Asset Poverty in Rural India

Swati Dutta - September 2012

Measuring Poverty

Poverty is usually measured in terms of income or consumption expenditure. The main reason for selecting income as one of the indicators of poverty is that it gives a target certain standard of living. Moreover information on income is easily available and we can readily calculate the number of people whose standard of living is below a pre- determined level of income. However this does not tell us anything about who the poor are, why they are poor and whether their situation will improve in the long run. In India a major policy objectives towards poverty relief has been to cater to the consumption needs of the poor and not necessarily to enable them to get out of poverty. This type of benefit merely addresses certain aspects of a social and economic problem like a natural disaster, an unexpected negative income shock, or economic stagnation. It is focused only on the consequences of poverty, but not the root causes of the poverty.

However, if a household faces any temporary income shock then if they forced their children to withdraw from school they thereafter permanently hamper human capital formation. If human capital formation is affected in the first stages of life it will affect the ability to build financial capital and physical capital. Therefore a poor person also needs insurance against certain types of shocks which affects basic asset formation.

In India, the government has tried to reduce poverty by increasing economic growth which is solely measured by raising per capita gross domestic product.  However to improve the policy effectiveness in order to reduce poverty, we need to focus on a variety of deprivation measures instead of concentrating on income poverty only.  The main distinguishing feature of this paper is to measure a new area of poverty based on a rural household’s access to basic assets across Indian States.  To be poor means not only having a low income but it also means a lack of assets.  An asset based approach to poverty reduction focuses on assisting the poor to develop their stock of wealth and to use it effectively to achieve sustainable improvements in their lives. 

An asset based poverty reduction programme is effective because an analysis of a varied and complementary set of assets can address the different and interactive causes of poverty.  Additionally, the approach allows for intervention which addresses the particular needs of the poor with a focus on the household in a microeconomic context.  Reducing risk and vulnerability and fostering resilience has become an important concern in recent poverty reduction programs.  Directed pro-poor policies are as effective as targeted policies. Examples are keeping children at school, reforming property rights, providing access to drinking water or just any other social benefit which could be income enhancing or asset enhancing.  This research will contribute towards understanding how the level of the poor’s assets give a better understanding of their vulnerability and with this understanding how pointed policies can be devised to decrease persistent poverty across Indian States.


Poverty Literature Review

The most common approach to poverty measurement relies on household expenditure/consumption or income data at a single point in time (Foster et al., 1984).  Once a money metric poverty line is defined, the population can be divided into poor and non-poor and standard measures of headcount ratios and standard Foster, Greer, and Thorbecke (FGT) [1] measures can be calculated to determine the extent and depth of poverty within an economy (Foster et al. 1984).  Baulch and Hoddinott (2000) report detailed studies of poverty dynamics based on panel data from ten African countries and in a survey Hoddinott (2003) found that the number of panel studies of African poverty had risen substantially.  A common finding across all of these studies is that transitory poverty comprises a rather large share of overall poverty.  The large share of transitory poverty is based on income or expenditure data which highlighted the inherent stochastic or random nature of flow-based measures of welfare.  People seem to be better off in one period compared to another without any significant change in their underlying circumstances, particularly with respect to the stock of their productive asset.  This result was basically due to random price and yield fluctuations and from irregular and stochastic earnings from remittances, gifts and lotteries.

In India, the major focus of poverty research has been on the pattern of regional variation and on the incidence of rural poverty and its determinants.  Datt and Ravallion (2002) showed that India has maintained its general progress towards poverty reduction, but that there exists large differences across States.  Poverty fell during the 1990s, but not as much as the economic growth rate would have predicted.  According to Datt and Ravallion, growth did not occur in the States where poverty was highest as would be expected in accordance with the idea of convergence.  Certain types of initial inequalities impede the prospects for growth-mediated poverty reduction, such as asset inequality (land) and education.  The States with low levels of human capital and low farm productivity have a lesser capacity to reduce poverty in response to economic growth.  This shows that growth cannot be enough to reduce poverty in the Indian States, as causality is also shaped by inequalities in human capital and disparities between rural and urban areas.  As noted by Deaton (2004), economic development has been increasingly conceived from a policy perspective as aimed at poverty reduction rather than simply economic growth enhancing.  Mehata (2003) found spatial estimates at various levels of disaggregation which reflect convergence of deprivation in multiple dimensions or multidimensional poverty across all India districts.  There are some States which are performing extremely well in terms of all development indicators such as health, education, and infrastructure, but there may be districts within these that are most deprived in the country too.  Hence he concludes that no single indicator can capture the complexities of development in a country particularly such a complex phenomenon such as poverty. 

