Journal of Social and Economic Development

https://doi.org/10.1007/s40847-023-00238-z

RESEARCH PAPER

Economic well‑being of middle‑aged and elderly adults

in India: variations by household composition

Sanjay K. Mohanty1 · P. Arokiasamy2 · Itishree Nayak3 · Prashant Shekhar4

Accepted: 31 January 2023

© The Author(s), under exclusive licence to Institute for Social and Economic Change 2023

Abstract
Using data on 42,949 households from the recently conducted Longitudinal Ageing Study
of India, this paper examined the economic well-being of middle-aged and elderly adults
in India. All households were classified into three mutually exclusive groups: households
with only elderly members (60+), households with both elderly and non-elderly members,
and households with no elderly members. Economic well-being was assessed using subjec-
tive well-being and a composite index that comprised per capita consumption expenditure,
monthly per capita income, and wealth index. The mean value of the economic well-being
index of middle-aged and elderly adults was 53.8 (95% CI 53.3–54.4). It was 51.6 among
households with only elderly members, 53.5 among households with both elderly and
non-elderly members, and 54.9 among households without any elderly members. Health
expenditure accounted for 20% of the consumption expenditure among households with
only elderly members compared to 13% among households with both elderly and non-
elderly members and 12% among those with no elderly members. Controlling for socio-
demographic characteristics, households with only elderly members and those with both
elderly and non-elderly members had a lower economic status compared to households
with no elderly members. Subjective well-being was positively and significantly associ-
ated with the objective measures of well-being as measured by the composite index. When
the economic well-being was measured using per capita consumption expenditure alone,
households with middle-aged and elderly adults were found to be better-off than house-
holds with non-elderly members. However, when we measured the economic condition
using the comprehensive economic measure, we found the elderly households to be poorer
than the non-elderly households. Economic independence is key for the elderly living inde-
pendently, but the poor elderly have no choice except to live with their children for their
survival. These findings highlight the need for strengthening social security and ensuring
health protection for the elderly in India.

Keywords Elderly · Non-elderly · Economic well-being · LASI · India

* Sanjay K. Mohanty
[email protected]
Extended author information available on the last page of the article

13
Vol.:(0123456789)
 Journal of Social and Economic Development

Introduction

In the developing countries, the demographic transition is leading to fast increase in

median age, size, and share of the elderly population, widowhood, and old-age dependency
(WHO 2011; Fuster 2017; Isherwood et al. 2017). These demographic changes are caus-
ing societal changes and increasing the economic insecurity and the disease burden among
elderly. They are also associated with a rise in nuclear families, an increase in individu-
alism, and growth in mobility (Bianchi 2014). While these changes affect everyone, the
elderly population is especially vulnerable due to retirement from workforce, reduced or no
income, deteriorating health, rising disease burden, increasing familial responsibility, and
social isolation (UN 2013; Bloom et al. 2003; Sheiner 2014; Maestas et al. 2016; Kämp-
fen et al. 2018). Disease burden and financial catastrophe also have a strong age gradient
(Mohanty et al. 2014; Prince et al. 2015; Arsenijevic et al. 2016).
Socio-economic development and urbanisation act as catalysts of the decline in inter-
generational co-residence worldwide (Szołtysek et al. 2011; Hughes and Waite 2002).
With increasing levels of development, the traditional family system has weakened and
nuclear families have become mainstream (Chakravorty et al. 2021). An increasing propor-
tion of middle-aged and elderly adults are now living alone or with their spouses or with
their unmarried children (Ruggles and Heggeness 2008). The living arrangements of older
adults are associated with their economic well-being, physical and psychosocial health,
and life satisfaction (Zimmer and Das 2014). Rising incomes, increasing coverage of social
support system, growing mobility, and falling dependency on agriculture are driving the
trend of independent living arrangements among the elderly (Szołtysek et al. 2011). At the
same time, social and demographic factors like privacy among younger couples, internal
migration (Taylor and Bain 2005), social networking, and changes in patterns of marriage,
cohabitation, and divorce are changing the household composition (Kaur and Singh 2013).
The life satisfaction of the elderly is associated with their financial well-being and social
capital (Yeo and Lee 2019). Household income is another key determinant of the mental
health and well-being of the elderly (Jeon et al. 2007).
India had 136 million older persons (60+) in 2020, accounting for 10% of the country’s
population. It is the second largest country in the world in terms of size and growth of the
elderly population (MoHFW 2019). The growth rate of the elderly population is 3.5% com-
pared to 0.07% of children aged 0–14 years in India (MoHFW 2019).
The longevity of the elderly has been increasing across the states and socio-economic
spectrum of India. But half of the elderly are financially dependent, and two-fifths do not
have any source of income (Kulkarni et al. 2016). Less than 10% of India’s total workforce
works in the organised sector, and the coverage of old-age pension is low and inadequate
(Maestas et al. 2016). The extent of poverty is higher among the elderly households com-
pared to the non-elderly ones (Srivastava and Mohanty 2012). Hospitalisation and out-
of-pocket expenditure (OOPE) have a strong age gradient, with elderly households being
more likely to have high OOPE and catastrophic health spending (Pandey et al. 2018; Kas-
tor and Mohanty 2018). Despite that, the insurance sector in India systematically excludes
the elderly and the chronically sick adults from its ambit.
Large-scale population-based health surveys, such as the NSSO and the NFHS, provide
large amounts of data in India. Yet neither of them provides comprehensive information
on the economic well-being of households. The National Sample Survey (NSS) collects
consumption expenditure data regularly but does not collect information on household
income, assets, and debt. The National Family and Health Survey (NFHS), on the other

13
Journal of Social and Economic Development

hand, collects information neither on consumption expenditure nor on household income

but, instead, uses economic proxies (the wealth index) that do not adequately capture the
economic well-being of households, especially households with elderly persons. Filling
this void, the Longitudinal Ageing Study in India (LASI) provides detailed data on the con-
sumption, income and assets, and debts of older adult households. Using data from LASI,
we present a comparative assessment of the economic well-being of households with only
elderly members, households with both elderly and non-elderly members, and households
with no elderly member in India.

