LPG Project 3 PDF

Report on the use of LPG
as a domestic cooking fuel option

in India
Prepared by
Antonette D’Sa and K.V.Narasimha Murthy
June 2004
International Energy Initiative

25/5 Borebank Road, Benson Town,
Bangalore 560 046
E-mail: [email protected]
Website: www.iei-asia.org
Contents:
page
List of Tables iii

List of Figures iv
List of Annexes v
Summary vi
1. Introduction 1
1.1 Background 1
1.2 Why LPG? 2
1.3 Objectives of this study 5
1.4 Methodology 6
2. Demand for LPG 9

2.1 Domestic use of cooking fuels in India 10
2.2 Estimation of domestic requirement of LPG 12
2.2.1 Extent of dependence on LPG 12
2.2.2 Consumption levels 16
2.3 Estimated future requirement of LPG 19
3. Supply of LPG 22
3.1 Current availability of LPG in India 22
3.1.1 In country refining capacity 23
3.1.2 Imports 26
3.1.3 Transport 28
3.1.4 Storage and distribution infrastructure 32
3.1.5 Marketing 35
3.2 Supply-demand balances 38
3.2.1 Regional balances 38
3.2.2 Sensitivity to demand scenarios 39
3.2.3 Estimated increases in supply to meet projected
requirements 41
4. Challenges to extending LPG for domestic use 41

4.1 Ensuring adequate supply and accessibility 41
4.2 Increasing affordability 42
4.3 Pricing policies 43
4.4 Poverty issues 45
5. Experiences of LPG programmes 46

5.1 Experiences in other developing countries 46
5.2 Experiences of an LPG programme in India 48
6. Issues for Indian domestic cooking fuels 49

Antonette D’Sa & K.V.Narasimha Murthy i
International Energy Initiative, Bangalore
6.1 Choice of fuels 50
6.2 Providing LPG 50
6.2.1 Pricing 50
6.2.2 Marketing (financing and packaging) schemes 53
6.2.3 Public awareness 53
6.2.4 Supply security 53
6.2.5 Dependable distribution network 53
6.2.6 Regulation 53
6.3 Non conventional alternatives 55
Annexes 56
Bibliography 75
Antonette D’Sa & K.V.Narasimha Murthy ii

List of tables page
Table 1: Indoor concentration of health-damaging pollutants from a typical

wood-fired cooking stove 2
Table 2: IPCC default (uncontrolled) emission factors for residential
fuel combustion (g/kg) 3
Table 3: State-wise use of LPG as fuel for cooking in the year 2000-01 12
Table 4: Increase in India’s total LPG consumption and the number of
consumers and distributors 16
Table 5: Reported monthly household consumption of LPG, 1999-2000 17
Table 6a: Nominal monthly expenditure on LPG as primary cooking
fuel in rural India, (NSS) 1999-2000 18
Table 6b: Nominal monthly expenditure on LPG as primary cooking
fuel in urban India, (NSS) 1999-2000 18
Table 7: Domestic dependence on LPG in the year 2001 19
Table 8: Estimated number of households using LPG in the years
1993-94 and 1999-2000 20
Table 9: Business-as-usual scenario - Projected domestic LPG requirement
based on current growth rates and use per household 21
Table 10: Scenario 2: Projected domestic LPG requirement based
on increased rural dependence but current use per household 22
Table 11: LPG Production in India (in million tonnes or mmt) 23
Table 12: Average refinery yields of Indian refineries
(based on 2001-02 production) 25
Table 13: Importance of crude oil and petroleum product (POL) imports 26
Table 14: Import facilities for petrol/diesel and LPG 27
Table 15: Estimated movement of petroleum products through the Railways 29
Table 16: Indian petroleum product pipeline capacities (in mmtpa),
as on 1st April 2002 29
Table 17: Fuel storage capacity (effective tankage) in the country 32
Table 18: Current state-wise distributors as on 1.4.2003 35
Table 19: Current state-wise consumers as on 1.4.2003 36
Table 20: LPG Infrastructure Costs 38
Table 21(a): LPG demand-supply balance, with special reference to
auto-fuels in India, projected till 2007 40
Table 21(b): LPG demand-supply balance in India projected till 2007 40
Table 21(c): LPG demand-supply balance in India projected till 2010-11 40
Table 22: Properties of LPG 59
Table 23: Fuel chemical composition, moisture content, and net energy 61
Antonette D’Sa & K.V.Narasimha Murthy iii

List of figures
Figure 1: Comparison of annualised costs of stoves 4

Figure 2: Range of demand scenarios 8
Figure 3: World-wide end-use demand for LPG - in the year 2000
and estimates for the year 2005 10
Figure 4a: Indian urban household dependence on various cooking fuels
in 2001 (the figures indicate the proportion of all urban households
using a particular fuel) 11
Figure 4b: Indian rural household dependence on various cooking fuels in 2001
(the figures indicate the proportion of all rural households using a
particular fuel) 11
Figure 4c: Indian rural household dependence on various cooking fuels in 2001 12
Figure 5a: Historical progression of primary cooking fuel choice in rural India
(Comparison of 1993-94 and 1999-2000 NSS data) 14
Figure 5a: Historical progression of primary cooking fuel choice in urban India
(Comparison of 1993-94 and 1999-2000 NSS data) 14
Figure 6: Petroleum refineries in India 24
Figure 7: Petroleum and product pipelines in India 31
Figure 8: Crude oil refining process 56
Antonette D’Sa & K.V.Narasimha Murthy iv

page
List of Annexes
Annexe 1: Technical details 56

Annexe 2: Fuel analysis 61
Annexe 3: A comparison of the annualised costs of cook-stoves (in India) 63
Annexe 4: India’s Administered Pricing Mechanism (APM) for the
petroleum sector 64
Annexe 5: Lessons from India’s improved stove programme 74
Antonette D’Sa & K.V.Narasimha Murthy v

Summary
The purpose of this study has been to examine the domestic use of liquefied
petroleum gas (LPG) in India. LPG is being considered because it is one of the relatively
clean and efficient cooking-fuel options currently available in the country. After
estimating current and potential increases in the domestic demand for LPG, we have
considered the possibility of meeting these demands, in view of several problems, and
then listed policy issues that could help surmount the barriers.
Demand (Section 2)
The current (primary)1 cooking fuel use patterns (Census of India, 2001) reveal
that LPG is used by 33.6 million (or 17.5% of the total) homes. In urban areas, the most
commonly used fuel is LPG (47.96%), followed by firewood (22.74%), and kerosene
(19.16%). However, in rural areas, 90% of rural homes still depend on some traditional
form of biomass, with firewood by far, the most important fuel (64.10%), followed by
crop residues (13.10%), and cow-dung (12.80%). The use of LPG (5.67%) is now
increasing in importance. Factors like income, (urban/rural) location, and the availability
and price of alternatives appear to have affected the choice of fuels.
Based on estimates derived from the Census figures, the average annual rate of
increase of LPG-dependent households in the 1990s’ has been about 11.8% in urban and
6.8% in rural areas2. Corresponding to the increase in LPG dependence, the urban
proportion of homes dependent on firewood and kerosene has fallen. Urban families have
shifted away from these fuels to LPG, possibly because of the easier accessibility, lack of
other fuel options, and more regular cash incomes.
If a business-as-usual scenario were assumed, that is if the current rates of

population-derived increase in the number of homes and the above rates of adoption of
LPG were projected, LPG would be used by over 90% of urban homes by the year 2008,
but less than 9% of rural homes. Such growth rates could be projected to later years;
however, enough data has not been obtained to gauge the adoption curves and the present
positions along it, so that such projections may not be reasonable.
From the current country-wide average use per household, based on total sales,
and weights for rural and urban differences (based on National Sample Survey estimates),
we have found the annual LPG use to be about 101.4 kg/rural household and 119.3
kg/urban household. These estimates have been assumed for future demand estimation.
(The lower rural use could be due both to difficulties in obtaining fuel refills and to the
availability of biomass for back-up/supplementary use). At this level of use, the LPG
required for domestic cooking would rise from about 3.87 million tonnes (mmt) in 2000-
01 to 6.46 mmt in 2005-06 and 9.10 mt in 2010-11.
1
Some households use more than one fuel; these figures pertain to the main source.
2
There are even higher estimates of household adoption of LPG, based on point-to-point growth
rates obtaining from a comparison between specific rounds of the National Sample Survey (NSS,
2001).
Antonette D’Sa & K.V.Narasimha Murthy vi
Apart from business-as-usual, enhanced-rural growth scenarios have been
projected, but these may not be practicable, considering the number of families living at
the subsistence level and unable to afford payment for fuel.
In addition, provision for other users must be included in the allocation of supply,
particularly the rapidly increasing use for automobile fuelling – by consumer choice in
the four-wheeler category and through a mandatory requirement in the three-wheeled
auto-rickshaw segment.
Supply (Section 3)
India’s indigenous production of LPG has not been able to keep pace with
increasing demand. Production rose from 2.150 mmt in 1990-91 to 7.273 in 2002-03, but
imports were required throughout the period. Of the total LPG supply in 2002-03, 4.903
mmt were from crude oil refineries, 2.370 mmt from natural gas, and 1.073 mmt (13% of
the total) were imported. With the average yields obtaining at present at Indian refineries,
LPG accounts for only 4.5% of the crude oil processed. Hence, in spite of the recent
discoveries of gas and the major refinery projects being undertaken, estimates from the
central Ministry of Petroleum and Natural Gas (MoP&NG) indicate a continuing shortage
of LPG, at least in the near future. By the year 2006-07, indigenous LPG production
would be 8.10 mmt, but total demand would be 11.48 mmt with current usage patterns
and 13.40 mmt, in a higher auto-fuel3 demand scenario. (Enhanced domestic demand
scenarios, like those our study, were not published).
Regarding the cost of imports, in recent years, the LPG import bill has amounted
to only 1.4%-3.4% of the net oil (POL)4 import bill, so that this source of supply has been
relied upon. However, the Asia–Pacific region has a shortage and dependence on the
Middle East that may not be strategically wise.
Even when available at the main ports and scattered refineries, LPG has to be
effectively transported, stored and distributed all over the country, if it has to be a viable
domestic fuel. Production is concentrated in the western region; pipeline capacity and
railway-tank-wagons are inadequate. There are also regional imbalances of demand and
supply that have to be addressed. Improvements are being made, but considering the
geographical spread of the country, the available infrastructure is still inadequate, for
example, the northern region has continually been a deficit area. More importantly,
although private distributors have entered the market, they have not extended services to
rural areas that seem to have been left a Public Sector concern.
Challenges to effective provision of domestic LPG (Section 4)
The need for using cleaner fuels has already been established. However, numerous
challenges are faced when considering the increased use of LPG; these include ensuring
adequate supply and accessibility, increasing affordability, effective pricing policies, and
reaching the people now dependent on collected biomass.
3
Here, 20% of petrol (gasoline or motor spirit) would be replaced by LPG.
4
POL = petroleum, oil, and lubricants
Antonette D’Sa & K.V.Narasimha Murthy vii
• Ensuring reliable supply and accessibility – The country needs not only additional
LPG production capacity, in the face of increased demand from the domestic and
auto-fuelling sectors, but also the development of adequate transportation (pipelines
and rail-tank-wagons), and storage installations. There has to be a reliable
distribution system running to local distributors even in rural areas, to prevent
refilling inconveniences that seem to counteract the advantages of using LPG.
• Increasing affordability – The economically disadvantaged face the problems of
high first costs of LPG (connection and equipment), and the lumpiness of relatively
high refilling bills, and loans are difficult to service without financial returns from the
investment.
• Appropriate pricing policies – These are a challenge, particularly because of the
subsidies already offered. The subsidies do not reach most of the poor as they are not
yet users of LPG, there is diversion of subsidised LPG from domestic to other uses,
and there is also a heavy burden on the central exchequer.
• Poverty issues – While the use of LPG is beneficial for health and the quality of life,
there is no direct impact on poverty alleviation without a link with income generation.
Further, questions regarding how the inherent benefits of LPG or other clean fuels can
be extended to the poor remain unanswered.
Lessons from other LPG experiences (Section 5)
Experiences in several other developing countries have been studied; the

following factors appear to have helped extend the domestic use of LPG (including lower
income households):
• Lower prices of LPG through cross subsidies from other distillates,
• favourable relative prices of LPG (in relation to competing cooking-fuels like
kerosene),
• special assistance for LPG purchase directed to lower income households,
• initial cost financing (deferred/instalment payments for the purchase of stove and
cylinder deposit),
• smaller cylinders/bottles to target (lower income) households through lower
periodic/incremental refuelling bills,
• special subsidies to these smaller cylinders/bottles – intended for lower income
groups,
• restriction on the supply of competing fuels (e.g. kerosene), and
• dependable distribution systems.
From the Deepam scheme implemented for households below the poverty line in
the state of Andhra Pradesh (in south-east India), one can get some more insights. For
example, although the scheme aimed at those below the poverty line, some of these
dropped out from it, while 80% of those above the specified income limit managed to be
included. Secondly, implementation bottlenecks -- limited choice, inability of suppliers
to supply equipment on time, co-ordination problems at the local level for the supply
arrangements, and irregular “commissions” for fuel refills -- contributed to dissatisfaction
among the recipients.
Antonette D’Sa & K.V.Narasimha Murthy viii

Issues for Indian domestic fuels (Section 6)
In the context of the provision of appropriate cooking fuels, Indian decision

makers would have to first consider the choice of fuels. LPG appears to be the preferred
option for those able to afford the initial and refill costs. If the use of LPG were to be
encouraged even for middle/low income households, there would be issues concerning
appropriate pricing and financing schemes, and dependable supply and delivery.
Provision of LPG
On the demand side, one would have to consider pricing (in particular, the
question of subsidies), financing options, and public awareness, and on the supply side,
security of supply, effective distribution/delivery, and regulation.
• Pricing issues
• Choice of LPG subsidies: With a subsidy provided for domestic users of LPG
even after the dismantling of the Administered Pricing Mechanism (APM),
any decisions regarding domestic LPG provision would have to begin with
pricing. Subsidy-options would also have to be decided upon – either on the
initial costs of connections/stoves, or on the fuel, through funds from cross-
subsidies or budgeted from the exchequer, and so on. Subsidising initial costs
helps to overcome the first-cost sensitive, and seems preferable to fuel (or
refill) subsidies because the latter could be diverted to other uses/users.
However, first-cost subsidies leave possibilities for dropouts from those who
cannot afford the fuel costs, resulting in “dead” investments.
• Operating (fuel) subsidies: If LPG refill subsidy is to be continued, some
precautions have to be taken:
• rationing/quotas (quantitative limits) for the subsidised fuel (as with
ration cards) and/or coupons (as with food stamps);
• differentiated containers (say, smaller cylinders, and/or cylinders
painted another colour) for specific purposes (as with subsidised
kerosene currently being coloured blue), to prevent use by those
outside the scope of the planned benefits;
• use-based subsidies (as with baseline tariffs for electricity) with
prices increasing with the level of consumption, thereby helping only
the minimum-level users and restricting “subsidy capture”.
• Cross subsidies from other distillates: This has been the Indian practice for
many years, but would need to be weighed against the disadvantages of higher
costs of transport (from higher priced auto-fuels).
• Funding of subsidies: The source of funds for the subsidies would have to be
one/more from among:
• LPG companies themselves - through a mandate of the government,
requiring the providers to sell below their costs, as in the present Indian
situation, but this has to be temporary or else there could be financial
disasters (as happened with the State Electricity Boards);
Antonette D’Sa & K.V.Narasimha Murthy ix

• regulated cross-subsidies from one consumer category to another -
effective as long as the funding category’s price elasticity is not too high
as to curtail sales;
• progressive tariffs (with the price per unit increasing with the amount
consumed): Here, the more affluent customers who use more, pay more.
This would work if the upper segment were large enough to support the
lower segments and could be considered akin to cross-subsidies from
higher income consumers to the others.
• Pricing of competing fuels: When evaluating the pricing of LPG, one has to
consider the relative prices of these fuels, and whether or not inter-fuel shifts
are desirable.
• Reducing/removing the subsidy on kerosene could make LPG relatively
cheaper, without a burden on the exchequer. (However, in the near term, or
as long as homes are not electrified, subsidies to kerosene have to merit
consideration because it is the source of lighting for about 43% of the
population).
• If the relative costs of LPG vis-à-vis other fuels were reckoned after
accounting for their calorific values and the efficiencies of the related
stoves; it can therefore be argued that LPG subsidies are not required.
• Direct cash benefits instead of subsidised fuel: There could be schemes through
which LPG is priced at its full cost, but targeted households get some pre-determined
compensation (as in the case of electricity for irrigation, in the state of Tamil Nadu).
This would avoid careless use of the fuel, while assisting the economically
disadvantaged. Such programmes would require funding from the government - with
transfer payments directly to the poor, but the better the targeting, the higher the
administrative costs. Also, earlier experiences with such below-BPL schemes have
not been very successful.
• Marketing (financing and packaging) schemes: Instalment payments for the cost of
connection and stove, and each fuel refill in much smaller containers (e.g. 2 – 5 kg,
instead of the regular 14.2 kg cylinders), will reduce the “lumpiness” of successive
cash outlays. (The latter option has been launched on a small scale by the three main
Public Sector distributors, but needs to be extended beyond limited areas).
• Public awareness: Awareness of the adverse impacts on health of indoor pollution

and the benefits of “cleaner” fuels would increase their popularity and thereby, the
willingness to pay.
• Supply security: Dependable supply of LPG requires -

• adequate and well dispersed import facilities,
• indigenous processing plants (refineries and natural gas fractionating
plants),
• availability of storage capacities throughout the country, and
• multi-mode transport facilities for moving LPG from alternative
destinations.
Antonette D’Sa & K.V.Narasimha Murthy x

• Dependable distribution network:
• The problems of distributors -- who face unfavourable economies of scale
when demand is low or dispersed, and those of consumers -- whose
location precludes them from LPG use, can be addressed through
extension of the distribution network beyond urban and semi-urban areas.
• Complementary infrastructure – roads, equipment suppliers, repair
services, etc. – should be built up in tandem, to facilitate the smooth
operation of the system (analogous to the rationale for improving rural
infrastructure along with electrification).
• Regulation - the government’s role: The government/regulator would have to set

standards to maintain safety and avoid corruption, impose measures for ensuring that
the cylinders are checked for their user-worthiness and are properly filled, and
provide consumer protection. (With a large numbers of operators and poor
enforcement of standards, accidents and commercial malpractice can occur).
While the government has to be involved, at least through its policies, in helping
to provide energy services to the economically disadvantaged, there has also to be a
suitable environment for the private sector to cater to those who can pay for their
needs. Subsidies will continue to be necessary for a while, but have to be applied
with care. Development assistance/grants – from aid agencies, etc. could help only
small fractions of the population; which means that the government and market forces
have to handle the rest and their extent and effectiveness have to be expanded to meet
current and growing needs.
Other options
There are other important alternatives to traditional cooking fuels, in particular,

biomass-based fuels already in use in a few places in the country, for example, biogas
(through animal dung and/or fibrous crop residues), and those not yet in use in the
country, such as synthetic LPG. These have been projected, as local sources of
petroleum-based products like LPG are limited, and international sources could be
adversely affected by political problems and price volatility. Renewable sources would
obviously be preferable, as long as they were used in a sustainable manner. Therefore,
the use of LPG can be considered as a short/medium term option i.e. a transition fuel (or a
complement) to sustainable fuels.
Antonette D’Sa & K.V.Narasimha Murthy xi

1. Introduction
1.1 Background:
Of the two billion people in the world currently dependent on biomass energy
(chiefly wood, and also dung and crop residues), some 700 million are estimated
to live in India alone (ESMAP, 2001). According to the Census of India, 2001,
about 91% of rural and 31% of urban5 homes depend chiefly on traditional fuels --
fuel-wood, animal and crop waste and charcoal -- for cooking.
Dependence on traditional forms of biomass adversely affects human

productivity particularly when time is increasingly spent farther and farther afield
for diminishing fuel-wood sources and if the health of those exposed is
endangered by high concentration of particulate matter from inefficiently burnt
domestic fuels. While individuals (mainly women and girls) are exposed to the
injurious effects (of smoke inhalation, the emission of unburned hydrocarbons
through traditional stoves, and soot deposits when washed off vessels, etc.) and
also have to spend time on fuel gathering, the community as a whole is adversely
affected both by the ambient pollution created by simultaneous cook-fires and
through land degradation in cases where fuel-wood is gathered in an unsustainable
manner6.
While Agenda 217 specifically recognized the challenge of providing access

for rural households to modern energy sources and called for “a rural energy
transformation”, efforts have focused chiefly on electricity generation. This has
meant that the need for cleaner and more efficient cooking fuels has not been
adequately addressed.
Trends in household fuel use can also be viewed along an “energy ladder”,
from simple biomass fuels -- twigs/shrubs, dung, crop waste -- at the lowest
levels, to fuel-wood, charcoal, and kerosene, and finally to LPG and electricity.
The fuel-stove combinations become cleaner and more efficient, but also increase
in capital costs as the ladder is ascended (OTA, 1992). Therefore, as household
income increases, people are able to move up the energy ladder, affording
seemingly more expensive but more efficient sources of energy, if they are
accessible8.
5
“Urban” is defined by the Census of India as any place with a municipality, corporation,
cantonment board or notified town area committee, or one satisfying the following three criteria
simultaneously: (i) a minimum population of 5,000, (ii) at least 75% of the male working
population engaged in non-agricultural pursuits, and (iii) a density of population of at least 400
per km2.
6
Actually, forests have been cleared for other reasons such as expanding settlements, roads, etc.
7
Agenda 21 is a comprehensive plan of action of the UN Division for Sustainable Development;
originally adopted at the UN Conference on Environment and Development in 1992, its
implementation was reaffirmed at the World Summit on Sustainable Development in 2002.
8
The energy ladder concept has been proven in studies of specific areas, for example, for a
sample of households in the city of Bangalore India (Reddy, B.S., 1995, 1996a).
Antonette D’Sa & K.V.Narasimha Murthy 1
Unfortunately, while households around the world have moved to higher
quality rungs of the ladder, in developing countries9 many are still dependent on
fuel-wood or have been forced down by local wood shortage to crop residues or
even shrubs and grasses (UNDP, 1998). It therefore is pertinent to assess the
current use of various domestic cooking fuels and the possibility of shifting to
cleaner and more efficient options. One of these options is liquefied petroleum
gas (LPG)10.
However, the likelihood of enhancing supplies of LPG and a distribution

network to meet increasing domestic demand have also to be considered.
Juxtaposed with the household demand must be the competing demand from the
automobile sector. This necessitates an assessment of the supply-side
requirements – from refinery capacity to transport, bottling and distribution, and
the associated constraints.
1.2 Why LPG?

Given the extensive use of firewood for cooking in India, studies have
been made on emissions from biomass-based stoves, including a detailed study of
greenhouse gases from small-scale combustion devices in developing countries –
with special reference to household stoves (Smith et al., 2000a, b). Table 1 shows
the indoor concentration of health damaging pollutants from a typical wood-fired
cooking stove while Table 2 indicates the default emission levels for carbon
monoxide (CO), methane (CH4), non-methane organic compounds and nitrous
oxide (N2O), through various residential fuel options.
Table 1: Indoor concentration of health-damaging pollutants from a typical

wood-fired cooking stove
1 kg of wood per hour
in 15ACH 40m3 kitchen
Carbon
monoxide Particles Benzene 1,3-Butadiene Formaldehyde
3 3 3 3
150 mg/m 3.3 mg/m 0.8 mg/m 0.15 mg/m 0.7 mg/m3
3 3 3
(10 mg/m ) (0.1 mg/m ) (0.002 mg/m ) (0.0003 mg/m ) (0.1 mg/m3)
3
The numbers in parentheses indicate typical standards set to protect health.

