Public Sector Employment

What has changed? ∗

R Nagaraj
Abstract

This paper, first, reports the trends in the size and composition of public sector employment – a crucial
strata of the working class in many a political economic explanations of contemporary India. Since 1997,
as per official figures, its size has shrunk by 10 per cent – expectedly, with the market-oriented reforms.
But the NSS estimates show a boom in public employment in 2000s, growing annually at 3.4 per cent
(compared to 1.5 per cent in the economy as a whole). In reconciling the contradictory trends, one realizes
how little things have changed. Why is it so, despite an official ban on fresh recruitment? The reasons, the
paper contends, seem to lie in the demands placed on public employees by the electoral politics.

Indira Gandhi Institute of Development Research

A K Vaidya Marg
Goregaon East
Mumbai 400065.

November 1, 2014

Email: nag@igidr.ac.in

∗
Paper prepared for the Conference on Political Economy of Contemporary India, being held at Indira Gandhi
Institute of Development Research, Mumbai during November 20-21, 2014. This is part of the trilateral
collaboration between UC Berkeley, King’s College London, and IGIDR, Mumbai under UK-India Educational
Research Initiative (UKIERI). This is first draft. Please do not quote or cite, comments welcome.
 Picture

“Why has no one turned up at the office today? Is a cricket match on somewhere in the world?”

“Don’t you fellows read newspapers or listen to ministers speak? India is shining and feel-good factor is spreading
all over, and you chaps have come with good old demands?”

R Nagaraj 2 Public Sector Employment

 Public Sector Employment
What has changed?
R Nagaraj
No careful account of India’s political economy is complete without reckoning with the size and
composition of public employment, invariably described as a leviathan, bloated, dysfunctional
and inefficient – and, as caricatured in R K Laxman’s countless cartoons in Time of India for
decades (see Picture). Public employees form a vital constituency in the electoral politics: most
political parties have a trade union wing representing the white-collar workers. Moreover, the
unity of this section of the working class – cutting across party lines – is also evident in how it
protects its interest in securing periodic pay rise every decade under the Central Pay
Commission.

That the white-collar workers form a significant constituency is widely accepted in varied
analyses of India’s political economy. If they were part of the “intermediate regimes” in K N
Raj’s (1973) account, they were also a constituent of Bardhan’s (1984) “professional in the
public sector” as a propertied classes; they were the salariat in Patnaik and Rao’s (1977)
formulation of a mixed underdeveloped economy (emphasis as in original); and, they were
among the rent-seekers in T N Srinivasan’s (1985) neo-classical economic story.

Scholars, in line with their preferred view, have all too readily accepted one of the above
characterizations as self-evident, without examining the evidence on the size, structure and
evolution in public employment. Perhaps it was Ashok Desai and Ena Desai – interestingly, from
a libertarian political perspective – made the only careful and critical analysis – in a multi-
country volume edited by Gus Edgren (1986) for the ILO-ARTEP, titled The Growing Sector.
Piecing together evidence from diverse sources, Desai and Desai explained the growth in public
employment as determined by India’s ability to tax and spend.

More recently, if there was one aspect of the market-oriented reforms that had wide appear
cutting across theoretical perspectives and political persuasions, it was to curb the size of the
bureaucracy – as a means of reigning in the fiscal deficit. While the political right believed in
shrinking the state as an end itself, most others advocated modernization and decentralization of
bureaucracy for better services delivery, by getting rid of the “unproductive” army of unskilled
and semi-skilled workers that is mostly concentrated in the capital cities.

Nearly three decades after Desai and Desai’s account of the “growing sector”, and over two
decades of the market-oriented reforms, where does the public employment stand now? Has the
leviathan been tamed? Has it been restructured responding to people’s rising aspirations for
better public services delivery, in tune with the changing state’s role in the economy? Has the
bureaucracy become more productive with office automation? How does the reform of public
employment in India compare with similar efforts elsewhere?

This short paper seeks to address some, but not all, of these questions. Section I presents a
factual account of the long-term trends in employment and wages in public sector, drawing from
many studies and official sources. Section II presents a contrasting statistical picture for the last

R Nagaraj 3 Public Sector Employment

decade using the NSS sample survey data. Reconciling the divergent trends, Section III offers a,
tentative, political economic explanation. Summarizing the main findings Section IV concludes
the study.

