ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Demographic Dividend Revisited: The Mismatch between Age and Economic Activity-Based Dependency Ratios

Demographic Dividend Revisited: The Mismatch between Age and Economic Activity-Based Dependency Ratios

This paper dramatises some of the problems with the idea behind the demographic dividend by comparing theoretical dependency ratios (derived from population age distributions) with actual dependency ratios (based on work participation rates) in India. It finds that once one looks at actual dependency - the ratio of non-workers to workers - not only is this still very high, it has probably worsened in recent times. Moreover, it might worsen further if we achieve the highly desirable goal of eliminating child labour and the ultimately justifiable goal of reducing old-age labour, unless there is a major rise in working age labour force participation and/or productivity.

SPECIAL ARTICLE

Demographic Dividend Revisited: The Mismatch between Age and Economic Activity-Based Dependency Ratios

Alaka Malwade Basu

This paper dramatises some of the problems with the idea behind the demographic dividend by comparing theoretical dependency ratios (derived from population age distributions) with actual dependency ratios (based on work participation rates) in India. It finds that once one looks at actual dependency – the ratio of nonworkers to workers – not only is this still very high, it has probably worsened in recent times. Moreover, it might worsen further if we achieve the highly desirable goal of eliminating child labour and the ultimately justifiable goal of reducing old-age labour, unless there is a major rise in working age labour force participation and/or productivity.

Alaka Malwade Basu (ab54@cornell.edu) is with the department of development sociology, Cornell University, Ithaca, New York and currently at the Centre for the Study of Regional Development, School of Social Sciences, Jawaharlal Nehru University, New Delhi.

T
he title of this paper should have been expanded to say something like: For the nth time! For if there is one demographic idea that has captured the imagination in recent years and is constantly visited and revisited, it is the idea of the demographic dividend (DD henceforth, in the interests of trees as well as of typing speed) and the vocabulary created around this idea. Phrases like the “age transition dividend”, the “demographic bonus”, the “window of opportunity”, the “youth bulge” and “ dependency ratios” have come to acquire a life of their own, sprinkling the advice-filled documents from international organisations, the cynical outpourings of academic hair-splitters and the breathless jabbering of journalistic reportage. If I add to the din on the subject here, it is only to restate in simple language what the fuss is all about and then to ask if the fuss needs to be taken down a bit in ways other than has been done so far.

Idea behind the DD

What then is the DD? Baldly put, the term refers to the economic potential temporarily unleashed by changes in the age structure of a population in the direction of increasing the proportions of those of working age. Contrary to some (mis)interpretations, the DD is not about increased numbers of those of working age; i e, a growing labour force on its own does not create a DD (even if that too is in some ways good for a growing economy), this growth must change the balance between those of labour force, age and those outside this age range. When this happens, as explicated by Bloom and Canning (2001) whose work is most frequently associated with the entry of the term into modern academic and popular discourse, what we get is a beneficial change in the ratio of net consumers to net producers in an economy or society. In other words, what we get is a beneficial change in what is called the dependency ratio, whereby there are fewer net consumers to be supported by each net producer, leaving more resources available for productive investment.

Such a change typically occurs at the start of a fertility transition in a high fertility society. As fertility falls (and the more rapidly this happens, the more dramatic the DD; but a shorter duration of the DD as well, because low fertility cohorts enter the l abour force that much more quickly and soon reduce the relative size of the working age population), there is a fall in the size of the unproductive younger generation, relative to the size of previous generations. These previous generations, which are the product of the high fertility as well as the declining mortality of

Economic & Political Weekly

EPW
september 24, 2011 vol xlvi no 39

SPECIAL ARTICLE

the past, are entering and passing through the productive ages at this time. As fertility continues to fall, this advantage of relative size of the working age groups will last as long as these generations of a high fertility past are of working age. Once they exit this age, not only do they become net consumers, there are now, thanks to the fertility decline, fewer net producers of later generations now of working age to support them. Even worse, these new net consumers (that is, the old) are probably going to live longer than their high fertility parents, so that the dependency burden they impose is even greater than it would have been if they left this world in a more timely manner!

All this means that very soon, the falling dependency ratios (DRs) that fertility decline sets off begin to rise again; hence the “window” of opportunity before things get disappointingly back to normal. But while this window is open, the rise in the proportion of the population that is of working age can do wonders for an economy if it is properly harnessed. It seems to have done such wonders for the east Asian economies of the 1970s and 1980s (Bloom and Williamson 1998; Bloom, Canning and Malaney 2000),

Table 1: Trends in the Age Distribution of the Indian Population

R eference Bureau (PRB) in 2007. We rely on the PFI/PRB projections rather than the United Nations (UN) projections for two reasons:

(1) The PFI/PRB projections are more “realistic” in their future fertility assumptions; instead of assuming, as the UN projections do, that eventual total fertility rates (TFRs) will be 1.85 in the e ntire country, they assume in one set of projections (Scenario A, that is used here) that states that currently have above replacement level fertility will stabilise at replacement levels (i e, a TFR of 2.1), whereas those that have below replacement level fertility will reach TFRs of 1.85 if they are currently above 1.85 but below

2.1 and those that already have TFRs below 1.85 (such as Kerala and Tamil Nadu) will remain at current levels. The first part of this assumption certainly makes sense; it is difficult to imagine states like Uttar Pradesh and Rajasthan going well below replacement level fertility in the next several decades. Whether Kerala and Tamil Nadu will or will not continue further fertility decline is however more debatable.