The most recent trend in poverty research is to define poverty within an asset space. A prominent early asset poverty measure was perhaps offered by Oliver and Shapiro, (1990) and by  Sherraden, (1991).  These authors define a household as asset poor if its net financial asset value is zero or negative.  A review of current asset based approaches, however, shows that there is no consensus for a single analytical framework for this measure indeed the asset based approach has generated a wide range of studies that can be classified into three broad categories.  The first deals with an asset vulnerability framework (Moser 1998), which highlights the relationship between vulnerability, risks, and asset ownership, identifying not only the risks but also resilience in resisting or recovering from the negative effects of a changing environment (Zimmerman and Carter, 2003; CPRC, 2004).  The second category of studies helps to understand the asset-based approach to poverty reduction (Carter and May, 1999 and 2001; Carter and Barrett, 2006; Carter et al. 2007).  Asset-based approaches have been developed to address the causes and dynamics of longer-term persistent structural poverty primarily in rural Africa and in Asia (Sahan and Stifel, 2000; Naschold, F. 2005 and 2009). The approach helps us to understand the root causes of chronic and transient poverty.  The third category includes community based asset building strategies for poverty reduction (Hulme, 2006).

An asset based poverty evaluation is an important issue in the Indian context, but there has not been much rigorous work done using nationally representative household data.  There are some studies which discuss the wealth inequality in Indian States and measurement issues such as under representation, under- reporting and mis-valuations of wealth and how to overcome those problems by using large samples, appropriate survey techniques and identifying proper price deflators (Subramanian and Jayaraj, 2006; Jayadev, 2007a and 2007b). The latter study found that the wealth gap is strong in the enrolment of children across Indian States (Flimer, Pritchett, 2001). However these studies do not discuss the role of assets in poverty reduction. This paper based on a study by the author attempts to bridge this gap. The objective is to analysis the trends in asset poverty in Indian rural States and to see how the individual States have been performing over time in reducing asset poverty.



In order to construct a Composite Poverty Index the author applied the technique known as Multiple Correspondence Analyses (MCA) to create an asset index for all Indian States based on data from Demographic and Health Survey (DHS) of India for the years 1992, 1998 and 2005. MCA is part of a family of descriptive methods (such as clustering & factor analysis, and , Principal Components Analysis) which reveal patterning in complex data sets  MCA allows an analysis of the pattern of relationships of several categorical dependent variables, ie, variables that can take on one of a limited, and usually fixed, number of possible values. (Asselin, 2002).  There are several studies which have used the MCA score to generate a composite poverty index (Moser, C. and Felton, A. 2007, Filmer, D. and K.Scott 2008). A composite asset index for each household CPI for each state was calculated. [2]

In using the asset indices to consider the evolution of assets over time, it is necessary to construct asset indices that are comparable over time.  The author constructed the asset index is constructed by using pooled weights obtained from the application of MCA techniques to all three surveys for all the common States.  The CPI constructed using MCA has a tendency to be negative in populations of the lowest deciles.  To obtain positive asset values required for the further analysis, a value equal to the greatest negative value is added to each of the asset index values, so that the lowest observed values become zero (Asselin, 2002).  A small further magnitude is also added to make the lowest value non-zero.  We used this composite asset index to estimate poverty for each Indian States using the appropriate household survey weights. 

