Content uploaded by Jayati Ghosh
Author content
All content in this area was uploaded by Jayati Ghosh on Feb 01, 2016
Content may be subject to copyright.
The “Demographic Dividend” and Young India’s Economic Future
C.P. Chandrasekhar, Jayati Ghosh and Anamitra Roychowdhury
Jawaharlal Nehru University, New Delhi
Consistent and moderately high GDP growth rates during most years since 1980 have
encouraged optimistic projections about India’s future growth potential. The argument
is as follows: just as India transited from the “Hindu rate of growth” of around 3.5 per
cent during the first three decades of planned development to a higher growth
trajectory of close to 6 per cent over the last 25 years, it can now move on to a new
growth trajectory along which growth could average as much as 9 per cent per annum.
Needless to say, such projections must be based on an assessment identifying
potential sources of the new dynamism. One such assessment turns on conventional
sources of dynamism in developing countries during the post-War years. It points to
recent improvements in India’s merchandise export performance and the growing
attractiveness of India as a destination for global foreign direct investment (FDI)
flows. Attributing these to the effects of economic liberalisation it argues that once
this process crosses some critical threshold level, India could replicate the experience
of many East Asian countries and China.
A second assessment points to new sources of growth in financial, software and IT-
enabled services, with leadership performance in the latter two. The strength of this
source of dynamism is seen to be reflected in the fact that even at its relatively low
level of per capita income, services account for more than 50 per cent of GDP. India,
it is argued, is experiencing an altogether new trajectory of growth intensive in
intangibles such as knowledge and skills, because of its unique advantages in the
current conjuncture. To quote a study by the World Bank (2005: 10), India “has a
critical mass of skilled, English-speaking knowledge workers, especially in the
sciences. It has a well-functioning democracy. Its domestic market is one of the
world’s largest. It has a large and impressive Diaspora, creating valuable knowledge
linkages and networks.” It combines this with other key ingredients including:
“macroeconomic stability, a dynamic private sector, institutions of a free market
economy, a well-developed financial sector, and a broad and diversified science and
technology (S&T) infrastructure.” In this view: “Building on these strengths, India
can harness the benefits of the knowledge revolution to improve its economic
performance and boost the welfare of its people.”
A third assessment, which is the concern of this paper, while not necessarily denying
the importance of the other two, suggests that the effects of these are likely to be
amplified because of the demographic advantage that India currently has relative to
the developed countries and also countries like China. India is and for some time will
remain one of the youngest countries in the world. A third of India’s population was
below 15 years of age in 2000 and close to 20 per cent were young people in the 15-
24 age group. The population in the 15-24 age group grew from around 175 million in
1995 to 190 million in 2000 and 210 million in 2005, increasing by an average of 3.1
million a year between 1995 and 2000 and 5 million between 2000 and 2005. In 2020,
the average Indian will be only 29 years old, compared with 37 in China and the US,
45 in West Europe and 48 in Japan. The demographic process this implies would
create a large and growing labour force, which is expected to deliver spin-offs in
terms of growth and prosperity through a number of routes. The most obvious positive
1
effect is that a higher growth trajectory is not likely to run into bottlenecks set by
labour inadequacy. Of course, this assumes that the growing workforce of youth can
be trained to acquire the skills needed by the newer and technologically more
dynamic industries.
This notion of a “demographic dividend” overturns the older popular perception that a
large and “excess” population is a problem rather than a benefit from an economic
point of view (see for example Coale and Hoover 1958). For those in search of
simplistic explanations of underdevelopment, excess population has always provided
an easy way out. Large populations are seen to result in high levels of aggregate
consumption at even low levels of per capita income, resulting in small surpluses.
Since these surpluses must be spread thin across the population, their effects in terms
of growth of employment and income are seen as limited. Moreover, with limited
resources thinly spread because of large numbers, the tasks of feeding the population,
ensuring universal access to education and health and delivering basic services like
water and sanitation are seen as near impossible. The conclusion then is that the
growth of population has to be controlled if economic growth has to be triggered and
the quality of life improved.
The pithy response to this argument was that every mouth to be fed comes with two
hands that can be put to work. But in addition, a range of arguments have been
advanced against various versions of neo-Malthusianism (Crenshaw et. al. 1977;
Simon 1977; Kelley 1988). The problem of development has been seen as that of
employing more workers in more productive activities that can yield larger surpluses,
without depriving them of the basket of goods they currently consume. If the larger
surpluses thus garnered are invested, growth will accelerate. The economic problem
in poor countries then becomes that of identifying and implementing a strategy that
can make this happen. If the production of wage goods falls when workers are shifted
to new activities, or if those remaining in the wage goods sector increase their self-
consumption, the process of structural diversification can run into a wage goods
bottleneck. Once that problem is resolved, the concern shifts to ensuring the
availability of the capital goods needed for the labour employed in newer and more
productive activities.
The Demographic Dividend
More recently, however, a view has gained ground that what matters is not the size of
the population, but its age structure. A population “bulge” in the working age groups,
however large the total population, is seen as an inevitable advantage characterised as
a “demographic dividend”. This is why India - which is beginning to be characterised
by such a bulge - is seen as advantaged, despite its large population. This has
provided one more “macroeconomic” argument to those who see India emerging as a
regional (or even global) power in the not too distant future, possibly even displacing
China as potential world leader.
In this view, the demographic dividend is also expected to resolve the problem of
garnering surpluses over consumption needed for investment. A nation’s population
can be divided into those in the labour force (say, the 15-64 age group) and those
outside it. Given the availability of work and the resulting employment, the division
broadens to include those outside the labour force, those available for work but
unemployed and those in the actual workforce. Since those outside the work force
2
would be consuming part of what is produced by currently employed workers, the
ratio of those outside the workforce to those in it (the dependency ratio) would be
among the factors influencing the surplus available for investment after current
consumption. Hence, everything else remaining the same, the higher the share of
workers to non-workers, the larger would be the surplus. For given unemployment
rates, the higher the ratio of those in the labour force to those outside it, the larger
would be the surplus.
