate.
2 2 33 2 x 4 5 3 3
3 48 3 3 2 2 8 a
2 5 g 3 3 5 y 3
i) cod - sal N N N N
Fig. 4.8 Behavior of Complete Model Incorporating Economic
Growth. Market Clearing, Population Growth and Oil
Sectors
As seen in Figure 4.8, income grows, however, the income
growth rate is less than the population growth rate. This is
represented by a decline in income per capita throughout the
simulation. General price level, interest rate, wage rate, and
optimal capital labor-ratio are also depressed. Unemployment
rate rises since the growth in income is unable to absorb the
additional labor supply. Net exports are positive indicating
that there is excess inventory. Propelled by the need to
support higher outlays as income rises, oil exports rise but
are limited by the declining oil reserves toward the end of the
simulation
5. THE SEARCH FOR AN APPROPRIATE GROWTH STRATEGY
Since different growth strategies may yield different
growth patterns, the pattern of growth sought is the main basis
in determining the alternative strategies of growth (Lewis
1984).
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 647
The pattern being sought in our analysis is one which
might overcome the problems of transition from an oil-
dependent economy to an independent one. The appropriate
strategy, in the long run, has to maintain a sustained growth
and encourage the country to become self-reliant. It is also
important that the instruments of change fall within the
purview of a normal interventionist role of government and not
call for suppressing laissez faire. Last but not least, these
policies may not transfer costs of growth, into the future.
This form of intervention can be easily realized if the entry
point for the government is the market. Examples of policies
to influence market are establishment of a minimum wage rate
for workers or a ceiling on interest rates (Herrick &
Kindleberger, 1984),
We experimented with the following well known policies
using the model to understand how they propagate through the
system: (1) increasing government purchases, (2) encouraging
investment through cutting taxes, (3) encouraging investment
through increasing saving propensity, (4) restricting imports
and expanding exports (trade policies), (5) supporting wage and
interest rates, and (6) subsidizing domestic oi1 consumption.
The policy of increasing government purchases is simulated by
stepping up the fractional government purchases to 20 percent
of average income as compared with 10 percent in the Base Run.
The policy to encourage investment through tax cuts, is
implemented through stepping down the fractional gross profits
tax by 50 percent. The policy to encourage investment through
increasing saving propensity, is implemented by reducing the
propensity to consume by 10 percent. Trade policies include
making the foreign trade respond faster to changes in excess
inventory, introducing 10 percent positive non-oil exports, and
doubling oil exports. Support of wage rate is implemented by
holding it constant at initial value. Oil subsidy is simulated
by stepping down the oil price to domestic consumers by 50
percent. This policy has been seen as one in which the poor
benefit from the oil bonanza. However, several studies have
suggested that it helps the higher income group considerably
more (Down, 1983). The Base Run is made with fractional
government purchases at 10 percent of average income,
population growth rate at 2 percent per year, and absence of
all development policies.
We also explored a set of policies which indirectly
influence factor prices and technological mix. These include
(4) stepping down interest rate normal (long-run interest rate
by 50 percent, (2) stepping up wage normal (long-run wage
rate) by 50 percent, and (3) increasing gradually the goal of
capital-labor ratio to be almost twice its initial value by
2030.
The simulation results are shown in the following tables
giving the figures of income per capita, unemployment rate
value of net exports, and capital-labor ratio respectively in
2030. Those values in 1970 were respectively 43.82 thousand
money units/year/million persons, 10 percent, nil, ard 130.4
thousand capital units/million persons. This initial capital-
labor ratio and other parameters, provide the elasticity of
648 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
capital of 0.17 and the elasticity of workers of 0.83. As
compared with the US economy in which the elasticity of capital
is 0.25 (Forrester, Nathan B., 1982), the Indonesian economy
was more labor intensive in 1970
Table 5.1 shows the simulation results of Base Run and six
development policies mentioned above,
When development policies are absent (Base Run) the
introduction of flexible interest rate and technology in
addition to flexible prices and wages not only decreases income
per capita, it also makes the system more labor intensive while
increasing unemployment rate. When the indicated development
policies are implemented, the presence of such flexibility
improves behavior mainly for the first, the second, and the
fourth policy, while it worsens behavior for the rest
Apparently, the presence of such a flexibility in the face
of population growth makes the system move toward greater labor
intensity which is concommitant with a considerable weakening
of the accelerator and multiplier effects. However, when
demand is stimulated, the same flexibility allows capital
intensity to increase which strengthens multiplier and
accelerator effects
It should be noted that stimulation of investment only at
the cost of limiting demand greatly weakens the growth engine
while supporting ‘wage rate also greatly discourages
investment, Subsidization of domestic oil consumption produces
results similar to the base runs while it also limits exports
A comparison of the figures of Table 5.1 for the various
computer runs shows thatthe capital intensity in the economy
must rise for increasing its growth potential through
multiplier and accelerator effects, while at the same time
demand must remain high to stimulate investment.
