The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1750-6220.htm

IJESM
1,3 Modelling for policy assessment
in the electricity supply sector
of Pakistan
240
Hassan Qudrat-Ullah and Mustafa Karakul
School of Administrative Studies, York University, Toronto, Canada

Abstract
Purpose – The purpose of this paper is to provide a long-term assessment of Pakistan’s electricity
policy in the context of both environmental and resource constraints. To increase the sustainability of
energy supply, the Government of Pakistan introduced a series of reforms in the electricity supply
sector during 1990-1995. In response to these policy incentives, most of the independent power
producer offers included coal, oil, and/or gas-based power plants. Considering that Pakistan produces
only up to 40 percent of its oil demand domestically and thermal power generation causes CO2
emissions, there is a great need for an assessment of the existing electricity policy.
Design/methodology/approach – Drawing on system dynamics methodology, this study presents
and utilizes a dynamic simulation model that captures the dynamics of the sectors underlying the
electricity supply system including investments, capital, production, resources, financial resources,
and the environment.
Findings – The key findings of this study are: policy incentives encouraged thermal-based
generation at the potential expense of hydro power generation; and the evolution of electricity supply
related CO2 emissions exhibits an exponential growth.
Research limitations/implications – While there are other emissions related to the electricity
supply system with potentially severe environmental concerns, for example SO2, this study focuses
only on CO2 emissions.
Originality/value – The paper offers a system dynamics model and provides some useful policy
insights for the electricity supply sector of Pakistan.
Keywords Electric power generation, Environmental management, Resource management, Pakistan
Paper type Research paper

1. Introduction
The purpose of this paper is to provide a long-term assessment of the electricity policy
in the context of both environmental and resource constraints. To this end, several
issues are relevant:
.
Is the policy encouraging or discouraging the use of the local resource base?
.
To what extent is the electricity supply sector becoming dependent on fuel
imports that inherit the uncertainties of fuel prices and availability? and
.
How will this policy contribute to CO2 emissions scenarios?

International Journal of Energy Sector

Understanding of such complex policy issues necessitates the use of system simulation
Management (Ford and Bull, 1989; Olaya and Dyner, 2005). Therefore, to address these issues,
Vol. 1 No. 3, 2007
pp. 240-256 we present and utilize a dynamic simulation model that captures the dynamics of the
q Emerald Group Publishing Limited
1750-6220
sectors underlying the electricity systems including investments, capital, production,
DOI 10.1108/17506220710821125 resources, financials, and environment.
 The energy sector, especially of developing countries, poses serious challenges to Modelling for
the policy makers. On the one hand, they face financial and technological constraints in policy
their efforts to add new energy capacity. On the other hand, the rising global concerns
for environment have enormous implications for their economies (Sahir and Qureshi, assessment
2007). Pakistan, with medium GDP growth rates faces more than 10 percent annual
increase in its energy consumption for the foreseeable future (Qudrat-Ullah and
Davidsen, 2001; Siddiqui, 2006). To increase the sustainability of energy supply, in 241
1990-1991, the Government of Pakistan introduced reforms and provided incentives for
private (including foreign) sector investments, particularly in the electricity generation
capacity. The key incentives were (PEY, 1992):
.
the government’s guarantee for power purchases for Water and Power
Development Authority (WAPDA) and Karachi Electricity Supply Corporation
(KESC), both state-owned units, for 30 years;
.
reduced local earning investment requirements and simplified procedures; and
finally
.
attractive bulk power tariffs of 6.5 US cents per kilo watt-hour.

