An investigation of multidimensional energy poverty in Pakistan.
Awan, Rehmat Ullah ; Sher, Falak ; Abbas, Akhtar 等
This study uses Alkire and Foster's (2007) methodology to
measure Multidimensional Energy Poverty in Pakistan using Pakistan
Social Living standard Measurement (PSLM) data for 2007-08. This measure
also satisfies the property of decomposability. Multidimensional Energy
Poverty Index (MEPI) has been calculated. Value of MEP headcount for
rural Pakistan is 71 percent as compared to 29 percent in urban areas of
Pakistan. Multidimensional Energy Poverty headcount for rural Pakistan
is 71.4 percent and 28.6 percent of the households residing in rural
areas of Pakistan are energy poor. Households of Pakistan are most
deprived in cooking fuel dimension (55 percent) while deprivation is the
least in dimension of home appliances ownership (15 percent). In urban
areas of Pakistan, households are more deprived in dimension of cooking
fuels (23 percent) followed by indoor pollution (19 percent). Almost one
third households of rural Pakistan are deprived in dimension of indoor
pollution (69 percent). Contribution of indoor pollution (32 percent) to
multidimensional energy poverty headcount in Pakistan is the highest
followed by the cooking fuels dimension (31 percent) and collectively
these two dimensions contribute up to 63 percent in overall
Multidimensional Energy Poverty headcount for Pakistan. Study further
concludes that households deprivation in energy decreases with the
increase in value of cut offs. Overall indoor pollution, cooking fuel
and entertainment appliances are the three major contributors, to
overall MEP headcount not only as a whole but region wise also. Based on
results, the study established that there is significant and higher
incidence and severity of energy poverty in rural areas as compared to
urban areas in Pakistan.
JEL Classification: 100,13,132, Q00, Q4, Q40, Q49
Keywords: Multidimensional Energy Poverty, Energy Poverty, Poverty
Measurement, Multidimensional Energy Poverty Index (MEPI),
Decomposability, Deprivation, Intensity of Poverty
1. INTRODUCTION
Poverty is an alarming problem all over the world. It is one of the
severe challenges today faced by not only the developing nations but by
the developed nations also. However, the problem is worst in developing
countries [United Nations and IEA (2010)]. All these countries face
poverty in different forms such as food poverty, energy poverty,
shortage of natural resources, shortage of agricultural products, lack
of shelter and clothing among others. It is persuasive to correlate
poverty with lack of energy consumption also. Such a correlation
identifies that poor use energy very inadequately [Pachauri, et al.
(2004)]. Energy helps societies to move from one development stage to
another. Worldwide energy demand is increasing while supply is
decreasing due to increase in the world population, emerging economies
and economic development. In current day to day life energy has become
an essential requirement. For all of us energy is required for lighting,
transportation, cooking, health services, and to fulfill many of our
basic needs. Electricity access at household level enhances
telecommunication, entertainment, and knowledge via radio, television,
and computer etc.
World Economic Forum (2010) defines energy poverty as "the
lack of access to sustainable modern energy services and products".
The energy
poverty is defined as a situation where the absence of sufficient
choice of accessing adequate, reliable, affordable, safe and
environmentally suitable energy services is found. In simple words,
energy poverty is the lack of access to suitable traditional (fire wood,
chips, dung cakes etc.) and modern energy services and products
(kerosene, liquefied petroleum, gas etc.). For development of any
country, energy is the first step. A person is considered to be energy
poor if he or she does not have access to at least (a) the equivalent of
35 Kg per capita per year LPG for cooking from liquid and/or gas fuels
or from improved supply of solid fuel sources and improved (efficient
and clean) cook stoves and (b) 120KWh electricity per capita per year
for lighting, access to most basic services (drinking water,
communication, improved health services, education improved
services and others) plus some added value to local production.
To enhance livelihood opportunities for all, electricity plays a
major role. To change the poor's life in a better way, clean and
efficient energy resources are required. Firewood collection for cooking
consumes a lot of women's time. Clean energy sources for cooking
like electricity, and gas etc. mean improvement in living standards and
time saving also. The income poor could also be energy poor, however not
all of the energy poor are income poor. Energy scarcity and poverty go
hand in hand and show a strong relationship. Welfare of masses is
affected by the level of energy consumption. There is a negative
correlation between access to modern energy services and energy poverty.
