Better data needed on job scenario

It's essential for India to have a 21st-century labour market assessment system for analysing trends

"There are lies, damned lies, and statistics,” the 19th century British Prime Minister Benjamin Disraeli famously said. Today, his description of questionable data may be applied, somewhat facetiously, to Indian labour statistics. The problem is not that they are “fake”, but rather that they give only a partial and sometimes inaccurate view of India’s job situation. Not surprisingly, the publication of these statistics has stirred up a heated debate as to whether India is undergoing a period of jobless growth.

At face value, Labour Bureau data are quite alarming. Take a closer look, however, and we find some significant gaps. Quarterly surveys of enterprises suggest that India’s labour market had meagre growth in the range of 150,000 to 400,000 jobs each year from 2013 to 2016. But these surveys do not provide an accurate aggregate national view. Most enterprises have fewer than 10 employees, but the quarterly surveys only track companies with more than 10; they make up less than two per cent of the national total.

Annual surveys, conducted over a larger base of households, suggest that fewer than two million jobs are being created annually. But that conclusion misses a crucial change: a marked structural shift in the workforce away from agriculture and towards the non-farm sector, particularly construction, trade, and transport. The data indicate that employment in agriculture may have shrunk by around 26 million between 2011 and 2015, while non-farm jobs appear to have risen by 33 million (exhibit).

That non-farm job growth outpaced the decline in agricultural employment is a positive sign. Yet, the data suggest that the labour force participation rate fell by three percentage points, from 55.4 per cent in 2011 to 52.4 per cent in 2015. These findings need further validation, including by testing the statistical robustness, since falls in labour force participation are not usually discernible over such short periods. Even if labour participation is declining, we need to understand more clearly why that is the case. For example, it may indicate that more young people have stayed in education, or that more women from households, once in extreme poverty, are entering the middle class. Both of these are welcome social developments rather than a sign of labour market weakness.

As India’s economy evolves and modernises, it is essential for the country to design and implement a 21st-century labour market assessment system that enables more relevant, timely, and accurate analysis of the trends, which is currently under active discussion in the government under a task force on employment data headed by NITI Aayog Vice-Chairman Arvind Panagariya. Moreover, India now needs to focus on the quality of jobs, rather than simply their quantity. A new paper from the McKinsey Global Institute argues that this should be done through a new emphasis on “gainful employment” that encourages the creation of better, safer, higher paid, and more productive work. It is a more holistic way of thinking about employment — but we need to be able to measure it.

All figures in million

Note: Years are financial years from April to March; thus 2011 is FY2011, from April 2010 to March 2011. Numbers may not sum due to rounding

Source: Labour Bureau; UN Population Division (medium variant); McKinsey Global Institute analysis

New statistical tools we could introduce include:

*A full-fledged labour market employment survey conducted on a quarterly basis, with a much larger sample size, and validated by supplementary data from enterprises, job search portals, sales revenue, and tax-related data systems of the government. The recent announcement that NITI Aayog will conduct a quarterly household survey across urban areas, and an annual one across both urban and rural areas, to estimate the number of employed and unemployed people is a step in this direction.

*Wages and incomes could be more comprehensively measured and harmonised with the consumption surveys. Longitudinal studies can be carried out using a constant panel of households to measure social mobility.

*Time-use surveys are essential to capture how much work is obtained and how time is allocated, including across specific tasks, paid work, and unpaid work. A long-overdue modernisation of occupational definitions would be needed to reflect changes in the job market.

*Specific labour segments can be covered more deeply; for instance, an annual higher education graduate survey could cover graduates of selected disciplines.

*State-of-the-art data collection methodologies and surveying tools can be deployed in the labour field, with digitally enabled registration, data recording, verification processes, and speedy release of unit-level open data.

These are just some ideas for ways to enhance our understanding of how India’s labour market is changing. Indians aspire to higher pay, better, and more productive working conditions, and safer, cleaner, and more stimulating work. Better statistics are an important starting point to achieving those aspirations.
 

Rakesh Mohan, a former executive director at the IMF and deputy governor of the Reserve Bank of India, is a distinguished fellow at Brookings India and Senior Fellow at the Yale Jackson Institute for Global Affairs. Anu Madgavkar is a partner at the McKinsey Global Institute.