In response to these policy incentives, many independent power producers (IPPs) made
substantial investment contracts with the government. Most of the IPP offers included
coal, oil, and/or gas-based power plants. Interestingly, hydroelectric generation, despite
its rich indigenous resource base, could not attract any significant investment. As a
result, the government introduced special incentives for hydroelectric generation,
commonly known as “hydel policy 1995” (PEY, 1995). However, incremental IPP
investment in hydroelectric generation capacity reached only a meagre 198 MW by the
year 2004. By contrast, IPPs continued to invest in thermal (coal, gas and/or oil based)
capacity amounting to 5.5 GW in the same time period. It is to note that the major share
of IPPs’ investments in thermal capacity went to oil-fired power plants. Considering
that Pakistan produces only 40 percent of its oil demand domestically and thermal
power generation causes environmental problems, especially in terms of CO2
emissions, there is a great need for an assessment of the existing electricity policy and,
eventually, a decision by the policy makers to either reinforce or possibly introduce a
shift.
Based on our simulation results, the key policy insights are:
.
the government’s purchase rate signal, by itself, will not provide a stable signal
for investments rather may promote risk averse behaviour;
.
the oil and gas-based electricity generation experiences an increase of 68 and
51 percent, respectively, and the hydro-based generation faces a decrease of
27 percent (compared to the pre-policy case);
.
the environment is exposed to the additional 167 million tons of CO2;
.
the economy endures the increased (additional 1,919 million barrels) oil imports
dependency; and
.
contrary to the end-of-the-pipe measures for emissions reductions, the program
involving efficiency improvement and the competitive substitution for the least
emitting technology may, in the long run, cause the economy to evolve with the
reduced CO2 emissions and the least imports dependency.
IJESM The remainder of this paper is organized as follows: in Section 2, the modeling
1,3 approach for energy policy analysis and assessment is described. The dynamic model,
in terms of its causal structures, is presented in Sections 3 and 4 illustrates the model
behavior and the policy assessment. Conclusions are presented in Section 5.

2. Modeling approach for energy policy assessment

242 The policy analysis of economy-wide energy decisions requires, in the first place, an
integrated modeling approach capable of representation of the dynamics of energy,
economy, and the environment. In the literature, mostly large-scale optimization and
econometric techniques have been used to serve the planning needs of energy policy
decision makers (Bunn and Dyner, 1996; Qudrat-Ullah, 2005). However, a critical look
at these traditional techniques reveals a number of methodological limitations and
constraints including:
.
Inter-temporal interactions (and feedback) between energy system variables
such as demand, investment, production capital, indigenous resource use,
production, environment, and finances are not modeled explicitly in and across
various sectors.
. The disequilibrium framework for modeling is missing. For example, the
adjustments in the need for new electricity production technologies that arise in
response to changes in resource price that in turn are reacting to new equilibria
typically create transient behavior.
.
The variability in the elasticity of substitution between the competing
production technologies is not reflected.
.
Time delays and other distortions in perceiving the true value of the energy
system variables are not explicitly modeled.
.
Desired and actual variables’ magnitudes (e.g. rate of production of electricity
and cost efficiency) are rarely distinguished from real magnitudes in the model.
.
Nonlinear responses to actions (e.g. effect of import dependency on investment
share of the technologies) are not explicitly represented.

To provide useful policy support, the required modeling approach needs to address the
above mentioned limitations. Researchers, especially from system dynamics
community, have found system dynamics modeling capable of overcoming these
methodological limitations and to be useful in providing policy insights into the
dynamics of energy, economy, and environment. For instance, system dynamics
models have been developed and applied to national energy policy design and
evaluation (Ford, 1983; Naill, 1992); energy investments and uncertainty (Ford, 1985;
Qudrat-Ullah, 2005); conservation policy analysis (Ford and Bull, 1989); effects of
agents on utility performance (Geraphty and Lyneis, 1985); inter-fuel substitution in
OECD-European electricity production (Moxnes, 1990); privatization of electricity
industry (Bunn and Larsen, 1992; Bunn et al., 1997); energy efficiency and electricity
substitution by gas in the residential and industrial sectors (Dyner et al., 1995); and
policy assessment in the natural gas industry (Olaya and Dyner, 2005).
In the following sections, we will explain the system dynamics model that has been
developed to better understand the energy policy assessment for electricity supply
system of Pakistan within the environmental and resource constraints.
3. The dynamic model Modelling for
The model developed consists of numerous cause-effect relationships among the policy
variables of electricity supply system of Pakistan. Figure 1, adopted from Qudrat-Ullah
and Davidsen (2001), shows a simplified overview of these causal relationships. assessment
The model focuses on the IPP investments and their interactions with the external
environment of the electricity supply sector. Both, positive and negative feedback loops
are present. We discuss the behaviour as influenced by the individual loops in Figure 1. 243
3.1 Electricity demand sector
The positive loop (R) shown in Figure 2 manages electricity demand sector. Beginning
with an increased electricity demand, more resources (fuel) are required.
To meet the increased consumption of a resource, resource storage capacity must
be increased to retain a sufficient level of inventory. It takes time to build this storage
capacity and this time varies across the technologies. Once the required storage
capacity is acquired, the on-site fuel-supply can be increased to the capacity limit.
Thus, this feedback loop is a loop that manages a potential limit to the expanding
behaviour caused by the real operating positive loop shown in Figure 3, i.e. it lifts the
resource limit. But this loop will have no effect unless the loop in Figure 3 calls for
expansion.