Source: Smith et al., 2000b
9
The term “developing countries” is usually used for lower income countries that are members of
the G-77, and China.
10
Liquefied petroleum gas consists mainly of propane (C3H8) and butane (C4H10). Annexe 1
contains more technical details.
Table 2: IPCC default (uncontrolled) emission factors for residential
fuel combustion (g/kg)
CO CH4 TNMOC* N2O
Gas1 2.0 0.2 0.2 0.005
Oil2 0.9 0.4 0.2 0.030
Wood 80.0 5.0 9.0 0.060
Charcoal 200.0 6.0 3.0 0.030
Dung/agricultural wastes3 68.0 4.0 8.0 0.050
Source: IPCC, 1997 (quoted in Smith et al., 2000b)
1. Determined using IPCC emission factors given for “Natural gas” and the net calorific
value given for “LPG”.
2. Determined using the IPCC emission factors given for “Oil” and the net calorific
value given for “Kerosene”.
3. Determined using the IPCC emission factors given for “Other Biomass and wastes”
and the average of the net calorific values given for “Dung” and “Agricultural
wastes”.
* Total non-methane organic compounds
There have been studies correlating fuel use and personal activity patterns
with health concerns, based on the use of biomass, and types of stoves, and in
particular, for specific parts of the country. For example, a sample study of
58,768 individuals in 10,265 rural households in 118 villages from 18 districts in
the north-Indian states of Uttar Pradesh, Himachal Pradesh and Rajasthan (Parikh,
et al., 2003) found correlation between the incidence of respiratory ailments and
the use of biomass-based fuels; the effects of health damaging pollutants through
the present cooking fuels was established, although this was exacerbated by
factors such as kitchen location and limited ventilation.
Among “cleaner” fuels, biogas, kerosene and LPG, the first depends on the
availability of cattle, and between the latter two, LPG has been found from
complete life-cycle environmental assessments (burden associated with the entire
product/package) to be a preferable option. A comparison was made between
kerosene and LPG (Jungbluth, 1995) in terms of the entire product/package, i.e.
on the basis of the total life-cycle impact from the extraction of crude oil and
natural gas, to processing in refineries and fractionating plants, product transport
and distribution, and finally cooking. For a majority of the indicators, it was
concluded that LPG had an ecological advantage over kerosene.
For the purpose of comparing the total costs of each alternative, we have
made a comparison (in Indian Rupees) of the annualised life-cycle costs (ALCC)11
of the commonly used stoves, at a discount rate of 12% per year. (In the case of
kerosene LPG, there is a difference in the price per unit between the administered
11
Annualised life-cycle cost = the annual equivalent value of the total costs incurred (initially and
during the working life of the equipment) = [Kx(CRF) + A], where K is the capital or initial
purchase cost, CRF = capital recovery factor = i÷[1-(1+i)-n], with i = interest or discount rate/year
and n = operating life of the equipment (in years), and A is the average annual operating cost =
the sum of fuel and maintenance costs. The costs that could result from adverse health effects
have not been considered.
price at which refills are purchased through the Public Sector12 oil companies and
the market price; hence two options each have been considered). These ALCCs
include both the initial costs and the operating costs, the latter varying with the
amount of fuel required (dependent on the energy content of the fuel and the
efficiency with which it is used) and the prevailing prices of the fuel. (Annexe 3
shows the prices and efficiencies of stoves and the prices of each fuel used for the
computation).
As Figure 1 indicates, the life-cycle costs increase from ordinary fuel-

wood stoves to LPG and electricity stoves. It is important to note that the
constituents of total life-cycle cost vary, with fuel comprising a much higher
proportion in the case of the less efficient options like fuel wood and conversely
the stove (capital) cost contributing much more to the higher-efficiency options
like LPG. Therefore, a larger investment made in the present for acquiring a more
efficient carrier system13 is compensated for by the long-term saving in fuel costs.
Figure 1:
Comparison of annualised costs of

stoves
4500
Rupees per year
4000
3500 Maintenance
3000
2500 Fuel
2000
1500
1000 Capital
500
0
Wood- Wood- Kero- Kero- LPG- LPG- Electric
tradn impr. PDS market subs. market (regular
coil)
Types of stoves
This is based on the authors’ computation, using market prices/subsidies prevailing in

the year 2003 (as shown in Annexe 3).
LPG can therefore be recommended both for its higher efficiency and lower
environmental impact than the alternatives. The human labour avoided and time
saving achieved through convenient cooking fuels have not been imputed with a
value, but need to be considered too.
There are other alternatives to traditional cooking fuels. Renewable sources

would obviously be preferable, as long as they were used in a sustainable14
manner. In particular, biogas (through animal dung and/or fibrous crop residues)
12
A company is termed “Public Sector” when the government owns a 51% or greater
shareholding in the organisation.
13
In addition to the cost of the LPG stove, one has to pay for the initial LPG “connection”.
14
We use the internationally accepted definition of sustainability as “meeting the needs of the
present without compromising the ability of future generations to meet their own needs”.
has been found to be the most efficient among the currently available “clean”
cooking fuels (Smith, et al., 2000). But the use of biogas is restricted by the
availability of cattle. New renewable options not yet in use the country, such as
di-methyl ether (DME), methanol, and synthetic LPG (syn-LPG) have also to be
considered.
Since LPG is a petroleum-based fuel, it can be argued that increased use of

this fuel should not be advocated; local sources of petroleum-based products are
limited, and international sources are adversely affected by political problems and
price volatility. On the contrary, it should also be considered that people in
developing countries, particularly in the lower income categories should be
allowed the choice of such a fuel, because their contribution to greenhouse gas
(GHG) emissions has been miniscule and constraints should therefore not be
imposed on them in the name of climate change. A poor person in India is said to
emit only 50 to 60 kg of carbon, compared to the world average of 1,100 kg and
5,000 kg in the USA (Parikh and Denton, 2002).
Therefore, the use of LPG is being considered as a short/medium term

option i.e. a transition fuel (or a complement) to biomass-based fuels.
1.3 Objectives of this study:

The purpose of this study is to examine:
1. the domestic use of cooking fuels in India, particularly that of LPG

2. the growth in domestic use of LPG in India particularly -
a. in continuation of the recent trend,
b. in excess of the trend,
3. the requirements – in terms of supply and distribution – to meet the
increased demand for LPG (in 2),
4. the challenges that are likely to be faced (for the implementation of 2 and
3),
5. experiences elsewhere, from which lessons could be learnt, and
6. the policies that could help surmount the barriers (in 4).

1.4 Methodology
The method followed for the subsequent sections is briefly being described
below.
Demand estimation:
Current requirement -
As data are not available in the form required, some computation has to be
done (using assumptions where required) to obtain estimates of the relevant
variables.
The service-based energy-use of any category of users for any period can
be described as the product of two variables, namely, (1) the total number of users
(an indicator of the spread of, or access to, that energy source), and (2) the energy
requirement for each user during that period (an indicator of the magnitude of
energy required to enjoy the service derived from that energy source).
For (1) the number of households using LPG for cooking, there are several
numbers available, namely, the decennial Census of India (2001) and various
estimates based on the aggregate number of domestic connections served by the
main Public Sector Undertakings in the petroleum sector. The Census information
is being considered the most reliable and hence the year 2000-01 is being taken as
the base year for the estimation.
For (2), the average LPG requirement per household, we are dividing the
estimated total annual use by “domestic” connections, by the estimated number of
such domestic connections (through all the public sector and private
distributors)15. This is not strictly correct because “domestic” LPG is known to be
diverted to automobiles and even small industries and commercial establishments.
This can be taken as a proxy for the “requirement per home”, because the actual
requirement for cooking based on the food cooked at each meal and the number of
meals for which LPG was the cooking fuel (in cases where more than one fuel is
used), are difficult to obtain for the country as a whole.
For the base-year, the total LPG use M1 is therefore the product of n1, the
number of households using LPG, and m1, the average annual use per
household (as a proxy for the strict requirement based on actual heat used for a
specified level of cooking). Then:
n1 x m1 = M1
[number of households [specific annual fuel use] [total LPG
requirement]
(say, in thousands) (kg per household) (thousand kg or
tonnes)
15
As the question of privatisation of (or government “dis-investment” from) Public Sector
undertakings is currently being debated, oil corporations have not been forthcoming about details.
It is important to avoid the “per capita” consumption figures usually
published, as these represent total use divided by the total number of homes in the
entire population and are therefore incorrect when applied only to LPG-using
homes.
Estimates of future requirement -
In this case, future use of a particular fuel, is estimated on the basis of the
base-year data.
Given the base-year number of LPG-dependent households and the

average use, the total LPG requirement Mk in any future year k can be estimated
according to the expected rates of change (growth rates) gn and gm, of the number
of users n1, and their average annual fuel use16 m1, respectively,
i.e., nk x mk = Mk,
where
nk = n1 x (1 + gn)k-1
and mk = m1 x (1 + gm)k-1
These growth rates gn and gm, could be based on past trends or on new
growth rates, g’n and g’m, depending on the policies likely to be implemented. For
example, if cleaner more efficient fuel use is to be encouraged in the domestic
sector, an increased growth rates of household LPG connections would be called
for, so that g’n > gn. (These rates of growth of consumers could vary over the
period considered).
Similarly, the average fuel use per consumer could also be expected to be
either constant, or change (increase/decrease). A focus on improved efficiency of
energy, say, through improved stoves, if possible, would result in lower fuel use
per household for the same level of energy service, i.e. m’k < mk. Even with
stove-efficiency constant, there could be changes in the average use because of the
level of services derived, for instance, where people shift from a
complementary/back-up fuel to using it for all their cooking/heating needs, the
requirement per household would increase, i.e. m’k > mk.
Simplifying the required steps from the above generalisation, one could
consider only two options for each variable -- number of households and fuel use
per household – in future scenarios:
The number of households would change over time either:
according to the current (business-as-usual) annual rate of growth gn, (leading to
nk), or
a new suggested rate of growth g’n (leading to nk).
The unit use per household, could:
continue at the current level, i.e. mk = m1 without any change (i.e. gm=0), or
change by some determined amount to m’k.
16
Ideally, at any point of time, one would have to consider, not merely an average fuel use per
consumer for the entire population, but several consumer segments, each with a particular usage
pattern.

Hence, as shown in Figure 2, there are four possible outcomes: business-as
usual (nk.mk), user-development-focused (n’k.mk), use-intensity-altered (nk.m’k),
and combined user and intensity altered (n’k.m’k).
Figure 2: Range of demand scenarios
Focus on consumer-population
Current New
growth rate growth rate
Current
End-use Unit user-development-
business-as-usual
(per consumer) Use focused
nk.mk
orientation n’k.mk
New use-intensity-
Unit altered user-& intensity-
Use altered
nk.m’k
n’k.m’k
Even without strictly working out growth rates in relation to a base-year, one
can consider scenarios with different proportions of the expected population
dependent on LPG for their main cooking requirements; one could also consider a
restriction (ceiling) on the dependence on LPG.
Supply assessment:
When assessing the possibility of meeting the requirement, one has to consider
both the quantity of LPG needs and the system for effective domestic delivery.
Quantity
Current production and the proposed refinery increases and production pattern
will give the estimated in-country supply; this includes production both directly
from natural gas and from distillation yield at refineries. To these one must add
imports; here there are problems of the country’s debt burden from the import bill,
depending on the international prices and currency exchange rates, and also on the
political situation.
Infrastructure
Supply to the consumer has to be analysed further, considering other

necessities of transport, bottling and distribution infrastructure, as well as the
regional bottlenecks and other problems. Marketing facilities must also be

considered. In the absence of corporation/company specific details (not revealed
for strategic reasons), an overall picture is being presented.
Challenges:
There are obvious problems regarding increased LPG use, particularly with
regard to accessibility – particularly in rural areas, affordability – of the initial
costs and fuel, and availability – in terms of the supply, transport, storage and
distribution network. These have to be looked at systematically, so that a solution
can be suggested to tackle each challenge.
Other experiences:
The experiences with (i) the expansion of LPG use in other countries and (ii)
LPG programmes in India are also being used to derive factors that would either
help or inhibit the successful implementation of LPG use programmes.
Suggested:
Finally, based on the situation described in the demand and supply sections,
the barriers to enhanced supply, and the lessons that could be learnt, suggestions
are being made regarding the policies through which the problems encountered
can be overcome.
2. Demand for LPG

While the worldwide average growth rate for LPG demand was about
3.7% per year during the 1990s, this varied between about 2% in Western Europe
and 3% in North America and about 6% in Asia (Purvin and Gertz, 2000). In
particular, China exhibited an average annual growth of over 19% and India,
9.5%. It is estimated that India’s annual growth will be over 11% between 1999
and 2005. In addition, India’s dependence on LPG, at 7.8% of its consumption of
all refined petroleum products, is one of the highest in the Asia Pacific region17
(MoPN&G, 2003b).
Worldwide, the end-use demand for LPG has been as shown in Figure 3.
However, while half of all LPG used East of the Suez was consumed by the
residential-commercial sector in 1985, this use is expected to increase to about
60% by the year 2005 (Purvin and Gertz, 2000). Growth of the residential-
commercial sector LPG demand is also expected to vary from region to region,
varying from a barely positive growth rate in Europe to over 5% for Asia during
1999-2005. The largest growth rates in this category will be in China and India;
in 1985, 5% of the total world residential-commercial LPG consumption was in
these two countries, but by 2005, this consumption will rise to more than 20% of
17
Conversely, India’s dependence on petrol (gasoline or motor spirit) is one of the lowest in the
region.
the world total. This would result in a deficit in the Asia Pacific region, further
necessitating imports from the Middle-east.
Figure 3: World-wide end-use demand for LPG - in the year 2000 and estimates for
the year 2005
Source: Purvin and Gertz, 2000

R/C = residential and/or commercial
2.1 Domestic use of cooking fuels in India
Several estimates of household use of cooking fuels in India have been

obtained (for example, IIFM, 1999; Malhotra, et al., 2001; Natarajan, 1990;
NSSO, 1992). However, the most exhaustive information appears to be from the
recent decennial Census of the Indian population (Census of India, 2001). Figures
4a and b (constructed from this information) show the proportion of households
using each type of cooking fuel, in urban and rural areas, respectively. In urban
areas, the most commonly used fuel is LPG (47.96%), followed by firewood
(22.74%) and kerosene (19.16%), with much lower dependence on other fuels. In
the rural areas, in contrast, firewood is, by far, the most important fuel (64.10%).
Other sources of biomass – crop residue (13.10%) and cow-dung (12.80%), are so
far the main alternatives, although LPG (5.67%) is now increasing in importance.
However, 72% of the country’s households live in rural areas. Thus, the
countrywide picture, shown in Figure 4c, indicates that traditional biomass
(firewood, crop waste, and dung) constitutes the main source of cooking fuels.

Figure 4a: Indian urban household dependence on various cooking fuels in
2001 (the figures indicate the proportion of all urban households using a
particular fuel)
0.6%
0.2% Firewood
0.4%
Crop residue
0.3% 22.7%
Cowdung cake
2.1%
Coal/lignite/charcoal
Kerosene
LPG
48.0% 2.0%
4.6% Electricity
19.2% Biogas
Any other
No cooking
Figure 4b: Indian rural household dependence on various cooking fuels in

2001 (the figures indicate the proportion of all rural households using a
particular fuel)
0.2%
0.8%
0.5%
Firewood
1.6% 0.1% Crop residue
5.7%
1.1% Cowdung cake
12.8%
Coal/lignite/charcoal
Kerosene
LPG
13.1% Electricity
64.1%
Biogas
Any other
No cooking

Figure 4c: All India household dependence on various cooking fuels in 2001
0.3%
0.6%
0.4%
Firewood
0.2%
Crop residue
17.5%
Cowdung cake
6.5% Coal/lignite/charcoal
Kerosene
LPG
2.0% 52.5%
Electricity
9.8%
Biogas
10.0%
Any other
No cooking
2.2 Estimated domestic requirement of LPG
2.2.1 Extent of dependence on LPG
The Census reveals that in the year 2001, there were 33.6 million or 17.5%
of the households in the country using LPG as their primary cooking fuel. These
comprised 7.845 million homes (or 5.67 % of the population) in rural areas and
25.752 million (or 47.96 % of the population) in urban areas. From the
information on individual states and union territories in the country, as shown in
Table 3, the dependence varied from over 50% in the (chiefly urban) union
territories to under 15% in the eastern states.
Table 3: State-wise use of LPG as fuel for cooking in the year 2000-01
State/Union Territory Total number Households LPG-using

of households using LPG proportion
(%)
All-India 191,963,935 33,596,798 17.5

Delhi 2,554,149 1,737,730 68.0
Chandigarh 201,878 126,146 62.5
Goa 279,216 145,453 52.1
Daman & Diu 34,342 17,304 50.4
Pondicherry 208,655 83,326 39.9
Mizoram 160,966 60,600 37.6
Punjab 4,265,156 1,435,648 33.7
Uttaranchal 1,586,321 531,076 33.5
Haryana 3,529,642 1,067,110 30.2
Maharashtra 19,063,149 5,656,425 29.7
Gujarat 9,643,989 2,746,018 28.5
Himachal Pradesh 1,240,633 348,727 28.1
Jammu & Kashmir 1,551,768 343,052 22.1
Dadra & Nagar Haveli 43,973 9,595 21.8
Manipur 397,656 86,608 21.8
Arunachal Pradesh 212,615 42,994 20.2
Andaman & Nicobar 73,062 14,706 20.1
Islands
Tamil Nadu 14,173,626 2,703,970 19.1
Andhra Pradesh 16,849,857 3,200,615 19.0
Sikkim 104,738 19,718 18.8
Karnataka 10,232,133 1,874,198 18.3
Kerala 6,595,206 1,168,536 17.7
Rajasthan 9,342,294 1,437,023 15.4
Madhya Pradesh 10,919,653 1,483,947 13.6
Assam 4,935,358 652,306 13.2
Tripura 662,023 85,477 12.9
West Bengal 15,715,915 1,962,540 12.5
Lakshadweep 9,240 1,055 11.4
Uttar Pradesh 25,760,601 2,913,579 11.3
Nagaland 332,050 31,479 9.5
Meghalaya 420,246 32,520 7.7
Chhattisgarh 4,148,518 309,801 7.5
Jharkhand 4,862,590 327,624 6.7
Orissa 7,870,127 410,823 5.2
Bihar 13,982,590 529,069 3.8
Source: Census of India, 2001
Several factors such as household income, location, and availability and

prices of alternatives, appear to affect the choice of or dependence on LPG.
Household income
It is expected that the dependence on LPG would increase with the

income/expenditure level of the household, as income has been found to be an
important variable in the choice of household items. This has been proven by the
periodic National Sample Survey (NSS), conducted by the Government of India18,
that elicits household expenditure on a variety of items. Using the reported
household expenditure as a proxy for household income, the expenditure on each
commodity/category of commodities, can be analysed according to household
income levels.
The most recent information obtained is from the NSS 55th round
pertaining to the year 1999-2000. Figures 5a and b, based on NSS data (for 1993-
94 and 1999-2000), show the percentage of households dependent on each type of
cooking fuel in each expenditure decile of the sample.
18
The National Sample Survey Organisation (NSSO) is under the Ministry of Statistics and
Programme Implementation of the Government of India. Details about the Survey are included in
Annexe 5, part 6.
Figure 5a: Historical progression of primary cooking fuel choice in rural
India (Comparison of 1993-94 and 1999-2000 NSS data)
Figure 5b: Historical progression of primary cooking fuel choice in urban

India (comparison of 1993-94 and 1999-2000 NSS data)
The graphs indicate that as one proceeds upwards along the expenditure
(income) deciles, households shift to “better” (cleaner and more efficient) fuels.
Obviously, the top deciles consume a disproportionately higher share of these
better carriers than the poor. This could be because, as incomes rise, the
households’ capital resources also increase, so that they can more easily incur the
initial costs of more expensive energy carriers like LPG (for the stove,
connection). Further, with increasing income, the consumer discount rate falls as
consumers more easily forego present consumption in return for future earning.
In other words, with lower consumer discount rates, future saving19 would have
relatively higher present values, so that seemingly more expensive options like
LPG would be more attractive. Studies on household energy use, for example, a
study on Bangalore city (Reddy, B.S., 1996), have verified this.
Comparative prices of fuels
The use of each cooking fuel, as with commodities, is influenced by the

prevailing prices. As such, wherever fuel can be “freely” collected, it is the
preferred option, hence the high proportion of firewood (usually twigs, etc.) use in
rural areas. Where firewood is not collectible, the next available option is used.
Kerosene is usually the first modern fuel to be used, because the administered
price, when obtained through the Public Distribution System, is relatively low.
Availability
The availability of a particular type of fuel has a strong influence on the

householders’ choices; obviously, apart from the prices, the ease with which
substitutes or competing fuels can be obtained, would affect the amount of the fuel
used. For example, kerosene is more easily transported and stored than LPG, and
therefore easier to obtain. The following Section, dealing with the amount of LPG
used, indicates a lower average use of LPG in rural than in urban areas; this could
be the result of greater difficulty in obtaining refuelling (cylinder replacements) as
also the availability of biomass sources that could be used to complement the
supply of LPG. The distribution system is obviously more developed in urban
areas, thereby affecting availability. As Figures 5a and b indicate, the decline in
the graph of homes using any fuel is balanced by increases in those using the
available alternatives.
Location
As shown in Figure 5a on rural areas, the use of firewood is persistently

high except in the highest three deciles where it is partially replaced by LPG,
whereas in Figure 5b on urban areas, both purchased wood and kerosene are
increasingly replaced by LPG as one proceeds up the income ladder.
The demand for LPG has historically been higher in the urban areas,
probably because the higher costs of refills vis-à-vis other fuels necessitates
higher cash incomes and also because the absence/shortage of biomass forces a
dependence on other fuels. Moreover, LPG is more easily available in urban
areas, as discussed above. However, the “switch” between fuels is often found to
be incomplete, as many households use more than one fuel, partly because of
differences between the tasks undertaken – the main meal versus supplementary
or additional heating. Further, although there appear to be more choices (wood,
kerosene, LPG, electricity), the gaps in and uncertainty of supply of each lead to
dependence on more than one source, with families storing and using more than
one fuel simultaneously as a risk mitigation strategy.
19
The present value of any saving S, derived k years from the present, at discount rate i% per year
= S÷(1+i)k
In contrast, in rural areas, the continued availability of some type of biomass --
branches, twigs, fronds, grasses, crop field waste, -- even if further away from
home, has not pushed households to other options. But here too, shifts to better
fuels do not eliminate the use of a traditional carrier, as users distinguish between
cooking of the main meals and other uses such as water heating.
Social factors
In addition to ability to pay, increasing incomes and education also lead to

awareness of the adverse impacts of indoor pollution associated with each fuel
evidenced in the quick switching from wood and twigs to kerosene as a family
moves from a slum to a tenement (Gupta and Kohlin, 2001). Adoption of a
“better” fuel has also been perceived as a status symbol (NIRD, 2002).
Historical progression
There have been perceptible shifts between over time away from fuelwood
and kerosene and towards LPG. As shown in the Figure 4 series above, the shifts
are evident even between the six-year period 1994-2000. In particular, during the
last two decades, the demand for LPG as a convenient fuel for cooking has been
increasing, to the extent of there being waiting lists of households seeking
“connections” (implying access to one/two cylinders of LPG at a time) from
distributing agencies. Thus the shifts shown in the Figures could have been
blunted by the lack of availability. The increasing demand for LPG has provided
a consumer base for private distributors who have been permitted into the market
in 1996.
However, it must be noted that the use of LPG through domestic connections
may not have been only for household use but also for cooking in commercial
establishments (hotels, etc.), for fuelling vehicles, and for small industrial units.
2.2.2 Consumption levels
The estimated total number of consumers – domestic and others -- and their
corresponding use of LPG are shown in Table 4.
Table 4: Increase in India’s total LPG consumption and the number of consumers
and distributors
Years Total (all Number of Number of

sectors’) consumers distributors
consumption (millions)
(‘000 tonnes) (actual)
1980-81 405 3.3 1,105
1985-86 1,241 10.7 2,742
1990-91 2,415 17.0 3,930

