Printer Friendly

Poverty Inequality and Social Exclusion.

Byline: _

Although poverty in urban areas is substantial and increasing global poverty is still predominantly a rural phenomenon. According to IFAD (2011) about 70 per cent of the world's very poor people around one billion are rural and a large proportion of the poor and hungry amongst them are children and youth. The report specifies that Neither of these facts is likely to change in the immediate future despite widespread urbanization and ongoing or approaching demographic transitions across regions. Now and for the foreseeable future it is thus critical to direct greater attention and resources to creating new economic opportunities in the rural areas for tomorrow's generations".

The empirical literature suggests that rural areas require specific policies for poverty alleviation and rural development due to the distinctive characteristics of rural life: unfavorable demographic situation remoteness poor infrastructure meagre labour market opportunities low education level and inferior quality of institutions. These rural' characteristics may interact and generate vicious circles' which ultimately amplify the phenomenon of rural poverty.

In contrast it is observed in the context of developing countries that national economic and social policies are generally urban biased which may contribute to rural poverty by excluding the rural poor from the benefits of growth and development. According to Khan (2000) policy biases that generally work against the rural poor include:

Urban bias in public investment for infrastructure and provision of safety nets;

Implicit taxation of agricultural products through so-called support prices and an overvalued exchange rate;

Direct taxation of agricultural exports and import subsidies; Subsidies for Capital-Intensive technologies;

Favouring export crops over food crops; and Bias in favour of large landowners and commercial producers with respect to rights of land ownership and tenancy publicly provided extension services and access to (subsidised) credit.

Moreover social and economic deprivations of rural populations have been neglected and often remain invisible in official statistics documents and policy analyses. Two examples may be mentioned in the context of Pakistan. To determine poverty incidence the official poverty line is estimated at national level instead of using separate urban and rural poverty lines. Second the targeting of the largest social assistance programme (BISP) is based on the poverty score card. Here also a unique score card is used for identification of both urban and rural poor despite the distinct characteristics of each component/segment/group. This situation indicates a lack of public awareness as well as unconsciousness of policy makers around the understanding of sources and drivers of poverty and social exclusion of rural population.

This chapter partly fills the gap by profiling special features of consumption and multidimensional poverty and also evaluates the extent of social exclusion in terms of multiple deprivations.

CONSUMPTION POVERTY IN THE RURAL CONTEXT

Traditionally Household Integrated Economic Surveys (HIES) are used to estimate poverty in Pakistan. These nationally representative surveys are carried out by the Pakistan Bureau of Statistics (PBS) with a sample of around 16 to 18 thousand households across the country. Individual household level (unit record) data of HIES are used to estimate consumption poverty for rural areas.

Estimation of Consumption Poverty Line

Among the various approaches of defining income/consumption or traditional poverty calorific approach' is the most popular in developing countries due to its practicality. Almost in all studies of poverty in LDCs including Pakistan poverty level is defined in terms of food inadequacy which is typically measured by the lack of nutritional (calorie) requirements. Correspondingly the Government of Pakistan adopted this approach for estimating official poverty line. According to the Poverty Reduction Strategy Paper (PRSP-I GoP 2003a) the Planning Commission described the following definition for estimating the poverty line.

Calorific requirement approach wherein all those households (or individuals) are classified as poor who do not have income sufficient to allow a consumption pattern consistent with minimum calorie requirements (2350 calories per adult equivalent per day). It is also assumed that the households earning incomes equivalent to poverty line not only have sufficient food to meet the minimum nutrition requirements but also the non-food requirements".

However the Government of Pakistan does not estimate separate urban and rural poverty lines. As the rural lifestyle in general requires a greater consumption of calories than the urban lifestyle then for any given level of income rural households are likely to consume more calories on average than their urban counterparts. Thus poverty estimates derived from official methodology using unique poverty line for both urban and rural households underestimate rural poverty and overestimate urban poverty.

To get rid of this deficiency the Poverty Research Unit of Social Policy and Development Centre (SPDC) estimates separate urban and rural poverty lines using 2230 and 2550 calories per day per adult as the minimum calorie requirement1 for urban and rural areas respectively. Thus the rural calorie norm (minimum calorie requirement) recommended by SPDC is used here to estimate rural consumption poverty line.

To estimate household expenditures which are required for obtaining the minimum required calories Calorie-Consumption Function (CCF) is estimated. Poverty line is then computed by combining calorie norms and estimated coefficients of the CCF. Poverty can then be used to define the poor by total expenditure falling short of the poverty line; by the average dietary pattern the expenditure would translate into fewer calories than required.

Once a poverty line is defined and hence the individual/household poverty status determined through relating poverty line and household expenditure the question is how to aggregate this information into a single index to proxy the status of a group of individuals. The most popular measure namely the Headcount Index (incidence) assigns equal weights to all poor regardless of the extent of poverty. However there are other measures which are sensitive to distribution among the poor and combine both the incidence and intensity of poverty. Three aggregate measures/indices are estimated: Headcount Poverty Gap and Poverty severity. The formulae and the weights assigned to these indices are described in Box 6.1.

Latest Estimates of Rural Poverty

The estimated rural poverty line (Rs. 2298 per adult equivalent or Rs. 1926 per capita per month) from the latest available HIES data for the year 2010-11 is mapped on household per capita total expenditure for computing various poverty measures or aggregates. Chart 6.1 displays the estimated statistics of poverty incidence (headcounts). It is estimated that overall about 39 percent of the rural population of Pakistan was poor during the year 2011. As expected rural poverty is the lowest in the Punjab province and highest in Balochistan province. The magnitude of rural poverty is almost equal in Sind and Khyber Pakhtunkhwa while poverty in Balochistan province is relatively higher.

Table 6.1 Estimated Rural Poverty Measures - 2010-11

###Head###Poverty

###Count###Gap###FGT2

###Index###Index###Index

###[Incidence] [Depth]###[Severity]

Pakistan###38.66###6.92###1.84

Punjab###35.49###6.21###1.60

Sindh###43.18###7.67###2.01

Khyber Pakhtunkhwa###41.79###8.04###2.40

Balochistan###46.85###8.27###2.06

The information clearly conveys that the plight of the rural people may be obscured by ignoring the analysis of poverty and deprivation separately for the rural context. This is very much evident in the case of rural Sindh.

Table 6.1 summarises the famous FGT aggregate measures of rural poverty. Besides incidence or headcount no significant differences are observed across provinces in the Poverty Gap Index (PGI) or poverty depth. The PGI informs the required per capita contribution to lift poor people out of poverty (as a proportion of the poverty line). Nonetheless here too the magnitude is highest for Balochistan. Similar trends are evident in the measure of poverty severity. It is however worthy to note that poverty depth and severity indices are notional and are generally used to rank regions or territories or to track changes over time.

