research articles on poverty in india

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 17 January 2023

Small area variations in four measures of poverty among Indian households: Econometric analysis of National Family Health Survey 2019–2021

• Anoop Jain 1 ,
• Sunil Rajpal 2 , 3 ,
• Md Juel Rana 4 ,
• Rockli Kim 5 &
• S. V. Subramanian   ORCID: orcid.org/0000-0003-2365-4165 6 , 7  

Humanities and Social Sciences Communications volume  10 , Article number:  18 ( 2023 ) Cite this article

2339 Accesses

5 Citations

8 Altmetric

Metrics details

• Development studies
• Health humanities

India has seen enormous reductions in poverty in the past few decades. However, much of this progress has been unequal throughout the country. This paper examined the 2019–2021 National Family Health Survey to examine small area variations in four measures of household poverty. Overall, the results show that clusters and states were the largest sources of variation for the four measures of poverty. These findings also show persistent within-district inequality when examining the bottom 10th wealth percentile, bottom 20th wealth percentile, and multidimensional poverty. Thus, these findings pinpoint the precise districts where between-cluster inequality in poverty is most prevalent. This can help guide policy makers in terms of targeting policies aimed at reducing poverty.

Similar content being viewed by others

Inequalities in the prevalence of double burden of malnutrition among mother–child dyads in India

Health shocks and vulnerability to poverty in Congo

Investigation of multidimensional poverty in Pakistan at the national, regional, and provincial level

Introduction.

Income and wealth are measures of socioeconomic position (SEP) that have long been connected to health outcomes through myriad pathways and mechanisms (Adler et al., 1994 ; Braveman and Gottlieb, 2014 ; Galobardes et al., 2007 ; Oakes and Kaufman, 2006 ; S. V. Subramanian et al., 2002 ). Impoverished parents are often unable to provide children with adequate nutrition, safe drinking water, or improved sanitation (Karlsson et al., 2020 ; Victora et al., 2003 ). Poor households are also more likely to be in areas that lack access to healthcare, food security, and centralized waste management (Mosley and Chen, 1984 ; Victora et al., 2003 ), and are more vulnerable to the effects of climate change (Hallegatte and Rozenberg, 2017 ), which further exacerbates deleterious health outcomes (McMichael et al., 2007 ; Romanello et al., 2021 ). Poverty is also associated with adverse mental health outcomes (Lund et al., 2010 ; Patel and Kleinman, 2003 ).

In India, the Global Multidimensional Poverty Index found that 271 million Indians were lifted out of poverty between 2006 and 2016 (Initiative et al., 2019 ). However, much of this progress has been geographically varied throughout the country. For example, while national data show a falling poverty headcount ratio between 1983 and 1994, states such as Assam, Haryana, and Himachal Pradesh experienced increases (Himanshu, 2007 ). Additionally, while Andhra Pradesh experienced the greatest decline in multidimensional poverty between 1999 and 2006, Bihar’s reduction was the slowest during the same period (Alkire and Seth, 2015 ). Other studies have examined India’s 88 regions defined by the National Sample Survey Organization according to climate, language, and culture (Chauhan et al., 2016 ). While some of these regions, such as Tamil Nadu and Karnataka, experienced significant declines in poverty between 1993 and 2012, other regions in southern Odisha and Chhattisgarh continue lagging behind (Chauhan et al., 2016 ). Districts have also been targeted with poverty eradication policies, such as investing in industrial and agricultural growth, given significant inter-district disparities within states (Chandra, 2021 ; Chaudhuri and Gupta, 2009 ).

However, single-level analyses assume a certain degree of homogeneity within a given geography despite evidence pointing to significant intra-unit inequalities (Kapur Mehta and Shah, 2003 ; Singh et al., n.d.). Varying agricultural and ecological conditions, for example, are associated with disparate agricultural yields and thus poverty rates within states (Palmer-Jones and Sen, 2003 , 2006 ). These within-region variations can be seen when looking at certain health outcomes, such as child malnutrition, which is one indicator in the Multidimensional Poverty Index (MPI) (Initiative et al., 2019 ). A recent study showed that 93% of the variation in child stunting (height-for-age Z score), an anthropometric indicator of malnutrition, is attributable to between-individual variations (Mejía-Guevara et al., 2015 ). Similarly, 80–85% of the variation in child undernutrition was attributable to within-population differences in India (Mejía-Guevara et al., 2015 ; Rodgers et al., 2019 ). Such evidence points towards the importance of considering variation within geographical units, such as districts, while designing targeted strategies under maternal and child nutrition programs.

Similar types of within-population analyses of poverty throughout India have not been done. For example, the Indian government launched the Aspirational Districts Program (ADP), an initiative targeting the 112 least developed districts (Porter and Stern, n.d. ). While this program targets poverty eradication programs at the district level, it does not take into account the variations in poverty that might exist within districts and between communities. Understanding these small area variations in poverty is important given that previous research has shown how child malnutrition and dietary diversity, indicators of poverty, also vary significantly within districts and between clusters (Jain et al., 2022 ; Rajpal et al., 2021 ).

Given this background, the purpose of this paper was to better understand within-district and between-community variations in poverty in order to inform the effective targeting of poverty-eradication policies throughout India. Doing so is important considering that as per the most recent census data from 2011, almost 22% of people in India lived on less than USD 1.90 per day (GoI 2011 Census of India, 2011 ). Therefore, we examined these variations using four different measures of household poverty. These were (a) bottom 10th wealth percentile; (b) bottom 20th wealth percentile; (c) below the poverty line; and (d) the multidimensional poverty index. We used data from the fifth round of the National Family Health Survey (NFHS) from 2019 to 2021.

Data source and sample

This analysis was conducted using data from the fifth round of the National Family Health Survey (NFHS). These data were collected between 2019 and 2021. A two-stage cluster sampling strategy was employed for household selection. The primary sampling units (PSUs) were clusters, defined as groups of adjacent households. The first stage of sampling involved selecting rural and urban clusters. Clusters containing more than 300 households were divided into smaller groups from which households were selected in the second stage of sampling. No more than 22 households were selected from any given PSU. The NFHS includes data from 2,795,894 de jure household members, nested in 30,170 rural and urban clusters, in all 707 districts, and in all 36 states/union territories. The multilevel structure for the four measures of poverty is presented in Fig. 1 .

Green describes individuals, orange describes clusters, blue describes districts, and gray describes states.

Primary outcomes

We analyzed the small area variation of the percent prevalence of individuals in the bottom wealth decile and bottom wealth quintile. These ranges are constructed by assigning each member of the household the wealth index score. Individuals are then ranked against the entire population based on their scores. This distribution is then divided into equal bins. Both of these outcomes were dichotomized such that individuals were either in the bottom 10th or 20th percentiles, or above.

We also analyzed the small area variation of individuals that have a below-poverty line (BPL) card. These cards are distributed to poor households by state governments, entitling households to 25–35 kg of subsidized grain per month as per state government guidelines. Individuals in BPL households were considered BPL for this study.

Finally, we analyzed the small area variation of multidimensional poverty (MDP). The MDP captures deprivations across three dimensions, health, education, and standard of living. Under health, the MDP includes indicators of nutrition, mortality, and antenatal care. Under education, the MDP includes indicators of years of schooling and current school attendance. Under the standard of living, the MDP includes the type of cooking fuel, sanitation, drinking water, electricity, housing quality, assets, and bank account. We used these indicators to construct a deprivation score following the weighting process outlined in the NITI Aayog MDP baseline report (India National Multidimensional Poverty Index, 2021 ). Households with a score >0.33 are considered multidimensionally poor. Individuals in MDP households were considered MDP for this study.

Statistical analysis

The NFHS data are structured such that individuals at level one were nested in clusters at level two, districts at level three, and states at level four. Each of the outcomes included in our analysis was binary. Therefore, we estimated four four-level variance component models to decompose the proportion of geographic variation attributable to clusters, districts and states for individual i in cluster j , district k , and state l using Eq. ( 1 )

In this model, π ijkl is the log odds of the outcome for individual i . The random effects are the residual differentials for clusters ( u 0 jkl ), districts ( v 0 kl ), and states ( f 0 l ). Each of the residual differentials is assumed to be normally distributed with a mean of zero and a variance of u 0 jkl ~ N(0, \(\sigma _{u0}^2\) ), v 0 kl ~ N(0, \(\sigma _{v0}^2\) ), and f 0 l ~ N(0, \(\sigma _{f0}^2\) ) where the variances quantify the between-cluster, between-district, and between-state variation, respectively. The variance at level one (households) cannot be computed in models with binary outcomes (Kim et al., 2016 ).

The proportion of variation attributable to each geographic level—clusters, districts, and states—was calculated by dividing the variance of a given level by the total geographic variation (i.e., for the cluster level, \(\sigma _{u0}^2\) /( \(\sigma _{u0}^2\)  +  \(\sigma _{v0}^2\)  +  \(\sigma _{f0}^2\) ) × 100). We conducted this analysis in MLwiN 3.05 using the Monte Carlo Markov Chains method with a burn-in of 500 cycles and monitoring of 5000 iterations of chains, the same procedure used in previous studies (Jain et al., 2022 ; Rajpal et al., 2021 ).

Next, we generated precision-weighted estimates specific to each cluster for each outcome. This was done using Eq. ( 2 )

We calculated the standard deviations of these cluster values by district, which would be used to elucidate the small area variation for each outcome. Finally, we generated precision-weighted estimates specific to each district for each outcome. This was calculated using Eq. ( 3 )

Sample characteristics

Of the 2,795,894 individuals sampled in the NFHS-5, 258,808 were in the bottom 10th percentile of the wealth index, while 532,760 were in the bottom 20th percentile of the wealth index (Table 1 ). Of the 2,791,372 individuals living in households with complete BPL data, 1,366,554 were BPL. Finally, of the 441,293 individuals living in households with complete MDP data, 177,563 were multidimensionally poor. The percent prevalence for each outcome by state is presented in Table 1 .

Correlations between measures of wealth

We estimated the correlation values for the district means for each measure. We found strong positive correlations (0.93, p  < 0.001; 0.72, p  < 0.001) between the mean district percent estimates for individuals in the bottom 10th wealth percentile and individuals in the bottom 20th wealth percentile and MDP individuals. We also found a strong positive correlation (0.8, p  < 0.001) between individuals in the bottom 20th wealth percentile and MDP individuals. We found a positive correlation (0.29, p  < 0.001) between BPL individuals and bottom 10th wealth percentile individuals, and a positive correlation (0.34, p  < 0.001) between BPL individuals and bottom 20th wealth percentile individuals. Finally, we found a positive correlation (0.22, p  < 0.001) between MDP and BPL individuals. These results are presented in Fig. 2 .

A Bottom 10th wealth percentile and bottom 20th wealth percentile. B Bottom 10th wealth percentile and BPL. C Bottom 10th wealth percentile and MDP. D Bottom 20th wealth percentile and BPL. E Bottom 20th wealth percentile and MDP. F BPL and MDP.

Relative importance of geographic levels

We found that states were the largest source of variation for individuals in the bottom 10th wealth percentile (66%), the bottom 20th wealth percentile (63%), and BPL households (54%). Clusters were the largest source of variation for MDP individuals (50%). Districts were the smallest source of variation for all four outcomes. A summary of these values is presented in Fig. 3 . The variance estimates for each of the four measures of poverty are presented in Supplementary Table 1 .

Bar graph showing the geographic variance partitioning by clusters, districts, and states for the four measures of poverty.

Small area variation in household poverty

We computed the standard deviations of the predicted cluster wealth index scores by each district. These values can be interpreted as the within-district and between-cluster variations in individual poverty. We computed the standard deviations of the predicted percentage of individuals in the bottom 10th and 20th wealth percentiles in each cluster by the district. The within-district between-cluster standard deviations for individuals in the bottom 10th wealth percentile ranged from 0.0004 to 32.9 with a median value of 6.9. The within-district between-cluster standard deviations for individuals in the bottom 20th wealth percentile ranged from 0.0001 to 33.6 with a median value of 14.2. The within-district between-cluster standard deviation for multidimensionally poor individuals ranged from 0.0002 to 45.6 with a median value of 29.1. Finally, within-district between-cluster standard deviation for households with BPL cards ranged from 2.6 to 31.2 with a median value of 17.6. These ranges, along with the district mean ranges, are presented in Fig. 4 . The district-level predictions, between-cluster standard deviations by district, and cluster-level predictions are presented in Figs. 5 – 8 . We also show the cluster-level prevalence of each measure of poverty by state and Union Territory in Fig. 9 .

A Box plot of district-level percent of households by each poverty measure. B Box plot of the district-level distribution of cluster-level inequality by each poverty measure.

A Geographic prevalence of bottom 10th wealth percentile individuals across 640 districts in India. B District-level distribution of cluster-level inequality for bottom 10th wealth individuals across 640 districts in India. C Geographic prevalence of bottom 10th wealth percentile individuals across 30,170 clusters in India.

A Geographic prevalence of bottom 20th wealth percentile individuals across 640 districts in India. B District-level distribution of cluster-level inequality for bottom 20th wealth individuals across 640 districts in India. C Geographic prevalence of bottom 20th wealth percentile individuals across 30,170 clusters in India.

A Geographic prevalence of BPL individuals across 640 districts in India. B District-level distribution of cluster-level inequality for BPL individuals across 640 districts in India. C Geographic prevalence of BPL individuals across 30,170 clusters in India.

A Geographic prevalence of MDP individuals across 640 districts in India. B District-level distribution of cluster-level inequality for MDP individuals across 640 districts in India. C Geographic prevalence of MDP individuals across 24,416 clusters in India.

A Bottom 10th wealth percentile. B Bottom 20th wealth percentile. C Below poverty line (BPL). D Multidimensionally poor (MDP).

Correlation between district percent and cluster standard deviation

We calculated the associations between the predicted district-level percentages of individuals in the bottom 10th and 20th wealth percentiles and the cluster standard deviations. We found a significant positive correlation between the predicted district percentage of individuals in the bottom 10th percentile and the cluster standard deviation (0.75, p  < 0.001). We also found a significant positive correlation between the predicted district percentage of individuals in the bottom 20 th percentile and the cluster standard deviation (0.75, p  < 0.001). There was a significant positive correlation between the district percentage of multidimensionally poor households and the cluster standard deviation (0.24, p  < 0.001). Finally, there was a slight negative correlation between the district percentage of households with BPL cards and the cluster standard deviation (−0.17, p  < 0.001). These results are presented in Fig. 10 .

A Bottom 10th wealth percentile district-level prevalence and bottom 10th wealth percentile cluster-level standard deviation. B Bottom 20th wealth percentile district-level prevalence and bottom 20th wealth percentile cluster-level standard deviation. C District-level BPL prevalence and cluster-level BPL standard deviation. D District-level MDP prevalence and cluster-level MDP standard deviation.

This paper had four salient findings. First, we found null to moderate correlations between the district mean and SD values for all of the primary outcomes. Second, the largest share of geographic variation for each outcome was attributable either to states or clusters. Third, we found a wide range in the within-district between-cluster SD values for all four poverty measures. Furthermore, while our results show that poverty is generally clustered in north, central, and parts of east India, district-level clustering varies based on the wealth measure being analyzed. Finally, we found significant positive correlations between the percentage of individuals in the bottom 10th and 20th wealth percentiles by district and the cluster standard deviations. However, we found a significant negative correlation between the percentage of multidimensionally deprived individuals in a district and the cluster standard deviations.

There are two data limitations to this study. First, certain questions about household wealth in the NFHS are self-reported. Despite this being a possible source of measurement error, the NFHS data are widely considered to be of high quality (Corsi et al., 2012 ). Second, the precision-weighted estimates presented in this paper could potentially be biased by the fact that we did not adjust for any sociodemographic correlates of wealth, such as caste or household head education.

These findings could help inform anti-poverty policies in several ways. For example, our results point to the importance of considering even smaller geographic units in anti-poverty policy design. We show that a large share of the variation in poverty is attributable to clusters, highlighting the contextual influence these relatively small geographic units play on household-level outcomes. This is consistent with findings from prior studies that also show the critical role of clusters in shaping poverty outcomes in India (Kim et al., 2016 ). This has also been shown in the context of correlates of child undernutrition, a key indicator of household poverty (Jain et al., 2021 ). Thus, poverty-eradication policies such as the Aspirational Districts Program and the Mahatma Gandhi National Rural Employment Guarantee Act need to examine clusters within districts that need to be prioritized to ensure equitable advancement.

Furthermore, there is an extensive body of research documenting rising income and wealth inequality throughout India (Chancel and Piketty, 2019 ; Mishra and Bhardwaj, 2021 ; S. Subramanian and Jayaraj, 2013 ). Some of these studies elucidate between-district differences (Menon et al., 2018 ; Mohanty et al., 2016 ), while others have examined between-state disparities (Alkire and Seth, 2015 ; Anand and Thampi, 2016 ). Yet our findings clearly highlight the fact that variations in household wealth exist at a much smaller geographic scale. This is demonstrated by our analysis of MDP individuals and those in the bottom 10th/20th wealth percentiles, which shows that districts with a higher percentage of poor individuals tend to have greater small area variation. There are a few different explanations for widening wealth inequality between regions throughout India. Between-caste inequality, regional variations in agriculture and climate, and varying degrees of infrastructure investments are some of the possible explanations for persistent wealth inequality throughout India (Chauhan et al., 2016 ; Ghosh and De, 1998 ; Palmer-Jones and Sen, 2003 , 2006 ; Zacharias and Vakulabharanam, 2011 ). Future research should explore the extent to which these factors explain the small area variations in poverty found in this study. Additionally, future research should examine how anti-poverty policies and programs can be tailored to varying within-district and between-cluster conditions so as to avoid a one size fits all approach. Doing so is important given that household wealth is associated with factors such as whether a woman has a skilled birth attendant present at delivery (Kesterton et al., 2010 ), children’s educational outcomes (Bacolod and Ranjan, 2008 ; Cashman et al., 2021 ), and intimate partner violence (Ackerson and Subramanian, 2008 ).

When viewed through the lens of social epidemiology, our results point to the difficulty in accurately measuring wealth as an indicator of socioeconomic position and its impacts on health (Braveman et al., 2001 ; Howe et al., 2012 ; Kawachi et al., 2010 ; Oakes and Rossi, 2003 ). This is emphasized by the fact that not all of the measures are clustered in the same areas throughout India, making them all different in what they might be capturing. This highlights why selecting four different primary outcomes was important given that each one measures something different. Furthermore, our results also point to the importance of measuring area indicators rather than simply individual-level measures of wealth. Previous studies have established the fact that wealth disparities and inequality are strongly associated with health (McMichael, 1999 ; Wilkinson and Pickett, 2006 ). This is important when considering multidimensional poverty given that our findings highlight that districts with a higher percentage of multidimensionally poor individuals have a greater degree of inequality. Thus, our findings pinpoint the precise districts where between-cluster inequality in poverty is most prevalent. This can help guide policy makers in terms of targeting public health and social welfare policies.

Our analysis also underscores the importance of examining the small area variations of the composite indicators of wealth given that poverty is multidimensional and is an overall deprivation in terms of assets and housing quality. Indicators such as access to safe water and sanitation and electricity are important unto themselves (Jain and Subramanian, 2018 ). However, unequal access to these assets can lead to deleterious health and social outcomes. Access to safe drinking water and sanitation is important for child health and psychosocial outcomes among women (Caruso et al., 2018 ; Fink et al., 2011 ; Sahoo et al., 2015 ). Meanwhile, household electrification is associated with increases in women’s empowerment (Samad and Zhang, 2019 ; Standal and Winther, 2016 ), which is similarly associated with improved maternal health outcomes (Grown et al., 2005 ; Roy and Chaudhuri, 2008 ). As such, the unequal distribution of these essential assets within-districts and between-clusters, could help explain small area variations in wealth-based outcomes such as child health (Chalasani, 2012 ; Rajpal et al., 2021 ). Addressing the unequal distribution of these essential assets and goods across small areas in India is particularly important in the wake of the global COVID-19 pandemic, which more than doubled the number of people in India earing $2 or less from 60 million to 134 million between 2020 and 2021 (Kochhar, 2021 ).

In conclusion, previous research has elucidated the extent to which poverty varies between states and districts in India. There are a number of contextual factors that explain these differences. We build on this prior research to show that there also exist small area variations in poverty within districts and between clusters in India. Our results show that the degree of regional inequality in poverty depends on both the geographic level and measure of poverty being assessed. Policy makers need to be cognizant of both these factors when designing and implementing anti-poverty programs and initiatives. Doing so could help improve a number of health, social, and economic outcomes.

Data availability

The codes used for the current study are available from the corresponding author on reasonable request.

Ackerson LK, Subramanian SV (2008) State gender inequality, socioeconomic status and Intimate Partner Violence (IPV) in India: a multilevel analysis. Aust J Soc Issues 43(1):81–102. https://doi.org/10.1002/j.1839-4655.2008.tb00091.x

Article   Google Scholar  

Adler NE, Boyce T, Chesney MA, Cohen S, Folkman S, Kahn RL, Syme SL (1994) Socioeconomic status and health: the challenge of the gradient. Am Psychol 49(1):15–24. https://doi.org/10.1037/0003-066X.49.1.15

Article   CAS   Google Scholar  

Alkire S, Seth S (2015) Multidimensional poverty reduction in India between 1999 and 2006: where and how? World Dev 72:93–108. https://doi.org/10.1016/j.worlddev.2015.02.009

Anand, I., & Thampi, A. (2016). Recent trends in wealth inequality in India. Economic and Political Weekly, 59–67.

Bacolod MP, Ranjan P (2008) Why children work, attend school, or stay idle: the roles of ability and household wealth. Econ Dev Cult Change 56(4):791–828. https://doi.org/10.1086/588165

Braveman P, Cubbin C, Marchi K, Egerter S, Chavez G (2001) Measuring socioeconomic status/position in studies of racial/ethnic disparities: maternal and infant health. Public Health Rep 116(5):449–463. https://doi.org/10.1016/S0033-3549(04)50073-0

Braveman P, Gottlieb L (2014) The social determinants of health: it’s time to consider the causes of the causes. Public Health Rep 129(1_suppl2):19–31. https://doi.org/10.1177/00333549141291S206

Buddhadeb Ghosh, & Prabir De. (1998) Role of Infrastructure in Regional Development: A Study over the Plan Period. Economic and Political Weekly, 33(47/48), 3039–3048.

Caruso BA, Cooper HLF, Haardörfer R, Yount KM, Routray P, Torondel B, Clasen T (2018) The association between women’s sanitation experiences and mental health: a cross-sectional study in Rural, Odisha India. SSM—Popul Health 5:257–266. https://doi.org/10.1016/j.ssmph.2018.06.005

Cashman L, Sabates R, Alcott B (2021) Parental involvement in low-achieving children’s learning: the role of household wealth in rural India. Int J Educ Res 105:101701. https://doi.org/10.1016/j.ijer.2020.101701

Chalasani S (2012) Understanding wealth-based inequalities in child health in India: a decomposition approach. Soc Sci Med 75(12):2160–2169. https://doi.org/10.1016/j.socscimed.2012.08.012

Chancel L, Piketty T (2019) Indian income inequality, 1922–2015: from British Raj to Billionaire Raj? Rev Income Wealth 65(S1). https://doi.org/10.1111/roiw.12439

Chandra H (2021) District-level estimates of poverty incidence for the state of West Bengal in India: application of small area estimation technique combining NSSO Survey and Census Data. J Quant Econ 19(2):375–391. https://doi.org/10.1007/s40953-020-00226-8

Chaudhuri S, Gupta N (2009) Levels of living and poverty patterns: a district-wise analysis for India. Econ Political Wkly 44(9):94–110

Google Scholar  

Chauhan RK, Mohanty SK, Subramanian SV, Parida JK, Padhi B (2016) Regional estimates of poverty and inequality in India, 1993–2012. Soc Indic Res 127(3):1249–1296. https://doi.org/10.1007/s11205-015-1006-6

Corsi DJ, Neuman M, Finlay JE, Subramanian SV (2012) Demographic and health surveys: a profile. Int J Epidemiol 41(6):1602–1613. https://doi.org/10.1093/ije/dys184

Fink G, Günther I, Hill K (2011) The effect of water and sanitation on child health: evidence from the demographic and health surveys 1986–2007. Int J Epidemiol 40(5):1196–1204. https://doi.org/10.1093/ije/dyr102

Galobardes B, Lynch J, Smith GD (2007) Measuring socioeconomic position in health research. Br Med Bull 81–82(1):21–37. https://doi.org/10.1093/bmb/ldm001

GoI 2011 Census of India (2011). GoI. http://censusindia.gov.in/

Grown C, Gupta GR, Pande R (2005) Taking action to improve women’s health through gender equality and women’s empowerment. Lancet 365(9458):541–543. https://doi.org/10.1016/S0140-6736(05)17872-6

Hallegatte S, Rozenberg J (2017) Climate change through a poverty lens. Nat Clim Change 7(4):250–256. https://doi.org/10.1038/nclimate3253

Article   ADS   Google Scholar  

Himanshu. (2007) Recent Trends in Poverty and Inequality: Some Preliminary Results. Economic and Political Weekly, 42(6), 497–508.

Howe LD, Galobardes B, Matijasevich A, Gordon D, Johnston D, Onwujekwe O, Patel R, Webb EA, Lawlor DA, Hargreaves JR (2012) Measuring socio-economic position for epidemiological studies in low- and middle-income countries: a methods of measurement in epidemiology paper. Int J Epidemiol 41(3):871–886. https://doi.org/10.1093/ije/dys037

NITI Aayog (2021) India National Multidimensional Poverty Index. United Nations Development Programme.

Initiative O, Alkire S, Conceicao P, Barham A, Conconi A, Dirksen J, Carbajal F, Evans M, Hall J, Jahic A, Kanagaratnam U, Kivilo M, Kovacevic M, Kovesdi F, Mitchell C, Nogales R, oldiges C, Ortubia A, Walkey C (2019) Global Multidimensional Poverty Index 2019: illuminating Inequalities. United Nations Development Programme.