(2) The summary of age distributions in the PFI/PRB projections assumes old-age dependency to apply to persons above the

Proportion of the Census PFI/PRB-Projections UN Population Projections, 2008
Total Population 1 2 3
in Age-Group 1961 1971 1981 1991 2001 2011 2021 2031 2041 2051 2061 2071 2081 2091 2101 2010 2025 2050
0-14 41.0 42.0 39.5 37.2 35.3 30.7 28.1 25.7 23.7 22.1 20.8 19.9 19.2 18.7 18.3 30.6 24.7 18.2
15-64 55.9 54.6 56.6 58.1 59.6 64.1 65.7 66.4 66.2 65.4 64.1 62.9 61.7 60.5 59.7 65.2 67.9 68.0
65+ 3.1 3.3 3.8 4.0 4.8 5.2 6.2 7.9 10.1 12.5 15.1 17.2 19.1 20.8 22 4.2 7.3 13.7
Total 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100.0

Sources: (1) Registrar General of India (2001), Primary Census Abstract, Census of India, 2001.

  • (2) Population Foundation of India and Population Reference Bureau (2007), The Future Population of India – A Long-range Demographic View (Scenario A).
  • (3) United Nations (2009), World Population Prospects, The 2008 Revision Vol I. Comprehensive Tables; New York: UN, Department of Economic and Social Affairs.
  • Table 2: Trends in Actual and Projected Population Dependency Ratios

    1961 1971 1981 1991 2001 2011 2021 2031 2041 2051 2061 2071 2081 2091 2101 2010 2025 2050
    Youth dependency ratio 770 808 734 672 621 484 428 394 364 338 328 316 311 317 307 500 331 257
    Old-age dependency ratio 106 115 120 122 131 78 94 121 151 191 234 274 310 350 369 122 179 332
    Total dependency ratio 875 923 854 794 752 562 522 515 515 529 562 590 621 667 675 622 510 589

    Youth DR = (Number of persons aged 0-14) × 1,000/ (Number of persons aged 15-64). Old-Age DR = (Number of persons aged 65+) ×1,000/ (Number of persons aged 15-64). Total DR = Number of persons aged 0-14 and 65+) ×1,000/ (Number of persons aged 15-64). Source: Calculated from successive censuses of India for 1961 to 2001; from the PFI/PRB population projections for the 2011 to 2101 decades; from the UN 2008 population projections for 2011, 2025

    and 2050.

    for Ireland in the 1990s (Bloom and Canning 2003) and may account for at least some of the rapid economic growth that India is flaunting these days – according to recent estimates by Aiyer and Mody (2011), 40-50% of the rise in per capita incomes in India since the 1970s can be attributed to the ongoing DD. By the same token, the slower speed of the age transition and thus the slower appearance of the DD may partly explain low economic growth rates in sub-Saharan Africa.

    For more detailed (and possibly clearer) expositions of the idea behind the DD, the reader should go to Bloom and Canning (2001), Bloom et al (2003) and James (2008). In the meantime, some clarity is added here by Tables 1 and 2. Table 1 merely presents trends in the proportion of working age population. Table 2 converts these proportions to trends in dependency ratios. The last row of Table 2 depicts actual trends in population DRs in India from 1961 to 2001 and projected trends from 2011 (we do not as yet have the age distributions from the 2011 Census to be able to calculate actual DRs) to 2101 using the projections prepared by the Population Foundation of India (PFI) and the Population age of 65 rather than those above the age of 60 as is the current convention. This decision is reasonable – as retirement ages rise everywhere and as the increasing life expectancy comes with i ncreasing lengths of active life, assuming net dependency from the age of 60 is unnecessarily pessimistic. For the sake of consistency, the census-based DRs in Table 2 are also calculated, assuming a working age category of 15-64 years.

    In any case, for comparative purposes, the last three columns of Table 2 present DRs (again using the 15-64 working age classification) using the UN (2008) population projections for India. Because the UN medium projections assume that the entire country will progress towards a TFR of 1.85, eventual population sizes are smaller than in the PFI/PRB estimates and the fall in DRs is also sharper.

