Context
India is one of the few countries in the world where women’s work participation rates have fallen sharply — from 29 per cent in 2004-5 to 22 per cent in 2011-12 and to 17 per cent in 2017-18.
What could be the possible explanations for the decline?
- No consensus among economists: Trying to explain whether women are choosing to focus on domestic responsibilities or whether they are pushed out of the workforce has become a minor industry among economists.
- Can the quality of data be the explanation? Strangely, the one explanation we have not looked at is whether the declining quality of economic statistics may account for this trend.
- Our pride in the statistical system built by PC Mahalanobis is so great that we find it unimaginable that it could fail to provide us with reliable employment data.
- However, as challenges to economic statistics have begun to emerge in such diverse areas as GDP data and consumption expenditure, perhaps it is time to consider the unimaginable.
- Issue of data collection: Is the decline in women’s labour force participation real or is it a function of the way in which employment data are collected?
Anatomy of the decline in participation rates
- Driven by rural women: The anatomy of the decline in women’s work participation rates shows that it is driven by rural women.
- Data of the prime working-age group: In the prime working-age group (25-59)-
- Urban area data: Urban women’s worker to population ratios (WPR) fell from 28 per cent to 25 per cent between 2004-5 and 2011-12, stagnating at 24 per cent in 2017-18.
- Rural area data: However, compared to these modest changes, rural women’s WPR declined sharply from 58 per cent to 48 per cent and to 32 per cent over the same period.
- Among rural women, the largest decline seems to have taken place in women categorised as unpaid family helpers — from 28 per cent in 2004-5 to 12 per cent in 2017-18.
- This alone accounts for more than half of the decline in women’s WPR. The remaining is largely due to a drop of about 9 percentage points in casual labour.
- In contrast, women counted as focusing solely on domestic duties increased from 21 per cent to 45 per cent.
What are the explanations for this massive change?
- Data collection issue: It is the change in our statistical systems that drives these results.
- Change of workforce collecting data: The questionnaires through which the National Statistical Office (NSO) collects employment data have not changed, but the statistical workforce has, and the surveys that performed reasonably well in the hands of seasoned interviewers are too complex for poorly trained contract data collectors.
- How data is collected? The National Sample Surveys (NSS) do not have a script that the interviewer reads out. They have schedules that must be completed. The interviewer is trained in concepts to be investigated and then left to fill the schedules to the best of his or her ability.
- The NSS increasingly relies on contract investigators hired for short periods, who lack
- Need for redesigning the surveys: Do we need to return to the days of permanent employees or can we design our surveys to overcome errors committed by relatively inexperienced interviewers?
- A survey design experiment led by Neerad Deshmukh at the NCAER-National Data Innovation Centre provides an intriguing solution.
- In this experimental survey, interviewers first asked about the primary and secondary activity status of each household member, mimicking the NSS structure.
- They then asked a series of simple questions that included ones like, “do you cultivate any land?” If yes, “who in your household works on the farm?”
- Similar questions were asked about livestock ownership and about people caring for the livestock, ownership of petty business and individuals working in these enterprises.
- What was the result of survey experiment: The results show that the standard NSS-type questions resulted in a WPR of 28 per cent for rural women in the age group 21-59, whereas the detailed activity listing found a WPR of 42 per cent — for the same women.
- This is an easily implementable module that does not require specialised knowledge on the part of the interviewer.
Identifying the sectors from which women are excludes
- Missing the identification of sector: In our concern with ostensibly declining women’s work participation, we have missed out on identifying sectors from which women are excluded and more importantly, in which women are included.
- What data for men indicate? For rural men, ages 25-59, between 2004-5 and 2017-18, casual labour declined by about 6 percentage points.
- However, this decline is counterbalanced by regular salaried work which increased by 4 percentage points.
- Thus, it seems likely that men are exchanging precarious employment with higher-quality jobs.
- What data for women indicate? In contrast, women’s casual work has declined by 9 percentage points while their regular salaried work increased by a mere 1 percentage point.
- Moreover, the usual route to success, gaining formal education, has little impact on women’s ability to obtain paid work.
- The explanation for the disparity: Rural men with a secondary level of education have options like working as a postman, driver or mechanic — few such opportunities are open to women.
- It is not surprising that women with secondary education have only half the work participation rate compared to their uneducated sisters.
- Takeaway: The focus on employment for women needs to be on creating high-quality employment rather than getting preoccupied with declining employment rates.
Conclusion
It may be time for us to return to the recommendations of ‘Shramshakti: Report of National Commission on Self Employed Women and Women in the Informal Sector’ and develop our data collection processes from the lived experiences of women and count women’s work rather than women workers. Without this, we run the risks of developing misguided policy responses.