Effects of the Demographic Transition
Observed demographic trends suggest that both the size and age structure of the
population (and therefore the dependency ratio) in all countries tend to change over
time because of the nature of the demographic transition. The latter is characterised by
the fact that death rates tend to decline before declines in the birth rate set in. Initially,
the death rate falls because of declines in infant and child mortality resulting from
improved “public health interventions related to water and sanitation, and to medical
interventions such as vaccine coverage and the use of antibiotics” (Bloom and
Canning 2004: 11) Improved knowledge and reduced costs allow for these factors to
be exploited even at relatively low levels of per capita income so long as political
pressure or the political will to provide basic social services and rudimentary health
facilities exist. At a later stage the decline in the death rate and increases in average
life expectancy result from reduced death rates in the middle and older age groups
because of higher incomes, improved lifestyles and better and more expensive
medical technology.
As compared with this, birth rate reductions depend on the age of marriage and the
fertility rate. Both of these depend on the level of development to a far greater degree.
Development often leads to the dilution of social norms prescribing early marriage,
and fertility rates within marriage decline as higher child survival rates, female
education and labour market opportunities associated with development reduce the
desired family size. Of course, social policy can make a substantial difference to child
survival rates, and female education and family planning programmes can influence
the desired fertility rate. But in general, the observed decline in birth rates tends to
begin well after the decline in death rates sets in.
The difference in the relationship between death and birth rates on the one hand and
development on the other affects not just the rate of population growth but the age
structure of the population. The initial fall in infant mortality and improvement in
child survival results in a boom generation, with a larger number of people in the
youngest ages. After some time, the lagged fall in fertility rates reverses the baby
boom, resulting in a “bulge” in the younger ages. (However, the earlier baby boom
has its echo around 25 years later as the earlier baby boomers have their own
children.) As is to be expected, the bulge created by the baby boom moves up the age
structure, so that at some point the population in the working age (15-64) is much
higher than it was previously or will be subsequently. Finally, the bulge enters the old
age bracket, as is happening in the developed countries and epitomised by Japan
currently.
Implications for growth
3
Given the implications of the dependency ratio for surpluses available for investment
and growth, it should be obvious that this shifting age structure can have significant
consequences for economic growth. Periods characterised by low dependency ratios
would be characterised by higher growth, if the inducement to invest surpluses exists.
Conversely, periods characterised by high dependency ratios would be characterised
by a slowing of growth, unless productivity increases raise the output of a smaller
proportion of workers enough to neutralise the demographic deficit. But if the
window of opportunity available when the population bulge enters the working age
groups is to result in an acceleration in growth, the processes of development which in
part created this bulge must have been such as to ensure that the quality of those
entering the workforce is of the desired level and that these workers find employment
opportunities as and when they enter the labour force.
Despite the demographic determinism that characterises the work of those who
emphasise the significance of the demographic dividend, many of them admit that
there is no guarantee that the benefits of the “window of opportunity” created by the
population bulge will be exploited. To quote Bloom and Canning (2004: 22-23), “both
empirically and theoretically there is nothing automatic about the link from
demographic change to economic growth. Age distribution changes merely create the
potential for economic growth. Whether or not this potential is captured depends on
the policy environment”. Thus, while East Asia’s macroeconomic performance is seen
as being tracked quite closely by its demographic transition, with as much as a third of
its “miracle growth” estimated to be on account of the demographic dividend, Latin
America is seen to have “stumbled” during the 1950s and 1980s, when its
demographic trends resembled those in East Asia.
These caveats notwithstanding, there is indeed an element of automaticity about the
relationship between demographic trends and economic outcomes in the
“demographic dividend” school of thought. This element is built on a supply side
understanding of the determinants of economic growth, with a high savings rate seen
as a prerequisite and investments in education and health to ensure the quality of the
workforce seen as necessary facilitators. Beyond this, policy matters only inasmuch as
it should not privilege an inward-looking strategy, but rather encourage an export-
oriented one, which also is seen as a prerequisite for growth. That is, policy matters
only in the sense that it should not subvert the dividend, but rather facilitate the
realisation of its fruits.
The relation between demographic trends, savings and growth is illustrated, for
example, by the analysis of the Taiwanese success by Bloom and Canning (2004).
Their basic understanding of the East Asian miracle is (correctly) informed by the
view that increased productive investment and expenditure on education, rather than
increases in productivity, were the proximate determinants of the East Asian miracle.
Explanations of the East Asian model therefore must revolve around the factors that
resulted in higher investment. To some, the role of an interventionist State was crucial
to this outcome. But Blum and Canning attribute this to high savings rates, which are
presumed to be automatically invested in pre-Keynesian fashion.
Once savings are seen as determining investment (and not the other way around), the
demographic dividend argument rests on showing that high savings are related to
demographic factors. There are two steps through which this is achieved. First,
increased savings are seen as being due to rising life expectancy and the need to fund
retirement income. And the choice of retirement as an option when increased life
4
expectancy and health should encourage a lengthening of the working life span is
attributed to “mandatory or conventional retirement ages, coupled with the strong
financial incentives to retire that are inherent in many social security systems” (Bloom
and Canning 2004: 35). Taiwan is seen to have such a system.
Demography also affects educational investments, since greater longevity increases
the private incentive to invest in education by increasing the time span over which
such investments may be recouped. But all of this would be of little avail if the
government does not ensure a “flexible” economy that has the ability to absorb a
rapidly rising labour force. Thus the benefits of the demographic dividend depend on
“good policies”, captured by indices such as the openness of an economy or the
quality of government institutions. According to this literature, it is only when these
conditions do not prevail that the demographic potential remains unexploited.
Thus, the notion of the demographic dividend is used by its leading advocates as the
basis for advocating neoliberal economic policies, by using the argument that large
and young populations create a growth opportunity which can be exploited with open-
door strategies and flexible labour markets. This argument is supported with the
selective choice of empirical illustrations, despite numerous counter-examples. Thus,
despite rapid liberalisation of their economic policies, conscious export orientation
and the availability of a surplus workforce, Latin American countries went through a
Lost Decade in the 1980s and after a long period of typical neoliberal adjustment are
now striving to discard that framework. Governments in East Asian countries were
deeply protectionist and adopted strict interventionist measures, even when they
pursued mercantilist policies of export expansion involving discipline of the domestic
industrial class fostered by the State and supported with a range of incentives. So if
these countries are seen as having exploited the demographic dividend, this resulted
from the adoption of policies that were significantly interventionist.
To understand what kinds of policies can help exploit the window of opportunity
created by a demographic bulge in the working age groups, it is necessary to
recognise that the dependency ratio must be defined not as the ratio of the non-
working age to working age population but the ratio of actual non-workers to workers.