1995-96 3,849 25.7 5,165
1996-97 4,183 29.3 5,426
1997-98 4,581 33.7 5,538
1998-99 5,041 38.1 5,648
99-2000 6,029 47.3 6,161
2000-01 6,613 57.9 6,477
2001-02 7,310 63.5 7,486
2002-03* 8,157 69.8 7,910
* indicates provisional data
Sources of data: CMIE, 2003; MoP&NG, 2003c.
Given the total consumption of LPG and the number of connections, as

shown in Table 4, the average annual use of LPG per connection works out to
about 115 kg.
The NSS results can be used to verify this. Details from the 55th round
(1999-2000) on the reported average monthly household consumption of LPG
(shown in Table 5) indicate a cluster around 14.2 kilograms (one regular cylinder)
per month and another cluster around 7-8 kg (half a cylinder) per month; these are
equivalent to 170 kg and 85 kg per year, respectively. The averages from the
entire sample survey for rural and urban households were 11.3 kg per month
(135.6 kg/year) and 13.3 kg per month (159.6 kg/year), respectively.
Table 5: Reported monthly household consumption of LPG, 1999-2000

(Figures in parentheses indicate the percentage of households using LPG in the sample)
Quantity Rural Urban National

(kg/month) (%) (%) (%)
up to 2 4 3 4
2-4 5 1 2
4-6 7 3 4
6-7 6 3 4
7-8 14 8 10
8-9 1 1 1
9-10 8 9 9
10-11 3 3 3
11-12 2 3 2
12-13 1 1 1
13-14.2 6 6 6
14.2 31 42 39
14.2-15 6 6 6
15-16 2 2 2
16-18 1 2 2
18-20 1 1 1
20-25 1 3 2
25-30 1 2 2
30 or more 0 1 1
Source of data: NSS 55th Round

The corresponding nominal monthly expenditure on LPG and proportion
of the household’s expenses are listed in Tables 6a and 6b for rural and urban
households, respectively.
Table 6a: Nominal monthly expenditure on LPG as primary cooking fuel in rural
India, (NSS) 1999-2000
Expenditure Amount spent Proportion of
decile (Rupees) expenses (%)
1 53 4.8
2 91 3.9
3 84 3.9
4 102 4.9
5 138 5.5
6 141 4.8
7 137 4.8
8 152 4.4
9 148 4.1
10 153 3.3
Table 6b: Monthly expenditure on LPG as primary cooking fuel in urban

India, (NSS) 1999-2000
Expenditure Amount Proportion of

decile spent expenses (%)
(Rupees)
1 137 5.9
2 147 5.5
3 156 5.6
4 162 4.9
5 163 4.4
6 163 4.1
7 165 3.8
8 160 3.3
9 163 3.0
10 162 2.1
However, the authenticity of these estimates is based on each respondent’s

ability to recall and/or correctly estimate the family’s purchases and use of the
relevant commodity and there appears to be overestimation as compared with the
distributors’ estimates of sales, where available. The amount of LPG used for
cooking may also be overestimated because domestic buyers have been known to
use their quota for other purposes such as running cars. In this context, the LPG
use in rural areas can reasonably be considered lower than that in urban areas
because it is less likely that it is used for other services.
For future estimation of domestic LPG requirement, therefore, one needs the
true fuel requirement per household, based on efficiency of LPG-stoves and
cooking needs. However, cooking needs vary between families, in terms of
lifestyle patterns and the type of food cooked (depending on regional customs).
And, as indicated above, overestimation also occurs.
Hence, the assumption is being made that the average annual
consumption20 per connection is equivalent to the annual requirement per
household, but this single average is being weighted between rural and urban
areas in the ratio of the average NSS-reported household use, i.e. 11.3 kg per
month and 13.3 kg per month, and the number of Census-reported LPG-dependent
households -- 7.845 million and 25.752 million, in rural and urban areas,
respectively. Correspondingly, the aggregate annual average of 115.12 kg is
being disaggregated into 101.4 kg for rural areas and 119.3 kg for urban areas.
Then, for the average LPG requirement per household, as a first approximation for
the base year 2001, we are using these estimates of average LPG use per
household in rural and urban areas. Therefore, for the reported LPG-using
households, the total requirement would be 0.795 million metric tonnes (mmt) in
rural areas and 3.072 mmt in urban areas, as shown in Table 7. Further this
represents 58.5% of the total use of 6.613 mmt of LPG reported (MoP&NG, 2003)
for that year.
Table 7: Domestic dependence on LPG in the year 2001
For the base-year (2001): units Rural Urban Total

Census data: Total number of households
in the country million 138.272 53.692 191.964
Census data: Number of LPG-dependent
households million 7.845 25.752 33.597
=> Proportion of households using LPG % 5.67 47.96 17.50
Assumed average annual use per
household (based on derived All-India kg/year 101.4 119.3 115.1
average and National Sample Survey
results)
=> Estimated total domestic LPG use mmt 0.795 3.072 3.868
2.3 Estimated future requirement of LPG
As explained in the methodology, for the estimation of future domestic

LPG demand, one needs to consider the average LPG requirement per household,
and the projected increases (growth rates) for the number of LPG-dependent
households.
The average LPG requirement per household was estimated in Section

2.2.2 above. Growth rates would depend on the scenario envisaged.
For a “business-as-usual” scenario, the average requirement per

household is assumed to be the same as that in the base year and the projected
increases in the number of LPG-dependent households depend on the current rate
of growth of LPG-using households (or that obtaining in the recent past,
20
To obtain the average consumption per household, it is important to compute the average
obtaining among only the LPG-using households of the population; if the amount used in the
domestic sector were divided by the total households in the population, the “average” for the
country would be unrealistically low.
depending on accurate data availability). These growth rates have been estimated
as follows:
(1) The total number of households in the country, in rural and urban areas, in any
particular year, has to be estimated by interpolating between the decennial Census
figures. Then, with the National Sample Survey (NSS) proportions of the
population using a particular fuel, and the estimated total number of households,
the relevant number of households using the fuel in that year can be obtained.
Thus, the number of LPG-using households for the NSS years 1993-94 and 1999-
2000 was estimated. These numbers are shown in Table 8.
Table 8: Estimated number of households using LPG in the years 1993-94 and 1999-
2000
units Rural Urban Total

1993-94:
Estimated total number of millions 123.187 44.405 167.593
households
LPG-using proportion % 1.80 29.70 9.19
=> LPG-using households millions 2.217 13.188 15.406
Kerosene-using proportion % 1.90 22.90 7.46
=> Kerosene-using households millions 2.341 10.169 12.509
Firewood-using proportion % 80.10 30.30 66.90
=> Firewood-using households millions 98.673 13.455 112.128
1999-00:
Estimated total number of millions 136.009 52.255 188.264
households
LPG-using proportion % 5.40 44.10 16.14
=> LPG-using households millions 7.344 23.045 30.389
Kerosene-using proportion % 2.70 21.70 7.97
=> Kerosene-using households millions 3.672 11.339 15.012
Firewood-using proportion % 75.40 22.20 60.63
=> Firewood-using households millions 102.551 11.601 114.151
Please note:
(a) The total number of households in each year was estimated by interpolating between
the Census figures for 1991 and 2001.
(b) The proportion of households using each fuel in rural and urban areas is from the
National Sample Survey (NSS) in the given years.
(2) From the number of LPG-using homes so estimated21, the current (1999-2001)
average annual increase in users has been derived. These annual growth rates of
6.82% for rural areas and 11.75% for urban areas are being used for the business-
as-usual scenario.
Table 9 shows a business-as usual scenario, with the number of households

and the LPG requirement at 5-year intervals. Here, one must note that projecting
21
As a means of verifying these estimates, the same method was used to estimate the number of
kerosene-using households, because apart from new households, the increase in LPG using-
households would involve a fuel shift from households that paid for another fuel. In addition,
those purchasing firewood also incur costs that could stimulate a changeover to the LPG option.
the current increase in the number of LPG-using households would take the urban
dependence on LPG to around 90% of the projected number of households by the
year 2008. If one envisages that the urban dependence will not exceed 90%, the
rate of increase of households could, after that point, be reduced to the expected
population-determined increase of households (2.75% per year)22. Actually,
enough data has not been obtained to gauge the adoption curves and the relative
positions along it, so that annual-growth-rate based projections may not be
reasonable. However, with the current rates of LPG adoption, even in the year
2015-16, LPG would be used for cooking in only about 11.9% of rural homes.
For the country as a whole, LPG would account for about 36.4% of homes, with
the total requirement amounting to 10.8 mmt.
Table 9: Business-as-usual scenario - Projected domestic LPG requirement based on

current growth rates and use per household

2000-01 (base year):
Number of LPG-dependent
Growth rates projected %/year 6.82 11.75
2005-06:
Estimated total number of households millions 150.164 61.493 211.657
Estimated LPG-using households millions 10.910 44.873 55.783
=> Proportion of total households % 7.27 72.97 26.36
=> Estimated domestic LPG use mmt 1.106 5.354 6.460
2010-11:
=> Proportion* of total households % 9.30 90.00 33.64
2015-16:
*At the current rate of adoption of LPG for cooking, the urban dependence will reach
90% around 2008; thereafter the increase has been pegged at the average household
growth rate.
As an alternative, one could consider a scenario in which the rural

dependence is increased through doubling of the rate of increase of LPG-
connections from 2005-06 onwards. Even in this scenario, LPG would be used
for cooking in only about 22% of rural homes in the year 2015-16. For the
country as a whole, LPG would account for about 43% of homes, with the total
requirement amounting to 12.6 mmt. Other rural-enhanced-growth scenarios can
22
This was the average annual increase in the number of households in urban areas between 1991
and 2001; the corresponding rate for rural households was 1.66%. As a first approximation, these
rates are being projected for the estimation of the total number of households in the scenarios till
2016.
be projected, but these would not be practicable without substantial increases in
household incomes.
Table 10: Scenario 2: Projected domestic LPG requirement based on increased

rural dependence but current use per household

2000-01 (base year):
Number of LPG-dependent
Growth rates projected %/year 13.63 11.75
2005-06:
2010-11:
2015-16:
*At the current rate of adoption of LPG for cooking, the urban dependence will reach
90% around 2008; thereafter the increase has been pegged at the average household
growth rate.
3. Supply of LPG
Worldwide, the supplies of LPG are growing to meet demand. In 1985, world
supply was approximately 114 million tonnes; this is expected to increase to 240
million tonnes in 2005 (Purvin and Gertz, 2000), from enhanced processing of
natural gas and rising oil-refinery throughput. The growth in production of LPG
will probably outstrip that of most other oil products, since natural gas processing
– now the largest source of LPG -- is increasing more rapidly than crude oil
processing. Rising natural gas production will add to the amount of gas that is
processed and boost the supply of propane and butane. As markets develop,
reduced flaring of natural gas in many countries will also boost LPG supply;
Saudi Arabia and Nigeria, that flare gas the most, both plan to phase out the
practice (WB & WLPGA, 2002).
3.1 Current availability of LPG in India
Production of LPG in India grew steadily during the 1990s, both from crude
oil refining and from increased natural gas processing (Table 11). Imports also

increased during the 1990s’ as demand outstripped indigenous production, but fell
during 2000-02 due to the surge in Indian refinery output.
Table 11: LPG Production in India (in million tonnes or mmt)
Years From From Total Net

crude oil natural gas indigenous imports
refineries fractionators production
(a) (b) (a)+(b)
1990-91 1.221 0.929 2.150 0.329
1995-96 1.539 1.714 3.253 0.596
1998-99 1.724 1.914 3.638 1.173

99-2000 2.487 1.986 4.473 1.587
2000-01 4.088 2.045 6.133 0.853
2001-02 4.778 2.205 6.983 0.659
2002-03* 4.903 2.370 7.273 1.073
* indicates the Ministry’s provisional figures
Source: MoP&NG, 2003a,c; also www.indialpg.com
3.1.1 In-country refining capacity
India’s total refining capacity for all petroleum products (as on 1.4.2002)
was 116.07 million metric tonnes per annum (mmtpa) (MoP&NG, 2003a). As
shown in Figure 6, there are currently 18 refineries in operation in the country (16
in Public Sector, one in joint sector, and one in private sector). Of the 16 Public
Sector refineries, seven are owned by Indian Oil Corporation Limited (IOCL), two
by Chennai Petroleum Corporation Limited (a subsidiary of IOCL), two by
Hindustan Petroleum Corporation Limited (HPCL) and one each by Bharat
Petroleum Corporation Limited (BPCL), Kochi Refineries Limited (KRL) (a
subsidiary of BPCL), Bongaigaon Refinery & Petrochemicals Limited (BRPL) (a
subsidiary of IOCL), Numaligarh Refineries Limited (NRL) (a subsidiary of
BPCL) and Oil and Natural Gas Corporation Limited (ONGC). There is one
refinery Mangalore Refinery & Petrochemicals Limited (MRPL) in the joint
sector, (operated by HPCL), and one refinery in the private sector, at Jamnagar (in
the western state of Gujarat) belonging to Reliance Petroleum Limited (RPL).

Figure 6: Petroleum refineries in India
Bhatinda
Digboi
Panipat
Mathura Bongaigaon
Numaligarh
Barauni
Guwahati
Bina
Koyali
Jamnagar Haldia
Mumbai
Vizag Paradeep
Existing
Under Construction/Proposed
Subsidiary Companies
Mangalore Chennai
Cuddalore
Cochin Narimanam
Indian Oil Corporation Limited (IOCL) owns and operates seven refineries
in the country -- at Digboi, Guwahati, Barauni, (all the north east), Haldia (in the
east), Mathura and Panipat (in the north), and Gujarat (in the west) with a
combined installed capacity of 38.15 mmtpa; these achieved a total crude
throughput of 33.76 mmt (million metric tonnes) during 2001-2002. In addition,
its two subsidiaries, Chennai Petroleum Corporation Ltd. (with two refineries in
south India) and Bongaigaon Refinery and Petrochemicals Ltd. (with one refinery
in the north east), add another 9.35 mmtpa to its refining capacity.
The two refineries of the Hindustan Petroleum Corporation Limited

(HPCL) -- one on the west coast (in Mumbai) with a capacity of 5.5 mmtpa and
the other on the east coast (Visakhapatnam) with the capacity of 7.5 mmtpa --
produce a wide variety of petroleum products. During the year 2001-02, these
achieved a combined crude throughput of 12.33 mmt. The Corporation also
operates Mangalore Refinery & Petrochemicals Limited, with a capacity of 9
mmtpa.
During the year 2001-02, the Bharat Petroleum Corporation Limited

(BPCL) refinery at Mumbai achieved a throughput of 8.77 mmt; the throughput
achieved between April and December 2002 was 6.50 mmt.
Further, to keep pace with increasing consumption, 5 major refinery projects

are being implemented to add 40.5 mmtpa to refining capacity. Of these, the
construction of a 9 mmtpa refinery at Paradeep (a port in the eastern state of
Orissa) was commenced in May 2000 and that of another 9 mmtpa refinery at
Bhatinda (Punjab, north India) in June 2000. The first cross-country LPG pipeline
with a carrying capacity of 1.7 mmtpa and a total length of 1,270 km has also been
commenced. However, the costs of even expansion of refinery capacity are high,
with a recent addition of only 3 mmtpa estimated at Rs 23,603.8 million (US$
524.5 million) (MoP&NG, 2002, Section 3.4.4).
Given projected increases in capacity at specified refineries, one can

estimate the increase in LPG production through these refineries, because each
refinery has its own product slate/pattern depending on the configuration of its
processing units and it is not technically feasible to change the product slate
substantially. Table 12 gives the average refinery yields of Indian refineries. The
LPG yield from Indian refineries is about 4.5% of the total distillates.
Table 12: Average refinery yields of Indian refineries (based on 2001-02 production)
Product Percentage by weight of

crude oil processed
LPG 4.5
Naphtha 8.6
Petrol 9.1
ATF/Kerosene 11.5
Diesel 37.5
Lubes 0.6
FO/LSHS 11.5
Bitumen 2.4
Others 6.8
Fuel & Loss 7.5
Total 100.0
Source: Petroleum Planning & Analysis Cell (PPAC), MoP&NG, 2003b
Apart from the production at oil refineries, LPG is extracted from natural
gas (as was indicated in Table 11). This is currently the source of almost a half of
the LPG produced in the country. LPG is now being extracted from natural gas at
Duliajan and Lakwa in Assam (in the north-east), Bijaipur in Madhya Pradesh
(central India), Hazira and Vaghodia in Gujarat, and Uran and Ussar in
Maharashtra (all in the west), Pata in Uttar Pradesh (in the north) and
Nagapattinam in Tamil Nadu (in the south). In addition, a new plant is being set
up at Gandhar in Gujarat by the Gas Authority of India Limited (GAIL) and this
will have the processing capacity of 0.207 mmtpa (MoP&NG, 2002).
Ideally, this study should project estimates of future supply of LPG from
the various potential sources described so far. However, these estimates would be
subject to several assumptions, as the plans of the main firms dealing with the
supply of LPG (and other petroleum products) are not providing information on
the basis of which such estimates could be drawn. This appears to be mainly due
to the fact that structural changes in the sector are on the anvil, particularly dis-
investment of governmental holding in these undertakings.

3.1.2 Imports
Import of crude oil was made duty-free with effect from 1st April 2001.
Further, the Government decided in May 2001 to allow public sector oil
companies to exercise the option to import their crude oil requirement directly,
under the “actual user licensing policy” or through the largest Public Sector
Undertaking (IOCL).
In order to improve oil security, the oil companies made efforts towards
diversification of crude oil sourcing during 2002-03. IOCL had term contracts
with the national oil companies of Saudi Arabia, Kuwait, Abu Dhabi, Malaysia,
Libya & Nigeria. In addition, IOCL had a term contract with the national oil
company of Iran for supply of crude oil to MRPL. The remaining requirement was
procured through tenders. BPCL entered into term contracts with the national
companies of Kuwait, Saudi Arabia, Malaysia and Abu Dhabi to import crude oil
for its Mumbai refinery and KRL. Besides this, BPCL purchased crude oil of
Yemen, Egypt and some West African countries, on tender basis. BPCL is also in
the process of developing other sources of crude oil from countries like Angola
and Libya. For its Mumbai and Visakhapatnam refineries, HPCL entered into
term contracts during 2002-03 with the national oil companies of Saudi Arabia,
Abu Dhabi and Libya.
Import bill
Imports of petroleum products (petroleum, oil and lubricants or POL, in

export-import parlance) have constituted a significant proportion of the country’s
import bill over the years, contributing to the country’s unfavourable balance of
payments. As shown in Table 13, crude oil and petroleum product imports have
accounted for over 40% of the value of imports, although this contribution has
fallen to 15-20% of the total import bill. This “energy-debt nexus” has largely
been ignored, with discussions on the debt crises focussing on terms of repayment
rather than the role that prominent imports such as energy have played in
accentuating the problem (C.R.Reddy, et al., 1992).
Table 13: Importance of crude oil and petroleum product (POL) imports
Year Value of as a as a
imports of percentage of percentage
POL i.e. crude total imports of total
oil and (%) exports
petroleum (%)
products
(US$ million)
1970-71 180 8.3 8.8
1980-81 6,656 41.9 78.4
1990-91 6,028 25.0 33.2
1995-96 7,526 20.5 23.7

1996-97 10,036 25.6 30.0
1997-98 8,164 19.7 23.3
1998-99 6,433 15.4 19.1
Source: Directorate General of Commercial Intelligence and Statistics (DGCIS), Kolkata,
quoted in Table 7.2(A), Economic Survey: 1999-2000, (MoF, 2000)
POL = Petroleum, oil and lubricants
However, thus far, LPG has not contributed greatly to the total crude oil
and petroleum product (POL) import bill. LPG accounted for between about 1.4%
and 3.4% of the net POL bill over the last four years (’99 –’03)23. (During the last
three years, India has been exporting petroleum products like naphtha, motor
spirit, diesel and fuel oil, so that we are now net exporters of petroleum products
as a whole; however, the increasing imports of crude oil contribute to the growing
net import bill). Hence, it can be proposed that India import LPG to the extent of
the deficit of requirement over indigenous production.
Further, for LPG, in particular, there can be price differences on the basis of
the size of shipments that influence the landed costs; the larger the shipment, the
lower the cost per unit. For example, in West African markets, the shipping cost
of a 1,000 tonne shipment is at least 30% more on a per tonne basis than a 2,000
tonne shipment and at least three times the cost per tonne of a 12,000 tonne
shipment (WB&WLPGA, 2001).
Ports
IOCL is a promoter of Petronet LNG Limited (PLL) along with the Oil
and Natural Gas Commission (ONGC), Bharat Petroleum Corporation Limited
(BPCL) and Gas Authority of India Limited (GAIL). PLL is putting up terminals
at Dahej in Gujarat and Kochi in Kerala. The LPG import/export facility of the
joint venture Indian Oil Petronas Pvt. Ltd. at Haldia has been commissioned and is
terminalling LPG for public sector companies.
The existing infrastructure to receive imported crude oil and LPG are
given in Table 14. Although adequate for crude oil, the infrastructure at Indian
ports for LPG is inadequate to meet demand and is also not well dispersed. Over
75 per cent of indigenous LPG production comes from the sources located north
of Goa, and half the LPG import infrastructure is also located in that region. Due
to inadequate import facilities on the east coast, inland movement is required and
the costs are substantial.
Table 14: Import facilities for petrol/diesel and LPG

(in mmtpa)
Port Crude oil LPG
Kandla - 1.00
Vadinar/Sikka 48.60 0.10
Mumbai including JNPT 6.90 0.20
Ratnagiri - 0.20
Goa - -
Mangalore 9.60 0.60*
Kochi 7.60 -
23
The US$ was equivalent to about Rs 46 - 48, during the period.
Chennai 6.30 -
Tuticorin 0.80 0.20
Vizag 7.40 0.25
Paradeep - -
Haldia 9.10 0.60
Total 96.30 3.15
*
Can receive 1mmtpa by special measures
Efforts for increasing supply
To reduce the dependence on oil imports, the New Exploration Licensing

