Is direction of trade responsible for wage inequality in India? An empirical verification.
Tiwari, Aviral Kumar ; Aruna, M.
Abstract
The present study analyses the impact of direction of trade on wage
inequality in Indian economy and therefore attempts to empirically test
the validity of the HO-SS (Heckscher-Ohlin-Stopler-Samuelson) theorem
and the model given by Davis (1996). This study employs different
specifications of semi log linear multiple regression model for the
analysis. To minimize the problem of multicollinearity, principal
component analysis has been used. The empirical result of the present
study rejects the proposition of both the models. In other words
India's trade with developed and developing countries leads to
increase in wage inequality.
I. INTRODUCTION
Economic reforms were adopted in developed or developing countries
with a predetermined objective that via integration of the domestic
country with the rest of the world will result in increase in the growth
rate, increase productivity on one side and reduces intra and inter
country income inequalities as well as wealth inequalities on the other
side. It was thought that Hecksher-Ohlin-Stopler-Samuelson (HO-SS)
framework would be bringing egalitarian trend. Over the years,
liberalization has benefited all the countries in the form of expansion
of market, by increasing the inflow of capital, by creation of jobs by
Multi-National Companies/Trans-National Corporations (MNCs/TNCs) and
through transfer of knowledge and import of advanced technology.
However, it is seen that it has not only widened intra and inter country
inequality (See for detailed discussion O' Rourke 2002) but also
has created some negative social effect (2) Thus, it is very important
to give balanced view about benefits and the problems of the
liberalization. It is also necessary to figure out who is the winner and
who is the looser in order to make a balanced policy decisions. So the
present study is intended to address this real world economic problem.
A number of studies have been conducted in the past for developed
countries particularly for US, UK and Latin American countries to answer
the question, who is responsible for widening wage inequality between
skilled and unskilled labors (See for example Haskel and Slaughter
(2001, 2003); Tombazos (1999); Haskel and Heden (1999); Berman, Bound
and Machin (1998); Autor, Katz and Kruger (1998); Khan and Lim (1998);
Sachs and Shatz (1994); Haskel and Szymanski (1993); Bound and Johnson
(1992); and Katz and Murphy (1992) among others) ? Though, there is no
consensus among researchers, but most of them have found that Skilled
Biased Technological Change (SBTC) is mainly guilty and the effect of
trade on wage inequality is either negligible or zero; even if the
effect of trade on wages is high, it is found that these effects are
counter balanced by the effect of SBTC (See, Haskel and Slaughter
(2001); Berman, Bound and Machin (1998); Autor, Katz and Kruger (1998);
Khan and Lira (1998); and John bound and George Johnson (1992) among
others).
Recently, the focus of researchers has shifted towards developing
countries and a limited number of researches are available for India and
for other developing countries. These studies have applied econometric
techniques and have found conflicting results.
India offers interesting case to study the effects of
globalization- measured in the form of trade and Foreign Direct
Investment (FDI) inflow- on wage inequality due to the following
important reasons.
* Since mid 1980s, the Government of India (GOI) implemented a
number of far reaching economic policy reforms in the domestic sector as
well as external sector. Domestic sector reforms constitute industrial
deregulation, introduction of Competition Act (2002), Competition
(Amendment) Bill 2007, privatization, disinvestment, financial sector
reform etc. and external sector reforms constitute trade policy reforms,
exchange rate reforms, and investment reforms (i.e., FDI allowance).
* Although the reforms initiated (marginally) in mid 1980s were
endogenous, the major reforms implemented in 1991 were exogenous as
these were brought to solve the Balance of Payment (BOP) crisis under
International Monetary Fund (IMF) conditionalities.
I.I. India's Experience during 1980s and 2000s
India's experience during last three decades can be analyzed
as follows:
* Widening wage inequality between skilled and unskilled labor (but
in case of gender wage inequality there is no consensus among
researchers). For example study by Dutta and Reilly (2008) finds little
evidence that openness is an important determinant of the industry level
pay gaps but study by Menon and Rodgers (2008) finds that increasing
openness to trade is associated with gender pay gaps.
* Supply of skilled labors (i.e., number of educated peoples
inclusive of males and females) as well as demand (i.e., share in
employment) for skilled labors
has also increased. For example Bishwanath (2009) found the
increased employment elasticity of workers in post reform period that
shows that in post reform period demand for skilled labors has
increased.
* Wage inequalities among states, among urban and rural areas,
across job types, and between older and younger age groups who are not
college educated have increased (See for details Mukherjee (2007) for
occupational, spatial and across jobs wage inequality and for
inter-state disparity see Ghose and Roy (2007), and Aggarwal (2007)).
* The value and volume of trade has increased considerably, but
composition and direction of trade has got changed. (3)
* Inflow of FDI has also increased in both absolute and relative
terms (i.e. FDI as percentage of Gross Domestic Product (GDP)). (4)
* Percentage expenditure of firms on in-house R&D expenditure
has also increased both in absolute and relative terms (i.e., R&D
expenditure as percentage of sales). (5)
* Immigration also has increased (particularly illegal). For
example Yabuuchi and Chaudhuri (2007) and Chaudhuri (2008) have
discussed that developing countries including India are facing the
problem of illegal immigration of people (particularly unskilled labor)
from neighboring poor countries.
* Bargaining power of unions has decreased because of reduction in
the membership of unions (For detailed analysis see Mathur and Mishra
(2007)).
I.II. Objective of the Study
The main objective of this study is to identify the factors
affecting the wage inequality and to examine whether the direction of
India's trade is responsible factor for widening wage inequality.
I.III. Definition
In this study ex-post measurable and objective definitions (like
FDI inflow as a ratio GDP, total trade as a ratio of GDP etc.) of
globalization has been used and wage inequality has been defined using
occupational criteria.
II. LITERATURE REVIEW
This section is organised in to two subsections. First sections
discusses about the basic HO-SS model and second section discusses about
the alternative expiations put forward by the ecnomists recently.
II.I. Basic Theory
Under Hecksher-Ohlin (H-O) model, the Stopler-Samuelson (S-S)
theorem was the first theoretical formulation to explain the effects of
free trade on income distribution among productive factors. The basic
result of HO-SS model is that, protection increases the relative return
to the factor scarce in the country- labor in developing country and
capital in developed country or unskilled labor in developing country
and skilled labor in developed country. Therefore, when liberal trade is
allowed between developed and developing country, prices of unskilled
labor intensive (exported) products should increase the wages of
unskilled workers and the prices of skilled labor intensive (imported)
products should decrease the wages of skilled workers in developing
country. So, this implies that wage inequality should be reduced in the
developing country due to opening of the home country for trade with
developed country. This also implies that prices of skilled factor will
increase in developed country via increase in the prices of skilled
labor intensive commodity and prices of unskilled factor will fall due
to the same reasoning in developed country so opening of developed
country for trade will increase wage inequality. However, empirical
evidence has turned against this conventional wisdom and has drawn
attention to the alternative mechanisms through which the trade openness
has affected the wage inequality.
II.II. Alternative Explanations
An extension of the above (HO-SS) model considers capital,
unskilled and skilled labor as relevant factors of production. However,
in these types of model, capital skill complementary is one underlying
assumption. It was originally proposed by Rosen (1968) and Griliches
(1969) and has been recently explored by Goldin and Katz (1998), Machin
and Reenen (1998) and more recently by Krusell et al. (2000) and Acmoglu
(2003). This assumption is based on the argument of Wood (1995) that is
trade liberalization results in "defensive innovation". This
implies that greater competition from foreign firms in Less Developed
Countries (LDCs) will force the domestic firms either to engage in
R&D (Research and Development) or to adopt new and advance
technologies through import in order to secure their market share in the
domestic as well as international market.
Trade liberalization has encouraged the inflow of Foreign Direct
Investment (FDI). Thus, the effects of FDI on demand of the skilled
labor and on wage inequality are more direct. That is, FDI raises the
demand for skilled labor in both developed and developing countries and
increases wage inequality since FDI brings new technologies which are
skilled biased (Berman, Bound, and Machine 1998). Even if technologies
brought by FDI are factor neutral, the transition process of
transferring and installing new technologies are skill biased
(Pissarides 1997). So in all cases FDI will increase in wage inequality.
However, Figni and Gorge (1999) have proposed inverted U hypothesis for
FDI's effect on wage inequality.
Davis (1996) has developed a model in which the central hypothesis
was that the availabilities of a country's factor of production
should not be assessed in relation with the wider international economy
rather it should be assessed with a group of countries with similar
factor endowments. In this model, trade liberalization can increase the
demand for skilled labor in a developing economy as long as among the
countries of its cone, it has relatively high supply of skilled labor.
Therefore, a country from a cone where there is greater supply of
skilled labor can experience a reduction in wage inequality. The
reduction in the prices of products produced in developed country has no
effect on the prices of the factors of production in developing country,
since they do not produce the same goods.
Model developed by Feenstra and Hanson (1997) shows that the
increase in wage inequality in developed and developing country is
consistent with FDI flow from developed to developing country. Since,
FDI inflow changes the structure of the production; their model assumes
the production of a simple final good that requires a continuum of
intermediary goods with varying proportions of skilled and unskilled
labor. The cost of the production of the final good was assumed to be
smaller in developing countries than in the developed countries and
assuming that capital returns are higher in developing countries, when
trade liberalization takes place in developing countries there will be
transfer of capital or FDI from developed to developing countries, this
increases the demand of skilled worker in both countries and thereby
resulting in wage inequality.
Acemoglu (2002, 2003) has developed a model of endogenous
technological change. In this model, increasing supply of skilled labor
induces SBTC through the market size effect. That is greater demand for
skilled intensive good by consumers or educated labors will increase the
profitability of skilled intensive good and thereby encourage SBTC.
Domestic institutions like Job Security Regulations (JSRs) are
argued to be important determinants of industry performance like
productivity, profits, and employment during liberalization era.
However, the impact of these regulations on wage inequality has not been
addressed yet adequately. While these regulations can restrict the
firms' ability to adjust the skill mix in response to the trade
openness.
The study by Chamarbagwala (2006) has investigated the effects of
trade liberalization on wage inequality. Since the author used NSSO data
therefore in this case wage inequality is defined as the wage gap
between the high educated and the low educated workers. This study used
the approach developed by Katz and Murphy (1992). This approach is based
on the supply and demand framework.
The result of the study supports the argument that domestic and
external sector reforms have created more white-collar jobs. That is,
reforms have created jobs for those who were ready to upgrade their
skills.
Study by Berman, Bound and Griliches (1994) developed a different
decomposition approach in this area. Though they also found that within
industry component dominates the between industry components in both
cases of the employment and the wages. However, they found that for the
last period -1979 to 1987--between industry components plays a major
role. Next, the study decomposes within and between industry components
into within-between sectoral components. The authors have taken
consumption, export, import and defense as sectors. They found that most
of acceleration in the proportion of nonproduction workers/skilled
workers wages in the particular industry and acceleration in the share
of employment in the particular industry were due to within skill
upgrading. Thus the results found to be favoring SBTC hypothesis.
To validate the decomposition findings and give more insights they
carried out regression analysis on different specifications and found
similar results as they derived from decomposition approach.
Study by Kijima (2006) analyzed the changes in the overall wage
inequality distribution of urban in India during 1980s and 1990s.
Furthermore, it also identifies the causes of the changing wage
distribution in urban India during the period of 1983 to 1999. To
analyze the data the author has compiled the data in percentiles and the
wage inequality was measured by the wage differential between 90th and
10th percentiles of wage distribution. The author found that this
differential started increasing in 1980s but during the reform era of
the 1990s it has increased at rapid rate. It was also found that wage
inequality grew faster after 1993 than before 1993. This difference was
due to wage inequality above median deteriorated more rapidly after
1993. To explain the causes of these differentials the study followed
the approach developed by Juhn, Murphy, and Pierce (1993). The author
found an interesting result that the factor responsible for wage
inequality was different in two periods. The author found that
increasing wage inequality between observed skills (such as schooling
and working experience) was a major contributor to the increase in wage
inequality in 1980s on the other hand the rise in returns to observed
skills (particularly tertiary education) increased the wage inequalities
in 1990s. Using the given result of the decomposition, the author
examined the causes of increased skill premium. The author found that
the effect of demand shift was greater than the effects of supply shift.
This demand shift for skilled workers was due to SBTC measured by within
industry demand shifts. Trade reforms measured by between industry
shifts of skilled workers was only a minor contributor to increase in
demand for skilled workers.
In this study the author focused on wage distribution of urban male
workers by arguing that, since female workers participation in labor
force in only 20 percentages so, it will not affect the main result. But
if female workers participation in skilled workers is high it will
definitely affect the result (Dutta. 2008).
Kumar and Mishra (2008) used this approach to measure the impact of
trade reforms. They followed the approach of Goldberg and Pavnick (2004)
in order to explain variation in wages and trade policy measures (tariff
and other measures) across industry and over time to identify the impact
of trade on the wages. They used two stage estimation technique. They
found that the coefficient of tariff was negative and statistically
significant. The coefficient of trade was not found to be statistically
significant implying that it is not an important factor for increasing
wage inequality.
Further they carried out sensitivity analysis to check the
robustness (6) of the result reported for coefficient of tariff. They
found that effect of tariff on wage inequality is robust with inclusion
of exchange rate measure.
It is noteworthy that there are other ways through which trade can
affect the wage inequality, for example the unionization, the
contemporaneous real import and export flows, the minimum wage, and the
immigration as mentioned by Pinelopi Koujianou Goldberg and Nina Pavcnik
(2005). The effects of these ways on the wage inequality have not been
analyzed by this study.
Dutta (2007) followed the almost similar approach. She used the two
steps model for the selection and the wage determination. First, she
used the Multinomial Logit (MNL) model. Then she followed the two stage
regression technique used by Kumar and Mishra (2008) but then she used
two regression techniques namely Ordinary Least Squares (OLS) and
Weighted Least Squares (WLS). She found that the coefficient of tariff
was positive and significant which indicates that wage inequality is
increasing.
Further, she carried out the sensitivity analysis to check the
robustness of the result reported for the tariff. She found that the
coefficients of the tariff was robust with respect to the inclusion of
no-exemption tariff measure and the other alternative measures like- the
contemporaneous real import and export flows, the import and export
shares, and the import penetration and the export intensity.
The interesting thing is that where Kumar and Mishra (2008) finds
tariff as a variable which decreases wage inequality Dutta (2007) finds
tariff increases wage inequality.
Mullen and Panning (2009) used cost share equation as used by
Berman, Bound, and Griliches (1994) in spirit of Feenstra and Hanson
(1999). The period of analysis of this study was 1997-2002. The authors
modified cost share equation by adding two regressors' namely
technical progress and outsourcing. The same cost share equation was
used to explain variation in the employment also.
For the estimation in this study the Generalized Least Square (GLS)
technique has been used where the industries were weighted by the value
added shares of the industries. They have used different models in the
sense that each model either include or exclude different variables
and/or variable definitions.
They found that output increases the share of unskilled workers,
capital skill complimentarity do not exist i.e. more investment in
capital do not increase the share of skilled workers while off shoring
not only increases the wage inequality but also decreases employment of
unskilled workers. Also domestic investment on research and development
activities increases wage inequality.
II. METHODOLOGY, DATA AND MODEL DESCRIPTION
The present study employs the model developed by Berman, Bound and
Griliches (1994) and later it is used by Machin and Reenen (1997), and
Mullen and Panning (2009). This model is derived from translog cost
function. The cost function assumed to be quasi fixed as capital is
assumed to be fixed and both production and nonproduction workers are
treated as variable. Translog variable cost function can be written as:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)
where CV represents variable costs, Y is value added, the W's
represent unit costs of the variable factors and K represents capital.
Cost minimization implies share equations of the form
[S.sub.i] = [alpha.sub.i] + [rho][Y.sub.i]ln(Y) +
[summation.sub.j][[gamma].sub.ij]ln([W.sub.j]) + [[rho].sub.i] ln(K) (2)
Given that [summation][S.sub.i] = 1 and the symmetry and
homogeneity restrictions, only one of the two share equations required
to be estimated. Using the restrictions of constant returns to scale
assumption the wage bill share equation for unskilled workers can be
written as:
SUWW = [alpha.sub.1] + [alpha.sub.2]ln(Y) +
[alpha.sub.3]ln([w.sup.usk] / [w.sup.sk]) + [alpha.sub.4]ln + (K/Y) +
[epsilon] (3)
The specifications of the models used in the present study are:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)
where SUWW is share of unskilled workers to total wages, K/Y
denotes capital output ratio and TT/GDP measures ratio of total trade to
gross domestic product and finally FDI/GDP measures ratio of foreign
direct investment to gross domestic product. Similarly to analyze the
impact of direction of trade on wage inequality variable TT/GDP is
replaced by variable TTDCs/GDP and TTLDCs/GDP in next two models. Where
TTDCs denotes trade with developed countries and TTLDCs denotes trade
with developing countries.
This study has used time series data from 1980-81 to 2005-06. Data
Sources for the study are EPW-CD Rom (Vol-II), ASI Factory sector, Hand
book of statistics (RBI), Indian economy database (Vol-II), WDI, and
UNCTAD.
Wage inequality in this study is measured by share of unskilled
workers (and unskilled workers are taken as non production workers) in
total workers, output has been measured by ratio of net value added to
Gross Domestic Product, capital output ratio has been measured by a
ratio of Gross Fixed Capital Formation to net value added, openness has
been measured by ratio of total merchandise trade to Gross Domestic
Product and by ratio of inflow of foreign direct investment to Gross
Domestic Product.
However, principal component analysis has been done due to problem
of multicollinearity between ratio of total merchandise trade and Gross
Domestic Product and ratio of foreign direct investment to Gross
Domestic Product, between ratio of total trade with developed countries
to Gross Domestic Product and ratio of foreign direct investment to
Gross Domestic Product, and between ratio of total trade with developing
countries to Gross Domestic Product and ratio of foreign direct
investment to Gross Domestic Product. Variance inflation factor (VIF)
was ranging from 8.0 to 20.0 in some cases and condition index was
ranging from 30.0 to 100.007. By applying principal component analysis
and using Eigen value criteria (>1) and scree plot criteria two
factors were extracted, namely domestic factors and external factors and
to avoid cross loadings between components or factors Varimax rotation
with Kaiser Normalization was used.
IV. DATA ANALYSIS AND FINDINGS AND CONCLUSIONS:
Table 1 clearly shows that there may be problem of
multicollinearity among ratio of wages of unskilled workers to skilled
workers, TT/GDP and FDI/GDP and capital output ratio, and between
FDI/GDP and TT/GDP.
So, to avoid problem of multicollinearity principal component
analysis has been performed and the factors extracted are presented in
table 2.
Table 2 shows that total variance explained by the two factors
extracted is 86.168% which is good enough to carry out analysis. Table 3
shows rotated salutation of the components or factors. The values given
under the column of components are factor loadings. Factor loadings
values indicates that output, capital-output ratio, and ratio of wages
of skilled to unskilled workers are coming under component two; this
component has been named as domestic factors and ratio of FDI to GDP and
ratio of total trade to GDP is coming under factor one; this factor is
named as external factors.
One more important thing is that loading of variable output is high
domestic factor indicating that output is most important variable which
affects wage inequality similarly in external factor loading of ratio of
total trade to GDP is high indicating that it is most important variable
in external factor. The goodness fit of the model is presented in table
4(1), 4 (2).
The empirical estimate presented in table 5 indicates that the
external factors has not only negative impact on share of wages of
unskilled workers but also their negative impact is found to be
higher(-.041) than the positive impact of domestic factors(.024) on
share of unskilled workers.
The table 6 shows the indication of multicollinearity among same
set of variables.
So, to avoid problem of multicollinearity again principal component
analysis has been done and following two factors has been extracted
namely domestic and external factors respectively. The cumulative
variance explained by the two factors is 86.107% and presented in table
7.
Here, it is clear from table 8 that loading of variable TTLDCs/GDP
is comparatively higher than the previous case in factor two which has
been named as external factor. Loading of output has fallen marginally
and loadings of capital-output ratio and ratio of wages of skilled to
unskilled workers has increased marginally in factor two which has been
named as domestic factor.
Table 9 presents the goodness of fit of the model 9 (1) and 9 (2).
Table 10 reveals that negative value of external factor (-.035) is
higher than the positive value of domestic factor (.031). It can be
inferred that increase in wage inequality is due to external factors
which are over compensating the positive impact created by the domestic
factor on share of unskilled workers.
Table 11 gives an indication of the problem of multicollinearity
among the variables capital-output ratio and ratio of wages of skilled
to unskilled workers, ratio of FDI to GDP and ratio of total trade with
less developed countries to gross domestic product.
So, to avoid multicollinearity principal component analysis has
been performed and two variables have been extracted. Cumulative
variance explained by both factors is 86.852% as presented in table 12.
The rotated component matrix presented in table 13 shows that
loading of trade with less developed countries is comparatively higher
than the FDI in external factor (component one) and loading of output
variable is comparatively higher than other variables loaded in domestic
factor (component two).
Obtained two factors scores were regressed on the share of wages of
unskilled labors.
Goodness fit of the model is presented in table 14 (1) and 14 (2).
Table 15 gives a very interesting finding that score of external
factor (-.041) in case of third model was found: to be having similar
impact as was in case one. This indicates that trade with developing
countries found to be having larger negative impact on share of wages of
unskilled workers than the trade with developed countries.
V. CONCLUSIONS
This study finds that domestic factors are decreasing the wage
inequality while external factors are playing major role in increasing
the wage inequality. Though, it is difficult to interpret that in India,
capital-skill complimentarity do not exist but it can be said that even
if it exists its negative impact on share of unskilled workers is
overcompensated by output and by ratio of wages of unskilled to skilled
workers because the coefficient of domestic factor in all three cases
was found to be positive.
As far as direction of trade is concerned it is found that trade
with less developed countries or what we can call trade with developing
countries is major variable affecting the wage inequality than the trade
with developed countries. This empirical evidence contradicts the model
of Davis (1996).
The limitation of this study is that India's trade with all
developed countries and all developing countries has not been taken for
analysis. Further, this study can be extended in the direction of
industry wise and country wise analysis.
References
Acemoglu, Daron (2002), "Technological Change, Inequality, and
Labor Market" Journal of Economic Literature, Vol. 40, 7-72.
--.(2003), "Pattern of Skill Premia" Review of Economics
Studies, Vol. 70, 199-230.
Agrawal, Suresh Chand (2007), "Wage Discrimination in Indian
Manufacturing: A State-level Analysis," The Indian Journal of Labor
Economics, Vol. 50, No. 4, 9991-998.
Autor, David H., Lawrence F. katz, and Alan B. Krueger (1998),
"Computing Inequality: Have Computers Changed the Labor
Market?," The Quarterly Journal of Economics, Vol. 113, No. 4,
1169-1213.
Berman, Eli, Jhon Bound, and Stephen Machin (1998),
"Implications of Skilled-biased Technological Change: International
Evidence," The Quarterly Journal of Economics, Vol. 113, No. 4,
1245-1279.
--, and Zevi Griliches (1994), "Changes in the Demand for
Skilled Labor within US Manufacturing: Evidence from Annual Survey of
Manufactures," The Quarterly Journal of Economics, Vol. 107, No. 1,
367-397.
Bound, John, and George Johnson (1992), "Changes in Structure
of Wages in the 1980s: An Evaluation of Alternative Explanations,"
The American Economic Review, Vol. 82, No.3, 371-392.
Chamarbagwala, Rubiana (2006), "Economic Liberalization and
Wage Inequality in India" World Development, Vol. 34, No.12,
1997-2015.
Chaudhuri, Sarbajit (2008), "Wage Inequality in a Dual Economy
and International Mobility of Factors: Do Factor Intensities Always
Matter?," Economic Modelling, Vol. 25, 1155-1164.
Davis, D. R. (1996), "Trade Liberalization and Income
Distribution" NBER Working Paper, No. 5693.
Dutta, Puja Vasudeva (2007), "Trade Protection and Industry
Wages in India" Industrial and Labor Relation Review, Vol. 60, No.
2, 268-286.
--, and Barry Reilly (2008), "The Gender Pay Gap in Era of
Economic Change: Evidence for India, 1983 to 2004," The Indian
Journal of Labor Economics, Vol. 51, No. 3, 341-366.
Feenstra, Robert C. and Gordon H. Hanson (1997), "Foreign
Direct Investment and Relative Wages: Evidence from Mexico's
Maquiladoras. Journal of International Economics, Vol. 42, 371-393.
--(1999), "the Impact of Outsourcing and High Technology
Capital on Wages: Estimates for the United States, 1979-1990," The
Quarterly Journal of Economics, Vol. 114, No. 3, 907-940.
Figini, P. and H. Gorg (1999), "Multinational Companies and
Wage inequality in the Host Country: The Case of Ireland,"
Weltwirtshaftliches Archives, Vol. 135, 594-612.
Freeman, R. B. (1980), An Empirical Analysis of the Fixed
Coefficients Manpower Requirements Model, 1960-1970. The Journal of
Human Resources, 15 (2), 176-199.
Ghose, Arpita, and Satarupa Roy (2007), "Inter-state Disparity
in Growth and Determents of Wage Rate for Industrial Labor in
India" The Indian Journal of Labor Economics, Vol. 50, No. 4,
975-989.
Goldar, Bishwanath (2009), "Trade Liberalization and Labor
Demand Elasticity in Indian Manufacturing" Economic and Political
Weekly, Vol. 44, No. 34, 51-57.
Goldberg, Pinelopi Koujianou, and Nina Pavcnik (2005), "Trade,
Wages, and the Political Economy of Trade Protection: Evidence from the
Colombian Trade Reforms," Journal of International Economics, Vol.
66, 75-105.
Goldin, C., and Katz Lawrence F. (1998), "The Origin of
Technology-skill Complimentarity" The Quarterly Journal of
Economics, Vol. 113, 693-732.
Griliches, Zevi (1969), "Capital Skill Complimentarity"
Review of Economics and Statistics, Vol. 51, 465-468.
Haskel, Jonathan E., and Matthew J. Slaughter (2003), "Have
Falling Tariff and Transportation Costs Raised US Wage
Inequality?," Review of International Economics, Vol. 11, No. 4,
630-650.
--, and Stefan Szymanski (1993), "Privatization,
liberalization, Wages and Employment: Theory and evidence for the
U.K," Economica, Vol. 60, 161-182.
--, and Ylva Hede (1999), "Computers and the Demand for
Skilled Labor: Industry-and Establishment-level Panel Evidence for the
UK," The Economic Journal, Vol. 109, C68-C79.
Juhn, C., K. Murphy, and B. Pierce (1993 "Wag Inequality and
Rise in Returns to Skill" Journal of Political Economy, Vol. 101,
410-442.
Katz, Lawrence F., and Kevin M. Murphy (1992), "Changes in
Relative Wages, 1963-1987: Supply and Demand Factors," The
Quarterly Journal of Economics, Vol. 107, No. 1, 35-78.
Kevin H. O'Rourke (2002), "Globalization and Wage
Inequality: Historical Trends" Aussenwirtschaft, March, 65-101.
Khan, James A., and Jong-Soo Lim (1998), "Skilled
Labor-augmenting Technological Progress in U. S. Manufacturing" The
Quarterly Journal of Economics, Vol. 113, No. 4, 1281-1308.