Data and methods

Data

We used the unit data from wave 1 of the Longitudinal Ageing Study in India (LASI),
2017–18. LASI is India’s first and the world’s largest-ever comprehensive nationwide study
on the health, economic, and social well-being of older adults (45+). It was harmonised
with the Health and Retirement Survey (HRS) family of ageing studies for cross-country
comparison. LASI is a collaborative study of the International Institute for Population Sci-
ences (IIPS), Mumbai, the Harvard Chan School of Public Health (HSPH), USA, and the
University of Southern California (USC), USA, with financial support from the Ministry of
Health and Family Welfare, Government of India, the National Institute of Ageing (NIA),
USA, and the UNFPA-India. The LASI survey was canvassed among sample households
with at least one member aged 45+. A total of 42,949 households, comprising 72,250 indi-
viduals aged 45 years and above, across all the states and union territories except Sikkim
(the survey was under way in Sikkim at the time of the release of the LASI data), were
successfully interviewed using the stratified multistage probability cluster sampling design.
Detailed information on the survey design, contents, and process of LASI is available in
the public domain (IIPS, HSPH, and USC 2020).
LASI canvassed a household schedule, an individual schedule, a biomarker sched-
ule, and a community schedule from the eligible households and age-eligible respond-
ents. Figure 1 presents the flowchart of the detailed information collected on household
economic condition in the survey. The LASI household schedule collected information
on the demographics of each household member and on the housing condition, sanita-
tion, living conditions, consumption, income, wealth, debt and loans, and the subjec-
tive economic well-being of each household. An abridged version of the consumption
schedule, covering over 30 questions, was canvassed, which facilitated the estimation
of household consumption expenditure on food and non-food items for 30 days preced-
ing the survey. The information on non-food expenditure included expenditure incurred
on hospitalisation and outpatient services and was treated as being a part of household
consumption. Questions on income were canvassed carefully to capture annual income
all participating household members from all sources. A set of seven screening ques-
tions was asked to identify the sources of income. Annual household income by source
was estimated and aggregated to obtain total household income and then converted
into per capita income by dividing the total income by household size. Detailed ques-
tions were also asked on assets, debts and loans, and reasons for loans. The question
on the subjective economic well-being read as follows: How well, would you say, your

13
 Journal of Social and Economic Development

Fig. 1  Schematic representation of economic modules covered in LASI, wave 1

household is managing financially these days? We used all these economic variables to
provide a summary measure of the economic well-being of elderly households in India.

Methods

We used descriptive statistics, constructed composite indices, and performed multivar-

iate analyses to understand the economic well-being of middle-aged and elderly adults
in India. A brief description of these methodologies is given below.

13
Journal of Social and Economic Development

Construction of wealth index

The wealth index was computed from a set of variables on household consumer dura-
bles, ownership of house, and household amenities by using the principal component
analysis (PCA) (Filmer and Pritchett 2001; Rutstein and Johnson 2004; Rutstein 2015).
PCA generates as many principal components as there are variables, with each principal
component being the weighted sum of all the variables. The first principal component
was used in the estimation. A total of 28 variables for rural areas and 26 for urban areas
were used to compute the wealth index. We calculated the combined score using the
appropriate urban and rural factor scores, constants, and coefficients (Rutstein 2015).
The wealth index was classified into five quintiles: poorest, poorer, middle, richer, and
richest. Cronbach α was used to check the reliability of the variables used in the con-
struction of the asset-based wealth index.

Index of consumption, index of income, and wealth index

We measured the economic well-being by combining the monthly per capita consump-

ing the composite index, we used a minimum value of ₹100 each for per capita con-
tion expenditure, the monthly per capita income, and the wealth index. For construct-

capita income were truncated at 99 percentiles, with values of ₹14,179 and ₹31,562,
sumption and per capita income. The upper limits of per capita consumption and per

respectively. We then used a logarithmic transformation of the consumption and income

variables due to their skewed distribution. In the case of the wealth index, we made the
composite score begin from 0 by adding a constant term of 5.99 such that the composite
score ranged from − 5.99 to 8.61. The minimum and maximum values of these variables
are given in Table 2. The consumption, income, and wealth were normalised by using
the standard method.

Index of monthly per capita consumption expenditure (IMPCE)

The index was computed as follows:

( )
ln MPCEi − ln (100)
IMPCE = (1)
ln (14, 179) − ln (100)
where MPCEi is the monthly per capita expenditure of the ith individual, and ₹100 and
₹14,179 are, respectively, the minimum and maximum monthly expenditures of an
individual.

Index of per capita income (IPCI)

The index was computed as follows:

( )
ln PCIi − ln (100)
IPCI = (2)
ln (31, 562) − ln (100)

13
 Journal of Social and Economic Development

where PCIi is the monthly per capita income of the ith individual, and ₹100 and ₹31,562
are, respectively, the lower and upper limits.