Most of the analyses of poverty have been carried out at the aggregate rural level due to the sample design of HIES which provides statistically reliable estimates of poverty and other characteristics only at the national or regional (urban/rural) levels. Because of this obstacle very few studies attempted to provide variation in poverty at disaggregated levels especially in terms of agro-ecological differences2 in rural Pakistan. These studies found significant differences in poverty levels; nonetheless these estimates are not representative and not statistically reliable as they have been derived from a district representative survey and thus do not capture the inter-district differences in a particular agro- ecological or climatic zone.

The Poverty Research Unit of SPDC attempted for the first time to predict poverty with the help of non-income poverty correlates at subnational levels by applying small area estimation technique in the context of Pakistan3. The technique employs two surveys: a small survey which is representative at national and regional level (HIES) and a large district representative survey (PSLM). Both surveys are conducted by PBS. This technique is used for this study to estimate consumption poverty at the levels of agro-climatic zones4 of Pakistan (Box 6.2). Chart 6.2 highlights the estimated poverty headcount or incidence for the year 2010-11.

The highest incidence of consumption poverty is estimated for Low- Intensity Punjab" (mostly South Punjab and D.I. Khan of Khyber Pakhtunkhwa) zone followed by Rice-Other Sindh" zone. The estimated poverty incidence of Cotton/Wheat-Punjab" zone is also high. Again this zone consists of districts of south Punjab.

Box 6.2 Pakistan Agro-Climatic Zones

In contrast lowest poverty (15 percent) incidence is estimated for Barani" (rain-fed) zone of Punjab. Moreover about 47 and 41 percent poverty incidence is estimated for Balochistan and Khyber Pakhtunkhwa provinces respectively. These provinces have a very small share in agriculture GDP.

Despite methodological differences and other inconsistencies surprisingly the poverty trends are very much similar to earlier studies described in Malik (2005). High poverty levels are generally observed in Sindh and southern Punjab while lowest level of poverty is observed in barani areas of the Punjab province.

Trends in Rural Poverty

There is consensus among researchers and analysts that economic growth may not always be a sufficient condition for poverty reduction but it certainly is a necessary one. Chart 6.3 confirms this phenomenon by highlighting the inverse relationship between agriculture GDP and rural poverty incidence. A decline of 4 percentage points is observed during the periods 2001 and 2005. The principal factor for this decline in rural poverty was the remarkable growth of 7.5 percent in agriculture in 2004- 05 as against 0.1 percent in the fiscal year 2000-01. In contrast due to the decline in growth in agriculture GDP during 2005 and 2011 poverty level is reverting back and showing an upward trend with an increase of 8 percentage points during 2005-2011 periods.

Chart 6.4 portrays the trend in poverty incidence from 1987-88. All these poverty numbers are estimated using unit record household level data of HIES and by applying throughout a consistent and identical methodology for estimating poverty lines and poverty indices. The chart indicates a rising trend in rural poverty incidence up to the period 2000- 01. However rural poverty has dropped with an annual growth rate of 4 percent during 2001-2005. Again during 2004-05 and 2010-11 estimated poverty incidence has gone up with an annualised growth of 4 percent.

Socio-Economic Correlates of Consumption Poverty Understanding the key demographic and socio-economic characteristics of the poor is an essential prerequisite for the formulation of an effective and meaningful poverty alleviation strategy. An attempt is made to establish links between consumption poverty and social demographic and economic attributes of households. The demographic characteristics include household size dependency ratio age and gender of the head of the household. Access to asset endowments is assessed based on ownership of land and livestock as well as the educational attainment of the head and spouse of the household. Impact of remittances on poverty is evaluated by estimating separate poverty incidence for households which receive domestic or foreign remittances and which do not.

To establish the link between poverty and nature of work in the rural context occupational characteristics are also considered. The analysis is carried out by applying two different methods. First poverty incidences are estimated for various categories of household characteristics. For instance what would be the poverty level of households with less than five family members as compared with households with family size of more than nine This bi-variate analysis although it provides useful insights in terms of poverty determinants fails to provide the net impact of an attribute on poverty status after controlling the other characteristics. Thus a multivariate analysis is supplemented by estimating logistic regression function. The summary statistics of the logistic regression indicate a good-fit of the model with a high percentage of correct predictions and expected signs of all coefficients. The findings of these exercises are collated in Tables 6.2 and 6.3.

Family size and dependency ratio are important determinants of rural poverty. The incidence of poverty is increasing significantly with the increase in family size. About 19 percent households with a family size less than five are designated poor while the incidence is 47 percent of those households which have a family of more than 9 members. Similar differences are observed in the categories of dependency ratio. Very low magnitude of poverty incidence (10 percent) is evident in Table 6.2 for households which have less than 50 percent dependency ratio. Highly statistically significant coefficients of these two characteristics in the logistic regression (Table 6.3) corroborate the importance of population welfare programmes in alleviating rural poverty.

Female headship of households is considered a positive correlate of poverty. The experience of developing countries shows that as heads of households women face all kinds of cultural social legal and economic obstacles that men even poor men do not. However to understand the true impact of female headship on poverty it is essential to integrate the role of transfers and remittances into the analysis. By and large women in Pakistan acquire the status of head of a household in two eventualities. First when men migrate in search of better economic prospects and women temporarily take charge of the household. Such instances are particularly common in northern areas of Pakistan where the phenomenon of out-migration is prevalent. Second when the male head of household dies or departs from the household and woman provides for her family. The results of poverty incidence (Table 6.2) show that in the latter case the probability of the household being poor is high.

In the rural context it is assumed that the education of head or spouse of household does not play an influential role in the income generating activities and hence is not as important as the endowment of physical capital (land livestock machinery etc.). However the findings clearly demonstrate that education of the family head directly or indirectly influences poverty levels. The poverty incidence of households with illiterate head is 42 while it is as low as 8 in cases of households where head has intermediate or higher level of schooling. The findings of multivariate analysis also confirm the role of education of head as the coefficient associated with schooling is negative and statistically significant.