Policy (NELP) has been drawn up. Through this policy, exploration blocks, both
on land and offshore were awarded to bidders. A large gas discovery (named
Annapurna) was made in the Krishna-Godavari basin (in Andhra Pradesh).
Similarly, to encourage the exploration and production of new sources of
hydrocarbon resources, the Coal Bed Methane (CBM) policy has been formulated;
through this policy blocks for exploration and production in this category have
also been awarded. In addition, the Oil and Natural Gas Commission (ONGC)
has identified 15 major fields for implementing improved oil recovery plans.
3.1.3 Transport
LPG is moved from the point of production or import by pipelines, barges,

and rail and road tankers, to terminals, where it is stored under pressure. From the
terminals, it is transported as required to petrochemical plants, bulk depots or
cylinder filling plants; large users are supplied in bulk, while residential and small
commercial users receive pressurized cylinders through the distribution agents of
petroleum companies.
Considering the geographical spread of the country, the infrastructure for

movement of petroleum products is inadequate for handling the growing volume
of petroleum products. Pipelines are limited. Due to non-availability of tank-
wagons, oil movement is undertaken by road, which is not only hazardous and
polluting but also involves 15 to 20 times the specific energy use as through
pipelines and 5 times the energy use by rail24. In a country where oil is being
imported, expenditure on movement of POL products by road thus has been an
additional drain on foreign exchange. The actual losses due to road/rail
transportation are also 3 to 5 times higher than through pipelines.
Rail:
The Railways have been an important means of transportation, but the limiting
factor has been the availability of tank-wagons. Notwithstanding this fact, more
than 40% of the petroleum product transport is by rail. The available details are
listed in Table 15.
24
The average diesel used by trucks per tonne km of freight hauled in India has been 0.0341 litres,
whereas by rail it has been 0.0069 litres (Plan. Com., GoI, 1991)
Table 15: Estimated movement of petroleum products through the Railways
Year Freight Total Proportion

hauled by (million (%)
rail tonnes)
(million
tonnes)
1989-90 24.6 54.1 45.50
1990-91 25.1 55.0 45.60
1991-92 26.2 57.0 46.00
1992-93 26.5 59.0 44.90
1993-94 26.1 60.8 42.90
1994-95 28.6 65.4 43.70
1995-96 29.3 72.5 40.40
Source: MoP&NG, 2003a
The rail share of total petroleum product transport may, however, fall in
the years to come due to withdrawal of budgetary support. To overcome the
shortage of tank-wagons, especially for transportation of LPG, oil companies have
been financing railways under the "Own your tank-wagon scheme". The
Railways offer a rebate in freight with respect to products moved through tank-
wagons owned by oil companies. Since the depreciation on tank-wagons is
compensated for under the administered pricing mechanism (APM)25, oil
companies surrender the rebate so received to the Oil Coordination Committee
(OCC).
Pipelines:
Internationally, transport of products by pipelines is preferred to other

modes of transport for reasons of safety, operational convenience and
environmental benefits. In most cases, pipeline transport is also cheaper than rail
and road transport, but in India, only around 32% of petroleum product transport
is through pipelines. However, it is estimated that the share of pipelines in
product transportation may touch around 45% in a few years (MoP&NG, 2003a).
The region-wise petroleum product pipeline capacities in the country are listed in
Table 16.
Table 16: Indian petroleum product pipeline capacities (in mmtpa), as on 1st April
2002
Product No. Existing No. Proposed No. Total

capacity capacity capacity
(existing +
planned)
Petrol/diesel
West coast - 4 27.00 3 13.00 7 40.00

inland
25
The APM is explained in Annexe 4.
East coast - 3 6.70 1 1.40 4 8.10
inland
Others 5 8.15 5 6.02 10 14.17
Total 12 41.85 9 20.42 21 62.27

LPG
West coast - 1 1.70 1 0.80 2 2.50
inland
East coast - - - 1 1.16 1 1.16
inland
Total 1 1.70 2 1.96 3 3.66
To match the post Administered Pricing Mechanism (APM) scenario, the

Ministry of Petroleum and Natural Gas (MoP&NG) has issued guidelines26 for
laying petroleum product pipelines. The new guidelines for grant of right of user
(ROU) for petroleum products do not contemplate any restrictions or conditions
for grant of ROU for crude oil. Product pipelines have been categorised as
follows:
(i) Pipelines originating from refineries, whether coastal or inland, till a distance
of 300 kilometres from the refinery,
(ii) pipelines dedicated to supplying product to particular consumer, originating
either from a refinery or from the oil company’s terminal, and
(iii) pipelines originating from ports and pipelines originating from refineries
exceeding 300 km in length, other than those specified in (i) & (ii) above.
As per the guidelines, companies and investors will have complete freedom in
respect of the pipelines originating from refineries or meant for captive use of
companies for which ROU will be unconditional.
However, for pipelines exceeding 300 km in length and those originating from a
port location, grant of ROU will be subject to fulfilment of certain conditions27.
Figure 7 indicates the location of crude oil and product pipelines in India.
Indian Oil Corporation Limited (IOCL) has the country’s largest network, with a
combined length of 6,523 kms and a capacity of 43.45 mmtpa. IOCL’s pipelines
carried 40.36 mmt during 2001-2002. Petronet India Limited (PIL) a private
26
Vide notification F.No. P-20012/5/99-PP dated 20.11.2002
27
Some of these conditions are:
- Oil companies/investors interested in laying a product pipeline originating from a refinery or a
port would be required to publish the proposal inviting other interested companies to take capacity
in the pipeline.
- Any oil company interested in sharing the capacity of the pipeline, will be able to do so on
mutually agreed commercial terms and conditions. The proposer would then provide capacity for
such interested party also.
- The proposer company applying for the grant of ROU in land would need to provide at least
25% extra capacity for others.
- The pipeline will be owned and operated by the proposer company.
- The pipeline tariff will be subject to the control orders or the regulations that may be issued by
the Government under the appropriate law in force.
company, has so far implemented the Vadinar - Kandla pipeline and the Kochi -
Kurur pipeline projects. The Mangalore – Bangalore pipeline project (in the state
of Karnataka) is at an advanced stage of implementation.
Figure 7: Petroleum and product pipelines in India
Jalandhar
Bhatinda
Saharanpur
Meerut
Panipat Nahorkatiya
Delhi Tinsukia
Bongaigaon
Mathura Siliguri
Jodhpur Tundla Lucknow
Digboi
Chaksu Barauni Guwahati
Kot Kanpur
Sidhpur
Kandla Ahmedabad
Vadinar Koyali Budge

Salaya Navgam Haldia Budge
Manmad
Mumbai Vizag
Product
Proposed Product
Crude Oil
Vijayawada Proposed Crude
Chennai
Karur
Kochi Madurai
The new pipelines projects yet to be fully commissioned, or still under

construction, are:
• Kandla port (Jamnagar in western India) and indigenous production units in

Jamnagar, to Loni (in Uttar Pradesh in northern India), 1,246 km long and likely
to convey 2.5 mmtpa,
• Mumbai - Manmad pipeline, by Bharat Petroleum Corporation Ltd. (BPCL),
covering 270 km with an initial capacity of 3.30 mmtpa,
• Vizag - Vijayawada pipeline, by Hindustan Petroleum Corporation Ltd. (HPCL),
covering 380 km, with an initial capacity of 4.00 mmtpa, expected to be
commissioned by mid–1999.
However, the investment required may have hindered pipeline expansion, for
example, Gas Authority of India Limited (GAIL)’s 1,246 km LPG pipeline from
Kandla to Loni is estimated to cost Rs 12.295 billion (US$ 273 million), including
a foreign exchange component of Rs 3.867 billion. Acknowledging the
importance of creation of a pipeline grid, the Ministry of Petroleum & Natural Gas
(MoP&NG) of the Government of India has recently approved the setting up of an
apex holding company28 which will co-promote specific pipeline joint venture
companies (JVCs) to implement discrete sections of the grid.
Port traffic:
Currently, limited product movements take place between port locations.

Oil companies, at the direction of the Oil Coordination Committee (OCC), have
taken on charter-hire 27 tankers from shipping companies with an aggregate
tonnage of 0.638mn (MoP&NG, 2003a). In addition, the direct import of products
is also handled at port locations.
Road:
Nearly 30% of the total transport of petroleum products is by road. Unless

urgent measures are taken to improve the pipeline and rail infrastructure, road will
continue to be one of the key modes of transport.
3.1.4 Storage and distribution infrastructure
For storage and distribution, one has to consider installations, depots,

bottling and tankage capacity. Installations are large storage points attached to
refineries or to ports, serving as supply sources to locations in the region, while
depots are small storage and distribution centres that generally cater to the needs
of a city or town. At present, oil companies have installations in almost all major
cities and port locations and depots at all district headquarters.
Tankage
India usually has total storage capacity of about 16 days’ supply of LPG,
as shown in Table 17. Details on tankage of the industry are available for 1995
when the total tankage (all products’) capacity stood at 10.75 mmt.
Table 17: Fuel storage capacity (effective tankage) in the country

(in number of days requirement)
Product Marketing Refinery Total
name terminals/ tankage tankage
tankage
Petrol 47 17 64
Diesel 36 12 48
LPGa 10 6 16
a
Total storage for domestic and auto-fuel LPG
28
The holding company will be a non-governmental company in which the main public sector
companies IOCL, BPCL, and HPCL will hold 16% each and IBP will hold 2%. The remaining
50% will be offered to private sector oil companies and financial institutions. The holding
company shall subscribe to 26% of equity in each of the JVCs, 48% shall be offered to the public
and the remaining 26% shall be subscribed to by oil PSUs, financial institutions and private sector
oil companies.
The Ministry of Petroleum and Natural Gas (MoP&NG) intends
construction of an additional 530,000 tonnes capacity. For example, HPCL has
construction in progress for additional product tankage and allied facilities at
Pedapalli, Hassan and Irumpanam, in southern India.
A 60,000 tonne LPG cavern-storage project has just been initiated in

Vishakapatnam (in the state of Andhra Pradesh). A joint venture between HPCL
and the French company Total SA, it is being described as the safest method of
storing hydrocarbons (Business Line, 2004b). It will also help feed the
southeastern part of the country.
Containers
To meet the growing demand for LPG, the country is looking at quicker
ways of distributing imports. LPG is usually imported in large tankers and
unloaded into onshore storage tanks at ports. However, as India has only a few
ports large enough to berth modern LPG tankers, there remains the problem of
conveying LPG from theses few ports to the bottling plants at various locations.
Hence, the Ministry is considering the following option: Large tankers or

“mother” vessels will bring around 30,000 tonnes to the high seas and unload their
cargo into containers on smaller ships or “daughter” vessels of 14,000 tonnes each
(Petrowatch, 2003). These smaller vessels will then berth at ports that are too
small for the main carriers. The containers would then be unloaded and stored at
parking yards till they can be moved to bottling plants on especially designed
trucks.
Southern and eastern India – with an LPG deficit and therefore dependent
on imports – will benefit the most. Thus far, only Haldia (in West Bengal in
eastern India) and Vishakapatnam (in Andhra Pradesh, south-eastern India) have
facilities to berth regular LPG tankers, and these cannot economically supply the
southern peninsula region. Through this proposed container option, the expensive
option of constructing a large port along the southern part of the peninsula is
avoided and the existing smaller ports (such as Tuticorin, Tamil Nadu and
Kakinada, Andhra Pradesh, both on the south-eastern coast) can be utilised. The
risk will also be lower as transfer from mother to daughter vessels will take place
further from the shore.
LPG bottling plants
Four types of cylinders/bottles are currently being marketed by the Public

Sector Oil Companies – Indian Oil Company (IOCL), Hindustan Petroleum
Company (HPCL) and Bharat Petroleum Company (BPCL): the 14.2 kilograms
(kg), 19 kg, 47.5 kg and recently, 5 kg, each29. While the 19 and 47.5 kg cylinders
29
Each LPG cylinder marketed by the public and the private sectors is supposed to carry its
complete details including serial number, tare and gross weight, water capacity, ISI approval
monogram, test dates, manufacturer’s identification and year of manufacture. The cylinders have
to be manufactured only by the approved manufacturers, under the supervision of BIS inspectors
are meant for industrial and commercial customers, domestic consumers are
provided with the 14.2 kg cylinders and now 5 kg for low-income urban, as well
as semi-urban and rural homes.
Special facilities are needed to pack LPG in cylinders and LPG bottling
plants have been set up near the markets to facilitate the return of empty cylinders
and re-fuelling. It may be noted that manual bottling and distribution in small
carriers is cost-effective in developing countries, hence one can ignore the
economies of scale in bulk handling and distribution. However, safety standards
and reliability may not be as good as with automated filling plants.
The initial cost of new bottling plants is about Rs 2,600 (US$ 57.8) per
tonne per annum (tpa) capacity, with a plant of 70,000 tpa having been built at Rs
180 million (US$ 4 million) (MoP&NG, 2003, Section 3.7.2) and another of
138,000 tpa, at Rs 360 million (US$ 8 million) (MoP&NG, 2003, Section 4.2.2.1).
Regional distribution
Some regional distribution activities are worth noting.
The northern region:
GAIL has commenced work on a mega-project for laying a 1,264 km LPG

pipeline with a capacity to carry 2.5 mmtpa; the pipeline would run from
Jamnagar (in western India) to Loni (near Delhi) with receiving terminals to push
LPG into the pipeline, pumping stations, and boosters and delivery terminals for
supply to the marketing companies (Indiainfoline, 2002).
The western region:
There are arrangements between the organisations IOCL, HPCL and BPCL for
sharing infrastructure like depots, terminals and bottling plants. For instance,
HPCL is expanding its facilities at Loni so that BPCL does not have to invest in a
new plant of its own there and BPCL is sharing its facilities at Manmad with
HPCL.
The southern region:
Private players – Sri Shakthi, Caltex-SPIC and Mobil -- have a strong foothold
in the distribution market in the region, possibly due to inadequate supply from
the existing organisations. However, HPCL has a unique 60,000 million tonne
underground cavern storage facility at Vishakapatnam.
The eastern region:
This has so far been a region of relatively low demand, so that distribution
facilities are not increasing as in the other regions.
and are painted with a signal red colour; those from BPCL have a yellow ring around the bung,
those from HPCL a blue ring, and those from IOCL are fully red.
3.1.5 Marketing
In answer to demand, LPG marketing has historically been confined

largely to urban and semi-urban areas. Until recently there have been long
waiting lists for LPG “connections”, in spite of the extensive network of sales
points.
With the entry of private LPG distribution companies, the situation in

urban areas has eased considerably. The Petroleum Ministry of the Central
Government (MoP&NG) is also loosening its permissible marketing rules and has
proposed that private refiners be allowed to sell directly to bulk consumers after
meeting the demands of Public Sector companies that sell to domestic users
(Business Line, 2004a).
Tables 18 and 19 indicate the most recently available number of LPG

distributors and consumers served by Public and private companies, all over the
country. In recent years there have been noticeable attempts by Public Sector
companies to increase their supply to rural areas, but the tables do not distinguish
between urban and rural areas.
Table 18: Current state-wise distributors as on 1.4.2003
States Number of distributors

Andhra Pradesh 711
Arunachal Pradesh 28
Assam 212
Bihar 231
Chhatisgarh 94
Delhi 307
Goa 48
Gujarat 508
Haryana 256
Himachal Pradesh 97
Jammu & Kashmir 138
Jharkhand 106
Karnataka 455
Kerala 318
Madhya Pradesh 420
Maharashtra 908
Manipur 26
Meghalaya 30
Mizoram 23
Nagaland 22
Orissa 150
Punjab 400
Rajasthan 392
Sikkim 3
Tamil Nadu 510
Tripura 26
Uttar Pradesh 914
Uttaranchal 126
West Bengal 402
Union Territories
Andaman & Nicobar 2
Chandigarh 30
Dadra & Nagar Haveli 1
Daman & Diu 2
Lakshadweep 2
Pondicherry 12
TOTAL 7,910
Table 19: Current state-wise consumers as on 1.4.2003
States Number of consumers

(in thousands)
Andhra Pradesh 7504
Arunachal Pradesh 88
Assam 1247
Bihar 1564
Chhatisgarh 623
Delhi 3443
Goa 325
Gujarat 4115
Haryana 2315
Himachal Pradesh 929
Jammu & Kashmir 995
Jharkhand 674
Karnataka 3688
Kerala 3514
Madhya Pradesh 2809
Maharashtra 9362
Manipur 163
Meghalaya 72
Mizoram 141
Nagaland 90
Orissa 934
Punjab 3299
Rajasthan 2779
Sikkim 69
Tamil Nadu 6592
Tripura 164
Uttar Pradesh 7077
Uttaranchal 1154
West Bengal 3547
Union Territories
Andaman & Nicobar 32
Chandigarh 279
Dadra & Nagar Haveli 20
Daman & Diu 22
Lakshadweep 3
Pondicherry 173
TOTAL 69,805
Source: MoP&NG, 2003c
Public sector marketing network and schemes
IOCL has an extensive network of over 22,000 sales points backed for
supplies by 182 bulk storage points, and 78 LPG bottling plants. During the year
2002, IOCL has launched compact 5 kg cylinders for the benefit of the people in
rural and hilly areas.
During the year 2001-02, HPCL commissioned 178 retail outlets and 210
LPG distributorships and released 17.42 lakh new LPG connections. HPCL has
also introduced 5 kg cylinders in the states of Punjab, Uttar Pradesh and Jammu &
Kashmir, in the month of August ’02. One LPG bottling plant of 44-mmtpa
capacity at Kota, Rajasthan, and capacity augmentation of six existing plants (at
Kondapally, Mysore, Palghat, Gummudipundi, Unnao, and Jamshedpur) by a total
of 138 mmtpa, have been completed during 2002-’03 (till September ’02).
Construction work for the augmentation of an additional four LPG bottling plants
by a total capacity of 142 mmtpa is in progress and scheduled to be completed
during 2003.
HPCL now has a scheme called rasoi ghar (kitchen) for communal use of
LPG stoves in villages. Individual households would not have to invest on stoves
or pay a connection deposit, as with personal connections, but would have only to
pay for the use of the fuel and the facility, on the basis of the duration of usage. In
order to identify all the factors that can influence the effective operation of
HPCL’s rasoi ghar and to develop a viable model, a pilot project was taken up at
village Agwan, Tal Palghar, in Thane district (Maharashtra state). Accordingly,
the idea of a community kitchen was mooted to the panchayat of the village. The
pilot project was commissioned on 17.8.2002. Till November 2002, 49
community kitchens had been established in various parts of the country.
During the year 2001-02, BPCL commissioned 140 new retail outlets, 17
kerosene dealerships and 313 new LPG distributorships, and released 15.68 lakh
new LPG connections. In August 2002, BPCL has launched 5 kg cylinders at 33
selected rural markets in the State of Andhra Pradesh, Karnataka, Tamil Nadu,
Punjab, Rajasthan, Maharashtra, Gujarat, Madhya Pradesh, & West Bengal.
BPCL’s brand of LPG called Bharatgas, now has an online customer

service B2C (Business to Consumer) initiative in order to provide a direct channel
for Bharatgas customers to interact with BPCL. The online facility of booking
Bharatgas cylinders is currently available in the cities of Kolkata, Chennai,
Mumbai, Thane District, NCR Delhi (including Noida, Ghaziabad, Hapur, Meerut
and Sardana) Hyderabad/Secunderabad, Bangalore, Pune, Jaipur, Alwar, Dausa,
Bharatpur, Sikar, Lucknow and Nasik covering 5.2 million Bharatgas customers.
In order to reach far-flung rural customers, BPCL had introduced the Rural
Mobile Vehicle (RMV), in 1999, in the state of Punjab. Encouraged by this novel
method of reaching rural customers, BPCL has introduced 20 RMVs during the
year 2002-03.
Costs of LPG
Estimates of costs (Indiainfoline, 2002) show an import price varying

between Rs 15/kg and Rs 17/kg (including freight charges that also vary between
Rs 1.50/kg and Rs 3.00/kg, depending on volumes) so that, with port and terminal
charges, the cost would be Rs 21/kg to Rs 23/kg. The ex-refinery cost is
estimated to be Rs 17.50/kg. The costs of bottling as well as transport costs and
the distribution margin would have to be added to this.
If supply were to be extended into rural areas on a larger scale, there

would have to be more distribution agencies/vendors. Brazil is said to have
26,000 such vendors serving 35 million households (Barnes and Halpern, 2000).
In contrast, India has only about 12,000 distributors (WB&WLPGA, 2002),
serving about 33.3 million -- 7.8 million in rural areas and 25.8 million in urban
areas (Census of India, 2001). This is not intended to imply that the number of
distributing agencies is the reason for inadequate rural penetration, but a
successful distribution system would require many more rural-based market
players.
In the context of increasing LPG infrastructure in the form of cylinder

filling capacity, road tankers, storage tanks and cylinders, estimates of
international costs are as follows:
Table 20: LPG Infrastructure Costs
Item Capacity Cost

Additional cylinder filling capacity 100 fills/day US$ 2,500 – US$ 3,500
at an existing facility @ 12.5kg each
Small LPG road tanker 6 – 7 tonnes US$ 60,000 – US$ 70,000
Storage tank (at end-user site) 1 tonne US$ 1,000 – US$ 2,000
LPG cylinder (e.g. for residential 12.5 kg US$ 15 – US$ 20
Consumers)
LPG cylinder (e.g. for smaller 6 kg US$ 10 – US$ 15
Residential consumers)
Source: WLPGA, 2002.
In view of estimates of meeting future storage and distribution

requirements, more details regarding increases in bottling capacity, tankage, and
so on, are required. These have not been obtained for the reasons already
explained, but efforts will continue to be made to complete this aspect of the
analysis.
3.2 Supply-demand balances
3.2.1 Regional balances

The possible problems of zone-/region-wise imbalances between supply and
demand are being discussed below.
The northern region, consisting of the states of Jammu and Kashmir, Himachal
Pradesh, Haryana, Punjab, Delhi, Chandigarh, Uttaranchal, Uttar Pradesh and
Madhya Pradesh, consumes about 1.94 mmtpa or 33% of the country’s total LPG
use.
This has been a petroleum product deficit area (Indiainfoline, 2002). The
Indian Oil Corporation (IOCL)’s Mathura and Panipat refineries together
contribute about 0.5 mmtpa, and Gas Authority of India Limited (GAIL) supplies
LPG from its gas fractioning plants at Auraiya (0.3 mmtpa) and Vijaypur (0.4
mmtpa), with the balance met from the western region (including imports).
However, it is expected that the situation will be remedied in future through
increased transport from the western region via a cross-country pipeline and also
with the completion of new refineries. Hindustan Petroleum Corporation (HPCL)
is setting up a 9 mmtpa refinery at Bhatinda, while Bharat Petroleum Corporation
(BPCL) is due to construct a 7 mmtpa refinery at Bina and another 7 mmtpa
refinery in Lucknow.
The western region, comprising Rajasthan, Gujarat, Maharashtra and Goa, uses
about 1.77 mmtpa or 30% of the country’s consumption. However, this region
may have a surplus capacity with the commissioning of Reliance Petrochemical
Limited (RPL)’s 27mmtpa refinery, apart from the existing refineries of the three
main corporations, IOCL, HPCL and BPCL (Indiainfoline, 2002).
The southern region that includes Karnataka, Kerala, Andhra Pradesh and Tamil
Nadu, consumes about 1.47 mmtpa (or 25% of the country’s LPG use). The
public sector companies HPCL, Cochin Refineries Limited (CRL), Madras
Refineries Limited (MRL), and Mangalore Refineries and Petroleum Limited
(MRPL) together supply about 0.8 mmtpa. This total will increase with MRPL’s
expanded capacity (6 mmtpa) and the proposed expansion of HPCL’s
Vishakapatnam refinery (by 3 mmtpa). The private Nagarjuna Oil Corporation
has recently commissioned a 6 mmtpa refinery.
The eastern region comprises the states of Bihar, Jharkand, West Bengal, Orissa,
Chattisgarh, Sikkim, Assam, Meghalaya, Arunachal Pradesh, Nagaland, Manipur,
Tripura and Mizoram. It currently accounts for only about 0.67 mmtpa or 12% of
the country’s LPG consumption. This requirement is met mostly through the
refineries of the IOCL, although the port facilities of Haldia (in West Bengal) are
used for imports. At present, the geographical spread together with the low per
capita incomes in most areas make it unattractive except in the few cities (chiefly
Kolkata).
3.2.2 Sensitivity to demand scenarios
As indicated in several studies (some of which are quoted below), the current
shortage of LPG supply vis-à-vis demand is likely to worsen. The estimated LPG

shortage varies between 3.4 and 5.9 mmt by the year 2006-07 and increases to 7.6
mmt by 2010-11.
In addition, LPG is being increasingly used for auto fuel use (legalised
since 24th April 2001)30. This will be a competing source of demand on the
already insufficient supply, as indicated in the Ministry’s Scenario 2, shown in
Table 21(a).
Table 21(a): LPG demand-supply balance, with special reference to auto-fuels in

