ISSN (Online) - 2349-8846
-A A +A

Stagnant Employment Growth

Last Three Years May Have Been the Worst

Vinoj Abraham ( teaches economics at the Centre for Development Studies, Thiruvananthapuram.

Employment growth in India slowed down drastically during the period 2012 to 2016, after a marginal improvement between March 2010 and March 2012, according to the latest available employment data collected by the Labour Bureau. There was an absolute decline in employment during the period 2013–14 to 2015–16, perhaps happening for the first time in independent India. The construction, manufacturing and information technology/business process outsourcing sectors fared the worst over this period.

The author acknowledges research support by Pooja Padmanaban.

India’s employment growth, as recorded by the employment and unemployment survey (EUS) conducted by the National Sample Survey Office (NSSO), had been slowing down since 2004–05. From about 2% per annum between 1999–2000 and 2004–05, employment growth declined to around 0.7% per annum between 2004–05 and 2009–10, and further slowed down to around 0.4% per annum between 2009–10 and 2011–12 (Mehrotra et al 2014; Shaw 2013). However, since the last thick sample round of the EUS in 2011–12, there has been scant information on employment growth, but much speculation around it. In this scenario, the three available data sources, the Labour Bureau’s Employment–Unemployment Surveys (LB–EUS), the Quarterly Quick Employment Surveys (QES), and the revised QES that give information for more recent periods, will help us decipher labour market trends after 2011–12.1

The Labour Bureau had been conducting nationally representative annual EUS of households, the latest round being for 2015–16 with a sample size of more than a lakh households, which is greater than the sample size of NSSO’s EUS.2 The QES, usually conducted quarterly in a year, provides employment figures based on establishments with more than 10 employees, largely in line with the definition of the organised sector in India. The QES had covered only seven sectors of manufacturing enterprises and information technology/business process outsourcing (IT/BPO) with the limited sample size of about 2,000 firms till end of 2015. Since 2016, in the revised QES, the coverage has widened to include all important segments of the services sector as well3 and the sample size has increased to more than 10,000 enterprises. It may be noted that using these two sources, we get evidence from both the supply-side (LB–EUS household surveys) and demand-side (QES enterprise surveys). Based on these surveys, a brief overview of the Indian labour market is presented.

Employment Growth and Generation

As per the LB–EUS, estimates based on usual principal status (UPS) criteria for the age group 15 years and above, total employment in India shrunk by about 0.4% per annum, during the period 2013–14 to 2015–16 (Table 1). This corresponds to an estimated reduction in employment for about 37.4 lakh persons during the period 2013–14 to 2015–16. If we consider the total employment based on sectoral estimates then the drop would have been to the tune of 53 lakh jobs (see note appended to Table 1 for the two estimates). It seems that the deceleration of employment growth during the period 2004–05 to 2011–12, has not reversed, but has worsened during the period 2013–14 to 2015–16. Perhaps this is for the first time in independent India that we have an absolute decline in employment, though there had been phases of employment growth being low at around 1% during 1993–94 to 1999–2000 and below 1% during 2004–05 to 2011–12. Employment growth, nevertheless, was positive for rural males, while for all other segments, rural females, urban females and males, all had an absolute decline in employment. The decline in employment was more widespread in the urban areas, while in the rural areas, the drop in employment was limited to women.

From the demand-side, the QES also corroborates the dismal picture of employment creation in India visible in LB–EUS. Since January 2010, successive quarterly surveys show that employment creation had been slowing down (Figure 1). After October 2011, in no quarter was the addition in employment more than 2 lakh per quarter. Between 2014 and 2016, the labour market conditions further deteriorated to an absolute decline in employment in three of the 12 quarters. During the 24-month period from March 2010 to March 2012, the employment in the surveyed sectors increased by 18.15 lakh, while in the next 24 months from March 2012 to March 2014, the employment creation slowed down to just 6.2 lakh and in the following 19 months from March 2014 to December 2015, employment creation further slowed down to 5.92 lakh. The monthly average of employment creation, during 2010–12 was approximately 76,000 per month (Table 3, p 15). During March 2012–March 2014 and March 2014–December 2015, the average monthly employment creation declined to around 26,000 and 30,000 respectively. Within this last period, during March–December 2015, the average employment creation drastically fell to less than 8,000 jobs per month.

The QES since April 2016 are a revised series, with a much larger sample size and widened coverage of various sectors. However, even the revised series of QES confirms the trends of employment stagnation that are visible in both the large LB–EUS and the QES of the earlier version. The change in employment during April–June 2016 was 77,000, in the next quarter July–September 2016, 32,000 and October–December 2016, 1.22 lakh, with the monthly averages at 26,000, 11,000 and 41,000 respectively (Table 4, p 15). These numbers, both the absolute and averages, seem to show a continuum of the near stagnation in employment creation that had set in from 2012 for the organised sector.