3.2 Electricity production sector

Figure 3 shows the loop that concerns the production capacity expansion. We start
with an increase in electricity demand (electricity energy usage) which might be caused
by improved economic conditions. This will cause an increase in the demand for
production capital. The higher capital demand will result in an increased desired
capital, thereby causing IPPs to increase the investments in the production capital.
These investments add, after some delay (due to power plant construction), more
capital to the existing stock. An increased stock of capital generates more electricity.

ECONOMIC GROWTH

Energy supply
PRODUCTION SECTOR
DEMAND SECTOR
Revenue stream
Investment needs and incentives
Fuel supply

INVESTMENT SECTOR Fuel Costs RESOURCE SECTOR

Capital investments CO2 abatement costs

Capital depreciation
ENVIRONMENT SECTOR
Figure 1.
CO2 emissions
A sectorial overview of the
CAPITAL SECTOR dynamic model
Energy generation technologies
IJESM
+
1,3 +
Economic Growth

Electricity Intensity Electricity Demand

+
–
244
Resource Demand

Electricity Price R
+
+

Resource
Storage
Capacity
Electricity Cost

Electricity Production
Figure 2. –
On-site Resource
A casual loop diagram of Supply
electricity demand +
+

Economic Growth +

+
– Electricity Intensity
Electricity Demand
+
Electricity Price
Capital Demand
+
R

Electricity Cost Desired Capital

Electricity +
Production
Figure 3. IPPs Investments
A causal loop diagram of Electricity Generating
On-site Resource + +
electricity production Capital + Share of Technology
Supply +
In the case of hydro-based capacity development, the increased investments bring us Modelling for
into more expensive areas in which to build plants so that the “economies to scale” policy
benefit obtained by more production is counteracted by the marginally increasing costs
of entering new areas for electricity production. In the case of thermal capacity assessment
development, however, the efficiency of investment is assumed to be a constant. Thus,
in the case of thermal capacity development, the higher electricity production lowers
the average production cost, thereby causing the electricity price to fall. The 245
inexpensive electricity rate encourages the consumers to increase usage, thereby
reinforcing the initial rise in electricity demand.

3.3 Resource sector

The positive loop shown in Figure 4 manages the resource sector. Beginning with an
increased electricity demand, more resources are required. At first, the resources are
made available at the site indigenously. When the indigenous resources are exhausted
to the limit, the on-site resource supply is met through the imported resource. Note that
the resource imports are subject to the resource storage capacity. The increased on-site
resource supply boosts the potential for electricity production. The lower costs lead to
the lower electricity prices. The relatively lower electricity rates encourage higher
consumption, thereby casing further increase in electricity demand.

3.4 Investment and production capital sectors

The feedback loop concerning the capacity constraints is shown in Figure 5. Figure 6
shows stock-and-flow structure of the production capital.
An increased IPP investment in the preferred technology (coal, oil, hydro, and gas)
results in the higher capital stock for that technology. However, after the investment
decision has been made, the capital production capacity can only be raised after some
construction delay. Thus, the stock of operational production capacity of each technology
depends on the rate of investments by the IPPs as well as on the completion rate.