Index of wealth (IW)

The index was computed as follows:

Compositscorei − 0
IW = (3)
14.59 − 0

Index of economic well‑being (IEWB)

The index was computed by using the arithmetic mean of the indices of consumption,
income, and wealth.
1
Index of economic wellbeing = (IMPCE + IPCI + IW) × 100 (4)
3
The composite index varies in the range of 0 and 100. The closer the value to 100, the
better the economic well-being, whereas the closer the value to 0, the worse the economic
well-being.

Outcome variables

The composite index of economic well-being that combined the objective measures of
monthly per capita consumption expenditure, monthly per capita income, and wealth index
was the main outcome variable. The subjective well-being of a household was the other
outcome variable.

Independent variables

Households were classified into three mutually exclusive categories based on their com-
position. These categories were used as the main independent variable and included: (a)
households with only members aged 60 years and above, (b) households with both elderly
and non-elderly members and (c) households with no elderly members but a member aged
45+.
Descriptive statistics, ordinary least squares regression, and ordered probit regression
were used in the analyses. The ordinary least squares (OLS) regression equation was esti-
mated by using the composite index of economic well-being, taking MPCE as the depend-
ent variable. The basic OLS regression model can be expressed as:
Yi = 𝛼 + 𝛽Xi + 𝜖i (5)
where Yi is the composite index of economic well-being of the ith individual; 𝛼 and 𝛽 are
the parameters to be estimated; Xi is the vector of the explanatory variables that includes
age, sex, and marital status of the elderly, size of household, and caste, religion, place of
residence, and educational attainment of head of household; and 𝜖i is the disturbance term
assumed to be independently and identically distributed (i.i.d.).

13
Journal of Social and Economic Development

The subjective measure of economic well-being—a dependent variable—was re-catego-

rised on a 1–5 scale as follows: finding it very difficult (1), finding it difficult (2), just about
getting by (3), doing all right, (4) and living comfortably (5). The response was in the order
scale, so an ordered probit model was used in the analysis.
The mathematical form of an ordered probit model is given as:
�
y∗i = Xi 𝛽 + 𝜀i i = 1, 2, 3, 4, … , N (6)

where y∗i is the observed ordinal rating (level of subjective economic status on a scale of
1–5), X is a vector of the independent variables mentioned before, and β is a vector of the
parameters. The error term (εi) is assumed to be independently and identically normally
distributed, N (0, 1).

Results

Table 1 presents the summary measures of economic well-being and health spending by
type of households in India. About half of the households (50%) had both elderly and non-
elderly members, 9% had only elderly members, and 41% had no elderly members but a
member aged 45–59 years. Household size and median age varied considerably across
the three types of households. Per capita health expenditure accounted for one-fifth of the
household consumption expenditure in households with only elderly members, compared
to 13% in households with both elderly and non-elderly members and 12% in households
with non-elderly members. However, the Gini index was similar across each type of house-
hold. The composite wealth index score was 30.1 among elderly-only households, 43.2
among households with both elderly and non-elderly members, and 41.2 among house-
holds without any elderly member in India.
Figure 2 shows the comparison of MPCE with and without health expenditure by
type of households in India. Both MPCE (health expenditure inclusive) and MPCE less
of health expenditure were found to be the highest in households with only elderly mem-
bers, followed by households with no elderly members and households with both elderly
and non-elderly members. Figure 3 shows the cumulative distribution function (CDF) of
MPCE by type of households in India. The CDF of households without any elderly mem-
ber was above that of households with no elderly members and households with both
elderly and non-elderly members. This suggests that at each stage of the distribution, the
MPCE of households with both elderly and non-elderly members was the lowest and that
of elderly-only households was the highest. Figure 4 presents the CDF of per capita income
by type of households. At the upper end of the distribution, households with only elderly
members had the lowest MPCI as compared to the other two types of households. At each
stage of the distribution, the MPCI was the highest in households where both elderly and
non-elderly members resided. Table 2 presents the log transformation of the minimum and
maximum values of MPCE, MPCI, and the wealth index and their mean and standard devi-
ation for all sample households. The standard deviation was 0.69 in the consumption index,
1.14 in the income index, and 2.38 in the wealth index. Table 3 presents the mean of three
economic indices as the composite index by type of household composition in India. The
index of MPCE was 65.4, the index of MPCI was 54.7, and the wealth index was 41.1 for
all middle-aged and elderly adults in India.

13
13
Table 1  Summary indicators of economic well-being and health spending by type of households, India, LASI wave 1, 2017–2018
Variables Households with only Households with both elderly and Households without All
elderly members non-elderly members elderly members

Number of households 3818 21,383 17,738 42,939

Percentage share in households 8.89 49.8 41.31 100
Median age 69 32 26 30
Average household size 1.57 5.55 4.61 4.8
Percentage urban 23.43 31.92 33.48 31.64
Monthly per capita consumption expenditure (in ₹) 3868 2898 3001 2967
95% confidence interval [3760–3975] [2842–2954] [2965–3036] [2842–2954]
Non-food expenditure(in ₹) 1721 1459 1460 1468
95% confidence interval [1639–1804] [1409–1508] [1433–1487] [1409–1508]
Non-food expenditure as share of consumption expenditure 44.51 50.34 48.67 49.48
Per capita health expenditure (in ₹) 792 384 352 386
95% confidence interval [730–853] [372–396] [336–368] [372–396]
Health expenditure as a share of MPCE 20.47 13.24 11.73 13.01
MPCE less of health expenditure 3076 2514 2648 2581
95% confidence interval [2999–3153] [2461–2567] [2618–2679] [2549–2612]
Gini index (MPCE) 0.39 0.37 0.36 0.37
Mean years of schooling of household members (60+) 7.59 8.77 8.9 8.79
Wealth index* [95% CI] 30.1 [29.2–31] 43.2 [42.3–44] 41.1 [40.1–42.1] 41.1 [40.4–41.7]