To calculate the asset poverty ratio or asset headcount ratio we will use the Foster, Greer, and Thorbecke (FGT) measure discussed above.  The major challenge in using the FGT measure is to define the poverty line in a proper way in the Indian context.  In the case of a traditional money metric poverty line, the poverty line is derived from a per capita consumption expenditure level based on a minimum calorie intake requirement.  For an asset index, however, there is no indication of what would be an appropriate asset poverty line.  The poverty line for an individual has been chosen arbitrarily, instead of an official one.  The official poverty line has been under debate in India, and various alternative poverty lines have been proposed.  In 2005, as per the official poverty line defined by the Planning Commission, the poverty head count ratio (HCR) in India was 27%. However, at the US$1.25 and US$2 poverty line standards of the World Bank; the HCR in India was 41.6% and 75.6% respectively.  Moreover, at the Asian Development Bank standard poverty line of US$1.35, the estimated HCR was around 60% (Chen and Ravallion 2008; Himanshu 2009). In order to avoid the uncertainty encompassing the poverty line, the 26th percentile is taken as the poverty line for the purpose of this paper.  The justification for taking this figure can also be given by the argument that it is near the average of the Planning Commission’s poverty line (Mohanty and Ram, 2011) and equivalent to a US$1.02 poverty line given by the World Bank.


For comparison purposes across time and across Indian States, the author used the common poverty line, constant across time and across States.  The FGT measure of poverty around a static asset poverty line say A to calculate asset head count ratio, [3] so that if the asset level of any household Ai is less than the static asset poverty line, an aggregation gives a head count measure of the poor.


Data Employed

The author utilized secondary information mainly from Demographic and Health Survey (DHS) data for the period 1992, 1998 and 2005 on household’s various assets.  The DHS in India, known as the National and Family and Health Survey (NFHS), was first conducted in 1992-93 and the second and the third rounds were conducted in 1998-99 and 2005- 06 respectively.  Household assets are defined as stock of financial, physical, human, natural or social resources that can be acquired, developed, improved and transferred across generations (Ford, 2004).  In the current poverty-related development debates, the concept of assets or capital endowments includes both tangible and intangible assets, which are broadly identified as natural, physical, financial, human and social assets.  However, in this study we have not incorporated the social assets because DHS data do not report themNatural assets include agricultural land and livestock which helps to maintain the sustainable livelihood of the people in the rural areas.  Physical assets are generally defined as the stock of plant, equipment, infrastructure and other resources owned by individuals, business and the public sector (World Bank, 2000).  In this study, however physical assets include various types of consumer durables or household amenities and the quality of housing.  Housing is the most important component of physical assets.  The DHS presents data on the quality of houses based on the material used for construction of walls and roof separately.  If both the walls and roofs are made of pucca material, a house is classified as pucca. Pucca dwellings are ones that are designed to be solid and permanent. The term is applied to build substantial material such as stone, brick, cement, concrete, or timber. The term comes from the Hindi word pakkā, literally 'cooked, ripe.

If the wall and roof are made of kutcha material the house is classified as kutcha, from kaccā 'raw, unripe', referring to buildings of flimsy construction.  In all other cases the house is classified as semi-pucca.  A wall is considered kutcha if the material used includes grass, leaves, bamboo, mud, un-burnt brick or wood.  It is pucca when the material used is burnt brick, metal sheets, stone, cement or concrete.  Similarly, a roof is considered kutcha if the material used is grass, leaves bamboo, mud, un-burnt brick or wood.  It is considered pucca when the material used includes tiles, slate, corrugated iron, zinc or other metal sheets, asbestos, cement sheets, bricks, lime, stone and concrete.  As a proxy for the standard of living within households we also include the quality of the drinking water facility, the toilet facility, the type of cooking fuel, and various household amenities such as electricity, television, radio, bicycle, watch, fan, water pump, and the kitchen facility within household.  A financial or productive asset comprises savings, credit, jobs and employment opportunities, and non-earned income used by people to achieve their livelihood objectives and to invest in new livelihood assets.  However the DHS data source is limited in scope for that information.  In our study, productive assets count as financial assets because they represent a current or potential income stream.  In the context of Indian rural States, sewing machines, tractors, thresher, animal drawn carts are all key indicators for productive assets.  Human assets include skills, knowledge, labour and the capacity to work, which will vary according to household size, skill levels, education, leadership potential, health status, etc.  A human asset is a prerequisite for using the other types of livelihood assets.  Education is the foremost indicator of human assets, or human capital.  For this paper, we have considered the education level of the male members of a household, the average highest education of the women and children (aged between 5-14 years)enrolled at school.  The ealth status of the household is also measured with any women aged 15-49 years with low body mass index (BMI) – norm of low BMI being lower than 18.5 is identified as underweight.  The body mass index directly represents the nutritional state of a household.  BMI data are present only for the female in DHS data, which is not enough, but can be taken as a proxy indicator because of the importance of women’s health.  The study also includes information on any family member who suffer from severe diseases (e.g. tuberculosis, heart disease) in order to understand the health status of households.