Jain A, Rodgers J, Li Z, Kim R, Subramanian S (2021) Multilevel analysis of geographic variation among correlates of child undernutrition in India. Matern Child Nutr 17(3). https://doi.org/10.1111/mcn.13197

Jain A, Subramanian SV (2018) Intrinsic and instrumental perspectives to sanitation. SSM—Popul Health 5:267–269. https://doi.org/10.1016/j.ssmph.2018.07.005

Jain A, Wang W, James KS, Sarwal R, Kim R, Subramanian SV (2022) Small area variations in dietary diversity among children in India: a multilevel analysis of 6–23-month-old children. Front Nutr 8:791509. https://doi.org/10.3389/fnut.2021.791509

Kapur Mehta A, Shah A (2003) Chronic poverty in India: incidence, causes and policies. World Dev 31(3):491–511. https://doi.org/10.1016/S0305-750X(02)00212-7

Karlsson O, Kim R, Joe W, Subramanian SV (2020) The relationship of household assets and amenities with child health outcomes: an exploratory cross-sectional study in India 2015–2016. SSM—Popul Health 10:100513. https://doi.org/10.1016/j.ssmph.2019.100513

Kawachi I, Adler NE, Dow WH (2010) Money, schooling, and health: mechanisms and causal evidence: money, schooling, and health. Ann NY Acad Sci 1186(1):56–68. https://doi.org/10.1111/j.1749-6632.2009.05340.x

Kesterton AJ, Cleland J, Sloggett A, Ronsmans C (2010) Institutional delivery in rural India: the relative importance of accessibility and economic status. BMC Pregnancy Childbirth 10(1):30. https://doi.org/10.1186/1471-2393-10-30

Kim R, Mohanty SK, Subramanian SV (2016) Multilevel geographies of poverty in India. World Dev 87:349–359. https://doi.org/10.1016/j.worlddev.2016.07.001

Kochhar R (2021) In the pandemic, India’s middle class shrinks and poverty spreads while China sees smaller changes. Pew Research Center. https://www.pewresearch.org/fact-tank/2021/03/18/in-the-pandemic-indias-middle-class-shrinks-and-poverty-spreads-while-china-sees-smaller-changes/

Lund C, Breen A, Flisher AJ, Kakuma R, Corrigall J, Joska JA, Swartz L, Patel V (2010) Poverty and common mental disorders in low and middle income countries: a systematic review. Soc Sci Med 71(3):517–528. https://doi.org/10.1016/j.socscimed.2010.04.027

McMichael AJ (1999) Prisoners of the proximate: loosening the constraints on epidemiology in an age of change. Am J Epidemiol 149(10):887–897. https://doi.org/10.1093/oxfordjournals.aje.a009732

McMichael AJ, Powles JW, Butler CD, Uauy R (2007) Food, livestock production, energy, climate change, and health. Lancet 370(9594):1253–1263. https://doi.org/10.1016/S0140-6736(07)61256-2

Mejía-Guevara I, Krishna A, Corsi DJ, Subramanian SV (2015) Individual and ecological variation in child undernutrition in India: a multilevel analysis. J South Asian Dev 10(2):168–198. https://doi.org/10.1177/0973174115588841

Menon P, Headey D, Avula R, Nguyen PH (2018) Understanding the geographical burden of stunting in India: a regression‐decomposition analysis of district‐level data from 2015–16. Matern Child Nutr 14(4):e12620. https://doi.org/10.1111/mcn.12620

Mishra AK, Bhardwaj V (2021) Wealth distribution and accounting for changes in wealth inequality: empirical evidence from India, 1991–2012. Econ Change Restruct 54(2):585–620. https://doi.org/10.1007/s10644-020-09290-9

Mohanty SK, Govil D, Chauhan RK, Kim R, Subramanian SV (2016) Estimates of poverty and inequality in the districts of India, 2011–2012. J Dev Policy Pract 1(2):142–202. https://doi.org/10.1177/2455133316642338

Mosley WH, Chen LC (1984) An analytical framework for the study of child survival in developing countries. Popul Dev Rev 10:25. https://doi.org/10.2307/2807954

Oakes JM, Kaufman JS (eds) (2006) Methods in social epidemiology, 1st edn. Jossey-Bass.

Oakes JM, Rossi PH (2003) The measurement of SES in health research: current practice and steps toward a new approach. Soc Sci Med 56(4):769–784. https://doi.org/10.1016/S0277-9536(02)00073-4

Palmer-Jones R, Sen K (2003) What has luck got to do with it? A regional analysis of poverty and agricultural growth in rural India. J Dev Stud 40(1):1–31. https://doi.org/10.1080/00220380412331293647

Palmer-Jones R, Sen K (2006) It is where you are that matters: the spatial determinants of rural poverty in India. Agric Econ 34(3):229–242. https://doi.org/10.1111/j.1574-0864.2006.00121.x

Patel V, Kleinman A (2003) Poverty and common mental disorders in developing countries. Bull World Health Organ 81(8):609–615

Porter ME, Stern S (n.d.) An Assessment of Aspirational Districts Programme. Institute for Competitiveness

Rajpal S, Kim J, Joe W, Kim R, Subramanian SV (2021) Small area variation in child undernutrition across 640 districts and 543 parliamentary constituencies in India. Sci Rep 11(1):4558. https://doi.org/10.1038/s41598-021-83992-6

Article   ADS   CAS   Google Scholar  

Rodgers J, Kim R, Subramanian SV (2019) Explaining within- vs between-population variation in child anthropometry and hemoglobin measures in India: a multilevel analysis of the National Family Health Survey 2015–2016. J Epidemiol https://doi.org/10.2188/jea.JE20190064

Romanello M, McGushin A, Di Napoli C, Drummond P, Hughes N, Jamart L, Kennard H, Lampard P, Solano Rodriguez B, Arnell N, Ayeb-Karlsson S, Belesova K, Cai W, Campbell-Lendrum D, Capstick S, Chambers J, Chu L, Ciampi L, Dalin C, Hamilton I (2021) The report of the Lancet Countdown on health and climate change: code red for a healthy future Lancet 398(10311):1619–1662. https://doi.org/10.1016/S0140-6736(21)01787-6

Roy K, Chaudhuri A (2008) Influence of socioeconomic status, wealth and financial empowerment on gender differences in health and healthcare utilization in later life: evidence from India. Soc Sci Med 66(9):1951–1962. https://doi.org/10.1016/j.socscimed.2008.01.015

Sahoo KC, Hulland KRS, Caruso BA, Swain R, Freeman MC, Panigrahi P, Dreibelbis R (2015) Sanitation-related psychosocial stress: a grounded theory study of women across the life-course in Odisha, India. Soc Sci Med 139:80–89. https://doi.org/10.1016/j.socscimed.2015.06.031

Samad HA, Zhang F (2019) Electrification and women’s empowerment: evidence from rural India. World Bank Policy Researchh Working Paper. World Bank, Washington, DC.

Singh N, Bhandari L, Chen A, Khare A (n.d.) Regional inequality in India: a fresh look. Econ Political Wkly 38(11), 1069–1073.

Standal K, Winther T (2016) Empowerment through energy? Impact of electricity on care work practices and gender relations. Forum Dev Stud 43(1):27–45. https://doi.org/10.1080/08039410.2015.1134642

Subramanian S, Jayaraj D (2013) The evolution of consumption and wealth inequality in India: a quantitative assessment. J Glob Dev 4(2). https://doi.org/10.1515/jgd-2013-0018

Subramanian SV, Belli P, Kawachi I (2002) The macroeconomic determinants of health. Annu Rev Public Health 23(1):287–302. https://doi.org/10.1146/annurev.publhealth.23.100901.140540

Victora CG, Wagstaff A, Schellenberg JA, Gwatkin D, Claeson M, Habicht J-P (2003) Applying an equity lens to child health and mortality: more of the same is not enough. Lancet 362(9379):233–241. https://doi.org/10.1016/S0140-6736(03)13917-7

Wilkinson RG, Pickett KE (2006) Income inequality and population health: a review and explanation of the evidence. Soc Sci Med 62(7):1768–1784. https://doi.org/10.1016/j.socscimed.2005.08.036

Zacharias A, Vakulabharanam V (2011) Caste stratification and wealth inequality in India. World Dev 39(10):1820–1833. https://doi.org/10.1016/j.worlddev.2011.04.026

Download references

Acknowledgements

The authors would like to thank the Demographic and Health Surveys program for making the fourth National Family Health Survey data freely accessible.

Author information

Authors and affiliations.

Global Health & Social Medicine, Harvard Medical School, Boston, MA, 02115, USA

Interdisciplinary Program in Precision Public Health, Department of Public Health Sciences, Graduate School of Korea University, Seoul, South Korea

Sunil Rajpal

Department of Economics, FLAME University, Pune, India

G B Pant Social Science Institute, Prayagraj, India

Md Juel Rana

Division of Health Policy & Management, College of Health Science, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul, 02841, South Korea

Harvard Center for Population and Development Studies, Cambridge, MA, 02138, USA

S. V. Subramanian

Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health, Boston, MA, 02115, USA

You can also search for this author in PubMed   Google Scholar

Corresponding authors

Correspondence to Rockli Kim or S. V. Subramanian .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Ethical approval

This article does not contain any studies with human participants performed by any of the authors.

Informed consent

Additional information.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary file, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Jain, A., Rajpal, S., Rana, M.J. et al. Small area variations in four measures of poverty among Indian households: Econometric analysis of National Family Health Survey 2019–2021. Humanit Soc Sci Commun 10 , 18 (2023). https://doi.org/10.1057/s41599-023-01509-0

Download citation

Received : 16 August 2022

Accepted : 04 January 2023

Published : 17 January 2023

DOI : https://doi.org/10.1057/s41599-023-01509-0

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Regional variations in urban poverty in india: pattern and determinants.

• Amrita Shergill

GeoJournal (2023)

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Wiley - PMC COVID-19 Collection

Poverty in India in the face of Covid‐19: Diagnosis and prospects *

Hai‐anh dang.

1 Development Data Group, the World Bank, Washington District of Columbia, USA

Peter Lanjouw

2 Department of Economics, Vrije Universiteit, Amsterdam the Netherlands

Elise Vrijburg

India has been hard‐hit by the Covid‐19 pandemic. The virus has exacted a heavy toll in terms of lives lost and deteriorating health outcomes. The economic consequences of the pandemic have been similarly grim. In this paper we attempt an initial, interim, assessment of the impacts of the crisis on poverty. We review the growing literature that considers emerging poverty impacts, noting that there remain significant knowledge gaps due to limited evidence on current welfare outcomes. We analyze pre‐Covid survey data to examine the incidence of chronic poverty and downward mobility during a period of rapid economic growth and declining poverty. A profile of poverty during such a period might offer a plausible, partial, window on population groups currently at risk. We suggest that, notwithstanding the severe initial impacts of the crisis on poverty, there are grounds for expecting further consequences going forward. As the virus has spread out of the relatively affluent cities, and as economic stagnation persists, rural areas, with historically higher rates of chronic poverty and vulnerability, may see particularly sharp increases in poverty. While recent vaccination developments offer some grounds for optimism, there remains an urgent need to identify, implement and amplify effective policy alleviation measures.

1. INTRODUCTION

The year 2020 marks a major break in the progress achieved throughout the world in reducing global poverty. Sustainable Development Goal 1, announced by the international development community in 2015, had proclaimed the ambitious objective of ending extreme global poverty by 2030. Success in meeting the Millennium Development Goal of halving global poverty during the 1990–2015 interval had undoubtedly emboldened the architects of the Sustainable Development Goals to pursue a poverty objective that, while clearly aspirational, did not appear to be entirely beyond reach. The onset of the Covid‐19 pandemic in early 2020 has dampened these ambitions. Nowhere is this more evident than in India, a country where poverty reduction during the first decades of the 2000s had been remarkable, but where the impact of the Covid pandemic has been staggering in terms of both public health and economic livelihoods.

Given that the Covid‐19 pandemic remains in full swing, it is impossible to arrive at a complete assessment of its full impact on poverty in India and how it may evolve going forward. The economic consequences of the crisis are still working their way through the Indian economy, and policy measures aimed at addressing both the public health and economic fallout from the crisis continue to be formulated and rolled out. Empirical evidence on the actual impact of the crisis also remains highly fragmented and incomplete.

Yet, to preserve as much as possible of the progress that India undeniably achieved in recent decades, policies that mitigate the most severe consequences of the crisis must be introduced. To this end, there is a need to anticipate the likely consequences of the crisis for poverty. Not only is it necessary to track how poverty is evolving in the face of the crisis, but it is important to also identify and understand the circumstances of those who face a heightened risk of falling into poverty as the crisis continues to unfold.

In this study we attempt to provide an assessment of how the fight to end extreme poverty in India has been disrupted by the arrival of the Covid‐19 crisis. In particular, we are interested in informing policy‐making by providing insights into possible population groups that are likely to have been most seriously affected by the crisis, or to be particularly exposed to its most pernicious consequences in the months and years to come.

We start in the next section with a brief review of progress in poverty reduction prior to the onset of the Covid‐19 crisis. We are severely handicapped in this regard in that nationally representative poverty data that underpin official estimates of poverty in India are not available beyond the year 2011/12, when the last publicly available National Sample Survey was published. Nonetheless, we document that, in the period between 2004/05 and 2011/12, poverty reduction in India was indeed significant. It seems likely that a trajectory of declining poverty persisted until the Covid‐19 crisis hit in 2020.

We then proceed in Section  3 to briefly describe the spread and dimensions of the Covid‐19 crisis in India and provide a timeline of the policy measures that were introduced in response to the spread of the virus. Section  4 summarizes findings from a desk review of reports on the poverty impacts of the crisis. Absent systematic and current statistical evidence, we rely on a variety of sources and often anecdotal evidence to arrive at an initial impression of the poverty consequences of the crisis and its economic manifestations.

We then turn in Section  5 to a statistical analysis of household survey data from the first decade of the 2000s to investigate patterns of poverty dynamics that occurred during that time. We employ methods to construct synthetic panels from cross‐sectional survey data to inquire into the characteristics of the chronically poor in India during this period of rapid economic growth and poverty reduction. It is likely that those who remained poor during that interval are among those who are also at greatest risk of destitution during the sharp economic downturn ushered in by the Covid‐19 crisis. We also scrutinize those who, during the period 2004/05–2011/12, faced a heightened risk of falling into poverty. Again, it seems reasonable to suppose that those population groups that are most vulnerable during such a period of rapid progress are likely to count among the most vulnerable during a period of tremendous economic stress. In Section  6 we offer some concluding remarks.

2. PRE‐COVID POVERTY TRENDS

The evolution of poverty in India has long been the subject of close attention. In particular, with the introduction of economic reforms in the early 1990s, there has been a strong interest in seeing how poverty outcomes fared as economic growth accelerated (Datt & Ravallion,  2011 ; Ravallion,  2011 ). Figure  1 illustrates that poverty fell sharply during this period, falling from nearly 50 per cent of the population in 1987/88 to just over 20 per cent in 2011/12 (Dang & Lanjouw, 2018 ). Although growth started to pick up in the mid‐1990s, the evidence suggests that poverty reduction only started to gather pace in the early 2000s, after the 2004/05 round of the National Sample Survey (NSS) and, in particular, between the 2009/10 and 2011/12 rounds. Interestingly, the dramatic falls in poverty between 2009/10 and 2011/12 occurred when per capita growth rates were in fact lower than during the 2004/05–2009/10 interval (Figure  1 ).

Trends in poverty and gross domestic product per capita for India, 1987/88–2011/12. Source: Dang and Lanjouw ( 2018 )

A question of considerable importance concerns how poverty evolved from 2011/12 up to the onset of the Covid‐19 crisis. Unfortunately, the data needed to answer such questions are not available. The subsequent “quinquennial” round of the NSS survey after 2011/12, fielded during the 2017/18 survey year, has not been released by the National Sample Survey Organization (NSSO). There is considerable debate as to the likely evolution of poverty, with many commentators suggesting that the pace of poverty reduction slowed post‐2011/12, alongside the lower per capita income growth rates for India as a whole. What seems unambiguous is that with the onset of the Covid‐19 crisis, the decline in poverty is likely to have halted altogether.

3. THE COVID‐19 PANDEMIC

The Covid‐19 pandemic, to which India became exposed at the beginning of 2020, has left a deep mark on the country. In absolute terms, the spread of the virus has been striking – particularly during the surge in the early months of 2021. With over 27.1 million reported cases (as of May 26, 2021), India lagged behind only the United States in terms of the absolute number of Covid‐19 infected persons (John Hopkins University Center for Systems Science & Engineering,  2021 ). Compared to its vast population, however, India's Covid‐19 outbreak seems less catastrophic. The country ranked 94th out of all 199 investigated countries on the number of cumulative confirmed Covid‐19 cases per million people as of May 25, 2021 (Our World in Data,  2021a ). However, the rapid spread of Covid‐19 during the spring of 2021 was particularly alarming (Figure  2 ). In late May, India topped the World Health Organization's (WHO) worldwide list of newly reported Covid‐19 cases in the last 24 hr with an astounding 222,315 cases on May 24 (WHO, 2021 ). Worryingly, these data may well have been underestimated. 1

Time‐map of daily confirmed Covid‐19 cases in India as of May 25, 2021. Source: Our World in Data ( 2021b )

Beyond the health crisis, India has also faced severe socioeconomic challenges. As noted earlier, the Indian economy was already slowing down before the recent pandemic. The two nationwide lockdowns introduced shortly after the onset of the pandemic, which have been rated as among the most stringent in the world, 2 dealt the Indian economy an additional blow. The first lockdown, starting on March 25, 2020, and lasting for 21 days, involved a total closure of almost every type of commercial entity (except for grocery outlets) (Ministry of Home Affairs, 2020a ; Singh et al.,  2020 ). The second one comprised an extension of the first lockdown – initially for two weeks until May 3, followed by another 2 weeks, and finally extending until May 31 (Ministry of Home Affairs, 2020b , 2020c , 2020d ). During the second lockdown, districts were divided into three possible zones – red, yellow and green, depending on the spread of Covid‐19 in that area – that determined the severity of measures in those places (Ray & Subramanian,  2020 , p. 3). Within the red zones, the highest‐risk areas were furthermore classified as containment zones (Express News Service,  2020 ). Although states and Union territories carried the responsibility to demarcate containment zones and to declare appropriate measures, the measures were expected to involve at least a night curfew from 7 p.m. to 7 a.m. imposed by the national government (Home Secretary,  2020 ). Within the containment zones, the lockdown was further extended to June 30 and finally to July 31 (Ministry of Home Affairs, 2020e ).

While the lockdown was extended in the containment zones, the rest of India was slowly reopened in different unlock phases (Gangwar & Ray,  2021 , p. 435). More recently, many states reverted to lockdowns on their own as a response to the new Covid surge in early 2021.

The disruption to economic activity in India throughout the lockdown period has been investigated by Beyer et al. ( 2021 ) who scrutinize daily electricity consumption and nighttime light intensity for evidence of economic turmoil. These two indicators capture important inputs into a wide range of economic activities and data are available at short notice and at high resolution (Beyer et al.,  2021 , pp. 4–5). Their model, which can explain 90 per cent of the variance in electricity consumption and light intensity in normal times, reveals that on the day the first lockdown was imposed (March 23, 2020), electricity levels were 21 per cent lower than their predicted normal value. It was not until the end of June, after the lockdown measures had been lifted, that deviations from the predicted normal electricity levels were no longer statistically significant (Beyer et al.,  2021 , pp. 5–7). However, in an extended analysis for The Economist , Beyer and colleagues show that India's rapidly rising infections have reversed the economic recovery from June (Figure  3 ). Clearly, the second wave placed the Indian economy under further strain.

Deviations from predicted daily normal electricity consumption in India (%). Source: The Economist ( 2021 )

4. POVERTY IMPACTS: A DESK REVIEW

The socioeconomic consequences of the Covid‐19 crisis and the measures subsequently taken to address them are unlikely to be distributed uniformly among Indian citizens. It is thus useful to identify those population groups that are most exposed to the economic burden of the current crisis in India and to elaborate on the specific vulnerabilities they face. We begin by assembling insights gleaned from a review of research and media reports that have emerged since the onset of the pandemic.

One of the most widely observed vulnerable groups in India is informal workers. This segment of the labor force is vast. A report by the National Commission for Enterprises in the Unorganised Sector (NCEUS) shows that, based on sectoral statistics from 2004–2005, the informal sector contributed up to 50 per cent to India's national gross domestic product (NCEUS, 2009 , p. 23). What makes informal workers economically vulnerable is that they are excluded from social security benefits, thereby being left unprotected against sudden shocks such as illness or death (Unni & Rani,  2003 , p. 130). Moreover, informal employment often generates lower incomes than similar formal employment does (Al Dahdah et al.,  2020 ). Based on data from India's national 2011–2012 Employment and Unemployment Survey studying over 450,000 individuals across India, Narayanan ( 2015 ) estimates the formal–informal wage gap at different quantiles of the wage distribution and finds that this gap is significant for both Indian males and females across each quantile. He suggests that informal workers are consistently punished for being informal workers, even when they have similar characteristics and skills to their formal worker counterparts. Lastly, informal workers often possess individual characteristics that draw them to this type of work and create other vulnerabilities simultaneously – such as being female or belonging to a lower caste (Unni & Rani,  2003 ). We touch further upon these cross‐dimensional characteristics below.

The Covid‐19 crisis has accentuated the vulnerability of informal workers in several ways. First and foremost, informal workers saw an immediate decline in their earnings as soon as the lockdown put their jobs on hold (Al Dahdah et al.,  2020 ). Based on data from the 2017–2018 Periodic Labour Force Survey, Estupinan and Sharma ( 2020 , pp. 14–17) estimate that 89.5 per cent of those at risk of losing their job during the first national lockdown and 68 per cent of those at risk during the second national lockdown were informally employed. Moreover, they estimate that the average wage loss for informal employees throughout the entire lockdown period in 2020 was 22.62 per cent, compared to a relatively low 3.83 per cent for formal employees.

Although informal employment could technically be more easily restored after a lockdown than formal employment, the numbers disappoint. Using a combination of personal surveys, secondary reports from organizations engaged with informal employees and comprehensive fieldwork, a case study by Azim Premji University ( 2021a ) conducted in Bangalore showed that only 6 per cent of the informal employees interviewed had returned to their previous job with similar or higher wages, and 15 per cent had not returned to any sort of work by December 2020. Finally, one could expect a sudden loss of income to have a bigger impact on informal than on formal employees given their generally lower‐income status (Centre for Monitoring Indian Economy,  2020 ).

Beyond the loss of income, informal employees are vulnerable due to a lack of access to health care. With one of the largest private and unregulated health‐care insurance markets, only 37 per cent of the Indian population had any form of health insurance in the fiscal year 2017/18 (Chatterjee,  2016 ; Tikkanen et al.,  2020 ). Although two‐thirds of the insured rely on public health insurance, informal employees might especially miss out on such chances (Tikkanen et al.,  2020 ). The largest existing health insurance schemes in place – the Employees’ State Insurance Scheme and the Central Government Health Scheme – rely upon employment in the formal sector, and smaller initiatives meant to target informal employees have not quite lifted off yet due to their limited focus on surgical procedures and families below the poverty line (La Forgia & Nagpal,  2012 ). Informal employees thus risk facing high out‐of‐pocket payments when they fall ill with the coronavirus.

One particularly vulnerable subgroup of informal workers are migrant workers, who often reside in rural areas but travel intra‐ or inter‐state to urban areas for work (Rawal et al., 2020 ). In a paper by Srivastava ( 2020 , pp. 29–30), the total number of vulnerable migrant workers is estimated using data from the NSSO, the National Statistical Office and the Census of India. Defining vulnerable migrant workers as those workers who either migrate for a short duration (i.e. seasonal migrants) or longer‐term migrants from the four lowest consumption quantiles and five lowest occupational categories, 3 the paper estimates that there are about 128 million such migrant workers in India (Srivastava,  2020 , pp. 8–13).

It should be noted that even compared to other informal workers, migrant workers face a particularly high chance of losing employment. In the four‐hour interval following the announcement of the first lockdown in March 2020, most of these migrant workers were suddenly left without a job (Sengupta & Jha,  2020 , pp. 158–162). Systematic data on the effects of the two nationwide lockdowns on migrants and other vulnerable communities are scarce, but a purposive sampling study from 5,000 telephone interviews conducted in 12 Indian states by Kesar et al. ( 2020 ) reveals that 86.6 per cent of inter‐state migrant workers lost employment due to these lockdowns in 2020. Regression analysis further shows that migrant workers were significantly more likely to be laid off than similar non‐migrant workers – rural migrant workers were 73.9 per cent more likely and urban workers 50.7 per cent more likely to lose their job than their non‐migrant counterparts (Kesar et al.,  2020 , pp. 19–22).

Presumably, these job losses were only partially recovered during the unlock phase. A survey of 372 migrant workers in the automobile, construction and garment sectors revealed that, in October 2020, 60 per cent of the migrant workers were still fully out of work (Seth,  2020 , p. 1). A survey conducted by Azim Premji University among 2,778 individuals across 13 states revealed that around 20 per cent of the informal workers interviewed who lost their job during the first two nationwide lockdowns were still out of work in October and November (Azim Premji University,  2021b , p. 17, n.d. ).

Alongside a heightened risk of losing their job, migrant workers are especially vulnerable to getting infected with Covid‐19. It is generally known that the housing conditions for migrant workers in their area of employment are often rudimentary and cramped, making it difficult for them to self‐isolate (Sengupta & Jha,  2020 , pp. 158–162). A cross‐sectional study by Babu et al. ( 2017 , pp. 336–338), using a sample of almost 50,000 internal migrant households from 13 different Indian cities, revealed that the majority of those (43.4 per cent) lived in non‐registered slums, followed by registered slums (32.7 per cent) and dwellings at the work or construction site (11.7 per cent). The majority of these workers (70.6 per cent) further resided in single‐room shelters, had no access to a private water tap (44.6 per cent) or their own private toilet (63.8 per cent). Under such conditions, the chances of getting infected with Covid‐19, paired with facing high out‐of‐pocket payments, are substantial.

Confronted with income shortfalls and the risk of getting infected with the coronavirus, national and local lockdowns compel migrant workers to go back to their rural homes, sometimes under hazardous conditions (Sengupta & Jha,  2020 , pp. 158–162). The first nationwide lockdowns (from March 25 to May 31) caused a flood of migrants attempting to return home. Estimates by the economist and demographer Amitabh Kundu and his colleagues put the number of internal migrants who returned home at around 12 million (Chishti,  2020 ). The issue was prominent enough for the national government to help individual states transport about 9.9 million migrants by means of designated buses and trains between May and June (Iyer,  2020 ). Unfortunately, the most vulnerable immigrant workers were unable to access these transportation opportunities as they were stranded in peripheral locations and on work sites, leaving them no alternative but to walk home (Srivastava,  2020 , p. 18). A database of media‐reported deaths in India up to July 3 showed that, out of the 961 deaths that resulted from the lockdowns (deaths resulting from Covid‐19 infection not included), 209 were caused by walking long distances or during migration (Thejesh GN, 2020 ). Despite a large number of these migrants returning to their work sites in the unlock phase, news reports fear that the new wave Covid‐19 and subsequent local lockdowns will cause another migrant push to rural India (Agarwal & Bellman,  2021 ; Gulati et al., 2021 , p. 24; Naik,  2021 ; The Indian Express,  2021 ).

When returning home, migrants may find themselves in an even more vulnerable position due to the absence of social protection measures and health facilities in rural areas. First of all, media reports show that in rural areas there is often a lack of awareness of appropriate measures against Covid‐19 or of what it means to quarantine. Furthermore, local efforts to inform rural residents about these matters are inadequate (Agrawal,  2020 ). Even when returning migrants and villagers are well informed, the quarantine arrangements in rural areas frequently fall short: they either involve being placed in poorly equipped quarantine centers or going into self‐quarantine, for which many migrants lack the means (Sengupta & Jha,  2020 ).

Second, numbers show that rural areas severely lack health‐care facilities compared to urban areas. Based on a nationwide survey conducted with 14,746 households from 12 different states, a report by the IMS Institute from 2013 revealed that inhabitants in rural areas all across India have to travel disproportionately further to access a health‐care facility: 63 per cent of the rural population have to travel over 5 kilometers to access a hospitalization unit, compared to 26 per cent of the urban population (IMS Institute for Healthcare Informatics,  2013 , p. 17). Moreover, the number of hospital beds in rural areas may be critically low. With data from the Directorate General of State Health Services, a study by Ghosh and Dinda ( 2017 , p. 110) reveals that the states whose rural areas have the fewest hospital beds per hundred thousand inhabitants are Bihar (5.685), Chhattisgarh (7.762), Uttar Pradesh (9.947), Haryana, (14.864), Madhya Pradesh (19.065) and Maharashtra (18.360). Three of these six states are among the top states home to inter‐state migrants. Professor Kundu from the Research and Information System estimated that, based on the 2011 Census of India, Uttar Pradesh would rank first (accounting for 25 per cent of the inter‐state migrants), Bihar second (accounting for 14 per cent) and Madhya Pradesh fourth (accounting for 5 per cent) (Singh & Magazine,  2020 ).