Table 5.2 shows the simulation results of selected policy
runs when capital intensity is increased autonomously to almost
twice its initial value in 1970. The policy of encouraging
investment through tax cuts while allowing complete flexibility
of the model gives best results since it generates highest
multiplier and accelerator effects. Such a policy may also.
in the long run, lead the country to self-reliance. On the
other hand, the policy of increasing government purchases may
create large deficits and trade policies may increase
dependence on the international markets, even if they are able
to create some propulsion in the economy.
An autonomous increase in capital intensity, may, however
be hard to realize unless mechanisms -of intervention are
entified, The following policy runs attempt to identify
means of such intervention.
One way to increase capital intensity might be to
influence prices of production factors. This can be done by
keeping nominal interest rates at a low level and wage rate at
a high level. Table 5.3 shows results of such strategies
Simiation Results of the First Policy Runs
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 649
Table 5.1
8 ae “eR m | &Bon | Reo e | PYow |
iy a8 3g | 289 | S858 | b3g9
Ae
o Sou | BE. e | “oo, |
AE ~
BR | gees | J @ec2 | song | eeas | B8es |
u xs RB | sogS | Soag | ASgR
cio |
By | ee. az | 8803 | Bay | S8ey |
2, | aos aa | eoe8 | do¥g | aod
i ag RS 5 8 8
ms es a ere ne
2 a
4s ee
s | 2h candy a
3 aay BE | AGRE | a8
£
£
eles] | a
4 ear | 2
y 4 | it
g a A ~
650 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
Table 5.2 Simlation Results of the Second Policy Runs
H H
Variable | Full flexibility| Full flexibility
' 1
j H i
| Policy Run { i
H i | except for fixed}-of market t
H H | interest rates | clearing H
H i H | mechanisms |
H H H H i
H H i H \
| Base Run | income per capita = | 62.85 ' 60.29 {
H | unemployment rate = | 0.07 1 0.08 H
H | net exports i 648.9 H 699.4 i
' | capital-labor ratio | 250.4 ' 250.4
\ H i
' . H i H
ie Increasing gov't | dics per capi te | pie
| purchases | unemployment rate | .
H | net, exports | -1775.0
H ' capital-labor ratio H 250-4
12. Encouraging invest- | income per capita t 76.80
| ment through tax { unemployment rate t 0.05
1 cuts | net exports H -68.0
i | capital-labor ratio | 250.4
{ i |
i H Y
13+ Encouraging invest- | income per capita | 27.99
| ment through { unemployment rate = | 0.18
| imereasing saving | net exports i 1052.0
i ‘propensity H capital-labor ratio H 250.4.
H H H
Hl H H
'4- Restricting imports | income per capita t 81.88
| ‘and expanding | unemployment rate { 0.04
' exports { net exports ; { 1075.0
i capital-labor ratio | 250.4
H H i
H H H
|5+ Supporting wage | income per capita =| 63.89
| Tate | unemployment rate t 0.01
i | net exports Hl 751.0
' ' capital-labor ratio | 250.4
16. Subsidizing domestic; income per capita 58.19
| Oi) consumption | unemployment rate 0.08
H | net exports 508.2
H ! capital-labor ratio 25064
!
t 1
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 651
Table 5.3 Simulation Results of the Third Policy Runs
{ i
Income } Unemp- | Net Exports} Capital-
per | loyment
capita } rate
Encouraging invest-
ment through tax
cuts
Encouraging invest-
through increasing }
saving propensity
4. Restricting imports
& expanding exports
5. Supporting wage
rate
Subsidizing domes-
tic oil consumption’
Deficit spending, export promotion and encouraging
investment through tax cuts are indicated as best policies.
The first two increase demand, while the third improves the
ability to invest, although, as stated earlier, the first
policy may create large national debt and the second may
increase dependence on foreign markets in the long run,
However, further helping to increase capital intensity through
explicitly raising its ambient value may also propel demand
Figure 5.1 shows the growth behavior when low interest
rate and high wage policy is also coupled with increasing
ambient capital labor ratio. Unemployment rate rises at
first, however, afterwards it declines below its initial value
Income, income per capita, and wage rate rise steadily. When
,oil exports start to decline due to the exhaustion of oil
reserves, interest rate rises to encourage savings in order to
provide for the increased capital investment needed for
supporting existing level of consumption, This set of policies
appear to be the best for realizing self-reliant and
sustainable growth.
652
x
32.0
1
bon Ibe
‘THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
e a cc ao FA
= s3 2 x OF ROK o
2 aM TH Soe Se Madd Zetez Sewn ae
B EEDORE WZEH FERKH AAS aa
a Haas
i 22
6 33 ‘ 1 1 1
4 7
1 1 ) '
O41 exports 7
er
yu.t
23u.t
‘
Iqcome per
2 ey
3 2
AT Dea
MVS
n VSs3n9
= ASN
3
t
Has oNos
2 2°32
= S04
3
Ve70.