Section I
A Factual Account

Employment Size: In 2009-10, as per the Economic Survey, public sector employed 18 million
persons, aggregated over all levels and forms of government – constituting 4 per cent the
national work force, and their wages 8 per cent of GDP. Contrary to popular belief and scholarly
writings, compared to other countries the size of the public employment in India is not large,
relative to its population. Though, international data are hard to come by, the few available
figures attest this fact.

For instance, in the US, in 2012, all the levels of the government put together, employed 22
million persons constituting 15 per cent of the workforce (Rampell, 2012). As per a news report,
in 2011, for every 100,000 residents, India had 162.3 government servants; the figure for the US
was over four times larger at 768.1 (The Hindu, January 30, 2012). Considering that the size of
public sector in the US is small, relatively to other developed countries, one can reasonably infer
that public sector employment (relative to population) in the advanced countries is, in general,
much higher than in India.

How would India compare with developing economies? Available data is dated and scanty, but
the information put together in an IMF study does not suggest India to be a positive outlier
(Kraay and Rijckeghem, 1996). Therefore, India’s public employment – to get the facts right – is
not bloated, adjusted for population size, in a comparative perspective.

Figure 1 shows public sector employment in 2012 by type of government. State and local
governments accounted for about one-half of the employment: expectedly so, as the local
administration is responsible for service delivery – as in most countries. The central government
employs only 14 per cent of the total, the rest is in quasi-government that includes public sector
enterprises (PSEs), electricity boards, road transport corporations and so on.

However, public sector employment in India, as per the official estimates, has declined in recent
years. It rose from 7 million in 1960-61, to about 20 million in 1997, declined thereafter by about
10 per cent by 2012, to 17.6 million (Figure 2). The decline is however sharper relative to the
population: it fell by 36 per cent, from 2271.4 per thousand in 1991 to 1465 in 2012 (Figure 3).
Evidently, the reforms are responsible for the decline. It is similar to what has happened in the
advanced economies after the Thatcher-Reagan right-wing revolution since the late 1970s.

Employment Structure:

Table 1 shows the absolute numbers of employees, and percentage change in employment over
the years by types of government, for half a century (1961 to 2012). During the employment
boom, i.e., 1961 to 1981, most of the increase occurred in (i) quasi government broadly
representing the widening role of the state in the production of goods and services, followed by

R Nagaraj 4 Public Sector Employment

state governments to provide the welfare services. During the contraction (1991-2012), central
government bore the brunt – both in absolute and in percentage terms – followed by the quasi
governments.

Figure 4 depicting just the growth – netting out the effect of the levels by indexation – shows
employment rose most rapidly in the earlier decades, in quasi government. A similar picture also
emerges from the data just for the central PSEs – clearly related to growth in public investment
into ‘heavy’ industry initially, and later into energy sector (Figure 4). But interestingly this is
also the branch of government has witnessed most rapid decline after the reforms. In other
words, the long-term expansion and contraction in public employment is clearly related to the
changing perceptions of the economic role of the state.

Table 2 provides data on the sectoral distribution of public employment from 1961 to 2011.
Little over one-half of it has been in community, social and personal services (including public
administration and defense). Sectors that have lost employment share in recent times are
manufacturing, construction (state’s withdrawal from producing goods) and transport and
communication (technological change; move from coal to electric locomotives, and from
electromechanical telephone exchanges to electronic).

As is widely known, the pyramidical structure of government employment has a large base of
unskilled and semi-skilled employees – consisting mainly of clerks, peons, drivers and attendants
– supporting a small fraction of career bureaucrats and political executives at the top – largely a
colonial legacy. Liberal reforms during the last two decades have sought to rationalize the
bureaucratic structure, with modest efforts (improve). One suspects, much of the reduction in
employment is on account of contracting out of the auxiliary services like, transport,
maintenance, running canteens, security services etc – besides the natural attrition. 1

Data for the recent 6 years show that senior bureaucrats – that is, “gazetted” officers – remain a
small share (5 per cent) of total government (Table 3). The decline in central government
employees is mostly on account of Group D (or class IV) employees. In fact, after the recent pay
revision in 2009-10, the central government has, in principle, abolished this category of
employees, signaling the official commitment to restructure the lower end of the bureaucracy.