*Wealth index was created by taking combined wealth score

Journal of Social and Economic Development
Journal of Social and Economic Development

4500
4000 3785
3395
3500 3146
3054 3003
3000 2716
2500
2000
1500
1000
500
0
Households with only elderly Households with both elderly Households with no elderly
members and non elderly members members
MPCE MPCE less of health expense

Fig. 2  Comparison of MPCE and MPCE less of health expense among adult households (45+) by type of
households, in India, 2017–2018

Fig. 3  Probability distribution function of MPCE by only elderly, both elderly and non-elderly and non-
elderly households in India, 2017–2018

Table 4 presents the composite index by socio-economic and demographic characteris-

tics and by type of elderly households in India. At 54.9 (CI 53.9–55.9), the index value was
the highest for households with no elderly members, followed by 53.5 (CI 52.9–54.1) for
households with both elderly and non-elderly members, and 51.6 (CI 50.8–52.5) for house-
holds with only elderly members. The composite index was higher in the urban areas, and
the pattern was similar across different types of households. Households with a BPL card
had a lower index value compared to those that did not have the card across all three types

13
 Journal of Social and Economic Development

1
.9
.8
Cumulative Probability

.7
.6
.5
.4
.3
.2
.1
0
0 2000 4000 6000 8000 10000
Monthly Per Capita Income

Households without elderly members Households with only elderly members

Households with elderly and non-elderly members

Fig. 4  Probability distribution function of MPCI by only elderly, both elderly and non-elderly and non-
elderly households in India, 2017–2018

Table 2  Minimum and maximum Minimum Maximum Mean (SE)

values of MPCE, MPCI, and
wealth index, India, 2017–2018
Log (MPCE) 4.60 9.55 7.90 (0.69)
Log (MPCI) 4.60 10.35 7.76 (1.15)
Wealth index 0.00 14.59 6.20 (2.38)

Table 3  Index of MPCE, index of MPCI, wealth index, and composite index and 95% CI by household
composition among middle-aged adults and elderly in India, 2017–2018
Index Households with Households with elderly Households All households
only elderly mem- and non-elderly members without elderly
bers members

Index of MPCE 67.6 (66.6–68.6) 64.3 (63.7–64.9) 66.4 (65.3–67.5) 65.4 (64.9–66.0)
Index of MPCI 57.0 (55.7–58.3) 52.7 (52.0–53.5) 56.9 (55.9–57.9) 54.7 (54.1–55.3)
Wealth index 30.1 (29.2–31.0) 43.2 (42.3–44.0) 41.1 (40.1–42.1) 41.1 (40.4–41.7)
Composite index 51.6 (50.8–52.5) 53.5 (52.9–54.1) 54.9 (53.9–55.9) 53.8 (53.3–54.4)

of households. The composite index declined monotonically with increase in household

size. Female-headed households had a lower index value compared to the male-headed
ones across all three types of households. The composite economic index for all three types
of households increased as the level of education of head of household increased from no
education to higher secondary.

13
 Table 4  Composite index by households’ composition and socioeconomic characteristics in India, LASI wave 1, 2017–2018
Background characteristic Composite index and 95% CI
Households with only elderly Households with both elderly and non- Households without elderly All households
members elderly members members

BPL card
No 53.8 (52.4–55.1) 56.8 (55.8–57.7) 57.6 (56.0–59.3) 56.8 (56.0–57.6)
Yes 49.0 (48.3–49.8) 49.3 (49–49.7) 51.9 (51.4–52.3) 50.3 (50.0–50.6)
Place of residence
Rural 48.3 (47.6–48.9) 49.8 (49.5–50.1) 51.1 (50.9–51.4) 50.1 (49.9–50.3)
Urban 62.9 (61.3–64.4) 61.5 (60.3–62.8) 62.5 (60.4–64.5) 62.0 (61.0–63.0)
Journal of Social and Economic Development

Household size
1–2 51.6 (50.7–52.4) 54.5 (53.5–55.6) 58.3 (57.3–59.3) 54.1 (53.5–54.6)
3–4 56.1 (51.8–60.3) 55.6 (54.6–56.6) 58.0 (55.9–60.1) 56.8 (55.6–58.1)
5–6 56.1 (53.4–58.8) 54.5 (53.2–55.8) 52.5 (52.1–53.0) 53.8 (52.9–54.7)
7 and more – 49.9 (49.4–50.4) 48.1 (47.5–48.7) 49.4 (49.0–49.8)
Sex of head of household
Male 52.8 (51.9–53.7) 53.6 (53.0–54.2) 55.3 (54.1–56.5) 54.2 (53.6–54.8)
Female 49.1 (47.7–50.4) 53.0 (50.8–55.2) 53.1 (52.1–54) 52.2 (51.1–53.3)
Marital status of head of household
Currently married 53.1 (52.1–54) 53.8 (53.1–54.5) 55.4 (54.2–56.6) 54.3 (53.7–55.0)
Others 49.5 (48.3–50.6) 52.7 (51.3–54) 52.9 (52–53.9) 52.1 (51.4–52.9)
Educational level of head of household
No education 46.8 (46.2–47.5) 47.9 (47.5–48.3) 51.5 (48.4–54.5) 49.1 (47.8–50.3)
Primary 50.3 (48.1–52.5) 51.0 (50.4–51.5) 53.1 (52.6–53.7) 51.7 (51.3–52.1)
Secondary 59.1 (57.6–60.6) 55.6 (54.4–56.8) 56.3 (55.7–56.9) 56.0 (55.3–56.7)
Higher secondary 71.5 (69.7–73.4) 64.0 (62.3–65.7) 63.8 (63.0–64.5) 64.3 (63.3–65.4)
Caste of head of household
Scheduled tribe (ST) 49.8 (44.2–55.4) 46.3 (45.6–47.0) 48.5 (47.5–49.4) 47.6 (46.7–48.4)