The difference between the two is determined by the extent of absorption into work of
the available labour force, which must take account of underemployment besides
unemployment. Since unemployment and underemployment are typically the outcome
of demand-side constraints, even if the presumption that increased longevity would be
accompanied by higher savings rates is right, there could be scenarios in which
investment rates fall short of savings rates and result in deflation rather than growth.
What is more, even in periods when the population bulge is not in the working age
groups, we may have large scale unemployment and inadequate expenditure on
education both by the government and by households. This would only erode the
ability of countries to exploit the demographic dividend as and when it emerges.
The Indian case
India is indeed in the midst of a process where it faces the window of opportunity
created by the demographic dividend. During the first two decades of post
Independence development, while infant mortality rates fell significantly, the fertility
rate remained more or less stagnant. This would have increased the population of
young people significantly, merely because of greater child survival. In the three
5
decades since then, though the fertility rate has been declining, the infant mortality
rate has fallen quite sharply, with possibly the same effect. One consequence of these
trends is the sharper fall in the crude death rate than the birth rate, though declining
mortality in the higher age groups would have influenced this as well.
Table 1: Trends in the Dependency Ratio in India (Medium variant)
Dependency
Ratio
Child
Dependency
Ratio
Old-age
dependency
ratio
1950 73 67 6
1955 74 68 6
1960 76 70 6
1965 78 72 6
1970 79 72 7
1975 77 71 7
1980 74 67 7
1985 72 65 7
1990 69 62 7
1995 68 60 8
2000 64 56 8
2005 60 51 8
2025 48 36 12
2050 50 27 22
Source: Population Division of the Department of Economic and Social Affairs of the United Nations
Secretariat, World Population Prospects: The 2004 Revision and World Urbanization Prospects: The
2003 Revision, http://esa.un.org/unpp, accessed 27 December 2005.
The effect of these trends on the dependency ratio is presented in Table 1, which
shows that while the total dependency rose initially because of a rise in the child
dependency ratio and stagnation in the old-age dependency ratio, it began to fall from
79 in 1970 as the child dependency ratio fell with the baby boomer generation moving
into working age groups and with old-age dependency rising only marginally because
of reduced death rates in older age groups. It is estimated to have fallen to 64 in 2005.
Thus India had actually begun to reap the demographic dividend around the mid-
1970s. But the process is likely to extend well into this century with the dependency
ratio projected to fall to 48 in 2025 because of continued fall in the child dependency
ratio. It would then start rising to reach 50 by 2050 because of an increase in the old-
age dependency ratio as the bulge moves forward and the death rate in the older
income groups declines. This suggests that the window of opportunity offered by a
population bulge had clearly opened for India by the mid-1970s or early 1980s.
It is true that there are different projections of the future age structure of the
population. As Table 2 shows, the United Nations Population Division’s Population
Database expects the youth population (15-24 age group) to begin to decline only by
2025. However, the official Indian projections based on the 2001 Census and the SRS
suggest that the decline may set in earlier. Going by the Registrar General’s figures,
India’s youth population stood at 195.1 million in 2001 and was expected to increase
by an annual average figure of 5.4 million during 2001-06 and 3.5 million during
2006-11, before beginning to decline.
6
Table 2: Average annual increment in youth population
UN Population Database Registrar General of India
Total Male Female Total Male Female
2001-06 4028.8 2022.6 2006.2 5406.2 3025.2 2381.2
2006-11 2680.8 1330 1350.4 3532.4 1670.4 1861.8
2011-16 1312.6 621.2 691.8 -212.8 -343.6 131
2016-21 226.4 76 150.4 -1857.4 -777.2 -1080.2
2021-26 45.2 -8.8 54 -1112 -290.2 -821.6
Source: “World Population Prospects: The 2004 Revision” available at http://esa.un.org/unpp/
and Government of India, Planning Commission.
It must be noted that the beginnings of the population bulge in the younger age groups
in India did coincide with the shift to a new growth trajectory noted earlier in this
paper. But was this the result merely of the demographic dividend and its supply side
effects on the rate of savings? And is the ability to exploit the dividend the result of
the more open, export-oriented strategy adopted by the government?
Chart 1 provides estimates of gross domestic capital formation and gross domestic
savings as percentages of GDP from 1980-81 onwards. These point in three
directions. First, both savings and investment rates have generally increased over
time, as part of a trend of much longer duration, whereby savings and investment rates
have tended to increase with economic development, in an Engels curve type pattern
in which increased aggregate incomes also allow for a larger share for savings.
Second, this tendency is not so visible during the years of accelerated liberalisation in
the 1990s, when amidst much volatility the savings rate seems to have stagnated.
Third, there appears to be a clear return to the tendency for savings and investment
rates to rise during the first three years of this decade, when there seems to be a
reassertion of the broad historical tendency noted earlier. However, this very recent
increase in both rates has involved savings rates which are higher and growing faster
than domestic investment rates, reversing the historical pattern, and suggesting that
there are other factors constraining investment in India despite the increased
economic openness and the advantages of the population bulge.
Chart 1: Savings and investment rates
(Gross domestic savings and gross domestic investment as per cent of GDP
at current market prices)
7
Savings and investment rates
15
17
19
21
23
25
27
29
1980-81
1981-82
1982-83
1983-84
1984-85
1985-86
1986-87
1987-88
1988-89
1989-90
1990-91
1991-92
1992-93
1993-94
1994-95
1995-96
1996-97
1997-98
1998-99
1999-00
2000-01
2001-02
2002-03
2003-04
Savings rate Gross investment rate
Source : CSO National Accounts Statistics, various issues
So did the economic “reforms” of the 1990s help exploit the demographic dividend
and spur economic growth? Advocates of reform have often argued that, whatever
else may be said about the effects of the reform process, it cannot be denied that it has
helped the Indian economy move to a higher annual average rate of growth. But it can
be convincingly argued that the transition to a higher economic growth trajectory
occurred well before the reforms of the 1990s. (Chandrasekhar and Ghosh, 2004)
Thus, as indicated in Chart 2, the transition to a high rate of growth occurred during
the 1980s, when liberalization was limited and halting, and not in the 1990s, when the
pace of liberalization was substantially accelerated and was far more widespread.