Trends in Household Assets

 Pior to making an assessment of asset based poverty in India, it is instructive to report how household asset ownership has changed over the period 1992 -2005 (Table 1).  The proportion of households having an electricity facility has increased from 30% in 1992 to 65% in 2005.  Household ownership of radio has declined from 34% in 1992 to 28% in 2005, but the proportion of households owning a television has drastically increased from 8% in 1992 to 30%.  Also, although the proportion of pucca houses increased from 7% in 1992 to 40% in 2005, still 19% of the households lived in kutcha houses in 2005.  In the case of a drinking water facility, in rural areas mostly people depend on public water.  In 1992, 44% of households had a public water facility whereas by 2005 this had increased to 70%.  However, only 12 % of households had a piped water facility in 2005, which shows very little improvement from 7 % in 1992 and 9% in 1998 for the same category of asset.  Although, the proportion of households having flush toilet increased from 1% in 1992, 11% in 1998 and 23% in 2005, but still in 2005 72 % of the household’s do not have toilet facility within their houses. 

An important point is that, though household’s access to agricultural land decreased from 64% in 1992 to 58% in 2005, however, their ownership to livestock increased from 29% in 1992 to 66% in 2005.  Concerning human assets, child school enrolment increased from 34% in 1992, to 65% in 1998 and to 86% in 2005.  There is also significant improvement in the state of women’s education within the household.  The secondary education of women increased from 2% in 1992 to 16% in 2005.  The data reveals, therefore, that there has been an improvement in the household asset allocation between 1992 and 2005 indicating that household well-being has improved.



Poverty Analysis Using FGT Index

To consistently compare the asset poverty across all Indian States, all the three surveys (1992, 1998 and 2005) were pooled to estimate asset weights by using the Multiple Correspondence Analyses (MCA) technique and to construct household asset index.  The MCA based on the 23 variables (primary indicators as in Appendix Table A2 of the Survey) and 55 categories are used to construct the CPI for each household. The weights attributed to the variable categories are presented in Table 2.  Weights with smaller or negative numbers indicate lower welfare i.e. higher poverty, while larger numbers indicate higher welfare and lower poverty.  To use these weights, the CIP must be monotonically increasing for each primary indicator (Asselin, 2002).  This axiom means that if a household improves its situation for a given primary variable, then its CIP value increases so that its poverty level decreases (larger values mean less poverty or equivalently, welfare improvement).  The largest positive scores are observed to be associated with goods and services comfort whose access is limited to certain number of well - off households.  The richer the households, the more access they have to these goods and services which include, possession of a television, pucca house, piped water facility, flush toilet facility, modern source of cooking fuel such as LPG, sewing machine and acquisition of literacy of household members.  The categories associated with the largest negative scores on the first axis are the most accessible goods and services.  The poorer the households, the less they possess of such goods.  This is a situation in which households may lack a bicycle, have no access to safe drinking water and/or hygienic toilet, an illiterate household member and/or child not enrolled in school.


Before analysing the poverty index, it is useful to start with the descriptive statistics of the asset index score.  The statistics are presented in Table 3.  Once the asset index is constructed, the poverty line is chosen as the 26th percentile of the pooled distribution of the indices (see Table 3).  The FGT Index measure of poverty is then applied to each of the States for the period 1992, 1998 and 2005 in order to calculate the asset poverty ratio for each State and for all the three periods (Table 4).  It is seen that irrespective of the time period, asset poverty declined in all of the Indian States except Orissa, (Figure 1).