13
Scheduled caste (SC) 48.3 (47.2–49.4) 49.3 (48.7–49.9) 52.4 (51.7–53.2) 50.4 (50.0–50.9)
 Table 4  (continued)
Background characteristic Composite index and 95% CI
Households with only elderly Households with both elderly and non- Households without elderly All households

13
members elderly members members

Other backward class (OBC) 50.4 (49.3–51.5) 54.1 (52.9–55.3) 55.7 (53.5–57.8) 54.3 (53.2–55.3)
None of above 57.6 (56.1–59.0) 57.7 (56.9–58.4) 58.4 (57.9–58.9) 57.9 (57.5–58.4)
Religion of head of head of household
Hindu 51.6 (50.6–52.5) 53.6 (52.9–54.3) 54.3 (54.0–54.7) 53.6 (53.3–54)
Muslim 50.1 (47.6–52.5) 52.1 (49.6–54.5) 57.4 (51.4–63.4) 54.2 (51.1–57.4)
Christian 53.6 (51.3–55.9) 51.9 (50.5–53.3) 53.8 (52.8–54.9) 52.9 (52.0–53.7)
Other 55.2 (51.6–58.8) 57.8 (56.7–58.8) 58.6 (57.2–59.9) 57.8 (57.0–58.6)
Wages/salary received
Yes 51.9 (51.0–52.8) 53.7 (53–54.4) 55.5 (54.3–56.8) 54.4 (53.7–55.1)
No 51.6 (50.4–52.7) 53.1 (52.1–54.1) 53.1 (52.3–54.0) 52.8 (52.2–53.4)
India 51.6 (50.8–52.5) 53.5 (52.9–54.1) 54.9 (53.9–55.9) 53.8 (53.3–54.4)
Journal of Social and Economic Development
Journal of Social and Economic Development

Table 5  Estimated coefficients of ordinary least square (OLS) regression analysis of composite index with
selected socioeconomic characteristics, India, LASI wave 1, 2017–2018
Background characteristics Coeff. p-value 95% confidence interval

Household composition
No elderly members in h­ ousehold®
Only elderly members in household − 0.82 0.09 [− 1.74 to 0.11]
Both elderly and non-elderly members in − 0.55 0.06 [− 1.13 to 0.03]
household
Total members in house
Household size − 0.38 < 0.001 [− 0.48 to − 0.28]
Place of residence
Rural®
Urban 6.13 < 0.001 [5.29 to 6.96]
BPL card holder
No®
Yes − 3.89 < 0.001 [− 4.58 to − 3.2]
Caste
Schedule tribe (ST)®
Scheduled caste (SC) 2.20 < 0.001 [1.36 to 3.05]
Other backward class (OBC) 4.13 < 0.001 [3.28 to 4.97]
Others 6.09 < 0.001 [5.13 to 7.05]
Religion
Hindu®
Muslim − 0.65 0.39 [− 2.11 to 0.82]
Christian 0.48 0.32 [− 0.45 to 1.41]
Others 1.19 0.03 [0.12 to 2.26]
Sex of head of the household
Male®
Female − 0.43 0.49 [− 1.67 to 0.8]
Level of education
No ­education®
Primary 1.61 < 0.001 [0.79 to 2.44]
Secondary 4.07 < 0.001 [3.01 to 5.13]
Higher secondary 9.64 < 0.001 [8.45 to 10.83]
Marital status
Currently ­married®
Others − 0.39 0.22 [− 1.02 to 0.23]
Wages/salary received
Yes®
No − 2.46 < 0.001 [− 3.14 to − 1.79]

Table 5 gives the result of the regression analysis, with composite index as
the dependent variable. Households with only elderly members (− 0.82, 95% CI
− 1.74–0.11) and those with both elderly and non-elderly members (− 0.55, 95% CI
− 1.13–0.03) had significantly lower values compared to households with no elderly
members. With every additional member, the composite index value was likely to

13
 Journal of Social and Economic Development

reduce by 0.38, suggesting that household size was negatively associated with the eco-
nomic well-being of middle-aged and elderly households. Residents of urban areas
had higher index values than rural residents. The composite index was lower for BPL

Table 6  Ordered probit regression model estimates of effect of household and background characteristics
on economic condition of households, India, LASI wave 1, 2017–2018
Background characteristics Coeff. p-value 95% confidence interval