The other reason for the decline in public employment is natural attrition, on retirement: those
who joined the service in 1960s and the 1970s would have retired in the 1990s, and 2000s,
contributing to a decline in the pay roll. Moreover, after the reforms as a policy matter, central
government froze fresh recruitment, leaving posts vacant, amounting to 18 per cent of the
sanctioned positions in 2012.

Occupational distribution of public employees:

Population census data can be analysed for this purpose, as DGET has stopped providing the
information. Considering the preponderance of group C and Group D employees, it is safe to

1
Organizationally, PSEs were the old post-war style not just vertically integrated manufacturing enterprises, but also
townships (often located in remote places) providing most civic services. After the reforms PSEs were allowed to
contract out most of these services, thus reducing their direct employment.

R Nagaraj 5 Public Sector Employment

assume that majority of them perform unskilled and semi-skilled jobs (barring teachers and
nurses who also belong to group B and C employees). The fact that the distribution has remained
roughly the same (and that public schools and hospitals have declined relatively), it is safe to
infer that not much occupational restructuring in public employment has occurred [needs to be
confirmed].

Public sector wages:

Have public sector wages – as popularly believed – grown faster than those in the rest of the
economy? Considering that these workers are unionized, and form a vital constituency in the
electoral politics, there are reasons to believe that this segment of the organized workers have
managed to grab a growing large share of the national cake. But, if public employment is based
on meritocracy – that us, often selected on competitive examinations – it is reasonable to expect
such employees to receive efficiency wages (to compensate for qualification, skill, experience
etc), not competitive wages.

As public sector wages data are not readily available, it is difficult to come up with definitive
evidence. We have however put together all the available evidence to make an considered
assessment. Compiling the information from the central wage commission reports, Eapen (1980)
showed that the real earnings of officers has declined by about 40 per cent between 1960 and
1972, though the wages of peons has remained roughly constant. In a similar but wider exercise
for a longer period, from 1960 to 1983-84, Desai and Desai (1985) found that the wages of all
categories of public servants, have declined in absolute terms rather steeply: the indexed value of
real earnings fell from 100 in 1960-61 to 34 in 1983-83 for the highest category employees.
However, the lowest category employees’ real earnings rose to 135 during the same period.
Thus, during the early decades, public sector real wages declined in absolute terms, and the wage
differential between the lowest and the highest category of workers narrowed – indicating
progressivism in the wage setting mechanism.

For the more recent years, we have examined the data for six years from 2005-06 to 2011-12 for
central government employees – a period in which the 6th pay commission award was granted.
During this period, on average, for all categories of central government employees, real wages
rose at 7.9 per cent per year – compared to per capita income growth of 6.8 per cent per year.

Before drawing firm inference from the forgoing findings, however, the data limitations need to
be borne in mind. The wage data do not include imputed value of the non-wage benefits of
government employment – for instance, subsidized housing (for small fraction of employees),
access subsidised schooling, and health facilities; and above all, defined-benefit pension (until
recently) for the employee and his family. In an economy with practically no universal social
security and poor public services, monetary value of these non-wage benefits, therefore, could be
substantial, and hence the real earnings reported above could underestimate the true
compensation of public employment.

Yet another, though partial, measure of public sector earnings is to look at the wages in the
central public sector enterprises, as reported in the annual Economic Survey, available since
1972. This series may not exactly track the government wages, but is likely to be close to it.

R Nagaraj 6 Public Sector Employment

Over the 42 years, on a trend basis, the average earnings in central PSEs rose 5 per cent per year,
compared to per capita growth at 3.6 per cent. The apparent gain for PSEs’ workers may
however disappear considering their qualification, skill and experience. Moreover, such a
comparison overlooks the divergence in the trend during the periods of employment boom and
contraction. During the employment boom (from 1972 to 1988), earnings roughly kept pace with
per capita income growth; but during the contraction, earnings rose very steeply. Prima facie, the
finding vindicates the standard inverse relationship between wages and employment.

Or, does it? Whether the rising earnings represent the collective bargaining power of the
organized workers, or, if the rising wages reflect efficiency wage considerations are hard to
distinguish. But it is widely admitted that since the 1990s, organized workers are on the back
foot, as they lost the bargaining strength, with “reforms by stealth” taking hold in the labour
market. Therefore, it seems reasonable to infer that rising real earnings after the reforms
represent efficiency wage considerations. As many sectors reserved for public sector were
opened up, government had to concede higher wages to match the wages offered in the private
sector retain skilled and experience employees in relatively higher technology industries.

In sum, it seems hard to sustain the view that over the long run, real wages in public sector have
risen at a disproportionately faster rate than per capita income, albeit such an inference needs to
take into account the non-wage advantages, and the job security enjoyed by these workers, in an
economy with practically no social security benefits outside the formal sector.

Section II
An Alternative Estimate of employment

We now report alternative estimates of public employment, using the NSS sample surveys of
employment and unemployment (the thick rounds) for the years 1999-2000, 2004-05 and 2009-
10 (Table 4). 2 In 1999-2000, public employment was lower (at 16.2 million) than the comparable
official estimates (at 19.4 million) reported earlier – a clear difference of 20 per cent.

But while the official data show an 8 per cent contraction in employment in the last decade, the
NSS data reveal a whopping 40 per cent rise (to 22.6 million). In other words, the NSS reports an
annual employment growth rate of 3.4 per cent, more than twice faster than the economy-wide
employment growth at 1½ per cent per year. This is true of all major states (accounting for over
90 per cent of the total), barring Karnataka; Bihar, Jharkhand, Himachal Pradesh and Jammu and
Kashmir have in fact more than doubled their employment (Table 5). Interestingly, the Left front
ruled West Bengal reported a growth rate just about the average of the states, despite its avowed
political sympathy for the organized labour.

To get a sense of perspective on the above-mentioned NSS estimates, Table 6 shows decadal
compound growth of public employment since 1961 using the official estimates. Interestingly,
the boom in 2000s, is similar to those achieved in the 1960s and 1970s (in the heydays of public
sector).

2
We have restricted the analysis to a few broad aggregates, as the quality of the estimates for finer classification
may not be accurate, given the sample size and its design.

R Nagaraj 7 Public Sector Employment

If the NSS estimates are to be believed (most economists and policy makers do), what could
possibly account for the boom, especially in times of fiscal stringency, and an official ban on
fresh recruitment? Arguably, rapid economic growth improved public resources for the
expansion of employment, responding to citizens’ demand for better public services (with many
rights based benefits enshrined in the constitution lately). To check if such a proposition is
empirically valid, we estimated the simple correlation coefficient between the following
variables for the cross-section of the 17 major states, separately for 1999-2000 and 2009-10:

1. Public employment per 1000 population and per capita income,

2. Public employment per 1000 population and HDI

For 1999-2000, (1) above is statistically significant with a positive sign, for 2009-10 the
correlation coefficient is not significant. In (2), for 1990-2000, the correlation coefficient is zero;
in 2009-10, the correlation coefficient is positive but weak (significant at 90% level of
confidence). These findings, therefore, do not offer a clear evidence for the foregoing
proposition.

A suggested hypothesis:

The NSS data for 2004-05 and 2009-10 have information on the length of the job contract, as
given in Table 7. It shows a growing proportion of public employees have short-term contracts,
implying these are temporary workers.

The contradiction in the trends described above seems to tell a deeper story. While the
government pay roll has declined, employment in the sub-contracted (or out-sourced) services is
likely to have gone up. This probably is the underlying reason for the contradiction. As those
working for service providers usually have short-term contracts, which is perhaps captured in the
increasing share of workers with less than one-year contract, as shown above.

If so, however, one may wonder why workers employed by outsourced agencies are recorded as
public sector employees. Our impression – based on an earlier experience of studying sub-
contracting in manufacturing industries – is that the NSS surveys do not seem so finely designed
as to distinguish between those directly employed by government, and those by the service
provider for government. 3

Further, the fact that central government does not have any class IV employee on its rolls since
2009-10, it clearly indicates that the services rendered by these employees are now outsourced
with workers having short term labour contracts.

If the forgoing interpretation is valid, then the decline in public employment is factually correct,
but materially not. It is yet another instance of informalisation of formal employment, with a

3
This is consistent with what we observed in our survey of workers employed in small enterprises living in the
outskirts of Bangalore (Nagaraj, 1989). Semi-literate workers do not know, or would not want to reveal, that they
work for an outsourced agency. These employment arrangements tend to be informal with practically no formal job
contract, or even a letter of appointment. However, in terms of worker’s social standing, it is always better to tell
that he or she works for government or for a large, publicly recognizable organization.

R Nagaraj 8 Public Sector Employment

poorer service conditions. The decline in the official employment figures seems as deceptive as
the abolition of child labour in industries such as carpet weaving, bidi rolling or firecracker
manufacturing, where the employment is driven “underground”, into households away from
public glare, and official records.

What does forgoing discussion imply for the structure of public employment? Little restructuring
seems to have happened towards the much-needed skill intensive professional services, but
simply out-sourcing of a few basic services like housekeeping, security services, gardening,
transport services etc. Whether such a move has improved the service quality at reduced costs (as
popularly argued) remains to be seen.

Section III
Understanding the Size and Trends

With 4 per cent of the total work force, the size of public employment is not large – in fact, small
by the standards of the advanced economies. But, its composition seems archaic, quite in contrast
to advanced economy, where the bureaucracy is decentralized; mostly consisting of professionals
serving the social needs.

Official figures show a 10 per cent contraction in public employment since 1997, most of which
is in central government and in PSEs – not much in state and local government administration,
and some of it could simply represent retirement and VRS. However, impressionistically, the ban
on fresh recruitment in the central government after the reforms seem to have created shortage of
qualified professionals like statisticians, city planners, doctors and so on, without really denting
the archaic employment structure.

For improving the delivery of social services and meeting the growing demand, the need is
perhaps not for contraction, but of restructuring and modernization of public services – away
from the colonial pattern, towards a contemporary welfare needs, and being responsive to the
growing democratic aspirations. Now with many rights-based entitlements enshrined in the
constitution, the need to re-orient public employment is all the more greater.

But what has been accomplished is contracting out of a few basic civic services. Rising public
employment as per the NSS estimate during the last decade seem to support our argument. There
have been many official committees and commissions to reorganized public administration –
most recent one being the Veerappan Moily commission on administrative reforms – but no
government has had the inclination to implement them. Why so? The answer, we contend,
perhaps lies in the political economy – tied to the electoral politics – for two reasons.

But before coming to them, it is useful to dispel some popular and simplistic perceptions. That
Prima facie, the liberal reforms have tamed the bull of the organized public sector workers and
their unions. Such a view would gladden the hearts of many a libertarians, who would commend
the reformists for shrinking public employment even as the economic growth accelerated.

Yet, public sector workers and their unions may have the last laugh, since overall public
employment (as measured by the NSS) has in fact been rising, with practically little

R Nagaraj 9 Public Sector Employment

modernization or restructuring of the administrative set up. The central government has
abolished Class IV employees by en mass promoting them to the next level, without effectively
restructuring their job-content or service conditions. It is not known if a similar change has
occurred in states and local administration. Our hunch is, probably not: because there has
practically been no contraction in their employment.

In the intermediate regime thesis referred to in the introduction, the numerically significant
public servants and petty producers were valuable in electoral terms and were perceived as a
bulwark against the big business and foreign capital. But now, the erstwhile beneficiaries of the
state capitalism seem to have outgrown the need for the state support they received, and have
tightly embraced foreign capital and big business (NRIs are a big proxy of this group). If so, with
a right ward drift in politics, what prevents the political parties from dumping public employees?
My tentative explanation is as follows: In the age of crony capitalism that we seem to be in now,
public employees remain the backbone of the political set up – be it top civil servants or a
humble peon – needed to grease the wheels the patronage system.

1. As lower level government servants form the foot soldiers for conducting elections, they
probably play a vital role in subtly influencing electoral outcomes. Therefore, no political
party can afford to confront the non-gazetted officers (NGOs). For example, in the 1980s,
NT Rama Rao, the charismatic Telugu film idol and the founder leader of the Telugu
Desam Party (TDP) in Andhra Pradesh antagonized the lower bureaucracy by refusing to
concede their wage demands; pitting them against the masses of poor in public spending
(which he probably thought would fetch him more votes), but lost the elections badly,
apparently as the bureaucracy worked insidiously to defeat the TDP.