For rural areas, the threat of bad containment measures and inadequate health‐care provisions is severe. Data from the Backward Regions Grant Fund, dedicated to the development of backward (rural) districts, shows that the contribution of the 243 backward regions for which data were available to the number of infections has risen from 11.2 per cent during the first Covid‐19 wave to 16 per cent during the new second wave. Moreover, in absolute terms, the death toll from these regions has quadrupled compared to that of the first wave (Sinha,  2021 ; Ministry of Panchayati Raj, n.d. ). Though the new wave has initially spread most rapidly in urban areas, the consequences of the second wave may be hardest felt in the less well‐equipped and impoverished rural regions (Mehta & Jamkhandikar,  2021 ; Mitra et al.,  2021 ).

Aside from serious health risks, another issue that awaits migrants when returning home is further income losses. In rural areas, poor households are typically unable to generate sufficient income from farming. With the nationwide lockdown falling exactly within the winter harvest season in 2020, rural families faced a shortage of labor and equipment and saw their crops being left unsold (Maggo,  2020 ). Based on key informant interviews among 1,515 farmers in two contrasting Indian states (Haryana and Odisha), Ceballos et al. ( 2020 , pp. 1–3) found that, in both states, the majority (61 per cent and 74 per cent, respectively) could not sell their crops immediately upon harvest. Thus, migrant workers and their rural‐based families have experienced a double‐dip in income from both a lack of wages and salaries and a lack of produce sales.

Such a double‐dip in income might lead to concerning levels of indebtedness. In a mixed‐methods‐based study including both data from the Networks, Employment, Debt, Mobility and Skills in India Survey and individual in‐depth surveys from 2016 to 2017 distributed among 2,692 individuals across 15 rural villages in Tamil Nadu, Guérin et al. ( 2020 , p. 11) shows that 99 per cent of the households were indebted, with a median of four outstanding loans per family and an average value of Rs. 58,000 (almost US$800) per loan. Informal interviews from the same study revealed that following the first‐wave lockdown, families were pressured by lenders to pay back their outstanding debts and faced difficulties finding new sources of credit (Guérin et al.,  2020 , pp. 17–20).

4.1. Cross‐cutting dimensions

Certain dimensions, such as gender, religious status or caste, will affect one's vulnerability in the current crisis disproportionately. These cross‐cutting dimensions often coexist with one's (informal) employment situation to create an even more vulnerable profile (Ray & Subramanian,  2020 , pp. 46–47). We will touch upon three of these dimensions: gender, religion and caste.

First, women are likely to carry a higher burden during the current crisis than men. Despite only 20 per cent of all Indian women being employed, among those women who are employed, 90 per cent are involved in informal employment, thus facing the associated vulnerabilities (World Economic Forum,  2020 , p. 28; International Labour Organization,  2018 , p. 88). The current crisis is likely to disproportionately affect their employment status. Based on the Centre for Monitoring Indian Economy's Consumer Pyramids Household Survey, drawing on data from 174,405 households nationwide, Ashwini ( 2020 , pp. 3–4) found that women employed pre‐lockdown were approximately 20 per cent less likely to be employed post‐lockdown than pre‐lockdown employed men. Estimating the effect of being female in rural areas on the risk of employment loss, another study finds that being a rural‐based female increases one's chance of losing employment during the lockdown period by 75.7 per cent compared to being a rural‐based male (Kesar et al.,  2020 , p. 22).

The vulnerability of women extends beyond a disproportionate loss of income. The literature suggests that, due to their lower social status, women in India are likely to suffer disproportionately from food shortages and price hikes, thus having higher chances than men of going hungry (Asadullah & Raghunathan,  2020 ; Global Hunger Index,  2010 , p. 14). Moreover, women often suffer from additional health risks (and accompanying high out‐of‐pocket payments). Research suggests that Indian households are more likely to prioritize, and devote more resources to, the health‐care needs of males than of females (Barcellos et al.,  2014 ; Oster,  2006 ). Female‐headed households are also more likely to suffer economically from a health shock. A study by Dhanaraj ( 2014 , p. 15), which employs a logit model on a panel data set covering more than 3,000 households in the Indian state of Andhra Pradesh over 15 years, estimates that female‐headed households faced an 80 per cent higher risk of experiencing such a welfare loss relative to male‐headed households.

Muslims form another vulnerable group. First, they are more likely to be employed in vulnerable segments of the informal sector than other religious groups. According to the 2014 Post Sachar Evaluation Committee (PSEC) 4 report, a mere 23 per cent of the Muslim urban households earn their livelihoods through regular wage employment compared to 42 per cent of all urban households. Furthermore, their incidence of relying on non‐agricultural self‐employment is 25 per cent, compared to 14 per cent for Hindu households (PSEC, 2014 , p. 14). Moreover, Muslims often lack access to social resources available to other vulnerable communities and suffer from a lower distribution of public welfare services and benefits, making it more difficult to cope with a loss of earnings (Pandya,  2010 , pp. 16, 29). This could explain why, according to the Dhanaraj study ( 2014 , p. 15), Muslim households have an estimated 26.5 per cent higher chance of welfare loss due to serious illness or death than other religious groups. They are also more likely to sell assets or borrow money to cope with such a loss, which in the long run is believed to increase their economic vulnerability even further (Dhanaraj,  2014 , pp. 6, 17). Thus, the current Covid‐19 crisis might deteriorate their already low economic status.

Third, Scheduled Castes (SCs, more commonly known as Dalits) are a particularly vulnerable constituency. Households belonging to this group are often reliant on low‐paid informal jobs that require migrating or commuting to urban areas, thus exposing them to the vulnerabilities of informal employment (Ganguly,  2020 ). Estimates provided by the NSSO, based on surveys covering more than 2.5 million households in 2011/12, indicate that SCs account for the bulk of households dependent on income from casual labor (Ministry of Statistics & Programme Implementation,  2015 , pp. 21–25). In addition, SC members had limited access to public benefits even before the crisis, and are likely to be in even more desperate need of them now (Ganguly,  2020 ). The study by Kesar et al. ( 2020 , p. 34) revealed that 54 per cent of all SC members in the sample did not receive any type of cash transfer during the first nationwide lockdown, which is considerably higher than the percentage for the other socially disadvantaged castes, namely the Scheduled Tribes (46 per cent) and the Other Backward Classes (44 per cent). Unsurprisingly, then, SC members were more susceptible to going hungry than other groups. Using a telephone survey among 164 rural households across 13 states, Niyati and Vijayamba ( 2020 ) reveal that the proportion of households with less than usual food consumption in September 2020 was higher among SC members (56 per cent) than among households belonging to other castes (42 per cent).

5. POVERTY IMPACTS: INSIGHTS FROM HISTORICAL ANALYSIS OF POVERTY DYNAMICS

Documenting the poverty consequences of the Covid‐19 crisis with formal statistical analysis is difficult due to the non‐availability of the necessary household survey data. As noted above, some efforts have been made to produce interim results based on phone interviews, data on electricity consumption and via the recording of economic activity through night‐lights data (Beyer et al., 2020 ; Kesar et al.,  2020 ). However, insights are still very limited and piecemeal.

One additional potential direction is to analyze pre‐Covid data to identify population groups in India that face a heightened likelihood of being chronically poor (i.e. long‐term poor) or that face a heightened risk of falling into poverty even if currently non‐poor. The working hypothesis here, is that vulnerability and chronic poverty observed during a period of rapidly rising living standards may point to population groups that are deprived in a fundamental sense, and are thus particularly likely to be hard hit when general economic conditions deteriorate.

Studying such poverty dynamics at the all‐India level is difficult, however, as the data underpinning official poverty estimates in India come from cross‐section rather than panel surveys. Dang and Lanjouw ( 2018 ) attempt to overcome this constraint by applying a “synthetic panel” method to NSSO data, and study welfare dynamics based on estimates derived from this approach. Their approach is outlined in the companion paper by Garcés‐Urzainqui et al. ( 2021 ) included in this symposium issue. We build on their analysis to assemble potential insights into the population groups particularly exposed to the impacts of the Covid‐19 crisis.

As described in Garcés‐Urzainqui et al. ( 2021 ), a key parameter needed to produce point estimates of poverty transitions on the basis of the synthetic panel approach is the correlation over time of the error terms from consumption models estimated with the two cross‐sectional rounds of data. The procedure for estimating this correlation from cross‐section data is not uncontroversial, with some observers casting doubt on the stability and reliability of the estimate that results from it (Elbers,  2021 ; Herault & Jenkins, 2019 ). Fortunately, in an earlier exercise employing these methods, Dang and Lanjouw ( 2018 ) were able to validate NSS‐based estimates of poverty dynamics for India employed in the present study by comparing them with those obtained from the India Human Development Survey (IHDS), an actual panel data set also covering the 2004/05–2011/12 period. They indicate that the error term correlation that they approximate using only cross‐sectional NSS data is not far from that observed in the IHDS panel data. While the IHDS could directly support analysis of poverty dynamics, the overall sample size is smaller than that of the NSS, and the latter also underpin the official poverty estimates in India. For this reason it was preferred, in the present study, to apply the synthetic panel method to the NSS data rather than to base the analysis on the IHDS survey. Dang and Lanjouw ( 2018 ) show that at the national‐level estimates of poverty mobility in India based on the IHDS and NSS line up very closely.

As indicated in Garcés‐Urzainqui et al. ( 2021 ), poverty dynamics can be explored by dividing the population into two groups: one poor and the other non‐poor. But the analysis can be expanded further by disaggregating the non‐poor group into two groups: the “vulnerable” (those that are non‐poor but still face a significant risk of falling into poverty) and the “secure” (or “middle class”). These two groups can be distinguished on the basis of a “vulnerability line” that lies above the poverty line and that separates the two non‐poor groups from one another. A common, but rather ad‐hoc, approach is to arbitrarily scale up the poverty line by a certain factor to obtain such a vulnerability line. For example, in India, vulnerability has in the past been proposed to occur within a fixed income range between 1.25 times and twice the national poverty line in India (NCEUS, 2007 ). This approach has the advantage of being simple and easily communicated, but it appears to be based on no underlying rationale.

A recent approach proposed in Dang and Lanjouw ( 2017 ) instead derives the vulnerability line from a specified probability of the non‐poor falling back into poverty. This approach estimates a vulnerability line that lies above the poverty line and below which the non‐poor population faces an average risk of falling back into poverty equal to some predetermined level (designated by Dang and Lanjouw as the vulnerability index). The vulnerability index itself has to be specified upfront and can be based on various criteria, including budgetary planning, social welfare objectives, or relative concepts of well‐being. In contrast to Pritchett et al. ( 2000 ) and Chaudhuri ( 2003 ), this approach to estimating vulnerability considers as “vulnerable” a segment of the population that is currently non‐poor, treating this population segment as distinct from the currently poor (even though the poor are certainly also likely to be vulnerable in a deeper sense).

Dang and Lanjouw ( 2018 ) analyze the dynamics of poverty and vulnerability in India based on the synthetic panels procedure outlined above. Their analysis covers multiple rounds of data between 1987 and 2012. We report here their findings for the period 2004/05–2011/12 (Table  1 ). Dang and Lanjouw ( 2018 ) estimate a vulnerability line that is based on a vulnerability index of 20 per cent. In other words, they derive a vulnerability line such that the risk, on average, of those who are located between the poverty and the vulnerability line is 20 per cent. This analysis yields a vulnerability line based on the 2004/05–2011/12 synthetic panel interval equal to Rs. 770 in 2004/05 prices and can be compared to the rural India poverty line in 2004/05 of Rs. 447 per person per month.

Welfare transition dynamics based on synthetic panel data, India, 2004/05–2011/12 (percent)

The vulnerability line is that which corresponds to a vulnerability index of 0.2 in 2004/05–2011/12 (i.e. Rs. 770). All numbers are in 2004 prices for all rural India. The rural India poverty line is Rs. 446.68 for 2004/05. All numbers are estimated with synthetic panel data and weighted with population weights, where the first survey round in each period is used as the base year. Bootstrap standard errors in parentheses are estimated with 1,000 bootstraps, adjusting for the complex survey design. Household head's age range is restricted to between 25 and 55 for the first survey and adjusted accordingly for the second survey in each period. Estimation sample sizes are 91,751 and 75,159 for the first and second periods, respectively.

Table  1 points to a fair amount of consumption mobility between 2004/05 and 2011/12. However, transitioning out of poverty and directly into the “secure” category is a very rare occurrence; most “escapes” from poverty landed the poor into the category of the vulnerable. Between 2004/05 and 2011/12, just under 18 per cent of the population was estimated to be chronically poor (in the sense of being poor in both periods). Of the 37 per cent of the population estimated to be poor in 2004/05, therefore, roughly half (48.8%) was unable to escape from poverty. At the same time, while just over 40 per cent of the population could be considered vulnerable in 2004/05, an estimated 6.4 per cent had dropped back into poverty by 2011/12 and another 19.4 per cent continued to face a heightened risk of falling back into poverty. Given that the economy of India was growing particularly strongly during this interval, and that overall poverty fell markedly (from an estimated 37 per cent to 25 per cent for this subset of the population, comprised of households with household heads aged between 25 and 55), it seems reasonable to suppose that the chronically poor and vulnerable possess characteristics and attributes that would make them particularly likely to experience reversals during a period of generalized contraction. It is also plausible that many of these characteristics and attributes would persist over time. If so, acquiring a better sense of the characteristics of the chronically poor and the vulnerable between 2004/05 and 2011/12 might point to the population groups also most at risk during the current Covid‐related economic downturn. Under that hypothesis we undertake such a profiling exercise below.

5.1. Who are the chronically poor and the downwardly mobile?

Our interest here is in studying chronic poverty and vulnerability during a time of rapid economic growth and deriving from that exercise some insights into how the current Covid‐19 pandemic might expose particular population groups to a heightened risk of poverty. In order to base our estimates on the actual experience of households during the 2004/05–2011/12 period, we present our findings in terms of the odds of belonging to the chronically poor and the odds of actually being downwardly mobile between 2004/05 and 2011/12. 5 In doing the latter we depart slightly from our concept of vulnerability, in that we look at the characteristics of those households who actually dropped into the status of poor in 2011/12 from the status of vulnerable or secure in 2004/05, combined with those households who fell into the status of vulnerable in 2011/12 from the status of secure in 2004/05. From Table  1 we can see that the average odds of downward mobility were thus 19.8 per cent (12.6/63.5 = 0.198). while the average odds of chronic poverty were 48.8 per cent (17.8/36.5 = 0.488).

We consider first in Table  2 the association between education and the odds of belonging to the category of either the chronically poor or the downwardly mobile. Table  2 indicates that the education level of the household head is closely associated with both chronic poverty and downward mobility. Households in which the household head is uneducated, or has less than primary school completion, are more likely than average to comprise the chronically poor or to be downwardly mobile. Chronic poverty becomes less likely once the household head has primary schooling or higher. And as education levels rise, the likelihood of chronic poverty diminishes further. The odds of downward mobility, on the other hand, only diminish appreciably (falling below 1) once the household head has completed secondary schooling or higher. Even a little education thus seems to help protect against chronic poverty, but more than a minimum is needed to guard against a heightened risk of downward mobility. From this, one might infer that protection measures during the Covid‐19 crisis should not fail to target those with moderate education levels as well. This is especially so since it is likely that overall education levels are likely to have increased since 2011/12.

Profile of the chronically poor or downwardly mobile, 2004/05–2011/12: Education

Estimates show the difference between the probability of falling into each category relative to the mean chronic poverty and vulnerability rates of 48.8% and 19.8%.

Table  3 examines patterns of employment of heads of households and the odds of being chronically poor or downwardly mobile. Considering first the chronically poor, we see that on the whole rural workers have a higher than average odds of belonging to the chronically poor, while urban workers are less exposed. Among rural workers, only those employed in the “others” (non‐labor) category have a lower than average odds of being chronically poor. Agricultural laborers in rural areas are particularly strongly linked to chronic poverty, while urban wage workers are least exposed. As noted above, the economic crisis induced by Covid‐19 was particularly strongly felt, at least initially, by informal sector workers in urban areas. To the extent that the patterns prevailing in 2004/05–2011/12 continued to hold, it would appear that the poorest of the poor at the all‐India level were not those who were initially hit the hardest by the crisis.

Profile of the chronically poor or downwardly mobile, 2004/05–2011/12: Employment

See explanatory notes to Table  2 .

Looking at patterns of downward mobility in Table  3 , we see that households headed by agricultural laborers were also particularly likely to experience downward mobility between 2004/05 and 2011/12, a time when rural incomes and employment were rising sharply. It is difficult to imagine that this vulnerability of agricultural laborers would not persist when the Indian economy experienced a sharp reversal as a result of Covid‐19. In urban areas, urban wage workers had the lowest odds of downward mobility, consistent with the booming, urban‐led, growth trajectory under way in India between 2004/05 and 2011/12.

Table  4 considers the likelihood of chronic poverty and downward mobility among different social groups between 2004/05 and 2011/12. Given that the Indian economy was growing strongly during this period, it is not surprising that Scheduled Tribes – the social group least likely to be full participants in the modernizing Indian economy – experienced the highest odds of chronic poverty. Relative to the average, the odds of chronic poverty were more than 25 per cent higher for this group. Compared to the Scheduled Tribes, Scheduled Castes face lower odds of chronic poverty, but are still significantly more likely than average to experience such poverty. This ranking of deprivation between Scheduled Tribes and Scheduled Castes also carries over to downward mobility. Scheduled Tribes were particularly likely to experience downward mobility between 2004/05 and 2011/12, followed by the Scheduled Castes. The other caste groups in the Indian population (Other Backward Classes and other castes) faced lower than average odds of chronic poverty and downward mobility.

Profile of the chronically poor or downwardly mobile, 2004/05–2011/12: Caste

Table  5 indicates that Muslim households were only slightly more highly represented than Hindu households among the chronically poor during the 2004/05–2011/12 period. However, the likelihood of downward mobility among Muslim households was considerably higher than among Hindu households. This finding resonates with the earlier discussion of the vulnerable position of Muslims in India during the Covid‐19 pandemic.

Profile of the chronically poor or downwardly mobile, 2004/05–2011/12: Religion

Table  6 examines the odds of chronic poverty and downward mobility by population groups defined in terms of dependency ratio. Households with a dependency ratio above 50 per cent are both slightly more likely than average to be chronically poor, and rather more likely than average to experience downward mobility.

Profile of the chronically poor or downwardly mobile, 2004/05–2011/12: Dependency ratio

We turn, finally, to the spatial distribution of chronic poverty and downward mobility across Indian states between 2004/05 and 2011/12. Table  7 indicates that of the major states, chronic poverty was relatively pronounced in Rajasthan, West Bengal, Jharkhand, Orissa, Chhattisgarh and Madhya Pradesh, and to a lesser extent Bihar. These are not states where Covid‐19 outbreaks were initially recorded. Indeed, the states of Maharashtra and Delhi are often described as having been confronted early on by the Covid‐19 crisis, but, with respective odds of chronic poverty between 2004/05 and 2011/12 of 0.932 and 0.715, would appear to have been relatively well protected in this regard. This lends some further support to the contention that the crisis affected a relatively better‐protected segment of the population – at least initially – than might have been feared. Again, however, this observation would need to be tempered by the fact that the large‐scale migration back to rural areas following the lockdown originated largely from such cities as Delhi and Mumbai, and would thus drive rising poverty rates in rural areas of such poor sending‐states such as Bihar, Jharkand, and Orissa.

Profile of the chronically poor or downwardly mobile, India, 2004/05–2011/12: Major states

When we examine the odds of downward mobility and consider those states with particularly large rural populations – regions within which there are fears that the Covid‐19 virus has more recently been spreading – these are also the states with the highest chronic poverty rates during prosperous times: Rajasthan, West Bengal, Jharkhand, Orissa, Chhattisgarh, Madhya Pradesh, joined now also by Bihar, Uttar Pradesh and Himachal Pradesh. In these states, the odds of downward mobility were all greater than average during the 2004/05–2011/12 period. To the extent that our conjecture is correct – that scrutinizing poverty dynamics during good times offers a perspective on looming trouble‐spots and particularly vulnerable groups during an economic downturn – we may take from this that poverty impacts in India will become more pronounced as the coronavirus spreads away from its initial points of entry into regions and among population groups that appeared initially to have been spared.

6. CONCLUDING REMARKS

India is suffering deeply from the Covid‐19 pandemic. Although the country had achieved major successes in its efforts to reduce poverty in the decades up to 2020, there are grounds for suspecting that the combined health and economic consequences of the spread of the coronavirus have brought that progress to a halt. It is reasonable to fear that progress in fighting poverty may have been significantly reversed.

Assessing the full implications of the crisis for poverty is hampered by the lack of solid evidence on poverty in the period immediately preceding the onslaught of the Covid‐19 crisis. Systematic statistical evidence on the evolution of poverty in the face of the crisis is also still pending. However, this paper shows that the rapidly growing literature documenting the spread of the virus, the corresponding impact on health outcomes, and the economic repercussions of the lockdown measures aimed at slowing the spread, provides ample grounds for suspecting a significant impact on poverty. This literature also documents population characteristics and identifies population subgroups that are facing sharply falling living standards.

The numerous studies assembled and reviewed in this paper point to the particular vulnerability of informal sector wage workers in urban areas. The studies document the phenomenon of mass migration by laid‐off temporary workers in urban centers – where the spread of the virus was first documented – back to their villages in rural areas. The literature further indicates that among those who have suffered a loss of employment and falling earnings, women, Scheduled Castes (Dalits) and religious minorities (notably Muslims) are disproportionately represented.

Because the greatest number of coronavirus cases have been documented in large urban centers such as Delhi and Mumbai, and because the phenomenon of mass migration has generally been of laid‐off urban workers, one may be inclined to view the poverty impact of the Covid‐19 crisis as largely confined to urban areas. There are grounds, however, for caution in drawing such a conclusion. First of all, the second Covid‐19 wave has spread far more rapidly and widely into rural regions than was initially experienced in the spring of 2020. Further, it is widely acknowledged that documentation of the spread of the virus is less comprehensive in rural areas than in urban centers, due to fewer resources, lower capacity, and a greater unwillingness of the rural population to submit to testing and the possibility of quarantine. As a result, the current incidence of Covid infections is likely to be both surging, and undercounted, in rural areas. Furthermore, the consequences of a rise in infections may be more difficult to carry for these areas due to their underdeveloped health‐care infrastructure.

Our analysis of poverty dynamics in India – based on pre‐Covid experience during an episode of rapid economic growth and poverty decline – indicates, moreover, that the risk of chronic (long‐term) poverty and the likelihood of falling into poverty are particularly high in rural areas, and in those states to which a majority of Covid‐driven migrants are moving. Our analysis is premised on the notion that poverty persistence and downward mobility during periods of generalized prosperity can help point to likely “hotspots” and particularly at‐risk population groups when economic downturns occur. We suggest, therefore, that while the initial impact of the Covid‐19 pandemic was on sectors and population groups that were relatively less poor, as the crisis spreads and the economic consequences of the crisis continue to reverberate, it is likely that the poverty impacts will become increasingly acute. Absent comprehensive and effective policy interventions to combat these poverty impacts, we suggest that they will be increasingly visible in rural areas.

The world, and India, remain in the midst of an unprecedentedly severe health and economic crisis. Recent progress in the development and application of a vaccine against Covid‐19 provides grounds for some optimism but is unlikely to dramatically alter circumstances on the ground for some time to come. The final consequences of the crisis on poverty are still unclear. There is a pressing need to continue to assemble and analyze emerging evidence to document the scale of the challenge and to identify those who are in greatest need of assistance. Policies must learn from emerging lessons and adapt in response to the evolving picture on the ground. Successful experience must be digested and documented, as it is not unlikely that new pandemics, calling for renewed intervention, will emerge in the future.

Dang, H.‐A. , Lanjouw, P. , & Vrijburg, E. (2021). Poverty in India in the face of Covid‐19: Diagnosis and prospects . Review of Development Economics , 25 , 1816–1837. 10.1111/rode.12833 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]

* This paper is part of the Oxford Policy Management DEEP project undertaken on behalf of the Foreign Commonwealth and Development Office of the UK. We are grateful to two reviewers for comments as well to Finn Tarp, guest editor of this symposium issue, for guidance. We are also grateful to Jehovannes Aikaeli, Ines Afonso Roque Ferreira, Chris Elbers, David Garcés‐Urzainqui, John Hoddinott, Kenneth Mdadila, Tseday Mekasha, Gerton Rongen, and Vincenzo Salvucci for helpful discussions. All errors are our own.

1 It should be noted that, in contrast to other countries, the number of PCR tests in India could not be isolated from the total number of tests performed, indicating that its test rate could also be based on other, less reliable tests (Our World in Data,  2021c ). According to the Hindustan Times , the more unreliable rapid antigen test made up about 49 per cent of the tests in November 2020 (Hindustan Times, 2020 ). Furthermore, other news reports indicate that massive testing seems to be largely done in low‐profile areas just to meet the high‐testing target (Menon,  2020 ; Express Web Desk,  2020 ).

2 See also the Oxford Covid‐19 Government Response Tracker (OxCGRT, n.d. ), which includes an index for the stringency of the government response to Covid‐19. India's government response received a score of 86.57 out of 100 ‘stringency points’ the day the first lockdown was announced. The following days this score rose quickly to 100.

3 Srivastava employs the 2004 National Classification of Occupational categories as defined by the National Career Service project, which is based upon the level of skill and education necessary to perform the occupation. The lowest five out of nine categories are: (5) service workers and shop and market sales workers; (6) skilled agricultural and fishery workers; (7) craft and related trades workers; (8) plant and machinery operators and assemblers; (9) elementary occupations (Srivastava,  2020 , p. 10; Ministry of Labour & Employment,  2015 , pp. 4–14).

4 The PSEC is a committee set up by the Prime Minister's Office on Socio‐Economic and Educational Status of the Muslim Community in India to evaluate the socioeconomic conditions of Muslim Indians (PSEC, 2014). The PSEC uses national data sets (such as from the NSSO) to infer its results.