    To return to Table 2, the DD is clearly visible from 1981 o nwards. From this point on, total DRs fall steadily until close to the middle of the 21st century, when they begin rising again. This reversal will begin at different times in different parts of the country d epending on the time at which fertility decline begins as well as the speed of the decline.

    september 24, 2011 vol xlvi no 39

    SPECIAL ARTICLE

    However, the rise after 2050 is not due to rises in the youth DRs ones. But that kind of need to claim ownership need not detain us (the number of young dependents per 1,000 working age popula-here. Instead, we examine some old and some new notes of caution) – that ratio continues to fall, because all historical experi-tion which question the principles underlying the definition of a ence suggests that once fertility begins a sustained decline, it is DD. The central principle to be addressed here is the notion that unlikely to rise again in any significant way (although, of course, the DD can be captured by age-determined distributions of workhistory is rarely a good predictor of the future). Instead, rising ers and non-workers. total DRs are a consequence of rapid rises in the old-age DR as the larger working age cohorts of the past enter old-age as well as Age Transition as a Marker of the DD live longer than their previous cohorts. This disaggregation of The fundamental premise of the DD hypothesis is that there is the components of changing DRs is important at least partly something called a “working age population” or even something b ecause the nature of dependency among the young is very dif-called a “potentially working age population”. But how close is the ferent from that among the old; the old are, on the whole, more overlap in countries like India between the age categories we use expensive to maintain than the young. to define groups that are net consumers versus net producers and

    So that is the DD in a nutshell. But why do we revisit it here? actual groups of net consumers and net producers? Not very much, Partly to reiterate some of the cynicism (see, for example, Chan-it appears from Table 3a and 3b. Of course, one can have many drasekhar et al 2006) that has already been expressed about the questions about measures of both age and labour force participaexcitement the DD usually causes and partly to throw some new tion, but even treating both these variables as being crudely meascold (or at least cool) water on this excitement. And why are we ured, Table 3a suggests that a fair number of people in the 0-14 focusing only on the less palatable aspects of the DD here? We do categories are productive (if one assumes that having what the this partly because the positive aspects are already getting plenty census calls a “main” occupation makes one a net producer), while of free publicity and partly because such pride could sometimes Table 3b suggests that several others are probably self-sufficient come before a fall. even if they are not net producers (that is, if one assumes that those

    If one were to classify the apprehensions related to the DD in that the census classifies as “marginal workers” produce more or India, one way might be to separate out old arguments from new less enough to meet their own consumption needs even if they are not much use in underwriting anyone else’s consumption).

    Table 3a: Age-Specific Work Participation Rate (Main), 2001 (Proportion of Population in the Workforce) Table 3b: Age-Specific Work Participation Rate (Main+Marginal), 2001

    5 to 14 15-59 60+ 5 to 14 15-59 60+

    P M F P M F P M F P M F P M F P M F

    India 2.3 2.7 1.8 48.1 71.2 23.3 32.1 52.8 12.0

    India 5.0 5.1 4.8 61.0 80.6 40.0 40.3 60.2 20.9

    J&K 2.1 2.5 1.6 40.4 64.5 12.6 32.3 54.0 6.6

    J&K 6.7 6.7 6.7 56.9 76.2 34.7 43.4 64.0 18.9

    HP 1.0 1.0 1.0 48.5 65.1 31.7 33.0 48.3 18.0

    HP 8.1 7.6 8.6 71.7 79.8 63.4 50.6 63.4 38.1 Punjab 2.0 2.7 1.2 48.5 75.1 18.6 30.6 51.6 9.0

    Punjab 3.2 3.9 2.4 56.2 80.9 28.5 35.2 56.0 13.8 Uttaranchal 1.1 1.4 0.9 44.4 65.3 26.5 31.1 46.6 15.6

    Uttaranchal 3.3 3.2 3.4 58.8 78.1 43.1 43.4 58.9 27.9 Haryana 1.4 1.7 0.9 48.0 70.4 22.0 24.7 41.2 8.1

    Haryana 4.8 4.7 4.9 63.3 80.4 43.3 33.4 48.7 18.1 Delhi 1.1 1.7 0.4 48.0 75.4 13.2 20.5 35.7 4.7

    Delhi 1.4 2.0 0.6 50.5 78.6 14.7 21.7 37.4 5.5 Rajasthan 2.5 2.4 2.7 53.0 75.0 29.0 29.1 48.5 11.2

    Rajasthan 8.3 7.0 9.7 69.9 83.7 55.0 38.9 56.1 23.1 Uttar Pradesh 1.7 2.4 0.9 39.6 65.9 10.5 37.2 62.8 8.4

    Uttar Pradesh 4.1 4.7 3.3 54.0 78.1 27.3 46.5 71.1 18.8 Bihar 2.3 3.2 1.2 43.4 69.6 15.4 37.5 62.3 9.4

    Bihar 4.7 5.6 3.7 57.2 80.6 32.2 46.4 70.9 18.6 Sikkim 4.7 4.8 4.6 60.1 75.9 41.1 45.8 61.8 24.9

    Sikkim 12.0 11.9 12.2 71.5 83.1 57.6 56.5 70.6 38.2 Arunachal Pradesh 2.7 2.3 3.2 63.0 75.3 48.6 51.3 65.8 35.0

    Arunachal Pradesh 6.1 5.2 6.9 71.8 81.3 60.6 60.6 73.4 46.1 Nagaland 4.0 3.9 4.0 53.5 61.0 45.1 63.1 71.5 52.4

    Nagaland 8.5 8.3 8.7 63.1 68.9 56.7 70.6 77.8 61.4 Manipur 2.0 2.1 1.9 45.0 58.7 31.1 40.9 56.6 24.9

    Manipur 5.7 5.5 5.9 63.5 70.9 56.1 55.4 68.8 41.8 Mizoram 2.4 2.5 2.4 63.9 75.8 50.8 44.1 58.6 29.3