2. If corruption is the grease, cadres of bureaucracy are the wheels of the patronage system.
To put it starkly, winning elections require huge sums of money, mostly obtained by
corrupt means (in the absence of state funding). Bureaucracy is the conduit through
which corruption flows to politicians. Politics is a labour intensive activity; politicians
need to provide a regular source of income to their supporters (to reward loyalty). Doling
out public jobs is, as yet, a sure means of doing it, forming the pivots of the patronage
network. We would therefore contend, it is the political patronage system that keeps
public employment growing (or, keeps it from declining).

3. Fiscal constraints may have restrained the system at the central government, but probably
not at the lower levels. Interestingly, despite the hardening of the budget constraint after
the passage of the FRBM act, state governments have managed to keep employment
growing in all but one state. Regardless of the political party in power all of them have
preferred to reduce public investment (foregoing potential output and employment
growth), rather than risk shrinking public employment.

Ordinary citizens are often outraged by petty corrupt practices – a traffic cop collecting mamool,
or a sub-registrar collecting bribes for approving a house plan or, house registration. But what
remains unobserved from the common man’s eye is that often the cop or the sub-registrar is just
the front end (or the last mile) for a local politician or a minister, to whom the bribes get
routinely passed on without batting an eyelid. The front man probably gets to keep only a small

R Nagaraj 10 Public Sector Employment

fraction of the total collection, to recoup his cost incurred in getting the job. These practices are
quite institutionalized, not random occurrences – as many a careful observes of the ground
reality can attest. So, the entire network of lower-level bureaucracy is perhaps a necessity for the
patronage system to function. Hence, given the cost of running the risky business of electoral
politics, it is perhaps hard to expect the political executives to rationalize the system that could
cut the umbilical cord of electoral finance.

Surely, the form of corruption has evolved. It is no longer an entirely cash business requiring to
change hands physically. In the light of the recent spate of scams, which are of completely
different orders of magnitude, it is amply demonstrated how politicians have acquired business
acumen in demanding and receiving favour in terms of equity stakes in corporate entities (in
complex web of companies and trusts). Though the evolved form of corruption may have
become less “labour-intensive”, one suspects, it nevertheless requires labour services to
camouflage the operations, and to cover the tracks – lest the information trail (in an internet age)
may yet spill the beans. So, despite the more sophisticated forms of corruption, need for such
services as yet remains.

If the foregoing propositions (stated perhaps somewhat bluntly) merit attention, then this is what,
we would contend, explains the persistence of the “dysfunctional’ bureaucracy – without getting
modernized to meet the growing need of citizens. While the free-market reformists may take
credit for the modest fall in public employment, and trade unions may fret about anti-labour
policies under the “neo-liberal regime”, the reasons for the persistence of the size and structure
of public employment probably run deeper.

Section IV
Conclusions

Public sector employees, as part of the organized working class, form a vital strata in many an
account of India’s political economy. But widespread misconceptions about their size, structure
and growth persist. In seeking to dispel some of them, we noted the following:

1. Public sector employees of all levels and forms of government put together constitute 4
per cent of the national workforce, their wages 8 per cent of GDP.
2. The employment, relative to the population, is small, just one-fifth of that in the US
(which is widely known to have small public sector among the developed nations).
Hence, the popular view that India’s public employment is bloated is simply incorrect,
adjusted for the population size.
3. Its composition, however, is different from that in the advanced economies. Sector-wise,
a large share is in industry and public administration and defense. Skilled and
professional services (as proxied by class I and II employees) form just 5 percent of the
central government employment; most of public employment consists of low-level semi-
skilled workers. In the developed countries, public employment is in services and in
skilled occupations – providing health, education and social services.
4. Public sector wages in the early decades declined in real terms, and the wage dispersion
narrowed. The trend got reversed in the last decade. But it seems difficult to read very
much into the wage trends for two reasons. Public employees form the nation’s

R Nagaraj 11 Public Sector Employment

 meritocracy needing adequate compensation. Moreover, public sector jobs entitle many
non-wage benefits, which probably more than compensate for the decline in the earlier
decades and the modest rise in real wages.
5. Official figures show, public employment sector has declined since 1997 by about 10 per
cent. Most of the decline is in the central government and in quasi government. This is
mainly due to natural attrition (retirement), VRS, and increasing contracting out of basic
services of maintenance and logistics. Best evidence for it is the fact that now there are
no class IV employees on the rolls of the central government.
6. Those employed indirectly by the state (via service providers) has gone up, as shown by
the NSS data. In other words, government pay roll has declined, but those employed
indirectly has expanded substantially. However, little occupational restructuring towards
skill intensive services, office automation and decentralization seems discernible.
Because of the employment freeze in the central government, shortages of many kinds of
technically skilled staff seems to be adversely affecting the quality of public services.

Why has there been no reorganization of public employment (or the state’s the inability to do so)
to meet the growing demand for professional services, away from the colonial structure of the
bureaucracy. It points to some deeper political economic reasons for the status quo. This paper
has suggested two plausible explanations, both pointing to the nature of India’s democratic
polity.

One, no political party (or formation) would like to antagonize the non-gazetted officers (NGOs),
who are in-charge of conducting elections and hence could subtly influence the electoral
outcomes. Second, perhaps more important, reason is that the bureaucracy is the machinery or
the conduit that delivers the bribes to the political class, who in turn, create or sustain
bureaucracy by posting their key men at the suitable cross sections in public decision making.

Surely, modes of political corruption have increasingly become sophisticated – wholesale and
corporate – and, hence perhaps less “labour and cash intensive”, as the recent spate of scandals
have revealed. Yet bureaucracy is needed to avoid the electronic paper trail, and to cover the
tracks, and hence the perpetuation of the status quo. If these propositions are sound and
substantial, then the chances of the reorganization and modernization of public employment to
meet the felt needs of the citizens seems remote.

[Acknowledgements: I thank C Rammanohar Reddy for his comments and suggestions on an earlier draft of the
paper, Vikash Vaibhav for analyzing the NSS data and Sumit Mishra for generating tables based on the census data.]

R Nagaraj 12 Public Sector Employment

Table 1: Composition of public employment by levels of levels of government.
(in 00,000)
Year Central State Govt Local Govt Quasi Govt Total
Govt
1961 20.9 30.1 11.7 7.7 70.5
1971 27.7 (32.6) 41.5 (37.9) 18.8 (60.7) 19.3 (150.7) 107.3 (52.2)
1981 32.0 (15.5) 56.8 (36.9) 20.4 (8.5) 45.8 (137.3) 154.8 (44.3)
1991 34.1 (6.6) 71.1 (25.2) 23.1 (13.2) 62.7 (36.9) 190.6 (23.1)
2001 32.6 (-4.4) 74.3 (4.5) 22.6 (-2.2) 61.9 (-1.3) 191.4 (0.4)
2012 25.2 (-22.7) 71.8 (-3.4) 21.1 (-6.6) 58.0 (-6.3) 176.1 (-8.0)
Note: Figures in the brackets refer to percentage change
Table 2: Sectoral distribution of public employment.
(per cent)
Sector 1961 1971 1976 1986 1991 2000 2011
Agriculture & allied 2.6 2.6 2.7 3.0 2.9 2.7 2.8
Mining 1.8 1.7 5.4 5.5 5.2 4.8 6.3
Manufacturing 5.2 7.5 8.4 10.3 9.7 7.9 5.9
Elec. Gas and water 3.2 4.0 4.0 4.4 4.8 4.9 4.8
Construction 8.6 8.2 7.5 6.7 6.0 5.7 4.9
Trade 1.3 3.1 0.4 0.7 0.8 0.8 1.0
Transport, storage & 24.0 20.6 18.2 16.6 15.9 15.9 13.8
communication
Finance, insurance & 3.7 5.8 6.3 6.7 7.9
real estate
Community, social and 52.9 52.2 49.8 47.1 48.4 50.6 52.7
personal services
Total (in absolute no. in 190.6 191.4 172.7
100,000)

Table 3: Distribution of government employees by type

(Per cent)
Gazetted officers Non-gazetted officers
Total Class 1/ Class 2/ Class 2/ Class 3/ Class 4/
Group A Group B Group B Group C Group D
(1) (2) (3) (4) (5) (6) (7) (8)
1994 4023 5.1 1.8 3.2
2005-06 3109 5.1 2.5 2.6 1.9 66.6 26.3
2006-07 3109 5.1 2.5 2.4 2.0 66.8 26.3
2007-08 3119 5.0 2.5 2.4 2.0 67.1 26.0
2008-09 3107 4.9 2.6 2.6 2.5 67.4 24.9
2009-10 3068 5.2 2.7 2.6 3.1 91.6
2010-11 3082 5.2 2.7 2.5 3.8 91.0
2011-12 3085 5.3 2.9 2.5 3.7 91.0