So in order to alleviate energy poverty, improvement in the access to
modern energy services is very essential. Availability of cheaper energy
is essential. According to United Nations, lack of electricity and heavy
reliance on traditional biomass are hallmarks of poverty in developing
countries. Lack of electricity enhances poverty and contributes to its
persistence, as it prevents most industrial activities and the job
creation. [United Nations and IEA (2010)].
To meet their survival needs in absence of efficient energy using
technologies and adequate energy resources, majority of poor depend on
biomass energy, animal power and their own labour. To improve the level
of satisfaction of basic human needs and living standards of the people
and to eradicate poverty energy resources must be improved. For the
better health care facilities and education clean energy is required.
Achievement of efficient energy resources can lead to the attainment of
evenhanded, economically strong and sustainable development. Present
study aims to investigate the level of energy poverty in Pakistan and to
find the extent of energy poverty in rural and urban areas of Pakistan
along with the impact of different variables on energy poverty in
Pakistan.
Rest of the study is organised as follows. Section 2 gives review
of literature. Section 3 is about methodology and data. Results and
discussions are are presented in Section 4. Section 5 concludes the
study giving some policy recommendations based on findings.
2. REVIEW OF LITERATURE
Pasternak (2000) found that there is strong relationship between
measures of human well-being and consumption of energy and electricity.
A roughly constant ratio of primary energy consumption to electric
energy consumption was observed for countries with high levels of
electricity use and then this ratio was used to estimate global primary
energy consumption in the Human Development Scenario. They established
positive correlation between Human Development Index (HDI) and annual
per capita electricity consumption for 60 populous countries comprising
90 percent of the world's population. Results further showed that
HDI reached a maximum value when electricity consumption was about 4,000
KWH per person per year.
Bielecki (2002) by using a measurement of the existing state of oil
security pointed out that the threats of supply disruption had not
deminished. Outlook of the oil market for coming two decades advocate
that there is still need to take more steps for the oil security. It was
also found that with rising importance of universal demand and trade of
gas, the gas security is also becoming gradually more significant. They
claimed that different severe security alarms do exist and will probably
strengthen in the future. This indicates that there is no area for
gratification on energy security. The present oil crisis measures
require extension to cover up energy sources for developing nations and
for others.
Clancy, et al. (2003) found that Energy security has turned into a
central community issue along with concerns with sky-scraping energy
prices and the incidence of regional shortage of supply. 2.8 million
Households in England are classified as being in fuel poverty in 2007
(13 percent of all households). It is found that the fuel poverty in the
UK is not going to be of the same order or intensity as that of sub
Saharan Africa. NGOs and practitioners also point at complex processes
of energy exclusion and self-exclusion at the community, household and
family level, leading to distinct micro cultures of energy use.
Pachauri, et al. (2004) measured Energy Poverty for Indian
Households using a two-dimensional measure of energy poverty and energy
distribution that combine the elements of access to different energy
types and quantity of energy consumed. They found that there is
significant reduction in the level of energy poverty due to rapid
development in India.
Stephen, et al. (2004) studied present and future renewable energy
potential in Kenya to meet the electrification needs of the poor. They
limited the study to solar and hydro technologies owing to technical and
socio-economic hurdles. They assessed that present Rural Electrification
Fund (REF) in Kenya realises the solar and hydro electrification
potential for poor. The results showed that if there is 10 percent
increase in Rural Electrification Fund (REF), annual revenue from rural
electricity connections increases by 42 percent in Kenya. There exists a
relation between access and use of energy and poverty.
Pachauri, et al. (2004) presented different approaches for
measurement of energy poverty by using Indian household level data. They
found positive relation between well-being and use of clean and
efficient energy resources. They also concluded that use of access and
consumption of clean and efficient energy increases the well-being.
Catherine, et al. (2007) examined UK Government's devotions to
eradicate fuel poverty among vulnerable families by year 2010 and in the
common people by 2016. They explained the relations among this measure
of fuel poverty and the governmental objective definition, using an
exclusive data set and the Family Expenditure Survey. They recognised
the link between two measures. They investigated the characteristics of
households in each group, and how each measure is interrelated with
different household issues.
Tennakoon (2009) analysed energy poverty status of Sri Lanka. Two
approaches namely Quantitative approach and Pricing approach of
measuring energy poverty were used. Results of Pricing approach showed
that Sri Lanka is facing high level of energy poverty (83 percent energy
poverty) while results of Quantitative approach revealed that energy
poverty in terms of cooking is very high due to high inefficiencies of
cooking stoves.
Bamess, et al. (2010) explored the welfare impacts of household and
energy use in rural Bangladesh using cross sectional data. The result
showed that although modern and traditional sources improved energy
consumption of rural Bangladesh households but the impacts of modern
energy sources were high as compared to traditional energy services. 58
percent households in rural Bangladesh are facing energy poverty.
Shahidur, et al. (2010) studied energy poverty of urban and rural
areas of India. The estimates showed that in rural area of India, 57
percent households are energy poor and only 22 percent households are
income poor while in urban areas of India, energy poverty is 28 percent
and income poverty is 20 percent. The persons in energy poverty were
also facing income poverty.
Marcio, et al. (2010) analysed the impact of energy poverty on
inequality for Brazilian Economy using Lorenz Curve, Poverty Gap, Gini
coefficient and Sen Index. It is concluded that rural electrification
leads to improvement in energy equity.
Jain (2010) explored the problems related to energy consumption
faced by Indian rural and urban households. The results showed that
energy poverty in rural areas of India is about 89 percent and 24
percent in urban areas of India. It was also concluded that 56 percent
households in India has access to electricity facilities. Poor persons
spend almost 12 percent of their total income only on the energy. Energy
poverty disturbs all aspects of human welfare like agricultural
productivity, access to water, education, health care and job creation
etc. Energy poor persons have no access to clean water and electricity
and they spend a large portion of their income and time to get energy
fuel. This consumption pattern of the poor persons on energy leads to
the income poverty.
Mirza and Szirmai (2010) discussed the consequences and
characteristics of the use of different energy services using Energy
Poverty Survey (EPS) data from 2008 to 2009. They outlined that the
rural population of Pakistan uses variety of energy services like
firewood, plant waste, kerosene oil and animal waste. Despite these
sources of energy, the population of Pakistan has to face the energy
crises or energy poverty. Estimates show that 96.6 percent of rural
households have to face energy short fall. In Punjab province of
Pakistan, 91.7 percent of rural households of the total rural population
are facing severe energy poverty.
Nussbaumer, et al. (2011) reviewed appropriate literature and
talked about sufficiency and applicability of existing methods for
measurement of energy poverty for several African countries. They
proposed a new composite index, Multidimensional Energy Poverty Index
(MEPI). It captures the incidence and intensity of energy poverty and
focuses on the deprivation of access to modern energy services. Based on
MEPI for Africa, the countries are categorised according to the level of
energy poverty, ranging from sensitive energy poverty (MEPI>0.9; e.g.
Ethiopia) to modest energy poverty (MEPI<0.6; Angola, Egypt, Morocco,
Namibia, Senegal). It was concluded that the MEPI will only form one
tool in monitoring improvement and designing and executing good quality
policy in the area of energy poverty.
3. DATA AND METHODOLOGY
The study uses Pakistan Social and Living Standards Measurement
(PSLM) Survey (2007-08) as latest available data set. This data set
includes sample of 15512 households consisting of 1113 sample
community/enumeration blocks. A two-stage stratified sample design has
been adopted for this survey. Villages and enumeration blocks in urban
and rural areas, respectively have been taken as Primary Sampling Units
(PSUs). Sample PSUs have been selected from strata/sub-strata with
Probability Proportional to Size (PPS) method of sampling technique.
Households within sample PSUs have been taken as Secondary Sampling
Units (SSUs). A specified number of
3.1. Methodology
For the analysis and for the measurement of energy poverty in
Pakistan, study uses Multidimensional Energy Poverty Index (MEPI),
proposed by Nussbaumer, et al. (2011). The MEPI is created by Oxford
Poverty and Human Development Initiative (OPHI) with association of
United Nations Development Programme (UNDP). The technique utilised is
derived from the literature on multidimensional poverty measures, from
the Oxford Poverty and Human Development Initiative (OPHI) [Alkire and
Foster (2007); Alkire and Foster (2009); Alkire and Santos (2010)],
which is improved by Amartya Sen's contribution to the debate of
deprivations and potential. Fundamentally, MEPI takes into account the
set of energy deprivation that may have an effect on an individual. It
captures five dimensions of basic energy services with five indicators.
An individual or a household is considered as energy poor if the
combinations of the deprivations that are faced by an individual surpass
a pre-defined threshold. The Multidimensional Energy Poverty Index is
the result of a headcount ratio (share of people recognised as energy
poor) and the average intensity of deprivation of the energy poor.
Multidimensional Energy Poverty Index (MEPI) merges two features of
energy poverty. On one side is the incidence of poverty defined as the
percentage of people who are energy poor, or the headcount ratio (H) and
the other is the intensity of poverty defined as the average percentage
of dimensions in which energy poor people are deprived (A).
Let [M.sup.n,d] indicate the set of all n*d matrices, and y [member
of] [M.sup.n,d] stand for an achievement matrix of n people in
ddifferent dimensions. For every i = 1, 2, ..., n and j = 1, 2, .... d,
the typical entry [y.sub.ij] of y is individual i's achievement in
dimension j. The row vector [y.sub.i] = ([y.sub.i1], [y.sub.i2],...,
[y.sub.id]) lists individual i's achievements and the column vector
[.sub.j] = ([y.sub.1j], [y.sub.2j],..., [y.sub.nj]) gives the
distribution of achievements in dimension j across individuals. Let
[z.sub.j] > 0 represent the cutoff below which a person is considered
to be deprived in dimension j and z represent the row vector of
dimension specific cutoffs. Following Alkire and Foster's
(2007)'s notations, any vector or matrix v,[absolute value of v]
denotes the sum of all its elements, whereas [mu] (v) is the mean of v.
Alkire and Foster (2007) suggest that it is useful to express the
data in terms of deprivations rather than achievements. For any matrix
y, it is possible to define a matrix of deprivations [g.sup.0] =
[[g.sup.0.sub.ij]], whose typical element [g.sup.0.sub.ij] is defined by
[g.sup.0.sub.ij]] = 1 when [y.sub.ij] < [z.sub.j], and
[g.sup.0.sub.ij] = 0 when [y.sub.ij] [greater than or equal to]
[Z.sub.j] x [g.sup.0] is an nxd matrix whose [ij.sup.th] entry is equal
to 1 when person i is deprived in jth dimension, and 0 when person is
not. [g.sup.0.sub.i] is the [i.sup.th] row vector of [g.sup.0] which
represent person i's deprivation vector. From [g.sup.0] matrix,
define a column vector of deprivation counts, whose [i.sup.th] entry
[c.sub.i] = [absolute value of [g.sup.0.sub.i] represents the number of
deprivations suffered by person i. If the variables in y are only
ordinal significant, [g.sup.0] and c are still well defined. If the
variables in y are cardinal then we have to define a matrix of
normalized gaps [g.sup.1]. Flor any y, le [g.sup.1] = [[g.sup.1.sub.ij]
be the matrix of normalized gaps, where the typical element is defined
by [g.sup.1.sub.ij] = [z.sub.j] - [y.sub.ij])/[z.sub.j] when [y.sub.ij]
< [z.sub.j], and [g.sup.1.sub.ij] = 0 otherwise.
The entries of this matrix are non-negative numbers less than or
equal to 1, with [g.sup.1.sub.ij] being a measure of the extent to which
person i is deprived in dimension j. This matrix can be generalised to
[g.sup.[alpha]] = [[g.sup.[alpha].sub.ij]], with a > 0, whose typical
element [g.sup.[alpha].sub.ij] is normalized poverty gap raised to a
power.
A sensible start is to recognise who is poor and who is not? The
majority of identification techniques recommended in the literature in
general pursue the union/ intersection approach. A person is considered
poor according to union approach, if that person is deprived in only one
dimension. While according to intersection approach an individual i is
considered to be poor if that individual is deprived in all dimensions.
If the equal weights are given to all dimensions the technique to
recognise the multidimensionally poor suggested by Alkire and Foster
deprivations are compared with a cutoff level k. where k= 1,2,..., d.
Now we describe the recognition method [[rho].sub.k] such that
[[rho].sub.k]([y.sub.i],z) = 1 when [c.sub.i] [greater than or equal to]
k, and [[rho].sub.k]([y.sub.i],z) = 0 when [c.sub.i] <k . This shows
that an individual is known as multidimensionally poor if that
individual has deprivation level at least in k dimensions. This is
called dual cutoff method because [[rho].sub.k] depends upon [z.sub.j]
within dimension and across dimensions cutoff k. This identification
principle describes the set of the multidimensionally poor people as
[Z.sub.k] - {i:[[rho].sub.k]([y.sub.i];z) = 1). A censored matrix
[g.sup.0](k) is obtained from [g.sup.0] by replacing the i'th row
with a vector of zeros whenever [[rho].sub.k]([y.sub.i], z) = 0. An
analogous matrix [g.sup.0](k) is obtained for [alpha] > 0, with the
ijth element [g.sup.[infinite].sub.ij] (k) = [g.sup.[infinite].sub.ij]
if c, [greater than or equal to] k and [g.sup.[infinite].sub.ij](k) = 0
if [c.sub.i] < k.
On the basis of this identification method, Alkire and Foster
define the following poverty measures. The first natural measure is the
percentage of individuals that are multidimensionally poor: the
multidimensional Headcount Ratio H = H(y;z) is defined by H = q/n, where
q = q(y,z) is the number of people in set [Z.sub.k]. This is entirely
analogous to the income headcount ratio. This method has the advantage
of being easily comprehensible and estimable and this can be applied
using ordinal data.
4. RESULTS AND DISCUSSION
Table 1 shows different Dimensions, Indicators and the Cut-offs.
From a human development point of view, a poverty indicator must be
significantly and eventually measurable at the individual, household, or
community level. It must allow a classifying of these demographic units
as more or less poor. Present study uses five main dimensions and their
relevant indicators for the measurement of Multidimensional Energy
Poverty Index (MEPI) based upon the availability of nationwide data. All
the five dimensions are weighted equally. Figure 1 shows the results of
Multidimensional Energy Poverty head count for overall Pakistan at dual
cutoff equal to 2 i.e. K=2. The empirical results show that in Pakistan
almost 54.6 percent and 45.4 percent of households are multidimensional
energy poor and energy non poor, respectively.
Figure 2 shows the results of Multidimensional Energy Poverty head
count for urban Pakistan. It is clear from figure that only 29 percent
of the households are multidimensional energy poor in urban areas of
Pakistan, while remaining 71 percent of the households in urban areas
are energy non-poor.
Figure 3 depicts the results of Multidimensional Energy Poverty
headcount for rural areas of Pakistan. The incidence and severity of
energy poverty is significant in rural areas of Pakistan. Results show
that Multidimensional Energy Poverty headcount for rural Pakistan is
71.4 percent and 28.6 percent of the households residing in rural areas
of Pakistan are energy non-poor.
The analysis of breakdown of energy poverty by dimension for
overall Pakistan is shown in Figure 4. Results show that households of
Pakistan are most deprived in cooking fuel dimension (55 percent), while
deprivation is the least in dimension of home appliances ownership (15
percent). Results further show that 52 percent, 33 percent and 19
percent of the households in Pakistan are deprived in terms of indoor
pollution, entertainment appliances and electricity, respectively.
Figure 5 shows the breakdown of energy poverty by dimension for
urban Pakistan. The empirical results show that in urban areas of
Pakistan households are more deprived in dimension of cooking fuels (23
percent) followed by indoor pollution (19 percent). In urban areas of
Pakistan only 3 percent households are deprived in dimension of home
appliances ownership. In dimensions of entertainment appliances and
electricity households are deprived by 18 percent and 7 percent,
respectively.
Figure 6 shows the breakdown of energy poverty by dimension for
rural Pakistan. Almost one third households of rural Pakistan are
deprived in dimension of indoor pollution (69 percent). As shown in
Figure 6, 58 percent households are deprived in cooking fuels dimension
in rural areas of Pakistan. Situation is also critical in entertainment
appliances in the same region. Households' deprivation in terms of
entertainment appliances, electricity and home appliances are 44
percent, 29 percent and 22 percent, respectively.
Figure 7 shows the contribution of urban and rural deprived
households to Multidimensional Energy Poverty headcount for overall
Pakistan. Contribution of rural and urban deprived households to
multidimensional energy poverty in Pakistan is 71 percent and 29
percent, respectively.
Figure 8 shows contribution of selected dimensions in
multidimensional energy poverty headcount. In the paradigm of
multidimensional energy poverty in Pakistan contribution of indoor
pollution (32 percent) is the highest followed by the cooking fuels
dimension (31 percent). Collectively these two dimensions contribute up
to 63 percent in overall Multidimensional Energy Poverty headcount for
Pakistan. While electricity, home appliances and entertainment
appliances contribute to overall Multidimensional Energy Poverty
headcount for Pakistan 11 percent, 8 percent and 18 percent,
respectively.
Figure 9 shows percentage of households deprived in exact number of
deprivations in overall Pakistan. In overall Pakistan, 95 percent
households are deprived when we set k=1. Households deprivation in
energy decreases with the increase in value of cut offs.
5. CONCLUDING REMARKS AND POLICY RECOMMENDATIONS
Based on results, the study concludes that there is significant and
higher incidence and severity of energy poverty in rural areas as
compared to urban areas, in overall Pakistan. Value of MEP Headcount for
rural Pakistan is 71 percent as compared to 29 percent in urban areas of
Pakistan. Results show that Multidimensional Energy Poverty headcount
for rural Pakistan is 71.4 percent and 28.6 percent of the households
residing in rural areas of Pakistan are energy non-poor. Households of
Pakistan are most deprived in cooking fuel dimension (55 percent), while
deprivation is the least in dimension of home appliances ownership (15
percent). In urban areas of Pakistan households are more deprived in
dimension of cooking fuels (23 percent) followed by indoor pollution (19
percent). Almost one third households of rural Pakistan are deprived in
dimension of indoor pollution (69 percent). Contribution of rural and
urban deprived households to multidimensional energy poverty in Pakistan
is 71 percent and 29 percent, respectively. Contribution of indoor
pollution (32 percent) to multidimensional energy poverty headcount in
Pakistan is the highest followed by the cooking fuels dimension (31
percent) and collectively these two dimensions contribute up to 63
percent in overall Multidimensional Energy Poverty head count for
Pakistan. Study further concludes that households deprivation in energy
decreases with the increase in value of cut-offs. Overall indoor
pollution, cooking fuel and Entertainment appliances are the three major
contributors, to overall MEP Headcount not only as a whole but region
wise also.
Based on above findings, the study suggests taking special
initiatives to combat Energy Poverty in most deprived areas particularly
the rural areas on priority basis by allocating more funds to them.
Indoor pollution and cooking fuel being the major contributors to
overall multidimensional energy poverty in overall Pakistan and regions
also, energy poverty in these dimensions should be individually
addressed in order to reduce overall multidimensional energy poverty.
Provision of subsidized solar panels, bio-gas plants and modern cooking
stoves can help a lot in this regard.
Comments
This paper is a pioneer research in the arena of energy poverty
representing overall Pakistan with national representing data of PSLM
2007-08. The paper explores energy poverty in urban and rural areas with
tabulation and graphical representation. A high deprivation in energy is
seen in rural areas in all provinces.
My major concern with this paper is:
(1) The study uses 2007-08 data although new data set 2010-11 is
also available which will give latest estimates of energy poverty.
(2) The study uses equal weights or simple averages with reference
to Alkire and Foster, (2009). This can be appropriate when the
dimensions have been chosen to be of relatively equal importance as seen
in Alkire and Foster (2007) taking income, health, schooling and health
insurance. But in your methodology it is mentioned that the study uses
Multidimensional Energy Poverty Index (MEPI), proposed by Nussbaumer, et
al. (2011). This study had used weights on the bases of degree of
importance of variables from .13 to .2 for its different indictors. It
would be appropriate if the study uses appropriate weights because
access to electricity for lighting, access to gas/LPG for cooking had
more importance versus ownership of fridge or TV for entertainment.
(3) In Table 1 for indicator 4 and 5 some clarification is needed.
For indicator 4 fridge is used as variable but when you go for cutoff
points you mentioned both fridge or electric fan. Same with indicator 5,
radio/TV is used as variable but in cutoff point you also added
computer.
(4) Finally, you had computed incidence of energy poverty by using
5 variables but not severity of energy poverty. These estimates are only
for urban/ rural break down but not at provincial level but you had
mentioned all in your conclusion.
Finally, I would say that the provision of modern energy services
is recognised as a critical foundation for sustainable development, and
is central to the everyday lives of people. Effective policies to
dramatically expand modern energy access need to be grounded in a robust
information-base.
Rashida Haq
Pakistan Institute of Development Economics, Islamabad.
REFERENCES
Alkire, S. and J. Foster (2007) Counting and Multidimensional
Poverty Measures. Oxford Poverty and Human Development Initiative.
Oxford University. (Working Paper No. 7).
Alkire, S. and J. Foster (2009) Counting and Multidimensional
Poverty Measurement (Revised and Updated). Oxford Poverty and Human
Development Initiative, University of Oxford. (Working Paper No. 32).
Alkire, S., and M. E. Santos (2010) Acute Multidimensional Poverty:
A New Index for Developing Countries. Oxford Poverty and Human
Development Initiative, University of Oxford. (Working Paper No. 38).
Bielecki, J. (2002) Energy Security: Is the Wolf at the Door? The
Quarterly Review of Economics and Finance 42, 235-250.
Catherine, W., et al. (2007) Identifying Fuel Poverty Using
Objective and Subjective Measures (No. 07-11).
Clancy, J. S., M. Skutsch, and S. Batchelor (2003) Finding the
Energy to Address Gender Concerns in Development. UK Department for
International Development DFID Project CNTR 998521.
IAEA (2005) Energy Indicators for Sustainable Development:
Guidelines and Methodologies. Vienna: International Atomic Energy
Agency.
IEA (2010) World Energy Outlook 2010. Paris: International Energy
Agency.
IEA, UNDP, and UNIDO (2010) Energy Poverty-How to Make Modern
Energy Access Universal? Special early excerpt of the World Energy
Outlook 2010 for the UN General Assembly on the Millennium Development
Goals. Paris: International Energy Agency.
Jain, G. (2004) Alleviating Energy Poverty: Indian Experience.
Regulatory Studies and Governance Division. The Energy and Resources
Institute.
Marcio, G. P., M. Aure, L. V. Freitas, and N. F. Silva (2010) Rural
Electrification and Energy Poverty: Empirical Evidences from Brazil, 14,
1229-1240.
Mirza, B. and A. Szirmai (2010) Towards a New Measurement of Energy
Poverty: A Cross-community Analysis of Rural Pakistan. United Nations
University, Maastricht Economic and Social Research and Training Centre
on Innovation and Technology. (UNU-MERIT Working Paper Series 024).
Nussbaumer, P., M. Bazilian, V. Modi, and K. K. Yumkella (2011)
Measuring Energy Poverty: Focusing on what Matters. Oxford Poverty and
Human Development Initiative. Oxford University. (Working Paper No. 42).
Pachauri, S. and D. Spreng (2003) Energy Use and Energy Access in
Relation to Poverty. Switzerland: Centre for Energy Policy and
Economics. (CEPE Working Paper No. 25).
Pachauri, S. and D. Spreng (2004) Energy Use and Energy Access in
Relation to Poverty. Centre for Energy Policy and Economics Swiss
Federal Institute of Technology.
Pachauri, S., A. Muller, Kemmler, and D. Spreng (2004) On Measuring
Energy Poverty in Indian Households. World Development 32:12, 2083-2104.
Pasternak, A. D. (2000) Global Energy Futures and Human
Development: A Framework for Analysis. Lawrence Laboratory, U.S.
Department of Energy.
Shahidur, et al. (2010) Microfmance and Poverty: Evidence Using
Panel Data from Bangladesh. World Bank Economic Review 19:2, 263-86.
Stephen, K., J. Kimani, and S. Amenya (2004) Improving Energy
Access: Possible Contribution of RETs to Poverty Alleviation. Global
Network on Energy for Sustainable Development.
Tennakoon, D. (2009) Energy Poverty: Estimating the Level of Energy
Poverty in Sri Lanka. Practical Action, Intermediate Technology
Development Group. United Nations.
Rehmat Ullah <awanbzu@gmail.com> Department of Economics,
University of Sargodha, Pakistan. Falak Sher
<muhammadfalak@hotmail.com> Department of Economics, University of
Sargodha, Pakistan. Akhtar Abbas <akhtar.bhatti01@gmail.com> is
Graduate Student, Department of Economics, University of Sargodha,
Pakistan.
Table 1
Selected Indicators and their Cutoffs
Dimension/ Indicator Variable Cutoff (Situation
Indicator of Deprivation)
Cooking Modern Type of A household
cooking fuel cooking fuel considered poor/
deprived if using
any fuel beside
electricity,
liquefied
Petroleum Gas
(LPG), kerosene
oil, natural gas,
or biogas for
cooking purposes.
Indoor Indoor Food cooked A household
Pollution pollution on stove or considered poor/
open fire if deprived if not
using any using modern cook
fuel beside stove or use three
electricity, stone cook stove
LPG, natural or if using any
gas, or fuel for cooking
biogas beside
electricity,
liquefied
Petroleum Gas
(LPG), natural
gas, or biogas.
Lighting Electricity Has access There is no proper
access to data for lighting;
electricity therefore for the
purpose we use
electricity
access. A
household
considered
poor/deprived if
the household has
no electricity
connection or
access to
electricity
facilities.
Services Household Has a This dimension
provided by appliance fridge/ deals with
means of Ownership Electric fan ownership of
Household household
Appliances appliances. A
household
considered poor/
deprived if the
household has not
a fridge or
electric fan.
Entertainment/ Entertainment/ Has a radio/ This dimension
Education education television deals with
appliance ownership of
ownership Entertainment/
education
appliance. A
household
considered poor/
deprived if the
household has not
Radio or
Television or
Computer.
Fig. 1. Results of Multidimensional Energy Poverty Headcount
for Overall Pakistan at K=2
Deprived 54.6%
Non-Deprived 45.4%
Note: Table made from bar graph.
Fig. 2. Results of Multidimensional Energy Poverty Headcount
for Urban Pakistan at K=2
Deprived 29%
Non-Deprived 71%
Note: Table made from bar graph.
Fig. 3. Results of Multidimensional Energy Poverty Headcount
of Rural Pakistan at K=2
Deprived 71.35%
Non-Deprived 28.65%
Note: Table made from bar graph.
Fig. 4. Dimension wise Breakdown of Energy Poverty for Overall Pakistan
Deprived Non-Deprived
Cookingfuel 55% 45%
Indoorpollution 52% 48%
Electricity 19% 81%
H.appliance 15% 85%
E.appliance 33% 67%
Note: Table made from bar graph.
Fig. 5. Dimension wise Breakdown of Energy Poverty for Urban Pakistan
Deprived Non-Deprived
Cookingfuel 23% 77%
Indoorpollution 9% 81%
Electricity 7% 93%
H.appliance 3% 97%
E.appliance 18% 82%
Note: Table made from bar graph.
Fig. 6. Dimension wise Breakdown of Energy Poverty for Rural Pakistan
Deprived Non-Deprived
Cookingfuel 58% 42%
Indoorpollution 69% 31%
Electricity 29% 71%
H.appliance 22% 78%
E.appliance 44% 56%
Note: Table made from bar graph.
Fig. 7. Results of Contribution of Region-wise Deprived Households to
Multidimensional Energy Poverty Headcount for Overall Pakistan
Urban 2%
Rural 71%
Note: Table made from pie chart.
Fig. 8. Results of Dimension-wise Contribution to Multidimensional
Energy Poverty Headcount for Pakistan
Cookingfuel 31%
Indoorpollution 32%
Electricity 11%
H.appliance 8%
E.appliance 18%
Note: Table made from pie chart.
Fig. 9. Results of Percentage of Deprived Households at
Different Cut Offs
1 95%
2 54%
3 22%
4 12%
5 1%
Note: Table made from bar graph.