    Mizoram 12.3 11.9 12.8 78.4 84.8 71.4 59.4 70.1 48.3 Tripura 1.1 1.3 0.9 43.3 67.8 16.8 32.2 57.1 8.8

    Tripura 2.6 2.7 2.5 54.8 75.6 32.4 39.8 64.1 17.1 Meghalaya 3.7 4.3 3.2 55.3 70.1 40.1 50.2 68.0 31.7

    Meghalaya 8.0 8.4 7.6 69.4 80.4 58.0 60.6 76.2 44.3 Assam 2.0 2.7 1.2 42.5 66.9 16.1 34.5 59.2 8.4

    Assam 5.1 6.0 4.1 56.3 77.6 33.1 41.8 66.3 16.0 West Bengal 2.0 2.6 1.4 44.3 71.8 14.1 26.1 47.3 5.9

    West Bengal 4.5 5.1 3.9 56.2 81.7 28.2 32.1 53.8 11.3 Jharkhand 1.8 2.2 1.4 40.1 61.7 16.9 28.7 48.3 9.1

    Jharkhand 5.5 5.4 5.5 61.8 78.9 43.6 41.4 60.8 22.1 Orissa 1.2 1.5 0.9 39.8 65.4 13.4 28.3 50.5 6.6

    Orissa 4.3 4.1 4.5 58.6 79.2 37.4 39.5 61.7 17.8 Chhattisgarh 2.5 2.7 2.3 54.9 72.9 36.6 35.7 56.5 18.1 Chhattisgarh 6.9 6.2 7.7 73.9 83.7 64.0 48.0 65.3 33.3

    MP 2.4 2.8 2.0 52.6 73.6 29.4 35.2 56.3 15.0 MP 6.7 6.4 7.0 69.6 83.4 54.3 45.7 63.9 28.4

    Gujarat 1.9 2.3 1.4 52.0 78.9 22.6 27.0 47.8 8.9 Gujarat 4.3 4.0 4.6 64.3 84.2 42.6 33.1 51.3 17.2

    Maharashtra 1.8 2.0 1.6 54.9 73.4 34.4 32.4 48.4 18.4 Maharashtra 3.5 3.5 3.5 64.3 80.0 46.8 40.4 55.0 27.6

    Andhra Pradesh 5.3 5.2 5.4 56.9 75.4 37.9 33.8 52.3 16.9 Andhra Pradesh 7.7 7.0 8.4 67.8 83.4 51.8 40.7 59.1 24.0

    Karnataka 4.1 4.8 3.3 55.0 77.1 31.9 32.5 53.1 13.9 Karnataka 6.9 7.2 6.6 66.2 83.8 47.8 39.1 58.1 22.0

    Goa 1.1 1.1 1.0 44.6 66.4 21.3 18.1 32.2 7.0 Goa 1.8 1.9 1.8 54.4 75.7 31.6 24.9 40.0 12.9

    Kerala 0.3 0.4 0.2 37.8 61.3 15.9 17.4 32.2 5.5 Kerala 0.5 0.6 0.3 47.0 73.4 22.4 22.9 40.4 8.8

    Tamil Nadu 2.6 2.9 2.3 53.5 73.2 33.8 36.2 53.2 19.4 Tamil Nadu 3.6 3.8 3.4 62.6 81.0 44.2 43.0 60.2 26.0 Source: Census of India, 2001. Source: Census of India, 2001.

    Economic & Political Weekly

    EPW
    september 24, 2011 vol xlvi no 39 55

    SPECIAL ARTICLE

    These child labour estimates, which are probably biased downwards are bad enough and statewise variations in them are often stark, but what is more interesting is the large numbers of the “old” that may be classified as net producers (main workers) or at least net self-supporters (marginal workers). These proportions are often not much lower than the working proportions in the

    o fficially working ages that the DD celebrates, even after we use the age of 65 rather than the conventional 60 as the upper cut-off to define working age, as we do here.

    Both these sets of additional workers – among the young and among the old – should bring DRs down, if one estimates actual dependency (that is the ratio of non-workers to workers – let us call this the economic dependency ratio or the EDR rather than theoretical age-derived dependency. Table 4 does this and finds that this is however far from being the case. All-India EDRs in 2001 (i e, ratios based on actual work participation rates) are, at 2,286 dependents per thousand workers (column 1 in the second half of Table 4), almost four times the dependency ratio using the 15-64 group to denote workers (this ratio is 673, see Table 2). State-level EDRs range from lows around 1,500 (if one can call figures around 1,500 “low”) in the north-east to ratios above 3,000 in Uttar Pradesh and Jharkand. To put this in plain text, in Uttar Pradesh in 2001, a thousand workers were supporting 3,225 non-workers and marginal workers.

    If one assumes, probably unrealistically, that marginal workers are also net producers, then these EDRs fall, but are still, at 1,557 per thousand (column 4 in the second half of Table 4),

    Table 4: Economic Dependency Ratios 1991 and 2001

    a lmost three times higher than the age DRs. And the figure for Uttar Pradesh is still over 2,000.

    If one uses the conventional 15-59 age group to constitute working age, then the difference becomes less striking (an age DR of 752 compared to an EDR of 2,286), but it is only slightly less striking.

    Given that high old-age labour force participation rates should serve to reduce, not increase, DRs, the source of this increase then lies in the false assumption that the theoretically productive age group (15-64) actually consists of working individuals for the most part. And that when it does not, (1) those who are not working are only busy honing their skills, so that when they do join the labour force, they will be that much more productive; and

    (2) even if a significant part of this age group is not economically active, the rise in the relative size of this age group should overcome this constraint and reduce DRs.

    Both these assumptions may be questioned in India. First, it is not the case that the 30% of the non-working male and 70% of the non-working female population of working age is busy acquiring marketable skills. We know that in the 2001 Census, fully 9% of males and 52% of females aged 15-59 were neither students nor workers (main or marginal). Second, Table 5 (p 57) suggests that while EDRs did indeed fall between 1981 and 1991 and that this was probably at least partly an expression of the DD, between 1991 and 2001, while the DD has continued to work in our favour, EDRs have, in fact, worsened. Indeed, EDRs in 2001 were worse than they were in 1981. This worsening may or may not be a trend

    (1) Economic Dependency, 1991 (2) Economic Dependency, 2001
    1 2 3 4 5 6 1 2 3 4 5 6
    Total Male Female Total Male Female Total Male Female Total Male Female
    India 1,932.7 2,493.2 8,598.1 1,669.5 2,337.4 5,842.2 India 2,286.2 2,979.9 9,822.0 1,557.2 2,277.6 4,923.6
    HP 1,906.5 2,640.2 6,860.1 1,335.1 2,230.7 3,325.7 J&K 2,888.5 3,383.6 19,740.2 1,702.2 2,384.3 5,950.3
    Punjab 1,638.5 1,713.1 37,631.4 1,569.5 1,682.0 23,476.5 HP 2,094.8 3,085.4 6,524.8 1,031.1 1,829.3 2,362.9
    Haryana 2,489.0 2,757.1 25,594.8 2,226.0 2,652.9 13,832.0 Punjab 2,108.7 2,571.3 11,721.1 1,668.8 2,188.4 7,028.0
    Delhi 2,284.6 2,548.7 22,049.7 2,270.9 2,538.7 21,526.1 Uttaranchal 2,655.5 3,762.1 9,027.9 1,708.8 2,682.8 4,706.5
    Rajasthan 2,162.5 2,691.0 11,010.2 1,572.8 2,368.0 4,683.3 Haryana 2,387.8 3,021.1 11,391.1 1,524.0 2,233.8 4,796.2
    Uttar Pradesh 2,363.4 2,677.4 20,153.2 2,105.2 2,564.2 11,760.7 Delhi 2,208.0 2,512.4 18,221.6 2,047.3 2,349.8 15,902.8
    Bihar 2,370.4 2,822.8 14,792.6 2,108.6 2,704.1 9,575.1 Rajasthan 2,240.7 3,042.5 8,501.7 1,377.6 2,227.9 3,609.4
    Sikkim 1,472.5 2,200.7 4,449.9 1,409.0 2,143.1 4,113.8 Uttar Pradesh 3,224.9 3,694.4 25,375.2 2,078.7 2,738.2 8,630.3
    Arunachal Pradesh 1,211.3 1,902.9 3,332.7 1,162.6 1,859.2 3,102.9 Bihar 2,942.4 3,537.5 17,489.6 1,966.9 2,686.3 7,344.7
    Nagaland 1,364.7 2,331.2 3,291.8 1,343.0 2,307.1 3,213.9 Sikkim 1,540.4 2,237.9 4,941.7 1,056.1 1,676.4 2,854.4
    Manipur 1,593.8 2,721.4 3,846.6 1,370.8 2,500.8 3,033.8 Arunachal Pradesh 1,645.7 2,554.2 4,626.4 1,273.7 2,094.8 3,249.3
    Mizoram 1,375.9 2,243.9 3,556.7 1,044.7 1,822.4 2,447.9 Nagaland 1,826.8 3,027.4 4,606.8 1,347.3 2,336.0 3,183.3
    Tripura 2,437.6 2,934.4 14,399.8 2,211.6 2,816.3 10,300.3 Manipur 2,286.2 3,503.8 6,578.7 1,292.4 2,317.0 2,922.4
    Meghalaya 1,487.2 2,366.8 4,001.3 1,334.7 2,214.4 3,359.6 Mizoram 1,451.6 2,333.9 3,839.7 902.1 1,602.3 2,064.2
    Assam 2,205.7 2,734.4 11,408.1 1,771.0 2,485.1 6,162.7 Tripura 2,506.8 3,081.9 13,433.6 1,759.0 2,453.4 6,214.3
    West Bengal 2,307.8 2,640.3 18,327.7 2,106.5 2,529.2 12,603.1 Meghalaya 2,063.1 3,215.6 5,756.4 1,390.2 2,372.4 3,357.8
    Orissa 2,050.8 2,506.4 11,280.1 1,664.3 2,288.8 6,100.3 Assam 2,746.9 3,341.0 15,447.0 1,794.5 2,491.2 6,416.5
    Madhya Pradesh 1,654.2 2,336.8 5,663.1 1,335.2 2,112.9 3,627.8 West Bengal 2,482.4 2,931.7 16,196.9 1,719.5 2,264.4 7,146.5
    Gujarat 1,930.7 2,396.7 9,928.8 1,485.5 2,158.2 4,765.4 Jharkhand 3,179.7 3,992.8 15,615.8 1,665.5 2,528.1 4,881.4
    Maharashtra 1,545.3 2,290.6 4,749.9 1,327.3 2,113.9 3,567.3 Orissa 2,838.1 3,399.9 17,175.7 1,578.0 2,298.3 5,034.8
    Andhra Pradesh 1,338.1 2,047.0 3,863.7 1,219.6 1,953.1 3,247.6 Chhattisgarh 1,953.2 2,905.2 5,960.7 1,152.3 2,016.3 2,689.0
    Karnataka 1,601.0 2,253.5 5,529.5 1,381.4 2,101.7 4,030.8 Madhya Pradesh 2,159.2 2,934.3 8,173.4 1,339.7 2,133.7 3,599.7
    Goa 2,049.8 2,738.0 8,155.5 1,834.2 2,568.4 6,417.0 Gujarat 1,976.3 2,495.9 9,492.4 1,383.9 2,031.8 4,339.7
    Kerala 2,505.4 3,247.3 10,965.5 2,181.5 2,934.7 8,500.3 Maharashtra 1,788.0 2,544.6 6,013.6 1,352.9 2,074.5 3,889.7
    Tamil Nadu 1,450.1 2,082.6 4,774.3 1,308.8 1,984.3 3,844.3 Andhra Pradesh 1,624.2 2,424.5 4,920.9 1,184.1 1,907.3 3,122.5
  • (1) EDR = (Number of non-workers + marginal workers) × 1000/(number of main workers),
  • (2) EDR = (Number of non-workers) ×1000/(number of main workers + marginal workers). Source: Census of India 1991 and 2001.
  • september 24, 2011 vol xlvi no 39

    SPECIAL ARTICLE

    – age and economic activity data from the 2011 Census will tell us may be called “cultural” factors underlying Table 4, especially for
    if it is – but it is certainly no cause for a blind celebration of the DD. women. It is true that female labour force participation tends to
    Table 5: Economic Dependency Ratios over Time – All-India be, for various reasons, especially related to the kind of work
    (Non-Workers + Marginal Workers)/ (Non-Workers)/ women tend to do, underestimated more than male work rates.
    (Main Workers) *1,000 (Main Workers + Marginal Workers)*1,000 However, it also has long been acknowledged that female labour
    1981 1991 2001 1981 1991 2001 force participation rates in India are much lower than in other
    1,990.4 1,932.7 2,286.2 1,662.9 1,669.5 1,557.2 Source: Derived as in Table 4 from work participation rates in the censuses of 1981, 1991 and parts of the world, even allowing for factors like poverty and
    2001. poor opportunity. Using data from the Human Development Sur
    vey for India and the International Labour Organisation for
    DD Benefits in India China, Desai (2011), for example, finds women’s work participa-
    Several cynics about the automatic advantages of the DD benefits tion in China being twice that in India in the 15-34 age groups.
    in India have commented on the supply and demand constraints China is better off than India, it is true; but this kind of differen
    on realising these benefits (see, for example, Desai 2010), but tial also exists between India and several African countries.
    there is more to the problem than the absence or presence of These cultural underpinnings are also suggested by the wide
    skills and the opportunities to use these skills. To begin with, Ta r egional variations within India – women’s work participation
    ble 6 suggests that for the non-working population, there is not rates in the 2001 Census range from around 50% in the north
    much skill acquisition going on. Barely 24% of the non-working east to 30% in the south (except for the anomaly of Kerala with
    population aged 15-49 (or 9% of the total population in this age 15%) to the teens in the north.
    group) in the 2001 Census can be classified as “students”. This Trend data as well as cross-sectional differences take this pic
    means that some 35% of this age group is neither working nor ture further by showing that development, at least in the early
    studying; i e, it is neither supporting the 1-14 or 60 “dependent” stages, is probably no panacea. With education and income rises,
    population (for that matter, it is not even supporting itself), nor is female labour force participation tends to fall even further in
    it preparing itself for 21st century employment! most parts of the country – if this is true, then the rising EDRs
    Table 6: Participation in Education by Age Group, All-India, 2001 b etween 1991 and 2001 in column 3 of Table 4 are probably a
    Age Group Students as Per Cent of Non-Working Students as Per Cent of Total good sign – they reflect women’s withdrawal from the workforce
    Population Population Person Males Females Person Males Females for reasons that are positive for women even if they do not bode
    0-14 76.0 74.0 78.5 46.5 48.8 44.0 well for dependency ratios. But then, columns 6 in Table 4 tell a
    15-59 23.9 25.8 21.3 9.1 10.6 7.4 more ambiguous story – when one looks at main plus marginal
    60+ 0.2 0.1 0.2 0.5 0.5 0.5 workers, EDRs have fallen – suggesting that women’s withdrawal
    Source: Economic Tables, 2001 Census. from the labour force may be somewhat involuntary, that they
    These figures get even more dismal when one disaggregates are losing opportunities for “main” work and being pushed into
    them by gender. As Desai (2011) points out in a rejoinder to those “marginal” work because they still need “some” work.
    who celebrate India’s DD in comparison to China’s closing of the These results, also challenge the expectation that the falling
    window of opportunity, once one takes differences in female fertility that heralds a DD is invariably accompanied by rises in
    l abour force participation rates in the two countries into account, women’s work levels as fewer children and rising aspirations
    by 2030, when India is expected to have an age distribution should mean a greater ability as well as a greater desire to join
    a dvantage over China, dependency ratios in India will actually the labour force. Indeed, in many formulations (see, for example,
    be a good 30% higher in India. Bloom and Canning 2003 on Ireland and Bailey 2006 on the
    Table 3 illustrates this very well. Columns 2 and 3 (as well as United States), it is the rise in female labour force participation
    columns 5 and 6) in this table estimate EDRs that would prevail if rates that follow fertility decline which account for much of the
    either working males (in columns 2 or 5) or females (in columns 3 benefits of a DD. Conversely, low female labour force participa
    or 6) were solely responsible for supporting the non-working tion in high fertility countries is often attributed in the literature
    population because females and males respectively had abdicated to the incompatibility between women’s reproductive and pro
    any responsibilities on this front. That is, assume in column 2 ductive roles.
    that working women either do not exist or else they work merely But in many parts of India, the social status attached to not
    to support themselves. A thousand working men in such a situa working (again, at relatively low levels of affluence, college-
    tion would have 2,979 dependents; as if this were not bad enough, educated women probably rise in status with jobs) in many parts of
    in the parallel case, a thousand working women would have to the country could well trump the desire for more economic wealth.
    support 9,822 dependents! Obviously, these figures are a far, far It is in any case better achieved through using one’s education and
    cry from the hypothetical dependency ratio estimates based on family wealth to marry a richer man. In other words, without more
    age distributions in Table 2. rapid social change, culture may well trump economic compul-
    While a part of the blame for this absence of a complete over sions as well as economic aspirations and not exploit the release
    lap between working age population and working population can from childcare responsibilities for some time to come.
    be attributed to mismatches between job skills and jobs availabil- How then is the DD to be realised? How can we claim a larger
    ity and/or a mismatch between growth rates in employment and benefit from changing age structures, if the expected overlap
    growth rates of the 15-59 age group, there are also several what b etween rises in working age population and working population
    Economic & Political Weekly september 24, 2011 vol xlvi no 39 57
    EPW