Touu
desu.
Zu.
2uiv.
2u20.
£050.
Fig. 5.1 Behavior of Complete Model with the Proposed Policy
Set
6. CONCLUSIONS
This study has attempted to identify indirect instruments
for attaining self-reliant and sustainable growth in the oil
dependent economy of Indonesia. An important requirement of
such an exercise is to incorporate into the model an
appropriate “policy space." This meant that the market
mechanisms which are to be influenced through indirect means
were to be included.
The best policy set identified in this analysis aims at
increasing multiplier effects and acceleration in the economy.
This is possible by making sure that demand is not limited
while a high degree of flexibility is maintained in the
market. Also important is to increase capital intensity
through influencing factor prices and having an explicit
capital intensity goal instead of the existing tradition as a
basis for determining appropriate factor proportions.
The results of this study are generalizable to some degree
THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986 653
and may hold true also for non-oil economies. The analysis
shows that appropriate technology to be adopted by a developing
natien may not acquiesce in existing labor availability but
should take into account the growth patterns desired.
Although the suggested strategy seems to stimulate wage
rate, the exact implications for income distribution have not
been studied since the model is limited. A dual economy
framework in which wages are determined through bargaining
mechanisms has been suggested by Saeed (1983) for determining
income distribution in agrarian economies. Such a framework
appears quite appropriate for studying income distribution also
in the industrial sector
654 THE 1986 INTERNATIONAL CONFERENCE OF THE SYSTEM DINAMICS SOCIETY. SEVILLA, OCTOBER, 1986
REFERENCES
Arndt, H. (1975), Development and Equity : The Indonesian
Case. World Development, Vol. 3, Nos. 283, February -
March 1975, pp. 77-90
Barro, Robert J. (1984), Macroeconomics. John Wiley & Sons, Inc
Bhagwati, Jagdish N. (1984). Development Economics : What Have
We Learned ?, Asian Development Review, Vol. 2, No. 1,
Pp. 23-38.
Dick, Howard (1980). The Oil Price Subsidy, Deforestation and
Equity. Bulletin of Indonesian Economic Studies, Vol. XVI,
No. 3, November 1980, pp. 32-60.
Directorate General For Power (1981). Energy Planning for
Development in Indonesia. Ministry of Mines and Energy,
Republic of Indonesia
Down, S. (1983). Household Energy Consumption in West Sumatra
Implications for Policy Makers. Energy, Vol. 8, No. 14
pp. 821-833.
Forrester, Nathan 8B. (1982). A Dynamic Synthesis of Basic
Macroeconomic, Theory: Implications for Stabilization
Policy Analysis, PhD Thesis, Cambridge: MIT.
Hamberg, Daniel (1956). Economic Growth and Instability. W.H.
Norton & Company, Inc., New York.
Herrick, Bruce & Charles P. Kindleberger (1984). Economic
Development. McGraw-Hill Book Company, Fourth Edition.
Lewis, W. Arthur (1984). The State of Development Theory.
American Economic Review, Vol. 74, No, 1, March 1984, pp.
4-10.
Low, Gilbert W. (1980). The Multiplier-Accelerator Model of
Business Cycles Interpreted from a System Dynamics
Perspective. In Randers' Elements of the System Dynamics
Method. MIT Press.
McCawley, Peter (1983). Survey of Recent Developments. Bulletin
of Indonesian Economic Studies, “Vol. XIX, No. 1, April
1983, pp. 1-31.
Pitt, Mark M. (1985). Equity, Externalities and Energy
Subsidies: The Case of Kerosene in Indonesia. Journal of
Development Economics, 17, pp. 201-217
Randers, Jorgen (1980). Elements of the System Dynamics Method
MIT Press.
Saeed, Khalid (1985). Poverty, Hunger, and Development Policy
Proceedings of the Conference of the System Dynamics
Society, Keystone, Colorado, July 1985
Saeed, Khalid (1983). Worker Compensation and Income
Distribution in Agrarian Economies: Patterns and the
Underlying Organization. Dynamica, Winter 1983
Solow, Robert M. (1956). A Contribution to the Theory of
Economic Growth. In Stiglitz & Uzawa's Readings in the
Modern Theory of Economic Growth. MIT Press
Stiglitz, Joseph E. & Uzawa, Hirofumi (1969). Readings in the
Modern Theory of Economic Growth. MIT Press.
Tomkins, Raymond (1980), Structural Features of a System
Dynamics Model of the UK Economy. Proceedings of the 6th
International Conference on System Dynamics. University of
Paris ~ Dauphine, November 1980
CC BY-NC-SA 4.0