Household composition
No elderly members in h­ ousehold®
Only elderly members in household − 0.15 < 0.001 [− 0.24 to − 0.06]
Both elderly and non-elderly members in 0.11 < 0.001 [0.05 to 0.17]
household
Combined economic index
Combined economic index 0.04 < 0.001 [0.04 to 0.04]
Total members in house
Household size 0.06 < 0.001 [0.05 to 0.07]
Place of residence
Rural®
Urban 0.07 0.11 [− 0.01 to 0.15]
BPL card holder
No®
Yes − 0.12 < 0.001 [− 0.18 to − 0.06]
Caste
Schedule tribe (ST)®
Scheduled caste (SC) − 0.15 < 0.001 [− 0.25 to − 0.05]
Other backward class (OBC) − 0.07 0.16 [− 0.16 to 0.03]
Others − 0.08 0.10 [− 0.18 to 0.02]
Religion
Hindu®
Muslim − 0.15 < 0.001 [− 0.25 to − 0.05]
Christian − 0.04 0.39 [− 0.15 to 0.06]
Others − 0.13 0.03 [− 0.24 to − 0.01]
Sex of head of the household
Male®
Female 0.06 0.19 [− 0.03 to 0.15]
Level of education
No ­education®
Primary 0.09 0.01 [0.03 to 0.15]
Secondary 0.18 < 0.001 [0.11 to 0.25]
Higher secondary 0.44 < 0.001 [0.3 to 0.57]
Marital status
Currently ­married®
Others − 0.05 0.30 [− 0.14 to 0.04]
Wages/salary received
Yes®
No 0.32 < 0.001 [0.26 to 0.38]

13
Journal of Social and Economic Development

card holders and for those who were not receiving any wage/salary compared to their
counterparts. Compared to the scheduled tribe households (the reference category),
households belonging to the categories of scheduled caste, other backward classes, and
“others” had a higher composite index. The level of educational attainment of head of
household was found to be positively and significantly associated with the economic
condition. There were no significant differences in the overall economic well-being by
religion, marital status, and sex of head of household.
Table 6 presents the results of the ordered probit regression of households’ perceived
economic status (subjective well-being) by household composition, composite index, and
other sociodemographic factors. Households having only elderly members were less likely
to report better economic conditions (− 0.15, CI − 0.24 to − 0.06) compared to those with-
out an elderly member. However, households having both elderly and non-elderly members
were more likely to report better economic conditions (0.11, CI 0.05–0.17). The likelihood
of reporting higher economic conditions increased with the increase in the composite index
of the household (0.04, CI 0.03–0.04). Those having a BPL card were less likely to report
better economic conditions compared to those who did not have the card. Education was
found to be a significant predictor of perceived economic status. Marital status and sex
of head of household were not significant predictors of perceived economic status among
middle-aged and elderly adults in India.

Discussion

Households in India are the main caregivers to the elderly, and the economic well-being
of households is a key determinant of elderly health and overall well-being. The elderly
are a particularly vulnerable section of the population due to the rise in economic insecu-
rity, decline in health, increase in medical spending, weakening of family system, and low
social protection. In this context, this paper examined the economic well-being of middle-
aged and elderly adults by household composition in India. Households were classified into
three mutually exclusive groups: households with only elderly members, households with
both elderly and non-elderly members, and households with no elderly member. This clas-
sification was guided by the fact that middle-aged and elderly adults are not a homogenous
group and that their economic independence is one of the key determinants of their living
arrangement. In contrast to the conventional analyses that focus only on per capita con-
sumption as a measure of economic conditions, we measured economic well-being using a
composite index that included per capita consumption, per capita income, and household
assets and amenities. The following are the key findings of this study.
First, in the composite index of economic well-being, middle-aged and elderly adults
ranked 54 on a 0–100 scale. The average score of per capita consumption was higher
than that of income and wealth among middle-aged and elderly adults. This is possibly
due to the high share of health expenditure in the consumption basket of middle-aged and
elderly adults in India. The per capita health expenditure of middle-aged and elderly adults
accounted for 13% of the per capita consumption expenditure, which was higher than that
of the overall population. An earlier study estimated the per capita health expenditure at
6.2% of the consumption expenditure for all households in India (Mohanty et al. 2018).
Second, the overall economic well-being of households varied by household composi-
tion. The composite index of economic well-being of households with only elderly mem-
bers was lower than that of households with both elderly and non-elderly members and

13
 Journal of Social and Economic Development

of households without elderly members. The proportion of households with a wage or a

salary was the lowest among elderly-only households. Many of the elderly were out of the
workforce due to retirement or inability to work. Elderly households were also poorer in
terms of wealth accumulation compared to non-elderly households. The share of health
expenditure was also the highest among households with only elderly members compared
to households without any elderly members. Our findings suggest that the living conditions
of older and elderly adults have a strong age gradient.
Third, households with only elderly members were a heterogeneous group that included
single member widow/widower households that were economically poor. Elderly persons
live alone when their children are away, if they do not have offspring, if they prefer to live
separately because of financial independence, or if their children get them separated. A
large proportion of elderly persons have no income or pension, which compels them to live
with their children.
Fourth, among other factors, the composite index of economic well-being of rural
households, female-headed households, and households with low levels of income were
significant predictors of elderly subjective well-being. These findings are a strong confir-
mation of previous assessments that in many countries, in the absence of social protection
systems and due to the low asset holding, savings are not sufficient to guarantee adequate
income to the elderly persons until the end of their lives. This makes older persons particu-
larly vulnerable to economic insecurity as well as to poverty and poor health, with limited
options for escape (Kumar 2003; Bloom et al. 2010; UN 2013).
Our findings call for a strengthening of the social security measures for elderly house-
holds in India. It may be mentioned that various old-age pension schemes, as a part of
the social security programme, are being implemented in all the states of India. In 1995,
the Government of India initiated the National Old Age Pension Scheme as a social assis-
tance program for the poor older adults. It covered all persons over 65 years, and under this
scheme both men and women were given an amount of Rs.75 per month. The National Old
Age Pension Scheme was renamed as Indira Gandhi National Old Age Pension Scheme
(IGNOAPS) in 2007. The amount of pension under IGNOAPS was raised from Rs. 75 to
Rs. 200/-per month per beneficiary, and there was a provision where the state governments
may contribute over and above to this amount. The primary aim of this program was to
provide social security through proving financial assistance to its beneficiaries, including
senior citizens, widows, and disable people.
In 2011–2012, it reduced the upper age limit from 64 to 60 years. In 2021–2022, under
IGNOPAS, a total of 22.1 million beneficiaries were covered and a sum of 4864 crore were
disbursed under direct benefit transfer (NSAP 2023). The Indira Gandhi National Widow
Pension Scheme, implemented in 2009, provides pension to widows aged 40–59 years from
below poverty line households. In 2021–2022, a total of 6.7 million beneficiaries were cov-
ered under IGNWPS and a sum of 1453 crore were disbursed under direct benefit transfer.
The monetary assistance provided under these different schemes is too small and not regu-
lar enough to meet the needs of the elderly (Kohli et al. 2017). Studies suggest that elderly
persons finding it difficult to prove their age to avail the benefits (Barnhart and Peñaloza
2013).
A number of other pension schemes are also being implemented through various finan-
cial institutions that provide higher interest rates to working and self-employed profes-
sionals as a part of the pension schemes. These include the Pradhan Mantri Vaya Vandana
Yojana (PMVVY), the National Pension System (NPS), the Senior Citizen Savings Scheme
(SCSS), etc. However, unlike IGNOAPS, the beneficiaries need to invest money for getting
returns in old age. In addition, the government of India introduced the Annapurna scheme.