Economic Growth
+
+
– Electricity Intensity Electricity Demand
+

Electricity Price Resourcel Demand

R
+ Operating Costs
Indigenous
Resource
+ Price of Capital – +

Electricity Cost Resource Import

+
+
– On-site Resource Figure 4.
Electricity Production
Supply + A casual loop diagram of
+
resources
IJESM On-site Resource Electricity Cost
1,3 Supply –

+
–

246 Electricity Production

R
+ Share of –
Technology

Investment
Electricity Generating Incentive
Figure 5. Capital Premium
A causal loop diagram of
investment +
IPPs Investments +

Capital Operational
under Capital
Order Rate Construction Completion Rate

Figure 6.
Stock-and-flow structure
of the operational capital IPPs Investments in Average Construction
of each technology each Technology Delay for each Technology

With the higher production capacity, more electricity can be generated. The development
of thermal capacity, for which the investment efficiency is assumed to be a constant,
benefits from the higher production of electricity. Consequently, the production cost is
lowered. The lower production cost technology is preferred by the IPPs for investments,
thereby boosting the initial increased investment in that technology.
Both, the self-reinforcing and the constraining feedback structures are
simultaneously at work within the model as in real life. Which loop will dominate,
and during what period of time, depends upon the strengths of the causal feedback
relationships. The simulation of the model describes the effects of these feedback loops,
over a period of time (for details on the mathematical equations of the model please see
Qudrat-Ullah, 2005). In the next section, we describe both the simulation model
dynamics and how it was applied to policy assessment task.

4. The model dynamics and the policy assessment

As with any simulation model, our model was exposed to validity testing before its use.
In general, the validation framework for system dynamics simulation model consists of
the two types:
(1) structural validity – to ascertain that the model generates the “right output
behavior for the right reasons” (Barlas, 1996, p. 186); and
 (2) behavior validity – to test the agreement between the model-generated behavior Modelling for
and the real system (Barlas, 1989, 1996; Law and Kelton, 2000; Oliva, 2003) policy
Our model passed all the tests successfully (for details on theses test, please see assessment
Qudrat-Ullah, 2004). Based on results from our simulation model, we present
the dynamics of various feedback loops of the electricity supply system of Pakistan.
The simulation model is run from 1980 to 2030. 247
4.1 Dynamics of policy incentives
Figure 7 shows a balancing loop that explains the initial suppression of the share of a
technology in the IPPs investment due to the policy incentives. As the capital cost of
the power plants varies, the electricity production cost also reflects this cost variation.
Especially, if we did not take the life of the capital into consideration, the technology
with the most cost-intensive capital would be the least favored one. However, a longer
physical life of a power plant (i.e. operation life) may offset a huge initial cost. So,
depending on the way the investment is being depreciated financially, when the cost of
electricity generation from a certain technology is relatively high, whether it be initially
or temporarily, the price of electricity may also be set relatively high as a kind of a
risk-averting strategy (i.e. depreciation takes place over a relatively short estimated
lifetime relative to the real lifetime of the capital). This provides a little or no profit
incentive for the investor. This discouraging investment incentive caused by a
short-term perspective on investments directs the IPPs to invest in other, alternative
technologies, thereby rendering the initial investment to follow the reverse course: a
suppression of the share of investment for more capital intensive technologies. Thus,
investments may not reflect the real long-term costs of a technology due to this risk
premium. The risk premium may arise not only from within the energy sector
(e.g. influenced by the uncertainty of the oil-price), but also from uncertainties in the
socio-economic environment.

IPPs Investments
Electricity Generating
+ Capital
Discount Period B

Share of +
Govt. Purchase Price
Technology +
Electricity Cost
– Discount Rate
NPV of
Investments + +
– +
Investment Figure 7.
Incentive Electricity Price The casual model of policy
Premium incentives
–
Capital Price
IJESM The Pakistan’s promulgated energy policies attract IPP investments in the electricity
1,3 generating capital (Figure 8). There seem to be three phases of investments:
(1) for the period 1980-1990;
(2) the period 1990-2005; and
(3) the period 2005-2030.
248 In the initial phase, there is a rapid growth in the investments. The gap between the
demand for production capital and its existing stock drives the IPP’s actual investments.
The delays involved in the construction of the power plants and the investments
adjustment cause the initial increase in the capital demand. Thus, more investments are
made to meet the increased demand of the production capital. The investment trend
during the second phase (1990-2005) follows the IPPs’ preference to invest in the least
capital intensive projects. It is worth mentioning that these model-based results are well
in agreement with the actual data. As of June 30, 2005, the installed capacity of electricity
generation in Pakistan amounted to a total of 19.4 GW, with IPP’s share equal to 5.8 GW
(versus our model exhibits a total of 19.5 GW installed capacity of electricity generation)
in thermal generation and a mere 0.0003 GW in hydro power plants (PEY, 2005).
Thus, the capital-intensive technologies (coal and hydro-based) experience a
decreasing-increase trend of investments while the least cost capital technologies (oil
and gas-based) portray a steady growth pattern, as shown in Figure 9.
The investments in the third phase, after the year 2005 when the capital mix
substitution has been acquired in line with the policy incentives, follow a steady
growth pattern. At this time, the first replacement of the oldest capital unit begins.
Note that a longer physical life of capital resulting from investments in more capital
intensive technologies would postpone these secondary investments. Both, the
depreciated and the new demand of the production capacity are added to make up total
investments during this final phase.

4.2 Dynamics of endogenous pricing of electricity

The model assumes constant specific investment costs over the whole simulation
horizon, as we want to isolate the effect of endogenously generated electricity pricing
dynamics from those generated by external perturbations. The loop concerning

4,000
Promulgation of
Policy
3,000
Million $ / Year

Delays
Create
2,000
the Steady Growth to
Gap Less Capital Cost-
compensate both the new
intensive Capacity
1,000 Demand of Capital and the
Substitution
Depreciated Stock

Figure 8. 0
IPPs investments scenario 1980 1990 2000 2010 2020 2030
Time
 Production Capacity Modelling for
30 policy
25
assessment
Giga-watt / Year

20 Coal
Hydro
15 249
Oil
10 Gas

0 Figure 9.
1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 Impact of policy incentives
on production capital
Time

the suppression of the electricity demand, in response to the acquisition of the

capital-intensive technology, is shown in Figure 10.
The increased electricity demand, which might be caused by the improved economic
conditions, causes an increase in the demand for production capital. The higher capital
demand will result in an increased desired capital, thereby causing IPPs to increase the
investments in the production capital. These investments add, after the construction
delay, more capital to the existing stock. The capital costs, operating costs, and the fuel
costs determine the electricity production cost for each of the technology. The higher
the cost is, the higher the price is. The higher electricity price discourages the
consumers from increasing usage, thereby suppressing the initial rise in electricity
demand.
The price of electricity in the base year is 20.57 $/MWh, and follows a linear
decline path shown in Figure 11. The higher rate of production of electricity brings
the production cost, down. However, the effect of the supply/demand ratio modifies

Economic Growth

Electricity Intensity
–
+ +

Electricity Demand
Electricity Price +

Price of Capital
+ Share of Technology Capital Demand
B

+
Electricity Cost +
+ Figure 10.
Electricity Generating
Capital IPPs Investments The causal model of
+ Desired Capital electricity pricing
+ + mechanisms
IJESM Average Electricity Price
Least-cost
1,3 technology
30
substitution
25
20
$/MWh
250 15
Increasing
10 fuel imports
5
Figure 11.
0
Impact of policy on
1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030
electricity pricing
Time

this cost-based price. The higher the supply is relative to the demand, the lower the
price is. But after the year 1990, when the less capital cost-intensive technology has
been substituted, a further decrease in the electricity price occurs. Only the rising fuel
cost due to inadequate indigenous supply counteracts this price fall.
The electricity intensity (of GDP) represents an increasing consumption pattern,
shown in Figure 12.
In the earlier phase (during the period 1980-1990), there is a rapid growth of
electricity intensity. This is in response to the corresponding fall in the electricity price.
Thereafter, we see a linear growth of electricity intensity as a reflection of the linear
price fall pattern over the corresponding time span (1990-2030).

4.3 Dynamics of indigenous resource (fuel supply)

Figure 13 shows the casual model of impacts of the indigenous sources of fuel supply.
Beginning with an increase in electricity demand, which might be caused by a growth
in the economy, the resource (i.e. fuel) demand will increase for most technologies.
The fuel is consumed to generate electricity. However, some resources are renewable
(e.g. hydro) and some are non-renewable (e.g. coal, oil, and gas). The more of a
non-renewable resource that is consumed, the less becomes available for short-term
future needs until it is being re-supplied. In the long term, scarcity may turn out to be
a permanent state of affairs, in which case prices will remain high. The resource
availability affects its price; the abundance suppresses prices, while scarcity results in
Kilowatt-hour / Thousand $

2.5

2.0

1.5

1.0
Figure 12.
Impact of policy on 0.5
electricity intensity of
1980 1990 2000 2010 2020 2030
GDP
Time
 + + Modelling for
policy
Electricity Demand Economic Growth
Electricity Intensity
assessment
+
–
251
Resource Demand

B +

Operating Costs
Electricity Price Resource Imports –
Capital Costs

Indigenous
– Resource
+
+ Availability Figure 13.
+ Fuel Cost The casual model of
– impact of the indigenous
Electricity Cost + resources

a higher price. Thus, if higher consumption rates create an indigenous resource

scarcity, the resource price will rise. Also, when the resource demand is met through
the imports instead from the indigenous supply, the resource price rises. The higher
fuel price results in rise of the electricity production cost. The higher production cost,
ultimately, is shifted to the consumer in terms of a higher electricity price. To offset
this price hike, the consumers might reduce their electricity consumption. A reduced
tendency to consume eventually balances the initial increase in electricity demand.
The electricity supply in the base year, as shown in Figure 14, was mainly
dependent upon gas-based generation (41 percent) and hydro-based generation
(58 percent). The oil-based electricity production was merely one percent of the total
supply.

Electricity Supply Mix

1,40,000
1,20,000
1,00,000
Coal
GWh / Year

80,000 Hydro
60,000 Oil
Gas
40,000
20,000
0 Figure 14.
1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 Impact of policy on
electricity supply
Time
IJESM Almost 60 percent of this oil consumption due to electricity generation is met from the
1,3 imports. Therefore, we witnessed a relatively slow linear growth pattern in the oil
imports on account of electricity generation only for the period 1982-1990, as shown in
Figure 15. When the policy incentives are promulgated, oil-based power generation
technology gains a relatively higher share of IPP investments. Consequently, more oil
is needed. Therefore, the oil imports follow a fairly steep growth pattern during the
252 period 1990-2000. Afterwards, the oil imports go into a steady, linear growth path due
to the substitution of the least-cost technology (i.e. oil-based).

4.4 Dynamics of IPPs investments and CO2 emissions

The model and its simulation results show a direct impact of IPPs investment decisions
on the evolution of electricity supply related CO2 emissions. Figure 16 shows a
balancing loop that explains the effects of capital cost on the IPPs’ investment. Starting
with the IPPs’ investment, the increased investment in an electricity generating
technology adds more electricity generating capacity, but with a time lag equal to the
power plant construction time. As the capital costs of the power plants vary, in case of
the capital-intensive projects (e.g. construction of a dam for hydropower plant), a huge

Promulgation of
Policy Incentives
Million Barrels / Year

150

100

50

Figure 15. 0
Impact of policy on 1980 1990 2000 2010 2020 2030
resource imports (oil)
Time

Electricity Generating
+ Capital
+
IPPs Investments

Electricity Cost

+ B
+
Figure 16. Investment
The casual model of Incentive Share of
impact of the Premium Technology – Capital Price
capital-intensive
technology –
initial capital cost is required. This initial investment is the major concern of the IPPs. Modelling for
In general, if the economic conditions of the country are stable, they might invest for policy
the long-term benefits, e.g. they may opt to invest in a technology that has lower
operating cost and longer physical life. In this case, distributing the return on assessment
investment over the longer period of time might offset the initial cost. Otherwise, their
investment choice might be a technology that generates a fast income stream and is
less capital intensive. 253
The evolution of electricity supply-related CO2 emissions over the period under
investigation (1980-2030) is shown in the Figure 17, and exhibits an exponential growth
pattern. In the initial phase for the period from 1980 to 1990, gas-based generation is
responsible for the CO2 emission levels (i.e. for the linear growth part of the pattern). This
is because until 1990, when the policy incentives are introduced, the thermal generation
capacity constituted nearly 41 percent of the total supply with 40 percent gas-based and
1 percent combined coal and oil-based. After 1990, the policy incentives encourage
oil-based and gas-based generation that was less capital cost-intensive technology but
mainly at the expense of hydro-based generation (Figure 17).
Thus, the gas-based and oil-based generation constitute the major part of thermal
generation. Also, the lower production cost due to higher production accelerates the
substitution of thermal generation capacity. Carbon intensity of coal, oil, and gas-based
generation is 0.95, 0.81, and 0.73 (tons CO2/MWh), respectively. Consequently, CO2
emissions into the environment exhibit an exponential growth trajectory.

5. Conclusion
In this work, we have contributed with:
.
a system dynamics model; and
.
some useful policy insights for the electricity supply sector of Pakistan.

The developed system dynamics simulation model captures the dynamics of electricity
demand, investment, resource, environment, finance, and production sectors and

Substitution of oil-
based generation at
the expense of
1 1,100 hydro-based
Million Tons (CO2) / Year

1 880 Emissions
due to oil,
660 gas, and
1 Emission coal-based
due to generation
1 440 gas-based
generation
1 220
Figure 17.
Impact of policy on CO2
1 0
emissions
1980 1990 2000 2010 2020 2030
IJESM adequately captures the impacts of policy incentives, endogenous pricing of electricity,
1,3 indigenous resource, and capital intensive technology. Using our simulation results,
the key policy insights are:
.
In response to the policy incentives, the IPPs demonstrate a short-term
perspective on investments for gas because it generates early income streams.
Thus, the government’s purchase rate signal, by itself, will not provide a stable
254 signal for investments rather may promote risk-averse behaviour.
.
The policy incentive cause huge shift in the capacity mix for the electricity
generation technologies. The oil- and gas-based generation experiences an increase
of 68 and 51 percent, respectively. While the hydro-based generation faces a
decrease of 27 percent (compared to the pre-policy case). The environment is
exposed to the additional 167 million tons of CO2. The economy endures the
increased (additional 1,919 million barrels) oil imports dependency – a huge
uncertainty surrounds the sustainable fuel supply and consequently the availability
of electricity supply system;
.
To bring CO2 emissions to 1995 level, new policy incentives have to be
introduced. The model results show that the high-CO2 reduction targets for the
electricity sector of Pakistan are achievable but with significant economic effects
based on policy interventions, a 5 percent improvement in the efficiency together
with a CO2 tax proposal;
.
The evolution of electricity supply-related CO2 emissions over the period under
investigation (1980-2030) exhibits an exponential growth pattern – the policy
incentives encouraged oil-based and gas-based generation that was less capital
cost-intensive technology but mainly at the expense of hydro-based generation.
. To harness the rich local potential of hydro-power generation and to have fairly
environmental-friendly energy policy, our analysis suggests the need for policy
interventions encompassing long-term investment guarantees from the
Government of Pakistan.

These conclusions are in agreement with the results of an earlier study that was
done a decade ago (Qudrat-Ullah and Davidsen, 2001). In other words, the
government’s 1995 “hydel policy” could not achieve any significant reduction in
electricity supply-related CO2 emissions over the period under investigation (i.e.
2005-2030).
Regarding the design of alternative policy, our analysis suggests the government
should provide “long-term investment guarantees to IPPs” to attract investments in
hydropower generation projects. We did not investigate this issue further as this goes
beyond the focus of the paper. However, the developed simulation model appears to be
robust enough for study of such policy design issues (Qudrat-Ullah, 2005).

References
Barlas, Y. (1989), “Multiple tests for validations of system dynamics type of simulation models”,
European Journal of Operational Research, Vol. 42 No. 1, pp. 59-87.
Barlas, Y. (1996), “Formal aspects of model validity and validation in system dynamics”,
System Dynamics Review, Vol. 12 No. 3, pp. 183-210.
Bunn, D. and Dyner, I. (1996), “System simulation to support integrated energy analysis and Modelling for
liberalised planning”, International Transactions in Operational Research, Vol. 3 No. 2,
pp. 105-15. policy
Bunn, D. and Larsen, E. (1992), “Sensitivity reserve margin to factors influencing assessment
investment behavior in the electricity market of England and Wales”, Energy Policy,
Vol. 29, pp. 420-9.
Bunn, D., Larsen, E. and Vlahos, K. (1997), “Complementary modeling approaches for analyzing 255
several effects of privatization”, in Bunn, D. and Larsen, E.R. (Eds), Systems Modeling for
Energy Policy, Wiley, Chichester, pp. 303-25.
Dyner, I., Smith, R. and Pena, G. (1995), “System dynamics modeling for residential energy
efficiency analysis and management”, Journal of Operational Research Society, Vol. 46,
pp. 1163-73.
Ford, A. (1983), “Using simulation for policy evaluation in the electric utility industry”,
Simulation, Vol. 40 No. 3, pp. 85-92.
Ford, A. (1985), “Short lead time technologies as a defense against demand uncertainty”,
in Plummer, J., Oatman, E. and Gupta, P. (Eds), Strategic Management and Planning for
Electric Utilities, Prentice-Hall, Englewood Cliffs, NJ.
Ford, A. and Bull, M. (1989), “Using system dynamics for conservation policy analysis in the
Pacific Northwest”, System Dynamics Review, Vol. 15 No. 1, pp. 1-16.
Geraphty, D.M. and Lyneis, J. (1985), “Feedback loops: the effect of external agents on utility
performance”, in Plummer, J., Oatman, E. and Gupta, P. (Eds), Strategic Management and
Planning for Electric Utilities, Prentice-Hall, Englewood Cliffs, NJ.
Law, A.M. and Kelton, W.D. (2000), Simulation Modeling and Analysis, 3rd ed., McGraw-Hill,
Fairfield, CT.
Moxnes, E. (1990), “Interfuel substitution in OECD-European electricity production”,
System Dynamics Review, Vol. 6 No. 1, pp. 44-65.
Naill, R. (1992), “A system dynamics model for national energy policy planning”,
System Dynamics Review, Vol. 8 No. 1, pp. 1-19.
Olaya, Y. and Dyner, I. (2005), “Modeling for policy assessment in the natural gas industry”,
Journal of Operational Research Society, Vol. 56, pp. 1122-31.
Oliva, R. (2003), “Model calibration as a testing strategy for system dynamics models”,
European Journal of Operational Research, Vol. 151, pp. 552-68.
PEY (1992), Pakistan Energy Yearbook 1992, Ministry of Petroleum and Natural Resources,
Government of Pakistan, Islamabad.
PEY (1995), Pakistan Energy Yearbook 1995, Ministry of Petroleum and Natural Resources,
Government of Pakistan, Islamabad.
PEY (2005), Pakistan Energy Yearbook 2005, Ministry of Petroleum and Natural Resources,
Government of Pakistan, Islamabad.
Qudrat-Ullah, H. (2004), “Resources, pollution, and development of sustainable energy
policies”, in Quaddus, M.A. and Siddique, M.A.B. (Eds), A Handbook of Sustainable
Development Planning: Studies in Modelling and Decision Support, Cheltenham, Edward
Edgar.
Qudrat-Ullah, H. (2005), “MDESRAP: a model for understanding the dynamics of electricity
supply, resources, and pollution”, International Journal of Global Energy Issues, Vol. 23
No. 1, pp. 1-14.
Qudrat-Ullah, H. and Davidsen, P. (2001), “Understanding the dynamics of electricity supply,
resources, and pollution: Pakistan’s case”, Energy, Vol. 26 No. 6, pp. 595-606.
IJESM Sahir, M. and Qureshi, A. (2007), “Specific concerns of Pakistan in the context of energy security
issues and geopolitics of the region”, Energy Policy, Vol. 35 No. 4, pp. 2031-43.
1,3 Siddiqui, R. (2006), “The Pakistani economy: performance and challenges”, The Muslim World,
Vol. 96 No. 2, pp. 252-69.

256 Corresponding author

Hassan Qudrat-Ullah can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints