SPECIAL ARTICLES
Employment
and
Wages
in
ManufacturingIndustries
Tiends, Hypothesis and Evidence
R Nagaraj
Thtedecline in registeredmanufacturingemploymentthattookplace in the 80s is widelybelieved to reflect substitution
of capitalfor labour,as the wage rate reportedlyincreasedrapidlybecauseof growing rigiditiesin the labourmarket.How
valid isthisproposition?Withoutgettingintotheanalyticaldifferencesontherelationshipbetweenlabourmarketbehaviour
and econonmic
performance,this study examinesthtetrendsin wages, and (as a measureof labour marketdistortions)the
power of organised labour.Reassessing the postulatedrelationshipbetween earnings, capital intensityand employment,
the study suggests an alternative explanationfor the changes in employment.
The World Bank also offers a similar changes in the composition of industrial
THE decline in the growth of employment
in themanufacturingsector in the 80s andin explanation. Claiming that the real wage sector.4Second, the orthodoxview seems to
particularthe fall in the numberof workers rate increasedat 7.2 per cent per annumin ignore the effect of a reduction of wages,
in registered manufacturing are by now the first half of the 80s, the Bank argued: mediated by aggregate demand, on output
accepted facts. While the growth rates in "...employers respondedtto the increase in growth [Taylor 1988]. Ibird, evidence on
real net value added and real earnings per wage rate]by virtuallystoppingnew hiring the relationship between the labour market
worker (earnings, hereafter) in registered and retrenching existing workers to the institutions and economic performance is
manufacturingimproved,the absolutenum- extent possible". The study added: "The ambiguous, as discernible from the recent
ber of workers employed declined (Fig- estimates... point to a significant trade off European experience in general and the
ure 1). These tendencies arediscerniblein a between the higher real cost of labour and Scandinavianexperiencein particular,where
numberof majorindustrygroups(Table 1).' employment. This suggests that the faster corporatismseems to have positively influPerceiving them to be part of an economy- growth of real wages in -1980s indeed enced industrial relations and hence
wide phenomenon, the Planning Commis- did play an importantrole in slowing em- macroeconomic performance [Freeman
sion (1990) favouredserious changes in the ploymentcreation"[TheWorldBank 1989: 1988].
development strategy.2
109-10].
Given such ambiguity, both in theory as
Isher Ahluwalia, however, identified
These views areconsistent with the trade well in experience, a close look at the orthopolicy-induced rigidities in the labourmar- theoreticperspectiveon the principalprob- dox view could, therefore, be of value in
ket as the principalreason for the decline in lem withIndia'sindustrialpolicy. Analysing understanding the relationship between
employment, though she mentions other Indianexperience between 1960 and 1980, labourmarketbehaviourand outputperforpossible reasons like the growthof contract Robert E B Lucas attributeslow employ- mance in India's manufacturingsector. In
labourandspill over of employmentintothe ment growth to import substituting
doing so, we propose to examine (a) the
unregistered sector. Identifying consumer industrialisationin general and the policy- concept of wage rate/wages as used in the
non-durablegoods industries in registered inducedrigiditiesin non-agriculturallabour studies mentioned above (Section I),
manufacturingas accountingfor the bulkof markets in particular:"...[a] rising indus- (b) evidence on the reportedlygrowing rithedecline, she argued:"The sharpincrease trial wage has been a significant factor in gidities in the non-agriculturallabourmarin the capital-labourratio in the first half of the observed move toward more capital ket (Section II), and (c) the validity of the
the 80s was associated with a sharpincrease intensive techniques... Since the scope postulatedrelationshipbetweenwages, capiin the real wage rate during this period.... and coverage of these job security laws tal-labourratio and employment at a disagWhile the cause and effect can be debatedat has increased through time, it again gregated level (Section III). Before conlength, the data seem to suggest that the seems likely that they have contributed cluding, a provisional alternative explanaconsumernon-durablegoods sector experi- to the trend towards reliance on more tion for the observed decline in registered
enced the maximum increase in capital in- capital intensive techniques within many manufacturing employment in the 80s is
tensity as well as the maximum increase in manufacturing industries" [Lucas 1988: suggested (Section IV).
the real wage rate during this period" 189-90].
I
[Ahluwalia 1992: 82-84]. Associating the
These propositions-based on (simplistiincrease in the real wage rate with the re- cally) neoclassical percepts-form a seriDistinction between Wage Rate
portedlygrowing labourmarketrigiditiesous viewpoint, especially in the present
and Earnings
citing evidence on the increase in disciplin- context of policy reforms.3Theiranalytical
ary action against workers between 1960 and empirical validity can, however, be
Ahluwalia and the World Bank use Anand 1975 and the Indian worker's prefer- contestedon a numberof substantiveissues. nual Survey of Industries (ASI) data on
ence for job security than for higher earn- Firstly, these studies seem to have ignored 'wages to workers'to compute the wage rate.
ings-she further said: "Tlhis makes for the possible role of human capital vari- (or wages).5 As the ASI definition includes
inflexibility in hiringandfiringpossibilities ables-namely, skill, educationand experi- all payments made to workers-excepting
and rigidities in the labour market" ence-in increasingthewages as these could lay off payments not made by the employer
[Ahluwalia 1992: 841.
be considerable in a period of structural and the imputed value of benefits in kind-
Economic and Political Weekly January 22, 1994
Economic and Political Weekly, Vol. 29, No. 4 (Jan. 22, 1994), pp. 177-186
177
cent) (Table 2). At a disaggregated level,
growthin earningsper mandayrangesfrom
thatis, the basic wage plus dearness allow- (-)1.3 per cent in leather (NIC 29) to 3.7
ance, wherever applicable-but also all ad- per cent per annumin repairservices (NIC
ditional remuneration for workers' addi- 97), with no statistically valid trendfor (i)
tional effort.6 Conceptually, earnings per beverages,(ii) cottontextiles, and(iii) wool
workeris a function of the wage ratefor the and synthetic fibre textiles (NIC 22, 23
standard working day, remuneration for and 24).
additionalhours of work (and more shifts)
A comparisonof the trendsby use-based
and incentive income for more intensive industrialclassification for the same period
work that is linked to output. In the decade suggeststhatin consumernon-durablegoods
beginning 1979-80, while the employment industries, accounting for nearly half the
growth turned negative, total person days registered manufacturing employment,
(or mandays) worked in registered manu- while earningsperworkerincreasedat 3 pet
facturing-and hence mandaysper worker- cent, earnings per mandayrose by a mere
recorded a positive trend growth rate; sug- 0.5 per cent per annum,which is less than
gesting that the observed increase in earn- one-thirdthe averagefor registeredmanuings per workercould, at least partly,repre- facturing(1.6 percent).Theobserveddiversent his (or her) compensation for greater gence betweenthetwo measuresof wages in
it evidently refers to workers' total earnings, which covers not only the wage rate-
effort and may not necessarily
imply an
increase in the wage rate, as has been argued
(Figure 2).7
Admittedly, since the definition of workers in ASI includes contract (or casual)
workers,it is arguablethat all or most of the
additionalmandays worked may well have
been done by contractworkersat the market
clearing wage rate, with little additional
effort by the supposedly unionised permanent workers.This propositionis examined
using the labourstatisticsdatafor thecensus
sector of ASI.8The shareof casual workers
in the total employment more thandoubled
from 4.6 per cent in 1980-81 to 10 per cent
in 1986-87. However, as theirproportionin
the totalmandaysworkedin 1986-87 is only
8.9 per cent, contractworkers' contribution
to mandays worked does not seem to be
disproportionateto their share in employment. An estimate-assuming that the proportionof contractworkersin totalemployment in the census sector is the same for
ASI as a whole and that the share of
mandays workedbycasualworkersin 198081 is same as in 1986-87-suggests that
between 1980-81 and 1986-87 while the
mandays per worker for all workers increased by 16 per cent, the same for casual
and permanentworkers increased by 15.8
per cent and 16.7 per cent respectively.
Thereis, thus,little basis to arguethatcasual
workers account for all (or most) of the
observed increasein mandaysworkedin the
80s, though theirsharein totalemployment
has gone up.
We examine the trends in earnings per
worker and earnings per manday, although
the latteras a proxyfor thewage rateignores
additional remuneration due to plausible
intensification of the working day. While
earnings per workerin registeredmanufacturingincreasedat 3.2 per cent in the decade
beginning 1979-80, earnings per manday
increased at only 1.6 per cent per annum,
whiclhis less thanthecorrespondingrealper
capita GDP (hereafter per capita income)
growthrate during the same period (2.7 per
178
these industriesseems to be consistent with
the decline in employment at the rate of 1.6
per cent and the growth in mandays per
worker2.3 per cent per annum;corresponding figures for registeredmanufacturingare
(-)0.5 per cent and 1.5 per cent per annum
respectively.
Ahluwalia, as notedearlier,observedthat
inversemovementbetweenemploymentand
wage ratein these industrieswas associated
with an above average increase in capitallabourratio. The foregoing evidence, however, suggests thatwhile arningsperworker
undoubtedly went up in the 80s, it was
mainly on account of an above average
increase in the number of days worked per
worker.The rise in earnings per mandayin
consumer non-durablegoods industries, as
shown above, is considerablylower (0.5 per
TABLE1: TRENDS
IN EMPLoYMIENT
ANDEARNINGS
IN REGISTERED
1973-74 TO 1988-89
MANUFACrURING,
IndustryGroup
20-21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
97
Total
EmploymentGrowth in
1973-74to
1979-80
1980-81to
1988-89
0.3
4.5
(-)1.6
4.4
(-)1.6
3.8
(-)0.2
1.0
5.2
4.2
3.5
3.5
2.8
0.9
1.6
2.4
2.4
1.3
1.4
(-)3.8
(-)0.3
(-)3.6
2.9
(-)3.4
3.7
(-)I.5
(-)0.8
5.1
2.6
2.0
1.9
1.1
0.4
(-)0.1
1.6
(-)1.6
2.4
1.1
(-)0.6
Earnings Per Worker
1973-74to
1979-80
5.7
3.1
2.1
3.0
3.0
1.8
3.1
3.1
1.7
2.8
3.5
2.4
2.5
3.7
3.5
3.5
3.7
3.4
3.3
3.2
Share of
Employment
1980-81to
1988-89
8.9
5.1
2.0
1.2
2.9
1.2
3.7
2.7
(-)0.3
3.9
3.2
2.2
2.2
3.3
3.2
3.1
-3.4
3.7
4.3
3.6
13.2
4.6
20.0
3.2
5.7
1.3
1.4
4.5
0.9
2.0
5.4
5.5
8.1
3.2
5.5
4.2
7.1
1.2
2.8
100.0
Source: AS! Summary Resultsfor the Factory Sector, various issues.
TABLE 2: GROWTHRATES IN REGISTEREDMANUFACTURING,1979-80 TO
Employment
of Workers
Intermediategoods
Capitalgoods
Consumerdurablegoods
Consumernon-durable
goods
Registeredmanufacturing
0.5 *
(0.336)
(-)0.02
(0.0)
1.6 *
(0.791)
(-) 1.6 *
(0.706)
(-) 0.5 *
(0.301)
1988-89
MandaysPer
Worker
EarningsPer
Worker
1.0 *
(0.870)
0.3
(0.134)
0.7
(0.181)
2.3 *
(0.750)
l.5 *
(0.663)
2.6 *
(0.843)
3.0 *
(0.900)
2.7 *
(0.776)
3.0 *
(0.829)
3.2 *
(0.579)
EarningsPer
Manday
1.6 *
(0.751)
2.7 *
(0.965)
2.0 *
(0.588)
0.5 *
(0.653)
1.6 *
(0.800)
Notes: (1) * Indicatesstatisticalsignificance of the estimated coefficient at 90 per cent confidence
level and above.
(2) Employment growth rate for registeredmanufacturingis for the nino-yearperiod since
1980-81 as the same forthe period since 1979-80 is statistically significant at 80 percent
confidence level.
Source:ASI Suaninaty Resultsof FactorySector, various issuies.
Economic and Political Weekly January 22, 1994
cent per annum)than the average for registeredmanufacturing(1.6percentperannum).
If these findingsarevalid, thenthey notonly
question Ahliuwalia'sevidence of increase
in the wage rate but also perhapsthe postulated associations between the wage rate,
capital-labourratio and the decline in employment.
Though these findings seem interesting,
they remain partial since earnings per
mandayis not the same as wage rate due to
a possible intensification of the working
day. We examine Occupational WageSurveys9 (OWS) which take into account the
distinctionbetweenthe wage rateperworker
perday (hereafterwage rate)'0andearnings
per worker per day (hereafter earnings)."
As these surveys have been conductedfour
times since 1958-59, using consistent definitions and methodology covering almost
the entire registered manufacturingsector,
they could be useful to discernthe long-term
movements in wages, disaggregatedby induistries.Ourpreliminaryattempt,however,
ignores the variation in wage rate and
earnings on account of education, skill
and experience, thoughconsiderablescope
seems to exist to do a more detailed exercise. Moreover, the figures reported here
are simple averages of the minimum and
maximum wage rate/earnings reported in
ings increased much faster than the wage
rate.13While the increasein the wage rateis
II
Strength of Organised
Working Class
in the range of (-)O.1 per cent (iron and
steel) to 5 per cent (prime movers) per
annum, earnings has went up between 1.3
The increase in the wage rate that is
per cent (garments)and nearly7.9 per cent presumed to have taken place is seen as an
(electrical machinery)per annum.14If it is evidence of the growing rigidities in the
conceded that the definition of the wage labour market-namely, minimum wage
rate (by Ahluwalia) and wages (by the
World Bank) represent earnings per
TABLE 4: CORRELATIONCOEFICIENTS BEWEEN
worker, then these diverging trends seem
EARNINGS PER
GROWTHRATES oF EMPLOYMENT,
to suggest that much of the reported inWORKER AND 'CAPITAL-LABOURRATO
crease in the wage rate or wages noted by
Simple Correlation
them, represents additional remuneration
Coefficient between
for workers'extraeffort-which is likely to
GrowthRates of
get reflected in higher labourproductivity
CapitalEarnings/
(Figure 10)-and perhapsnot a substantial
Labour
Workerand
increase in the wage rate for the standard
Ratio and
CapitalEmployLabour
workingday.
Ratio
ment
To recapitulate:wage rate according to
the OWS and earnings per manday as per 1 Trendgrowth rate during
ASI data have not gone up disproportion(-) 0.362 *
1973-74 and 1986-87 0.142
ately as has been suggested by the studies 2 (i) Percentagechange
quoted earlier. The wage rate over a long
0.479 *
(-) 0.215
in sub-period I
(ii) Percentagechange
period in half the numberof industriesfor
0.014
0.001
in sub-period II
which data are available (OWS) and earnings per mandayfor registeredmanufactur- 3 (i) Average of annual
growth rates in
ing in general and consumer non-durable
(-) 0.108
0.193
sub-period I
goods industries in particularin the 80s
(ii) Average of annual
of
(ASI) has not kept pace with growth per
growth rates in
capitaincome.However,earningsper-worker
0.348 *
0.436 *
sub-period II
in
increased faster than per capita income
OWS.
1973-74
and
1979-80; and
Fourteen industries for which data are registeredmanufacturingas well as in con- Notes: (i) Sub-periodI:
sub-periodII: 1980-81 and 1986-87.
available for all the rounds since 1958-59 sumer non-durablegoods industries since
(ii) Correlationcoefficients are computed
accouA.t for about 40 per cent of registered workers have evidently worked a larger
for growth rates across 42 ,hree-digit
more
manufacturingemployment. In seven in- numberofdays.Although,analytically,
industry groups.
dustries-accounting for about a quarterof intensive work could also accountfor a part
(iii) * Refers to statisticallysignificant at
per
in
earnings
growth
in
of
the
discernible
manufacturing
employment
registered
90 per cent level and above.
1980-81-the compound growth of wage worker,we have no basis to test the propo- Source:ASI SummaryResults of Factory Sector,
various issues.
rate between 1958-59 and 1986-87 is lower sitioln.'
than the correspondingper capita income
growthof 1.3percentper annum(Table 3)12
TABLE 3: TRENDS IN WAGE RATE AND EARNINGS IN SELECrED MANUFACrURINGINDUSTRIES
In one industry,namely, bicycle manufacturing,the wagerate has, in fact, declined by Industry
CompoundGrowthRate
7 percent in 1986-87 comparedto 1958-59.
Wage Rate between I
Between III and IV OWS
and IV RoundOWS
Wage Rate
Earnings
In theremainingindustries,wage rateshave
grown in the range of 1.3 per cent (ship Cottontextiles
1.0
0.5
4.2
building)to 2.9 per cent per annum(electri- Jutetextiles
1.8
2.0
2.4
cal machinery).Thetentativeinferencefrom Woollen textiles
0.7
1.6
4.3
theselong-termtrendsis thatwhile thewage Silk textiles
0.2
1.1
2.8
rates across industries under consideration Garments
0.4
1.3
1.3
1.5
1.5
7.4
have not remained constant, increases in Ship building
3.4
2.0
3.8
themdo not appearto be disproportionately Locomotive
0.7
2.6
6.4
largecomparedto per capitaincome growth Motorvehicle
-0.3
0.2
2.3
in the same periodl.This result, consistent Bicycle
1.8
Aircraft
0.3
5.0
with the foregoing evidence on the trendsin Ironand steel
0.3
2.7
earnings per manday in the 80s for regis- Castingand forging
2.7
5.9
tered manufacturing based on ASI data, Agriculturalmachinery
0.9
1.6
4.0
seems to provide a basis to question Electricalmachinery
2.9
4.5
7.9
Ahluwalia's and the World Bank's finding Machinetools
1.9
2.0
6.7
5.0
7.6
of a disproportionateincrease in the wage Primemovers
Textile machinery
0.2
0.9
6.7
rate.
It is, however, interesting to discover Notes: (i) I Round: 1958-59, ll Round: 1%3-66, III Round 1974-75, IV Round 1986-88.
from the same table that between the Tlird
(ii) Consumerprice indices for industrialworkers for all India are used. In order to reduce
andthe FourthRoundof the OWS (between
yearly fluctuations,wehaveused three-yeara*verages aroundthe yearof surveyto
1975-76 and 1986-87) in almost all the 17
deflate the money wage rate and earnings.
industriesfor which dataareavailable,earnl- Source: OccutpatiotialWageSurvey,variousreports.
Economic and Political Weekly January 22, 1994
179
FIGURE 1: TRENDS IN EMPtoYMrNT, EARNINGSAND VALUE ADDED IN
REGISTEREDMANUFACrURING,1973-74 TO 1988-89
ANDMANDAYS
WORKED
IN
FIGURE2: NUMBER
OFWORKERS
EMPLOYED
REGISTERED
1979-80 TO 1988-89
MANU,FACTURING,
Index of log values
1 Index of log values
115
100.9
100.7
100.5
/
100.3 -
105
100.1
100
99.9
99.7
95
74
77
80
83
86
Fiscal year ending
- Number of workers
89
X
99.5
80
I
82
- Earnings/worker
e
'
I
84
86
Fiscal year ending
I
I
88
... Net value added
- Numberof workers
Source: ASI SummaryResults
of mandays
Souirce:ASI SummaryResults
legislation(growing)strengthof tradeunions disputes in registeredmanufacturing,when
and increasingly stringent job security normalisedfor mandaysworked,no statislaws-as reflected in the powerof organised tically valid trendis discernibleduringthe
labour to appropriatea share of outputdis- two decades beginning 1970-71. There are
proportionateto their contribution, at the only fluctuationsin therangeof 0a7percent
expense of additionalemployment genera- to about 2.3 per cent of the number of
tion and with a socially undesirablerise in mandays worked (Figure 3). Union dencapital intensity. Robert E B Lucas, for sity-the proportionof workers unionised
instance, showing an increase in the abso- in registered manufacturing-fell from
lute number of mandays lost in industrial around45 per cent in the late 70s to around
30 per cent towards the end of the 80s
disputes in the organised sector as a wlole,
(Figure4).17 The proportionof strikesin,the
argued: "... it is clear that the comparative
power of unionsin pressingfor wage settle- total number of mandays lost in the disments has grown substantiallyandthe inci- putes, ignoring the sharp dip during the
dence of strikes has consequently risen as 'emergency', has also steadily declined to
reflected in the numberof mandays lost in less than half in the latter half of the 80s.
ThIisimplies thatin therecentyearsa greater
strikes and lock-outs ..." [Lucas 1988: 189];
he reiteratesthis view in amore recent study share of industrial disputes has been on
accountof lock-outs by employers and not
as well [Fallon and Lucas 1,993].
Lucas' measureseems incorrectfor three because of strikes (Figure 5).18 Finally, dereasons. One, the absolute number of clining shareof mandayslost due to strikes
mandays lost in industrialdisputeshas little in the total is discernible in a period when
meaning when the number of mandays the total number of workers involved in
worked has also grown. Two, as industrial industrialdisputeshas droppedfrom 38 per
disputes could be due to either strikes or cent in 1973-74 to less than 10 per cent in
1988-89 (Figure6).19
lock-outs-which he evidently recognisesThe decline in the strengthof unionised
an increase in the total numberof mandays
lost does not necessarily imply an increase labour is accompanied by changes in the
in the proportionof strikes. Three, his data structureof employmentas well.20Theshare
refer to the organisedsector of the economy of registered sector in total manufacturing
and not to the registered manufacturing employmentfell fromaround30 per cent in
industrieswhich he has examined;the latter 1977-78 to about23 per cent a decade later
(Figure 7). Moreover, within the factory
is a subset of the former.
Following Lucas in examining the trends sector, employment growth is positive in
in union power as a proxy for thd alleged factories employing less than 1,000 workrigidities, our evidence questions hlis con- ers; and a statistically significant negative
tention.'6In his measure of mandayslost in correlation exists-withl Rt2of 0.66-be-
180
-Number
tween the size class of factories andemployment growthbetween 1979-80 and 1988-89
(Figure 8).21 Since the average wage rate is
lower in smaller sized factories and the
likelihood of power of unions is also lower,
a priorione could arguethatthese structural
changes would have furthercontributedtowardsreducing policy-induced rigidities in
the labourmarkets and exerting downward
pressure on wage rate and earnings within
the registered manufacturingsector.
Arguably, these changes reflect the response of firms to the labourmarketrigidities. While there could be some validity in
such a view, it would perhapsbe incorrectto
attribute the structuralchanges mainly or
entirely to the behaviourof the labourmarket, as there could be a number of other
reasons.These changes seem to formpartof
the long-term tendencies in the size structure of manufacturing to correct for the
historicalbias in favourof large sized plants
in India [Nagaraj 1985]. This could also
reflect the outcome of various policies to
promotesmall-scale manufacturingandinstitutional changes in financing industrial
development.Moreover,diffusion of skills,
entrepreneurship
andtechnologiescouldalso
have favouredmodern small- andmediumscale firms.
The foregoing evidence unambiguously
shows a decline in' the strength of the
organised labour. Hence, it seems reasonable to suggest that, in a periodof declining
bargainingpower of organisedworkersand
structuralchanges in employment within
registered sector towards smaller sized es-
Economic and Political Weekly
January 22, 1994
3: NUMBEROFMANDAYSLosT IN INDUSTRY
FIGURE
AS PERCENTOF
DISPUmES
OF MANDAYSWORKED,1973-74 TO 1988-89
NUMBER
IN
UNIONISED
OFWORKERS
4: UNIONDENSrry:PERCENTAGE
FIGURE
1973-74 TO 1987-88
MANTFACRTRING,
REGISTERED
Per Cent
Per Cent
45 F
2
40
/
1.5
\
35
I
3Q0.5
74
07
80
83
86
89
25,1
74
77
Source: Indian Labour Yearbook,various issues
The 'idence adducedby Ahluwalia, the
Wor
and Lucas on the postulated re1 tionshipbetweenthewagerate,capi-
Economic and Political Weekly
86
- Per cent of workers
- Number of mandays lost
HI
Testing the Hypothesis
83
Fiscal year ending
Fiscal year ending
tablishments, unionised labour is unlikely
to have securedXtdisproportionateincrease
in the wage rate, especially at a time when
the share of registered manufacturingemployment in the total is itself declining.
Whileregisteredmanufacturingaccounted
for about30 percent of total manufacturing
employmentin 1977-78, for consumernondurablegoods industries it was only about
24 per cent (Figure 7). Moreover, the average size of a factory in these industries
declined from around70 workers in 197374 to less than 55 workers per factory in
1988-89, as in registered manufacturing
(Figure 9). Since employment in consumer
non-durable goods industries is predominantly in the unregisteredsector and since
within the registered sector their average
factory size has declined considerably, it is
quite unlikely, contraryto what Ahluwalia
has suggested, that organised workers in
these industries could have succeeded in
secuing notable increases in the wage rate.
This is not to deny thatorganised workers
do not command any bargainingpower in
the labourmarket.They perhapsdo in large
firms, in the public sector and in some
locations. But at an aggregated level, the
foregoing evidence seems to question the
proposition of growing rigidities in nonagriculturallabourmarkets.
80
Source: Indian Labour Yearbook,various issues
tal-labour ratio and employment seems
to have an analytical shortcoming. They
have shown trends in the wages/wage
rate and the rigidities in the labour
market without considering the changes
in the cost of capital. Analytically, what
matters in the choice of technique is not
the wage rate per se but the wage-rental
ratio. Perhaps these studies implicitly assume the cost of capital to have remained
constant. Admittedly, its measurementis
fraught with numerous conceptual and
empirical problems.Thesimpleandwidely
used measure is real interest rates, which
increased steeply in the 80s, even ignoring the spikes in the series (Figure 11).22
Therefore there is a need to qualify the
findings of Ahluwalia and others in this
respect.
This section re-examines the postulated
associations between increase in earnings
per worker, increase in capital intensity
(capital-labourratio), and a decline in employment'growthin registeredmanufacturing. The cost of capital is not included to
minimisecomputationaldifficulties.Simple
correlationsacross42 three-digitindustries
for the period 1973-74 and 1986-87 are
calculated between the growth rates of:
(i) earnings per worker and capital intensity; and (ii) capital intensity and employment.23Three variants of associations are
explored: (i) growthover the entire period
of 14 years; (ii) increase in sub-period I
(1973-74 and 1979-80) andin sub-periodII
(1980-81 and 1986-87); and (iii) average
of the annualgrowth rates in sub-periodI
and averageof annualgrowtlhrates in sub-
January 22, 1994
period II, thus obtaining five sets of correlatiorls.24
Evidence suggests that although the results are partially consistent with the hypothesis, in none of the variants does one
find consistent and statistically significant
results with the expected sign for the coefficients (Table 4). For instance, in variantIthat is over the entire period-while the
correlationbetweenthegrowthtates of capital-labourratio and employment is statistically significant with the expected negative
sign, the correlation between the growth
rates of earningsper worker and the capital
labourratiois not statisticallydifferentfrom
zero. Similarly, in variantIll-that is, average of annualgrowthratesin sub-periodIIwhile the correlation coefficient between
the growthrates of earnings per workerand
the capital-labourratio is statistically significantwith the expected negative sign, the
correlationbetweenthe growthratesof capital-labourratio and employment is positive
andstatistically significant, which is ir<consistent with the hypothesis.
We have also computed the above-mentioned correlations separately for the time
series data of 42 industries. To minimise
autocorrelationproblems the first differences in each series are correlated. The
findings shows that in only nine out of 42
industriesthe postulatedrelationshipis statistically valid with theexpected sign for the
estimatedcoefficients. These results, therefore, seem to suggest that the relationship
between earnings per worker,capital intensity and employment is more complex'than
postulated by the hypothesis. Moreover, if
181
MANDAYS
5: SHAREOFSTRIIESINTOTALNUMBER
FIGURE
LOSTININDUSTRES
DIsPUTEs,1970-71 TO 1988-89
Per Cent
ININDUSTRIES
INVOLVED
OFWORKERS
FIGURE6: PERCENTAGE
DIOSnu,
1973-74To 1988-89
Per Cent
40
35
80
30~
/
251
t
40!-
20.5.
71
74
77
80
83
86
Fiscal year ending
89
'74
77
80
83
86
89
Fiscal year ending
- Share of strikes
-
Per cent of workers
Sources: IrndianLabour Yearbook,ASI Summary
-
Source: Indian Labour Yearbook,various issues
Results, various issues.
value added across 19 two-digit industry
groups, for the period 1980-81 to 1988-89,
is (-)0.51 and statistically significant. The
relatively fastergrowthrateof these industries, especially chemical and petroleumbased industriesperhapsreflects conscious
IV
efforts to importsubstitutetheseproductsin
Towards an Alternative
response to the oil pricehikes in the 70s and
Explanation
the exploitation of newly discovered minThe causal connection between the wage eral oil deposits to reduceadverleeffects of
rate, capital intensity and employment is, externaleconomic shocks; andperhapsless
admittedly, an equilibriumcondition. In a driven by the evolution of relative prices.
Overhangof Einployment:The upturnin
labour surplus dual economy, evolved under certain institutionalconditions andwit- outputgrowth(andthecorrespondingneganessing an upturn in growth after a pro- tive growth rate in employment) in manulonged period of 'relative stagnation', the facturingin the 80s, as is well known, was
observeddecline in employmentcould have witnessed after a prolongedperiod of relamultiple causes, not necessarilyoriginating tive stagnation since the mid-60s when
in the reportedincreasein wage rate."5In an employmentgrowthwas sustainedat over 3
attemptat exploring an alternativeexplana- percent per annumdespite adecelerationin
tion, without offering an analytical model, output growth, resulting in a narrowingof
we propose to identify a set of factors which the gap between the growthof earningsper
workerand labourproductivity.More specould have influenced employment.
Changing Compositionof Output:Partof cifically, duringthe seven-yearperiodsince
the explanation for the decline in the 80s 1968-69 while net value addedgrew at 2.6
seems to lie in the changingcomposition of percentperannum,employmentkeptgrowoutput in favour of\less labour intensive ing at 3.9 percent per annum;hence, labour
industries. Fo, instatce, industry groups productivitywas increasingat a lower rate
food products (20-21), rubber,plastic, pe- than earnings per worker, resulting in an
troleum and coal products (30), chemical unsustainable situation in the long run
and chemical products (31), non-metallic (Figure 10).
mineral products (32), and electrical maIntuitively, one could argue that when
chinery(36) witnessedaboveaveragegrowth demand picked up in the 80s firms would
in value added but have a below average have first used theirlexistingstock of labour
shareof wages in value added.The correla- (as well as capital)intensivelybeforedecidtion between (i) growth rate in real gross ing to employadditionalworkers.Thiscould
value added and share of wages in gross perhapspartly account for the lack of emwe accept thatreal interestrate has gone up
duringthe period it would furtlherdilute the
strength of the observed associations between the three variables.
182
ployment growth in registeredmanufacturing, despite an upturnin output growth.
Changing Market Conditions: Analytically, changes in the product and financial
markets are likely to affect the demand for
labour. Studying the interdependence between market structure and technology,
Desai (1985) suggested an increase in competition duringthe period 1963-64 to 197879.26 Considering that liberalisation was
initiated in the 80s one could hypothesise
that industrialmarkets have become more
competitive exerting firms to reduce costs.
We, however, do not have any evidence to
supportthe proposition.
In responseto the public sector's growing
borrowingrequirementon the one handand
the policy directive of priority sector lending on the other, the cost of credit for the
private (non-agricultural)sector seems to
have increased significantly in the 80s.
F,i,gure11 shows a steady increasein interest
cost as a proportion of total income in
medium andlargepublic limited companies
and the trendseems to be positively associatedwith an increase in the real interestrate
in the 80s.27
A priori,one would expect an increase in
the cost of capital in a period of increasing
competitionto resultin substitutionof labour
forcapital andhence an increase in employment. However,given thedifferentialwagerentalratiosin theregisteredandunregistered
sectors, firms would perhaps have found it
profitable to shift the production to
unregisteredsector adversely affecting employment in the former.
Economic and Political Weekly
January 22, 1994
FIGURE7: SHARE OF REGISTEREDMANUFACTURINGIN TOTAL MANUFACTURING
EMPLOYMENT,
1J77-78 TO 1987-88
FIGURE
8: ENpLOYImENT
INREGISTERED
GROWTH
MANUFACTURING
BY
EMPLoYMENT
SIzECLAS9,1979-80 TO1988-89
Per Cent
35
GrowthRate in Per Cent
4
30
_
_ _
25
-
-
_
_
20
-2__
15
_
-
-
-
-4
10
2
1
1977-78
1983
5
7
6
EmploymentSize Class
1987-88.
-
Year
Registeredmanufacturing M Consumernon-durables
Sources:NSS, ASI various issues.
FIGURE9: FACTORYSIZE I*REGISTERED MANUFACTURING
INDUSTRIES,
1973-74 TO 1988-89
8
Employmentgrowth
Employmentsize-class by numberof workers:1:0-49; 2:50-99; 3:100-199;
4:200-499; 5:500-999; 6:1000-1999: 7:2000-4999; 8:5000+
FIGURE 10: EARNINGSAND LABOUJRPRODUCTIVITYIN REGISTERED
MANUFACTURING,1960-61 TO 1988-89
No of workersper factory
Index Numbers
200
/
75
70
150
65
100
60
50
0_
5SJ.
'
74
76
78
-
8.0
82
84
86
i
88
Fiscal year ending
Manufacturing
-
64
67
70
73_, 76
79
82
85
Fiscal year ending
Consumernon-durable
Source: ASI, Summary. Results of Factory Sector, various issues.
Modalities of Adjtstinent: Based on field
experience [Nagaraj 1989], we would
hypothesisethe following plausible adjustment process. As demand picked up for
manufacturingoutput, firms (i) used their
workforce intensively by increasing incentives for workers at the shop floor level,
linking wage agreements to productivity
norms,(ii) reorganisedshop floor andworkers' job-content, (iii) farmedout manufac-
61
Real earnings/worker
Real VA/worker
Source:ASI SummaryResults
turingof relativelysimplercomponentsand justment behaviourof publip sector firmsis
products to secure economies of specia- likely to be more-constrained by policy.
lisation andused contractlabourfor labour- These modalitiesof rationalisationof labour
intensiveservices, and(iv) increasinglyem- use in registered manufacturingwere feaployed part-timeworkers especially small sible in the 80s as a result of diffusion of
firms. While capital investment and auto- skills, spread of infrastructureand various
mationformelementsof restructuring,these policies to promote small firms.
are more likely only after the other options
The foregoing explanation, however,
are fully explored, unless dictatedby tech- would imply an increase in the growth rate
nological considerationi.However, the ad- of employment in the unregisteredsector.
Economic and Political Weekly January 22, 1994
183
suggest the limitation of the orthodox view
in explaining the labour market behaviour
and hence question the policy implications
that follow from such a view.28
FIGURE 1 1: SHARE OF INTERESTCOST IN COST OF PRODUCTIONAND REAL INTERESTRATE,
1960-61 TO 1988-89
Per Cent
Per Cent
55, --
/
-.
4.5-1
[Followingthe usual disclaimers,I gratefully
acknowledgethe valuablecommentsreceived
fromSubirGokam,AshimaGoyal,ShikhaJha,
Sunil Mani, ManojPanda,KiritParikh,V K
Ramachandran,K V Ramaswamy,C Ram
ManoharReddy,M
HSuryanarayana,
Bagaram
Tulpule,andthe two refereeswithinthe institutefortheircommentson anearlierdraftof the
paper.SureshD Tendulkar's'unsolicited'response to my draftpaperwas particularly
helpful, for which I am deeplyindebtedto him.]
120
01
3.5.K
Notes
t4
1
61
64
-
67'
13 7q 71
Fiscal year ending
70
Interest cost
+-
110
I J
82,
85
88
Realinterestrates
Sources: RBIBulletin,National AccountsStatistics, various issues.
tor and within registeredmanufacturingtowards smaller sized factories. These longterm tendencies would have furtherundermined the power of the organised working
class to induce flexible use of labour. Finally, not muchof a systematicrelationship
could be discerned in the postulated relationship between earnings per worker, the
capital-labourratio and employment at a
disaggregatedlevel.
Exploring an alternativeexplanation, it
was arguedthat(i) mainly in response to the
macro-economic compulsions of the 70s
CONCLUSION
and the discovery of appropriatenatural
A reportedlysharprise in the wage ratein resources,thecompositionofoutputchanged
the 80s in registeredmanufacturing-due to in favourof less labourintensive industries,
increasingpolicy-induceddistortionsin the (ii) there was a overhang of employment
labour market-is widely hypotbesised to that had built up during the prolonged
have led firms to s ubstitute capi ta 1for labour,
period of relative stagnation in the 70s,
resultingin tbeobserved decline in employ- and, (iii) increasing competition in the
ment. Consciously avoiding getting into the product market due to domestic liberalitheoreticaldebateon the causal relationship sation and increase in the cost of borrowbetween the labour market and economic ed funds could account for the decline in
growth, this study provided evidence to employment in registered manufacturing
suggest thatthe wage rate (using OWS) and in the 80s. It was further hypothesised
earnings per manday (using ASI) did not that there was (a) a restriction on fresh
increase disproportionatelyas has been ar- employment in large factories so as to use
gued. However, earnings per worker in- the existing work force more intensively,
creased at a faster rate than per capita in- (b) shop-floor reorganisationand changes
come growth mainly due to an increase in in workers'job-content, (c) contractingout
the number of mandays per worker. Our of (mostly unskilled) labour-intensiveserfindings also suggest a distinctdecline in the vices and farming out of production, and
power of organised labour,used as a proxy (iv) an increasinguse of part-timeworkers
for the reportedrigidities in the labourmar- in small firms permitting flexible use of
ket. Moreover, structuralchanges are dis- labour.
cernible in enmploymentin manufacturing
Finally, if evidence and argumentsoutindustriesin favour of the unregisteredsec- lined above have any validity, they seem to
Since employment growth in unregistered
manufacturinghas also declined [Planning
Commission 1990], one is inclined to furtherhypothesise thatin the non-factorysector there is a muclhmore intensive use of
labour for longer hours and in a flexible
manner. Moreover, there appears to be an
increasing trendQfworkerstakingup more
thanonejob, especially amongskilled workers, which could further restrict employment generation.
184
1 Unless otherwisestated,growthratesreportedinthisstudyarecomputedusingloglineartrendequation.
2 Visaria and Minhas (1990) question the
PlanningCommission'sfindings since in
threeoutof the fouryearsin whichthe NSS
data were collected happen to be 'bad'
rainfallyearsandhencethe employmentin
these years could have been affectedadversely.However,sucha problemis likely
to be unimportant
formanufacturing
sector
since:(i)outputwitnessedsustainedgrowth
with very little yearlyfluctuation,and (ii)
bad agriculturalharvestis foundto affect
manufacturing
outputwith one yearlag.
3 Recentlypublishedofficial'DiscussionPaper' on economicreformsechoes the foregoing views: "... we must reviewand reform
thecurrentlegislationsforemploymentand
industrialdisputesto ensurethatexcessive
rigiditiesareremovedandlong-termgrowth
facilitated.The operationof the current
IndustrialDisputesAct (in particular,the
requirementfor priorapprovalby governmentforclosureof sick unitsandretrenchmentof labour)needs to be reviewedas it
affectsincentiveforinvestmentandaswell
asincreasedemployment....Rigid
nrleslimiting the flexibilitywith which labourcan
be hiredand retrenchedhave the effect of
pushingentrepreneursiito more capitalintensivetechnologyand reducethe number of workersthay have to deal with"
[Governmentof India 1993: 34].
4 Itis perhapspertinentto cite SouthKorean
examplein thiscontext.Despitestrictlaws
prohibitingtradeunionactivity,realwage
rate in that countryincreased at a trend
growthrateof S percent perannumovera
long period.Amsden(1990) attributesit to
intensivelearning,educationand skill of
the Koreanworkers.
5 "Wages are defined to include all remunerationcapableofbeingexpressedinmonetarytermsand also payablemoreor less
regularlyin each pay periodto workersas
compensationfor work done duringthe
accountingyear.Itincludes(a)directwages
and salary(i e, basic wages, salaries,paymentof overtime,dearnesscompensatory,
houserentandotherallowances),(b)remu-
Economic and Political Weekly
January 22, 1994
nerationfor periodnot worked,(c) bonus
andex-gratiapaymentpaidbothat regular
andless frequentintervals..."(ASI,198889: Suimnary Results of Factory Sector,
p78).
6 For conceptual clarity, MauriceDobb's
distinctionseems instructivein this context: "Whenwe speakof a wage rate, we
usually refer to the amount paid to the
workerperhourpernormalworkingdayor
forperformingacertainjob.Inotherwords,
we are speaking about what the worker
receivesin returnforagivenoutputof work
on his part....Whenwe talkaboutthe standardof life of theworkerandhis family,we
areconcernedwiththetotalearningsof the
family-unit.... We cannot juidge what is
happening to theformer [wage rate] solely
from the figures about the latter [earn-
ings];andforanumberof reasonsthesetwo
quantitiesmay changequite defferently"
[Dobb 1959: 27; emphasisadded].
7 Due to possible changesin the definition
andmethodology,dataon mandayspriorto
1979-80arenotcomparablewiththesubsequentyears.
8 Censussectorof ASIincludesall factories
employing50 or moreworkerswithpower
and 100 or moreworkerswithoutpower.
9 LabourBureau,Governmentof India in
orderto obtain "accurateand up-to-date
wage statistics" [Indian Labour YearBook
1992:45] has pcriodicallyconductedOc-
cuspationalWage Surveys. The four rounds
have been conductedso far in 1958-59,
1963-65,1974-79and1985-88.Thecoveragehassteadilyincreasedoverthesuccessive
rounds;from 44 plantation,mining and
industriesin the firstround
manufacturing
to 53 in the fourtb.These surveysinclude
firmsintheorganisedsectoroftheseindustriescoveredunderthe relevantacts.Completeresultsof thefourthroundareawaited.
10 ToquoteOWS:"Wagerateshouldinclude
basic wage rate,cost of living allowance
and other guaranteedand regularlypaid
allowances,butexclusiveof overtimepayment,bonuses,gratuities,familyallowance
andothersocialsecuritypaymentsmadeby
employers."
purposeof thissurvey,wagerate
'.!FDorthe
hasbeen definedas the sumof basic wage
anddearnessallowancein respectof workers who receive both these components
while forotherworkersthe actualconsolidatedamountearningsrepresenttheirwage
rate" [Occupational Wage Survey (Fourth
Round),ReporlonTextileIndustries(198687), LabourBureau1989:46].
11 "For the purposeof this survey,pay roll
earnings have been defined as all those
componentsof remunerationreceivedby
workersmoreorless on regularbasis.Thus
basic wage,DA, productionandincentive
bonus,attendancebonus,overtimeallowance, shift allowance,cycle/transportallowance, washing allowance,etc, are includedinthedataon payrollearnings.ney
also includemoney valueof benefits/concessions accruingto workierson more or
less regularbasis".[Occupationlal
Wag,e
debt,Eequitycapital,t taxrateandrinterest
Suirvey(FourthRound),Reporton Textile
rate. A few variantswere tried using diflrtdu.stries(1986-87),LabourBureau1989:
ferent measures of return on equity. In
106].
all the cases the cost of capitalis foundto
12 Thesearecottontextiles,woollentextiles,
haveincreasedon a trendbasis since 1970silktextiles,garments,bicycle,agricultural
71. I am gratefulto SubirGokamfor en-machineryandtextilemachinery.
lighteningmea greatdeal on the analytical
13 As the coveragehas expandedin succesand empiricalissues in computingcost of
sive roundsof OWS,a comparisonof more
capital.
recentroundsenablesonetoincludegreater
23 In this exercise, the criticaldata are estinumberof industriqs.
matesof capitalsto4kseries,whichwe have
14 The observedincreasesin wage rateand
takenfromAggarwal(1991). Since the 42
earningsperhapsget overstateddue to a
three-digitindustriesaccountfor 72 to 75
conceivablereductionin themduringthe
percentof netvalueaddedatcurrentprices
emergency.A close perusalof the data
in the entire period,our exercise can be
confinnsthe suspicion.
consideredreasonablycomprehensive.
15 RobinMarrisobserveda similarvariation
betweenwagerateandearningsfortheUK 24 We computedaveragesfortheyears197374/1974-75, 1978-79/1979-80 and 1985manufacturingindustriesin the 80s, by
86/1986-87 and the incrementsfor each
comparingthe official earningsenquiry
industry.These series were used for comwith employers' associations' statistical
puting correlations,as mentionedin the
series on 'settlements'which relates to
text.
inbasicratesforstandard
"agreedincreases
workunderstandardconditions"[Marris 25 Mazumdar(1989), for instante, is consciousof thecomplexityof the labourmar1987:48].
ketbehaviourindevelopingeconomiesand
16 Thoughthese are official figuresand are
seems to argue for a closer studyof indithe only source to discernthe long-term
vidual economies before arrivingat simtrends, their quality, accordingto some
plisticgeneralisations.
unionanalysts,leaves muchto be desired.