Table 6.2 Consumption Poverty Incidence By Household Characteristics

###[Percentage of Poor Rural Households 2010-11]

###Pakistan###Punjab###Sindh###Khyber###Balochistan

###Pakhtunkhwa

###Overall###Rural Poor Households###34.14###30.06###37.79###37.59###42.30

###Family Size###less than 5###18.86###18.63###26.54###11.70###15.84

###6-9###43.10###43.86###50.44###34.12###39.01

###greater than 9 Members###47.13###43.82###55.36###47.32###42.27

###Dependency Ratio###less than 50%###9.66###5.38###18.48###19.91###15.07

###50%-100%###35.94###35.64###44.57###28.83###32.75

###More than 100%###46.99###46.78###49.93###46.19###42.59

###Headship###Male Headship###35.25###34.08###42.65###31.71###32.51

###Female Headship

###- No Remittance###41.62###43.20###45.56###34.58###42.40

###- Domestic Remittance###24.01###24.58###56.15###20.53###.

###- Overseas Remittance###8.46###6.80###67.99###8.46###.

###Age of Head###less than 25###27.21###27.22###36.64###14.93###22.52

###25-45###35.83###35.53###41.81###31.16###30.44

###46-65###34.39###32.29###45.58###31.68###35.96

###Above 65 Years###28.32###28.93###37.91###19.23###32.50

###Schooling of Head###Illiterate###41.67###42.14###51.49###32.67###36.46

###1-5###34.33###32.29###44.73###25.58###34.61

###6-10###25.01###22.63###34.62###28.60###19.97

###11-12###13.03###8.15###17.88###16.56###22.46

###greater than 12 Years###8.16###2.12###13.01###17.05###6.49

###Schooling of Spouse Illiterate###36.86###36.47###45.12###30.93###33.05

###1-5###24.09###24.09###29.56###15.46###12.08

###6-10###16.43###15.32###22.65###19.24###.

greater than 10 Years###3.36###2.02###7.46###8.74###.

###Household Type###Land Ownership###21.12###21.49###18.44###21.41###20.66

###Share Cropper (Hari)###33.59###28.10###45.62###39.43###40.67

###Non-Farm###42.26###42.39###50.65###34.34###34.47

###Farm Size###Landless###41.57###41.14###50.25###34.74###34.58

###Small Farm (less than 13 Acres)###22.38###22.61###20.90###22.18###23.13

###Large Farm (greater than 13 Acres)###8.70###8.31###9.22###9.14###9.82

###Livestock###No Livestock###38.79###37.96###50.99###31.64###34.34

###Livestock Ownership###29.61###28.62###35.46###27.60###25.89

###Remittances###No Remittances###37.09###35.83###42.65###36.33###32.48

###Domestic Remittances###26.78###28.51###50.99###19.23###16.10

###Overseas Remittances###10.82###7.85###48.76###13.03###29.09

94###Source: Estimated from household level data of HIES 2010-11

Ownership of land livestock and non-residential property are all negatively correlated with poverty incidence. Further medium and large farmers (ownership of land greater than 13 acres) play a dominant role in distinguishing non-poor from poor households. Poverty incidences for landless households small farmers and large farm households are estimated at 42 22 and 7 percent respectively. With respect to type of rural households highest incidence is observed for nonfarm households while about 34 and 21 percent share-cropper and landowner households respectively are designated poor.

Table 6.2 also reveals that remittances especially from overseas are instrumental in improving the standard of living of recipient households. It is evident from the table that poverty incidence is only 11 for those rural households which receive overseas remittances as against percent households which do not receive such remittances.

Nonetheless the remittance variable did not work in the logistic regression model and appeared statistically insignificant with wrong sign perhaps due to the multicollinearity1 problem.

Table 6.3 Results of Logistic Regression

###[Dependent Variable Poor = 1 Non-Poor = 0]

###Estimated###Level of

###Coefficients###Significance

Family Size###.319###.000

Dependency Ratio###-.012###.000

Head Unemployed###.536###.095

Head Wage Employed###.324###.000

Nonfarm Household###.514###.000

Number of Earners###-.213###.000

Age of Head###.005###.036

Education Level of Head###-.044###.000

Education Level of Spouse###-.030###.029

Large Farm Households [More than 13 Acres]###.181###.612

Agriculture Land [Acres]###-.056###.000

Household Asset Score###-.279###.000

Ownership of Non-Residential Building###-.179###.250

Livestock Ownership###-.708###.000

Household Structure Pucca###-.110###.271

Landline phone [PTCL]###-.138###.032

Sindh Province###.610###.000

Khyber Pakhtunkhwa Province###.661###.000

Balochistan Province###1.344###.000

Intercept [Constant]###-1.853###.000

An important determinant of poverty status is the stock of household assets. This variable is constructed by assigning equal weight5 to each of the twenty assets6 listed in the HIES questionnaire. In the logistic regression asset-score" appears highly correlated with poverty status of households. The coefficient associated with asset score" is negative and highly significant.

Consumption Poverty and Micronutrient Deprivation Consumption poverty is based on the premise of food inadequacy in terms of minimum calorie (energy) requirements. To estimate the consumption poverty line or poverty cutoff point average dietary pattern is translated into calories and statistically correlated with household consumption. Nonetheless the impact of other micronutrient deprivations on health and especially on labour productivity cannot be overlooked. Moreover micronutrient deficiency is an important factor which contributes to the poverty trap besides other factors such as no access to credit environmental degradation bad governance poor education system inadequate infrastructure and lack of public health care. Below is an average picture of malnourishment in rural households portrayed by highlighting the extent of deficiency with respect to protein vitamin A iron iodine and zinc.

The intakes of these micronutrients are derived from the dietary pattern of rural households as evident from HIES 2010-11 data on food consumption.

Table 6.4 compares the average nutrient intake with the recommended daily allowance. The calorie intake in rural Pakistan is higher than the recommended requirement (2625 Kcl versus 2550 Kcl) in all provinces except in Sindh. Due to the differences in climatic work and living environment it is not surprising that average calorie intake is the highest in Khyber Pakhtunkhwa province. On the average no significant protein intake deficiency is observed in rural population except for Sind province. However an unpleasant picture emerges with respect to other micronutrient intakes. Average daily intakes of vitamin A Iron Iodine and Zinc are far off the mark as compared to the recommended daily allowance.

