Employment Trends in India: A Fresh Look at Past Trends and Recent Evidence

1
Employment Trends in India A Fresh Look at Past
Trends and Recent Evidence

• Himanshu
• Fellow, Economics
• Centre de Sciences Humaines

2
Workforce Participation Rates tend to decline
over time in rural areas. The exceptions in this
regard are 43rd round CDS estimates, 50th round
and 61st round.
3
Urban WPR have remained stable for most period,
except for sudden jumps in 50th and 61st round
4
Unemployment rate for rural males have been
rising for most periods except for 50th round and
43rd round CDS. For females, they show
fluctuations before 50th round but have been
rising since then.
5
Urban male unemployment rates have been coming
down over the periods with sharp fall for 50th
round and increase during 55th round. However CDS
estimates of 61st round show increase. For
females, unemployment rates increase sharply in
the 61st round.
6
The trend in LFPR is similar to the trend
observed in the case of WPR for rural areas.
7
Urban male LFPR has been stable till 43rd round,
jumps sharply for 50th round and even sharper
during 1999-2005. the trend for females is also
similar
8
Percentage of Self-employed workers among total
workers in rural areas was declining till 55th
round but saw a reversal of the trend for 61st
round. For females, the increase was sharp enough
to nullify the decline till 55th round.
9
The reverse is true for casual wage labour in
rural areas
10
Urban males also saw self-employment increase
sharply at the cost of wage labour during
1999-2005. for females, the trend of increasing
regular employment continued.
11
Primary sector employment among rural male
workers has been coming down steadily except for
50th round when there was almost stagnation. For
females, 50th round actually saw primary sector
employment increase. However, 61st round saw
sharp non-farm diversification for both males and
females.
12
Tertiary sector is the largest employer in urban
areas with share increasing since 38th round. But
saw a sharp fall in 61st round.
13
The decline in tertiary sector employment is
balanced by a huge increase in secondary sector
employment, which was otherwise declining till
55th round
14
Why do 43rd round daily status estimates behave
differently?

• 43rd round CDS employment estimates in rural
  areas appear out of place compared to estimates
  from CWS, PS and Usual status.
• More importantly, the fact that CDS estimates for
  43rd round are very close to CWS estimates for
  this round raise suspicion on the official
  estimates of 43rd round CDS.
• In general CDS and CWS estimates will never be
  close because of the way these are collected.
• Unit level data suggest that the official
  estimates are probably wrong
• The corrected estimates as obtained from unit
  records fit in line with the expected trend.

15
For 50th round, the large increase in LFPR and
WPR and sharp fall in unemployment rates appear
suspicious.

• This round also shows abnormal trends in terms of
  employment status and industrial distribution.
• For example, 50th round shows almost no non-farm
  diversification while all other rounds show
  non-farm diversification.
• This round also shows abnormal fall in share of
  regular employment
• Such large increase during initial years of
  reform where even the government conceded that
  employment may fall, appears suspicious

16
What happened during 50th round?Basically a
change in method of classifying usual status
activity

• In the earlier NSS quinquennial surveys the
  identification of usual status involved a
  trichotomous classification of persons into
  'employed', 'unemployed' and 'out of labour
  force' based on the major time criterion. In the
  50th round, the procedure prescribed was a two
  stage dichotomous procedure which involved a
  classification into 'labour force' and 'out of
  labour force' in the first stage and the labour
  force into 'employed' and 'unemployed' in the
  second stage.

17
More specifically, the change meant the following
Number of Months in Activity Number of Months in Activity Number of Months in Activity
Labour Force Labour Force Not in Labour Force Principal Usual Activity Status by 50th Round
Person Employed Unemployed Not in Labour Force Principal Usual Activity Status by 50th Round
A 5 4 3 Employed
B 4 5 3 Unemployed
C 4 3 5 Employed
D 4 1 7 Out of Labour Force
Note In case of C as per the procedure followed
in past rounds, he would have been categorised as
not in labour force whereas he/she is now
categorised as employed. Instructions to NSS
investigators for 50th Round of EUS, Section
Five, Item 5.4.11
18
Changes regarding weekly and daily status in the
50th round

• In the earlier surveys, the current weekly status
  (CWS) of a person was first assigned on the basis
  of the response to the questions relating to his
  participation in gainful activities (non-gainful
  activities) and thereafter the daily time
  disposition data was collected only for those in
  the labour force as per the CWS. In the 50th
  round, the daily time disposition was collected
  for all the persons surveyed and the CWS was
  determined based on the time disposition data so
  collected, without probing any further on this
  point.

19
Estimates from the 61st round

• 61st round estimates are also different from the
  general trend but so far there is no evidence
  that this could be result of any change in
  methodology.
• These are in fact also supported by trends from
  the intervening annual rounds.
• However, Sundaram and Tendulkar have argued that
  these annual rounds after 55th round are probably
  giving estimates which are biased and suffer from
  large RSE and non-sampling errors.
• However, in a curious argument they also suggest
  that the rural estimates of 59th round are less
  biased and hence probably right. But the urban
  are biased and have large non-sampling errors.
• Unni and Raveendran are not convinced by these
  estimates and suggest that the 55th round was an
  underestimate because it was not a normal
  agricultural year?
• However, these trends are consistent with actual
  events in rural and urban areas and the trends
  after 50th round are comparable to each other.

20
The decline in wage employment was mainly in the
agricultural sector. However, the increase in
self-employment was seen for both agriculture as
well as non-agriculture. But within non-farm,
rural growth was higher than urban growth, while
it was reverse in the previous period
21
But why should increase in LFPR and WPR be viewed
with suspicion?

• Basically two reasons
• The first relates to the demographic pattern.
  LFPR and WPR can increase simply because the
  share of working population in the total
  population increases. That is, even if all the
  age groups maintain their respective LFPR and
  WPRs, the overall LFPR and WPR will increase
  simply because the weight of high LFPR and WPR
  age group increases.
• Educational attendance, which drives down WPR and
  LFPR especially among young
• The net impact may depend on the balance of these
  two if nothing else happens.

22
Income effect?

• Then there is the income effect which ST argue
  was responsible for driving down the LFPR and WPR
  during 50-55th round.
• That is LFPR declines when income improves and
  conversely, may increase if income declines.
• This appears plausible for 61st round.
• There was overall deceleration in agricultural
  output.
• It was also accompanied by sharp fall in prices
  of agricultural products
• Moreover wage rate growth collapsed for both
  casual and regular workers
• In other words, agrarian crisis could have pushed
  some people in the labour force

23
By age group the growth in labour force for
25-59 age-group for males appears consistent with
their increased share in population. However,
1999-05 also shows increase in labour force for
5-25 age-group which was showing decline in
earlier period or growing slower. Same is true
for old age population.
24
Break up by land owned classes throws up another
interesting fact for males, the increase is
higher for the land-less but for the females it
is actually happening in households which own
more than 1 acre of land.
25
Break up by industry and land size class also
suggests that the shift towards non-farm
employment is essentially due to the shift of
land-less and marginal land owning classes.
26
But within non-farm employment, percentage of
informal employment has increased substantially
including among regular workers.
27
Nonetheless, total non-farm employment which was
almost stagnant in the 1990s saw employment
increasing very sharply during 1999-05.
28
Casual wage rate growth decelerated for all
categories and all sexes
29
Even for regular wages, the only class which saw
positive increase was graduates and above. All
others saw real wages decline
30
This is also seen from the ASI data for the
factory sector, where the growth of worker wages
which increased slowly in nominal prices but has
seen decline in real terms for both rural and
urban areas. However, it is also seen that
managerial emoluments which were growing at more
or less similar rate before the reforms increased
faster than worker wages after reforms and the
gap has been increasing faster since 1999-00.
31
ASI data also shows that the percentage of wages
and profits in Net Value Added in the factory
sector was fairly stable in the 1990s. But after
that, percentage share of profits almost doubled.
Moreover, the decline in wage share during
1999-05 was despite NVA/worker almost doubling
during the same period.
32
has Indian Economy become more Lewisian?

• It appears so
• Non-farm employment, led by manufacturing,
  increased substantially at almost constant wages.
• Wage share has declined and profit share
  increased allowing the capitalist class to invest
  more, also reflected in the growing investment
  rate
• This was also facilitated by the weakening of
  trade unions and workers bargaining power.

33
But is it sustainable?