13
Journal of Social and Economic Development

Under this scheme, people aged 65 years and above, those were not covered in Indira Gan-
dhi National Old Age Pension Scheme, are entitled to 10 kg of free rice every month in
the form of social help. Swadhar Scheme was launched by Union Ministry of Women and
Child Health Development in 2002. Under this scheme, shelter homes are provided to wid-
owed including elderly widowed. Food, clothing, counselling, legal help and training for
rehabilitation are other components of the scheme. In Vrindavan, three homes for widows
receive Swadhar assistance.
The Indian constitution and judiciary played a significant role in providing social and
financial security to older people from time to time. The issue of widowed elderly or elderly
staying alone has always been considered as one of the components of a multipronged pro-
gram. The Maintenance and Welfare of Parents and Senior Citizens (Amendment) Bill,
2019 stipulated that the older adults regardless of their marital status can demand food,
clothing, housing, safety and security, medical attendance, healthcare and treatment neces-
sary for the parents to lead a life of dignity from their biological or adapted children. This
bill gets rid of the upper limit of maintenance fees that was Rs. 10,000.
The central as well as the state governments need to enhance resource allocation for,
and utilisation of, the social security programmes. The amount of money allocated for the
social security of the elderly by various state and central governments is small, and the
allocated amounts are not always spent. The health insurance coverage is growing in India
but is still grossly inadequate for the financial protection of the elderly in low-income set-
tings. Providing comprehensive medicare facilities to the elderly can protect them from
financial catastrophe. In addition, it is crucial to increase, among elderly households, the
awareness of health protection schemes such as the Health and Wellness Centres (HWCs)
and the Pradhan Mantri Jan Arogya Yojana (PM-JAY) launched under the umbrella of
Ayushman Bharat in 2018.
Acknowledgements Authors thank Dr. M. H Suryanarayana, Professor at Indira Gandhi Institute of Devel-
opment Research (IGIDR), Mumbai, for guiding the economic section of LASI module and his thoughtful
comments on draft paper.

Author contributions SKM done conceptualisation and design of study and drafting the manuscript; SKM,
PS, and IN did analysis and interpretation of data;; PA and SKM contributed to critical revision of the
manuscript for important intellectual content.

Funding Not applicable.

Data availability Longitudinal Ageing Study in India (LASI) wave 1 data have been used in the current
study. Data are available for academicians, researchers and policy makers and can be requested online at:
https://​www.​iipsi​ndia.​ac.​in/​conte​nt/​data-​reque​st.

Code availability Not applicable.

Declarations
Conflict of interest The authors declare that they have no conflict of interest.

References
Arsenijevic J, Pavlova M, Rechel B, Groot W (2016) Catastrophic health care expenditure among older peo-
ple with chronic diseases in 15 European countries. PLoS ONE 11(7):e0157765
Barnhart M, Peñaloza L (2013) Who are you calling old? Negotiating old age identity in the elderly con-
sumption ensemble. J Consum Res 39(6):1133–1153