Table 6.4Average Nutrient Intake in Rural Pakistan 2011

###[Per Adult Nutrient Equivalent Unit]

###Calorie###Protein###Vitamin-A###Iron###Iodine###Zinc

###[Kcal]###[g]###[RE]###[mg]###[ppm]###[mg]

###Punjab###2636###59###558###16###52###10

###Sindh###2490###51###338###14###50###9

###Khyber Pakhtunkhwa###2703###55###426###17###47###10

###Balochistan###2700###57###332###17###68###11

###Overall###2625###57###487###16###52###10

###Recommended Daily Allowance 2550###57###750###20###150###15

Table 6.5 Extent of Nutrient Intake Deficiency in Rural Households 2011

###[Percentage of Household Reported Nutrient Consumption Below the Recommended Allowance]

###Calorie###Protein###Vitamin-A###Iron###Iodine###Zinc

All Households###[Kcal]###[g]###[RE]###[mg]###[ppm]###[mg]

All Households

Punjab###52.41###49.00###76.72###84.93###98.01###92.74

Sindh###57.99###60.43###95.22###92.02###99.44###97.24

Khyber Pakhtunkhwa###50.74###54.43###87.48###78.92###98.41###91.81

Balochistan###48.47###50.28###92.64###75.29###96.67###88.53

Overall###52.89###51.88###82.46###84.67###98.24###93.14

Poor Households

Punjab###90.34###82.92###92.70###97.95###99.78###99.14

Sindh###93.16###90.51###99.58###99.91###99.76###100.00

Khyber Pakhtunkhwa###89.09###88.67###97.06###97.48###100.00###99.31

Balochistan###85.71###82.46###99.50###97.54###100.00###99.10

Overall###90.53###85.31###95.12###98.28###99.82###99.35

Non-Poor Households

Punjab###33.68###32.25###68.84###78.51###97.13###89.58

Sindh###31.84###38.05###91.98###86.15###99.20###95.18

Khyber Pakhtunkhwa###34.63###40.04###83.45###71.12###97.74###88.67

Balochistan###30.84###35.04###89.40###64.76###95.10###83.53

Overall###33.40###34.58###75.91###77.63###97.43###89.92

To further elaborate the phenomenon of severe deprivations of micronutrient intakes Table 6.5 has been developed. The table reports the extent of nutrient intake deficiency with respect to recommended daily allowance in rural households. It is evident from the table that in more than 80 percent rural households daily consumptions of vitamin A Iron Iodine and Zinc are below the recommended daily allowance. According to the disaggregated information with respect to household consumption poverty status almost more than 95 percent poor households are deprived in terms of the above micronutrients. The phenomenon of severe deprivations of micronutrient intakes clearly necessitates direct nutritional intervention schemes for the poor to escape from the poverty trap. Simultaneously the dietary trend in non-poor households calls for enhancing the level of awareness regarding knowledge as well as sources of micronutrients.

Although the above exercise of determining household status in terms of deprivation in micronutrient intake is useful7 the formulation of policy for nutritional interventions requires estimates of anthropometric measurement and clinical and core biochemical assessment of micronutrients especially for target groups (children and women). Specialised nutrition surveys are useful tools that provide estimates of severity and geographical extent of malnutrition in terms of all important nutritional status indicators. These surveys assess the nutritional status of the individual or a representative sample of individuals within a population by measuring anthropometric biochemical or physiological (functional) characteristics to determine the individual status in terms of nourishment.

Table 6.6 Incidence of Malnutrition Rural Pakistan

###2011###2001

###Protein/Energy Malnutrition: [Anthropometric Measurement]

###Children Under Five###Underweight [Weight-for-Age]###33.1###42.3

###Stunted###[Height-for-Age]###45.9###32.5

###Wasted###[Weight-for-Height]###18.0###11.2

###Women###Normal BMI###56.6###56.2

###Nutritional Deficiencies: [Clinical and Bio-Chemical Assessment of Micronutrients]

###Mothers###Iron Deficiency###26.6###38.9

###Iron Deficiency Anaemia###20.5###28.6

###Zinc Deficiency###43.2###44.9

###Iodine Deficiency (Goitre Visible)###3.4###11.8

###Children Under Five###Iron Deficiency Anaemia###33.0###36.8

###Zinc Deficiency###36.4###40.2

###Children - School Age###Iodine Deficiency###35.9###64.0

The latest National Nutrition Survey (NNS) was conducted in 2011 by the Aga Khan University in association with the Pakistan Medical Research Council Nutrition Wing-Cabinet Division (Government of Pakistan) and UNICEF (Pakistan). Table 6.6 furnishes the prevalence of malnutrition among children and women from the findings of NNS 2011 which have been made public8 recently. To compare the inter-temporal changes the incidences of malnutrition are also collated from the previous National Nutrition Survey of 2001-2002 (GoP 2004).

According to the table nearly 33 percent of children under five are underweight 46 percent stunted 18 percent wasted 33 percent have iron deficiency anaemia and 36 percent have zinc deficiency in rural Pakistan during the survey year 2011. About 3 percent of the mothers had iodine deficiency with visible signs of goitre while almost 21 percent mothers have iron deficiency anaemia. Moreover about 36 percent school-going children still have iodine deficiency albeit significant improvement has been noted since 2002.

Table 6.7 Variables Used to Assess Multi-Dimensional Poverty

###Dimensions###Variables

###Human Poverty

###Illiterate Head of Household

###Illiterate Spouse

###No child of primary age (5-9 cohort) is in school

###No household member has completed five years of schooling

###Poor Housing

###Congested Household (Households with only one room)

###Congested Household (Person per room greater 2)

###Household with Inadequate Roof Structure

###Household with Inadequate Wall Structure

###Households with no electricity

###Households using unsafe (not covered) water

###Households with no telephone connection (landline or mobile)

###Households using inadequate fuel for cooking (wood coal etc.)

###Households without latrine facility

###Economic and household Assets Poverty

###Households with no home ownership

###Households with no physical household assets

###Unemployed Head of Household

The NNS 2011 concludes that very little has changed over the last decade in terms of core maternal and childhood nutrition indicators. The survey does point towards gains in iodine status nationally following the implementation of a universal salt iodization and promotion strategy but is counterbalanced by substantial deterioration in vitamin A status and little to no gains in other areas of micronutrient deficiencies".

MULTIDIMENSIONAL POVERTY

The traditional uni-dimensional approach which considers only one variable such as income or consumption is popularly used due to its practicality. Nonetheless it is extensively criticised in the literature of welfare and well-being. Critics argue that to understand the complex phenomenon of poverty or to evaluate household or individual well-being holistically a multidimensional exercise is imperative.

Although there has been progress in defining and measuring the multidimensional nature of poverty and ample literature is now available on the conceptual and measurement issues the challenges remain quite serious if the objective is to reach a degree of operationality (for multidimensional paradigm) comparable to that enjoyed by the income poverty paradigm" (Bourguignon 2003).