Chart 2: Average annual growth rate of GDP (at constant prices) during Plan periods
8
Annual growth rate of GDP at constant prices
during recent Plan periods (per cent)
3.4
5
5.5
5.8
3.3
6.8
5.6
0
1
2
3
4
5
6
7
8
IVth Plan
(1969-74)
Vth Plan
(1974-79)
VIth Plan
(1980-85)
VIIth Plan
(1985-90)
2 Annual
Plans
(1990-92)
VIIIth Plan
(1992-97)
IXth Plan
(1997-2002)
Source: Economic Survey, Government of India, 2005-06
Of course, it could be argued that the 1980s were also years of liberalization, which is
why aggregate income growth was higher in this decade as well. However, the
primary factor behind the accelerated growth of the 1980s was the fiscal stimulus
provided by the State, financed by growing internal and external public borrowing.
Exploiting the access to foreign exchange afforded by the rise to dominance of
finance internationally, the government chose to pump prime the system. Rising
government expenditure, however, was not accompanied by an increase in resource
mobilization through rising taxes. The fiscal stimulus was financed through rising
deficits, including a rising deficit on the revenue account of the government’s budget.
The demand stimulus resulting from such expenditure was serviced by domestic
industry with the help of imported capital goods, intermediates and raw materials,
imports of which were liberalized. This essentially meant that the import intensity of
domestic production rose. But such growth was not constrained by inadequate access
to foreign exchange, since it was accompanied by an increase in foreign borrowing
from the IMF, the international commercial banking system and non-resident Indians.
This expansion proved to be unsustainable, culminating in the crisis of 1990-91. The
subsequent expansion from the mid-1990s witnessed an almost parallel build up of
internal public debt and a balance of payments situation that was even more volatile
because of the greater presence of short term capital inflows. The very recent period
has been characterised by another anomaly: the co-existence of relatively high growth
rates with continuing slack in the economy, reflected in the ex ante excess of domestic
savings over domestic investment, which in turn is expressed in growing holding of
foreign exchange reserves by the Reserve Bank of India.
The Demographic Dividend and Employment
9
Moreover, even to the extent that growth has occurred, it has not been such as to
absorb the rapidly rising labour force being generated by the demographic dividend.
As Table 3 shows, the period between 1993 and 2000 showed a dramatic deceleration
in employment generation, with the lowest rate of growth of rural employment in
post-Independence history. Even in urban areas, the rate of growth of employment
was significantly lower than the previous periods since the early 1980s. Subsequently,
the most recent NSS large survey indicates a recovery, although still not to the rates
achieved in the period between 1987-88 and 1993-94.
Table 3 : Growth rates of employment
(per cent change per annum)
Rural Urban
1983 to 1987-88 1.36 2.77
1987-88 to 1993-94 2.03 3.39
1993-94 to 1999-2000 0.66 2.27
1999-2000 to 2004-05 1.97 3.22
Source : Based on NSS employment rates (NSS 38th, 43rd, 50th, 55th and 61st Rounds)
and Census population figures and projections
Note : Employment here and in subsequent tables refers to all workers (principal status and
subsidiary status).
This in turn reflects an increase in labour force participation rates for both men and
women, as evident from Table 4. This includes both those who are actively engaged
in work and those who are unemployed but looking for work. But it should be noted
that this aggregate increase incorporates declining rates of labour force participation
among the youth, that is the age group 15-29 years, and a rise for the older age
cohorts.
10
Table 4: Labour force participation rates
Usual status (PS+SS) Current daily status
1993-
94
1999-
2000
2004-
05
1993-
94
1999-
2000 2004-05
Rural males 56.1 54 55.5 53.4 51.5 53.1
Rural females 33 30.2 33.3 23.2 22 23.7
Urban males 54.3 54.2 57 53.2 52.8 56.1
Urban females 16.5 14.7 17.8 13.2 12.3 15
Source: NSSO, Employment and Unemployment Situation in India, Sept 2006
Despite the growth of employment, unemployment rates have also been increasing,
and are now the highest ever recorded. Unemployment measured by current daily
status, which describes the pattern on a typical day of the previous week, accounted
for 8 per cent of the male labour force in both urban and rural India, and between 9
and 12 per cent of the female labour force, which is truly remarkable in a country that
provides nothing in the form of unemployment benefit or insurance.
The most significant change has been in the pattern of employment. There has been a
significant decline in wage employment in general, which includes both regular
contracts and casual work. While regular employment had been declining as a share
of total usual status employment for some time (except for urban women workers),
wage employment had continued to grow in share because employment on casual
contracts had been on the increase. But the latest data suggest that even casual
employment has fallen in proportion to total employment. In fact, the share of casual
labour has fallen for all categories of workers – men and women, in rural and urban
India. The sharpest decline has been in agriculture, where wage employment in
general has fallen at a rate of more than 3 per cent per year between 1999-2000 and
2004-05. But even for urban male workers, total wage employment is now the lowest
that it has been in at least two decades, driven by declines in both regular and casual
paid work. For women, in both rural and urban areas, the share of regular work has
increased but that of casual employment has fallen so sharply that the aggregate share
of wage employment has fallen. So there is clearly a real and increasing difficulty
among the working population, of finding paid jobs in any form.
This is probably the real reason why there has been a very significant increase in self-
employment among all categories of workers in India. The increase has been sharpest
among rural women, where self-employment now accounts for nearly two-thirds of all
jobs. But it is also remarkable for urban workers, both men and women, among whom
the self-employed currently constitute 45 and 48 per cent respectively, of all usual
status workers.
Impact on the youth
While aggregate labour force participation rates have risen, the same is not true of the
youth. As Table 5 shows, labour force participation rates have fallen quite
substantially for male rural youth, and not increased for young women in rural areas
either. In urban areas, there is a slight recovery of labour force participation rates from
the low levels of 1999-2000, but only for young women in the age group 20-24 years
is there evidence of any real increase.