India, projected till 2007
(in million tonnes)
Year Demand Supply Surplus/(Deficit)
Scenario 1* Scenario 2** Scenario 1 Scenario 2
2002-03 8.42 8.42 7.60 (0.82) (0.82)
2004-05 9.89 10.77 8.02 (1.87) (2.75)
2006-07 11.48 13.40 8.10 (3.38) (5.88)
* considers the current pattern of use of automobile fuels
** considers substitution of 10% petrol demand by LPG by 2004-05, and 20% by 2006-
07
Source of data: MoP&NG, 2003b.
Table 21(b):LPG demand-supply balance in India projected till 2007

(in million tonnes)
Year Demand Supply (Deficit)
2003-04 9.528 7.989 (1.539)
2004-05 10.310 8.823 (1.487)
2005-06 11.123 8.749 (2.374)
2006-07 11.966 8.635 (3.331)
Source of data: Business Line, 2003b.
Table 21(c): LPG demand-supply balance in India projected till 2010-11

(in million tonnes)
Year Demand Supply Gap between Additional Import
demand and capacity required
supply
2006-07 10.2 4.7 5.5 2.1 3.4
2010-11 12.3 4.7 7.6 4.2 3.4
Source: Extract from Sundarajan Committee report on “Hydrocarbon Perspective-2010
AD”
The estimates in our study deal with only the domestic demand, but even
the business-as-usual domestic requirement of 9.1 mmt and 10.8 mmt in the years
2010-11 and 2015-16, could exceed indigenous supply. Given that the domestic
requirement accounted for only 58.5% in the base year 2001 and the use of LPG
for fuelling cars and auto-rickshaws has been increasing rapidly, the total demand-
supply gap would be even higher.
30
LPG use for automobiles is not only legal, but even mandatory for use in some cases (e.g. auto-
rickshaws in certain parts of the country).
3.2.3 Estimated increases in supply to meet projected requirements
The supply estimates listed thus far have not considered the recent results
of new exploration and the proposals of increasing LPG production by both
private and public sector organisations. If these proposals fructify, increased
delivery to currently under-supplied areas could be possible.
The costs of new refinery capacity are high31. For instance, the total
investment for the 27 million metric tonnes per annum (mmtpa) plant at Jamnagar,
(in the state of Gujarat, in western India) was reported to be US$ 6 billion32 (Rs
288 billion in 2002), and LPG constitutes a relatively small fraction of potential
refinery output (as shown in Table 12).
Supply increases have to be derived from such enhanced refinery capacity,

natural gas fractionators, or imports. In the last case, dependence on international
markets may not be strategically wise as the necessity of importing petroleum
products has makes the country vulnerable to increases in the international prices
of crude oil and its products (and any fall in the value of the rupee vis-à-vis other
currencies).
4. Challenges to effective provision of domestic LPG

The need for using cleaner fuels has already been established. However,
numerous challenges are faced when considering the increased use of LPG; these
range from ensuring adequate supply and accessibility, to increasing affordability,
effective pricing policies, and reaching the people now dependent on un-priced
biomass.
4.1 Ensuring adequate supply and accessibility
Adequacy of supply is obviously related to the magnitude of demand. But, in

addition, ensuring the availability and accessibility all over the country requires
not only adequate refining capacity and/or imports but also the development of
adequate storage installations and transport systems, a reliable distribution system,
and the avoidance of infrastructure bottlenecks. Storage and bottling facilities
outside the urban centres of high demand have been limited by whatever the
Public Sector Undertakings (PSUs) have been willing to invest.
Supply issues:
31
These are not disclosed by oil companies for strategic reasons; only the costs of a few projects
indicated in other contexts are mentioned in reports or news items.
32
This plant of Reliance Petroleum Limited boasts the world’s largest polypropylene complex
(0.6 mmtpa), largest fluid catalytic cracking unit, delayed coking plant and paraxylene complex
(1.4 mmtpa), and also estimates its cost at 30-40% lower on a per-tonne basis, than recent
refineries built in Asia (RPL, 2000).
• Present supply shortage: As indicated in Tables 21(a) to (c), LPG is in
short supply even at the present requirement. Growing domestic
consumption would lead to an ever-increasing imbalance.
• Competing demands: There are likely to be further problems of supply if
LPG is used increasingly for automobile fuelling – both because of four-
wheelers (private vehicles and taxis) being converted and due to
mandatory norms requiring the use of LPG, as in the case of three-
wheelers (auto-rickshaws).
• Indigenous production: The costs of new production infrastructure --
refinery capacity and gas fractionation, and bottling units – are already
high (as indicated in Section 3.1) and would be difficult to recover with the
current price structure.
• Imports: More importantly, the Asia Pacific region will have a sizeable
deficit in the supply of LPG that would have to be met by importing from
the middle-eastern countries, any interruption in the Arab Gulf region may
lead to disruption in physical supplies and price risks.
Distribution/delivery issues:
• Infrastructure: The existing infrastructure at Indian ports for LPG is
inadequate to meet present demand. Also, over 75 per cent of indigenous
LPG production is from the western region, and half the LPG import
infrastructure is also located there. Due to inadequate import facilities on
the east coast, inland movement is required and the costs are substantial.
Internationally, pipelines are the preferred mode of transport, but in India
only around 32% of such transport is through pipelines. Due to non-
availability of tank-wagons, 30% of oil product movement is undertaken
by road, which is not only hazardous and polluting but also involves 15 to
20 times the specific energy use as through pipelines and 5 times the
energy use by rail.
• Consumer problems: Currently, vast (rural) areas of the country are
located far from distribution centres, so that users have to pay for the extra
costs of cylinder supply. Moreover, for small and remote markets, refills
often take more than a week, so that for those without a second cylinder
there are gaps in fuel supply, requiring a standby fuel also. (And, signing
up for a second cylinder obviously increases the deposit cost, precluding
lower income households from investing).
• Distributor problems: For LPG dealers considering rural markets, the
small number of purchasers and low rate of consumption (and refills) lead
to poor economies of scale, that, along with poor road infrastructure make
it difficult to establish commercially viable distribution networks.
• Safety: LPG delivery (as in the case of other pressurised or gaseous fuels)
involves cylinder management; this necessitates more careful transport
than kerosene or firewood that in turn imposes additional requirements on
prospective dealers.
4.2 Increasing affordability
A lack of awareness about the effects on health and the relative thermal
efficiency of alternative fuels could hinder people from making the effort to obtain
cleaner and more efficient fuels, even without financial hurdles. However, the
problem of affordability affects the households’ decision in several ways.
• Relatively high initial cost: An LPG “connection” (deposit for the

pressurised cylinder/canister) and stove constitute a large upfront cost
(when compared with the equipment for other fuels), so that some who
could afford the fuel cannot make the initial investment.
• Household perception of future saving: Total annual cost = annual fuel
expenses + annualised equipment costs. With the poor using higher
discount rates, future savings (if any) would be less valuable than current
expenditure.
• Larger minimum quantities of LPG usually33 have to be bought at each
refill (as compared to kerosene, charcoal and wood), undermining the use
of LPG in low-income households.
• Repayment difficulties: Whereas micro-credit programmes and loans for
productive purposes are repaid even by poor households, particularly by
women (Grameen Shakti in Bangladesh, SEWA in India, Vietnam
Women’s Union), through the returns they obtain, it could be difficult to
repay a loan for household convenience alone. (Improved lighting
systems contribute to longer hours and improved working conditions for
household industries like tailoring and basket-making and service
industries like TV repair shops, the profits from which can be used to
service the loans; but improved cooking conditions may not warrant
payment). People pay for some conveniences; beyond this level, there
needs to be some productive outcome to justify the expenses.
• The kerosene->electricity shift for lighting is not replicable because the
costs of the former are higher than for the improved (more efficacious
lighting); here, LPG can be more expensive when the total (equipment +
fuel) cost is added, unless consideration is given for reduced pollution and
the resulting health effects.
• Currently, the poorest sections of the population who do not “pay” for
fuel because they depend on whatever they can collect, cannot even
consider it.
4.3 Pricing policies
Appropriate pricing policies also appear to constitute a challenge in India.

Till April 1998, the Indian oil & gas industry had been under state control vide
the Administered Pricing Mechanism. (Annexure 4 has more details). The
production pattern, capital expenditure and pricing of petroleum products were
all determined by the state. The deficit incurred on products priced lower than
costs – LPG, kerosene and diesel – was compensated for by the higher-than
cost prices of motor spirit (gasoline), aviation turbine fuel, naphtha, fuel oil,
etc. These inflows and outflows were handled by the Oil Pool Account that
was self-sufficient, so that no government support was necessary.
33
New cylinders of 5 kg or less are not available in most places.
Liberalisation is supposed to have taken root in the petroleum sector with
several policy changes: in 1987, private participation was allowed in joint
venture refining, in 1993, parallel marketing was allowed for LPG and
kerosene, to attract the private sector into distribution and thereby increase the
availability of the products, in 1998, phased dismantling of the Administered
Pricing Mechanism (APM) was initiated and in 2002, the APM was
dismantled. When decontrol measures were initiated, retention pricing for
refineries was abolished. But controls on the prices of 5 products (motor
spirit, diesel, aviation turbine fuel, kerosene, and LPG) that contribute 70% of
the volumes, were retained, while subsidies on LPG and kerosene were limited
to 15% and 33% of import parity prices. (Tariffs on crude and petroleum
products were reduced to 0-5% and 15% respectively).
The kerosene and LPG segments still enjoy subsidies; these subsidies were
scheduled to be reduced substantially by the time of downstream petroleum
sector deregulation in April 2002. However, what happened was that in the
fiscal year (April to March) 2002-03, these subsidies that had previously been
managed through cross-subsidies from other petroleum products using the Oil
Pool Account, were for the first time made explicit in the national Budget.
The Ministry of Finance allocated Rs 45 billion (approximately US$ 1 billion
in December 2003) for LPG and kerosene; due to the rising international
prices, the actual subsidy worked out to Rs 100 billion, of which the
Government paid Rs 63 billion (Business Standard, 2003). With the retail
prices fixed34, and the costs higher than expected, there was a shortfall that
had to be met by the four main state oil companies35. For example, based on
the international price of US$ 230/tonne (April 2003), the cost of LPG is Rs
80/cylinder higher than the permitted retail price to the domestic sector, but
the subsidy provides only 56% (Rs 45.17) of it (Gupta, 2003).
There appear to be several problems, particularly with the subsidy system:
• Heavy burden on the exchequer – The fuel subsidy imposes high

opportunity costs. For example, the (central) government’s total bill for
subsidies to kerosene and LPG together for the year 2002-03 (Rs 63
billion) was similar to the Central Plan allocation for education (Rs 62
billion), of which only Rs 43 billion was set aside for primary education in
that year! (The Tribune, 2003) And, the amount allocated for rural
employment programmes was only Rs 4 billion (The Hindu, 2002).
• Subsidies for fuel reduce the incentive for efficient use – By lowering
end-use prices, they reduce the users’ incentive to conserve or use energy
more efficiently, and if not reimbursed to producers, they reduce their
incentive and ability to invest in new infrastructure/technology.
34
Prices were raised on 16th June 2004, since this report was prepared.
35
An additional problem has recently arisen. The Finance Ministry has proposed that private
LPG distributors be given the same subsidy for domestic sales as the Public sector companies
(Business Line, 2004a), but the retail price restriction for them has not been specified, which is
likely to lead to further problems.
• Misuse of subsidies – Even though subsidies for refills are provided for
domestic connections, domestic cylinders are often diverted to running
vehicles. The average use calculated on the basis of the total quantity of
LPG in some areas, divided by the number of domestic connections works
out to over 200kg per connection per year; this is too high for cooking and
indicates an obvious diversion to other uses.
• Subsidies to LPG users for fuel purchase not justifiable – The usual
justification for subsidies (sunrise industry protection, increased
employment, access for the poor, redistribution of income, etc.) cannot
apply in general, but only to specific categories of consumers. (Perhaps
one could justify subsidies for increased access to a “cleaner” fuel or
avoidance of other inefficient/polluting fuel options?)
• Subsidies garnered chiefly by the urban rich - This is obvious from the
residential descriptions of consumers and also from specific studies. It has
been estimated that three-quarters of the LPG subsidy went to urban
households in 1999-2000, four-fifths of whom were in the top half of the
population, expenditure-wise36. For example, a study of a sample of
homes in the city of Hyderabad indicated that 90% of the urban rich were
utilising the subsidy meant for domestic LPG (UNDP&ESMAP, 2003).
Even if lower-income households are able to benefit from LPG through
subsidies, the relative financial value to them is relatively small as their
consumption is generally modest.
4.4 Poverty issues
The requirements of those who survive on collected (“free”) biomass do

not appear to be addressed by providing LPG even at the present subsidised
rates. In particular:
• Would improved cooking fuel options have any impact on poverty

alleviation? There is not much empirical evidence to convincingly
demonstrate the linkage between specific energy strategies and poverty
reduction (as opposed to merely widening access); these are available in
other sectors such as health (Cecelski, 2000). LPG and other modern fuels
would be more efficient (in terms of heat delivered from input) and also
more environmentally benign in comparison with traditional biomass-
based stoves; they would also enable labour and time saving, freeing
people for more productive pursuits, if these were available. However,
without direct linkages to income-generation, there is no obvious affect on
reducing poverty.
• Would improved cooking fuels benefit the poor less than the others? It
has been observed in the past that rural electrification has benefited people
36
In economic parlance, this is the problem of “inclusion”.
with higher income rather than lower income37. The explanation seems
straightforward as only those with sufficient resources for the initial
investment in the connection and the energy-using equipment will be in a
position to benefit from electricity or any energy supply (Jechoutek, 1992).
The same is likely to be observed with modern sources of fuel for cooking
such as LPG, where the poorest households are unable to afford even the
subsidised rates. Thus far, the “middle” and “upper” classes on the income
ladder have benefited the most, and, on the energy ladder, kerosene is
being replaced by LPG, but not “free” biomass. It seems unlikely that the
poor would leapfrog the lower rungs of the ladder unless “free” biomass is
no longer available, hence the drudgery of fuel collection and traditional
stove tending for the poorest has not been reduced.
• Other cooking fuel options? For the poorest people who cannot afford (to
purchase) LPG (or any other fuels), there obviously need to be options like
more efficient biomass-based stoves, but appropriate strategies for this that
are not being discussed in this report.
5. Experiences of LPG programmes

When considering the increased domestic use of LPG in India, lessons
could be learnt from the way LPG use was enhanced in other developing countries
and from regional programmes within India.
5.1 Experiences in other developing countries:
Brazil: In Brazil, although LPG distribution had begun with private

entrepreneurship, the entire production and import system was taken over by
Petrobrás, the state-owned national oil company in 1955. From 1975, LPG prices
have been cross-subsidised by higher gasoline and diesel prices. In addition, the
supply and distribution facilities were suitably enhanced. However, since
liberalisation of the sector in the 1990s, several international oil companies have
entered the market. Retail prices of LPG have been deregulated progressively
since 1998, although the Federal Government has retained its control over the
wholesale price at which Petrobrás sells LPG from its refineries, processing plants
and import terminals.
Brazil has been successful in providing LPG to about 90%38 of its

households. The main reason for this extent of adoption appears to be the
controlled price of LPG through cross-subsidies from other petroleum products.
This was proved in 2002, when de-regulation led to increases in LPG prices and
some lower-income rural households switched back to fuel-wood. To counteract
this, an assistance programme began, providing low-income families with
subsidies towards LPG purchase. In addition, smaller cylinders – of only 2 kg
each – have been made available, facilitating use among lower income households
37
These include Munasinghe, 1987; Cecelski, 1990; Jechoutek, 1992; Foley, 1990; Barnes, 1998.
38
This was computed, as a proportion of the total 46 million households (WRI, 2002) in Brazil.
(WLPGA and UNDP, 2002). Another important reason for Brazilian success in
replacing domestic fuel-wood use with LPG even in relatively remote areas is a
very dependable system of distribution and replacement of cylinders (UNDP, et
al., 2000, Chapter 10). However, as about 81% of Brazilian families reside in
urban areas (IBGE, 2001), the distribution problems in largely rural countries
would not be encountered.
Guatemala: In this Central American country, where the LPG market is

completely liberalized, instalment payment plans to cover the purchase of a
suitable stove and the cylinder deposit fee are common and are helping to
facilitate the adoption of this fuel by low/middle income families.
Indonesia: LPG for domestic use has been subsidised, but kerosene subsidies are
even higher, which undermines the competitiveness of LPG (WB&WLPGA,
2002).
West African region: 60% of the LPG consumption in this region is concentrated
in four countries -- Cameroon, Côte d’Ivoire, Ghana, and Senegal, where demand
has grown significantly during the 1990s. (The use of LPG in the other countries
of the region is considerably lower). Factors that have contributed to the increase
in demand in the case of Senegal, where the highest growth has been recorded,
include subsidised LPG to small cylinders39 (of 6 kg each), helpful for low-
income households, and also new participants in the market who have adopted
aggressive marketing strategies (WB&WLPGA, 2002). In both Senegal and Côte
d’Ivoire, price subsidies available to small cylinders have not been extended to
larger bottles, emphasizing the assistance to lower income households (WLPGA
& UNDP, 2002).
Vietnam: Market liberalisation including lifting of price controls in the early

1990s’40 resulted in a number of private distributors entering the LPG market.
Around 75% of sales are to the household sector.
The Philippines: The opening of the market in 1996 encouraged several oil
companies to invest there. Since 1997, more than 100 bottling plants have been
built and demand, almost entirely for the household sector, has risen by about
40% (WB&WLPGA, 2002).
China: In the People’s Republic of China, the shift up the energy ladder from
biomass-based fuels to LPG was spurred on by the restrictions on the supply of
kerosene (UNDP et al., 2000, Chapter 10). With liberalisation of the market, a
number of international oil companies have established distribution and marketing
operations, as joint ventures with the Chinese (WB&WLPGA, 2002).
Factors contributing to extension of LPG use:
39
Retail prices ranged (in 2000) from US$ 336/tonne to US$ 652/tonne.
40
Price ceilings were reintroduced temporarily between June 1999 and March 2001 in response to
a surge in import costs (WLPGA, 2002).
From the experiences summarised in this Section, the following factors
appear to have helped extend the domestic use of LPG (including lower income
households):
• Lower prices of LPG through cross subsidies from other distillates,
(particularly gasoline),
• favourable relative prices of LPG (in relation to competing fuels like
kerosene)
• initial cost financing (instalment payments for the purchase of stove and
cylinder deposit),
• smaller cylinders/bottles to target (lower income) households through
lower periodic/incremental refuelling bills
• special subsidies to these smaller cylinders/bottles – intended for lower
income groups
• restriction on the supply of competing fuels (e.g. kerosene)
• dependable distribution (reliable and more storage, bottling and refuelling
units)
5.2 Experiences of an LPG programme in India
An important scheme implemented for the expansion of domestic LPG use

has been the Deepam LPG scheme in the state of Andhra Pradesh. This project
was launched on the 9th July 1999 for the distribution of domestic connections to
women of below the poverty line (BPL)41 families in the rural areas of the state.
Each connection was accompanied by a one-off subsidy to the extent of the initial
cost, to overcome the barrier to fuel switching. It was meant to reduce
dependence on firewood, reduce the drudgery of collection of/cooking on
firewood, reduce pollution and improve the health of women. Salient features of
this scheme are:
! The scheme was administered by the State government Departments of Rural
Development and Civil Supplies and distributed through Public Sector Oil
Companies (Bharat Petroleum and Hindustan Petroleum).
! The High Court directed that the scheme be confined only to “white-
cardholders” (i.e. those below Rs 11,000/year/family).
! The Department of Rural Development identified the beneficiaries; a target of
1.154 million spread over 22 districts was indicated. Later, the numbers were
increased so that by 2002 about 1.724 beneficiaries (including some of the
urban poor) were listed.
! The lists were given to the LPG dealers of the oil companies, who were also
expected to ensure training of the allottees in the use of LPG stoves.
! The Department of Civil Supplies provided a one-time deposit of Rs
1,000/connection towards the cylinder and regulator.
! Results in terms of the number of connections allotted: till March 2002, 88%
of the urban target and 91% of the rural target had been met (NIRD, 2002).
Several lessons can be learnt from Deepam:
41
The Poverty Line is defined in terms of the cost of a certain basket of goods, in particular, a
specified level of calorie intake per capita in urban and rural areas in each state.
• The scheme was not very efficacious, because although all white-card holders
participated, over 80% of non-white card-holders in the region also did42!
• The retention rate was down to 85% in less than three years, in a sample of 52
villages and 18 municipal wards, because of cylinders having been given away
to relatives (including dowry to daughters), and being lent (!) to civil servants
in local areas (NIRD, 2002).
• Factors affecting the refill rate were: distance from distribution points, and the
season i.e., there is higher demand during the monsoons.
• (Participants’) perceived advantages of LPG were: timesaving, social status,
cleaner environment, and help during the monsoons. LPG was found useful
chiefly during the rainy season because of more employment (implying more
cash available for refuelling), more labour demand (and therefore less time for
firewood collection) and moisture making collection and preservation of
biomass difficult. (The scheme itself was considered attractive because of the
initial fee waiver).
• However, the perceived disadvantages were: implementation bottlenecks,
reduction in kerosene quota (in municipal areas), high refill costs (including
illegal commissions) of refills, and unwanted envy of non-beneficiaries.
Implementation bottlenecks within the scheme that contributed to
dissatisfaction included: limited choice, inability of suppliers to supply stoves
and accessories on time, co-ordination problems at the local level for the
supply arrangements, and irregularities with beneficiaries also having to incur
Rs 5 – 30 extra, per cylinder, for collection/delivery.
• Suggestions from local self help groups (SHGs) for improvement include:
credit for refills and reduction in cylinder size (reducing cash outflow per refill
although the cost/kg would increase).
• Most importantly, the fuel use pattern of Deepam beneficiaries has not
changed as much as intended: Wood remains the dominant fuel (for the main
meals), while LPG is used for additional cooking (tea, guests, etc.); crop
residues the third most important source, and kerosene the fourth -- used for
igniting the fire or in urban areas. LPG (average) use in these areas =
3kg/family/month and does not increase with the number of family members
and/or wage earners.
(In addition, Annexe 5 has information on the National improved stove

programme in India).
6. Issues for Indian domestic cooking fuels

In the context of the provision of appropriate cooking fuels, Indian
decision makers would have to first consider the choice of fuels. If the use of
LPG were to be encouraged, these would be issues concerning the provision/
delivery of LPG. For the longer term, alternative fuels would also have to be
considered.
42
Further, some “white-card holders” do not appear to be BPL, but that is a separate issue.
6.1 Choice of fuels
The advantages of LPG over the traditional biomass-based fuels are numerous
– reduced pollution and thereby improved health, reduced/avoided deforestation
and ecological damage, improved efficiency and reduced cooking time, and
reduced fuel collection time and effort. However, factors like the beneficial (or
reduction of harmful) effects on health are not being quantified or even included
in the households’ consideration (as it appears in the survey of households of the
Deepam scheme). Hence, that it would need some intervention or public
awareness drive to insert the “clean” fuel factor into the reckoning. Only obvious
cause-effect sequences like polluted water causing illnesses push people to pay for
alternatives (IEI, 2003).
Another phenomenon to be considered is that many households both in rural

and in urban areas use multiple energy sources for cooking. In these cases, the
social benefits of shifting to cleaner fuels in terms of improved health and time
saving accrue only partially, to the extent of the shift. The effects in these cases
have not been studied, but, in so far as a partial shift is a step towards a complete
shift, efforts to promote such action would be justified.
If the goal is to address the difficulty of obtaining fuel in rural areas where
biomass supply is getting scarce, then LPG promotion remains a worthwhile
strategy; further, with reduced demand for biomass from those able to shift to
another fuel, those still dependent on biomass for economic reasons, would be
helped.
However, the comparison need not necessarily be only with traditional fuels.
A study was made (CBA Energy Institute, 1996) chiefly for the comparison of
LPG with natural gas, but also for issues of urban air quality, etc, in Mexico, as
well as Brazil, China, and India. As LPG infrastructure can be more quickly
deployed and because LPG would be an improvement over wood and coal,
opportunities for increased LPG use were perceived.
6.2 Providing LPG
In view of the problems faced in the country (in Section 4) and the
experiences elsewhere (in Section 5), the following issues would have to be
considered when drawing up policies for the provision/delivery of LPG.
On the demand side, one would have to consider pricing (in particular, the
question of subsidies), financing options, and public awareness, and on the supply
side, security of supply, effective distribution/delivery, and regulation.
Demand issues
6.2.1 Pricing
6.2.1.1 LPG subsidies