All three sources of data, namely, LB–EUS, the previous version of QES, and the newer version of QES, point to the fact that employment growth has slowed down during the period 2012 to 2016. This is substantiated by data representing both the demand-side and supply-side. Drawing from the QES, it may be possible to infer that there is almost a secular decline in employment generation from quarter to quarter since 2010. However, the period January 2014 to December 2016 is marked by three instances of an absolute decline in employment.

A comparison of the growth trends in gross value added (GVA) shows that quarterly movements in employment creation for the sectors covered under QES seem to follow by and large the quarterly movements in GVA of the economy till mid-2014 (Figure 1). Between October 2008 and December 2009, both employment generation and GVA growth had accelerated. Between January 2010 and December 2013, both the variables had continuously decelerated to reach complete stagnation by end of 2013. It may be inferred that the slowdown in employment growth during this period is basically a scale effect, wherein the decline in GVA growth had led to the decline in employment generation.

However, since mid-2014 there is some divergence in the growth of employment and value added. During the period March 2014 to March 2016 while GVA growth started accelerating again, employment creation remained low. From January 2014 to March 2016 though GVA growth per annum was above 7% on an average, the employment additionally created continued to remain flat, realising “jobless growth.”

The growth of GVA had, however, declined since January 2016 and has continually fallen throughout the year, and by 2017, the first quarter results show an even further decline in GVA growth. With the GVA growth slowing down substantially to less than 6% during the quarter October–December 2016 into early 2017, it may be only realistic to assume that employment creation is going to slow down even further, at least in the organised sector.

Change in Employment by Industry

It can be seen from the LB–EUS that during 2013–14 to 2015–16, across various sectors, the largest decline in employment was experienced in the primary sector (Table 2, p 14). Employment declined by about 52.9 lakh in the primary sector, both in rural and urban areas. Nearly the entire decline in the primary sector was in agriculture, though other sectors such as mining and quarrying also experienced a decline in employment. It is to be noted that this decline in agricultural employment is mainly that of female employment (reference period as UPS) in the sector.

The decline in employment in agriculture during 2013–14 to 2015–16 is a continuum of the trends visible during the period 2004–05 to 2011–12. Himanshu (2011) and Thomas (2012) had observed an absolute decline in employment in the rural agricultural sector by about 200 lakh between 2004–05 and 2009–10 and Mehrotra et al (2014) accounted for a decline of about 130 lakh during 2009–10 and 2011–12. Much of the decline in employment during this period was due to a decline in female employment in the sector (Abraham 2013) and shifting of male employment to non-farm sectors. As developing economies undergo the Lewisian structural transformation, it is expected that agricultural employment declines and non-agricultural employment would rise due to productivity and wage differences between the two sectors. Structural transformation is also accompanied by an initial withdrawal of women from the labour force. Thus, the decline in the agricultural employment visible during 2013–14 to 2015–16 may be interpreted as partly due to the long-term structural shift in the economy from agriculture to non-agricultural sectors. However, this period is also accompanied by a slowdown in agricultural GVA growth. While during 2013–14, the agricultural GVA grew at 5.57% per annum, during 2014–15 it fell to -0.19% per annum and 0.69% in 2015–16. In 2016–17, the growth rate increased to 4.88% (DBIE 2017), but available data for the first quarter of 2017–18 again shows a fall to 2.3% (CSO 2017). Given the decline in agricultural GVA, that in employment may be partly due to negative scale effects as well, and not entirely due to long-term structural transformation.

The fall in employment in agriculture had been conventionally accompanied by increasing labour absorption in two industries, namely manufacturing and construction, along with other sectors such as trade and transport, albeit in the unorganised sector. However, what is worrisome in the recent period is the absolute decline in both manufacturing and construction sector employment, especially in urban areas.

In the secondary sector, employment decline was highest in urban manufacturing followed by construction. While manufacturing employment had increased by 98 lakh persons between 2009–10 and 2011–12 (Mehrotra et al 2014), between 2013–14 and 2015–16 there was a decline of 21 lakh jobs in this sector. Further, most of the decline in employment in this sector was concentrated in urban areas. The construction sector also experienced an absolute decline in employment of about 4.2 lakh. In the services sector, there was an increase in employment by 63.5 lakh in rural areas, while in urban areas, employment increased by just about 2.9 lakh. Within the services sector, the increase in employment was concentrated mostly in one sector, namely wholesale and retail trade, and motor repairing. In rural areas, this sector accounted for more than 40% of all employment generated in this period, in urban areas this accounted for more than 62.5% of all employment generated. In rural areas, public administration and education sectors also saw some increase in employment, in urban areas, while the wholesale and retail trade saw an increase in employment, there was a decline in employment in almost all other sectors within services.

Thus, it can be seen that in general, the decline in employment is a widespread phenomenon, affecting most labour-absorbing sectors with the singular exception of trade and motor vehicle repairs. The fact that the productive non-agricultural sectors of the economy, such as manufacturing, have shown negative growth in employment does not fit the structural change argument. The absolute fall in employment in sectors such as construction, which acted as an employer of last resort for the ones displaced from agriculture, implies the dismal state of employment creation in India. What we are experiencing currently in the Indian labour market, especially in the non-agricultural sectors, cannot be viewed as part of long-term structural change.

It is possible to make some supporting inferences regarding the sectoral performance of the organised sector from the QES as well. Between March 2010 and March 2012, the increase in employment in these eight sectors was to the tune of 18.15 lakh jobs, of which nearly 70% of the additional employment was generated in the IT/BPO segment. During the period 2012–14, the employment created fell sharply to 6.24 lakh new jobs, but this drop in employment was mainly owing to the drop in employment in the IT/BPO segment from 12.46 lakh jobs in 2010–12 to merely 1.96 lakh jobs in the period 2012–14. Thereafter, the IT/BPO segment has not shown any major signs of recovery in the subsequent period March 2014 to December 2015. During the period March 2014 to March 2015, five out of the eight sectors recorded a decline in employment. This deepening and widening of the employment crisis is visible from the monthly average employment created which declined to just 8,000 during March to December 2015. Corresponding to these periods, the coefficient of variation for quarterly employment creation declined from 0.68 to 0.61 to 0.34, implying the spread of the decline to nearly all sectors.

As noted earlier from the LB–EUS as well, manufacturing sector employment creation for the surveyed sectors of QES had been declining from 5.29 lakh to 4.38 lakh, to 3.31 lakh during the period 2010 to 2012, 2012 to 2014, and March 2014 to December 2015, respectively. However, it may be noted that overall, these organised sector industries show positive growth in the aggregate. So it may now be possible to argue that the brunt of the absolute decline in employment is in the unorganised sector, at least in case of these sectors, while the organised manufacturing sector is seeing a dramatic slowdown in the employment growth.

With the revised QES for 2016, it can be seen that the stagnation in employment creation continues well into 2016 though signs of marginal recovery is visible in the last quarter of 2016, during October–December 2016, at least for the organised sector. For the period April to December 2016, the total employment created in these sectors was 2.3 lakh. The employment decline that was recorded in the manufacturing sector continues till April–June 2016, thereafter there seems to be recovery in employment creation in July–September 2016 and further during October–December 2016. However, the decline in employment creation in construction sector continues throughout the year. The decline in IT/BPO employment also seems to continue till April–June 2016 and subsequently a marginal revival in the later part of the year. Trade, which had shown robust growth during the period 2013–14 to 2015–16 as per the LB–EUS also seems to have stagnated by mid-2016.

Absolute Decline

Thus, the picture that emerges from these data sources is that of an absolute decline of employment in India, with much of it probably in the unorganised sector, while the organised sector is seeing a sharp decline in the growth of employment. This stagnation in employment is widespread, affecting almost all sectors of the economy, as is visible from three different data sets. This slowdown in employment growth was marked since 2004–05, however, the period 2012–13 and after is marked by absolute declines in employment.

Even though employment creation had stagnated for the manufacturing sector, GVA growth had been high at above 7% rates during this period, resonating the “jobless growth” phenomenon for the past years, wherein employment elasticity of manufacturing value added had remained at less than 0.10 during 1999–2000 to 2009–10 (GoI 2013). However, in the construction sector, the decline in employment was accompanied by sharp fall in GVA growth as well. This is also the case with the real estate, finance, and business services sector wherein both GVA growth and employment creation slowed down. These two sectors had the highest employment elasticities during 1999–2000 to 2009–10, at 1.13 for construction and 0.66 for real estate finance and business services. The decline in growth of GVA of these two labour absorbing sectors has greatly impeded employment creation in the past few years. Since the IT/BPO segment also falls within business service, the decline in the growth of this sector, and in turn, its contagion on construction sector may well have possibly played a key role in reducing employment growth. The slowdown in growth rates of GVA during the first quarter of 2017–18, especially in manufacturing, construction and business services, may only aggravate the employment crisis further.

