One striking feature of the latest National Sample Survey round results is the apparent decline in female employment in 2009-10 compared with 2004-05. This edition of MacroScan examines this feature in more detail to isolate the important processes and factors at work.
There has already been much discussion about the low rates of employment growth in India that are reflected in the latest large sample round to the NSSO surveys.
One of the features that have contributed to this decline in employment growth in the most recent five-year period is the slump in female employment, which can be considered as one of the more important elements in the overall deceleration of employment generation.
As was shown in a previous edition of MacroScan, applying the participation rates of the NSSO survey of 2009-10 to interpolated population figures from Censuses 2001 and 2011 shows that total female employment actually declined at an annual rate of 1.72 per cent between 2004-05 and 2009-10, while male employment (mostly in casual work) showed a slight increase, albeit much lower than in the previous period, at the rate of 1.72 per cent.
Clearly, this is a significant and potentially very disturbing result, especially given that women's work participation rates are already quite low in India compared to most other parts of the developing world.
It should be borne in mind, of course, that work participation rates as described by official surveys are not really good indicators of the productive contributions of women.
This is particularly so in large parts of India, where much of the economic activity of women, whether in the home or outside, is simply not recognised as such by other household members and even by the women themselves.
A significant part of women's work is not just unpaid, therefore: it is also socially unrecognised. This is true of not just social reproduction, but other economic activity where women's work is rendered invisible by social perceptions.
That is why many social scientists take women's work participation rate as one of the proxy indicators of women's overall status in society and of gender empowerment.
It is not just because paid work provides income individually to women rather than to male members of the household. It is also because the productive contribution of women is typically less recognised in societies where women are undervalued in general.
Chart 1 indicates the movement of female work participation rates in India from the late 1970s. Several features of interest emerge from this chart. First, rural participation rates are nearly three times the urban rates, though they still remain relatively low at only around 30 per cent compared to around double that in most East Asian countries.
Second, the longer term trend appears one of gentle decline even within these relatively low rates, which is truly remarkable in a rapidly growing economy. The experience of most other developing countries in phases of rapid growth has been that of rapid and often substantial increase in female overt work participation, as dynamic capitalism tends to draw in women to expand domestic labour supply for paid employment.
The third feature that seems to come out from Chart 1 is that if anything, 2004-05 is a bit of an outlier in terms of increasing female work participation, whereas 2009-10 indicates a reversion to the longer term trend of gradual decline. If this is indeed true (which of course requires further examination) then it suggests that inadequate capture of women's work in the latest sample round may not be the most important reason for the evident decline. Rather, the more pressing question should be why women's work participation rates have been so low in India and have remained low despite rapid economic growth and many other changes in society.
(In any case, that explanation — of poor investigative behaviour during the survey - does not solve the basic conundrum: if changing labour demand results in more demand for women in paid work, this is more likely to be captured by investigators than home-based work that is economically productive but unrecognised. Obviously, there has not been such a trend of evident increases in demand for women in paid employment – and that is a real paradox.)
Suppose we even consider that for whatever reason, 2004-05 was an unusual year within this dataset. So let us consider the change in women's employment status for the decade as a whole, that is, from 1999-2000 to 2009-10. Table 1 provides the details for women in the age group of 15 years and above, and even these are quite startling.
Remember that the 2000s were a decade of unprecedented rapid GDP growth for the Indian economy. In this decade, the number of women aged 15 years or more increased by 86.5 million.
But only 8.9 per cent of them joined the labour force, and only 7.5 per cent of them were described as gainfully employed. This relative lack of increase in the number of working women in a period of major economic expansion is not just unusual, it is also hard to explain in terms of most standard economic approaches.
More in education
The really large increase – accounting for more than 20 per cent of the increase in number of women overall – was in education. Of course this was dominantly confined to younger women, but it is clearly a process to be greatly welcomed. Chart 2 shows that the most recent period has experienced the biggest increase in the number of young women in education. This is particularly prevalent for urban girls aged 15-19 years, nearly 70 per cent of whom are now studying as their principal activity. But the recent relatively fast increase in education for girls even in rural areas is a good sign.
When we look at the patterns of employment of the relatively few women who are recognised as gainfully employed, even in this aspect the apparent lack of change appears to be more striking than any dynamism. Chart 3 shows the distribution of women workers by type of contract. Once again, the year 2004-05 appears as an outlier, when there was apparently a big increase in self-employment and an associated decline in the share of casual work.
Obviously, we will need more disaggregated data to delve into the actual processes at work in this case, but a quick assessment suggests that in rural areas the medium term process has been of decline in self-employment and increase in casual work of women, with regular employment persistently occupying a negligible space.
Real urban story
In urban areas, as shown in Chart 4, a somewhat similar process seems to have been at work in terms of the relative shares of self-employed and casual workers. But the real story for urban women is the increase in regular employment, which has continued into the most recent years. When this was first observed for 2004-05, it created much joy in official circles, until it was pointed out the largest increase in urban regular employment of women was in the form of domestic service – as maids, cooks and cleaners, hardly the most desirable or dynamic forms of work. This accounted for 3 million more urban women workers in the period 1999-2000 to 2004-05, far exceeding the increase in “export-oriented” sectors such as garments, leather and IT-enabled activities.
We do not yet have the detailed data that will allow us to examine whether a similar process was at work in the most recent quinquennium. But in any case, that experience should give us pause in interpreting the current evidence, and not give rise to undue or premature celebration.
The evidence on female unemployment rates adds to the uneasiness involved in including 2004-05, which seems more and more to have been an unusual year. Even excluding that year, Chart 5 suggests that open unemployment rates for women have remained at fairly high levels over the period of high economic growth. Throughout this period, more than 7 per cent of urban women who count themselves as in the labour force were unemployed and actively looking for work but not finding it, as their usual principal activity. In terms of current daily status – the activity that they pursued on an average day in the previous week – the rate of open unemployment has been even higher, at around 10 per cent. In rural areas, as expected, both rates were lower but still current daily status unemployment was significantly high at more than 8 per cent, and show a major increase compared to the late 1980s.
The other depressing feature that emerges from the latest round of NSS data is that economic growth has still not generated a process of employment diversification for women. Even more than men, and substantially so, women workers remain stuck in low value-added but arduous work in agriculture. Around two-thirds of women workers are still employed in agriculture as their principal economic activity, while the share for men workers has fallen to less than half.
For both men and women, this is actually an appalling rate of employment diversification. Every single development success that we have seen in history has been associated with a movement of workers away from primary activities to more value-added work in secondary and tertiary sectors.
The stubborn domination of agriculture as the primary source of work for most of our workers (especially women) is a particular problem given the agrarian crisis that has persisted for nearly two decades in the countryside, which makes involvement in such work increasingly fraught and financially unviable. All in all, these patterns of women's work – and indeed therefore of men's work – amount to a severe indictment of the Indian growth story, in terms of what it is delivering to the bulk of Indian workers.