5 The data that support the findings of this study are available from the corresponding author upon reasonable request.

The data that support the findings of this study are available from the corresponding author upon reasonable request

• Agarwal, V. , & Bellman, E. (2021). India’s migrants flee to their villages as Covid‐19 prompts new lockdown . Retrieved from https://www.wsj.com/articles/indias‐migrants‐flee‐to‐their‐villages‐as‐covid‐19‐prompts‐new‐lockdown‐11618392602 [ Google Scholar ]
• Agrawal, P. (2020). Bihar’s migrants return to face stigma, under‐prepared medical facilities . Retrieved from https://www.indiaspend.com/bihars‐migrants‐return‐to‐face‐stigma‐under‐prepared‐medical‐facilities/ [ Google Scholar ]
• Al Dahdah, M. , Ferry, M. , Guérin, I. , & Venkatasubramanian, G. (2020). The Covid‐19 crisis in India: A nascent humanitarian tragedy . Retrieved from https://booksandideas.net/The‐Covid‐19‐Crisis‐in‐India.html [ Google Scholar ]
• Asadullah, M. N. , & Raghunathan, K. (2020). Fighting gender inequality in India during the COVID‐19 pandemic . Retrieved from https://thediplomat.com/2020/06/fighting‐gender‐inequality‐in‐india‐during‐the‐covid‐19‐pandemic/ [ Google Scholar ]
• Ashwini, D. (2020). The Covid‐19 pandemic and lockdown: First order effects on gender gaps in employment and domestic time use in India . GLO Discussion Paper no. 607, Global Labor Organization (GLO), Essen. Retrieved from https://www.econstor.eu/handle/10419/222416 [ Google Scholar ]
• Azim Premji University (n.d.). COVID19: Analysis of impact and relief measures . Retrieved from https://cse.azimpremjiuniversity.edu.in/covid19‐analysis‐of‐impact‐and‐relief‐measures/#clips [ Google Scholar ]
• Azim Premji University (2021). Impact of COVID‐19 on livelihoods of informal sector workers and vulnerable groups in Bengaluru: A view from the ground . Retrieved from the Azim Premji University website: https://azimpremjiuniversity.edu.in/SitePages/pdf/COVID19‐Impact‐on‐Informal‐Sector‐Bengaluru.pdf [ Google Scholar ]
• Azim Premji University (2021). Reviving employment and livelihoods in India: COVID‐19 and after . Retrieved from the Azim Premji University website: http://publications.azimpremjifoundation.org/2531/1/CII%20Report%20‐%20Employment%20%20Livelihoods.pdf [ Google Scholar ]
• Babu, B. , Kusuma, Y. S. , Sivakami, M. , Kumar Lal, D. , Marimuth, P. , Babu Geddam, J. , Khanna, A. , Agarwal, M. , Sudhakar, G. , Sengupta, P. , Borhade, A. K. Z. , Kerketta, A. S. , & Akoijam, B. S. (2017). Living conditions of internal labour migrants: A nationwide study in 13 Indian cities . International Journal of Migration and Border Studies , 3 ( 4 ), 328–351. [ Google Scholar ]
• Barcellos, S. H. , Leandro, S. C. , & Lleras‐Muney, A. (2014). Child gender and parental investments In India: Are boys and girls treated differently? American Economic Journal: Applied Economics , 6 ( 1 ), 157–189. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Beyer, R. C. M. , Franco‐Bedoya, S. , & Galdo, V. (2021). Examining the economic impact of COVID‐19 in India through daily electricity consumption and nighttime light intensity . World Development , 140 ( 2021 ), 1–13. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Beyer, R. C. M. , Jain, T. , & Sinha, S. (2020). Lights out? Covid‐19 containment policies and economic activity . Policy Research Working Paper No. 9485. the World Bank. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Ceballos, F. , Kannan, S. , & Kramer, B. (2020). Impact of a national lockdown on smallholder farmers’ income and food security: Empirical evidence from two states in India . World Development , 136 ( 2020 ), 1–5. [ Google Scholar ]
• Centre for Monitoring Indian Economy (2020). An unhealthy recovery . Retrieved from the Centre for Monitoring Indian Economy website: https://www.cmie.com/kommon/bin/sr.php?kall=warticle&dt=2020‐08‐18%2011:02:19&msec=596 [ Google Scholar ]
• Chatterjee, M. (2016). Universal health coverage and occupational health and safety for informal workers: A view from India. In: R.M. Taylor, (Rapporteur), Forum on public‐private partnerships for global health and safety (pp. 30–39). Forum conducted at the meeting of the Board on Global Health; Institute of Medicine; National Academies of Sciences, Engineering, and Medicine. The NAtional Academies Press. [ PubMed ] [ Google Scholar ]
• Chaudhuri, S. (2003). ‘Assessing vulnerability to poverty: Concepts, empirical methods and illustrative examples’, mimeo . Department of Economics, Columbia University. Retrieved from https://www.researchgate.net/profile/Detlef_Brem/post/What_is_the_relationship_between_vulnerability_and_poverty_as_well_as_gender_and_poverty/attachment/59d6310fc49f478072ea0cff/AS:273621778403334@1442247923533/download/Assessing+Vulnerability+to+Poverty+theory+2003.pdf [ Google Scholar ]
• Chishti, S. (2020). Explained: How many migrant workers displaced? A range of estimates . Retrieved from https://indianexpress.com/article/explained/coronavirus‐how‐many‐migrant‐workers‐displaced‐a‐range‐of‐estimates‐6447840/ [ Google Scholar ]
• Dang, H. , & Lanjouw, P. (2017). Welfare dynamics measurement: Two definitions of a vulnerability line and the empirical application . Review of Income and Wealth , 63 ( 4 ), 633–660. [ Google Scholar ]
• Dang, H.‐A.H. , & Lanjouw, P. (2018). Welfare dynamics in India over a quarter‐century: Poverty, vulnerability, and mobility, 1987–2012 . WIDER Working Paper 2018/175. UNU‐WIDER. [ Google Scholar ]
• Datt, G. , & Ravallion, M. (2011). Has India’s economic growth become more pro‐poor in the wake of economic reforms? World Bank Economic Review , 25 ( 2 ), 157–189. [ Google Scholar ]
• Dhanaraj, S. (2014). Health shocks and coping strategies: State health insurance scheme of Andhra Pradesh, India (WIDER Working Paper 2014/003) . Retrieved from the UNU‐WIDER website: https://www.wider.unu.edu/publication/health‐shocks‐and‐coping‐strategies [ Google Scholar ]
• Elbers, C. (2021). A note on synthetic panels in poverty mapping, mimeo . School of Business and Economics, Vrije Universiteit. [ Google Scholar ]
• Estupinan, X. , & Sharma, M. (2020). Job and Wage Losses in Informal Sector due to the COVID‐19 Lockdown Measures in India . Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3680379 [ Google Scholar ]
• Express News Service (2020). Explained: What are containment zones, how are they demarcated? Retrieved from https://indianexpress.com/article/explained/coronavirus‐cases‐india‐containment‐zones‐6487494/ [ Google Scholar ]
• Express Web Desk (2020). Coronavirus May 19 highlights: Record 1,08,233 samples tested in a day, says Health Ministry . Retrieved from https://indianexpress.com/article/india/coronavirus‐in‐india‐news‐live‐updates‐covid‐19‐state‐wise‐cases‐tracker‐latest‐news‐lockdown‐4‐guidelines‐6416396/ [ Google Scholar ]
• Ganguly, S. (2020). India’s coronavirus pandemic shines a light on the curse of caste . Retrieved from https://theconversation.com/indias‐coronavirus‐pandemic‐shines‐a‐light‐on‐the‐curse‐of‐caste‐139550 [ Google Scholar ]
• Gangwar, H. S. , & Ray, P. K. C. (2021). Geographic information system‐based analysis of COVID‐19 cases in India during pre‐lockdown, lockdown, and unlock phases . International Journal of Infectious Diseases , 105 ( 2021 ), 424–435. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Garcés‐Urzainqui, D. , Lanjouw, P. , & Rongen, G. (2021). Constructing synthetic panels for the purpose of studying poverty dynamics: A primer . Review of Development Economics . [ Google Scholar ]
• Ghosh, D. , & Dinda, S. (2017). Chapter 6: Health infrastructure and economic development in India. In Das R. C. (Ed.), Social, health, and environmental infrastructure for economic growth (pp. 99–119). IGI Global. [ Google Scholar ]
• Global Hunger Index (2010). The Challenge of Hunger: Focus on the Crisis of Child Undernutrition . Retrieved from the International Food Policy Research Institute site: https://www.ifpri.org/publication/2010‐global‐hunger‐index‐challenge‐hunger [ Google Scholar ]
• Guérin, I. , Michiels, S. , Natal, A. , Nordman, C. J. , & Venkatasubramanian, G. (2020). Surviving debt, survival debt in times of lockdown (CEB Working Paper, no. 20–009) . Retrieved from https://ideas.repec.org/p/sol/wpaper/2013‐309493.html [ Google Scholar ]
• Gulati, A. , Jose, S. , & Singh, B. B. (2021). COVID‐19: Emergence, spread and its impact on the India economy and migrant workers (ZEF Working Paper 207) . Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3834328 [ Google Scholar ]
• Herault, N. , & Jenkins, S. (2019). How valid are synthetic panel estimates of poverty dynamics? Journal of Economic Inequality , 17 , 51–76. 10.1007/s10888-019-09408-8 [ CrossRef ] [ Google Scholar ]
• Hindustan Times . (2020). Rapid antigen tests account for close to 50% of Covid‐19 tests in India: Govt data . Retrieved from the Hindustan Times site: https://www.hindustantimes.com/health/rapid‐antigen‐tests‐account‐for‐close‐to‐50‐of‐covid‐19‐tests‐in‐india‐govt‐data/story‐WLh6mbgmXsZ92EBAA3zzBM.html [ Google Scholar ]
• Home Secretary . (2020). D.O. No. 40‐10/2020‐DM‐1(A) . Retrieved from https://prsindia.org/files/covid19/notifications/5976.IND_Implementation_of_Lockdown_May_21.pdf [ Google Scholar ]
• IMS Institute for Healthcare Informatics (2013). Understanding healthcare access in India: What is the current state? Retrieved from https://docshare04.docshare.tips/files/25555/255552955.pdf [ Google Scholar ]
• International Labour Organization (2018). Women and men in the informal economy: A statistical picture , 3rd ed. ILO. [ Google Scholar ]
• Iyer, M. (2020). Migration in India and the impact of the lockdown on migrants . Retrieved from https://www.prsindia.org/theprsblog/migration‐india‐and‐impact‐lockdown‐migrants [ Google Scholar ]
• John Hopkins University Center for Systems, Science and Engineering . (2021). COVID‐19 dashboard by the center for systems science and engineering (CSSE) at John Hopkins University . Retrieved from https://gisanddata.maps.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6 [ Google Scholar ]
• Kesar, S. , Abraham, R. , Lahoti, R. , Nath, P. , & Basole, A. (2020). Pandemic, informality, and vulnerability: Impact of COVID‐19 on livelihoods in India . CSE Working Paper 2020‐01. Retrieved from the Centre for Sustainable Employment site: https://www.research‐collection.ethz.ch/handle/20.500.11850/428008 [ Google Scholar ]
• La Forgia, G. , & Nagpal, S. (2012). Government‐sponsored health insurance in India: Are you covered? . International Bank for Reconstruction and Development/The World Bank. [ Google Scholar ]
• Maggo, D. (2020). Impact of COVID‐19 on smallholder farmers – insights from India . Retrieved from the World Business Council of Sustainable Development site: https://www.wbcsd.org/Overview/News‐Insights/WBCSD‐insights/Impact‐of‐COVID‐19‐on‐smallholder‐farmers‐in‐India [ Google Scholar ]
• Mehta, T. , & Jamkhandikar, S. (2021). COVID‐19 spreads to rural India, villages ill‐equipped to fight it . Retrieved from https://www.reuters.com/world/india/india‐sees‐record‐daily‐rises‐covid‐19‐infections‐deaths‐2021‐05‐06/ [ Google Scholar ]
• Menon, S. (2020). Coronavirus: Is India’s test and tracing strategy working? Retrieved from the BBC news site: https://www.bbc.com/news/world‐asia‐india‐55083825 [ Google Scholar ]
• Ministry of Labour & Employment (2015). National classification of occupations‐2015 (Code Structure): Vol‐I . Retrieved from https://www.ncs.gov.in/Documents/National%20Classification%20of%20Occupations%20_Vol%20I‐%202015.pdf [ Google Scholar ]
• Ministry of Panchayati Raj (n.d.), Backward regions grant fund: Programme guidelines . Retrieved from https://mizorural.nic.in/file/BRGF/BRGF_GUIDELINES.pdf [ Google Scholar ]
• Ministry of Statistics & Programme Implementation (2015). Employment and unemployment situation among social groups in India (NSS Report No. 563(68/10/4) . Retrieved from http://mospi.nic.in/sites/default/files/publication_reports/nss_rep_563_13mar15.pdf [ Google Scholar ]
• Mitra, E. , Talreja, S. , & Yeung, J. (2021). India’s second wave is ravaging villages without hospitals or even doctors to fight it . Retrieved from https://edition.cnn.com/2021/05/11/india/india‐covid‐gujarat‐village‐intl‐hnk/index.html [ Google Scholar ]
• Naik, M. (2021). As the pandemic’s second wave rages, migrants trudge home again . Retrieved from https://indianexpress.com/article/opinion/columns/as‐the‐pandemics‐second‐wave‐rages‐migrants‐trudge‐home‐again‐7297303/ [ Google Scholar ]
• Narayanan, A. (2015). Informal employment in India: Voluntary choice or a result of labor market segmentation? Indian Journal of Labour Economics , 58 ( 1 ), 119–167. 10.1007/s41027-015-0009-9 [ CrossRef ] [ Google Scholar ]
• National Commission for Enterprises in the Unorganised Sector (2009). The challenge of employment in India: An informal economy perspective (Volume 1‐Main Report) . Retrieved from the Development Commissioner Ministry of Micro, Small and Medium Enterprises website: http://dcmsme.gov.in/The_Challenge_of_Employment_in_India.pdf [ Google Scholar ]
• Niyati, S. , & Vijayamba, R. (2020). Impact of the Covid‐19 pandemic on food security and indebtedness in rural India . Retrieved from https://www.researchgate.net/profile/Niyati‐Singaraju/publication/351082698_Impact_of_the_Covid‐19_Pandemic_on_Food_Security_and_Indebtedness_in_Rural_India/links/6083fce9907dcf667bbdff6c/Impact‐of‐the‐Covid‐19‐Pandemic‐on‐Food‐Security‐and‐Indebtedness‐in‐Rural‐India.pdf [ Google Scholar ]
• Ministry of Home Affairs (2020a). Order. No. 40‐3/2020‐DM‐I(A) . Retrieved from https://prsindia.org/files/covid19/notifications/144.IND_Citizens_Guidelines_Lockdown_Mar_24.pdf [ Google Scholar ]
• Ministry of Home Affairs (2020b). Order. No. 40‐3/2020‐DM‐I(A) . Retrieved from https://prsindia.org/files/covid19/notifications/2737.IND_Extension_Lockdown_Order_April_14.pdf [ Google Scholar ]
• Ministry of Home Affairs (2020c). Order. No. 40‐3/2020‐DM‐I(A) . Retrieved from https://prsindia.org/files/covid19/notifications/IND_Extension_Lockdown_May_1.pdf [ Google Scholar ]
• Ministry of Home Affairs (2020d). Order. No. 40‐3/2020‐DM‐I(A) . Retrieved from https://prsindia.org/files/covid19/notifications/IND_MHA_Lockdown_Extension_upto_May31_17052020.pdf [ Google Scholar ]
• Ministry of Home Affairs (2020e). Order. No. 40‐3/2020‐DM‐I(A) . Retrieved from https://prsindia.org/files/covid19/notifications/MHAOrder_29062020.pdf [ Google Scholar ]
• National Commision for Enterprises in the Unorganised Sector (NCEUS) (2007). Report on conditions of work and promotion of livelihoods in the unorganised sector . New Delhi, India. [ Google Scholar ]
• Oster, E. (2006). Does increased access increase equality? Gender and child health investments in India (NBER Working Paper No. 12743). Retrieved from the NBER site: http://www.nber.org/papers/w12743 [ Google Scholar ]
• Our World in Data (2021a). Cumulative confirmed COVID‐19 cases per million people, May 25, 2021 . Retrieved from https://ourworldindata.org/coronavirus‐data‐explorer?tab=map&zoomToSelection=true&country=GBR~USA~ESP~ITA~BRA~IND~KOR&region=World&casesMetric=true&interval=total&hideControls=true&perCapita=true&smoothing=0&pickerMetric=location&pickerSort=asc [ Google Scholar ]
• Our World in Data (2021b). Daily new confirmed Covid‐19 cases . Retrieved from https://ourworldindata.org/explorers/coronavirus‐data‐explorer?zoomToSelection=true&hideControls=true&Metric=Confirmed+cases&Interval=New+per+day&Relative+to+Population=false&Align+outbreaks=false&country=~IND [ Google Scholar ]
• Our World in Data (2021c). Daily tests per thousand people (7‐day smoothed) . Retrieved from https://ourworldindata.org/coronavirus‐testing [ Google Scholar ]
• OxCGRT (n.d.). Relationship between number of COVID‐19 cases and government response . Retrieved from https://covidtracker.bsg.ox.ac.uk/stringency‐scatter [ Google Scholar ]
• Pandya, A. A. (2010). Muslim Indians: Struggle for Inclusion . Retrieved from the Stimson Regional Voices Website, https://www.stimson.org/wp‐content/files/file‐attachments/Muslim_Indians‐Complete_1.pdf [ Google Scholar ]
• Post Sachar Evaluation Committee (2014). Post Sachar evaluation committee report . Retrieved from https://www.sabrangindia.in/sites/default/files/audio_listing_images/kundu_commission_report_0.pdf?698 [ Google Scholar ]
• Pritchett, L. , Suryahadi, A. , & Sumarto, S. (2000). Quantifying vulnerability to poverty: A proposed measure, applied to Indonesia . World Bank Policy Research Working Paper No. 2437, The World Bank. [ Google Scholar ]
• Ravallion, M. (2011). A comparative perspective on poverty reduction in Brazil, China and India . World Bank Research Observer , 26 ( 1 ), 71–104. 10.1093/wbro/lkp031 [ CrossRef ] [ Google Scholar ]
• Rawal, V. , Kumar, M. , Verma, A. , & Pais, J. (2020). COVID‐19 Lockdown: Impact on agriculture and rural economy. SSER Monograph 20/3 . Retrieved from http://archive.indianstatistics.org/sserwp/sserwp2003.pdf [ Google Scholar ]
• Ray, D. , & Subramanian, S. (2020). India’s lockdown: An interim report . Retrieved from https://debrajray.com/wp‐content/uploads/2020/05/RaySubramanian.pdf [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sengupta, S. , & Jha, M. K. (2020). Social policy, COVID‐19 and impoverished migrants: Challenges and prospects in locked down India . The International Journal of Community and Social Development , 2 ( 2 ), 152–172. 10.1177/2516602620933715 [ CrossRef ] [ Google Scholar ]
• Seth, A. (2020). The post‐lockdown state of labour: Heading towards an institutionalization of exploitation and inequality . Retrieved from https://drive.google.com/file/d/1baXB35TNNlp‐ap6fMZtg4eIFg4Yxe8Mk/view [ Google Scholar ]
• Singh, K. D. , Goel, V. , Kumar, H. , & Gettleman, J. (2020). India, Day 1: World’s largest coronavirus lockdown begins . Retrieved from https://www.nytimes.com/2020/03/25/world/asia/india‐lockdown‐coronavirus.html [ Google Scholar ]
• Singh, S. , & Magazine, A. (2020). Explained: Indian migrants, across India . Retrieved from https://indianexpress.com/article/explained/coronavirus‐india‐lockdown‐migran‐workers‐mass‐exodus‐6348834/ [ Google Scholar ]
• Sinha, A. (2021). In second wave, Covid hits rural India: Cases and deaths quadruple . Retrieved from https://indianexpress.com/article/india/covid‐second‐wave‐rural‐india‐coronavirus‐cases‐7306422/ [ Google Scholar ]
• Srivastava, R. (2020). Understanding circular migration in India: Its nature and dimensions, the crisis under lockdown and the response of the state (Centre for Employment Studies WP 04/2020) . Retrieved from the Institute for Human Development website: http://www.ihdindia.org/working‐papers/2020/IHD‐CES_WP_04_2020.pdf [ Google Scholar ]
• The Economist . (2021). Tracking the economic impact of India’s second covid wave . Retrieved from The Economist website: https://www.economist.com/finance‐and‐economics/2021/05/01/tracking‐the‐economic‐impact‐of‐indias‐second‐covid‐wave [ Google Scholar ]
• The Indian Express (2021). Daily Briefing: Massive spread of Covid‐10 in rural India; nearly 100 bodies found floating in Ganga; and more . Retrieved from https://indianexpress.com/article/india/daily‐briefing‐covid‐19‐vaccination‐drive‐second‐wave‐may‐12‐7311703/ [ Google Scholar ]
• Thejesh, G. N. (2020). Non virus deaths . Retrieved from https://thejeshgn.com/projects/covid19‐india/non‐virus‐deaths/ [ Google Scholar ]
• Tikkanen, R. , Osborn, R. , Mossialos, E. , Djordjevic, A. , & Wharton, G. A. (2020). International health care system profiles: India . Retrieved from the Commonwealth Fund Website https://www.commonwealthfund.org/international‐health‐policy‐center/countries/india [ Google Scholar ]
• Unni, J. , & Rani, U. (2003). Social protection for informal workers in India: Insecurities, instruments and institutional mechanisms . Development and Change , 34 ( 1 ), 127–161. [ Google Scholar ]
• World Economic Forum (2020). Global gender gap report 2020 . Retrieved from the World Economic Forum Website: https://www.weforum.org/reports/gender‐gap‐2020‐report‐100‐years‐pay‐equality [ Google Scholar ]
• World Health Organization (2021). WHO coronavirus (COVID‐19) dashboard: Situation by Region, Country, Territory & Area . Retrieved from https://covid19.who.int/table [ Google Scholar ]

Advertisement

I.  Introduction

Ii.  the expenditure surveys, iii.  nsso versus nas expenditure estimates, iv.  the official poverty lines, v.  controversies regarding poverty lines 3, vi.  poverty at the national level, vii.  poverty in the states: rural and urban, viii.  poverty in the states by social group, ix.  poverty in the states by religious group, x.  inequality, xi.  concluding remarks, a comprehensive analysis of poverty in india.

• Cite Icon Cite
• Open the PDF for in another window
• Permissions
• Article contents
• Figures & tables
• Supplementary Data
• Peer Review
• Search Site

Arvind Panagariya , Megha Mukim; A Comprehensive Analysis of Poverty in India. Asian Development Review 2014; 31 (1): 1–52. doi: https://doi.org/10.1162/ADEV_a_00021

Download citation file:

• Ris (Zotero)
• Reference Manager

This paper offers a comprehensive analysis of poverty in India. It shows that regardless of which of the two official poverty lines we use, we see a steady decline in poverty in all states and for all social and religious groups. Accelerated growth between fiscal years 2004–2005 and 2009–2010 also led to an accelerated decline in poverty rates. Moreover, the decline in poverty rates during these years has been sharper for the socially disadvantaged groups relative to upper caste groups so that we now observe a narrowing of the gap in the poverty rates between the two sets of social groups. The paper also provides a discussion of the recent controversies in India regarding the choice of poverty lines.

This paper provides comprehensive up-to-date estimates of poverty by social and religious groups in the rural and urban areas of the largest 17 states in India. The specific measure of poverty reported in the paper is the poverty rate or headcount ratio (HCR), which is the proportion of the population with expenditure or income below a pre-specified level referred to as the poverty line. In the context of most developing countries, the poverty line usually relates to a pre-specified basket of goods presumed to be necessary for above-subsistence existence.

In so far as prices vary across states and between rural and urban regions within the same state, the poverty line also varies in nominal rupees across states and between urban and rural regions within the same state. 1 Similarly, since prices rise over time due to inflation, the poverty line in nominal rupees in a given location is also adjusted upwards over time.

The original official poverty estimates in India, provided by the Planning Commission, were based on the Lakdawala poverty lines, so named after Professor D. T. Lakdawala who headed a 1993 expert group that recommended these lines. Recommendations of a 2009 expert committee headed by Professor Suresh Tendulkar led to an upward adjustment in the rural poverty line relative to its Lakdawala counterpart. Therefore, while the official estimates for earlier years were based on the lines and methodology recommended by the expert group headed by Lakdawala, those for more recent years were based on the line and methodology recommended by the Tendulkar Committee. Official estimates based on both methodologies exist for only two years, 1993–1994 and 2004–2005. These estimates are provided for the overall population, for rural and urban regions of each state, and for the country as a whole. The Planning Commission does not provide estimates by social or religious groups.

In this paper, we provide estimates using Lakdawala and Tendulkar lines for different social and religious groups in rural and urban areas in all major states and at the national level. Our estimates based on Lakdawala lines are computed for all years beginning in 1983 for which large or “thick” expenditure surveys have been conducted. Estimates based on the Tendulkar line and methodology are provided for the three latest large expenditure surveys, 1993–1994, 2004–2005, and 2009–2010.

Our objective in writing the paper is twofold. First, much confusion has arisen in the policy debates in India around certain issues regarding poverty in the country—for instance, whether or not growth has helped the poor (if yes, how much and over which time period) and whether growth is leaving certain social or religious groups behind. We hope that by providing poverty estimates for various time periods, social groups, religious groups, states, and urban and rural areas, this paper will help ensure that future policy debates are based on fact. Second, researchers interested in explaining how various policy measures impact poverty might find it useful to have the poverty lines and the associated poverty estimates for various social and religious groups and across India's largest states in rural and urban areas readily available in one place.

The literature on poverty in India is vast and many of the contributions or references to the contributions can be found in Srinivasan and Bardhan ( 1974 , 1988 ), Fields ( 1980 ), Tendulkar ( 1998 ), Deaton and Drèze ( 2002 ), Bhalla ( 2002 ), and Deaton and Kozel ( 2005 ). Panagariya ( 2008 ) provides a comprehensive treatment of the subject until the mid-2000s including the debates on whether or not poverty had declined in the post-reform era and whether or not reforms had been behind the acceleration in growth rates and the decline in poverty. Finally, several of the contributions in Bhagwati and Panagariya ( 2012a , 2012b ) analyze various aspects of poverty in India using the expenditures surveys up to 2004–2005. In particular, Cain, Hasan, and Mitra ( 2012 ) study the impact of openness on poverty; Mukim and Panagariya ( 2012 ) document the decline in poverty across social groups; Dehejia and Panagariya ( 2012 ) provide evidence on the growth in entrepreneurship in services sectors among the socially disadvantaged groups; and Hnatkovska and Lahiri ( 2012 ) provide evidence on and reasons for narrowing wage inequality between the socially disadvantaged groups and the upper castes.

To our knowledge, this is the first paper to systematically and comprehensively exploit the expenditure survey conducted in 2009–2010. This is important because growth was 2–3 percentage points higher between 2004–2005 and 2009–2010 surveys than between any other prior surveys. As such, we are able to study the differential impact accelerated growth has had on poverty alleviation both directly, through improved employment and wage prospects for the poor, and indirectly, through the large-scale redistribution program known as the National Rural Employment Guarantee Scheme, which enhanced revenues made possible. In addition, ours is also the first paper to comprehensively analyze poverty across religious groups. In studying the progress in combating poverty across social groups, the paper complements our previous work, Mukim and Panagariya ( 2012 ).

The paper is organized as follows. In Section II , we discuss the history and design of the expenditure surveys conducted by the National Sample Survey Office (NSSO), which form the backbone of all poverty analysis in India. In Section III , we discuss the rising discrepancy between average expenditures as reported by the NSSO surveys and by the National Accounts Statistics (NAS) of the Central Statistical Office (CSO). In Section IV , we describe in detail the evolution of official poverty lines in India, while in Section V we discuss some recent controversies regarding the level of the official poverty line. In Sections VI to Section IX , we present the poverty estimates. In Section X , we discuss inequality over time in rural and urban areas of the 17 states. In Section XI , we offer our conclusions.

The main source of data for estimating poverty in India is the expenditure survey conducted by the NSSO. India is perhaps the only developing country that began conducting such surveys on a regular basis as early as 1950–1951. The surveys have been conducted at least once a year since 1950–1951. However, the sample had been too small to permit reliable estimates of poverty at the level of the state until 1973–1974. A decision was made in the early 1970s to replace the smaller annual surveys by large-size expenditure (and employment–unemployment) surveys to be conducted every 5 years.

This decision led to the birth of “thick” quinquennial (5-yearly) surveys. Accordingly, the following 8 rounds of large-size surveys have been conducted: 27 (1973–1974), 32 (1978), 38 (1983), 43 (1987–1988), 50 (1993–1994), 55 (1999–2000), 61 (2004–2005), and 66 (2009–2010). Starting from the 42nd round in 1986–1987, a smaller expenditure survey was reintroduced. This was conducted annually except during the years in which the quinquennial survey was to take place. Therefore, with the exception of the 65th and 67th rounds in 2008–2009 and 2010–2011, respectively, an expenditure survey exists for each year beginning 1986–1987.

While the NSSO collects the data and produces reports providing information on monthly per-capita expenditures, it is the Planning Commission that computes the poverty lines and provides official estimates of poverty. The official estimates are strictly limited to quinquennial surveys. While they cover rural, urban, and total populations in different states and at the national level, estimates are not provided for specific social or religious groups. These can be calculated selectively for specific groups or specific years by researchers. With rare exceptions, discussions and debates on poverty have been framed around the quinquennial surveys even though the other survey samples are large enough to allow reliable estimates at the national level.

For each household interviewed, the survey collects data on the quantity of and expenditure on a large number of items purchased. For items such as education and health services, where quantity cannot be meaningfully defined, only expenditure data are collected. The list of items is elaborate. For example, the 66th round collected data on 142 items under the food category; 15 items under energy; 28 items under clothing, bedding, and footwear; 19 items under educational and medical expenses; 51 items under durable goods; and 89 in the other items category.

It turns out that household responses vary systematically according to the length of the reference period to which the expenditures are related. For example, a household could be asked about its expenditures on durable goods during the preceding 30 days or the preceding year. When the information provided in the first case is converted into annual expenditures, it is found to be systematically lower than when the survey directly asks households to report their annual spending. Therefore, estimates of poverty vary depending on the reference period chosen in the questionnaire.

Most quinquennial surveys have collected information on certain categories of relatively infrequently purchased items including clothing and consumer durables on the basis of both 30-day and 365-day reference periods. For other categories, including all food and fuel and consumer services, they have used a 30-day reference period. The data allow us to estimate two alternative measures of monthly per-capita expenditures that refer to the following: (i) a uniform reference period (URP) where all expenditure data used to estimate monthly per-capita expenditure are based on the 30-day reference period, and (ii) a mixed reference period (MRP) where expenditure data used to estimate the monthly per-capita expenditure are based on the 365-day reference period in the case of clothing and consumer durables and the 30-day reference period in the case of other items.

With rare exceptions, monthly per-capita expenditure associated with the MRP turns out to be higher than that associated with the URP. The Planning Commission's original estimate of poverty that employed the Lakdawala poverty lines had relied on the URP monthly per-capita expenditures. At some time prior to the Tendulkar Committee report, however, the Planning Commission decided to shift to the MRP estimates. Therefore, while recommending revisions that led to an upward adjustment in the rural poverty line, the Tendulkar Committee also shifted to the MRP monthly per-capita expenditures in its poverty calculations. Therefore, the revised poverty estimates available for 1993–1994, 2004–2005, and 2009–2010 are based on the Tendulkar lines and the MRP estimates of monthly per-capita expenditures.

We note an important feature of the NSSO expenditure surveys at the outset. The average monthly per-capita expenditure based on the surveys falls well short of the average private consumption expenditure separately available from the NAS of the CSO. Moreover, the proportionate shortfall has been progressively rising over successive surveys. These two observations hold regardless of whether we use the URP or MRP estimate of monthly per-capita expenditure available from the NSSO. Figure 1 graphically depicts this phenomenon in the case of URP monthly per-capita expenditure, which is more readily available for all quinquennial surveys since 1983.

NSSO Household Total URP Expenditure Estimate as % of NAS Total Private Consumption Expenditure

Precisely what explains the gap between the NSSO and NAS expenditures has important implications for poverty estimates. For example, if the gap in any given year is uniformly distributed across all expenditure classes as Bhalla ( 2002 ) assumes in his work, true expenditure in 2009–2010 is uniformly more than twice of what the survey finds. This would imply that many individuals currently classified as falling below the poverty line are actually above it. Moreover, a recognition that the proportionate gap between NSSO and NAS private expenditures has been rising over time implies that the poverty ratio is being overestimated by progressively larger margins over time. At the other extreme, if the gap between NSSO and NAS expenditures is explained entirely by underreporting of the expenditures by households classified as non-poor, poverty levels will not be biased upwards.

There are good reasons to believe, however, that the truth lies somewhere between these two extremes. The survey underrepresents wealthy consumers. For instance, it is unlikely that any of the billionaires, or most of the millionaires, are covered by the survey. Likewise, the total absence of error among households below the poverty line is highly unlikely. For example, recall that the expenditures on durables are systematically underreported for the 30-day reference period relative to that for 365-day reference period. Thus, in all probability, households classified as poor account for part of the gap so that there is some overestimation of the poverty ratio at any given poverty line. 2

The 1993 expert group headed by Lakdawala defined all-India rural and urban poverty lines in terms of per-capita total consumption expenditure at 1973–1974 market prices. The underlying consumption baskets were anchored to the per-capita calorie norms of 2,400 and 2,100 in rural and urban areas, respectively. The rural and urban poverty line baskets were based on different underlying baskets, which meant that the two poverty lines represented different levels of real expenditures.

State-level rural poverty lines were derived from the national rural poverty line by adjusting the latter for price differences between national and state-level consumer price indices for agricultural laborers. Likewise, state-level urban poverty lines were derived from the national urban poverty line by adjusting the latter for price differences between the national and state-level consumer price indices for industrial laborers. National and state-level rural poverty lines were adjusted over time by applying the national and state-level price indices for agricultural workers, respectively. Urban poverty lines were adjusted similarly over time.

Lakdawala lines served as the official poverty lines until 2004–2005. The Planning Commission applied them to URP-based expenditures in the quinquennial surveys to calculate official poverty ratios. Criticisms of these estimates on various grounds led the Planning Commission to appoint an expert group under the chairmanship of Suresh Tendulkar in December 2005 with the directive to recommend appropriate changes in methodology for computing poverty estimates. The group submitted its report in 2009.

In its report, the Tendulkar committee noted three deficiencies of the Lakdawala poverty lines (Government of India 2009 ). First, the poverty line baskets remained tied to consumption patterns observed in 1973–1974. But more than 3 decades later, these baskets had shifted, even for the poor. Second, the consumer price index for agricultural workers understated the true price increase. This meant that over time the upward adjustment in the rural poverty lines was less than necessary so that the estimated poverty ratios understated rural poverty. Finally, the assumption underlying Lakdawala lines that health and education would be largely provided by the government did not hold any longer. Private expenditures on these services had risen considerably, even for the poor. This change was not adequately reflected in the Lakdawala poverty lines.

To remedy these deficiencies, the Tendulkar committee began by noting that the NSSO had already decided to shift from URP-based expenditures to MRP-based expenditures to measure poverty. With this in view, the committee's first step was to situate the revised poverty lines in terms of MRP expenditures in some generally acceptable aspect of the existing practice. To this end, it observed that since the nationwide urban poverty ratio of 25.7%, calculated from URP-based expenditures in the 2004–2005 survey, was broadly accepted as a good approximation of prevailing urban poverty, the revised urban poverty line could be anchored to yield this same estimate using MRP-based per-capita consumption expenditure from the 2004–2005 survey. This decision led to MRP-based per-capita expenditure of the individual at the 25.7 percentile in the national distribution of per-capita MRP expenditures becoming the national urban poverty line.

The Tendulkar committee further argued that the consumption basket associated with the national urban poverty line also be accepted as the rural poverty line consumption basket. This implied the translation of the new urban poverty line using the appropriate price index to obtain the nationwide rural poverty line. Under this approach, rural and urban poverty lines became fully aligned. Applying MRP-based expenditures, the new rural poverty line yielded a rural poverty ratio of 41.8% in 2004–2005 compared with 28.3% under the old methodology.

It is important to note that even though the method of pegging the national urban poverty line in the manner done by the Tendulkar committee left the national urban poverty in 2004–2005 originally measured at the Lakdawala urban poverty line unchanged, it did impact state-level urban poverty estimates. The methodology required that the state-level rural and urban poverty lines be derived from the national urban poverty line by applying the appropriate price indices derived from the price information within the sample surveys. In some cases, the state-level shift was sufficiently large to significantly alter the estimate of urban poverty. For example, Lakdawala urban poverty line in Gujarat in 2004–2005 was Rs541.16 per-capita per month. The corresponding Tendulkar line turned out to be Rs659.18. This change led the urban poverty estimate in 2004–2005 to jump from 13.3% based on the Lakdawala line to 20.1% based on the Tendulkar line.

An important final point concerns the treatment of health and education spending by the Tendulkar Committee in recommending the revised poverty lines. On this issue, it is best to directly quote the Tendulkar Committee report (Government of India 2009 , p. 2):

Even while moving away from the calorie norms, the proposed poverty lines have been validated by checking the adequacy of actual private expenditure per capita near the poverty lines on food, education, and health by comparing them with normative expenditures consistent with nutritional, educational, and health outcomes. Actual private expenditures reported by households near the new poverty lines on these items were found to be adequate at the all-India level in both the rural and the urban areas and for most of the states. It may be noted that while the new poverty lines have been arrived at after assessing the adequacy of private household expenditure on education and health, the earlier calorie-anchored poverty lines did not explicitly account for these. The proposed poverty lines are in that sense broader in scope.

We address here the two rounds of controversies over the poverty line that broke out in the media in September 2011 and March 2012. The first round of controversy began with the Planning Commission filing an affidavit with the Supreme Court stating that the poverty line at the time had been on average Rs32 and Rs26 per person per day in urban and rural India, respectively. Being based on the Tendulkar methodology, these lines were actually higher than the Lakdawala lines on which the official poverty estimates had been based until 2004–2005. However, the media and civil society groups pounced on the Planning Commission for diluting the poverty lines so as to inflate poverty reduction numbers and to deprive many potential beneficiaries of entitlements. For its part, the Planning Commission did a poor job of explaining to the public precisely what it had done and why.

The controversy resurfaced in March 2012 when the Planning Commission released the poverty estimates based on the 2009–2010 expenditure survey. The Planning Commission reported that these estimates were based on average poverty lines of Rs28.26 and Rs22.2 per person per day in urban and rural areas, respectively. Comparing these lines to those previously reported to the Supreme Court, the media once again accused the Planning Commission of lowering the poverty lines. 4 The truth of the matter was that whereas the poverty lines reported to the Supreme Court were meant to reflect the price level prevailing in mid-2011, those underlying poverty estimates for 2009–2010 were based on the mid-point of 2009–2010. The latter poverty lines were lower because the price level at the mid-point of 2009–2010 was lower than that in mid-2011. In real terms, the two sets of poverty lines were identical.

While there was no basis to the accusations that the Planning Commission had lowered the poverty lines, the issue of whether the poverty lines remain excessively low despite having been raised does require further examination. In addressing this issue, it is important to be clear about the objectives behind the poverty line.

Potentially, there are two main objectives behind poverty lines: to track the progress made in combating poverty and to identify the poor towards whom redistribution programs can be directed. The level of the poverty line must be evaluated separately against each objective. In principle, we may want separate poverty lines for the two objectives.

With regard to the first objective, the poverty line should be set at a level that allows us to track the progress made in helping the truly destitute or those living in abject poverty, often referred to as extreme poverty. Much of the media debate during the two episodes focused on what could or could not be bought with the poverty-line expenditure. 5 There was no mention of the basket of goods that was used by the Tendulkar Committee to define the poverty line.

In Annex E of its report (Government of India 2009 ), the Tendulkar Committee gave a detailed itemized list of the expenditures of those “around poverty line class for urban areas in all India.” Unfortunately, it did not report the corresponding quantities purchased of various commodities. In this paper, we now compute these quantities from unit-level data where feasible and report them in Table 1 for a household consisting of five members. 6 Our implicit per-person expenditures on individual items are within Rs3 of their corresponding expenditures reported in Annex E of the report of the Tendulkar Committee.

Source: Authors’ calculations using unit-level data (supplied by Rahul).

We report quantities wherever the relevant data are available. In the survey, the quantities are not always reported in weights. For example, lemons and oranges are reported in numbers and not in kilograms. In these cases, we have converted the quantities into kilograms using the appropriate conversion factors. The main point to note is that while the quantities associated with the poverty line basket may not permit a comfortable existence, including a balanced diet, they allow above-subsistence existence. The consumption of cereals and pulses at 50.9 kilograms (kg) and 3.5 kg compared with 48 kg and 5.5 kg, respectively, for the mean consumption of the top 30% of the population. Likewise, the consumption of edible oils and vegetables at 2.7 kg and 23.9 kg for the poor compared with 4.5 kg and 35.5 kg, respectively, for the top 30% of the population. 7 This comparison shows that, at least in terms of the provision of two square meals a day, the poverty line consumption basket is compatible with above-subsistence level consumption.

We reiterate our point as follows. In 2009–2010, the urban poverty line in Delhi was Rs1,040.3 per person per month (Rs34.2 per day). For a family of five, this amount would translate to Rs5,201.5 per month. Assuming that each family member consumes 10 kg per month of cereal and 1 kg per month of pulses and the prices of the two grains are Rs15 and Rs80 per kilogram, respectively, the total expenditure on grain would be Rs1,150. 8 This would leave Rs4,051.5 for milk, edible oils, fuel, clothing, rent, education, health, and other expenditures. While this amount may not allow a fully balanced diet, comfortable living, and access to good education and health, it is consistent with an above-subsistence level of existence. Additionally, if we take into account access to public education and health, and subsidized grain and fuel from the public distribution system, the poverty line is scarcely out of line with the one that would allow exit from extreme poverty.

But what about the role of the poverty line in identifying the poor for purposes of redistribution? Ideally, this exercise should be carried out at the local level in light of resources available for redistribution, since the poor must ultimately be identified locally. Nevertheless, if the national poverty line is used to identify the poor, could we still defend the Tendulkar line as adequate? We argue in the affirmative.

Going by the urban and rural population weights of 0.298 and 0.702 implicit in the population projections for 1 January 2010, the average countrywide per-capita MRP expenditure during 2009–2010 amounts to Rs40.2 per person per day. Therefore, going by the expenditure survey data, equal distribution across the entire country would allow barely Rs40.2 per person per day in expenditures. Raising the poverty line significantly above the current level must confront this limit with regard to the scope for redistribution.

It could be argued that this discussion is based on data in the expenditure survey, which underestimates true expenditures. The scope for redistribution might be significantly greater if we go by expenditures as measured in the NAS. The response to this criticism is that the surveys underestimate not just the average national expenditure but also the expenditures of those identified as poor. Depending on the extent of this underestimation, the need for redistribution itself would be overestimated.

Even so, it is useful to test the limits of redistribution by considering the average expenditure according to the NAS. The total private final consumption expenditure at current prices in 2009–2010 was Rs37,959.01 billion. Applying the population figure of 1.174 billion as of 1 January 2010 in the NSSO 2009–2010 expenditure survey, this total annual expenditure translates to daily spending of Rs88.58 per person. This figure includes certain items such as imputed rent on owner-occupied housing and expenditures other than those by households such as the spending of civil society groups, which would not be available for redistribution. Thus, per-capita expenditures achievable through equal distribution, even when we consider the expenditures as per the NAS, is likely to be modest.

To appreciate further the folly of setting too high a poverty line for the purpose of identifying the poor, recall that the national average poverty line was Rs22.2 per person per day in rural areas and Rs28.26 in urban areas in 2009–2010. Going by the expenditure estimates for different spending classes in Government of India ( 2011a ), raising these lines to just Rs33.3 and Rs45.4, respectively, would place 70% of the rural population and 50% of the urban population in poverty in 2009–2010. If we went a little further and set the rural poverty line at Rs39 per day and the urban poverty line at Rs81 per day in 2009–2010, we would place 80% of the population in each region below the poverty line. Will the fate of the destitute not be compromised if the meager tax revenues available for redistribution were thinly spread on this much larger population?

Before we turn to reporting the poverty estimates, we should clarify that while we have defended the current poverty line in India for both purposes—tracking abject poverty and redistribution—in general, we believe a case exists for two separate poverty lines to satisfy the two objectives. The poverty line to track abject poverty must be drawn independently of the availability of revenues for redistribution purposes and should be uniform nationally. The poverty line for redistribution purposes would in general differ from this line and, indeed, vary in different jurisdictions of the same nation depending on the availability of revenues. This should be evident from the fact that redistribution remains an issue even in countries that have entirely eradicated abject poverty. 9

Official poverty estimates are available at the national and state levels for the entire population, but not by social or religious groups, for all years during which the NSSO conducted quinquennial surveys. These years include 1973–1974, 1977–1978, 1983, 1987–1988, 1993–1994, 2004–2005, and 2009–2010, but not 1999–2000, as that year's survey became noncomparable to other quinquennial surveys due to a change in sample design. The Planning Commission has published poverty ratios for the first six of these surveys based on the Lakdawala lines and for the last three based on the Tendulkar lines. These ratios were estimated for rural and urban areas at the national and state levels.

In this paper, we provide comparable poverty rates for all of the last five quinquennial surveys including 2009–2010 derived from Lakdawala lines. For this purpose, we update the 2004–2005 Lakdawala lines to 2009–2010 using the price indices implicit in the official Tendulkar lines for 2004–2005 and 2009–2010 at the national and state levels. We provide estimates categorized by social as well as religious groups for all quinquennial surveys beginning in 1983 based on the Lakdawala lines and for the years relating to the last three such surveys based on the Tendulkar lines at the national and state levels.

While we focus mainly on the evolution of poverty since 1983 in this paper, it is useful to begin with a brief look at the poverty profile in the early years. This is done in Figure 2 using the estimates in Datt ( 1998 ) for years 1951–1952 to 1973–1974. The key message of the graph is that the poverty ratio hovered between 50% and 60% with a mildly rising trend.

The Poverty Ratio in India, 1951–1952 to 1973–1974 (%)

This is not surprising, as India had been extremely poor at independence. Unlike economies such as Taipei, China; the Republic of Korea; Singapore; and Hong Kong, China, the country then grew very slowly. Growth in per-capita income during these years had been a mere 1.5% per year. Such low growth coupled with a very low starting per-capita income meant at best limited scope for achieving poverty reduction even through redistribution. As argued above, even today, after more than 2 decades of almost 5% growth in per-capita income, the scope for redistribution remains limited. 10

We are now in a position to provide the poverty rates for the major social groups based on the quinquennial expenditure surveys beginning 1983. The social groups identified in the surveys are scheduled castes (SC), scheduled tribes (ST), other backward castes (OBC), and the rest, which we refer to as forward castes (FC). In addition, we define the nonscheduled castes as consisting of the OBC, and FC. The NSSO began identifying the OBC beginning 1999–2000. Since we are excluding this particular survey due to its lack of comparability with other surveys, the OBC as a separate group begins appearing in our estimates from 2004–2005 only.

In Table 2 , we provide the poverty rates based on the Lakdawala lines in rural and urban areas and at the national level. Four features of this table are worthy of note. First, poverty rates have continuously declined for every single social group in both the rural and urban areas. Contrary to common claims, growth has been steadily helping the poor from every broad social group escape poverty rather than leaving the socially disadvantaged behind.

FC = forward castes, NS = non-scheduled, OBC = other backward castes, SC = scheduled castes, ST = scheduled tribes.

Source: Authors’ calculations.

Second, the rates in rural India have consistently been the highest for the ST followed by the SC, OBC, and FC in that order. This pattern also holds in urban areas but with some exceptions. In particular, in some years, poverty rates of scheduled tribes are lower than that of scheduled castes, but this is not of great significance since more than 90% of the scheduled tribe population live in rural areas.

Third, with growth accelerating to above 8% beginning 2003–2004, poverty reduction between 2004–2005 and 2009–2010 has also accelerated. The percentage point reduction during this period has been larger than during any other 5-year period. Most importantly, the acceleration has been the greatest for the ST and SC in that order so that at last, the gap in poverty rates between the scheduled and nonscheduled groups has declined significantly.

Finally, while the rural poverty rates were slightly higher than the urban poverty rates for all groups in 1983, the order switched for one or more groups in several of the subsequent years. Indeed, in 2009–2010, the urban rates turned out to be uniformly higher for every single group. This largely reflects progressive misalignment of the rural and urban poverty lines with the former becoming lower than the latter. It was this misalignment that led the Tendulkar Committee to revise the rural poverty line and realign it to the higher, urban line.

Table 3 reports the poverty estimates based on the Tendulkar lines. Recall that the Tendulkar line holds the urban poverty ratio at 25.7% in 2004–2005 when measuring poverty at MRP expenditures. Our urban poverty ratio in Table 3 reproduces this estimate within 0.1 of a percentage point.

The steady decline in poverty rates for the various social groups in rural as well as urban areas, which we noted based on the Lakdawala lines in Table 2 , remains valid at the Tendulkar lines. Moreover, rural poverty ratios turn out to be higher than their urban counterparts for each group in each year. As in Table 2 , the decline had been sharpest during the high-growth period between 2004–2005 and 2009–2010.

Finally and most importantly, the largest percentage-point decline between these years in rural and urban areas combined had been for the ST followed by the SC, OBC, and FC in that order. Given that scheduled tribes also had the highest poverty rates followed by scheduled castes and other backward castes in 2004–2005, the pattern implies that the socially disadvantaged groups have achieved significant catching up with the better-off groups. This is a major break with past trends.

Next, we report the national poverty rates by religious groups. In Table 4 , we show the poverty rates based on Lakdawala lines of rural and urban India and of the country taken as a whole. Three observations follow. First, at the aggregate level (rural plus urban), poverty rates show a steady decline for Hindus, Muslims, Christians, Jains, and Sikhs. Poverty among the Buddhists also consistently declined except for 1983 and 1987–1988. With one exception (Muslims in rural India between 1987–1988 and 1993–1994), the pattern of declining poverty rates between any two successive surveys also extends to the rural and urban poverty rates in the case of the two largest religious communities, Hindus and Muslims.

Second, going by the poverty rates in 2009–2010 in rural and urban areas combined, Jains have the lowest poverty rates followed by Sikhs, Christians, Hindus, Muslims, and Buddhists. Prosperity among Jains and Sikhs is well known, but not the lower level of poverty among Christians relative to Hindus. Also interesting is the relatively small gap of just 5.8 percentage points between poverty rates among Hindus and Muslims.

Finally, the impact of accelerated growth on poverty between 2004–2005 and 2009–2010 that we observed across social groups can also be seen across religious groups. Once again, we see a sharper decline in the poverty rate for the largest minority, the Muslims, relative to Hindus who form the majority of the population.

This broad pattern holds when we consider poverty rates by religious groups based on the Tendulkar line, as seen in Table 5 . Jains have the lowest poverty rates followed by Sikhs, Christians, Hindus, Muslims, and Buddhists. With one exception (Sikhs in rural India between 1993–1994 and 2004–2005), poverty had declined steadily for all religious groups in rural as well as urban India. The only difference is that the decline in poverty among Muslims in rural and urban areas combined between the periods 2004–2005 and 2009–2010 had not been as sharp as that estimated from the Lakdawala lines. As a result, we do not see a narrowing of the difference in poverty between Hindus and Muslims. We do see a narrowing of the difference in urban poverty but this gain is neutralized by the opposite movement in the rural areas due to a very sharp decline in poverty among Hindus, perhaps due to the rapid decline in poverty among scheduled castes and scheduled tribes.

Before we turn to poverty estimates by state, we should note that in this paper, we largely confine ourselves to reporting the extent of poverty measured based on the two poverty lines. Other than occasional references to the determinants of poverty such as growth and caste composition, we make no systematic effort to identify them. Evidently, many factors influence the decline in poverty. For instance, the acceleration in growth between 2004–2005 and 2009–2010 also led to increased revenue that made it possible for the government to introduce the National Rural Employment Guarantee Scheme under which one adult member of each rural household is guaranteed 100 days per year of employment at a pre-specified wage. The employment guarantee scheme may well have been a factor in the recent acceleration in poverty reduction.

In a similar vein, rural–urban migration may also impact the speed of decline of poverty. Once again, rapid growth, which inevitably concentrates disproportionately in urban areas, may lead to some acceleration in rural-to-urban migration. If, in addition, the rural poor migrate in proportionately larger numbers in search of jobs, poverty ratios could fall in both rural and urban areas. In the rural areas, the ratio could fall because proportionately more numerous poor than in the existing rural population migrate. In the urban areas, the decline may result from these individuals being gainfully employed at wages exceeding the urban poverty line. Migration may also reinforce the reduction in rural poverty by generating extra rural income through remittances. Evidence suggests that this effect may have been particularly important in the state of Kerala.

We now turn to the progress made in poverty alleviation in different states. Though our focus in this paper is on poverty by social and religious groups, we first consider poverty at the aggregate level in rural and urban areas. India has 28 states and 7 union territories. To keep the analysis manageable, we limit ourselves to the 17 largest states. 11 Together, these states account for 95% of the total population. We exclude all seven union territories including Delhi; the smallest six of the seven northeastern states (retaining only Assam); and the states of Sikkim, Goa, Himachal Pradesh, and Uttaranchal. Going by the expenditure survey of 2009–2010, each of the included states has a population exceeding 20 million while each of the excluded states has a population less than 10 million. Among the union territories, only Delhi has a population exceeding 10 million.

A.  Rural and Urban Populations

We begin by presenting the total population in each of the 17 largest states and the distribution between rural and urban areas as revealed by the NSSO expenditure survey of 2009–2010 (Table 6 ). 12 The population totals in the expenditure survey are lower than the corresponding population projections by the registrar general and census commissioner of India (2006) as well as those implied by Census 2011. 13 Our choice is dictated by the principle that poverty estimates should be evaluated with reference to the population underlying the survey design instead of those suggested by external sources. For example, the urban poverty estimate in Kerala in 2009–2010 must be related to the urban population in the state covered by the expenditure survey in 2009–2010 instead of projections based on the censuses in 2001 and 2011. 14

As shown in Table 6 , 27% of the national population lived in urban areas, while the remaining 73% resided in rural areas in 2009–2010. This composition understates the true share of the urban population, revealed to be 31.2% in the 2011 census. The table shows 10 states having populations of more than 50 million (60 million according to the 2011 census). We will refer to these 10 states as the “large” states. They account for a little more than three-fourths of the total population of India. At the other extreme, eleven “small” states (excluded from our analysis and therefore not shown in Table 6 ) have populations of less than ten million (13 million according to the Census 2011) each. The remaining seven states, which we call “medium-size” states, have populations ranging from 36 million in Orissa to 22 million in Chhattisgarh (42 million in Orissa to 25.4 million in Chhattisgarh, according to the 2011 census).

Among the large states, Tamil Nadu, Maharashtra, Gujarat, and Karnataka, in that order, are the most urbanized with a rate of urbanization of 35% or higher. Bihar is the least urbanized among the large states, with an urbanization rate of just 10%. Among the medium-size states, only Punjab has an urban population of 35%. The rest have urbanization rates of 30% or less. Assam and Orissa, with an urban population of just 10% and 14%, respectively, are the least urbanized medium-size states.

B.  Rural and Urban Poverty

We now turn to the estimates of rural and urban poverty in the 17 largest states. To conserve space, we confine ourselves to presenting the estimates based on the Tendulkar line. We report the estimates based on the Lakdawala lines in the Appendix. Recall that the estimates derived from the Tendulkar line are available for 3 years: 1993–1994, 2004–2005, and 2009–2010. Disregarding 1973–1974 and 1977–1978, which are outside the scope of our paper, estimates based on the Lakdawala lines are available for an additional 2 years: 1983 and 1987–1988.

Table 7 reports the poverty estimates with the states arranged in descending order of their populations. Several observations follow. First, taken as a whole, poverty fell in each of the 17 states between 1993–1994 and 2009–2010. When we disaggregate rural and urban areas within each state, we still find a decline in poverty in all states in each region over this period. Indeed, if we take the 10 largest states, which account for three-fourths of India's population, every state except Madhya Pradesh experienced a consistent decline in both rural and urban poverty. The reduction in poverty with rising incomes is a steady and nationwide phenomenon and not driven by the gains made in a few specific states or certain rural or urban areas of a given state.

Second, acceleration in poverty reduction in percentage points per year during the highest growth period (2004–2005 to 2009–2010) over that in 1993–1994 to 2004–2005 can be observed in 13 out of the total 17 states. The exceptions are Uttar Pradesh and Bihar among the large states and Assam and Haryana among medium-size states. Of these, Uttar Pradesh and Assam had experienced at best modest acceleration in gross state domestic product (GSDP) during the second period while Haryana had already achieved a relatively low level of poverty by 2004–2005. The most surprising had been the negligible decline in poverty in Bihar between 2004–2005 and 2009–2010, as GSDP in this state had grown at double-digit rates during this period.

Finally, among the large states, Tamil Nadu had the lowest poverty ratio followed by Andhra Pradesh and Gujarat. Tamil Nadu, Karnataka, and Andhra Pradesh—all of them from the south—made the largest percentage-point improvements in poverty reduction among the large states between 1993–1994 and 2009–2010. Among the medium-size states, Kerala and Haryana had the lowest poverty rates while Orissa and Jharkhand made the largest percentage-point gains during 1993–1994 to 2009–2010.

It is useful to relate poverty levels to per-capita spending. In Table 8 , we present per-capita expenditures in current rupees in the 17 states in the 3 years for which we have poverty ratios, with the states ranked in descending order of population. Ideally, we should have the MRP expenditures for all 3 years, but since they are available for only the last 2 years, we report the URP expenditures for 1993–1994. Several observations follow from a comparison of Tables 7 and 8 .

MRP = mixed reference period, URP = uniform reference period.

First, high per-capita expenditures are associated with low poverty ratios. Consider, for example, rural poverty in 2009–2010. Kerala, Punjab, and Haryana, in that order, have the highest rural per-capita expenditures. They also have the lowest poverty ratios, in the same order. At the other extreme, Chhattisgarh and Bihar have the lowest rural per-capita expenditures and also the highest rural poverty ratios. More broadly, the top nine states by rural per-capita expenditure are also the top nine states in terms of low poverty ratios. A similar pattern can also be found for urban per-capita expenditures and urban poverty. Once again, Kerala ranks at the top and Bihar at the bottom in terms of each indicator. Figure 3 offers a graphical representation of the relationship in rural and urban India in 2009–2010 using state level data.

Poverty and Per-capita MRP Expenditure in Rural and Urban Areas in Indian States, 2009–2010

One state that stands out in terms of low poverty ratios despite a relatively modest ranking in terms of per-capita expenditure is Tamil Nadu. It ranked eighth in terms of rural per-capita expenditure but fourth in terms of rural poverty in 2009–2010. In terms of urban poverty, it did even better, ranking a close second despite its ninth rank in urban per-capita expenditure. Gujarat also did very well in terms of urban poverty, ranking third in spite of the seventh rank in urban per-capita expenditure.

Finally, there is widespread belief that Kerala achieved the lowest rate of poverty despite its low per-capita income through more effective redistribution. Table 8 entirely repudiates this thesis. In 1993–1994, Kerala already had the lowest rural and urban poverty ratios and enjoyed the second highest rural per-capita expenditure and third highest urban per-capita expenditure among the 17 states. Moreover, in terms of percentage-point reduction in poverty, all other southern states dominate Kerala. For example, between 1993–1994 and 2004–2005, Tamil Nadu achieved a 27.4 percentage-point reduction in poverty compared to just 19.3 for Kerala. We may also add that Kerala experienced very high inequality of expenditures. In 2009–2010, the Gini coefficient associated with spending in the state was by far the highest among all states in rural as well as urban areas.

In this section we decompose population and poverty by social group. As previously mentioned, the expenditure surveys traditionally identified the social group of the households using a three-way classification: scheduled castes, scheduled tribes, and nonscheduled castes. However, beginning with the 1999–2000 survey, the last category had been further subdivided into other backward castes and the rest, the latter sometimes referred to as forward castes, a label that we use in this paper.

We begin by describing the shares of the four social groups in the total population of the 17 states.

A.  Population Distribution by Social Group within the States

Table 9 reports the shares of various social groups in the 17 largest states according to the expenditure survey of 2009–2010. We continue to rank the states according to population from the largest to the smallest.

Source: Authors’ calculations from the NSSO expenditure survey conducted in 2009–2010.

Nationally, the Scheduled Tribes constitute 9% of the total population of India according to the expenditure survey of 2009–2010. In past surveys and the Census 2001, this proportion was 8%. The scheduled castes form 20% of the total population according to the NSSO expenditure surveys, though the Census 2001 placed this proportion at 16%. The OBC are not identified as a separate group in the censuses so that their proportion can be obtained from the NSSO surveys only. The figure has varied from 36% to 42% across the three quinquennial expenditure surveys since the OBC began to be recorded as a separate group.

The scheduled tribes are more unevenly divided across states than the remaining social groups. In so far as these groups had been very poor at independence and happened to be outside the mainstream of the economy, ceteris paribus, states with high proportions of ST population may be at a disadvantage in combating poverty. From this perspective, the four southern states enjoy a clear advantage: Kerala and Tamil Nadu have virtually no tribal populations while Andhra Pradesh and Karnataka have proportionately smaller tribal populations (5% and 9% of the total, respectively) than some of the northern states which had high concentrations.

Among the large states, Madhya Pradesh, Gujarat, and Rajasthan have proportionately the largest concentrations of ST populations. The ST constitute 20%, 17%, and 14% of their respective populations. Some of the medium-size states, of course, have proportionately even larger concentrations. These include Chhattisgarh, Jharkhand, and Orissa with the ST forming 30%, 29%, and 22% of their populations, respectively.

Since the traditional exclusion of the SC has meant they began with a very high incidence of abject poverty and low levels of literacy, states with high proportions of these groups also face an uphill task in combating poverty. Even so, since the SC populations are not physically isolated from the mainstream of the economy, there is greater potential for the benefits of growth reaching them than the ST. This is illustrated, for example, by the emergence of some rupee millionaires among the SC but not the ST during the recent high-growth phase (Dehejia and Panagariya 2012 ).

Once again, at 9%, Kerala has proportionately the smallest SC population among the 17 states listed in Table 9 . Among the largest 10 states, West Bengal, Uttar Pradesh, Bihar, Rajasthan, and Madhya Pradesh have the highest concentrations. Among the medium-size states, Punjab, Haryana, and Orissa in that order have proportionately the largest SC populations.

The SC and ST populations together account for as much as 40% and 35%, respectively, of the total state population in Madhya Pradesh and Rajasthan. At the other extreme, in Kerala, these groups together account for only 10% of the population. These differences mean that, ceteris paribus, Madhya Pradesh, and Rajasthan face a significantly more difficult battle in terms of combating poverty than Kerala.

The ST populations also differ from the SC in that they are far more heavily concentrated in rural areas than in urban areas. Table 10 illustrates this point. In 2009–2010, 89% of the ST population was classified as rural. The corresponding figure was 80% for the SC, 75% for the OBC, and 60% for FC.

An implication of the small ST population in the urban areas in all states and in both rural and urban areas in a large number of states is that the random selection of households results in a relatively small number of ST households being sampled. The problem is especially severe in many of the smallest states where the total sample size is small in the first place. A small sample translates into a large error in the associated estimate of the poverty ratio. We will present the poverty estimates in all states and regions as long as a positive group is sampled. Nevertheless, we caution the reader on the possibility of errors in Table 11 that may be associated with the number of ST households in the 2009–2010 survey.

B.  Poverty by Social Group

We now turn to poverty estimates by social groups. We present statewide poverty ratios based on the Tendulkar line for the ST, SC, and nonscheduled castes in Table 12 . We present the ratios for the OBC and FC in Table 13 . As before, we arrange the states from the largest to the smallest according to population. Separate rural and urban poverty estimates derived from the Tendulkar lines and Lakdawala lines are relegated to the Appendix.

NS = non-scheduled, SC = scheduled castes, ST = scheduled tribes.

FC = forward castes, OBC = other backward castes.

With one exception, Chhattisgarh, the poverty ratio declines for each group in each state between 1993–1994 and 2009–1010. There is little doubt that rising incomes have helped all social groups nearly everywhere. In the vast majority of the states, we also observe acceleration in the decline in poverty between 2004–2005 and 2009–2010 compared to between 1993–94 and 2004–2005. Reassuringly, the decline in ST poverty among scheduled tribes and scheduled castes and SC poverty has sped up recently with the gap in poverty rates between these groups and the nonscheduled castes narrowing.

The negative relationship between poverty ratios and per-capita expenditures that we depicted in Figure 3 can also be observed for the social groups taken separately. Using rural poverty estimates by social group in the Appendix, we show this relationship between SC poverty and per capita rural expenditures in the left panel of Figure 4 and that between the ST poverty and per capita rural expenditures in the right panel. Figure 4 closely resembles Figure 3 . The fit in the right panel is poorer than that in the left panel as well as those in Figure 3 . This is partially because the ST are often outside the mainstream of the economy and therefore less responsive to rising per-capita incomes. This factor is presumably exacerbated by the fact that the number of observations in the case of the ST has been reduced to 11 due to the number of ST households in the sample dropping to below 100 in six of the 17 states.

Scheduled Caste and Scheduled Tribe Poverty Rates and Per-capita MRP Expenditures in Rural Areas, 2009–2010

For years 2004–2005 and 2009–2010, we disaggregate the nonscheduled castes into the OBC and FC. The resulting poverty estimates are provided in Table 13 . Taking the estimates in Tables 12 and 13 , one can see that on average poverty rates are at their highest for the ST followed by SC, OBC, and FC in that order. At the level of individual states, ranking of the poverty rates of scheduled castes and scheduled tribes is not clear-cut, but with rare exceptions, poverty rates of these two groups exceed systematically those of other backward castes, which in turn exceed rates of forward castes.

An interesting feature of the poverty rates of forward castes is their low level in all but a handful of the states. For example, in 2009–2010, the statistic computed to just 3.9% in Punjab, 5.9% in Kerala, 6.5% in Haryana, 6.9% in Tamil Nadu, and 10.5% even in Rajasthan. In 14 out of the largest 17 states, it fell below 25%. The states with low FC poverty rates generally also have low OBC poverty rates making the proportion of the SC and ST population the key determinant of the statewide rate.

This point is best illustrated by a comparison of poverty rates of Punjab and Kerala. Poverty rates for the nonscheduled caste population in 2009–2010 was 7.3% in Punjab and 10.4% in Kerala, while those for scheduled castes stood at 29.2% and 27.4%, respectively, in the two states. But since scheduled castes constitute 39% of the population in Punjab but only 9% in Kerala, statewide poverty rate turned out to be 15.8% in the former and 12% in the latter.

The caste composition also helps explain the differences in poverty rates between Maharashtra and Gujarat on the one hand and Kerala on the other. In 2009–2010, statewide poverty rates were 24.8% and 23.2%, respectively, in the former and 12% in the latter (Table 10 ). In part, the differences follow from the significantly higher per-capita expenditures in Kerala, as seen from Table 11 . 15 But Maharashtra and Gujarat also face a steeper uphill task in combating poverty on account of significantly higher proportions of the scheduled tribe and scheduled caste populations. These groups account for 17% and 11%, respectively, of the total population in Gujarat, and 10% and 15% in Maharashtra. In comparison, only 1% of the population comprises scheduled tribes in Kerala, while just 9% comprise scheduled castes (Table 9 ).

Finally, we turn to poverty estimates by religious group in the states. India is home to many different religious communities including Hindus, Muslims, Christians, Sikhs, Jains, and Zoroastrians. Additionally, tribes follow their own religious practices. Though tribal religions often have some affinity with Hinduism, many are independent in their own right.

Table 14 provides the composition of population by religious group as well as the rural–urban split of each religious group based on the expenditure survey of 2009–2010. Hindus comprise 82% of the population, Muslims 12.8%, Christians 2.3%, Sikhs 1.7%, Jains 0.3%, and Zoroastrians 0.016%. The remaining comprises just 0.3%.

Together, Hindus and Muslims account for almost 95% of India's total population. With 34% of the population in urban areas compared with 26% in the case of Hindus, Muslims are more urbanized than Hindus. Among the other communities, Jains and Zoroastrians are largely an urban phenomenon. Moreover, while Muslims can be found in virtually all parts of India, other smaller minority communities tend to be geographically concentrated. Sikhs cluster principally in Punjab, Christians in Kerala and adjoining southern states, Zoroastrians in Maharashtra and Gujarat, and Jains in Gujarat, Rajasthan, Karnataka, and Tamil Nadu.

Given their small shares in the total population and their geographical concentration, random sampling of households in the expenditure surveys yields less than 100 observations for minority religious communities other than Muslims in the vast majority of the states. Indeed, as Table 15 indicates, only 13 out of the 17 largest states had a sufficiently large number of households even for Muslims to allow poverty to be reliably estimated. Orissa, Haryana, Punjab, and Chhattisgarh each had fewer than 100 Muslim households in the survey. Thus, we attempt poverty estimates by religious groups in the states separately for Hindus and Muslims only. We do provide estimates for the catch-all “other” category but caution that, in many cases, these estimates are based on less than 100 observations and therefore subject to large statistical errors.

As before, we present the estimates for statewide poverty of the religious groups using the Tendulkar line, placing the more detailed estimates for rural and urban areas and estimates based on the Lakdawala lines in the Appendix. Table 15 reports the estimates for Hindus, Muslims, and other minority religion groups for the years 1993–1994, 2004–2005, and 2009–2010.

Religious groups replicate the broad pattern seen in the context of poverty by social group. Poverty has fallen in every single state between 1993–1994 and 2009–2010 for Hindus as well as for Muslims, though the change is not always monotonic. While the level of poverty in 2009–2010 is higher for Muslims than Hindus in the majority of the states, the reverse is true in Bihar, Tamil Nadu, Madhya Pradesh, and Karnataka. An anomaly is the marginal increase in the poverty rate between 2004–2005 and 2009–2010 in Bihar for Hindus and in Gujarat for Muslims. The observation is particularly surprising since we simultaneously observe a significant decline in poverty during the same period for Muslims in Bihar and for Hindus in Gujarat. Interestingly, as documented in the Appendix, poverty rates for both Hindus and Muslims decline in both states based on the Lakdawala lines between 2004–2005 and 2009–2010.

Although the focus of this paper is on poverty, we find it useful to briefly report the evolution of inequality at the state and national levels in rural and urban areas. At the outset, it is important to note that the issue of inequality is complex partly because it can be measured in numerous ways. 16 The potential list of measures is almost endless, and there is no guarantee that these different measures will move in the same direction. Therefore, it is quite easy to show simultaneously that inequality has risen as well as fallen depending on the choice of measure.

In this paper, we use one measure of overall inequality based on the same expenditure survey data we used to report poverty measures in the previous sections: specifically, the Gini coefficient of household expenditures in rural and urban areas in the 17 states and in India as a whole using URP expenditures in 1983, 1993–1994, 1999–2000, 2004–2005, and 2009–2010. Table 17 and Table 18 report the Gini coefficient in rural and urban areas, respectively. As before, we arrange the states in descending order of population size.

Source: Planning Commission website (accessed 4 February 2013).

An immediate observation from Tables 17 and 18 is that, with rare exceptions, rural inequality tends to be lower than urban inequality. At the national level in 2009–2010, the Gini coefficient was 0.291 in rural areas and 0.382 in urban areas. These values reflect a difference of 9 percentage points. This is not surprising. The vast majority of the villagers are small farmers or wage laborers. As a result, variation in their incomes and therefore expenditures are not large. In contrast, cities serve as home to much of the industry and formal sector services as well as to a large informal sector which attracts migrant workers. This results in greater variation in incomes and expenditures.

The tables show no clear trend in the Gini in rural areas but do show a tendency for it to rise in urban areas. At the national level, rural Gini fell between 1983 and 1999–2000, rose between 1999–2000 and 2004–2005, and fell again between 2004–2005 and 2009–2010, with a small net decline over the entire period. In contrast, the urban Gini has climbed steadily.

This is hardly surprising since rapid growth, which can produce increased inequality, is concentrated in urban areas. In the Indian case, a dualism of sorts exists within urban areas. Output growth has been concentrated in the formal sector, while employment has been disproportionately concentrated in the informal sector. Unlike the Republic of Korea and Taipei, China in the 1960s and 1970s and the People's Republic of China more recently, employment in the formal sector has not grown in India due to the poor performance of labor-intensive sectors. Growth in India has been concentrated in skilled labor and capital-intensive sectors.

The data do not support the hypothesis that high levels of poverty reflect high levels of inequality. At least in the Indian case, the two outcomes are at best unrelated and at worst negatively associated. For example, at the national level, rural inequality has remained more or less unchanged and urban inequality has risen, while both rural and urban poverty have steadily and significantly declined over time.

Looking at a cross section of the data, Kerala offers the most dramatic example. In 2009–2010, it had the lowest levels of rural and urban poverty and by far the highest rural and urban Gini coefficients. At the other extreme, Bihar had the second lowest rural Gini coefficient but the highest rural poverty ratio during the same period.

At a more aggregate level, the left panel in Figure 5 plots the rural Gini against the rural poverty ratio, while the right panel plots the urban Gini against the urban poverty ratio. The exponential trend line has a negative slope in each case, though the fit is poor. In other words, there is no evidence of a positive relationship between poverty and inequality, but there is some evidence of a negative relationship.

Gini Coefficients and Poverty Ratios in Rural and Urban Areas in Indian States, 2009–2010

In this paper, we have provided a comprehensive analysis of poverty in India along six different dimensions: across time, across states, between rural and urban areas, across social and religious groups, and based on two different poverty lines (Lakdawala and Tendulkar). To keep the exposition manageable, we have concentrated on estimates based on the Tendulkar line except when we discuss poverty at the national level. In the latter case, we report estimates in rural and urban India derived from both the Lakdawala and Tendulkar lines. Our detailed estimates by social and religious groups, by rural and urban areas, and by state based on both the Lakdawala and Tendulkar lines are provided in the Appendix.

The following are some of the key conclusions of the paper. First, poverty has declined between 1993–1994 and 2009–2010 along every dimension. Indeed, poverty has fallen for every social and religious group in every state and in rural and urban areas, separately as well as jointly. Estimates based on the Lakdawala line show that the decline can be observed steadily since 1983 for all social and religious groups in all 17 large states.

Second, acceleration in growth rates between 2004–2005 and 2009–2010 has been accompanied by acceleration in poverty reduction. Poverty rates have fallen rapidly for all major social and religious groups at the national level. This phenomenon also holds true for most states across various social and religious groups.

Third, for the first time, poverty reduction between 2004–2005 and 2009–2010 has been larger for the scheduled castes and scheduled tribes than the upper caste groups. Thus, the gap in poverty rates between the socially disadvantaged and upper caste groups has narrowed over time. This pattern provides clear evidence to refute the claim that reforms and growth have failed to help the socially disadvantaged or that they are leaving these groups behind. A continuation of this trend, bolstered by further reforms and higher growth rates, would help eliminate the difference in poverty rates between the historically disadvantaged and the privileged.

Fourth, interstate comparisons reveal that the states with large scheduled castes and scheduled tribe populations face a steeper climb in combating poverty. The point is most forcefully brought out by a comparison of Punjab and Kerala. When we compare poverty rates in 2009–2010 by social group, the two states have very similar poverty rates. But because the poverty rates for the scheduled castes are higher than those for the nonscheduled castes in both states and the scheduled castes account for a much larger proportion of the population, the aggregate poverty rate in Punjab turns out to be significantly higher.

Finally, we find that in the case of India, there is no robust relationship between inequality and poverty. Indeed, to the extent that such a relationship exists, this would suggest that more unequal states enjoy lower levels of poverty. Kerala offers the most dramatic example. It has had one of the highest Gini coefficients for rural as well as urban areas and also one of the lowest poverty ratios for both regions. In 2009–2010, its Gini coefficients were by far the highest among the large states in both rural and urban areas, while its poverty ratios were the smallest.

Given space limitations, we have deliberately limited ourselves to providing one specific indicator of poverty—the headcount ratio—in different states and for different social and religious groups based on the two official poverty lines. There are at least two broad complementary directions in which the work in this paper can be extended.

First, it may be desirable for certain purposes to estimate alternative indicators of poverty such as the poverty gap or its close cousin, the Foster-Greer-Thorbecke index. Such an index allows one to gauge the resources needed to bring all those below the poverty line to a level above it. In a similar vein, we have focused on progress in combating poverty among social and religious groups that are more vulnerable. Alternatively, we could focus on a different dimension of vulnerability such as male-headed versus female-headed households and evaluate the progress in combating poverty among female-headed households.

The second direction in which the work of this paper could be extended is towards explaining the determinants of poverty. Within this broad category, we have left many questions unanswered. For instance, it would be useful to separate the contributions of growth and redistribution policies in explaining the decline in poverty. Likewise, we may want to know what role, if any, rural-to-urban migration may have played—directly as well as through remittances. Similarly, we might ask what role the division of population among various social and religious groups plays in determining the progress in combating poverty. Finally, we might also wish to study the role that education plays in bringing down poverty. The recent work by Hnatkovska and Lahiri ( 2012 ) shows that education has indeed been pivotal in bridging the wage gap between scheduled castes and scheduled tribes on the one hand and nonscheduled castes on the other. This suggests an important role for education in eradicating poverty.

a Calculated by adjusting the 2004–2005 lines using the index implicit in the official Tendulkar lines for 2004–2005 and 2009–2010.

Source: Planning Commission, Government of India, Data Tables.

SC = scheduled castes, ST = scheduled tribes, URP = uniform reference period.

a Delhi is 95% urban. The SC and ST estimates in this case are based on too few households and therefore subject to substantial sampling errors.

FC = forward castes, NC = nonscheduled castes, OBC = other backward castes.

a Only 5% of Delhi by population is rural. SC and ST estimates in this case are based on too few households and therefore subject to substantial sampling errors.

FC = forward castes, NS = nonscheduled castes, OBC = other backward castes, URP = uniform reference period.

MRP = mixed reference period, SC = scheduled castes, ST = scheduled tribes.

Source: Authors’ calculations

FC = forward castes, MRP = mixed reference period, NS = nonscheduled castes, OBC = other backward castes.

URP = uniform reference period.

MRP = mixed reference period.

The views expressed in the paper are those of the authors and not of the World Bank. We thank an anonymous referee, P. V. Srinivasan, and participants of the first 2013 Asian Development Review conference held on 25–26 March 2013 at the Asian Development Bank headquarters in Manila, Philippines.

Prices could vary not just between urban and rural regions within a state but also across subregions within rural and subregions within urban regions of a state. Therefore, in principle, we could envision many different poverty lines within rural and within urban regions in each state. To keep the analysis manageable, we do not make such finer distinctions in the paper.

We do not go into the sources of underestimation of expenditures in NSSO surveys. These are analyzed in detail in Government of India ( 2008 ). According to the report (Government of India 2008 , p. 56), “The NSS estimates suffer from difference in coverage, underreporting, recall lapse in case of nonfood items or for the items which are less frequently consumed and increase in nonresponse particularly from affluent section of population. It is suspected that the household expenditure on durables is not fully captured in the NSS estimates, as the expensive durables are purchased more by the relatively affluent households, which do not respond accurately to the NSS surveys.” Two items, imputed rentals of owner-occupied dwellings and financial intermediation services indirectly measured, which are included in the NAS estimate, are incorporated into the NSSO expenditure surveys. But these account for only 7–9 percentage points of the discrepancy.

This section is partially based on Panagariya ( 2011 ).

See, for example, the report by the NDTV entitled “Planning Commission further lowers poverty line to Rs28 per day.” Available: http://www.ndtv.com/article/india/planning-commission-further-lowers-poverty-line-to-rs-28-per-day-187729

For instance, one commentator argued in a heated television debate that since bananas in Jor Bagh (an upmarket part of Delhi) cost Rs60 a dozen, an individual could barely afford two bananas per meal per day at poverty line expenditure of Rs32 per person per day.

We thank Rahul Ahluwalia for supplying us with Table 1 . The expenditures in the table represent the average of the urban decile class including the urban poverty line. Since the urban poverty line is at 25.7% of the population, the table takes the average over those between the 20th and 30th percentile of the urban population.

The consumption figures for the top 30% of the population are from Ganesh-Kumar et al. ( 2012 ).

These amounts of cereal and pulses equal or exceed their mean consumption levels according to the 2004–2005 NSSO expenditure survey.

Recently, Panagariya ( 2013 ) has suggested that if political pressures necessitate shifting up the poverty line, the government should opt for two poverty lines in India—the Tendulkar line, which allows it to track those in extreme poverty, and a higher one that is politically more acceptable in view of the rising aspirations of the people.

The issue is discussed at length in Bhagwati and Panagariya ( 2013 ).

Although Delhi has its own elected legislature and chief minister, it remains a union territory. For example, central home ministry has the effective control of the Delhi police through the lieutenant governor who is the de jure head of the Delhi government and appointed by the Government of India.

Our absolute totals for rural and urban areas of the states and India in Table 6 match those in Tables 1A-R and 1A-U, respectively, in Government of India ( 2011b ).

The Planning Commission derives the absolute number of poor from poverty ratios using census-based population projections. Therefore, the population figure underlying the absolute number of poor estimated by the Planning Commission are higher than those in Table 6 , which are based on the expenditure survey of 2009–2010.

This distinction is a substantive one in the case of states in which the censuses reveal the degree of urbanization to be very different from that underlying the design of the expenditure surveys. For example, the expenditure survey of 2009–2010 places the urban population in Kerala at 26% of the total in 2009–2010, but the census in 2011 finds the rate of urbanization in the state to be 47.7%.

This is true in spite of significantly higher per-capita GSDP in Maharashtra presumably due to large remittances flowing into Kerala. According to the Government of India ( 2011a ), one in every three households in both rural and urban Kerala reports at least one member of the household living abroad.

For instance, inequality could be measured as the ratio of the top 10% to bottom 10% of the population, the ratio of rural to urban per-capita incomes, the ratio of skilled to unskilled wages (or formal and informal sector wages), and through the Gini coefficient (nationally or across states).

Email alerts

Affiliations.

• Online ISSN 1996-7241
• Print ISSN 0116-1105

A product of The MIT Press

Mit press direct.