However, the data also shows that the ranking of the States are different in different periods (Table 5).  The States Jammu &Kashmir, Kerala, Goa, Delhi and Punjab have stood on top of the nineteen States in the three years 1992, 1998 and 2005.  One notable change is also the movement of Kerala to the second position from the fifth position.  Uttar Pradesh, Bihar, Orissa, Madhya Pradesh and Karnataka were at the bottom in terms of ranking in asset based poverty in 1992.  However the States of Madhya Pradesh and Karnataka have shown positive movements over the years as shown in the Table 1 and have moved to eighth and tenth position respectively in 2005.  Notable among other better performing States are Gujarat and Rajasthan as indicated by the Table 5.  Between 1992 and 2005 we however find that Tamil Nadu, Maharashtra, West Bengal and Andhra Pradesh have loosed their position in terms of asset based poverty.  All other States have remained almost the same in relative ranking.




It is also instructive to compare the asset poverty ratio with income poverty (given by Planning Commission’s measurement) for all India States for the period 2004-05. One way of classifying States into four categories is to compare the asset poverty ratio and income poverty ratio (Planning Commission) of all States.  Figure 2 presents this classification. The vertical and horizontal grid lines present the All India income poverty ratio and asset poverty ratio respectively. The North-East quadrant represents the States where both asset poverty ratio and poverty head count ratio are above the all India average. These states are both chronically income poor and asset poor. States like Bihar and West Bengal fall into this this category. The North West quadrant indicates those States where asset poverty ratio is below all India average but income poverty ratio is above all India average. States like Madhya Pradesh and Maharashtra are in this category. It can be noticed that the South West quadrant presents States where both asset poverty and income poverty ratio are below the all India average. On the other hand in South East quadrant States are those with above all India asset poverty ratio but below all India income poverty ratios.



Assets serve as a buffer to reduce vulnerabilities of poorer households from becoming poorer. This study reveals from the data analysed that there was an improvement in the household asset allocation from 1992 to 2005 which indicates that household well-being in India has improved. It is also seen that irrespective of the time period, asset poverty declined in all the States except Orissa, however the ranking of the States are different in different time periods. Between 1992 and 2005 we find that Tamil Nadu, Maharashtra, West Bengal and Andhra Pradesh have worsened their position in terms of asset based poverty.  However, States like Rajasthan and Kerala have improved their position. The study has categorized States on the basis of income poverty and asset poverty. States like Bihar and West Bengal are both chronically income poor and asset poor. The major policy implication for these States is that, the government needs to focus on building the basic requirements for households there..  Therefore, a priority should be given to build up assets through a combination of financial and community building programmes.  This could be in the form of free quality school education for the children, pucca house construction, and access to health enhancing facilities etc.

States like Madhya Pradesh and Maharashtra provide examples of high income poverty ratio, but mild asset poverty ratio. Therefore, the major policy implication for these states is to secure the income generating process and smooth their consumption stream.  However States like Orissa, Uttar Pradesh, Andhra Pradesh and Tamil Nadu are asset poor but not income poor. To build up their assets governments can provide microfinance loan facilities.  Microfinance loans can help to build animal husbandry and other small scale business which can help households build up productive assets and improve their standard of living.  The main policy crux of the paper which uses available data constructively is that to reduce poverty more effectively we need a different subset of policy instruments for various types of poverty. Poverty takes several forms and must be identified correctly, and once identified it needs the right solution to activate change.



Asselin, L.M. (2002). “Composite Indicator of Multidimensional Poverty”, CECI, June.

Baulch, B. and Hoddinott, J. (2000). Economic Mobility and Poverty Dynamics in Developing

Countries, London, Frank Cass.


Carter, M. R. and May, J. (1999). “Poverty, livelihood and class in rural South Africa”, World Development, 27(1), 1–20.


Carter, M. R. and May, J. (2001). “One kind of freedom: the dynamics of poverty in post-apartheid South Africa”, World Development, 29(12),1987–2006.


Carter, M. R & Barrett, C (2006). “The Economics of Poverty Traps and Persistent Poverty: An Asset Based Approach”, Journal of Development Studies, 42(1): 178-199.


Carter, Michael R., Peter. D. L., Tewodaj. M. and Workneh. N. (2007). “Poverty Traps and Natural Disasters in Ethiopia and Honduras”, World Development, 35 (5), 835-56.


Chen, Shaohua and Ravallion, Martin (2008). “The Developing World is Poorer than we Thought, but no Less Successful in the Fight against Poverty”, Policy Research working paper 4703, Washington DC, World Bank.


