The third annual round of the Periodic Labour Force Survey (PLFS) data conducted during July 2019-June 2020 was released recently. The PLFS captures key indicators of the labour market such as the labour force participation rate (LFPR) — the proportion of population working or seeking work; worker-population ratio (WPR) — the proportion of population that is working; and the unemployment rate (UR) — the proportion of population in the labour force that is seeking but unable to find work. It also provides data on the earnings of different segments of workers.
The PLFS 2019-20 was expected to provide official estimates of the labour market distress that followed dwindling GDP growth and a lockdown following the novel coronavirus pandemic that brought several economic activities to a standstill. The data, however, show a decline in the unemployment rate to 4.8% in 2019-20 — the lowest in three years. While the headline numbers may seem pleasing, a detailed analysis paints a rather disconcerting picture.
Falling unemployment rate
The LFPR, WPR and UR are measured using two approaches — usual status and current weekly status. The usual status considers the activity of an individual over a relatively long period during the last 365 days, whereas the current weekly status is based on activity performed during the reference period of the last seven days.
The unemployment rate, as measured by the usual status, fell from 6.1% in 2017-18 to 4.8% in 2019-20. This is because even as the LFPR increased from 36.9% to 40.1%, the WPR increased from 34.7% to 38.2% during the same period. In other words, while there was an increase in the share of the population in the labour force over the last three years, there was an even higher increase in the share of those who were able to find work, and hence unemployment fell.
A fall in the unemployment rate would be heartening, except, it seems puzzling as it comes at a time of unprecedented economic distress. The quarterly GDP growth declined for successive quarters, sliding from 8.2% in January-March 2018 to 3.1% in January-March 2020, after which the economy contracted by 23.9% during April-June 2020.
Workforce composition
How were more people able to find jobs when economic activities were slowing down? The answer lies in the changing composition of the workforce.
The PLFS categorises the workforce into self-employed (which includes own account workers, employers and unpaid helpers in family enterprises); regular wage/salaried workers and casual labourers. Own account workers run small enterprises without hiring any labour but may take help from family members, while employers hire workers. Of all the worker categories, only the proportion of unpaid family workers has gone up significantly in the last three years. In fact, between 2018 and 2019, while the workforce increased by 2.9%, the proportion of all other employment categories in the workforce declined, except unpaid family helpers.
Over the same period, almost the entire rise in the workforce was accommodated by agriculture. Agriculture continues to perform the function of a sink — absorbing the workforce that cannot find remunerative employment elsewhere.
There is also a gendered dimension to the changing composition of the workforce. The category of unpaid family workers is dominated by women. The story of the declining unemployment rate can largely be explained by a movement of women from primarily being engaged in domestic work to agriculture and other petty production activities as unpaid family helpers, possibly in the hope of increasing family income in the times of unprecedented distress and lack of alternative employment opportunities.
The usual status is based on a loose definition of work that underestimates open unemployment. This is where the alternative measure of unemployment is relevant. Using the current weekly status approach, the unemployment rate was estimated to be 8.8%, unchanged during the last three years.
Impact of the lockdown
The PLFS survey for April-June 2020 overlapped with the national lockdown. The current weekly status unemployment rate in this quarter was 14%, and the urban unemployment rate was around 20%. Corrected for inflation, the average monthly income for the salaried increased by 2% in April-June 2020 over April-June 2019. The monthly earnings of the self-employed declined by 16% and the daily wage for casual workers declined by 5.6% over the same period. The real monthly per capita consumer expenditure declined by 7.6%.
The rise in the average income of salaried workers and the muted impact on consumer expenditure, as estimated from the PLFS, do not concur with other data for the lockdown period. Private final consumption expenditure declined by 26.7% in April-June 2020 over the same quarter in 2019. Numerous small-scale surveys also reported massive earnings loss during the lockdown. There is overwhelming evidence to suggest that the PLFS data may underestimate the loss of earnings and fall in consumption during the lockdown. This is a missed opportunity for the official survey to capture the labour market dynamics during the lockdown.
Strengthen statistical system
There is no official data on poverty after 2011-12 or on farm income after 2013, and no recent data on migrant workers. While the consumer expenditure data for 2017-18 was buried, the data on situation assessment of agricultural households are not yet released, despite being conducted between January-December 2019, before the latest PLFS.
Minor tweaks in future PLFS surveys can fill the data gaps. Currently, the PLFS captures incomes from agriculture and monthly consumer expenditure, but the questions on these aspects lack credibility. The predecessor to the PLFS, the National Sample Survey employment and unemployment surveys, collected data on consumer expenditure using a detailed schedule. There is no reason why the PLFS cannot do the same. Adding questions on costs and returns from cultivation and related activities can also capture more accurate data on agricultural incomes. Lengthening the questionnaire has its costs — but the costs of the absence of reliable and timely data on important policy-relevant indicators are far higher.
Ishan Anand is Assistant Professor, Jindal Global Law School, O.P. Jindal Global University, Sonepat, Haryana