Table 4: Alternative estimates of public employment

(in ,000)
Year NSS Official Difference
(%)
1999-00 16,148 19,415 (-) 20.2
2004-05 19,784 18,197 8.0
2009-10 22,563 17,795 21.3

R Nagaraj 13 Public Sector Employment

Table 5: State-wise distribution of public employment, for 1999-2000 and 2009-10
(per 000)
%
Principal States 1999-00 2009-10 change
AP 1,404 1,635 16.5
Assam 411 567 37.9
Bihar+Jharkhand 523 1,094 109.3
Gujarat 937 1,004 7.2
Haryana 378 591 56.2
HP 161 446 176.4
J&K 174 478 174.8
Karnataka 918 857 -6.6
Kerala 579 874 51.1
MP+Chathisgarh 1,251 1,942 55.2
Maharashtra 2,276 2,711 19.1
Orissa 625 892 42.8
Punjab 476 682 43.3
Rajasthan 820 1,546 88.4
TN 1,296 1,791 38.1
UP+Uttarkhand 1,640 2,655 61.9
WB 998 1,451 45.3
Sub-total 14,867 21,215 42.7
% of total 92.1 94.0

Table 6: Annual Growth in Public Employment

(Per cent)
Years Growth
rate (per
cent per
year)
1961-70 3.9
1971-80 3.5
1981-90 1.9
1991-00 0.1
2001-10 (-) 0.7
2000-10 3.4
Note: Compound growth rates from 1961 to 2012 computed using official data, for 2000-10, the NSS estimates are
used.

Table 7: Length of job contract

(Per cent)
Type of job contract 2004-05 2009-10
Less than a year 26.2 70.0
More than 3 years 37.7 59.9

R Nagaraj 14 Public Sector Employment

 Figure 1: Public Sector employment in 2012

Central govt
14%
Quasi govt
33%

Stat e govt
Local govt 41%
12%

Central govt Stat e govt Local govt Quasi govt

Figure 2: Total public sector employment

250

200
No. hundred thousand

150

100

50

0
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011

Year ending

Total public sector employment

Figure 3: Public Sector Employment per 1000 population

2500

2000

1500

1000

500

0
1961 1971 1981 1991 2001 2012
Year

Public employment /1000 pop

R Nagaraj 15 Public Sector Employment

 Figure 4: Employment Growth in various levels of Government

900
800
700
600
Index

500
400
300
200
100
0
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
1991
1994
1997
2000
2003
2006
2009
2012
Year

Central Govt State and Local Govt Quasi Govt

Figure 5: PSEs employment

25

20

15
Nos. in lakh

10

0
1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

Year ending

PSEs employment

R Nagaraj 16 Public Sector Employment

References

Bardhan, Pranab (1984): The Political Economy of Development in India, Oxford University
Press, New Delhi.

Eapen, Mridul (1980): “Trend in Public Sector Employment and Earnings”, Part I and II,
Social Scientist, Vol. 8, No. 8 and 9.

Desai, Ashok, and Ena Desai (1986): “India”, The Growing Sector, Studies in Public Sector
Employment in Asia, edited by Gus Edgren for ILO-ARTEP, New Delhi.

Kaary, Aart and Caroline van Rijckeghem (1995): Employment and Wages in Public Sector: A
Cross-Country Study, IMF Wp/95/70, Washington DC.

Nagaraj, R (1989): Sub-Contracting in Indian Manufacturing Industries: The Bangalore

Experience, unpublished doctoral thesis, Centre for Development Studies, Trivandrum,
Jawaharlal Nehru University, New Delhi

Patnaik, Prabhat and S K Rao (1978): “Explanation of Crisis in Mixed Underdeveloped

Economy”, Economic and Political Weekly, Annual Number, February.

Raj, K N (1973): “The Politics and Economics of “Intermediate Regimes”, Economic and
Political Weekly, July 7,

Rampell, Catherine (2012): ‘Big Government’ Isn’t So Big by Historical Standards. It’s Also
Shrinking”, The New York Times, July 27.

Srinivasan, T N (1985):”Neo-classical Political Economy, State and Economic Development”,

Asian Development Review, Vol. 3, No. 2.

R Nagaraj 17 Public Sector Employment