• About MIT Press Direct

Information

• Accessibility
• For Authors
• For Customers
• For Librarians
• Direct to Open
• Open Access
• Media Inquiries
• Rights and Permissions
• For Advertisers
• About the MIT Press
• The MIT Press Reader
• MIT Press Blog
• Seasonal Catalogs
• MIT Press Home
• Give to the MIT Press
• Direct Service Desk
• Terms of Use
• Privacy Statement
• Crossref Member
• COUNTER Member  
• The MIT Press colophon is registered in the U.S. Patent and Trademark Office

This Feature Is Available To Subscribers Only

Sign In or Create an Account

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Prevalence and correlates of multidimensional child poverty in India during 2015–2021: A multilevel analysis

Contributed equally to this work with: Jalandhar Pradhan, Soumen Ray, Monika O. Nielsen, Himanshu

Roles Conceptualization, Investigation, Methodology, Supervision, Writing – review & editing

* E-mail: [email protected]

Affiliation Department of Humanities and Social Sciences, National Institute of Technology Rourkela, Odisha, India

Roles Conceptualization, Methodology, Writing – review & editing

Affiliation UNICEF State Office, Bhubneshwar, Odisha, India

Roles Conceptualization, Supervision, Writing – review & editing

Roles Conceptualization, Data curation, Formal analysis, Methodology, Software, Writing – original draft

• Jalandhar Pradhan, 
• Soumen Ray, 
• Monika O. Nielsen, 

• Published: December 22, 2022
• https://doi.org/10.1371/journal.pone.0279241
• Peer Review
• Reader Comments

Despite increasing research and programs to eradicate poverty, poverty still exists and is a far greater concern for children than adults, leading child poverty to become a political, economic, and social issue worldwide and in India. The current study aims to find variations in the prevalence of child poverty and associated factors in India during 2015–21. In the current study, we used two consecutive rounds of the National Family Health Survey (NFHS-4, 2015–16 & NFHS-5, 2019–21) to estimate child poverty (aged 0–59 months) using the Alkire-Foster method. The multilevel logistic regression analyses were performed to find the important cofounder and cluster level variation in child poverty. The results show that about 38 percent of children were multidimensionally poor in 2015–16, which reduced to 27 percent in 2019–21. The decomposition analysis suggests that contribution of nutrition domain to child poverty increases over time, whereas the standard of living substantially declines from NFHS-4 to NFHS-5. The multilevel analysis results show that the age and sex of the child, age and years of schooling of the mother, children ever born, religion, caste, wealth quintile and central, northeast, north and west regions are significantly associated with child poverty over time. Further, the variance participation coefficient statistics show that about 12 percent of the variation in the prevalence of child poverty could be attributed to differences at the community level. The prevalence of child poverty significantly declines over time, and the community-level variation is higher than the district-level in both surveys. However, the community-level variation shows increases over time. The finding suggests a need to improve the nutritional status and standard of living of most deprived households by promoting a child-centric and dimension-specific approach with more focus on PSU-level intervension should adopt in order to lessen child poverty in India.

Citation: Pradhan J, Ray S, Nielsen MO, Himanshu (2022) Prevalence and correlates of multidimensional child poverty in India during 2015–2021: A multilevel analysis. PLoS ONE 17(12): e0279241. https://doi.org/10.1371/journal.pone.0279241

Editor: Faisal Abbas, National University of Sciences and Technology, PAKISTAN

Received: July 13, 2022; Accepted: December 3, 2022; Published: December 22, 2022

Copyright: © 2022 Pradhan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: https://www.dhsprogram.com/methodology/survey/survey-display-541.cfm .

Funding: The author(s) received no specific funding for this work.

Competing interests: The information of this document, however expresses author’s personal observation and opinions and does not necessarily represent UNICEF’s or NITR’s position. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Introduction

Globally about 356 million children (0–17 aged) live in impoverished households, consisting of 107 million children under the age of five in 2017. Again, children are even more susceptible to extreme poverty, where 17.5 percent of them live in extreme poverty, compared to an estimated 7.9 percent of adults [ 1 ]. The Sustainable Development Goal (SDGs) target 1.2 suggests the international community "reduce at least half the proportion of men, women and children of all ages living in poverty in all its dimensions through national definitions by 2030" [ 2 ]. The significance of SDG target 1.2 is important, as children are first time, included in the poverty goal worldwide; focus on the multidimensional nature of poverty and poverty goal clearly to the national definition.

The global distribution of poverty is unequal, and defining poverty is a significant challenge. Although it’s extensive economic success, Asia remains the world’s poorest continent, with more than half of the world’s impoverished people living there. Further, due to deep inequality in the south Asia region, the children are trapped in the vicious cycle of poverty and discrimination at different levels and phases, such as nutrition, health, sanitation and lag behind universal education.

The history of measuring or identifying poverty is very old for developed and developing nations [ 3 , 4 ]. The traditional poverty measure was the unidimensional measure of well-being and was solely based on the minimum income or expenditure needed to maintain a subsistence level. Academic research conducted by sociologists and economists demonstrates that poverty is more than related to insufficient to feed someone or family [ 5 ]. Further, the Amartya Sen capability approach (1997) introduces the principle of social justice and well-being, a major contribution to identified poor (based on development). Sen’s approach to well-being consisting two significant components 1) functioning in regards to states and actions in which individuals wish to live and 2) capacity, which refers to the possibility that the person is equipped to exercise their freedom of choice concerning different possible runs [ 6 ].

Despite these advancements, many national and international poverty measures depend on the minimum absolute measure. However, notable changes have been made to the definition and measurement of poverty in terms of the complex nature of poverty to use non-monetary or relative measures of poverty in low and middle-income countries [ 7 – 9 ]. There has been great discourse overuse of welfare outcome indicators presented as living standards in the context of poverty. Measurement as dwelling quality, overcrowding, access to water, sanitation, healthcare and education are utilised constantly to define poverty [ 10 , 11 ]. The latest development in measuring poverty is the Multidimensional Poverty Index (MPI), constructed by the Oxford Poverty and Human Development Initiative (OPHI) in collaboration with United Nations Development Program (UNDP). The MPI covers more than 100 developing countries by using individual and household level data about health (child mortality in households and nutrition), education (school attendance and years of schooling), and standard of living (electricity, flooring, drinking water, sanitation, cooking fuel and assets) [ 10 ].

In line with the provided method for measuring poverty, child poverty measurement is still in the development process [ 12 – 22 ]. Child poverty is often considered the children living in income or consumption-based below-poverty-line households. However, it is widely recognised that the household-based monetary indicator cannot capture child poverty [ 23 ]. Research has concluded that the unidimensional approach to majoring in poverty can not capture the depth of child poverty as the child’s need is different from adults [ 24 ]. The United Nations Convention on Child Rights (CRC) also maintains this idea for children’s well-being for the betterment and adequate standard of living [ 25 ]. The multifaceted approach is necessary to measure child poverty with CRC welfare dimensions and indicators.

Children in extreme poverty are affected differently from adults, mainly by inadequate nutrition, exposure to stress, and lack of early learning, resulting in lifetime poverty. Further, the adults have direct access to many things that may help overcome the poverty state, whereas the children solely depend on their adult family members for support, care and satisfaction of their basic needs [ 26 ]. Mounting evidence shows that healthy children are more likely to become healthy adults. Many child deprivation indexes were constructed using different domains and indicators in the developed nations, including material well-being, health, education, crime, housing, environment, family economic well-being, social relationship, economic security, exposure to risk and risky behaviour of children in need [ 27 – 29 ].

UNICEF developed the Multiple Overlapping Deprivation Analysis (MODA) to provide instruments for the multidimensional aspect of child poverty in terms of deprivation. MODA adopt a holistic approach to child well-being, which cannot be tackled in the sector (e.g. health, nutrition, and education), as deprivation is multifaceted and interrelated and has more adverse effects. MODA mainly focused on the child as the unit of analysis rather than the household and kept the child’s life cycle approach for a different child group has another need. In MODA, there are two steps for calculating deprivation. In the first stage, the deprivation has been identified further multiple overlapping deprivations calculated to find the combination of deprivation faced by the child. However, MODA has its weight limitation due to assuming equal weight to all dimensions. In that way, a child would consider deprived in a particular dimension if he/she has been deprived of any one indicator from that dimension.

However, little known about child poverty in the context of developing countries [ 13 – 15 , 20 , 22 ]. Gordon et al. have identified eight domains of severe child deprivation, including food, safe drinking water, sanitation, health, shelter, education, information and access to services [ 29 , 30 ]. A study based in Burkina Faso identified seven domains for child poverty for children aged 5–18. Countries analyse multidimensional poverty using various indicators and unit of analysis depending on how poverty is perceived [ 31 – 37 ].

Despite the growth in child poverty research around the globe, few studies have been conducted primarily focused on child deprivation, child poverty, or child well-being in India. Dutta (2020) utilised the MODA framework with nine dimensions, including nutrition, health, education, child protection, water, sanitation, housing, indoor air pollution and information for estimating child deprivation in India and Bangladesh, keeping the lifecycle approach [ 38 ]. In contrast, Chaurasia A. R. (2016) constructs a child deprivation index using five domains. This analysis is based on the Rapid Survey of Children (RSoC) 2013–14 for children below age 18, and the domains were survival, growth, education, protection and environment [ 39 ].

Even though child poverty research has gained attention, still child poverty is considered for children below the age of 18 [ 13 – 15 , 20 , 22 ]. However, it is well documented that every child group has their own need, and this need changes with the child’s growth. Children under the age of five constitute a considerable proportion of India’s population [ 40 ]. However, hardly any study explores the changes over time in child poverty and PSU-level variation in the multidimensional child poverty estimated using the standard measurement. Hence, this study aims to measure child poverty in India for 2015–16 and 2019–21 using Alkire and Foster’s multidimensional approach and later explore the associated factor with cluster effect on multidimensional child poverty among children under five years.

Data and methods

The present studies utilized two solely independent cross-sectional data from National Family Health Survey, namely NFHS-4 (2015–16) and NFHS-5 (2019–21). The survey provides essential information on crucial population and health indicators, including fertility, mortality, maternal, child and adult health, women and child nutrition, family welfare, and emerging issues like non-communicable diseases for India and each States/Union Territories. NFHS-4 survey first time provided district-level estimates for many crucial indicators, which expanded the sample size nearly six-fold than NFHS-3, and collected information from 601509 households, 699686 women and 103525 men. NFHS-4 adopted a two-stage sampling design in rural and urban areas of India to provide district-level estimates from 28583 primary sampling units (PSU) composed of village rural areas and census enumeration blocks (CEB) in urban areas from 640 districts of India [ 41 ].

NFHS-5 survey also provided district-level (707 districts) estimates for many crucial indicators and aligned with Sustainable Development Goals (SDGs) for preparing the database for monitoring government programmes and their progress toward achieving the SDGs by 2030. NFHS-5 collected information from 636699 households, 724115 women and 101839 men. NFHS-5 adopted a two-stage sampling design in rural and urban areas of India to provide district-level estimates from 30198 primary sampling units (PSU) composed of village rural areas and census enumeration blocks (CEB) in urban areas from 707 districts of India [ 42 ]. The present research utilised data from childbirth that took place five years before the survey date. To fulfil the study’s overall objective, we exclusively analyzed data of 213623 (NFHS-4, 2015–16) and 192292 (NFHS-5, 2019–21) children aged 0–59 months after eliminating pairwise missing information.

Outcome variable

The primary outcome variable for this study is the multidimensional poverty index, as poverty can not be measured with conventional methods based on money. A three-dimension consisting of nine indicators was used to measure multidimensional child poverty and deprivation based on the SDG’s goal and the availability of relevant data for the countries. The dimension are health, nutrition and living standard, and indicators are wasting underweight, immunization, child mortality in households, housing, water, sanitation, clean fuel, and information presented below Table 1 with their relevant weight.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0279241.t001

The nutrition recommended nutritional assessment was constructed using WHO-Anthro to convert weight, height and child age (months) in weight-for-age (WAZ) z-score for underweight and weight-for-height Z-score (WHZ) for wasting. Children whose WAZ and WHZ were -2 standard deviations (-2 SD) from the median of the reference population were identified as underweight and wasting, respectively.

Concerning the health domain, two indicators were used immunization and child mortality. A child is considered to be deprived if he did not receive all basic vaccination, including (BCG, DPT, polio and measles), living in a household that reported under-five child mortality in the past five years prior to the survey. Both surveys estimated the full immunisation rate for 12–35 month children.

The standard of living indicators was identified as the third dimension for MPI calculation, which included five indicators: housing, drinking water, sanitation, cooking fuel, and information.

Three items measured the housing indicators: material used in the construction of the floor, wall, and roof. It is evidenced that housing condition has profound health implications on children; in some cases, it is found more than in adults. A child is considered deprived if they live in a house with dirt, sand or dug, floor or wall and roof made of natural or rudimentary materials.

The provision of clean drinking water is one of the most basic indicators for improvement in marginalized or impoverished families [ 43 ]. A child is considered deprived if the household uses an unsafe drinking water source (unprotected well, unprotected spring, and surface water of river or lake) and the water source takes more than 30 minutes to collect and return home.

Poor and unimproved sanitation plays a role in deteriorating child health, which may result in premature death in some cases [ 43 ]. Children are deprived if they have no toilet facility, share toilet, use unimproved pit latrines, or practice open defecation.

Indoor air pollution is a potential source of health risks, such as acute respiratory infections in childhood and chronic obstructive pulmonary disease. A child is considered deprived if he or she is exposed to indoor air pollution caused by solid and fossil cooking fuels inside the home.

Children and individuals need media and information to enhance their intellect and identify information sources. Therefore, it is necessary for a child should live in a household with access to mass media exposure. A child is classified as deprived in information indicators if the child lives with a mother and has not been exposed to mass media, including radio/newspaper, television and radio.

Construction of MPI.

The multidimensional poverty indices are estimated by Alkire and Foster (AF) method. This approach provides data on various demographic accomplishments without requiring sorting or prioritization. Instead, they complement each other. The AF method offers many key decisions to the researcher, including identifying the unit of analysis, dimension, deprivation cutoff (for determining when an individual is deprived in a dimension), weights (for indicating the relative importance of various deprivation), and poverty cutoff (determining when a person is considered poor based on the amount of deprivation they experience). Due to its adaptability, the methodology can be applied in a wide variety of contexts, although its primary use has been in assessing multidimensional forms of poverty [ 21 , 44 ]. The AF method utilized a dual cutoff method to recognized the poor for each indicator and aggregated them according to different dimensions. As a result, the multidimensional poverty index can be decomposed into specific dimensions and indicators, which will help support evidence-based planning by focusing on specific dimensions and indicators. The weight of the dimension is equal, and then each indicator within each dimension is equally weighted. Thus, three types of estimates are generated- the percentage of headcount poverty (H), the intensity of poverty (A) and the multidimensional poverty index (M 0 ).

Where ’q’ represents the number of multidimensionally poor people, and ’n’ represents the total population of the study.

Where ’c’ is the poor experienced deprivation score and the intensity of poverty is a weighted average deprivation experienced by the multidimensionally poor.

Thus, MPI results from the proportion of multidimensionally poor and the intensity of poverty.

Cofounders.

The independent variable for the present study consists of socioeconomic, household, child, and mother-level factors. These factors included child age in months, child sex, female-headed household, age of mother, education level of the mother, children ever born, place of residence, religion, caste, wealth quintile (based on asset holding of household), and region (29 states and 8 UTs divided into total six regions).

Statistical analysis

The multidimensional child poverty prevalence and association with relevant cofounders have been presented in percentages. The chi-square test with a 5% significance level has been used to show the statistical association between a categorical variable and multidimensional child poverty. A multilevel logistic regression model with a random intercept was used to understand the clustering of the respondents within the district and the PSUs or the ’community’ level. Multilevel models are particularly appropriate and used for research designs where data are structured at more than one level, for example, village level, community level and state level [ 45 ]. The P-value of <0.05 is considered statistically significant. The multicollinearity test was conducted prior to multilevel analysis, and the variance inflation factor (VIF) value was found under the permissible limit of two.

Multilevel Logistic Regression (MLR).

The multistage sampling design is characterized as the sample drawn from such a population with a hierarchical structure. Therefore, the stratified multistage sample became the norm in the sociological and demographic survey mainly due to cost, time and efficiency. For such type of sample, the data clustering should be taken into consideration during data analysis. In the present study, NFHS-4 & NFHS-5 datasets, the individual (level 1) are nested within the PSU (Level 2), which is nested within the district (Level 3). Multilevel analysis with three levels has been utilized to identify the important cofounders of child poverty at the individual, PSU and district levels (i.e. Fixed part). The multilevel logistic regression analysis allows for partitioning the variation in the outcome variable (i.e. poor child) measured at the individual level. The variance can be attributed to individual variations at the PSU and district levels [ 46 – 48 ].

Ethics statement.

The NFHS-5 survey was conducted by the International Institute for Population Sciences (IIPS), Mumbai and received necessary ethical approval from the relevant ethics boards. We did not obtain additional ethical approval or informed consent because we accessed the anonymized NFHS-5 data available in the public domain at https://dhsprogram.com/data/available-datasets.cfm .

Descriptive statistics and prevalence of child poverty

The socioeconomic and demographic characteristics of children 0–59 months used in the analysis are presented in Table 2 . More than 60 percent of children belong to the 24–59 month age group, and 48 percent of the sample was a girl in both surveys. About 12 percent (NFHS-4) and 15 percent (NFHS-5) of children live in a female-headed household. About 31 percent of women were illiterate in NFHS-4 as compared to 22 percent in NFHS-5. The sampling distribution is quite similar in both surveys.

https://doi.org/10.1371/journal.pone.0279241.t002

Overall, about 38 percent child was multidimensional poor (MDP) by using a global 33 percent cutoff in 2015–16, which declined to 27 percent in 2019–21, showing an 11 percent points decline over time. The MPI was estimated at 0.178 in NFHS-4 and 0.120 in NFHS-5 ( S1 Table ). Forty-five percent (NFHS-4) and 31 percent (NFHS-5) of the children were MDP in the 12–23 month age group, and the MDP is higher among male children in both surveys. In addition, the child’s MDP was significantly higher in the female-headed household.

The child’s MDP significantly declined with higher maternal education levels in both surveys, and illiterate mothers reported higher MDP children (57% and 45% in NFHS-4 & NFHS-5, respectively). Similarly, children whose mothers have 6-plus children reported higher MDP (60% and 44% in NFHS-4 and NFHS-5, respectively). The MDP was about two-fold in rural residing children than in urban. Similarly, scheduled tribe children reported much higher MDP than any other castes in both surveys. The MDP declined with a higher wealth quintile and was lower in the richest wealth quintile (13% in NFHS-4 and 11% in NFHS-5). While making the regional comparisons, the measured MDP was found to be higher in the central region (47%), followed by the east (46%) and northeast region (37%), whereas the south region (22%) showed lower MDP children in NFHS-4. In contrast, the pattern in a slight change in NFHS-5 found the east (36%) and northeast (31%) regions children showing higher poverty, whereas the north (17%) and south (15%) regions reported lower child MDP.

State differential in multidimensional child poverty in India

The state differential with changing prevalence of multidimensional child poverty in two rounds of the national survey, i.e. NFHS-4 and NFHS-5, is portrayed in Fig 1 . Data illustrate that overall, child poverty declined significantly from NFHS-4 to NFHS-5. It shows that 27 percent of the child was MDP in NFHS-5 (2019–21), which was about 38 percent in NFHS-4 (2015–16) survey at the national level, which showed about 11 percent point decline in child poverty between these two surveys. The lowest level of child poverty was measured in the state/UTs of Puducherry (7.6%), followed by Sikkim (8%), Mizoram (8.8%), Punjab (9.7%) and Delhi (10.3%), where the highest level of child poverty was measured in the states Bihar (42.5%), followed by Jharkhand (41.3%), Assam (35.3%), Madhya Pradesh (33.1%), and Gujrat (29.8) in the lasted NFHS-5 (2019–21) survey. However, in the NFHS-4, Sikkim (8.6%) state reported the lowest child MDP, followed by Kerala (9.3%) and Chandigarh (10%), whereas Jharkhand (58%) reported a higher MDP child followed by Bihar (52.3%) and Madhya Pradesh (52.2%).

https://doi.org/10.1371/journal.pone.0279241.g001

The most significant change has been observed in Rajasthan, where child poverty declined about 19 percent points from 41.7 percent to 22.4 percent in NFHS-4, followed by Madhya Pradesh (19.1 percentage points), Jharkhand (17 percentage points), Uttar Pradesh (16.1 percentage points), and Chhattisgarh (15.1 percentage points). Similarly, the lowest change has been observed in Sikkim (0.7 percentage points), followed by Goa (1.4 percentage points) and Mizoram (2.6 percentage points). However, some states/Uts like Chandigarh (2.1 percentage points), Kerala (1.9 percentage points), Lakshadweep (1.0 percentage points), Nagaland (0.7 percentage points), and Himachal Pradesh (0.3 percentage points) showed a slight increase in child poverty during NFHS-4 to NFHS-5.

Decomposition of multidimensional child poverty in India

Multidimensional child poverty was decomposed to understand the contribution of the domain and various indicators in multidimensional child poverty in India in NFHS-4 & NFHS-5. Table 3 shows that among nine indicators, the underweight contributed the highest (about 30%) to multidimensional child poverty, followed by wasting (21%) in NFHS-5. A similar pattern was also observed in the NFHS-4 survey, where the highest contributor to multidimensional poverty was underweight (27.5%), followed by wasting (17%). Drinking water and child mortality in households contributed the least to multidimensional poverty at 2 percent and 2.3 percent, respectively, in NFHS-5.

https://doi.org/10.1371/journal.pone.0279241.t003

The domain-wise contribution shows that the nutrition domain (44.5%) contributed the most, with a slightly lower contribution from the standard of living domain (43.6%), whereas the health domain (11.9%) contributed the least to multidimensional child poverty in NFHS-4. Interestingly, a similar pattern was observed in the NFHS-5 survey, where the nutritional domain contributes the highest (51%) to multidimensional poverty, followed by living standard (39%) and the least from the health domain (10%).

Predictor of multidimensional child poverty in India

The results from multilevel logistic regression show correlate of multidimensional poverty among children 0–59 months by socioeconomic and demographic characteristics are presented in Table 4 . Results show that that child age, children ever born by mother, Muslim religion, caste, and north, central and west region are more likely to have multidimensional poverty compared to their reference categories. Children from 12–23 months [AOR: NFHS-4 = 2.22; NFHS-5 = 1.71] and 24–59 month [AOR: NFHS-4 = 1.37; NFHS-5 = 1.20], were more likely to experience MDP child than 0–11 month children. The rural residence children [AOR = 1.12] are more likely to have multidimensional poverty than urban residence children in NFHS-4. Similarly, children belonging to scheduled caste [AOR: NFHS-4 = 1.15; NFHS-5 = 1.14], scheduled tribes [AOR: NFHS-4 = 1.28; NFHS-5 = 1.22] and other backward castes [AOR: NFHS-4 = 1.09; NFHS-5 = 1.08] were more likely to be MDP compared to others caste children. additionally, children belonging to central region [AOR: NFHS-4 = 1.57; NFHS-5 = 1.41], east region [AOR: NFHS-4 = 1.13; NFHS-5 = 1.41] and west region [AOR: NFHS-4 = 1.58; NFHS-5 = 1.92] were more likely to be MDP compared to the north region children.

https://doi.org/10.1371/journal.pone.0279241.t004

Moreover, female children, mother age, maternal education, belonging to other religions and higher wealth quintile and south region are showing lower odds of having child poverty compared to their respective reference group. The female children [AOR: NFHS-4 = 0.93; NFHS-5 = 0.89] have showing 7% and 11% less likely to be MDP compared to male children in NFHS-4 & NFHS-5, respectively. Similarly, children from the northeast region [AOR: NFHS-4 = 0.72; NFHS-5 = 0.90] region were28% and 19% less likely to be MDP compared to the north region of India in NFHS-4 & NFHS-5, respectively.

Later the multilevel analysis finds the variation in child poverty between district and PSU levels in India. It observed that the child poverty variation is declining during NFHS-4 to NFHS-5, i.e. variance partition coefficient (VPC) at the district and PSU level, which contributes 4.4% and 10.7 percent, respectively, in NFHS-4 to the total variation in the child poverty prevalence. The variance partition coefficient (VPC) in child poverty prevalence during NFHS-5 declined to 2.7 percent at the district level, whereas it increased at the PSU level (11.9%).

Research on multidimensional poverty has been conducted globally and is also available in India [ 51 ]. However, this research mainly focused on all age groups and treated children as poor if they belonged to deprived households. Limited research has been conducted on the child poverty aspect in India. Although some workers have been carried out for children 0–17 age group using the MODA framework [ 38 ], MODA has its limitation as it depends upon the deprivation in the dimension where any child could be deprived if any indicator from a dimension deprived it considers that the child would be deprived on that dimension too.

Using the Alkire Foster method, this is the first study to estimate child poverty (under-five age group). The data of 213623 and 190916 children aged 0–59 months from the NFHS-4 (2015–16) and NFHS-5 (2019–21) survey has been utilized to obtain the prevalence and pattern of child poverty over time. The study estimates multidimensional child poverty using nine indicators from three dimensions and employs the Alkire-Foster (AF) method. Children are very vulnerable, and MPI-based research has shown that poverty among children is higher than that of adults [ 21 ]. Later the multilevel analysis was employed to obtain the district and PSU-level variance contribution to overall child poverty prevalence.

First, at the national level, about 27 percent of children were multidimensional poor in the latest NFHS-5 survey, which was 38 percent in NFHS-4. The Global report on MPI statistics for children (0–9 years) shows a higher child poverty in the neighbouring countries ranging from 28–60 percent compared to India (38% in 2015–16 and 27% in 2019–21). For instance, children from Afganistan (60% in 2015–16), Bangladesh (32.5% in 2019), Bhutan (45% in 2010), Nepal (27.9% in 2019) and Pakistan (48.6% in 2017–18) experiencing much higher multidimensional poverty. However, it also noted that the Global report on MPI calculates MPI at the household level and differentiated them by age-group whereas our analysis was performed at individual level with relevant child indicators [ 52 ]. In addition to this, other monetary measures of poverty, such as World Bank poverty estimated based on less than $1.90 per day, failed to capture the intensity and depth of poverty among children, as the children have different needs compared to other household members. In comparison, we solely use the child-related indicator to estimate child poverty, which directly and, in some cases, indirectly affects child well-being. Result also suggests that about 42.5 percent of children from Bihar were multidimensional poor, followed by Jharkhand (41%), Assam (35%), Madhya Pradesh (33%) and Gujrat (30%) in NFHS-5. Tripathi & Yenneti [ 32 ] have also observed these states with higher multidimensional poverty.

Moreover, Puducherry, Sikkim, Mizoram, Punjab, Delhi, Tamil Nadu, Kerala, and Chandigarh had the lowest levels of multidimensional child poverty, with about 7 to 12 percent of the children being multidimensionally poor in NFHS-5. The central and northern region state like Bihar, Jharkhand, Chattisgarh, Madhya Pradesh, and Uttar Pradesh constitute a significant proportion of the Indian population [ 40 ], and have the potential to alter the multidimensional child poverty at the national level. The results indicate that many states and union territories are improving their child poverty prevalence over time. States and Union territories like Puducherry, Rajasthan, Arunachal Pradesh, Tamil Nadu, Haryana, and Uttarakhand have reduced 41–50 percent child poverty over time from 2015–16 to 2019–21. In the same tenure at the India level, child poverty was also found to decline by 11 percentage points (29%), showing improvement in many child-related indicators [ 41 , 42 ]. However, some state and union territories like Chandigarh, Kerala, Lakshadweep, Nagaland and Himachal Pradesh have increased child poverty over time. Perhaps this is due to the fact that child nutrition indicators in many states and union territories have increased from NFHS-4 to NFHS-5 [ 41 , 42 ]. Additionally, India’s latest Global Hunger Index ranking (94 out of 107 countries) in 2020 supports this undernourished situation. Programmes such as Integrated Child Development Services Schemes (ICDS), Midday Meal Programme, and Iodine Deficiency Disorders Control Programme are targeted nutritional programmes that uplift the nutritional standard and play an important role in combating nutritional deficiencies, especially among women and children over the decade.