Despite difficulties and arbitrariness in the measurement and aggregation of household multiple deprivations a multidimensional approach to define poverty has been adopted in many developed and developing countries. The United Nations Development Programme (UNDP) has since 1990 challenged the primacy of GDP per capita as the measure of progress by proposing the Human Development Index (HDI) which combines income with life expectancy and educational achievement. Recently a global exercise was carried out by the Oxford Poverty and Human Development Initiative (OPHI) to develop Multidimensional Poverty Index (MPI) for more than 100 countries with the help of 10 non-income deprivation indicators of education health and standard of living9. The results in terms of countries' ranking and magnitude of poverty have been published in UNDP Human Development Report 2011. However there are some concerns regarding the subjectivity in selecting cut-off points for individual indicators as well as for overall index.

Moreover weights to indicators and sectors are also arbitrarily assigned for developing a composite index.

Due to these shortcomings and subjectivity the Poverty Research Unit of SPDC adopts a somewhat different methodology for estimating multidimensional poverty. Non-income deprivation indicators are combined through Categorical Principal Component Analysis (CATPCA) multivariate statistical technique. Consequently this research follows the methodology10 adopted in Jamal (2012b) to estimate rural multidimensional poverty aggregates. These estimates are derived from PSLM survey data enumerated during 2010-11 2008-09 and 2004-05.

Components of Multidimensional Poverty

The selection of dimensions or components to derive multidimensional poverty is purely based on the appropriate data available in the household surveys. Table 6.7 provides a schematic view of the dimensions and component variables integrated for the estimation of indices of multidimensional poverty. All these variables are binary. A value of 1 is assigned to poor households and 2 to non-poor households.

The extent of human poverty in the household is represented by current and future levels of education deprivations. Two measures illiteracy (head of household and spouse) and children out of school are included in this dimension11. Children between the ages of 5 to 9 who are not attending school are taken to compute out-of-school children at the primary level. Moreover another indicator of education deprivation is included. Households in which no household member has completed five years of schooling are considered poor.

No information regarding infant or child mortality and malnourishment is available in PSLM surveys. The dimension of health deprivation is therefore missing from the multidimensional poverty analysis due to absence of required information.

The housing quality dimension identifies people living in unsatisfactory and inadequate housing structures. It is represented by a series of variables. The housing structure is treated as inadequate if un- baked bricks earth bound materials wood or bamboo are used in the construction of a wall or the roof. Housing congestion is represented by households with only one room and if the number of persons per room is greater than 2. Access to basic utilities is an important aspect of everyday lives of people. Deprivation in this respect includes households with no electricity households using wood or kerosene oil as cooking fuel households with no safe drinking water availability and households with no landline or mobile telephone facility. Households which are lacking essential facilities such as kitchens bathrooms and toilets are also seen as an important poverty dimension.

Due to data constraints only households lacking a toilet facility are included in the poor housing' dimension of multidimensional poverty.

To capture the poverty in endowments non-ownership of house and non-ownership of essential household assets12 are added to the list of variables used to assess the household multidimensional poverty. Further category of households with unemployed head is also treated as poor and included in this dimension.

Estimates of Multidimensional Poverty

Table 6.8 presents national and provincial estimates of multidimensional poverty for the year 2010-11. Multidimensional poverty is estimated with the help of component/object scores. These scores are derived after adjusting with mean and standard deviation (standardising). Thus the estimates reflect relative poverty (or inequality) with reference to mean and should not be interpreted as an absolute poverty.

According to the table 44 percent of rural people of Pakistan were in a state of multiple deprivations in the year 2010-11 and living in desperate condition and eventually being socially excluded. As expected highest incidence is observed in Balochistan province where about 75 percent rural population is multi-dimensionally poor followed by rural Sindh with an estimate of 57 percent. It is however important to

Table 6.8 Multi-Dimensional Rural Poverty Trends

###[Percentage of Multi-Dimensionally Rural Poor Population]

###Head Count Index###Poverty Gap Index###FGT2 Index

###[Incidence]###[Depth]###[Severity]

###Pakistan###43.97###11.72###4.89

###Punjab###36.77###9.82###4.23

###Sindh###57.07###15.32###6.14

###Khyber Pakhtunkhwa###44.05###9.59###3.28

###Balochistan###75.17###26.04###12.61

reiterate the phenomenon which is also observed in the case of consumption poverty. The table reveals that the level of multidimensional poverty of rural Sindh is significantly higher than the poverty estimated for rural Khyber Pakhtunkhwa province.

Chart 6.5 shows inter-temporal changes in the incidence of multidimensional poverty. The estimates show a slight decline (3 percentage points) in rural multidimensional poverty during 2005-2011 periods. Somewhat similar trends are evident in other provinces. The highest (6 percentage points) drop in rural multi-dimensional poverty is observed in Khyber Pakhtunkhwa province.

For policy perspectives it is worth highlighting that consumption or income poverty measure only advocates the case for transfer policies and social safety-nets that alleviate poverty in the short-run whereas multidimensional deprivation measures (literacy enrolment household wealth housing conditions child mortality etc.) remain stagnant in the short-run and document the recommendation of structural socio- economic policies that could alleviate the intergenerational poverty in the long-term. Therefore consumption poverty and multidimensional poverty are not a substitute for each other for policy formulation. Both provide different information in differing contexts.

GEOGRAPHICAL INDICES OF MULTIPLE DEPRIVATIONS

One of the approaches13 of studying social exclusion is through the construction of deprivation indicators often with the purpose of informing and guiding resource allocation among regions or of supporting a case for resource targeting in a particular region. In the context of Pakistan empirics on poverty an additional tool referred to as Index of Multiple Deprivations (IMD) is used for mapping spatial or geographical deprivations. Unlike multidimensional or consumption poverty indices which first determine household status in terms of poverty before developing aggregate measures the IMD is estimated by aggregating indicators at a particular geographical level. For instance to arrive at the tehsil district or provincial estimate of deprived or socially excluded population in terms of any specific indicator both numerator and denominator are correspondingly aggregated at tehsil district or provincial levels.

Moreover multidimensional poverty described above provides an estimate of relative poverty14 and deprivations whereas IMD provides the extent of absolute level of multiple deprivations. In developing or underdeveloped countries where both absolute and relative poverty (inequality) are prevalent it is the absolute level of welfare which is preferred by development planners and policy makers because of urgency associated with starvation malnutrition social exclusion and other afflictions.

Components of IMD

IMDs are made up of separate types or sectors of deprivation each of which contains various indicators in order to give a broad measure of that type of deprivation. This exercise is based on the Pakistan Social and Living Standard Measurement (PSLM) survey datasets. Depending on the data availability in PSLM the attempt is to choose indicators that reflect the poorest segment of society; thus the IMD measures the extent of socially excluded population.