    SPECIAL ARTICLE

    does not exist? One can try to improve this overlap by efforts to raise male and (especially) female labour force participation with economic incentives as well as social advocacy. In addition, one can also increase investments in productivity-generating skills (which the age structure of the DD should allow more easily now that there are fewer investments needed in the youngest age groups, since there are fewer members in these groups) so that the substantially higher incomes of those who do work make up for the dependent status of the increasing numbers of female non-workers as development proceeds. Aiyer and Mody (2011) suggest that something like this has already happened – that the economic and social developments which lead to fertility d eclines are the very same ones that lead to productivity i ncreases as well.

    The overlap between age and labour force participation should also be increased by a more rapid withdrawal from the workforce of those ages below 15 (which is a good and necessary thing for many reasons), as well as a gradual decline in labour force participation by the “older” old (those above 70, for example), even if this raises DRs because women do not step into the workforce to make up for it. No development worth its name should tolerate 5% of those aged 5-14 in the 2001 Census (and these are certainly underestimates) being economically active, when they should be in school or at play. And rural grandparents need to be playing with these freed children rather than slaving in the fields.

    That is too summary a consideration of policy options, for sure. But that is because the aim of this paper is more to temper the enthusiasm of the DD optimists than to undertake an exhaustive analysis of solutions.

    References

    Aiyar, S and A Mody (2011): “The Demographic Dividend: Evidence from the Indian States”, IMF Working Paper No WP/11/38 (Washington DC: International Mone tary Fund).

    Bailey, M (2006): “More Power to the Pill: The Impact of Contraceptive Freedom on Women’s Labour Supply”, Quarterly Journal of Economics, Vol 121.

    Bloom, D and D Canning (2003): “Contraception and the Celtic Tiger”, The Economic and Social Review, Vol 34.

    Bloom, D E and D Canning (2001): “Demographic Change and Economic Growth: The Role of Cumulative Causality” in N Birdsall, A C Kelley and S W Sinding (ed.), Population Matters: Demographic Change, Economic Growth and Poverty in the Developing World (New York: Oxford University Press).

    Bloom, D E, D Canning and J Sevilla (2003): The Demographic Dividend: A New Perspective on the Economic Consequences of Demographic Change (Santa Monica: The Rand Corporation).

    Bloom, D and J Williamson (1998): “Demographic Transitions and Economic Miracles in Emerging Asia”, World Bank Economic Review, Vol 12.

    Bloom, D, D Canning and P Malaney (2000): “Demographic Change and Economic Growth in Asia”, Population and Development Review, Vol 26.