13
 Journal of Social and Economic Development

Bianchi SM (2014) A demographic perspective on family change. J Fam Theory Rev 6(1):35–44
Bloom DE, Mahal A, Rosenberg L, Sevilla J (2010) Economic security arrangements in the context of pop-
ulation ageing in India. Int Soc Secur Rev 63(3–4):59–89
Bloom D, Canning D, Sevilla J (2003) The demographic dividend: a new perspective on the economic con-
sequences of population change. Rand Corporation, California
Chakravorty S, Goli S, James KS (2021) Family demography in India: emerging patterns and its challenges.
SAGE Open 11(2):21582440211008176
Filmer D, Pritchett LH (2001) Estimating wealth effects without expenditure data—or tears: an application
to educational enrollments in states of India. Demography 38(1):115–132
Fuster V (2017) Changing demographics: a new approach to global health care due to the aging population.
J Am College Cardiol 69(24):3002–3005
Hughes ME, Waite LJ (2002) Health in household context: living arrangements and health in late middle
age. J Health Soc Behav 43(1):1
IIPS, HSPH and USC (2020) Longitudinal ageing study in India (LASI), wave1, 2017–18, India Report.
International Institute for Population Sciences, Mumbai
Isherwood LM, King DS, Luszcz MA (2017) Widowhood in the fourth age: support exchange, relationships
and social participation. Ageing Soc 37(1):188–212
Jeon GS, Jang SN, Rhee SJ, Kawachi I, Cho SI (2007) Gender differences in correlates of mental health
among elderly Koreans. J Gerontol B Psychol Sci Soc Sci 62(5):S323–S329
Kämpfen F, Wijemunige N, Evangelista B (2018) Aging, non-communicable diseases, and old-age disability
in low-and middle-income countries: a challenge for global health. Int J Public Health 63:1011–1012
Kastor A, Mohanty SK (2018) Disease-specific out-of-pocket and catastrophic health expenditure on hos-
pitalization in India: Do Indian households face distress health financing? PLoS ONE 13(5):e0196106
Kaur G, Singh S (2013) Changing patterns of marriage in Indian society. Indian J Econ Dev 9(3):261–267
Kohli C, Gupta K, Banerjee B, Ingle GK (2017) Social security measures for elderly population in Delhi,
India: awareness, utilization and barriers. J Clin Diagn Res JCDR 11(5):LC10
Kulkarni S, Raju S, Bammidi S (2016) Social security for the elderly in India. Building knowledge base on
ageing in India: Increased awareness, access and quality of elderly services thematic paper, 1
Kumar SV (2003) Economic security for the elderly in India: an overview. J Aging Soc Policy
15(2–3):45–65
Maestas N, Mullen KJ, Powell D (2016) The effect of population aging on economic growth, the labor force
and productivity (No. w22452). National Bureau of Economic Research, Cambridge
Mohanty SK, Chauhan RK, Mazumdar S, Srivastava A (2014) Out-of-pocket expenditure on health care
among elderly and non-elderly households in India. Soc Indic Res 115(3):1137–1157
Mohanty SK, Kim R, Khan PK, Subramanian SV (2018) Geographic variation in household and cata-
strophic health spending in India: assessing the relative importance of villages, districts, and states,
2011–2012. Milbank Q 96(1):167–206
MoHFW (2019) Report of the technical group on population projection. Population projection for India and
States, 2011–36
National Social Assistance Programme (NSAP), Ministry of rural development, government of India.
https://​nsap.​nic.​in/. Accessed 8 Jan 2023
Pandey A, Kumar GA, Dandona R, Dandona L (2018) Variations in catastrophic health expenditure across
the states of India: 2004 to 2014. PLoS ONE 13(10):e0205510
Prince MJ, Wu F, Guo Y, Robledo LMG, O’Donnell M, Sullivan R, Yusuf S (2015) The burden of disease
in older people and implications for health policy and practice. Lancet 385(9967):549–562
Ruggles S, Heggeness M (2008) Intergenerational coresidence in developing countries. Popul Dev Rev
34(2):253–281
Rutstein SO, Johnson K (2004) The DHS wealth index. DHS comparative reports no. 6. ORC Macro,
Calverton
Rutstein SO (2015) Steps to constructing the new DHS wealth index. ICF International, Rockville, MD
Sheiner L (2014) The determinants of the macroeconomic implications of aging. Am Econ Rev
104(5):218–223
Srivastava A, Mohanty SK (2012) Poverty among elderly in India. Soc Indic Res 109(3):493–514
Szołtysek M, Gruber S, Zuber-Goldstein B, Scholz R (2011) Living arrangements and household formation
in an industrializing urban setting: Rostock 1867–1900. In: Annales de démographie historique, No. 2.
Belin, pp 233–269. Cairn/Softwin
Taylor P, Bain P (2005) ‘India calling to the far away towns’ the call centre labour process and globalization.
Work Employ Soc 19(2):261–282
UN. (2013) Inequality matters: report of the world social situation 2013. United Nations, New York

13
Journal of Social and Economic Development

United Nations (2013) Income poverty in old age: an emerging development priority. Department of Eco-
nomic and Social Affairs programme on ageing (UNDESA): social.un.org/ageing @UN4Ageing
WHO (2011) Global Health and Aging. https://​www.​who.​int/​ageing/​publi​catio​ns/​global_​health.​pdf
Yeo J, Lee YG (2019) Understanding the association between perceived financial well-being and life satis-
faction among older adults: Does social capital play a role? J Fam Econ Issues 40(4):592–608
Zimmer Z, Das S (2014) How skip-generation and other older person household compositions in Sub-Saha-
ran Africa relate to material well-being

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

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such publishing agreement and applicable
law.

Authors and Affiliations

Sanjay K. Mohanty1 · P. Arokiasamy2 · Itishree Nayak3 · Prashant Shekhar4

P. Arokiasamy
[email protected]; [email protected]
Itishree Nayak
[email protected]
Prashant Shekhar
[email protected]
1
Department of Population and Development, International Institute for Population Sciences,
Govandi Station Road, Deonar, Mumbai 400088, India
2
International Institute for Population Sciences, Govandi Station Road, Deonar, Mumbai 400088,
India
3
National Institute of Medical Statistics, Indian Council of Medical Research, Main Ring Road,
Ansari Nagar East, AIIMS Campus, New Delhi 110029, India
4
Population Research Centre, The Maharaja Sayajirao University of Baroda, Vadodara,
Gujarat 390002, India

13