The selected sectors and indicators in constructing indices of multiple deprivations are schematised in Table 6.9 while a brief methodology for developing the composite index is furnished in Box 6.4. Following Jamal (2012a) this study considers 17 indicators to cover a range of social housing and economic deprivations.

Estimated Indices of Multiple Deprivations

Table 6.9 Indicators used to represent Sectoral Deprivations

###Dimensions###Variables

###Education:###Illiteracy Rate (10 years and above) Female

###Illiteracy Rate (10 years and above) Male

###Out of School Children (5-9 Years) Female

###Out of School Children (5-9 Years) Male

###Health:###Lack of Immunization

###No Prenatal Health Care

###No Postnatal Health Care

###Did not Receive Tetanus Toxoid Injection

###Housing Quality:###Household with Inadequate Roof Structure

###Household with Inadequate Wall Structure

###Congested Household (Households with only one room)

###Households without latrine facility

###Housing Services:###Households with no electricity

###Households using unsafe (not covered) water

###Households with no telephone connection (landline or mobile)

###Households using inadequate fuel for cooking (wood coal etc.)

According to Chart 6.6 which displays the extent of rural deprivations overall 38 percent population of rural Pakistan is deprived or multi- dimensionally poor in terms of selected indicators and dimensions (education health housing quality housing services and economic). The provincial phenomenon is very much similar to the trends observed in consumption and multidimensional poverty. About 33 percent rural population of Punjab is deprived followed by Khyber Pakhtunkhwa where the level of deprivation is 36 percent. The highest 54 percent deprived population is estimated for Balochistan Province.

The extent of rural deprivation across agro-climatic zones is displayed in Chart 6.7. Similar to multidimensional poverty the lowest deprivation is estimated for rain-fed (Barani) Punjab. Across agro-climatic zones of Punjab the highest magnitude of IMD is observed in low- intensity' followed by cotton/wheat' Punjab. Major parts of both zones consist of districts of south Punjab. Almost equivalent magnitude (42-43 percent) is estimated for two agro zones of Sindh. The phenomenon indicates that cropping patterns and other agricultural practices in different zones do not impact the standard of living in Sindh province. Again the level of multiple deprivations in Khyber Pakhtunkhwa is less than the levels of deprivation observed in Sindh and Balochistan provinces.

Indices of Multiple Deprivations are also derived from PSLM datasets for the year 2009 and 2005. Table 6.10 furnishes the estimated IMDs for these years. A declining trend is evident throughout the period in the table. It is also evident that the inter-provincial gap in terms of rural IMDs has declined somewhat mainly due to the fact that the rate of decline in Punjab IMDs is lower than that of other provinces especially in the period 2009-2011.

Table 6.10 Inter-Temporal Trends in Rural Deprivations

###2011###2009###2005

###Pakistan###37.7###39.3###48.2

###Punjab###32.7###33.5###40.8

###Sindh###42.6###46.6###57.7

###Khyber Pakhtunkhwa###35.9###38.3###48.4

###Balochistan###53.6###56.6###67.6

INCOME INEQUALITY

Income inequality and poverty affect each other directly and indirectly through their link with economic growth. These interact with one another through a set of two-way links (see Chart 6.8). Some of these links can be explored separately but often one influences another causing indirect effects. For instance inequality can indirectly influence poverty as inequality affects growth and growth in turn influences poverty.

Small changes in income distribution can have a large effect on poverty. A simple arithmetic example can help visualise this. Imagine that the share of national income that goes to the poorest 20 percent of Pakistan's population increases from 7 percent to 7.25 percent. A change in income distribution of one quarter of one percent would barely affect the Gini coefficient but for the poor this represents about 4 percent increase in their total income. Such a small redistribution would have the same effect on poverty as doubling the annual growth (distribution neutral) of national income from 4 percent to 8 percent.

Various summary measures of inequality are furnished in Table 6.11 in order to describe the extent and nature of inequality in rural Pakistan. The Gini concentration ratio is the most widely used measure of inequality. The Gini provides an estimate of resource inequality within a population. It is the most popular and well-known measure of inequality and summarises the extent to which actual distribution of resource differs from a hypothetical distribution in which each person/unit receives an identical share. Gini is a dimensionless index scaled to vary from a minimum of zero to a maximum of one; zero representing no inequality and one representing the maximum possible degree of inequality.

The Gini coefficient for rural Pakistan is 0.37 for the year 2010-11 indicating a high level of income inequality. Provincially Punjab has the most unequal distribution of rural income followed by Khyber Pakhtunkhwa. Interestingly Balochistan the province with the lowest income level in the countryhas comparatively the most equal income distribution.

Table 6.11 Per Capita Income Inequality in Rural Pakistan

###[Gini Coefficients and Income Shares]

###2011###2009###2005

Gini Coefficients

Pakistan###0.357###0.347###0.373

Punjab###0.365###0.373###0.403

Sindh###0.325###0.284###0.278

Khyber Pakhtunkhwa###0.349###0.300###0.347

Balochistan###0.295###0.287###0.230

Income Share of the Lowest 20% of the Population

Pakistan###8.0###8.5###8.1

Punjab###7.2###7.5###7.2

Sindh###8.9###9.3###10.1

Khyber Pakhtunkhwa###8.1###9.0###8.1

Balochistan###9.4###9.5###10.3

Income of the Highest 20% of the Population

Pakistan###43.2###43.4###45.8

Punjab###44.5###45.4###48.3

Sindh###41.9###38.0###38.5

Khyber Pakhtunkhwa###44.1###39.4###43.6

Balochistan###38.8###38.8###34.3

Ratio of the Highest to the Lowest

Pakistan###5.5###5.2###5.7

Punjab###6.2###6.1###6.7

Sindh###4.7###4.1###3.8

Khyber Pakhtunkhwa###5.5###4.4###5.4

Balochistan###4.1###4.1###3.3

2011 2009 2005

The high level of income inequality in Punjab is apparently a consequence of regional contrasts within the province. Middle Punjab has long been regarded as the first region to have adopted agricultural innovations and was the site of the beginnings of the 1960s green revolution in Pakistan. It is however also a region characterised by high population density and declining land-labour ratios. It has the lowest proportion of the workforce involved in agriculture with relatively high landlessness; the workforce is primarily absorbed in the industrial sector (both large- and small-scale). Lower Punjab is mainly agricultural however unlike middle Punjab there continues to be a presence of powerful landlords with high unequal distribution of land. Land distribution patterns and non-agricultural development in lower (south) Punjab are similar to that of rural Sindh.