Second, the decomposition of multidimensional poverty by indicator suggests that among ten indicators, underweight (30%) made the most significant contributor to multidimensional poverty, followed by wasting (21%) in NFHS-5, which needs to be further addressed to improve underweight and wasting among children to reduce multidimensional child poverty in India. Two-fifths of children over the age of 0–59 months suffer from malnutrition in India, which is considered a serious problem in the face of public health [ 41 ].

The nutrition dimension contribution has increased over time (44.5% in NFHS-4 to 51% in NFHS-5), whereas the standard of living dimension decreased by 4.4 percent-points between NFHS-4 to NFHS-5 (43.6% to 39%, respectively). The results imply that the government programme improved more standard of living indicators such as improved drinking water, toilet building under Swacch Bharat Abhiyan (Clean India Movement) for improved sanitation, government aid for constructing pakka houses under different central and state government schemes, and promotion of clean cooking fuel through Ujjawala Yojana. One study finding from Ghana reveals that living standards are the most significant contributor to child poverty [ 53 ], coinciding with our study of NFHS-4, but not for NFHS-5. Different study settings, sample sizes, research designs, and survey times might explain the difference ( S1 Fig ).

The present study also examines the amount of variability in multidimensional child poverty using multilevel analysis for the effect of each level [ 48 , 50 ]. The variation in the prevalence of multidimensional child poverty has been presented with the help of VPC. The smaller value of VPC in NFHS-4 (4.4%) demonstrates a modest variance at the district level, and a more significant variation is observed at the community or PSU level (10.7%). This implies that the many socio-economic indicators vary from district and PSU levels. Further, the result shows that the variation in multidimensional child poverty has declined over time (4.4% in NFHS-4 to 2.7% in NFHS-5) at the district level but increased (10.7% in NFHS-4 to 11.9% in NFHS-5) at the PSU level in same duration.

The higher number of children born by mothers is positively associated with child poverty. Our finding is in line with other studies where a higher number of children in the family comprises well-being of the child and quality of care [ 31 , 54 , 55 ]. Perhaps this is because Indian families still favour sons, and to fulfil this, couples have many offspring to obtain their desired sex composition, even in a small family. Further, a higher number of children reduces parental attention and sometimes increases parental stress with a higher number of children. Moreover, it is advisable to consider children’s age as the different stages of life associated with diverse levels of expenditure required by childcare. The Rural reside children reported higher poverty compared to urban reside children, which is in line with previous studies [ 29 , 30 , 53 , 56 , 57 ], where rural children experienced more poverty compared to urban mainly due to the information constraints in the rural area. Mother age, maternal education, and belonging to a wealthier quintile were significantly associated with child poverty. The higher mother’s age, education, and wealth are major factors that improve the mother’s and child’s overall health. Similarly, these indicators are directly associated with the 3-A: affordability, availability and accessibility of programmes and treatment of any communicable diseases in childhood which may curb the illness and improve the child’s nutritional status. The socially disadvantage population sub-group, namely the scheduled tribe children, are more likely to be multidimensionally poor compared to general caste children confirming the previous research [ 34 , 35 ].

The present study has a few limitations. First, this study is based on cross-sectional information, so any causal relationship between multidimensional child poverty and its predictor could not be established. The indicator and dimension selection was challenging; however, the dimension and indicators of child poverty estimates have been selected based on prior work and research. Lastly, the present study focused on children under the age five. Moreover, some individual child samples were removed from the final analysis as the anthropometric measurements were missing (child’s height/weight measurements are out of plausible limits), the pairwise removal of missing information.

The study has estimated under-five child poverty in India, using Alkire and Foster’s multidimensional approach. The finding of this paper reveals the significant contribution of the nutrition dimension to child poverty in India, as poverty has a negative impact on growth and educational performance at a later age and has laid a weak foundation for the future. So attempt to improve the nutritional status of children by providing healthy dietary food and an effective public distribution system (PDS) including diverse and nutritious food grains. It also stipulates the inclusion of early interventions to improve the child’s nutritional status for a better future and lower child poverty. Recently Government of India launched ‘POSHAN Abhiyan’ in 2018 to reach the most deprived region in India with the primary aim of bringing a significant drop in country’s overall national deprivation. Such policy changes by the government of India show that the government is still working hard to reach the SDG goal of ending hunger and ensuring everyone has enough food. However, such programmes have been going on for a long time with the same goal and methods, they have helped reduce the poverty indicator over time.

There are significant regional differences in experiencing multidimensional child poverty. Children in the southern regions are estimated to have lower multidimensional poverty, while those in the east, northeast and central regions of India have the highest multidimensional poverty. Therefore, more focus should be given to states and regions with higher incidence and intensity to improve the condition of overall child poverty and achieve equitable and inclusive growth in the country. Notably, being underweight, wasting, immunization, clean cooking fuel, housing condition, and sanitation are significant sources of early childhood deprivation. Finally, there were significant variations in MDP at district and PSU levels, and at PSU levels, it increased compared to district-level MDP over time. The finding of this research support an in-depth assessment to expose the causes of poverty at the district and PSU levels to improve policymaking. Additionally, the interventions may be modified in such manner that enhance their target and take into account the differences between state, district, and PSU levels.

Therefore, efforts should be made to enhance the nutritional status and standard of living of most deprived households by promoting a child-centric and dimension-specific approach and focusing on PSU-level intervention so that child poverty can be minimised and eliminated in India.

Supporting information

S1 table. headcount ratio (h), intensity (a) and multidimensional poverty index (m 0 ), india and states..

https://doi.org/10.1371/journal.pone.0279241.s001

S1 Fig. Percentage of children deprived in each indicator during 2015–2021.

https://doi.org/10.1371/journal.pone.0279241.s002

Acknowledgments

The authors are grateful to the Department of Humanities and Social Sciences, National Institute of Technology (NIT) Rourkela and UNICEF, Odisha, for their support and encouragement, which helped improve this research paper.

• View Article
• Google Scholar
• 2. United Nation. Transforming our world: The 2030 agenda for sustainable development. New York. 2015. Available from: https://stg-wedocs.unep.org/bitstream/handle/20.500.11822/11125/unepswiosm1inf7sdg.pdf?sequence=1
• 3. Naoroji D. Poverty and Un-British Rule in India. Vol. XXX, Oxford University. 1901. 60 p.
• 9. G Gordon D, Nandy S. Measuring child poverty and deprivation. In Global Child Poverty and Well-Being 2012 Feb 29 (pp. 57–102). Policy Press.
• 15. Alkire S. Child Poverty in Bhutan: Insights from Multidimensional Child Poverty Index (C-MPI) and Qualitative Interviews with Poor Children. National Statistics Bureau,[Government of Bhutan]; 2016.
• 18. Oxford Poverty and Human Development Initiative, Oxford University. CHILD MULTIDIMENSIONAL POVERTY IN THAILAND, 2019 https://www.unicef.org/thailand/media/3171/file/Child%20Multidimensional%20Poverty%20in%20Thailand.pdf .
• 20. Alkire S, Ul Haq R, Alim A. The state of multidimensional child poverty in South Asia: a contextual and gendered view. OPHI Working Paper 127, University of Oxford; 2019.
• PubMed/NCBI
• 29. Gordon D, Nandy S, Pantazis C, Townsend P, Pemberton SA. Child poverty in the developing world. Policy Press; 2003 Oct 21.
• 30. Minujin A, Nandy S, editors. Global child poverty and well-being: Measurement, concepts, policy and action. Policy Press; 2013 Jan 28.
• 40. Census of India. Census of India. Office of the Registrar General & Census Commissioner, Governement of India. 2011.
• 41. IIPS & ICF. National Family Health Survey (NFHS-4) 2015–16 India. 2017. Available http://rchiips.org/Nfhs/NFHS-4Reports/India.pdf
• 42. IIPS & ICF. National Family Health Survey (NFHS-5), 2019–21: India. 2022. Available http://rchiips.org/nfhs/NFHS-5Reports/NFHS-5_INDIA_REPORT.pdf
• 44. Alkire S, Roche JM, Ballon P, Foster J, Santos ME, Seth S. Multidimensional poverty measurement and analysis. Oxford University Press, USA; 2015.
• 45. Sophia Rabe-Hesketh, Skrondal A. Multilevel and Longitudinal Modeling Using Stata. In Texas: STATA press; 2012.
• 50. Snijders TA, Bosker RJ. Multilevel analysis: An introduction to basic and advanced multilevel modeling. SAGE; 2011 Oct 30.
• 57. Peng J, Liu W, Qin X, Wu H. An Analysis of the Factors Affecting Rural Multidimensional Poverty. In8th International Conference on Social Network, Communication and Education (SNCE 2018) 2018 May (pp. 1149–1155). Atlantis Press. Available from: https://doi.org/10.2991/snce-18.2018.239

India’s poverty debate needs to move on: Let’s adopt new norms

• We ought to acknowledge India’s success in relieving poverty by norms framed back in the 1970s, but it’s also time to revise our poverty cut-off in accordance with what we consider a ‘decent standard of living’ today.

The National Sample Survey Office (NSSO) released summary results for the Household Consumer Expenditure Survey (HCES) earlier this year. This release re-ignited discussions on poverty, with commentators raising a range of issues. Using the Tendulkar Committee methodology, an SBI report estimated that poverty in India would be around 6.3% in 2022-23. Using the Rangarajan Committee methodology, C. Rangarajan and Mahendra Dev suggested that the all- India poverty ratio would be around 10%.

In either case, these estimates suggest that there has been a sharp reduction in poverty since 2011-12. Partly in response to these articles, other commentators have suggested revisiting the poverty line. They argue that changes in survey methodology in the recent HCES renders the application of earlier methodologies to HCES data inappropriate. These types of arguments were comprehensively rebutted by Surjit S. Bhalla in a column in this paper on 27 March 2024.

None of these discussions, however, examined the appropriateness of the existing methodologies for tracking poverty using consumption data.

Consumption-based poverty measurement in India traces its roots to a Working Group constituted by the Planning Commission in 1962, which was subsequently re-visited and revised by a task force in 1979. This task force provided a careful description of the reasoning it employed to arrive at a poverty line for India. 

In brief, it defined the poverty line as a per-capita consumption expenditure level which could meet an average per capita daily calorie requirement of 2,400 Kcal in rural areas and 2,100 Kcal in urban areas, along with the associated quantum of expenditure on non-food items. This average calorie norm was based on an analysis of the demographic and activity-based composition of the population at that time.

The monetary value of this norm formed the basis of poverty lines in all subsequent revisions. However, the basic approach underlying this calculation was never seriously revisited until the Tendulkar and Rangarajan committees. Both of them noted that changes in the demographic and activity composition since the task force required changes in the calorie norm and expenditure level. 

However, while they revisited the expenditure and nutritional dimensions of a poverty norm, they did not adequately address its non-food components. The essence of their argument was to assert that if spending in an expenditure class is adequate to meet nutritional norms, it will also be normatively adequate to meet the associated non-food components.

The problem with this assumption is that India has changed significantly since the task force on poverty was set up in the 1970s. Life expectancy at birth, which was 49.7 years in 1970 has risen to 69.4 in 2018. The population above 60 has risen from 6.1% to above 10.1% (as of 2021). India’s gross enrolment ratio (GER) at the primary level has risen from about 62% in 1971 to universal enrolment today. In higher education, our GER has risen from below 6% then to around 28% today.

These changes have implications for out-of-pocket expenditures on education and health. NSS surveys reveal this fact by showing the significant rise in expenditure on these accounts over the years. The reduction in mortality rates and rise in life expectancy have increased age heterogeneity in our population, with a rising share of the elderly. Changes in household composition (with the rise of unitary families) have also increased the proportion of the elderly who are living on their own. It is well documented that the medical requirements of the elderly are different from those of the young. 

The implications of this changing age profile on the normative health expenditure requirements of this emerging group have thus far not been carefully addressed. Given that the proportion of the elderly is projected to rise even more sharply in the coming years, the adequacy of norms derived for a much younger population will become even more questionable.

The inclusion of a poll promise to extend the coverage of Ayushman Bharat to the elderly in the Bharatiya Janata Party manifesto is an indicator that the political class has understood this problem better than the academic community that specializes in studying poverty.

The success of policies that have resulted in a rise in enrolment ratios both at primary and higher education levels has meant that competition for aspirational jobs has become stiffer. This increased competition has led to increased out-of-pocket expenditure on supplementary tuition classes, as indicated by the mushrooming of coaching institutions across the country. 

The burden of expenditure on private coaching is a major factor in creating what we may call ‘education poverty’ among families with young children. Addressing this problem is not merely about making financial resources available, but about changing the approach to education, as has at least been signalled in the National Education Policy, 2020.

The deeper problem is as follows. Consumption- based poverty measures are essentially capturing the average attributes of the population. While an average is a reasonable way to compare population attributes over time when the underlying population structure is stable, increased heterogeneity in the population makes the use of an average to describe attributes such as poverty seriously problematic.

For example, an elderly household may have enough income to meet the expenditure requirements of a nutrition norm, but not its healthcare requirements. Similarly, we could have a household with young children that can meet its food needs but not its aspirational requirements of private tuition. 

Such examples reveal that households may exceed the thresholds of extreme poverty, but yet continue to lack resources for a ‘decent standard of living.’ In order to meet the Sustainable Development Goal of ending poverty in all its forms, we should acknowledge India’s success in relieving poverty by norms framed in the 1970s and move on, so that we can define fresh norms appropriate for Amrit Kaal.

MINT SPECIALS

Wait for it….

Log in to our website to save your bookmarks. It'll just take a moment.

You are just one step away from creating your watchlist!

Oops! Looks like you have exceeded the limit to bookmark the image. Remove some to bookmark this image.

Your session has expired, please login again.

Congratulations!

You are now subscribed to our newsletters. In case you can’t find any email from our side, please check the spam folder.

Subscribe to continue

This is a subscriber only feature Subscribe Now to get daily updates on WhatsApp

Open Demat Account and Get Best Offers

Start Investing in Stocks, Mutual Funds, IPOs, and more

• Please enter valid name
• Please enter valid mobile number
• Please enter valid email
• Select Location

I'm interested in opening a Trading and Demat Account and am comfortable with the online account opening process. I'm open to receiving promotional messages through various channels, including calls, emails & SMS.

The team will get in touch with you shortly

Main navigation

• Clinical Education
• O'Brien Fellowships
• Steinberg Fellowships
• CHRLP Article Lab

Gender Disparities In Organ Donation And Transplantation In India: A Call For Equality

• Add to calendar
• Tweet Widget

The 2021 statistics report of the National Organ and Tissue Transplant Organisation (NOTTO) of India has brought attention to a significant gender disparity: 80% of organ recipients were male, while, 75% of organ donors were female. This phenomenon has become entangled in the complex existing web of intersectional discrimination in India, which deviates from the principles of justice and equality enshrined in Articles 14 and 15 of the Indian Constitution. Moreover, the gender inequality in organ donation and transplantation in India reflects discriminatory practices and is a manifestation of patriarchal ideologies that marginalize women’s accessibility to healthcare facilities. This gender imbalance constitutes a violation of women’s right to health and medical care, which were recognized as essential aspects of the right to life under Article 21 of the Indian Constitution in the landmark case of Francies Coralie Mullin v Union Territory of Delhi.

Female organ donors are influenced by sociocultural factors and financial dependency. Financial independence may reduce their inclination to donate. Additionally, one of the psychological factors contributing to the gender gap in organ donation is women’s prior experiences, like childbirth, which enhance their trust in the medical system. Mothers are the primary organ donors among parents accounting for 73% of donations, while wives contribute 91% of donations in married couples. Regarding this gender inequality, Dr Pranjali Modi, the convenor of the State Organ and Tissue Transplant Organisation (SOTTO), has emphasized that patriarchal beliefs play a significant role in the general reluctance to accept organ donations from men, who are often considered to be the primary breadwinners in families. Moreover, when men donate organs, financial difficulties might arise during the evaluation, surgery, and recovery phases, which can result in the coercion of women to undergo organ donation surgery.

Relatedly, there is a subconscious prioritization of men over women for life-saving treatments, such as organ transplantation, based on the perceived economic importance of their roles. These biases are also apparent within medical institutions and professions, potentially affecting doctor-patient interactions and resulting in gender inequalities in organ transplant procedures. However, a notable scientific reason for the gender disparity in organ transplantation is that pregnancy sensitizes women’s immune systems, potentially diminishing the compatibility of their organs with those of their blood relatives, which are those individuals who are most inclined to donate organs. Further, women waiting to receive organ donations, due to their typically smaller stature, encounter a greater likelihood of being declined organ offers, leading to a higher risk of death or removal from the waitlist compared to small-stature men. This inherent bias in healthcare exacerbates gender inequalities in organ transplantation access for women. These factors, altogether, contribute to unequal access to organ transplantation, a critical healthcare intervention, as evidenced by a longitudinal study spanning 15 years which revealed that a mere 15% of female individuals were organ recipients in the nation.

This healthcare outcome contradicts the stipulations outlined in Article 12 of the International Covenant on Economic, Social and Cultural Rights (“ICESCR”), which acknowledges the entitlement of every individual to achieve “ the highest attainable standard of physical and mental health .” Additionally, the importance of guaranteeing equitable allocation of health facilities, goods and services among both men and women has been underscored in the remarkable case of Karukola Simhachalam vs Union of India and Ors (WP PIL No 164 of 2019). The case further emphasized the necessity of counteracting any hindrances to the full realisation of the right to health and eradicate discriminatory practices for all people. Nevertheless, any practice embedded in gender bias not only perpetuates disparities for women but also undermines the very bedrock of equality principles. This practice is vulnerable to legal scrutiny under the Universal Declaration of Human Rights , which accentuates the significance of legal egalitarianism, the right to health, and well-being.

It is imperative to confront gender imbalances to foster a society characterised by equality, inclusivity and the protection of women’s rights. Although the Transplantation of Human Organs Act, of 1994 ostensibly maintains a gender-neutral stance, the actual implementation unveils a fabric intricately woven with evident gender disparities and prejudices. The dissonance between legal intent and application of the legislation in practice underscores the need to erase gender stereotypes and inequalities within the organ transplantation and donation domain.

Addressing the issue of gender disparities in organ transplantation necessitates raising awareness as a foundational step. To mitigate disparities among disadvantaged groups, it is crucial to implement transplant centers and community initiatives targeting rural and marginalized areas and establish transplant centers in those places as well. The pre-transplant assessment should encompass psychosocial and economic dimensions to identify any potential negative motivations for donation, such as underlying abuse, coercion, threats and violence. Conducting research studies and collecting accurate statistics at the grassroots level is also imperative for developing policies that promote equality and fair access to organ transplantation. Additionally, overcoming biological barriers in transplantation procedures requires the implementation of science-backed policies.

Lastly, providing medical assistance to organ donors for the transplantation process and offering medical support to organ recipients from vulnerable communities are essential steps that Indian government should take. However, addressing gender imbalance requires more than just focusing on education and poverty alleviation. It necessitates a deeper examination of traditional gender roles and the roles of women within their families. Policies must be implemented to ensure that women receive equitable access to transplantation services, regardless of their societal status or financial standing within the family.

Department and University Information

Centre for human rights and legal pluralism.

• Faculty of Law
• Law Admissions - BCL/JD
• Law Admissions - graduate programs
• Law Student Affairs Office
• Law Career Development Office
• Nahum Gelber Law Library
• Focus online
• CHRLP Facebook page
• Business Law Platform
• Centre for Intellectual Property Policy
• Fortier Chair in Int'l Arbitration & Commercial Law
• Institute of Air & Space Law
• Jean Monnet Chair in International Economic Integration
• Labour Law and Development Research Laboratory
• Oppenheimer Chair in public international law
• Paul-André Crépeau Centre for Private & Comparative Law
• Peter MacKell Chair in Federalism
• Private Justice and the Rule of Law
• Research Group on Health & Law
• Rule of Law and Economic Development
• Stikeman Chair in Tax Law
• Wainwright Fund

Advertisement

Poverty Among Elderly in India

• Published: 06 September 2011
• Volume 109 , pages 493–514, ( 2012 )

Cite this article

• Akanksha Srivastava 1 &
• Sanjay K. Mohanty 1  

1778 Accesses

42 Citations

Explore all metrics

Using consumption expenditure data of the National Sample Survey 2004–2005, this paper estimates the size of elderly poor and tests the hypotheses that elderly households are not economically better-off compared to non-elderly households in India. Poverty estimates are derived under three scenarios—by applying the official cut-off point of the poverty line to household consumption expenditure (unadjusted), consumption expenditure adjusted to household size and consumption expenditure adjusted to household composition. Results show that an estimated 18 million elderly in India are living below the poverty line. On adjusting the consumption expenditure to household size and composition, there are no significant differences in the incidence of poverty among elderly and non-elderly households in India. This is in contrast to the notion that elderly households are better off than non-elderly households in India. Based on the findings, we suggest that the age dimension should be integrated into social policies for evidence based planning.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

Poverty Among The Elderly: An Application of a Collective Consumption Model in Brazil

Income and Expenditure Mismatch of Poorest of the Poor: An Analysis of Financial Requirement of Slum Dwellers

Measurement and Analysis of Poverty in Rwanda

Alam, M., & Barrientos, A. (2010). Changing demographic landscape of south Asia and emerging issues of employment, ageing, and old age security. In M. Alam & A. Barrientos (Eds.), Demographics, employment and old age security (pp. 1–22). India: Maclillan.

Google Scholar  

Barrientos, A., Gorman, M., & Heslop, A. (2003). Old age poverty in developing countries: Contribution and dependence in later life. World Development, 31 (3), 555–570.

Article   Google Scholar  

Buhmann, B., Rainwater, L., Schmauss, G., & Smeeding, T. (1988). Equivalence scales, well-being, inequality, and poverty: Sensitivity estimates across ten countries using the Luxembourg Income Study (LIS) database. Review of Income and Wealth, 34 , 11542.

Coale, A. J., & Hoover, E. M. (1958). Population growth and economic development in low income countries . Princeton: Princeton University Press.

Deaton, A., & Paxson, C. (1995). Measuring poverty among the elderly. NBER working paper no. 5296, Cambridge, Massachusetts.

Deaton, A., & Zaidi, S. (2002). Guidelines for constructing consumption aggregates for welfare analysis. LSMS working paper 135.

Dreze, J. (1990). Widows in rural India. London School of Economics, Development Economics Research Programme DEP 26.

Dreze, J., & Srinivasan, P. (1997). Widowhood and poverty in rural India: Some inferences from household survey data. Journal of Development Economics, 54 (2), 217–234.

Gasparini, L., Alejo, J., Haimovich, F., Olivieri, S., & Tornarolli, L. (2007). Poverty among the elderly in Latin America and the Caribbean. Background paper for the World Economic and Social Survey.

Hurd, M. D. (1990). Research on the elderly: Economic status, retirement, and consumption and saving. Journal of Economic Literature, XXVIII , 565–637.

International Institute for Population Sciences (IIPS), & ORC MACRO. (2007). National Family Health Survey (NFHS -3), 2005–2006: India (Vol. I). Mumbai: IIPS.

Knodel, J., Chayovan, N., & Siriboon, S. (1992). Impact of fertility decline on familial support for the elderly: An illustration from Thailand. Population and Development Review, 18 (1), 79–103.

Kumar, S., Sathyanarayana, K. M., & Omer, A. (2011). Living arrangements of elderly in India: Trends and differentials. Unpublished work. Paper presented at international conference on challenges of population aging in Asia, March 14–15, 2011, INSA, New Delhi.

Mahal, A., & Berman, P. (2001). Health expenditures and the elderly: A survey of issues in forecasting, methods used, and relevance for developing countries. Research paper 01.23 . Cambridge, MA: Harvard Burden of Disease Unit.

Meenakshi, J. V., & Ray, R. (2002). Impact of household size and family composition on poverty in rural India. Accessed through: http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V82-460M93D-2& .

Mohanty, S. K., & Sinha, R. K. (2010). Deprivation among the elderly in India. In M. Alam & A. Barrientos (Eds.), Demographics, employment and old age security (pp. 301–315). India: Maclillan.

Mujahid, G., Pannirsalvem, J., & Doge, B. (2008). The impact of social pensions: Perceptions of Asian older persons UNFPA Country technical services team for East and South East Asia . Thailand: Bangkok.

National Sample Survey Organisation. (2006). Level and pattern of consumer expenditure, 2004–2005. NSS 61st round, Report no 508 (61/1.0/1).

Nelson, J. A. (1988). Household economies of scale in consumption: Theory and evidence. Econometrica, 56 (6), 1301–1314.

Office of the Registrar General, Census Commissioner, India. (1999). Sample Registration System . New Delhi: Ministry of Home Affairs.

Office of the Registrar General, Census Commissioner, India. (2004). Sample Registration System . New Delhi: Ministry of Home Affairs.

Office of the Registrar General, Census Commissioner, India. (2006). Population projection for India and States (2001–2026) . New Delhi: Ministry of Home Affairs.

Office of the Registrar General, Census Commissioner, India. (2008). Sample Registration System . New Delhi: Ministry of Home Affairs.

Office of the Registrar General, India (RGI). (2009). Report on causes of death in India 2001–2003 . New Delhi: RGI.

Pal, S., & Palacios, R. (2006). Old age poverty in the Indian States: What the household data can say? Report no. 16, discussion paper, South Asia: Human Development Sector, World Bank.

Pal, S., & Palacios, R. (2008). Understanding poverty among the elderly in India: Implications for social pension policy. Discussion paper no. 3431 . Germany: Brunel University and IZA.

Panda, P. K. (1998). The elderly in rural Orissa: Alone in distress. Economic and Political Weekly, 33 (25), 1545–1550.

Pandey, M. K. (2009). On ageing health and poverty in Rural India . Delhi, India: Institute of Economic Growth.

Planning Commission, Government of India. (2007). Poverty estimates for 2004–2005. Accessed through: http://planningcommission.nic.in/news/prmar07.pdf .

Sherlock, P. L. (2000). Old age and poverty in developing countries: New policy challenges. World Development, 28 (12), 2157–2168.

Treas, J., & Logue, B. (1976). Economic development and the older population. Population and Development Review, 12 , 645–673.

Whithead, M., Dahlgren, G., & Evans, T. (2001). Equity and health sector reforms: Can low income countries escape the medical poverty trap? The Lancet, 358 (928), 833–836.

World Bank Report. (2006). Accessed through: http://siteresources.worldbank.org/INTANNREP2K6/Resources/2838485-1158333614345/AR06_final_LO_RES.pdf .

WHO (World Health Organisation). (2010). Accessed through: Program. http://www.who.int/mediacentre/news/notes/2010/noncommunicable_diseases_20100514/en/index.html .

Zimmer, Z. (2008). Poverty, wealth inequality and health among older adults in rural Cambodia. Social Science and Medicine, 66 , 57–71.

Download references

Acknowledgments

The authors thank Dr. Rajesh K. Chauhan, Joint Director, Population Research Centre, University of Lucknow, for his help in data decoding.

Author information

Authors and affiliations.

International Institute for Population Sciences, Govandi Station Road, Deonar, Mumbai, 400088, India

Akanksha Srivastava & Sanjay K. Mohanty

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Akanksha Srivastava .

Rights and permissions

Reprints and permissions

About this article

Srivastava, A., Mohanty, S.K. Poverty Among Elderly in India. Soc Indic Res 109 , 493–514 (2012). https://doi.org/10.1007/s11205-011-9913-7

Download citation

Accepted : 24 July 2011

Published : 06 September 2011

Issue Date : December 2012

DOI : https://doi.org/10.1007/s11205-011-9913-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Household size
• Household composition
• Non-elderly
• Find a journal
• Publish with us
• Track your research