Chronic Poverty Research Centre (CPRC). (2004). “The Chronic Poverty Report 2004-5”, University of Manchester: Institute for Development Policy and Management.


Datt, Gaurav & Ravallion, M, (2002). “Is India’s Economic Growth Leaving the Poor behind?” Policy Research Working Paper Series 2846, The World Bank.


Deaton, (2004). “Measuring Poverty”, Working Paper 170. Princeton University, Woodrow Wilson School of Public and International Affairs, Research Program in Development Studies. 



Flimer, D and Lant, P. (2001).  “Estimating Wealth Effects without Expenditure Data-or Tears: An Application to Educational Enrollments in States of India”, Demography,38(1),115-132.


Ford Foundation. (2004). “Building Assets to Reduce Poverty and Injustice”. New York: Ford Foundation.


Foster, J., J. Greer, E. Thorbecke (1984). “A Class of Decomposable Poverty Measure”,

Econometrica, 52(3): 761-766


Himanshu (2009).  “New Global Poverty Estimates: What do these mean?”


Hoddinott, J. (2003). “Pathways from poverty in sub-Saharan Africa”, International Food Policy Research Institute Working paper.


Hulme, D. (2006). “Asset-based Approaches to Poverty Reduction in Bangladesh, and BRAC’s Ultra poor Programme”.



Jayadev, A; S. Motiram and V.Vakulabharanam (2007a). “Patterns of Wealth Disparities in India During the Era of Liberalization”, Economic and Political Weekly, 42 (39), 3853-63.


Jayadev, A; S. Motiram and V.Vakulabharanam (2007b). “Imagined Problems in Computing Wealth Inequalities”, Economic and Political Weekly, 42 (51), 69-71.


Mehta, A.K. (2003). “Multidimensional Poverty in India: District level Estimates”, CPRC-IIPA Working Paper No. 9, Chronic Poverty Research Centre (CPRC), Manchester, UK.



Mohanty, K.S and F. Ram (2011). “Pattern of Poverty Reduction and Fertility Transition in India”, IIPS Working paper series 5.


Moser (1998). “The Asset Vulnerability Framework: Reassessing Urban Poverty Reduction Strategies”, World Development, 26 (1):1-19.


Naschold, F. (2005). “Identifying Asset Poverty Thresholds – New methods with an application to Pakistan and Ethiopia”, Ithaca, Cornell University.



Naschold, F.(2009). “Poor stays poor: Household asset poverty traps in rural semi-arid India”, Selected Paper prepared for presentation at the Agricultural & Applied Economics Association 2009 AAEA & ACCI Joint Annual Meeting, Milwaukee, Wisconsin, July 26-29, 2009.


Oliver, Melvin L., and Thomas M. S. (1990). “Wealth of a Nation: A Reassessment of Asset Inequality in America Shows at Least One-Third of Households Are Asset Poor”, American Journal of Economics and Sociology 49: 129–151.


Sahn, D.E and D.C. Stifel (2000). “Poverty Comparisons Over Time and Across Countries in Africa”, World Development, 28(12):2123-2155.

Sherraden, M (1991).  Assets and the Poor. A New American Welfare Policy Armonk, NY: M.E. Sharpe.


Zimmerman, F. J & Carter, Michael R. (2003). “Asset smoothing, consumption smoothing and the reproduction of inequality under risk and subsistence constraints”, Journal of Development Economics 71(2):233-260.


Subramanian, S and D. Jayaraj, (2006). “The Distribution of Household Wealth in India”,

UNU WIDER Research Paper No 116.


[1] The Foster-Greer-Thorbecke (sometimes referred to as FGT) metric is a generalized measure of poverty within an economy which measures the outfall from the poverty line, but is also considers inequality among the poor. The FGT measure was developed by Professor Erik Thorbecke, his former student Professor Joel Greer, and another graduate student at Cornell University at the time, Professor James Foster.


[2] Where CPIi is the ith household’s composite poverty indicative score.  Iij is the response of household i to category j and Wj is the weight which we will derive from MCA.  K is the total number of primary indicators. The equation is:





Where Ai is the asset stock of household i and the binary indicator variable  reflects whether the household i’s asset stock falls below the static poverty line. α = 0 gives head count measure of the poor.