Between 2002 and 2005 the Gini coefficient for rural Pakistan shows no change in rural income inequality. However a significant deterioration in rural income inequality is observed during the period 2005-2011. The rural Gini coefficient for per capita income has increased approximately 10 percent from 0.35 to 0.37. It is worth noting that consumption poverty has also significantly increased during this period. The provincial trend is somewhat different. Barring Punjab provinces a downward trend in income inequality is observed during the period 2002- 2005. For the period 2005-2011 the Gini shows an upward trend in Punjab and Khyber Pakhtunkhwa provinces while slight decline is observed in Sindh and Balochistan Provinces.

A limitation of the Gini coefficient as a measure of inequality is that it is most sensitive to the middle part of income distribution rather than to that of extremes because it depends on the rank order weights of income recipients and on the number of recipients within a given range. Thus to capture small changes in extreme parts of income distribution the lowest and highest quintile income shares are also computed to supplement the estimates of the Gini coefficient.

Table 6.11 also provides information regarding the share of income accruing to the lowest 20 percent (i.e. the lowest quintile) and to the highest 20 percent (i.e. the highest quintile) of the population. Statistics with respect to income shares show that in 2004-05 the lowest quintile obtained just about 8.5 percent of the national income while the highest quintile obtained 43.4 percent of the income. By 2010-11 the share of the lowest quintile had declined to 8.1 percent and that of the highest quintile increased to 45.8 percent. As a result the ratio of the highest to the lowest quintile has increased from 5.2 to 5.7. Like the Gini the increase in the ratio of highest to lowest overall rural income share clearly indicates deterioration in the rural income distribution during the period 2005-11.

Table 6.12 Distribution of Rural Households Across Primary Activity Groups

###Pakistan###Punjab###Sindh###Khyber###Balochistan

###Pakhtunkhwa

###Non-Agriculture###49###50###54###39###23

###Agriculture###51###50###46###61###77

###Livestock###17###14###26###11###24

###Farm###34###36###20###50###53

Income Inequality across Farm and Nonfarm Households Rural households are generally distinguished in accordance with their access to agricultural land. According to the latest Pakistan Agriculture Census 2010 only 34 percent of rural households are engaged in the crop sector. However this percentage is somewhat higher in Khyber Pakhtunkhwa and Balochistan provinces; the lowest proportion is observed in Sindh province (Table 6.12). Thus it is worth estimating separate levels of income inequality across farm and nonfarm households. The inequality coefficients for diverse sources of income associated with the nature of primary activities will provide some clue regarding the sources of overall income inequality in rural Pakistan.

Table 6.12 furnishes per capita income inequality in terms of Gini coefficients for farm and nonfarm rural households. Interesting observations emerge from the table. High magnitudes of Gini are observed in farm households except in Khyber Pakhtunkhwa province. The difference in the level of inequality is quite significant in Punjab and Sindh provinces the agriculture heartland of the country. In contrast insignificant differences with respect to Gini coefficients are observed in the provinces which have a tiny share in national agriculture value added

Table 6.13 Per Capita Income Inequality Across Farm

###versus Nonfarm Households [Gini Coefficients for 2010-11]

###Farm Household###Non-farm Households

Pakistan###0.419###0.313

Punjab###0.451###0.319

Sindh###0.315###0.247

Khyber Pakhtunkhwa###0.326###0.363

Balochistan###0.244###0.224

The significant disparities in the magnitude of income inequality as evident in Table 6.13 clearly indicate the necessity for formulating a different set of policies for farm and nonfarm households to alleviate poverty as well as to improve income distribution.

Land Distribution Profile

Among the various sources and determinants skewed land distribution is a major constituent part of rural income inequality. According to Adams and He (1995) agricultural income makes the largest contribution to overall inequality. Depending on the year agricultural income accounts for between 35 and 45 percent of overall income inequality. This is largely because agricultural income is strongly correlated with landownership which is distributed quite unevenly both in the area of the report and in rural Pakistan as a whole". Their study was based on a rich panel of data of rural households of four districts of Pakistan. Naschold (2009) who also worked on the above panel dataset concluded that land ownership is a key to explaining the level of inequality but not its (inter-temporal) changes". Therefore to observe the level as well as changes in the pattern of distribution of land ownership in rural Pakistan Tables 6.14 and 6.14 have been developed from agriculture census data.

Table 6.14 which furnishes the size analysis of farm holdings on top and bottom tails of land distribution points towards the highly unequal distribution of land. On the lower tail 68 percent of farms are holdings of less than five acres and the total area under such farms comprises 21 percent of total farm area. In comparison only one percent farms have 50 acres or more: they hold 21 percent of total farm area. The land distribution in Punjab province seems relatively better than that of Sind Province as one percent farms with 50 acres or more hold only 8 percent of total farm areas of the province. As expected the distribution is quite different in Khyber Pakhtunkhwa and Balochistan provinces which possess more or less a phenomenon of subsistence agriculture. The Khyber Pakhtunkhwa province has the highest percentage (83 percent) of farm holdings of less than 5 acres while in Balochistan only 7 percent farms hold 63 percent of total farm area of the province.

Table 6.14 Land Ownership Percent of Farms and Area

###Less Than 5 Acres###50 Acreas and More

###Farms###Area###Farms###Area

###Pakistan

###1990###54###13###2###28

###2000###62###17###2###23###POVERTY INEQUALITY AND SOCIAL EXCLUSION

###2010###68###21###1###21

###Punjab

###1990###53###14###2###27

###2000###62###19###1###15

###2010###68###27###1###8

###Sindh

###1990###36###8###5###41

###2000###43###10###4###29

###2010###51###12###3###23

###Khyber Pakhtunkhwa

###1990###72###25###1###16

###2000###81###33###1###17

###2010###83###37###1###11

###Balochistan

###1990###26###3###10###57

###2000###30###4###8###49

###2010###40###4###7###63

Although the size analysis of farm holdings presented in Table 6.14 gives useful insights a summary measure of inequality in land ownership facilitates a quick comparison of distribution across regions and over time. The famous and widely used Gini coefficient of inequality15 is applied to the data on proportion of farms and land area owned. The estimated magnitudes of Gini are furnished in Table 6.15. Although the estimated Gini for Pakistan is stagnant at the level of 0.63 since 1990 significant variations across provinces are evident. The table also reveals a decreasing trend in Punjab and increasing trends in Sindh and Khyber Pakhtunkhwa provinces. The highest inequality in land ownership in terms of Gini coefficient is observed in Balochistan province.