Unemployment Rates

It seems that the slowdown in employment growth had aggravated the problem of unemployment (Table 5). The UPS-based overall unemployment rates increased from 38 per thousand in 2011–12 to 50 per thousand. Unemployment rates had been gradually increasing in India, at least since 1999–2000 as per the NSSO’s EUS.4 The gradually rising unemployment rates for the rural areas continue to show a rising trend till 2015–16 in the LB–EUS as well. The urban unemployment rates also show a rising trend till 2012–13, thereafter there seems to be a marginal decline. Given that these are open unemployment rates pertaining to the major-time criteria, which has conventionally shown very low figures, owing to the lack of alternatives for the poor, these increasing rates of open unemployment may also be indicative of the worsening underemployment, a more pertinent problem for India.


From the analysis, it seems to be clear that since 2011–12 the labour market in India is facing a severe crisis with employment growth stagnating across almost all sectors, and unemployment rate is worsening. This trend of employment stagnation is visible from multiple sources of data. The period since mid-2014, wherein there was a change of government, does not show any improvement in the labour market conditions. If anything, these indicators seem to have got worse since mid-2014. There is an absolute decline in employment during the period 2013–14 to 2015–16, which is perhaps happening for the first time in independent India. The stagnation in employment growth seems to be largely a scale effect, where the slowdown in economic growth, especially of labour-absorbing sectors such as agriculture, construction and business sector, has constricted employment growth. Though some of these trends, such as decline in agricultural employment, have their roots in structural transformation of the economy, it is the loss of growth dynamism of highly employment elastic sectors, such as construction and finance, and business services, that have aggravated the stagnation in employment creation. While laudable efforts like the “Make in India” programme aimed at enhancing manufacturing growth in India may have positively affected manufacturing, its conventionally low labour-absorbing capacity seems to have not reaped benefits for the working class. Given that GVA growth rates have fallen continuously during the last quarter of 2016–17 and 2017–18, arguably a fallout of demonetisation and the goods and services tax, it is possible that employment growth may further remain subdued in the near future. Much of the employment decline is probably concentrated in the unorganised sector, while the organised sector has experienced a slowdown in growth, thus the weakest among the working class are bearing the brunt of the employment decline.


1 Both the Labour Bureau data sets have been criticised in the past for their lack of representativeness. While the LB–EUS had small samples initially, the QES suffered from low sample size and poor coverage in terms of sectors. For instance, see BusinessLine (2017).

2 The 68th round of NSSO on employment and unemployment and second round of Labour Bureau’s Employment–Unemployment were conducted in 2011–12 with approximately the same reference period. Indicators of labour market such as the total labour force participation rates (LFPR) and workforce participation rates, based on usual principal status (UPS) criteria for age 15 and above, estimated by both these surveys match reasonable well. This gives us some confidence is using LB–EUS in the absence of NSSO’s EUS for recent periods.

3 Labour Bureau states these selected industries represent more than 80% of all industrial employment in industries with more than 10 employees.

4 As can be seen from Table 5 the LB–EUS results for the same year and category of population, shows a higher unemployment rates compared to NSSO’s EUS. So the levels shown in NSSO’s EUS and LB–EUS are not comparable for the years. But the trends in both the surveys can be interpreted together.


Abraham, Vinoj (2013): “Missing Labour or Consistent ‘De-feminisation,” Economic & Political Weekly, Vol 48, No 31, pp 99–108.

BusinessLine (2017): “Panagariya to Head Task Force to Measure Job Creation,” BusinessLine, 9 May,

CSO (2017) “Press Release of Ministry of Programme Implementation,” 31 August, Central Statistics Office, Ministry of Statistics and Programme Implementation, Government of India.

DBIE (2017): “Database on Indian Economy,” Reserve Bank of India, viewed on 10 August 2017.

GoI (2013): Twelfth Five Year Plan, Government of India, Vol 3, p 136.

Himanshu (2011): “Employment Trends in India: A Re-examination,” Economic & Political Weekly, Vol 43, No 59, pp 43–59.

Mehrotra, Santosh, Jajati Parida, Sharmistha Sinha and Ankita Gandhi (2014): “Explaining Employment Trends in the Indian Economy: 1993–94 to 2011–12,” Economic & Political Weekly, Vol 49, No 32, pp 49–57.

Shaw, Abhishek (2013): “Employment Trends in India: An Overview of NSSO’s 68th Round,” Economic & Political Weekly, Vol 48, No 42, pp 23–25.

Thomas, Jayan Jose (2012): “India’s Labour Market during the 2000s: Surveying the Changes,” Economic & Political Weekly, Vol 48, No 51, pp 39–51.

Updated On : 27th Sep, 2017


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top