When discussing the pricing of LPG in India, the most important issue is that
of the prevailing subsidies. Market forces are being recommended in most sectors
nowadays, but these affect affordability of LPG among lower income households.
If subsidies could be justified for this purpose, policy makers need to consider
several specific issues regarding the choice of subsidies and their funding.
• Choice of LPG subsidies: Choices have to be made from among the many
subsidy-options – either on the initial costs of connections/stoves, or on the
fuel, and either cross-subsidies or budgeted from the exchequer. In particular,
the following aspects should be considered:
o Initial (first-cost) subsidies – Subsidising initial costs seem preferable to

fuel (or refill) subsidies because the latter could encourage inefficient use
or could be diverted to other uses/users. A one-off fee-waiver on the
connection/stove makes sense when the barrier to adoption is the high
initial cost. However, first-cost subsidies leave possibilities for dropouts
from those who cannot afford the fuel costs, resulting in “dead”
investments, as noted in the case of the Deepam scheme in the state of
Andhra Pradesh.
o Operating (fuel) subsidies – If LPG refill subsidy is to be continued, some

precautions have to be taken:
• There could be rationing/quotas (quantitative limits) for the subsidised

fuel (as with ration cards) and/or coupons (as with food stamps).43
• There could be differentiated containers (say, smaller cylinders,
and/or cylinders painted another colour) for specific purposes (as with
subsidised kerosene currently being coloured blue), to prevent use by
those outside the scope of the planned benefits.
• Subsidies could be use-based (as with baseline tariffs for electricity)
with prices increasing with the level of consumption, rather than
across-the-board reduction in price that results in “subsidy capture”
(WEC, 2001) by wealthier sections of the population.
o Cross subsidies from other distillates – This has been the Indian practice
for many years, but would need to be weighed against the disadvantages of
higher costs of transport.
• Evaluation of subsidies: Even when justified for social/environmental

benefits, subsidies should be appropriate. Before introduction, subsidies
should be evaluated in terms of efficiency (cost-benefit analysis of welfare
gained versus the distorting effects and the costs of the subsidy), efficacy
(targeting success in reaching those for whom it is intended, avoiding errors of
inclusion of those who should not be benefited and exclusion of those who
should), and cost-effectiveness (i.e. administrative costs should not be
prohibitive) (WB, 2000).
43
There could also be time-limits (sunset clauses) for such subsidies, but this may not be
practicable as it is often politically infeasible to remove such benefits.
• Funding of subsidies: The source of funds for the subsidies would have to be
one/more from among -
• LPG companies themselves, through a mandate of the government,
requiring the providers to sell below their costs, as in the present Indian
situation, but this has to be temporary or else there could be financial
disasters (as happened with the State Electricity Boards);
• regulated cross-subsidies from one consumer category to another -
effective as long as the funding category’s price elasticity is not too high
as to curtail sales;
• progressive tariffs with the price per unit increasing with the amount
consumed: the more affluent customers who use more, pay more, but this
would need the upper segment to be large enough to support the lower
segments and could be considered akin to cross-subsidies from higher
income consumers to the others.
6.2.1.2 Pricing of competing fuels
When evaluating the pricing of LPG, one has to consider the relative
prices of these fuels and whether or not inter-fuel shifts are desirable.
• Subsidies to kerosene: Reducing/removing the subsidy on kerosene could

make LPG relatively cheaper, without a burden on the exchequer. Thus far,
subsidies have been higher for kerosene than for LPG; in 1998 when the APM
dismantling was initiated, LPG subsidy was about 32% while the kerosene
subsidy was more than 50% (MoP&NG, 2003a). However, in the near term,
or as long as homes are not electrified, subsidy to kerosene has to merit
consideration because it is the source of lighting for about 43% of the
population (according to the household data from the Census of India, 2001).
• Relative efficiencies: If the relative costs of LPG vis-à-vis other fuels were
reckoned after accounting for their calorific values and the efficiencies of the
related stoves, LPG would not seem as expensive44 (as was shown in Figure
1).
6.2.1.3 Direct cash benefits instead of subsidised fuel
There could be schemes through which LPG is priced at its full cost, but
targeted households get some pre-determined compensation. This would avoid
careless use of the fuel (and may also be an incentive for fuel efficiency), while
assisting the economically disadvantaged. Such programmes would require
funding from the government - with transfer payments directly to the poor, but the
better the targeting, the higher the administrative costs, and experiences with BPL
schemes have shown that those not entitled manage to get themselves included.
44
With lighting improvement, payments for the improved source (electricity) are less than those
for the earlier source (kerosene lamps) because of the much greater efficacy of electric lighting.
6.2.2 Marketing (financing and packaging) schemes
There are several marketing schemes that encourage the purchase of consumer
durables by lowering the amount of each cash outflow. Similar methods could be
used to help lower income households in the case of LPG. Instalment payments
for the cost of connection and stove, and each fuel refill in much smaller
containers (e.g. 2 – 5 kg, instead of the regular 14.2 kg cylinders), will reduce the
“lumpiness” of successive cash outlays. The latter option has been launched by
the Public Sector companies but needs to be extended beyond limited areas.
6.2.3 Public awareness
Awareness of the adverse impacts on health of indoor pollution and the

benefits of “cleaner” fuels would increase their popularity and thereby the
willingness to pay.
Supply issues
6.2.4 Supply security
Supply security implies uninterrupted availability of LPG. Since various

deficiencies exist in the present system, we require:
• adequate and well dispersed import facilities,
• well dispersed indigenous LPG processing plants (refineries and natural
gas fractionating plants),
• storage capacities throughout the country, and
• multi-mode transport facilities for moving LPG from alternative
destinations.
6.2.5 Dependable distribution network
The LPG distribution network also needs to be improved – or else bottlenecks

hamper the delivery flow:
• Distributors face unfavourable economies of scale when the demand is low

or dispersed. The problems of consumers whose location precludes them
from enjoying the facility have to be addressed through extension of the
distribution network beyond urban and semi-urban areas.
• There should be complementary infrastructure – roads, equipment
suppliers, repair services, etc. – built in tandem, to facilitate the smooth
operation of the system. This would be analogous to the rationale for
improving rural infrastructure along with electrification.
6.2.6 Regulation
The government’s roles in setting standards to maintain safety and avoid

corruption are essential. In Brazil, the LPG industry and the government had
to introduce a code of practice in 1996 to improve the quality of service and
safety of the system particularly with respect to the standard of the cylinders.
Measures for ensuring that the cylinders are checked for their user-worthiness
and are properly filled have to be in force. Consumer protection has to be
provided, particularly as with a large number of operators (distributors) and
poor enforcement of standards, accidents and commercial malpractice can
occur.
Currently, the UNDP and the World LPG Association (WLPGA) have a
partnership/initiative called the LPG Challenge to address concrete barriers to
meeting the thermal energy needs of rural and peri-urban populations through
the expanded use of LPG (UNDP, 2002)45. Additional factors identified
through the project could be included.
The government has to be involved, at least through its policies, in

helping to provide energy services to the economically disadvantaged. But
there has also to be a suitable environment for the private sector to cater to
those who can pay for their needs. Subsidies will continue to be necessary for
a while, but have to be applied with care. Development assistance/grants –
from aid agencies, etc. could help only small fractions of the population;
which means that the government and market forces have to handle the rest
and their extent and effectiveness have to be expanded to meet current and
growing needs.
6.3 Non-conventional alternatives
It is important to reiterate that LPG is a fossil-based fuel and as such

cannot be considered a sustainable source in the long term; it is being
recommended as a part of the transition to renewable energy sources such as
modern biomass-based fuels, till such times as these technologies become
affordable, accessible, available and acceptable. Some of these biomass-based
fuel options are listed below:
Biogas from animal waste: In areas where cattle are kept extensively (for
example for dairying), biogas (CH4 and CO2, in the ratio 3:2) can be generated
from cattle dung, if adequate amounts can be supplied daily to the digester.
India’s largest biogas plant has been running since April 1987 in the village of
Methan (Sidhpur tehsil, Patan district, in the state of Gujarat); the plant,
consisting of eight digesters, has a total capacity of 630 m3, and caters to the
main cooking fuel requirements of 320 families (Jamwal, 2003). However,
the supply of dung by villagers to the plant has been found to be inadequate to
meet the village cooking needs in many other villages; in fact, it has not been
enough even for running a dual-fuel (biogas-diesel) generation plant for the
electricity and water-pumping needs of the village (IEI, 2003). Family-size
biogas plants are operating successfully and constitute a feasible option
(wherever the cattle owned are adequate). About 3.482 million family-size
biogas plants have been constructed in the country (MNES, 2003), of which an
45
UNDP has initiated pilot projects in different regions. Specific targets will have two basic
categories – affordability and availability. UNDP hopes to undertake this partnership in not less
than 10 countries.
estimated 80% are operating successfully (AFPRO-CHF, 1997). The
efficiency of biogas stoves has been found to be higher than other available
alternatives (Smith, et al., 2000).
Other options:
Gasification: Where adequate crop residues are available from the crops
cultivated in the area, crop-residue can be gasified to produce carbon
monoxide and hydrogen, combustible gases that could be used for cooking or
for power generation, (Mukunda et al., check, Henderick and Williams, 2000,
Shyam, 2002). Even where crop residues are not normally available,
plantations can be started on fallow/degraded lands (to avoid competing with
agriculture), for biomass generation (Larson and Kartha, 2000).
New options (not yet field-tested in India): The amount of household cooking
fuel that could be produced from the biomass assigned for the purpose
depends on the particular fuel considered and the conversion technologies
employed; possibilities include ethanol, di-methyl ether (DME), and synthetic
LPG, but these have yet to be tried in India.
In general, the country’s strategies of fostering economic growth and

employment opportunities need to be focused on and accelerated because they
would bring in the collateral benefits of the use of better domestic fuels.

Annexe 1:
Technical details of LPG
LPG consists of hydrocarbons that are gaseous at normal atmospheric pressure,

but can be condensed to the liquid state at normal temperature, by the application of
moderate pressure. LPG is derived from two sources: from the processing of natural
gas streams produced either alone or in association with crude oil, and from crude oil
refining. Worldwide, natural gas processing currently accounts for roughly 60% of
total marketed LPG supply and crude oil refining for the remaining 40%
(WB&WLPGA, 2002).
Figure 8: Crude oil refining process
Distillation is the first step in the processing of crude oil and it takes place in a tall
steel tower called a fractionation column. The inside of the column is divided at
intervals by horizontal trays. The insulated column is kept very hot at the bottom, but
as different hydrocarbons boil at different temperatures, the temperature gradually
reduces towards the top, so that each tray is a little cooler than the one below.
The crude oil needs to be heated up before entering the fractionation column and
this is done at first in a series of heat exchangers where heat is taken from other

process streams that require cooling before being sent to rundown. Heat is also
exchanged against condensing streams from the main column. Typically, the crude
will be heated up in this way up to a temperature of 200 - 280 0C, before entering a
furnace.
As the raw crude oil arriving contains quite a bit of water and salt, it is normally
sent for salt removing first, in a piece of equipment called a desalter. Upstream from
the desalter, the crude is mixed with a water stream, typically about 4 - 6% on feed.
Intense mixing takes place over a mixing valve and (optionally) as static mixer. The
desalter, a large liquid full vessel, uses an electric field to separate the crude from the
water droplets. It operates best at 120 - 150 0C, hence it is conveniently placed
somewhere in the middle of the preheat train.
A part of the salts contained in the crude oil, particularly magnesium chloride, are
hydrolysable at temperatures above 120 0C. Upon hydrolysis, the chlorides get
converted into hydrochloric acid, which will find its way to the distillation column's
overhead where it will corrode the overhead condensers. A good performing desalter
can remove about 90% of the salt in raw crude.
Downstream from the desalter, crude is further heated up with heat exchangers,
and starts vaporising, which will increase the system pressure drop. At about 170 -200
0
C, the crude will enter a 'pre-flashvessel', operating at about 2 - 5 bar, where the
vapours are separated from the remaining liquid. Vapours are directly sent to the
fractionation column, and by doing so, the hydraulic load on the remainder of the
crude preheat train and furnace is reduced (smaller piping and pumps).
Just upstream the preflash vessel, a small caustic stream is mixed with the crude,
in order to neutralise any hydrochloric acid formed by hydrolysis. The sodium
chloride formed will leave the fractionation column via the bottom residue stream.
The dosing rate of caustic is adjusted based on chloride measurements in the overhead
vessel (typically 10 - 20 ppm).
At about 200 - 280 0C the crude enters the furnace where it is heated up further to
about 330 -370 0C. The furnace outlet stream is sent directly to the fractionation
column. Here, it is separated into a number of fractions, each having a particular
boiling range.
At 350 0C, and about 1 bar, most of the fractions in the crude oil vaporise and rise
up the column through perforations in the trays, losing heat as they rise. When each
fraction reaches the tray where the temperature is just below its own boiling point, it
condenses and changes back into liquid phase. A continuous liquid phase is flowing
by gravity through 'downcomers' from tray to tray downwards. In this way, the
different fractions are gradually separated from each other on the trays of the
fractionation column. The heaviest fractions condense on the lower trays and the
lighter fractions condense on the trays higher up in the column. At different elevations
in the column, with special trays called draw-off trays, fractions can be drawn out on
gravity through pipes, for further processing in the refinery.

At the top of the column, vapours leave through a pipe and are routed to an
overhead condenser, typically cooled by air fin-fans. At the outlet of the overhead
condensers, at temperature about 40 0C, a mixture of gas, and liquid naphtha exists,
which is falling into an overhead accumulator. Gases are routed to a compressor for
further recovery of LPG, while the liquids (gasoline) are pumped to a hydrotreater
unit for sulphur removal.
A fractionation column needs a flow of condensing liquid downwards in order to

provide a driving force for separation between light and heavy fractions. At the top of
the column this liquid flow is provided by pumping a stream back from the overhead
accumulator into the column. Unfortunately, a lot of the heat provided by the furnace
to vaporise hydrocarbons is lost against ambient air in the overhead fin-fan coolers. A
clever way of preventing this heat lost of condensing hydrocarbons is done via the
circulating refluxes of the column. In a circulating reflux, a hot side draw-off from the
column is pumped through a series of heat exchangers (against crude for instance),
where the stream is cooled down. The cool stream is sent back into the column at a
higher elevation, where it is been brought in contact with hotter rising vapours. This
provides an internal condensing mechanism inside the column, in a similar way as the
top reflux does which is sent back from the overhead accumulator. The main objective
of a circulating reflux therefore is to recover heat from condensing vapours. A
fractionating column will have several (typically three) of such refluxes, each
providing sufficient liquid flow down the corresponding section of the column. An
additional advantage of having circulating refluxes is that it will reduce the vapour
load when going upwards in the column. This provided the opportunity to have a
smaller column diameter for top sections of the tower. Such a reduction in diameter is
called a 'swage'.
The lightest side draw-off from the fractionating column is a fraction called
kerosene, boiling in the range 160 - 280 0C, which falls down through a pipe into a
smaller column called 'side-stripper'. The purpose of the side stripper is to remove
very light hydrocarbons by using steam injection or an external heater called 'reboiler'.
The stripping steam rate, or reboiled duty is controlled such as to meet the flashpoint
specification of the product. Similarly to the atmospheric column, the side stripper has
fractionating trays for providing contact between vapour and liquid. The vapours
produced from the top of the side stripper are routed back via pipe into the
fractionating column.
The second and third (optional) side draw-offs from the main fractionating
column are gas oil fractions, boiling in the range 200 - 400 0C, which are ultimately
used for blending the final diesel product. Similar as with the kerosene product, the
gas oil fractions (light and heavy gas oil) are first sent to a side stripper before being
routed to further treating units.
At the bottom of the fractionation column a heavy, brown/black coloured fraction

called residue is drawn off. In order to strip all light hydrocarbons from this fraction
properly, the bottom section of the column is equipped with a set of stripping trays,
which are operated by injecting some stripping steam (1 - 3% on bottom product) into
the bottom of the column.

LPG produced from straight distillation consists of “saturated” hydrocarbons, i.e.
propane and butane, whereas LPG produced by both cracking and reforming
processes has, in addition to hydrocarbons, some quantities of unsaturated
hydrocarbons also (i.e. propylene and butylene). There is also moisture and some
impurities (such as sulphur compounds) that -are removed by suitable treatment at the
refinery. LPG burns cleanly, producing no particulate matter, with low emissions of
CO, unburned hydrocarbons and NOx, and less CO2 than most other fossil fuels and
less than unsustainable biomass. The exact composition of LPG can vary but it
usually consists predominantly of propane (C3H8) and butane (C4H10), with a small
proportion of propylene (C3H6) and butylene (C4H8). Commercial LPG also contains
traces of lighter hydrocarbons like ethane (C2H6) and ethylene (C2H4) and heavier
hydrocarbons like pentane (C5H12). LPG marketed in India conforms to Indian
Standard Specification IS-4576.
Table 22: Properties of LPG
Propane Butane
Chemical formula C3H8 C4H10
Liquid Density 0.505 0.575
Gas Density 1.5 1.95
Ratio Gas/liquid 274 230
Atm. Boiling ptc. -42 -2
Specific heat liquid 0.60 Btu/deg. 0.58 Btu/deg
Latent heat Vaporization 358 kj/kg. 372 kj/kg
Flammability limit 2.2 - 9.5% 1.8 - 8.5%
Auto temp ign 470ºC 410ºC
44.10
Mole Weight 58.12
kg/k/mole
Freezing Point -187.7ºC -138.4
Critical temp 96.7ºC 152.1ºC
Critical Press 42.5 bar 38.0 bar
Litres/tonne 1965-2019 1723-1760
Octane number <100 92
Relative density of liquid 537-543 406-431
Maximum flame temperature 1980 1990
Ratio of gas volume to liquid volume 274 233
Soluble in water Slight Slight
Colour Colourless Colourless
Normally used as gas, LPG is stored and transported as liquid under pressure for
convenience and ease of handling; liquid LPG evaporates to produce about 270 times
its volume of gas. This facilitates storage and transportation in relatively small
containers. In addition, unlike traditional fuels and other liquid fuels, LPG has an
indefinite shelf life, not deteriorating over time. (Adapted from Cheresources, 2002)
LPG at about 45.5GJ/tonne, has a higher energy content than the fuels currently in
use for cooking – kerosene (43.2 GJ/tonne), fuel-wood (about 15 GJ/tonne), crop
residues (13 – 14 GJ/tonne) and dung (12.5 – 13 GJ/tonne). In addition, the higher
efficiency of LPG stoves (about 65%) as compared with traditional stoves (about
15%) and even “improved” models of biomass-based stoves (up to 45%), makes the
relative efficiency considerable.

Annexe 2:
Fuel Analysis
Solid fuels and kerosene were analysed for carbon, ash, sulphur, nitrogen and
hydrogen content using standard methods (BIS 1987). For LPG, the energy content
was given by Bharat Petroleum Company Ltd. (BPCL). The chemical composition,
moisture content and net (low heating value) energy of the fuels are given in Table 23,
using the method described below.
Table 23: Fuel chemical composition, moisture content, and net energy
Fuel Moisture Net Energy

content (kJ/kg) Carbon Nitrogen Ash H2 Sulfur
(%)
LPG - 45837 86.0
Biogas - 17707 39.6 6.5
(kJ/M3)1
Kerosene - 43116 84.3 0.02 0.0 14.2 0.04
Eucalyptus 6.1 15333 45.4 0.14 0.4 6.4 0.02
Acacia 6.5 15099 41.8 0.35 2.89 6.3 0.01
Root fuel 5.7 15480 51.8 1.18 7.0 4.5 0.08
Charcoal 1.7 25715 80.0 0.69 7.4 1.8 0.06
Char- 7.2 15928 50.3 0.25 40.0 3.2 0.05
briquette
Mustard 5.9 16531 42.1 0.36 2.7 6.3 0.01
straw
Rice straw 8.8 13027 38.1 0.40 15.6 6.2 0.05
Dung cake 7.3 11763 33.4 0.90 52.2 3.9 0.07
1
standard temperature and pressure
Basis of calculation:
Moisture content (wet basis): To determine the moisture content of any fuel it is
necessary that it should be of small particle size. The wood was sawed to make
sawdust in such a way that the whole area, including cell wall, was included. About
five pieces of the fuel samples taken from different places were sawed and the
sawdust obtained were mixed properly and used for moisture content measurement.
These steps were all carried out in triplicate.
A known quantity of sample was taken in a crucible and kept in an oven maintained at
105 o C till the weight stabilizes. The weight loss was measured and the moisture
content of the sample was estimated as follows.
% Moisture Content (M.C.) = (W1 – Wf)/ (W1-Wc) *100
W1 = initial weight of sample

Wf = final weight of sample
Wc = weight of crucible

Calorific value: Calorific value (energy content) of a fuel was determined by
calorimetry.
Benzoic acid was used to standardize the bomb calorimeter. One gram of sample was
taken in a crucible and made into a pallet and the initial weight was noted. It was
placed in the bomb, which was pressurized to 18 atm of oxygen. The bomb was
placed in a vessel containing a measured quantity of water. The ignition circuit was
connected and the water temperature noted. After ignition the temperature rise was
noted every minute till a constant temperature was recorded. The pressure was
released and the length of unburned fuse wire was measured. The calorific value was
calculated as:
((tc x w) - (m+n))/weight of sample(g) = kj/kg = Hw

tc = temperature rise ( C)
w = apparent heat capacity by benzoic acid (J)
m = calorific value of thread (J)
n = calorific value of Nichrome ignition wire (J)
The apparent heat capacity by benzoic acid (w), calorific value of thread (m), and the
calorific value of Nichrome ignition wire were provided by the instrument supplier.

Annexe 3:
A comparison of the annualised costs of cook-stoves (in India)
<-------------------- all at 12% discount rate ----------------------------->

wood/crop waste <-- kerosene ----> <--------- LPG ---------> Elect.
trad. improved PDS mket subsidised market
fuel fuel fuel fuel
STOVE PRICE (a) (Rs) 10 150 400 400 1800 1800 1500
USEFUL LIFE OF EACH 3 3 5 5 15 15 10
STOVE (years)
DEPOSIT OR ONE-TIME 750 750
PAYMENT (Rs)
INTEREST (discount) RATES 12 12 12 12 12 12 12

(%)
CAPITAL RECOVERY 0.416 0.416 0.277 0.277 0.147 0.147 0.177

FACTOR
ANNUALISED CAPITAL COST 4.16 62.45 110.96 110.96 264.28 264.28 265.48
(b) (Rs)
ENERGY CONTENT OF THE 15 15 35 35 45.5 45.5 3.6

FUEL (MJ per kg, litre, or kWh)
EFFICIENCY OF STOVE (c) 15% 30% 45% 45% 60% 60% 71%
ANNUAL FUEL USAGE 1395 698 199 199 115 115 1223
(litres/yr., kg/yr., kWh/yr.) (d)
PRICE OF FUEL (Rs/litre, Rs/kg, 1.00 1.00 11.00 16.50 18.52 27.65 3.00
Rs/kWh) (e)
ANNUAL FUEL COST (Rs) 1,395 698 2,193 3,289 2,130 3,179 3,669
ANNUAL MAINTENANCE 0.00 0.00 25.00 25.00 75.00 75.00 0.00

EXPENSES assumed nil (Rs)
=> TOTAL ANNUALISED 1,399 760 2,329 3,425 2,469 3,519 3,935
COSTS PER STOVE (Rs)
------------------------------------
Please note:
US$ = Rs 45 (November 2003)
(a) Stove prices refer to the market prices prevailing in Bangalore.

(b) Annualised cost = cost x capital recovery factor (CRF), where CRF = i/{(1+ 1/i)}n ; discount rate (i) here
is assumed = 12%
(c) The efficiencies of stoves are from "Bioenergy: Direct Applications in Cooking" by G.S.Dutt and
N.H.Ravindranath, (Table 10, p.676) in Renewable Energy, 1993 and from NCAER's "Energy Demand in
Greater Bombay", 1975, quoted in TEDDY, 1996-97.
(d) The annual fuel usage was entered for LPG connections (= average usage per connection according to
the oil companies' sales figures) and that of the other fuels was derived thus:
(MJ/kg x efficiency x kg/year)LPG / (MJ/kg x efficiency)other = (kg/year) other
(e) Market-level fuel prices are also from Bangalore;
subsidised prices of kerosene through the PDS (public distribution system) through which specified amounts
of fuel per household are provided, are limited to 24 litres per family per year for regular card holders and
120 litres per “green card” holder.
Annexe 4:
India’s Administered Pricing Mechanism (APM) for the Petroleum Sector
(based on information from MoP&NG, 2003a)
Evolution of APM
Till 1939, there were no controls whatsoever on the pricing of petroleum

products. Between 1939 and 1948, the oil companies themselves maintained pool
accounts for major products without any intervention by the government. In 1948,
an attempt was made to regulate prices through Valued Stock account procedure.
Under this procedure, realisation of oil companies was restricted to the import
parity price of finished goods (with Ras Tanura as the basing point), plus excise
duties/ local taxes/ dealer margins and agreed marketing margins of each of the
refineries. Any excess realization was surrendered to the Government. The
Shantilal Shah committee, set up in 1969, did not favour the import parity price
being set as a benchmark for domestic pricing as domestic refining capacity had
significantly increased by then. In 1976, the Oil Pricing Committee (OPC)
recommended the discontinuance of the import parity principle on the following
grounds:
• About 90% of the total demand of POL products were met by indigenous
production and no major shortfall was anticipated.
• Prices of finished products and crude oil did not necessarily move in tandem.
• Import parity did not take into account inter-refinery differences in terms of
product pattern, type of crude used, location and scale of operation.
• The structure of West Asian product prices, which was the basis of
determining prices in India, did not necessarily reflect the cost pattern and
operations of Indian refineries.
The OPC therefore suggested that the domestic cost of production should be
the determining factor for pricing of petroleum products.
The Administered Pricing Mechanism (APM) in existence until 1998, was

evolved on the recommendation of the OPC and came into existence on December
16, 1977. The smooth implementation of APM was possible, as by then, all the
foreign oil companies were acquired by the Government of India.
Rationale for APM
One of the important drawbacks of the import parity pricing was that the
indigenous cost of production was totally overlooked while determining producer
prices. This issue was addressed through Retention Pricing Mechanism, by which
refiners were allowed to "retain" out of the sale proceeds,
• Cost of crude
• Refining cost and
• Reasonable return on investment.
The same mechanism was extended to marketing & distribution companies as
well. The Government of India also fixed the pricing of finished products and the
returns of oil companies were de-linked from the price at which the goods were
finally sold. With the administration of pricing of products by the government, the

retention mechanism also came to be known as the Administered Pricing
Mechanism or APM.
The scheme was administered under the aegis of the Ministry of Petroleum &
Natural Gas through its executive wing "Oil Co-ordination Committee" (OCC)
with its secretariat at New Delhi.
Objectives of APM
• To optimise the utilisation of refining and marketing infrastructure by treating

the facilities of all the oil companies as common industry infrastructure, the
access of which would be available to all the oil companies by hospitality
arrangements, thus eliminating wasteful duplication of investment.
• To make available all products at uniform price ex-all refineries so as to
minimise cross-haulage of products & associated energy costs.
• To ensure continuous availability of products/ crude to refiners by recognising
import needs wherever there are deficits in indigenous production.
• To ensure that the returns to oil companies are reasonable, in line with
operational efficiencies as also generation of sufficient resources to enable
industry to set up facilities to meet the growing needs.
• To ensure stable prices by insulating domestic market from the volatility of
prices in the international market.
• To achieve socio-economic objectives of the Government by ensuring
availability of certain products at subsidized rates for weaker sections of the
society and priority sectors in the industry through cross-subsidization of
products.
Functioning of APM
The basic principles on which the edifice of APM was built can be
summarised thus:
• Raw materials were made available at a pre-determined fixed price at the

manufacturing point (Delivered Cost of Crude) on a sustained & continuous
basis to refiners. Similarly, finished products were made available to
marketing companies at pre-determined prices (Ex-refinery prices).
• Refining/ conversion/ marketing costs were reimbursed as per certain pre-
determined criteria.
• Compensation for investments in fixed assets and working capital was given
as per laid down norms.
• Rewards and penalties were built into the system to encourage efficiency.
Retention price
The oil companies were reimbursed in addition to the cost of crude oil
• Operating costs
• Return on capital employed

The Oil Coordination Committee (OCC) undertook a cost updating study of
each of the oil companies, once every Pricing Period of three years. The first year
in the pricing period was called the base year. The exercise was normally
undertaken in the middle of the pricing period and completed at the end of the
pricing period. The costs incurred during the said period, including projections
for the pricing cycle, were collated for each of the oil companies and ad-hoc
margins were worked out first and thereafter replaced by final margins. It may be
noted that not all costs were reimbursed and the expert committee of OCC
moderated the actual costs. The margins for the pricing period were worked out
by pro-rating the aggregate costs over the standard throughput/ sales volumes as
per the sales plan entitlement (SPE) to arrive at operating cost per unit. The
operating cost so arrived would be static during the pricing period excepting for
permitted escalations which were considered for reimbursement on the merits of
each case (e.g., increases in salaries/wages on account of long term settlements,
increases in the direct variable costs such as chemicals, catalysts, utilities, etc.).
The companies were also eligible for a return on their total capital
employed, consisting of average net fixed assets and normative working capital.
APM for refineries
Standards were laid down for each refinery with respect to throughput,
product pattern, fuel and losses. The standard throughput was fixed after taking
into account the crude availability, the primary/ secondary/ offsite facilities, intake
capacity and other technical factors. For a new refinery, the standard was 60% of
the installed capacity in the first year of operation and 90% in the second year of
operation.
Based on the aggregate operating cost (OC) and return on capital employed
(ROCE) standards so set, the OC and ROCE per unit of crude throughput was
worked out for each of the refineries, for the relevant pricing period.
The Delivered Cost of Crude (DCC) for imported crude was worked out
for each of the refineries, on the basis of pooled free-on-board (FOB) cost of
crude, freight, insurance, ocean loss, wharfage/other landing charges and customs
duty.
The difference between the landed cost of crude and the DCC could be claimed
from the Pool account, subject to the following restrictions.
• Actual cost of insurance was limited to a maximum premia based on free
particular average clause including war risk premia.
• Ocean loss for imported crude was taken as 0.5% of the bill of lading quantity,
0.2%/ 0.3% of Bombay High custody transfer quantity by West Coast / East
Coast refineries. Variation in actual quantity of losses vis-à-vis the norms
would benefit or adversely affect the refineries.
The retention price per tonne of crude for each of the refineries was thereafter
worked out by cumulating the DCC, the operating cost (OC) and the ROCE.
While working out the operating cost, the following amounts were reduced as the
same was recovered from the marketing companies separately in addition to the
ex-refinery prices.
• Rs 50/tonne of LPG filled in bulk.
• Rs 200/tonne of LPG filled in packed cylinders.
The total amount of reduction was worked out by multiplying the aforementioned
rates with the standards set for the pricing period.
The retention price per tonne of crude so computed was then pro-rated over
various products, as laid down in the standard product pattern through a set of
indices laid down by OCC. While calculating the retention prices of the products,
the cost of fuel and loss was spread over all the products, based on indices
developed after taking note of the current supply & demand position. These were
the prevailing international prices of various petroleum products, need to
encourage production of deficit products and conversely to discourage production
of surplus products, and other factors affecting the distribution and allocation
efficiency. The role of the indices was limited to determination of the product
prices of refineries; this had little bearing on the final consumer prices.
APM for marketing and distribution – with reference to LPG
Marketing of petroleum products is done by oil companies through a

network of storage and distribution facilities which include installations, depots,
LPG bottling plants, airfield stations (AFS), retail pump outlets (RPOs) and sales
offices spread across the country.
Operating costs till ex-storage level:
Under APM, the operating costs to be reimbursed up to the ex-storage level

were broken up as:
• Installation cost
• Distribution cost
• Administration cost
The installation & distribution cost were disaggregated into common costs and
specific costs.
Specific costs represented the cost of product losses incurred at the

installation and distribution stage and were determined as per the given norms.
For example, for LPG, distribution losses of 0.25% were permissible. Specific
costs were computed by multiplying the aforementioned percentage to the sum of
ex-refinery price and excise duty of each product. Specific costs were uniform for
all the oil companies and therefore if a company was able to reduce its incidence
of loss, it would gain. On the contrary, if its losses were more than the norm, it
would lose.
All operating costs other than specific costs were categorized as common
costs. Since the common costs were bound to be different from one company to
another, the actual reimbursement would differ from one company to another.
The allowable costs for the pricing period were collated and the total cost was pro-
rated over volumes to arrive at a per kilolitre (kl) cost or per metric tonne (mt)
cost.

In line with the procedure for the return on capital employed at the refinery
stage, the return on capital employed up to the ex-storage stage was worked out.
Capital employed to the extent of net worth would earn 12% post-tax return and
balance if any would be treated as deemed borrowings on which the weighted
average cost of borrowings would be given.
The marketing margin at the ex-storage point would thus include the
installation, distribution (both specific & common), and administration costs and
the return on capital employed and this would be the retention margin per selling
unit. The weighted average marketing margin of all the oil companies was
computed and included in the selling price, and the oil companies would adjust the
differential between the retention margin and the marketing margin included in
the selling price in the Pool account.
LPG filling, cylinder compensation & LPG pricing
Packed LPG is being marketed in cylinders of several sizes - 14.2 kg, 19

kg and 50 kg from the Public Sector Units (PSUs) and 12 kg and 17 kg from
private sector distributors. While 14.2 kg cylinders are supplied for domestic
consumers only, the others are for non-domestic consumers. The selling prices of
LPG for domestic consumption are subsidised, but for other uses the selling price
is determined on an import parity basis.
For each refinery, standard LPG filling norms were set. For all fillings up
to the standard, each refinery would be entitled to a uniform filling margin of Rs
200 per mt for packed LPG and Rs 50/ mt for LPG sold in bulk. If LPG filling
exceeded the standard, the refineries were eligible to retain Rs 50/ mt of
incremental LPG packed and the balance amount of Rs 150/ mt was surrendered
to Pool account. There is no penalty however for not filling up to the standard.
As stated earlier, the filling margin recovered on LPG was deducted from
the refining cost while computing retention margins, and the amount so deducted
was restricted to the standard. Thus the additional margin of Rs 50 per mt would
accrue as an incentive for the refining companies.
In respect of bottling plants other than refineries, operating cost excluding

depreciation was reimbursed uniformly on the basis of industry average cost.
Depreciation cost and return on capital employed were computed for each of the
refineries and the retention margin was worked out for each of the oil companies
by aggregating the operating cost, depreciation cost and return on capital
employed. The weighted average filling margin of refining & marketing
companies was built into the selling price and the difference between the margin
recovered through the selling price and the retention margin would be adjusted in
the Pool account. With respect to sales effected out of the bottling done by
refining companies, the difference between the margin recovered through the
selling price and the margin paid to refining companies was surrendered to Pool
a/c. Thus, marketing companies whose operating cost other than depreciation were
below the industry average were bound to gain and those whose costs were above
the industry average would lose.

In addition to the filling margin, marketing companies were entitled to a
uniform marketing margin of Rs 640/ mt of LPG packed, recovered through
selling price, to cover the following expenses:
• Depreciation on LPG cylinders/ regulators: Rs 252/ mt
• Return on Net investment in cylinder/ regulators: Rs 196/ mt (Net investment
meaning actual procurement cost minus deposits received from consumers)
• Repairs & maintenance: Rs 192/ mt
The depreciation included in the marketing margin represented 1/15th of

cylinder cost/ regulators, while 100% depreciation for cylinder/ regulator was
charged off in the accounts. To compensate oil companies for this depreciation
cost, companies were permitted to claim the differential between procurement
cost, depreciation & return element included in the marketing margin from the
pool a/c. Each new enrolment would bring in a deposit (of Rs 900, Rs 1,500 and
Rs 2,000, for 14.2, 19 kg and 50 kg cylinder, respectively) and Rs 100 (per
pressure regulator). Also, 100% of the cost of cylinders qualified for depreciation
under the Income-Tax rules, hence actual cash inflow to the oil companies for
every new enrolment was nearly 2.35 times the actual cash outflow. For example,
for every Re 1 invested in a cylinder, Re 1 from the Pool a/c towards depreciation,
approximately Re 1 from consumers in the form of deposits, and Re 0.35 being
the tax saving on account of depreciation. Since the depreciation cost reimbursed
was treated as an income, the net cash flow after reducing the impact on such
income was twice that of the investment. In respect of replacements, as no deposit
was received, the cash inflow was equal to the cash outflow. Thus the LPG
business was the most lucrative among the APM products, both in terms of profit
and cash generation.
To ensure uniform pricing, the commission payable to the distributors is

determined by the Government of India. The formula for calculation of
distributor's commission as on April 1, of every year is the same as applicable to
other (petrol/diesel) dealers, except for the slabs and factors which were, for 14.2
kg domestic (pkd) cylinders:
• Slab I - till 3000 refills per month: factor = 0.31
• Slab II - beyond 3000 refills per month: factor = 0.33
No such bifurcation regarding slabs was made for 19 kg and 50 kg refills.

Unlike the case of petrol and diesel, distributor commission is not revised with
changes in administered prices.
For LPG supplied at centres other than refinery points, notional rail freight
(NRF) applicable for bulk LPG was recovered through the selling price from the
nearest refinery to the bottling plant located in the upcountry centre. The
difference between the actual freight and NRF could be claimed by the oil
companies from the Pool a/c. Even if the product were supplied from a point
other than the contiguous refinery point, the difference in transportation charges
could also be claimed from the Pool a/c.

Surcharges
In addition to claims/ surrenders that are self-balancing, oil companies were

entitled to several other claims like crude oil price differential, imported product
price differential, differential freight etc. The oil pool had to generate funds to
meet these claims and the same was done through levy of surcharges such as Cost
& Freight surcharge, Freight surcharge pool surcharge, Retail pump outlet
surcharge and State surcharge.
Product price adjustment
In addition to these surcharges, the Government of India tried to achieve

its objective of ensuring availability of certain products at subsidised rates for
weaker sections of the society and priority sectors in the industry through cross-
subsidisation of products. The cross subsidisation was done through product price
adjustment (PPA) by which a higher PPA was recovered from products which
were expected to bear the loading and a lower or a negative PPA was recovered
from the price of products which were to be subsidised. Kerosene and LPG
supplied to domestic consumers and naphtha, and fuel oil supplied to fertilizer
units were subsidised through a lower / negative PPA. The bulk of this subsidy
was borne by petrol (motor spirit), aviation turbine fuel (domestic airlines), LPG
(other than domestic), and naphtha, and fuel oil supplied to industries other than
fertilizer.
Standard LPG filling norms
For all refineries, the filling quantity was fixed. Any quantity filled in
excess of standard would entitle the refineries for an additional amount of Rs 50/
mt that would be a straight addition to contribution margins. No penalty was
applied for filling below the standard.
For all marketing bottling plants, the cost reimbursement was uniform
based on industry average. Therefore companies whose operating cost was lower
than the industry average were bound to gain and companies whose operating cost
was higher were bound to lose, to the extent of differential cost. It may also be
noted that as regards marketing plants, no standards were set and the actual
contribution was a multiple of actual quantity filled and the per unit retention
margins. Hence, there was a tremendous incentive for LPG filling at these plants.
The additional contribution, earned by filling marketing bottling plants, was
significantly higher than Rs 50/mt for additional filling in refinery bottling plants.
Summary of the APM
The Administered Price Mechanism (APM) was thus based on a retention

or cost-plus formula, whereby oil companies were allowed to recover their
operating costs and earn a post-tax return on net assets. The Central government’s
Oil Co-ordination Committee (OCC) controlled the prices of each product; it also
computed an ex-refinery price applicable across the country. For each distributor,
a margin was calculated, based on actual operating costs and a return on assets,

this margin being added to the ex-refinery price to reach the gross selling price.
The price would then be adjusted according to the subsidy set by the OCC, to
arrive at the final selling price (including an excise duty set by the Government);
the OCC adjusted prices and subsidies about once a year.
The Oil Industry Pool Account mechanism was used to subsidise and
cross-subsidies certain oil products; financial inflows from collection of
surcharges on the sale of some products were meant to offset the outflows for
compensating for the shortfall in revenue on other products. The Pool Account
was meant to be in balance over the long run without budgetary support from the
Central Government. However, during the 1990s the Pool Account fell into
deficit when adjustments failed to keep pace with changes in import prices; this
led to shortfalls in disbursements to the oil companies.
Dismantling of the APM
While the APM ensured a degree of price stability, it failed to provide

adequate incentive for companies to minimise their costs and use capital
efficiently. In addition, Pool Account deficits undermined the public distributors’
ability to invest in distribution infrastructure.
In 1998 the (Central) Government initiated a phased dismantling of the

APM, to bring prices in India in line with international prices (but inclusive of
duties); refinery-gate prices, including that if LPG, were set at the level if import
prices. LPG subsidy was reduced from 68% to 33% at the beginning of 2001-02.
In March 2002, the APM was dismantled, with all major products decontrolled
and the Pool Account wound up. However, subsidies for kerosene and LPG will
continue (while being reduced in a phased manner) at least till March 2005. The
Government is financing this subsidy directly.
Current (2003-04) status of LPG subsidies
The Finance Ministry has provided (Public Sector Unit) oil firms a subsidy
on LPG cylinders for domestic use, at Rs 67.75 per cylinder during 2002-03, and
will provide Rs 45.17 per cylinder during 2003-04; the subsidy per cylinder is
likely to drop to Rs 22.58 during 2004-05. This subsidy was not earlier available
to private LPG marketing companies, but from the year 2003-04 is likely to be
given to them too. However, there remains a difference between the cost and the
retail price per cylinder, even after taking into account the subsidy. To counter
this, the central Government has put together an intricate system of cross-
subsidisation by which retailing firms and LPG producers share the under-
recoveries in the case of Public Sector Units; thus far, this mechanism has not
been made available to private companies.
Gas pricing
Till the 1970s’, gas prices were based on the recommendations given by
expert committees. In the early 1970s’, gas prices were set on a negotiated basis,
resulting in different gas prices for different consumer segments. In the mid 70s’,
the price of natural gas was determined by the producers themselves, based on the
thermal equivalence of substitute fuels and the opportunity cost to the consumer.
In 1986, a decision was taken by the Government of India to fix uniform

prices for natural gas on a year-to- year basis. This policy was followed till 1991.
From January 1, 1992, the prices of natural gas were fixed for a period of four
years. This pricing was based on the recommendation of the Kelkar Committee,
set up by the Government to examine natural gas prices.
Post December 1995, the consumer price for non-North-East areas was
fixed by the Government at Rs 1,850/tcm (exclusive of royalty @ 10 per cent and
class tax varying from 0 to 19 per cent), for a calorific value of 9,000 kilocalories.
The corresponding figure for North East India was Rs 1,000/tcm with a provision
for further discounts. In January 1996, the Government appointed a Committee
under the chairmanship of Mr T.L.Sankar to review the pricing of natural gas.
Based on the recommendations of this Committee, Government fixed a price band
of 2,150 Rs/tcm as the lower limit and 2,850 Rs/tcm as the ceiling for the
consumer price. Producer Price actually payable to the producer (ONGC) was
pre-determined at an amount lower than the consumer price so that the difference
between the Consumer Price and Producer Price could be credited to a Gas Pool
Account. This Account was established in order to encourage the development of
the gas industry in India by partly compensating exploration and development
companies for the low margins received in the development and sale of gas, at
prices fixed by the government.
In addition to the price as fixed above, royalty, taxes, duties and other
statutory levies on the production and sale of natural gas are payable by the
consumers. The royalty on gas, as fixed under the Oilfields Development Act, is
10 per cent of the wellhead price. For privately operated fields, the royalty is
fixed on the negotiated wellhead prices. There is no cess on natural gas (unlike
crude oil) although a cess could be levied under the law. There is no excise duty
on natural gas or on crude oil, as these are minerals, although excise duty is
charged on petroleum products. A sales tax is leviable at state rates if the sale is
within the state or at the central rate of 4 percent for inter-state sales. The sales
tax rates vary from state to state ranging from zero to 22 per cent. It may be noted
that Gas Authority of India Limited (GAIL) does not get a margin on merchant
sales; it is allowed a return only on its investment in the pipeline. In order to
encourage investment in the exploration of oil and gas, the Government has
allowed contractors freedom to market oil and gas produced under New
Exploraton Licencing Policy (NELP). Accordingly, oil and gas produced under
NELP blocks are not covered under the Administered Gas Pricing Mechanism and
the producers are free to market gas at the market-determined prices.
On July 23, 2003, a Group of Ministers, represented by producer and user

Ministries, met and recommended that:
• natural gas prices be increased on an ad-hoc basis with immediate effect, as the
prices have remained static since October 1999;

• a Tariff Committee be appointed to study the cost structure of ONGC and OIL,
and suggest a reasonable price, within six months, for the period till complete
deregulation of the gas prices is brought about;
• the price of gas be raised from 2,850 Rs/tcm to Rs 3,200/tcm, a rise of 12.28 per
cent;
• the gas produced by the joint venture of Tapti and Panna-Mukta of about 8
MSCMD be sold by GAIL/producer at market-determined price; however, 1
MSCMD of gas from Ravva joint venture field in Krishna-Godavari basin could
be taken by GAIL and adjustment for the higher cost made as per the existing
arrangement;
• the Gas Pool Account be limited to Rs 1 billion per annum as per the actual
requirement of compensation for concessional gas prices in the northeast region
and other purposes;
• gas produced by ONGC and OIL from new gas fields be sold at a price
determined in terms of NELP contracts, to provide a level-playing field between
these oil sector PSUs and other players;
• the price of gas for northeastern region be pegged at 60 per cent of the revised
price for general consumers;46
The gas transportation charges along the HBJ pipeline system were fixed at
1,150 Rs/tcm with effect from October 1, 1997 based on the recommendations of
the Sankar Committee.
GAIL also uses natural gas internally, as a fuel for operating the compressors
required to ensure desired pressure of gas in the HBJ pipeline system. There are a
total of six compressors stations along the HBJ system of which two compressors
were commissioned after October 1, 1997. Further, two compressor stations at
Jhabua and Hazira were augmented after October 1, 1997. As a result, the total
quantum of natural gas used internally as fuel by GAIL has increased.
Simultaneously, the gas price has also increased from the level considered during
HBJ tariff fixation by Sankar Committee. Therefore, the cost of transportation has
been raised to 1,160 Rs/tcm. At the meeting of Committee of Secretaries (CoS) in
May 2003, ministry of Petroleum and Natural Gas (MoP&NG) had suggested that
the gas prices be increased from Rs 2,850/tcm to Rs 3850/tcm whereas the
Ministry of Power and Department of Fertilisers indicated Rs 3250/tcm as their
acceptable price for gas. On July 23, 2003, Group of Minister (GoM), represented
by producer and user Ministries met and recommended an increase in natural gas
prices of Rs 350/tcm. They have also suggested that the Gas Pool Account to be
limited to Rs 1 billion per annum as per the actual requirement of compensation
for concessional gas price in northeast region and other purposes. However,
MoP&NG is yet to take a decision on these recommendations.
46
At present, the consumer price for general consumers is 2,850 Rs/tcm whereas for north-eastern
consumers the corresponding price is 1,700 Rs/tcm which works out to be 60 percent of the
general consumer prices. The difference between the producer price and the consumer price in the
northeastern region may be reimbursed to OIL from the Gas Pool Account as is being done under
the existing arrangement.
Annexe 5:
Lessons from India’s improved stove programme
Lessons could also be learnt from India’s national improved stove (chulha)
programme.
In 1984-85, the Ministry of Non-Conventional Energy Sources (MNES) of the

Union Government of India had initiated the national programme on improved
chulhas47 (NPIC) for the promotion of research and dissemination of improved
chulhas among biomass-using households. Till April 2002, when this programme
was disbanded, about 34 million improved chulhas had been installed in homes
Mahapatra, 2003) in 23 States and 5 Union Territories of the country. The
programme had two components: R&D and target fulfilment. While the R&D
component was handled at the state level by independent government or academic
bodies, the targets were to be met by agencies of the state government primarily as
a welfare activity (Hanbar and Karve, 2002). The lessons that could be learnt
from the programme and the assessments particularly by the National Council for
Applied Economic Research (NCAER):
• Participation of the users is essential. Lack of perception of
improvements resulted in few wanting the improved stoves. A study
of 9,867 chulha owners (who acquired the stoves between 1996 and
2001) and 1,979 non-owners in 24 states, revealed that only 38.8%
demanded them, while the rest had to be persuaded by implementing
agencies.
• Target installation numbers can be distorted by corruption. (In some
places stoves were shown to be working when they were never
installed).
• Training cannot be ignored. The NCAER study reported an average of
only 27.2% of households receiving training, with some regions having
no training at all. There were better results where states took an
interest.
• Maintenance after installation is also essential. Around 89% of users
did not know whom to contact when repairs were needed and only
17% reported the availability of adequate hardware in nearby markets.
• Standards have to be maintained. The promised fuel efficiency was
not experienced with 35% complaining that cooking time was longer.
The NCAER report found that although women had been instrumental
in taking the decision to install the new stove, their disillusionment
adversely affected the continued use of the stoves; only 16.6% showed
willingness to reinstall the chulha, if broken. However, when offered a
new version with longer life, no smoke and less fuel, 87% were very
keen.
• There are categories of users who have more than one stove; the chulha
is used for cooking regular meals and a “superior” fuel – LPG or
kerosene - for quick additions.
47
Chulha = stove
Bibliography:
• AFPRO-CHF, 1997. Biogas: The Indian NGO Experience, Action for Food
Production, Canadian Hunger Foundation and Tata Energy Research Institute,
New Delhi.
• Barnes, D.F., 1998. “Best Practices and Grid Rural Electrification: Preliminary
Evidence from Selected Case Studies”, Village Power ’98: Scaling Up Electricity
Access for Sustainable Rural Development, October 6-8, Washington D.C.
(NREL/World Bank)
• Barnes, D.F. and Halpern, J., 2000. “The Role of Energy Subsidies’, in Energy
Services for the World’s Poor, ESMAP Energy and Development Report.
• Barnett, A., 1999. “Increasing Access to Sustainable Energy Sources: A
Summary of Recommendations Made to the Sustainable Energy Programme of
the Shell Foundation”, 20th September.
• BIS, 1987. Indian Standard Methods of Tests for Coal and Coke (Part I:
Proximate Analysis) [IS: 1350-1987], Bureau of Industrial Standards, New Delhi.
• Business Line (Bureau report, New Delhi), 2004a. “Government to provide
subsidy on LPG sold by private companies”, Business Line, Bangalore, 2nd
January 2004, p.5.
• Business Line (Bureau report, New Delhi), 2004b. “Minister to lay foundation
stone for LPG cavern storage project”, Business Line, Bangalore, 19th February
2004, p.5.
• Business Standard, 2003. “More Oil Cross-Subsidy Now”, 4th March, available at
www.business-standard.com/index.asp.
• CBA Energy Institute, 1996. LPG Markets in Mexico: Analysis, Issues and
Cross-National Comparisons, a CBA Energy Institute White Paper, prepared by
M.M.Foss, F.García, G.T.Westbrook and J.Campos, CBA Energy Institute,
University of Houston, Houston, Texas, available at
www.uh.edu/energyinstitute/pdf/lpgsum.pdf.
• Cecelski, E., 1990. Linking Electrification with Rural Development in Asia,
UNDP and ESCAP.
• Cecelski, E., 2000. “Enabling Equitable Access to Rural Electrification: Current
Thinking and Major Activities in Energy, Poverty and Gender” – Briefing Paper
for a Brainstorming Meeting on Asia Alternative Energy Policy and Project
Development Support: Emphasis on Poverty Alleviation and Women, Asia
Technical Unit for Alternative Energy (ASTAE), The World Bank, 26-27 January,
Washington, D.C., available at
www.sms.utwente.nl/energia/download/resources/current_thinking_energypovert
ygenderz.pdf
• Census of India, 2001. Houses, household amenities and assets: India, States and
Union Territories – 2001 Census, Office of the Registrar General, Census of
India, Government of India, New Delhi, available at
www.censusindia.net/2001housing/housing_tables_main.html
• CMIE, 2003. Energy, Centre for Monitoring Indian Economy, May 2003,
Mumbai.

• Cheresources Inc., 2002. Information on Crude distillation, Chemical Engineers’
Resource Page, Cheresources Inc., available at
www.cheresources.com/refining.html
• Dutt, G. and Ravindranath, N.H., 1993. “Bioenergy: Direct Applications in
Cooking” in Renewable Energy: Sources for fuels and electricity, Johansson,
T.B., et al., (Ed.), Island Press, pp.653-698.
• Energy and Development Research Centre (EDRC), 1998. Rural Electrification
in South Africa, based on the Project on the role of electricity in the integrated
provision of energy to rural areas of South Africa, EDRC, University of Cape
Town.
• ESMAP, 2001. “Household Energy and Poverty”, in Indoor Air Pollution –
Energy and Health for the Poor, Issue No.4, UNDP/World Bank Energy Sector
Management Assistance Program, May.
• Fischer, S.L., 2001. “Biomass-derived liquid cooking fuels for household use in
rural China: potential for reducing health costs and mitigating greenhouse gas
emissions”, Energy for Sustainable Development, V (1), pp. 23-30.
• Foley, G., 1990. Electricity for Rural People, Panos, London.
• G8-RETF, 2001. Strategic Pathways Analysis – Annex 1, G8 Renewable Energy
Task Force, July, available at www.renewabletaskforce.org/pdf/G8_annexes.pdf
• Gupta, G. and Köhlin, G., 2001. Options and Choices of Domestic Fuel in Urban
Households: The case study of Kolkata, India, available at
www.handels.gu.se/econ/EnvironmentalEconomicsUnit/fuelpapers.pdf
• Gupta, Uttam, 2003. “Few gains from notional decontrol”, Editorial in Business
Line, 5th July 2003, Bangalore, p.8.
• Hanbar, R.D. and Karve, P., 2002. “National Programme on Improved Chulha
(NPIC) of the Government of India: an overview”, Energy for Sustainable
Development, VI (2), pp. 49-55.
• Henderick, P. and Williams, R.H., 2000. “Trigeneration in a northern Chinese
village using crop residues”, Energy for Sustainable Development, IV (3), pp. 26-
42.
• IBGE, 2001. Demographic Census 2000 – Results of the universe, Instituto
Brazileiro de Geografia e Estatïstica, available at www.ibge.gov.br
• IEA, 2004. Oil Market Report, International Energy Agency, Paris, 12th May,
2004, available at www.omrpublic.iea.org/currentissues/full.pdf
• IEI, 2003. Report on IEI’s Rural Electricity and Water Supply Utility (REWSU)
project with special reference to the utility at Mavinakere, prepared by
K.V.Narasimha Murthy, Chandru, B.T., and Antonette D’Sa, International Energy
Initiative, October, available at (http://www.iei-asia.org/IEIBLR-
REWSUReport.pdf).
• IIFM, 1999. Woodfuel Trade in India, Indian Institute of Forest Management,
Bhopal.
• Indiainfoline, 2002. Information on the oil and gas industry, available at
www.indiainfoline.com
• IPAN online, 2000. News Bulletin - Oil & Natural gas sector: September 2000,
Indian Public Affairs Network, available at
www.ipan.com/REVIEWS/ARCHIVES/dec2000oil.htm

• IPCC, 1997. Guidelines for National Greenhouse Gas Inventories: Reference
Manual, Volume 3, United Kingdom: Blacknell.
• Jamwal, N., 2003. “Full Circle: India’s largest biogas plant produces more than
just cooking gas”, Down to Earth, March 15, 2003, p.42.
• Jechoutek, K.G.,1992. “Selection and Economic Viability of Rural Electrification
Projects”, in Rural Electrification Guidebook for Asia and Pacific, G.Saunier
(Ed.), Asian Institute of Technology, Bangkok.
• Jungbluth, N., 1995. Restricted Life-cycle Assessment for the Use of Liquefied
Petroleum Gas and Kerosene as Cooking Fuels in India, thesis submitted to the
University of Berlin, December, available at www.esu-services/download/DA-
India.pdf.
• Larson E.D. and Jin, H., 1999. “Biomass conversion to Fischer-Tropsch liquids:
preliminary energy balances”, in Overend, R.P., and Chornet, E., (eds.), Biomass,
A Growth Opportunity in Green Energy and Value-added Products: Proceedings
of the 4th Biomass Conference of the Americas, Vol., 1, Elsevier Science, Oxford,
UK, pp. 843-853.
• Larson, E.D. and Kartha, S., 2000. “Expanding roles for modernized biomass
energy”, Energy for Sustainable Development, IV (3), pp. 15-25.
• Mahapatra, R., 2002. “Phulmai’s walk”, Down to Earth, December 15, 2002, pp.
25-34.
• Mahapatra, R., 2003. “Up in smoke”, Down to Earth, January 31, 2003, pp. 25-28.
• Malhotra, P., I.H.Rehman, P.Bhandari, R.Khanna, and R.Upreti, 2001. “Rural
energy data sources and estimations in India”, Tata Energy Research Institute,
New Delhi, available at
www.pesd.stanford.edu/docs/2002mtg_TERI/papers/REHMAN_data_ppr.pdf
• MNES, 2003. “Cumulative Achievements of Renewable Energy Programmes in
India – 31st March 2003”, Ministry of Non-Conventional Energy Sources,
Government of India, available at www.mnes.nic.in/ach1.htm
• MoF (Ministry of Finance, Government of India), 2000. Economic Survey: 1999-
2000, Ministry of Finance – Economics Division, Government of India, New
Delhi.
• MoP&NG (Ministry of Petroleum and Natural Gas, Government of India), 2002.
Natural Gas, available at http://petroleum.nic.in/ngbody.htm
• MoP&NG, 2003a. Annual report for the year 2002-03, available at
www.petroleum.nic.in/annrep2003.pdf
• MoP&NG, 2003b. “Availability, Demand and Security of Supply of Autofuels”,
Chapter 7 of Auto Fuel Policy Report, Petroleum Planning & Analysis Cell
(PPAC), Ministry of Petroleum and Natural Gas, Government of India, available
at www.petroleum.nic.in/ch_7.pdf
• MoP&NG, 2003c. Petroleum Statistics, Petroleum Planning & Analysis Cell
(PPAC), Ministry of Petroleum and Natural Gas, Government of India, available
at www.petroleum.nic.in/Petstat.pdf
• Mukunda, H.S., Dasappa, S., and Shrinivasa, U., 1993. “Open-top wood
gasifiers”, in Renewable Energy: Sources for fuels and electricity, Johansson,
T.B., et al., (Ed.), Island Press, pp.699-728.
• Munasinghe, M., 1987. Rural electrification for development: Policy analysis and
applications, Westview Press, Boulder, 1987.

• Natarajan, I., 1990. Fuelwood Balance in the Nineties, National Council of
Applied Economic Research (NCAER), New Delhi.
• Natarajan, I., Demand Forecast for Biofuels in Rural Households, in Section III:
Biomass Energy Analysis and Projections, available at
www.iea.org/pubs/proc/files/bioend/S3P2-NAT.PDF.
• NSSO, 2001. National Sample Survey, 55th Round: July 1999 – June 2000.
National Sample Survey Organisation, Ministry of Statistics and Programme
Implementation, Government of India, New Delhi.
• NSSO, 1992. Sarvekshana, Volume XVI, No.1, National Sample Survey
Organisation, Department of Statistics, Government of India, July-September.
• NIRD, 2002. Presentation of the Salient Findings of the Evaluation of the
Deepam Scheme in Andhra Pradesh, National Institute of Rural Development-
World Bank Research Study Team available at
www.worldbank.org/wbi/cleanair/caiasia/topics/iap/20workshop/iap_documents/
Rajakutty.pdf.
• OTA, 1992. Fueling Development: Energy Technologies for Developing
Countries, Office of Technology Assessment, OTA-E516, Washington D.C.
• Pachauri, S., and Spreng, D., 2003. “Energy use and energy access in relation to
poverty”, CEPE Working Paper Number 25, Centre for Energy Policy and
Economics, Swiss Federal Institutes of Technology, Zürich, June available at
www.cepe.ethz.ch/download/cepe_wp/CEPE_WP25.pdf.
• Parikh, J. and Denton, F., 2002. “Networking around the world: Report on
Gender and Climate Change Event at COP8”, Energia News, Volume 5, Issue 4,
December.
• Parikh, J., Biswas, H., and Karmakar, S., 2003. “Cooking with biofuels: Risk
factors affecting health impact on rural women”, Special article in Economic and
Political Weekly, Vol. XXXVIII, No.26, June 28-July 4, pp. 2681-2692.
• Petrowatch, 2003. Volume 7 Issue 15, 8th October.
• Plan.Com, GoI, 1991. Sectoral Energy Demand in India, Planning Commission,
Government of India, in cooperation with the Economic and Social Commission
for Asia and the Pacific, the UNDP and the Government of France, Regional
Energy Development Programme (RAS/86/136), August.
• Purvin and Gertz (A.Chandra, R.Gist, K.Otto & C.Whitley), 2000. “Continued
LPG Demand Growth Changes Historical Trade Patterns” in Oil & Gas Journal,
26th June, available at www.purvingertz.com/ogj062600.html
• Reddy, B.Sudhakara, 1996. “Consumer discount rates and energy carrier choices
in urban households”, International Journal of Energy Research, Vol. 20, pp.187-
195.
• Reddy, C. Rammanohar, D’Sa, Antonette and Reddy, Amulya K.N., 1992. “The
Debt-Energy Nexus: A Case Study of India” in Economic and Political Weekly,
Vol. XXVII, No.27, 4th July, pp. 1401-1415.
• RPL, 2000. “Dikhana hai” available at www.indianmarketwatch.com/uploaded-
files/research/rpl.htm
• Sarma, E.A.S., Maggo, J.N., and Sachdeva, A.S., 1997. India’s Energy Scenario
in 2020, (based on the authors’ study “Environmentally-Constrained Alternative
Scenarios for Energy” for the Asian and Pacific Development Centre, Kuala
Lumpur) available at

www.worldenergy.org/wec.geis/publications/default/tech_papers/17th_congress/1
_1_27.asp
• Shyam, M., 2002. “Agro-residue-based renewable energy technologies for rural
development”, Energy for Sustainable Development, VI (2), pp. 37-42.
• Smith, K.R., 2000. “National Burden of Disease in India from Indoor Air
Pollution”, Inaugural article, National Academy of Sciences, 97(24): pp.13286-93.
• Smith, K.R., Uma, R., Kishore, V.V.N., Lata, K., Joshi, V., Zhang, J., Khalil,
M.A.K., 2000a. “Greenhouse implications of household stoves: an analysis for
India”, Annual Review of Energy and Environment, 25, pp.741-763.
• Smith, K.R., Uma, R., Kishore, V.V.N., Lata, K., Joshi, V., Zhang, J., Rasmussen,
R.A., Khalil, M.A.K., 2000b. Greenhouse gases from small-scale combustion
devices in developing countries: Phase IIA – Household stoves in India, prepared
for USEPA, Office of Research and Development, Washington D.C., (EPA-
600/R-00-052) available at www.epa.gov/appcdwww/apb/R-00-052.pdf.
• Sustainable Alliance, 2002. Peri-Urban and Rural Energy Services (PRES) in
Ethiopia, Kenya and Uganda, Phase 1 Project Report, The Shell Foundation
Sustainable Energy Programme (SEP), January, available at
www.pres.energyprojects.net/links/PRES_project_to_stakeholders.pdf.
• UNDP, et al., 2000. World Energy Assessment: Energy and the Challenge of
Sustainability, (Ed.) J. Goldemberg, United Nations Development Programme
(UNDP), United Nations Department of Economic and Social Affairs, World
Energy Council, September.
• UNDP, 2002. The LPG Challenge- Discussion Paper, UNDP Sustainable Energy
Programme, available on www.undp.org/seed/eap
• UNDP and ESMAP, 2003. India: Access of the Poor to Clean Household Fuels,
United Nations Development Programme and the World Bank Energy Sector
Management Assistance Programme, July, available at
www.wbnl0018.worldbank.org/esmap/site.nsf/files/263-03+India.pdf
• WB and WLPGA (World Bank and World LP Gas Association), 2001. West
Africa LPG Market Development Study, 2001, available at
www.worldgas.com/mainpages/publications/publications.php.
• WB and WLPGA (World Bank and World LP Gas Association), 2002. The Role
of LP Gas in Meeting the Goals of Sustainable Development, The World Bank and
World LP Gas Association, Paris, January, available at
• WB (World Bank), 2000. Energy Services for the World’s Poor: Energy and
Development Report 2000 available at
www.worldbank.org/html/fpd/esmap/energy-report2000
• WEC (World Energy Council), 2001. Pricing Energy in Developing Countries,
World Energy Council, London, June, available at
www.worldenergy.org/wec.geis/global/downloads/pedc/PEDC.pdf
• WLPGA, 2002. Guidelines for LPG Usage in Rural Energisation, March refer to
• WLPGA & UNDP (World LP Gas Association and United Nations Development
Programme), 2002. The LP Gas Rural Energy Challenge – Draft Concept Paper,
available at http://www.undp.org/wec-geis/wssd/lpgchallenge.pdf
• WRI, 2002. Brazil: The Structure of Poverty, World Resources Institute,
available at www.povertyprofit.wri.org/pdfs/hammond_brazil_profile_xls.pdf


LPG Project 3 PDF

Uploaded by

Copyright:

Available Formats

LPG Project 3 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

LPG Project 3 PDF

Uploaded by

Copyright:

Available Formats

Report on the use of LPG

as a domestic cooking fuel option

International Energy Initiative

List of Tables iii

2. Demand for LPG 9

4. Challenges to extending LPG for domestic use 41

5. Experiences of LPG programmes 46

6. Issues for Indian domestic cooking fuels 49

Antonette D’Sa & K.V.Narasimha Murthy ii

Table 1: Indoor concentration of health-damaging pollutants from a typical

Antonette D’Sa & K.V.Narasimha Murthy iii

Figure 1: Comparison of annualised costs of stoves 4

Antonette D’Sa & K.V.Narasimha Murthy iv

Annexe 1: Technical details 56

Antonette D’Sa & K.V.Narasimha Murthy v

If a business-as-usual scenario were assumed, that is if the current rates of

Challenges to effective provision of domestic LPG (Section 4)

Lessons from other LPG experiences (Section 5)

Experiences in several other developing countries have been studied; the

Antonette D’Sa & K.V.Narasimha Murthy viii

In the context of the provision of appropriate cooking fuels, Indian decision

Antonette D’Sa & K.V.Narasimha Murthy ix

• Public awareness: Awareness of the adverse impacts on health of indoor pollution

• Supply security: Dependable supply of LPG requires -

Antonette D’Sa & K.V.Narasimha Murthy x

• Regulation - the government’s role: The government/regulator would have to set

There are other important alternatives to traditional cooking fuels, in particular,

Antonette D’Sa & K.V.Narasimha Murthy xi

Dependence on traditional forms of biomass adversely affects human

While Agenda 217 specifically recognized the challenge of providing access

However, the likelihood of enhancing supplies of LPG and a distribution

1.2 Why LPG?

Table 1: Indoor concentration of health-damaging pollutants from a typical

The numbers in parentheses indicate typical standards set to protect health.

As Figure 1 indicates, the life-cycle costs increase from ordinary fuel-

Comparison of annualised costs of

This is based on the authors’ computation, using market prices/subsidies prevailing in

There are other alternatives to traditional cooking fuels. Renewable sources

Since LPG is a petroleum-based fuel, it can be argued that increased use of

Therefore, the use of LPG is being considered as a short/medium term

1.3 Objectives of this study:

1. the domestic use of cooking fuels in India, particularly that of LPG

Antonette D’Sa & K.V.Narasimha Murthy 5

Estimates of future requirement -

Given the base-year number of LPG-dependent households and the

Antonette D’Sa & K.V.Narasimha Murthy 7

Figure 2: Range of demand scenarios

Supply to the consumer has to be analysed further, considering other

Antonette D’Sa & K.V.Narasimha Murthy 8

2. Demand for LPG

Source: Purvin and Gertz, 2000

2.1 Domestic use of cooking fuels in India

Several estimates of household use of cooking fuels in India have been

Antonette D’Sa & K.V.Narasimha Murthy 10

Figure 4b: Indian rural household dependence on various cooking fuels in

Antonette D’Sa & K.V.Narasimha Murthy 11

2.2 Estimated domestic requirement of LPG

2.2.1 Extent of dependence on LPG

State/Union Territory Total number Households LPG-using