    Chandrasekhar, C P, J Ghosh and A Roychodhury (2006): “The ‘Demographic Dividend’ and Young India’s Economic Future”, Economic & Political Weekly, Vol 41.

    Desai, N (2010): Demographic Dividend or Debt?, Eleventh JRD Tata Memorial O ration (New Delhi: Population Foundation of India).

    Desai, S (2011): “The Other Half of the Demographic Dividend”, Economic & Political Weekly, Vol 45.

    James, K S (2008): “Glorifying Malthus: Current Debate on ‘Demographic Dividend’ in India”, Economic & Political Weekly, Vol 43.

    EPW Research Foundation (A UNIT OF SAMEEKSHA TRUST)

    YYYGRYTHKP YYYGRYTſVUKP +PFKC 6KOG 5GTKGU

    #HGYOQPVJUCIQ'294(KPVTQFWEGFCPQPNKPGFCVCDCUGUGTXKEGEJTKUVGPGFCUŎ+PFKC6KOG5GTKGUŏYYYGRYTſVUKP6JGRTQLGEVGPXKUCIGFRTQXKUKQPQHFCVC KPVYGNXGOQFWNGUFKURNC[KPIVKOGUGTKGUQPCYKFGTCPIGQHOCETQGEQPQOKECPFſPCPEKCNUGEVQTXCTKCDNGUKPCOCPPGTEQPXGPKGPVHQTTGUGCTEJCPFCPCN[VKECN YQTM6JKUKUVCTIGVGFVQDGPGſVRCTVKEWNCTN[UVWFGPVUTGUGCTEJUEJQNCTURTQHGUUKQPCNUCPFVJGCECFGOKEEQOOWPKV[DQVJKP+PFKCCPFCDTQCF

    6JKUQPNKPGUGTXKEGKUCRCTVQHVJGRTQLGEVHWPFGFD[VJG7PKXGTUKV[)TCPVU%QOOKUUKQP
    7)%CPFGZGEWVGFD[VJG6CVC+PUVKVWVGQH5QEKCN5EKGPEGU 6+55

    /WODCKCPFVJG'EQPQOKECPF2QNKVKECN9GGMN[ '29

    #HVGTVJGUWEEGUUHWNNCWPEJQHVJGſTUVBſPCPEKCNOCTMGVUOQFWNGŏKP,CPWCT['294(KURNGCUGFVQKPVTQFWEGVYQOQTGOQFWNGU$CPMKPI5VCVKUVKEU=$CUKE

    5VCVKUVKECN4GVWTPU
    $54?CPFQOGUVKE2TQFWEVQH5VCVGUQH+PFKC0KPGOQTGOQFWNGUYKNNDGCFFGFUQQP

    6JGFGOQXGTUKQPECPDGCEEGUUGFD[HTGGTGIKUVTCVKQP6JGGZKUVKPIOGODGTUCNTGCF[TGIKUVGTGFYKVJWUCPFCEEGUUKPIOGODGTUGTXKEGUCVYYYGRYTHKPYKNN TGSWKTGPQHTGUJTGIKUVTCVKQP6QICKPHWNNCEEGUUVJGUWDUETKRVKQPTCVGUCTG

    +PKVKCN 5WDU ETKRVKQP4CVGU2GT#PPWO
    +PFKXKFWCNU 7PKXGTUKVKGUGGOGF 7PKXGTUKVKGUQT%QNNGIGU 1VJGT+PUVKVWVKQPU
    #EEQTFKPIVQ0WODGTQH%QPEWTTGPV7UGTU #EEQTFKPIVQ0WODGTQH%QPEWTTGPV7UGTU
    7RVQ VQ /QTGVJCP 7RVQ VQ /QTGVJCP
    (KPCPEKCN/CTMGVU OQFWNG +PFKC
    KP4U
    (QTGKIP
    KP75
    $CPMKPI5VCVKUVKEU $CUKE 5VCVKUVKECN4GVWTPU $54 +PFKC
    KP4U
    (QTGKIP
    KP75
    QOGUVKE2TQFWEV QH5VCVGUQH+PFKC +PFKC
    KP4U
    (QTGKIP
    KP75

    9JGPUWDUETKRVKQPKUOCFGHQTVYQQTOQTGOQFWNGUCFKUEQWPVQHYKNNDGCXCKNCDNGQPVQVCNKPKVKCNUWDUETKRVKQPCOQWPV +PKVKCN5WDUETKRVKQPKUXCNKFHQTQPG[GCTCPFTGPGYCNUQPCPCPPWCNDCUKUYKNNDGCXCKNCDNGQPCFKUEQWPVQXGTKPKVKCNUWDUETKRVKQPCOQWPV (QTCP[HWTVJGTFGVCKNUQTENCTKſECVKQPURNGCUGEQPVCEV

    KTGEVQT '29 4GUGCTEJ (QWPFCVKQP % #MWTNK +PFWUVTKCN 'UVCVG #MWTNK 4QCF -CPFKXNK
    'CUV /WODCK
    RJQPG QT OCKN VQ GRYTHXUPNEQO

    september 24, 2011 vol xlvi no 39

    Dear reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here

    Comments

    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top