It is believedthattradeunionsoftendo,not 26 To quoteDesai: "Manyindustrialmarkets
in Indiaarecompetitiveandaccommodate
file returnson time,especiallythe smaller
a large numberof firms...Ratherthan a
ones andfirm-basedindependentunions.
tendencyfor monopoly,the marketstruc17 However,at a disaggregatedlevel, trend
tures display a tendencytowardsintense
declinein uniondensityis statisticallysigcompetition..."[Desai 1985:168].
nificantinonlyeightoutof l8 majorindustrygroupsandinonlyoneindustry,namely, 27 Real interestrate is equal to [(1+interest
rate)/(1+inflationrate)]. We have used
cottontextiles(NIC23)thetrendis positive
scheduledcommercialbanks' depositrate
andstatisticallysignificant.Inthe remainfor one yearandabove maturityandinflaing cases,due to considerableyearlyfluctionrateis derivedfromimplicitGDPdeflathetrendsare
tuationin unionmembership,
tor.
notstatisticallysignificant,thoughhavethe
28 Freeman's(1993a and 1993b) assessment
expectednegativesign.
of the evidence on the validity of the
18 Althoughthisisawidelyused,andis a valid
views
'distortionists'and'interventionists'
simplificationatanaggregatelevel, we are
on the labour market,against the backconsciousthat,at a micro-economiclevel
groundof structuraladjustmentin a large
lock out by managementmay well reprenumberof developingcountries,seemsinsent a pre-emptivemeasureto preventa
structive:
strikeby workersor vice versa.
Inthe80s manyeconomistscameto view
19 Eventhisfigureis perhapsanoverestimate
institutionalinterventionsin the labour
sinceit is notuncommonfor workersto be
marketas majorimpedimentsto growth
membersof morethanone union.
and employment...Concernswere exconstitutes
20 Thatregisteredmanufacturing
of thetotal
pressedthatthese and relatedintervena smallanddecliningproportion
tions would preventlabour marketadmanufacturingemployment is not admany.Forinstance,
justmentsnecessaryforstabilisationand
equatelyappreciatedby
structuraladjustment.
theWorldBank(1989),unableto reconcile
Forthesubsetof countriesthathavebeen
the employmentfiguresfromthe ASI and
studiedintensely, the evidence dispels
thepopulationcensusseemstoinferthatthe
fearsthatgovernmentorunionintervenlatterarean overestimate.
tions invariablyimpede labour-market
21 Informationon employmentby size class
of factoriesis availableonly for factory
adjustments...
Inshortwhenyouarrivein developiaand
accountsfor
sector.Since manufacturing
find the economy a mess, don't look to
overfour-fifthsof thefactoryemploymnent,
governmentor union labour-market
our findingis unlikelynot incorrect.
in-.
22 Thoughwidelyused,realinterestrateas a
terventionas the prime cause of the
trouble[Freeman1993b:404-051.
measureof the cost of capital has many
limitations,since it ignores tax rate and
References
depreciationrules.We estimatedtrendsin
thecostof capitalfortheRBI's mediumand Aggarwal,AshokK (1991):Estimnates
of Fixed
large non-financialpubliclimitedcompaCapitalStockin RegisteredManufacturing
nies usingtheequation:(DID+ E) (1 - t) r
Sector in India,W P No 937, IndianInsti+ (did + E) (returnon equity),where D is
tute of ManagementAhmedabad.
Economic and PHitical Weekly January 22, 1994
18S
Ahluwalia, Isher Judge (1992): Productivity
anidGrowth in InidianManufactu'ring, Oxford University Press, Delhli.
Amsden, Alice (1990): 'South Korea's Record
Wage
Rates:
Labour
Late
in
Industrialisation', Industrial Relations,
Vol 29, Nol, Winter.
Desai, Ashok (1985): 'Market Structure and
Technology: TheirInterdependence in Indian Industry', Research Policy, Vol 14.
Dobb, Maurice (1959): Wages, Cambridge
Economic\Handbooks, Cambridge University Press.
Fallon, PeterRand Lucas, Robert E B (1993):
'Job Security Regulations and the Dynamic Demand for Industrial Labour in
India and Zimbabwe', Journal of Developtne~t Ecotnomnics,Vol 40.
Freeman, Richard B (1988): 'Labour Market
Institutions and Economic Performance',
Ecotnomic Policy, Vol 3, No 1, April.
-(1993a): 'Labour Market Institutions and
Policies: Help or Hindrance to Economic
Development?', Annual Conference on
Development Economics, 1992, Supplement to the World Banik Research Observer, March.
-(1993b): 'Labour Markets and Institutions
in Economic Development', American
Economic Review, Volume 83, No 2, May.
Government of India (1993): Econtomic Reforms: Two YearsAfterand the TaskAhead,
Department of Economic Affairs, Ministry of Finance.
Lucas, Robert E B (1988): 'India's Industrial
Policy' in Robert E B Lucas and Gustav F
Papanek (eds), The Indiani Economty:Recent Developmetnt atndFutitfreProspects,
Oxford University Press, Delhi
Marris, Robin (1987): 'Does Britain Really
Have a Wage Problem?', LlYods Banik
Review, April.
Mazumdar, Dipak (1989): Micro-economic
Issues of Labouir Markets in Developing
Countries: Analysis antd Policy hnmplications, EDI Seminar Paper No 40, World
Bank, Washington DC.
Nagaraj, R (1985): 'Trends in Factory Size in
Indian Industry, 1950 to 1980', Economic and Political Weekly, Review of
Management, February.
-(1989): Sub-Conttracting in Indian Manufactuiring Industries: The Banigalore Experience, PhD Thesis (Centre for Development Studies, Trivandrum), Jawaharlal
Nehru University, New Delhi.
Planning Commission (1990): Emiployment:
Past Tretds and Prospects for 1990s,
Working Paper, New Delhi.
Lance (1988):
Taylor,
Varieties
of
Stabilisation Experienice: Towards Sensible Macro-economics in the T7hird World.
Clarendon Press, Oxfordl.
Visaria, Pravin and Minhas B S (1991): 'Evolving an Emlployment Policy for the 1990s:
What Do the Data Tell Us?', Ecottomic
and Political Weekly, April 13.
World Bank (1989): Imidia:Poverty, Employment amid Social Services, Washington
DC.
186
RELEASESFROM-SEGMENT OOS
RURAL CREDI
Roleof InfrmnaSoctr
by
Dr. Dinesh Chandr
forewordby:
Prf. A.M Khusro
ISBN:81-85330.204
Price:Rs.40.00
STRUC URAL ADJUSTMIENT
PROGRAMME:
Puttingtle FaustThing, fvry
-and
Envionmat Last
An
Analysis
by
Prof.AMoTAVAMUKHERJEE
Profeor. of Plaoning& o
s
LB.S. NadonalAcadn of Adminmstadon.
ISBN: 81-85330-21.-2
Price:Rs. 450.00
DEVELOPMENTCHALLENGES
needfor a newvaluesystem
by
Padmanabhan
forewordby
Prof. Hans. W Singer
ISBN: 81-85330-18-2
Price:Rs. 330.00
Pleasesandyourordersto:
SegmentBook Distributors
B/23 & 25, KailashColony,New Delhi-110048
Phone: 644-3013
Fax No: 011-6470472
Economic and Political Weekly January 22, 1994