11
Table 5: Labour force participation rates of youth,
according to Usual Status (principal and subsidiary activities)
15 – 19 years 20 – 24 years
Rural Males
1987-88 63.0 91.8
1993-94 59.8 90.2
1999-00 53.2 88.9
2004-05 52.9 89.1
Rural Females
1987-88 41.5 48.4
1993-94 37.1 46.9
1999-00 31.4 42.5
2004-05 33.1 43.5
Urban Males
1987-88 42.9 79.2
1993-94 40.4 77.1
1999-00 36.6 75.2
2004-05 38.1 76.9
Urban Females
1987-88 16.9 22.5
1993-94 14.1 23.0
1999-00 12.1 19.1
2004-05 14.4 25
Source: NSSO, Employment and Unemployment Situation in India, Sept 2006
It is certainly possible that falls in labour force participation among the youth result
from their delayed entry into the work force, partly because they are extending their
years of education. If this is so, it would be a positive sign, indicating a greater degree
of skill formation in the young labour force of the future. However, Chart 3 indicates
that this is not the dominant reason. Except for rural females, where the ratio was very
low to start with, there has been very little increase in the proportion of those
reporting themselves as usually engaged in education. For young urban females, there
was actually a decline in such a proportion.
12
Chart 3: Youth in education, according to usual status
Share of those in education in youth aged 15-29 years
(per cent)
19
10
30
23
20
13.4
30.5
20.3
0
5
10
15
20
25
30
35
Rural males Rural females Urban males Urban females
1999-2000
2004-05
Source: NSSO, Employment and Unemployment Situation in India, 55th and 61st Rounds
What is more disturbing is that, despite the fact that labour force participation rates
among the young population have decreased or not increased much (except for urban
women in the age group 20-24 years), open unemployment rates have increased.
Table 6 reveals that youth unemployment was substantially higher than
unemployment across all the working age population, and what is more it also
increased across all categories of young people – men or women, rural or urban. So
the youth are far more prone to be actively seeking work and not finding it. Given that
open unemployment by “usual status category” has generally been low in India
because of the absence of any sort of social protection for the unemployed, it is
disturbing to note that as many as 6-8 per cent of young rural males and 12-14 per
cent of urban male youth describe themselves as available for work and seeking it but
not finding it. The proportions of young women describing themselves as usually
unemployed are even larger.
13
Table 6: Unemployment rates among young people and overall population
Rural India
Urban India
15-19 20-24 All 15+ 15-19 20-24 All 15+
Males
Usual
Status
1993-94 3.3 4.9 2.0 11.9 12.6 5.4
1999-00 5.5 5.2 2.1 14.2 12.8 4.8
2004-05 7.9 6.2 2.1 14 12.5 4.4
Current
Daily
Status
1993-94 9.0 10.3 5.6 16.2 17.0 6.7
1999-00 13.1 11.7 7.2 19 17.1 7.3
2004-05 15 12.9 8.0 18.4 15.8 7.3
Females
Usual
Status
1993-94 1.9 2.8 1.3 12.8 21.7 8.3
1999-00 3.2 4.9 1.5 13.2 19.4 7.1
2004-05 6.7 9.3 3.1 15.6 25.8 9.1
Current
Daily
Status
1993-94 8.3 8.2 5.6 18.6 28.5 10.4
1999-00 12.8 12.1 7 18 25.9 9.4
2004-05 12.6 14.9 8.7 16.4 27.3 11.6
Source: NSSO, Employment and Unemployment Situation in India, 50th, 55th and 61st Rounds
The current daily status criterion describes the nature of activity on a typical day of
the reference week, and therefore can be thought of as a “flow” measure of work
possibilities. By this indicator, open unemployment levels for the young are truly
alarming, accounting for nearly 20 per cent of urban young men in the age group 15-
19 years and 30 per cent of urban women in the age group 20-24 years. These
numbers translate into an estimated 36 million young people of between 15 and 29
years who were “usually unemployed” at the start of 2005, and as many as 58 million
young people who were unemployed on any particular day.
If this is a true description of labour markets in India at present, it has significant
implications. One concern relates to the possibility of missing the window of
opportunity provided by a large young population, because the economic growth
process simply does not generate enough jobs to employ them productively. Another
important concern follows from this, in terms of the negative social impact of growing
numbers of young unemployed. If the economy does not generate adequate
employment of a sufficiently attractive nature, the demographics could deliver not a
dividend but anarchy.
The question of employability
This naturally brings to the forefront not only issues relating to increasing demand for
labour, but also the nature of the labour force and its employability. In this context,
the quantity, quality and relevance of education are all crucial. Several indicators
suggest that, during the liberalisation years, there has been a setback on the literacy
and education fronts, improvements in which are seen as not just conducive but even
necessary to exploiting the demographic dividend. As Chart 4 shows, the spread of
literacy has been slow during the years of globalisation and even in 2004-05 the
country was far short of achieving total literacy even in the more developed urban
14
areas. Indeed, the pace of improvement in literacy rates appeared to have decelerated
further in the first part of this decade.
Chart 4: Literacy rates
Literacy rates (per cent)
0
10
20
30
40
50
60
70
80
90
100
1983 1987-88 1993-94 1999-00 2004-05
Rural Males Rural Females Urban Males Urban Females
Source: NSSO, Employment and Unemployment Situation in India, 55th and 61st Rounds
In 2004-05, only 21.1 per cent of rural males and 10.2 per cent of rural females of 15
years and above had a minimum education of secondary school and above. In urban
areas, the education level was slightly better with 48.3 per cent of urban males and
35.6 per cent of urban females with at least that much education. However, only
around 1.5 per cent of persons aged 15 years or more in urban areas and less than 5
per cent in urban areas had technical qualifications of even the most rudimentary kind.
By no stretch of imagination can India currently be characterised as a knowledge
economy in any meaningful sense.
There is the further problem that even those who have been educated find it hard to
get jobs, whether these jobs are appropriate to their skills or otherwise. Educated
employment declined slightly for men between 2000 and 2005, but was still around 6
per cent for those with secondary school degrees and 7 per cent for graduates.
Unemployment among educated women was much higher and also got worse,
reaching rates of 34 per cent for rural female graduates, and 20 per cent for urban
women with high school and above.1
Vocational training appears to be doing little to resolve this problem. To begin with,
even in 2004-05 only a very small proportion of youth, less than 4 per cent, had
received any sort of vocational training. But also most such training apparently does
not increase employability. As Table 7 shows, the proportion that has received some
sort of vocational training is significantly higher among the unemployed than the
employed youth, by all categories.
1 Figures from NSSO, Employment and Unemployment Situation in India, September 2006.
15
Table 7: Per cent of youth aged 15-29 years who have received vocational training
Among all
youth
Among
employed
Among
unemployed
Among those not
in the labour
force
Rural males 2.7 2.8 9.6 1.4
Rural females 2.3 4.8 17.4 1.3
Urban males 6.5 7.2 16.6 4.2
Urban females 4.7 15.8 24 3.1
Source: Same as Table 5
The Gender Dimension
As has been repeatedly recognised but never adequately addressed, the challenge set
by a young population is all the greater in the case of females. The principal problem
is that participation in gainful economic activity is typically less for young women
than for young men. In addition, there are the problems of less access to education
and opportunities for skill development for young women.
Female labour force participation according to the usual activity criterion was just 33
per cent and 43 per cent in the 15-19 and 20-24 age groups in rural India in 2004-05.
This compared with 53 and 89 per cent respectively in the case of young men. The
situation was even worse in urban India with the percentages at 14 and 25 per cent
respectively for females, as compared with 38 and 77 per cent for males, although in
the urban case there is some evidence of increase in the rate for young women aged
20-24 years compared to the previous period. There are other sources of concern.
Despite lower labour force participation rates than men, there is higher incidence of
unemployment among young women. For young women in the age-group 20-24, open
unemployment rates by current daily status are as high as 16 per cent in rural areas
and 27 per cent in urban areas.
The greater divergence between male and female youth labour participation in urban
as compared with rural areas, points to the role that attitudes and perceptions play in
the determination of the role of women. Their greater presence in the rural labour
force is possibly the result of the availability of opportunities around the home and
farm, whereas their relative absence from the urban labour market must be influenced
by the common view that women should not be exposed to workplace environments
and that the woman’s role is not that of a breadwinner but an (unpaid) home worker.
Such perceptions tell on attitudes towards female education as well. The gender gap in
both literacy and education remains large, especially in some states, even though it
has been narrowing to some extent in recent years. This in turn affects employment
possibilities, although even for educated unemployment, young women show higher
rates than young men. The attitudes that underlie these outcomes in turn impinge on
the female experience with adolescence and youth itself. Gender-based expectations
of future prospects and potential employment typically put girls at a disadvantage in
many ways, such that these expectations become self-fulfilling.
IT as the solution
16
It is often held that the rapid growth of modern IT-driven services in India offers an
opportunity to exploit the demographic dividend. In fact there is an increasingly
popular perception that India would be able to encash the demographic dividend
through the growth of its IT and IT-enabled services sector. Of course, there is not
doubt that both in absolute and relative terms the size of the IT sector in India is now
impressive. The National Association of Software and Services Companies
(NASSCOM) estimated the size of the software and IT-enabled services industry in
2004-05 at $22.6 billion, comprising of $4.8 billion of domestic revenues, $13.1
billion of software and services export revenues and $4.6 billion of revenues from
exports of IT-enabled services and business process outsourcing (BPO).2 Placed in the
context of the economy as a whole, the sector’s revenues now amount to around 4.5
per cent of GDP. This makes it an important segment of the non-agricultural sector.
By way of comparison, the gross revenues from IT services was in 2004-05 about 20
per cent higher than the GDP generated in India’s construction sector and almost three
times as much as the GDP in mining and in electricity, gas and water supply. What is
more, gross revenues from IT services exceeded 12 per cent of GDP generated in
India’s services sector as a whole, which accounts for more than 50 per cent of the
nation’s GDP. Thus, even though the software and IT-enabled services sector started
from a small or negligible base a decade back, its rapid expansion at an annual
compound rate of more than 30 per cent per annum between 1998-99 and 2004-05 has
ensured that it is today an important presence in the economy.3
The fact that the rise to maturity of this sector has been driven predominantly by
external demand is also well recognised now. Exports of software and IT-enabled
services have risen at a compound annual rate of 38 per cent a year since 1997-98,
and overwhelmingly explain the rapid rise of the sector. In 2004-05 exports of
software and services as estimated by the Reserve Bank of India was, at $17.7 billion,
equal to about a fifth of India’s merchandise exports and higher than one of India’s
principal commodity exports, viz. textile and textile products (including carpets).
This has made IT services exports an important component of India’s total
(merchandise and non-merchandise) exports. The ratio of IT services to merchandise
exports has risen from 14 per cent in 2000-01 to an estimated 22.5 per cent in 2005-
06. Further, the ratio of net IT services export earnings to total net invisible earnings
rose from 53 to 59 per cent between those two years.4
However, the sector’s contribution to employment does not compare with its role in
the generation of income and foreign exchange. The only available estimates here are
those from NASSCOM, which indicate that employment rose from around 285,000 in
1999-2000 to just 1,287,000 in 2004-055, or at a compound rate of about 35 per cent
per annum. This is indeed remarkable given the fact that rate of growth of
employment during 1999-2000 to 2004-05 as per NSS statistics amounted to just 1.97
per cent in rural areas and 3.22 per cent in urban areas. However, these growth rate
figures conceal the low base from which employment has grown, making the absolute
contribution of the sector to employment minimal.
2 Figures from Indian http://www.nasscom.in/Nasscom/templates/NormalPage.aspx?id=28485,
accessed 18 November 2006.
3 Computed from figures from NASSCOM and CSO.
4 Computed from figures of national accounts and balance of payments statistics available in Reserve
Bank of India, Handbook of Statistics on the Indian Economy.
5 Figures from http://www.nasscom.in/Nasscom/templates/NormalPage.aspx?id=28487, accessed 18
November 2006.
17
The total IT industry, including both hardware and software elements, as well as IT-
enabled services, still employs only slightly more than 1 million workers, out an
estimated total work force in India of more than 415 million, and urban work force of
around 110 million. Total employment in this sector is far short of even the annual
increment in the youth workforce. This mismatch between the sector’s contribution to
GDP and its contribution to employment does suggest that, despite its high growth,
this sector can make only a marginal difference to employment even of the more
educated groups in urban areas.
Therefore, as of now, the expectation that the demographic dividend would itself
trigger processes that would help exploit its benefits does not seem to be warranted in
the Indian case. While the labour force is indeed expanding, thus far, the task of
absorbing an increasingly youthful workforce has been postponed rather than
undertaken. If the challenge is not met soon, the dividend can prove a liability.
Implications for health status of the population
Discussions on the economic implications of the demographic dividend often also
presume that the desired quality of the population bulge entering the workforce in
terms of health is ensured. While the changing age distribution of the population can
eventually lead to an increase in the supply of working age population, it may not
necessarily lead to an increase in productivity (or lower rates of absenteeism), without
significant improvements in the health status of the population (both of the working
and the non-working age). Secondly, from the view point of the savings-growth
causality, increased longevity in itself may not be accompanied by increased savings
of the working age population, because of the increase in disease burden across all
age groups (despite longer life spans) and rise in healthcare expenditures.
Although India accounts for 16.5 per cent of the world’s population, it contributes to a
fifth of the world’s share of diseases. The National Commission on Macroeconomics
and Health (2005) has classified health conditions into three major categories namely:
communicable diseases, maternal and child health conditions; non-communicable
diseases; and accidents and injuries. It has been projected that with a few exceptions
such as leprosy and blindness, incidence of the majority of health conditions
belonging to all the three categories will rise significantly by 2015. Diseases under the
first category accounted for nearly half of India’s disease burden in 1998. While the
burden on account of most of these diseases and deaths on account of malaria, TB,
diarrhoea, and other infectious diseases will reduce and leprosy will be eliminated,
HIV/AIDS and TB and drug-resistant malaria are projected to increase.
What needs to be noted is that the age distribution of prevalence varies with disease.
In some cases like asthma and tuberculosis prevalence is concentrated in the younger
and/or older age groups. In others like jaundice and malaria the distribution is more
even. But in areas like reproductive health, HIV/AIDS and possibly even mental
health, the prevalence of health problems is likely to be concentrated in the bulge.
The other health challenge posed by the specific phase of the demographic transition
that India is going through is that relating to infant mortality and reproductive health.
This is inevitable given the fact that the proportion of women in the reproductive age
group is projected to rise to a peak of nearly 55 per cent in 2016 and remain around
that level for the next decade. The increase in this proportion will also require
18
increased efforts to reduce not only maternal mortality but also the number of infant
and child deaths. This would require investment in infrastructure to ensure safer and
better facilities for child birth and advocacy to ensure utilisation of these facilities.
The nutritional problems associated with effective child-bearing remain significant
and widespread. Poor nutrition for girls during adolescence (and resultant anaemia
and inadequate calcium intake) contributes to future obstetric risk, adversely affects
the reproductive process and affects the health and development prospects of the
children as well. Yet under-nutrition and deficiencies of micro-nutrients may be
described as characteristic features of the condition of women and girl-children in
most parts of India. While there has been some improvement in nutritional conditions
over time, gender disparities in this area remain significant and have probably even
widened over time.
All this suggests that both expenditures related to both nutrition and reproductive
health will have to go up substantially, if there is to be any social and material benefit
from the demographic bulge.
Comparison with China
The evolving employment, education and health situation therefore does not warrant
great confidence in India’s ability to exploit the demographic “window of
opportunity”. Despite this, the demographic argument is routinely invoked to suggest
that India is likely to overtake China in economic performance. A typical example is a
recent survey of China by The Economist (Miles 2006: 12) which notes: “Goldman
Sachs, an investment bank, reckons that China’s “demographic bonus” of a large
working age population with a small number of dependents (helped by the one-child-
per-couple policy introduced in the late 1970s) will shortly run out as the number of
young workers starts declining. The country’s dependency ratio will begin to rise by
2010, whereas India, on current trends, will not reach that point until 2040.” It is also
often suggested that China’s shrinking workforce would soon raise wages to an extent
where it would lose its export competitiveness (Barboza 2006).
These suggestive remarks are based on three presumptions, each of which need to be
examined closely. The first is that China’s high growth in the period since the early
1980s can be explained in terms of the demographic dividend. According to Feng and
Mason (2005), the support ratio in China—defined as ratio of the effective number of
workers, or the population weighted by age-specific productivity weights, and the
effective number of consumers, or the population weighted to allow for variation in
consumption by age—increased by 28 per cent between 1982 and 2000 or at an
average annual rate of 1.3 per cent. During the same period the purchasing power
parity-adjusted GDP grew at an annual rate of 8.4 per cent. On this basis, it is held
that the demographic dividend accounted for 15 per cent of China’s economic growth.
Besides the fact that even this computation accounts for only 15 per cent of China’s
heady growth during the concerned period, what is important to note is that the
computation is based on the growth of the actual workforce and not the labour force.
As Table 8 indicates, the rate of growth of the workforce was robust in urban areas
and started slackening in rural areas only since the mid-1990s. Since this was
accompanied by significant migration to urban areas, the workforce participation rate
rose in urban areas till the mid-1990s and in rural areas throughout the recent period.
19
Thus China’s success is partly related to its ability to maintain relatively reasonable
employment growth rates even during the reform years, excepting for the more recent
period when the reform of state-owned enterprises has resulted in a substantial loss of
jobs (estimated at 20-30 million by The Economist and at 14 million over the five
years ending 2003 by the OECD). As noted above, this was not true of India where
the reform has been accompanied by a sharp deceleration in employment growth. This
would mean that despite the rise in the share of population in the working age group,
the “support ratio” in India may not increase as expected. However, an area of
concern for China is the decline in the urban employment-to-population ratio since the
early 1990s, as apparent from Chart 5.
Table 8: Employment growth in China
(average annual rates per cent)
Urban Rural
1984-87 4.04 2.71
1987-1994 4.91 3.67
1994-2000 3.84 0.02
2000-04 3.43 -0.15
Source: CEIC Database
Chart 5: Work force participation rates in China
Work force participation rates in China (per cent of population)
40
45
50
55
60
65
70
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Urban Rural
Source: CEIC Database
The second presumption often made in the kind of India-China comparisons noted
above is that investments in human capital in India would match that of China’s,
making the latter exploit the benefits of its growing workforce in the coming years, as
China did in the recent past. On the surface India and China seem to doing equally
well on the educational front, with the ratio of educational spending to GDP in India
matching that in China, and the public share in India being larger. But outcomes have
20
varied significantly. As compared with India’s record discussed above, as early as
1986, 85 per cent of the relevant age group in China was receiving six years of
primary education and this had risen to 100 per cent by 2000 (OECD 2005). The
graduation rate of junior secondary school rose from 41 per cent in 1986 to 85 per
cent in 2003. And the participation in education after the nine year level, has been
rising even if not as rapidly as for junior secondary schools. Going further, in 1999,
the government declared its intention to double the size of tertiary education
institutes, especially in technical fields of study. As a result the number of graduates
from higher education doubled between 2000 and 2003. A further 50 per cent increase
seems likely by 2006, based on enrolment rates since 2002. The emphasis on
technical subjects has meant that graduates in science and engineering amounted to
5.25 per cent of the relevant age group in 2003 and this ratio is rapidly rising.