Impact of Agriculture Prices on Income Distribution

Although Government involvement in the market for important food and cash crops has changed substantially over time it still intervenes to stabilise prices of major crops and agriculture inputs. Recently during the last five years a spike in the commodity prices especially cotton and rice has been observed with the government claiming that it will not only boost production but will also improve the income of growers. It is also argued that subsequently the increase in rural income will not only support the industrial and service sectors through higher consumption but will also benefit the poor through trickle-down phenomenon.

The higher commodity prices provide incentive to growers to bring more acreage under cultivation; generally there exists a direct and positive correlation between procurement support or expected crop prices and the supply. A rough picture16 of the relationship between support/procurement prices and crop production is portrayed in Box 6.5 by plotting crop production and one year lagged real support prices. The correlation coefficients are also computed to provide a summary of the statistical relationship. The highest price responsiveness with the correlation coefficient of 0.79 is observed in case of wheat crop while the lowest (0.42) is estimated for rice crop.

Nonetheless the pertinent concern here is to explore how benefits of rising crop prices are distributed among rural households. Due to the paucity of relevant panel micro-level farm data no systematic study is available to verify the general perception that the policy of support price deteriorates rural income distribution; eventually income disparity in rural areas has widened as a result of rising crop prices. It is argued that:

Incomes from the crop sector are roughly proportional to the distribution of land which is quite skewed and as such any favour or bias towards the crop sector would help large landlords more than small farmers. Only 34 per cent of the rural population is engaged in the crop sector and a vast majority of them are small landholders. This means that only a small proportion of population in the rural areas stands to gain from increasing crop prices.

The transfer of additional cash has widened income disparity in rural society even if many small farmers have also benefited from the soaring crop prices because the trickle-down" has been uneven and limited.

In case of wheat crop the contention of marketable surplus' is often cited to strengthen the argument of worsening rural income distribution due to rising prices. Pakistan Agricultural Prices Commission (APCOM) has conducted a survey in the major wheat surplus districts in Sindh in 1997 and in Punjab in 1998. According to this study (Dorosh and Salam 2006) only 8 and 11 percent share in total sale of wheat crop goes to small farmers (less than 12.5 Acres) in Sindh and Punjab provinces respectively.

Table 6.15 highlights the share in sale of wheat across farm size. Dorosh and Salam (2006) did not disaggregate the share of farmers with land up to 5 acres which is in fact the target group for poverty reduction strategies.

Table 6.15 Trend in Land Ownership Inequality Gini Coefficients

###Pakistan###Punjab###Sindh###Khyber###Balochistan

###Pakhtunkhwa

###1990###0.63###0.59###0.57###0.61###0.66

###2000###0.63###0.58###0.59###0.63###0.65

###2010###0.63###0.55###0.60###0.62###0.75

Table 6.16 Sale of Wheat by Farm Size

###Sindh###Punjab

###less than 12.5 Acres###8###11

###12 to 25 Acres###11###22

###25 to 50 Acres###16###23

###More than 50 Acres###65###44

An attempt is also made to explore the trickle down phenomenon in terms of rural wages. Pakistan Labour Force Surveys (LFS) report wages in overall agriculture (Agriculture Livestock Hunting Forestry Logging and Fishing) sector as well as wages of market oriented skilled and subsistence agricultural and fishery workers. To monitor the trend in rural wages since 1991 LFS data is used for plotting monthly nominal and real (adjusted with CPI) wages. Charts 6.9 and 6.10 furnish the trend for overall agriculture sector and for skilled workers respectively.

According to these charts real wages for overall agriculture sector have declined in the 90s and since then are almost stagnant. However an upward trend is observed in case of skilled agriculture workers in the first half decade of 2000s while in the later half a slight declining trend is evident. Thus the initial analysis of trends in rural wages apparently does not indicate the existence of the trickle down phenomenon.

Notes:

1. The justifications of taking these minimum requirements are described in Jamal (2002). The paper also provides other technical details in term of methodological choices and options available to estimate consumption poverty line.

2. A summary of these studies is provided in Malik (2005).

3. For technical details and poverty estimates at the sub-national levels see Jamal (2007) and Jamal (2013).

4. Box 6.2 provides details in terms of boundaries and districts for each agro-climatic zone.

5. A constant 1 is assigned to each of the assets owned by the household and the assets score is obtained by summing up across all assets at the household level. Of course uniform allocation of score irrespective of the asset characteristics tends to smooth out the distribution of assets across households. To the extent that these assets have different values and all exhibit different rates of depreciation uniform allocation might even increase the distortion in the distribution of household assets. But what actually matters in this construction is the ownership of assets by a household and not so much the values of the asset which are difficult to estimate accurately from surveys. The maximum asset score is 20 and the minimum is 0 for poorest households which possess none of the assets listed.

6. These assets are; iron fans sewing machine video/cassette player tables/chairs clocks TV VCR/VCPVCD refrigerator air-conditioner air cooler computer bicycle motor cycle car tractor mobile Cooking Range Stove/Burner and Washing machine.

7. According to UNICEF (1998) there are two possible ways to assess the adequacy of food and nutrition and to detect the presence of inadequacy in food intake among individuals or population groups: the first measures nutritional intake and the second assess nutritional status"

8. Humanitarian Response Pakistan (http://www.pakresponse.info) http://pakresponse.info/LinkClick.aspxfileticket=scqw_AUZ5Dw%3Dandtabid=117andmid=752

9. For detail see Alkire and Santos (2010) and Alkire and Foster (2007).

10. The methodology is very briefly described in Box 6.3. For detailed methodology see Jamal (2012b)

11. Literacy is defined as the ability of a person to read and write in any language with understanding"

12. These assets are Iron Fan Sewing Machine Radio TV Chair/Table and Watch/Clock.

13. Social exclusion is generally studied from one of three contrasting perspectives: a predominantly structuralist approach; an experiential approach informed particularly by cultural geography; and a more instrumental approach based on statistical indicators.

14. A measure of relative poverty defines poverty" as being below some relative poverty threshold. For example the statement that households with an accumulated income less than 50% of the median income are living in poverty" uses a relative measure to define income poverty.

15 Gini coefficients for this exercise are computed from the grouped data of Agricultural Censuses and hence the magnitudes of coefficients might be different if compared with the Gini computed from individual farm-level data. Due to aggregation bias the estimates from grouped data in general are higher. The standard formula for computing Gini for grouped data is furnished below.
COPYRIGHT 2013 Asianet-Pakistan
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:Annual Review Social Development in Pakistan
Article Type:Report
Geographic Code:9PAKI
Date:Dec 31, 2013
Words:9585
Previous Article:The State of the Health Sector in Rural Pakistan.
Next Article:Social Protection for the Rural Population.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters