The National Sample Survey (NSS) has just released some key characteristics of employment and unemployment in India based on the survey conducted between July 2009 and June 2010. Discussion on these topics in India often loosely uses language of employment and labour markets from the more developed world, with misleading and confused conclusions.
Also See | Workforce Estimates (PDF)
To illustrate the problem, consider the following: one employment measure commonly used in India is the so called Usual Principal and Subsidiary Status (UPSS) —a measure of employment based on time spent in the year before the survey. Principal status being the activity accounting for majority of his/her time over the year and subsidiary status, refers to activity other than the principal activity undertaken on a short-term basis. If one uses this measure and computes the working population by multiplying the worker population ratio (WPR)—the total number of workers in the country’s population—with the population estimates made by the census for the years corresponding to the last three surveys carried out by NSS every five years, one will observe that in the first five years of the decade, the workforce increased by approximately 60 million, whereas the corresponding increase in the second half of the decade is only of two million.
This number has been used to conclude that employment growth has slowed. It also raises concerns about the quality of economic development.
This last conclusion is based on the assumption that all forms of employment in the UPSS measure are equally desirable. This is not correct because this bunched number hides many distinct structural components. Three of these are worth detailed assessment. These are: “Subsidiary status employment”, employment of the young, and employment of women.
The UPSS measure includes, in the definition of employment and workforce, both principal and subsidiary status activities. The principal activity status (PS) on which a person spent a relatively longer time (major time criterion) during the 365 days preceding the date of survey is considered the usual principal activity status of the person. A person may have, in addition to his principal status activity, pursued some other economic activity for 30 days or more during this time. Any such activity is termed subsidiary economic activity status (SS). The two measures are used to determine the workforce and the number of persons who are working and available for work (labour force). We should expect that improvements in principal status employment or household well-being can and should lead to reductions in subsidiary employment, a feature that emerges in the data almost consistently from 1983.
Employment in the young:
The overall WPR and the labour force participation ratio—the number of persons in the labour force in the population—are for all age groups. Absence from the labour force in different ages is for different reasons and has different social implications. Thus, absence of individuals who are younger than 24 is due largely to their attending school. This feature is sharpest in males and in females below 15, but remains significant in females over 15. As we see in the accompanying table, these age groups see sharp declines in their WPR. This decline then can be attributed to an increase in school attendance and a decline in child labour, an objective promoted consistently in government policies.
Employment of women:
A third hidden dimension in the data relates to the employment of women. Since the 1980s there has been a near-consistent decline in WPR for women. This has been commented upon earlier and there may be social and cultural reasons for this trend.
However, disentangling these distinct trends leads to very interesting comparisons in the employment growth depicted in the table.
The table shows that if we focus only on male PS employment in ages 25 or more then, far from a slowdown in growth of workers since 2004, we actually see a sharp increase.
In contrast, when looking at the employment among females, young and in subsidiary status, we actually see reductions between 2004 and 2010. Thus, the slow growth is due to the way in which this data has been combined. There is one other point not fully revealed in this table. Comparison over a longer period, from, say, the 1980s, shows a sustained decline in these categories of employment. We can see from the table that employment on account of subsidiary status, females and the young in 2004-05 shows an increase from 1999-2000, marking a break in this long-term pattern of decline. If we omit the data for 1999-2000, then the broad pattern of decreasing WPR’s in these categories is maintained.
What confounds the picture is the fact that in 1999-2000 we have a much sharper fall in WPR in these categories, leading to a correction, which yields an apparent rise in WPR between 1999-2000 and 2004-05. This phenomenon has been explained by some commentators as arising due to an employment slowdown in the 1990s, or due to the fact of 1999 being an abnormal year on account of a recession. In either case the results suffer from an illusion created by the 1999-2000 survey.
There are other problems in comparing Indian data and that from developed countries such as the US. These are due to the presence of self-employment, seasonal fluctuations, among others, all of which make it difficult to compare Indian numbers. In addition, there are other facets in the detailed results presented in NSS reports on wages and skills, which taken together paint an interesting picture of employment, one that is not at odds with high growth seen in recent years.
Graphic by Ahmed Raza Khan/Mint
T C A Anant and Rajiv Mehta are the chief statistician of India and additional director general of the National Sample Survey, respectively
Comments are welcome at firstname.lastname@example.org