Physical access to health facilities has also been improving rapidly in China.
According to the OECD (2005), China’s total health expenditure at 5.3 per cent of
GDP in 2002 is only slightly lower than in some OECD countries. The share of
private financing is higher in China than in most OECD countries, but substantially
lower than in India, which (at more than 80 per cent) has one of the highest ratios of
private to total health spending in the world. However, as in India, China during the
reform years has displayed a rising proportion of private spending in total health
expenditure. The share of public health outlays in on-budget spending in China has
fallen from 4.2 per cent in 1994 to 2.9 per cent in 2002 and represented only 0.6 per
cent of GDP. Thus this is an area in which both countries need to make significant
advances, though with much better physical access, a rapidly rising per capita income
and slower rate of population increase China has some advantages in this respect.
Finally, the presumption is that India’s growth is of a kind that would deliver
productivity increases per worker comparable to that in China. It is true that GDP per
worker has been rising in India, partly because employment growth has decelerated
even as GDP growth has accelerated. But a disconcerting feature of GDP growth in
India is the dominance of services in total growth. Of the cumulative increase in GDP
between 1990 and 2004, while 55 per cent was accounted for by manufacturing in the
case of China, as much as 60 per cent was accounted for by services in the Indian
case. Given the technological trajectory, it should be expected that the potential for
increases in productivity is far greater in industry than in services. This suggests that
distorted growth of incomes in the services sector, illustrated by the discussion on the
IT and IT-enabled services, explains the paradoxical acceleration of GDP per worker
in India at a time when GDP in services is rising rapidly. This does raise questions
about the sustainability of India’s productivity growth trajectory.
For all these reasons, the Chinese experience does not necessarily establish the
opportunity open to India in the coming decades on account of demographic
advantages. Nor is there any certainty about predictions that India would catch-up
with or overtake China in the near future. What the comparison does suggest is that
India can learn from the interventionist policies adopted by the Chinese government
in the early years of the reform, and China needs to be wary of the implications of
some of its reform policies, especially for employment and health.
Conclusion
21
In sum, the evidence suggests that India faces a major deficit in the areas of education
and health which could adversely affect the conversion of a growing labour force into
an effective workforce offering quality, low-cost labour. Further, the changing age
structure of the population is likely to change the pattern of the disease burden
substantially. The existing situation in areas that affect the population in the “bulge”
age-group suggests that the disease burden is likely to rise, leading to a deficit in
health capital. And there are reasons to believe that neoliberal economic reforms in
India since the 1990s are weakening the ability of the State to intervene successfully
and undertake the necessary investments in these and other areas, that would give
India the wherewithal to benefit from the new opportunities that the global economy
supposedly affords.
The implications are clear. Just as the “excess population” argument failed to
recognise the benefits that can be garnered if these excess workers could be put to
work, the “demographic dividend” argument ignores the fact that available workers
are not automatically absorbed to deliver high growth. Strategies exist to exploit the
demographic window of opportunity that India has today, but they need to be adopted
and implemented. In addition, the challenge of meeting a range of goals relating to
education and health are bound to increase. Focusing on the automatic “gains” to be
delivered by the demographic dividend may result in an under emphasis of the effort
needed to meet the new challenges that the current phase of the demographic
transition brings. In particular, emphasis on liberal and open door policies and
excessive fiscal prudence may militate against the adoption of appropriate policies.
India’s experience during the liberalisation years does suggest that markets do not
ensure that these problems are resolved, and this can result in wasting of the
opportunities that the country’s demographic transition temporarily offers.
References
Barboza, David (2006), “Labour Shortage in China May Lead to Trade Shift”, The
New York Times, April 3.
Bloom, David E. and David Canning (2004), Global Demographic Change:
Dimensions and Economic Significance, National Bureau of Economic Research
Working Paper No. 10817 available at http://papers.nber.org/papers/w10817.pdf
accessed February 2006.
Bloom, David E. and Richard B. Freeman (1988), "Economic Development and the
Timing and Components of Population Growth", Journal of Policy Modeling, 10:57-8
1.
Chandrasekhar, C. P. and J. Ghosh. (2004) The Market That Failed: Neoliberal
Economic Reforms in India, Second edition, New Delhi: Leftword Books.
Coale, Ansley J. and Edgar M. Hoover (1958), Population Growth and Economic
Development in Low Income Countries, Princeton, NJ: Princeton University Press.
Crenshaw, Edward M., Ansari Z. Ameen and Matthew Christenson, (1977),
“Population Dynamics and Economic Development: Age-Specific Population Growth
22
Rates and Economic Growth in Developing Countries, 1965 to 1990”, American
Sociological Review, Vol. 62, No. 6.
Feng, Wand and Andrew Mason (2005), Demographic Dividend and Prospects for
Economic Development in China, Paper Prepared for UN Expert Group Meeting on
Social and Economic Implications of Changing Population Structures, Mexico City,
Mexico 31 August – 2 September 2005, UN, Population Division, Department of
Economic and Social Affairs.
Kelley, Allen C. (1988), “Economic Consequences of Population Change in the Third
World”, Journal of Economic Literature, Vol. 26, No. 4.
Miles, James (2006), “Balancing Act: A Survey of China”, in The Economist, March
25th-31st 2006.
OECD (2005), OECD Economic Surveys: China, 2005, Paris: OECD.
Reserve Bank of India (RBI) (2001), Annual Report 2000-01, Mumbai: Reserve Bank
of India.
Simon, Julian L. (1977), The Economics of Population Growth, Princeton, NJ:
Princeton University Press.
Simmons, Ozzie G. (1988), Perspectives on Development and Population Growth in
the Third World, New York: Plenum.
United Nations Population Fund (UNFPA) (2005), State of World Population 2005:
The Promise of Eqality, Gender Equity, Reproductive Health and the Millennium
Development Goals, New York: United Nations Population Fund.
World Bank (2005), India and the Knowledge Economy: Leveraging Strengths and
Opportunities”, Washington D.C.: World Bank, April
23