Kijima, Yoko. (2006), "Why Did Wage Inequality Increase?
Evidence from Urban India 1983-99," Journal of Developed Economics,
Vol. 81, 97-117.
Krusell, P., Ohanian, I, E., Rios-Rull, J. V., and Violante. G. L.
(2000), "Capital Skill Complimentrity and Income Inequality; a
Macroeconomics Analysis" Econometrica, Vol. 68, 1029-1053.
Kumar, Utsav, and Prachi Mishra (2008), "Trade liberalization
and Wage Inequality: Evidence from India," Review of Development
Economics, Vol. 12, No. 2, 291-311.
Machin, Stephen, and John Van Reenen (1998), "Technology and
Changes in Skill Structure: Evidence from Seven OECD Countries,"
The Quarterly Journal of Economics, Vol. 113, No. 4, 1215-1244.
Mathur, Ashok, and Sunil Kumar Mishra (2007), "Wages and
Employment in the Indian Industrial Sector: Theory and Evidence,"
The Indian Journal of Labor Economics, Vol. 50, No. 1, 83-110.
Menon, Nidhiya, and Yana Van Der Mettlen Rodgers (2009),
"International Trade and the Gender Wage Gap: New Evidence from
India's Manufacturing Sector," World Development, Vol. 37, No.
5, 965-981.
Mukherjee, Dipa (2007), "Post Reforms Trends in Wage
Differentials; a Decomposition Analysis for India" The Indian
Journal of Labor Economics, Vol. 50, No. 4, 955-965.
Mullen, John K., and James Planning (2009), "Foreign Sourcing
of Intermediate Inputs: Impacts on Unskilled Labor in US Manufacturing
Industries," Eastern Economic Journal, Vol. 35, 160-173.
Pissarids, C. A. (1997), "Learning by Trading and Returns to
Human Capital in Developing Countries" World Bank Economic Review,
Vol. 11, 17-32.
Robbins, D. J. (1996), "HOS Hits Facts: Facts Win; Evidence on
Trade and Wages in the Developing Countries," Development
Discussion Paper, No. 557.
S., Rosen (1968), "Short-term Employment Variation on Class-1
Railroads in the US, 1947-63" Econometrica, Vol. 36, 11-29.
Sachs, Jeffrey D., and Howard J. Shatz (1994), "Trade and Jobs
in U. S. Manufacturing" Brooking Study's on Economic Activity,
Vol. 1, 1-84.
Tombazos, Christis G. (1999), "The Role of Imports in
Expanding the Demand Gap between Skilled and Unskilled Labor in the
US" Applied Economics, Vol. 31, 509-516.
Wood, Adrian (1995), "How Trade Hurt Unskilled Workers"
Journal of Economic Perspectives, Vol. 9, No. 3, 57-80.
Yabuuchi, Shiggemi, and Sarbajit Chaudhuri (2007),
"International Migration of Labor and Skilled-unskilled Wage
Inequality in Developing Economy" Economic Modelling, Vol. 24,
128-137.
Notes
(1.) This study acknowledges to Prof. A.K.Rao, Prof. J. M.Reddy,
Prof. V.N.Reddy for their valuable suggestions and finally Ms. Kavita
Joshi for her kind help.
(2.) Increase in poverty, income and wealth inequality, wage
inequality (measured in terms of either education qualification or
occupational classification), and gender wage pay gaps etc.
(3.) Value of India's export and import has increased
considerably over the period of planning. Export has increased from
$1,269 million in 1950-51 to $ 8, 486 million in 1980-81 and further to
$1, 55, 512 million in 2007-08. Import during the same period rose from
$1,273 million to $15,869 and further to $2, 35, 911 million.
India's trade with OECD was 78.0% in 1960-61 of the total trade
which came down to 40.1% in 2006-07.
(4.) FDI inflow has increased from $129 million in 1991-92 to $7,
722 million in 2005-06.
(5.) Research and Development (R&D) intensity has increased to
0.33 percentage in 1996-97 from 0.05 in 1990-91 and the payment for
royalty and technical fees has increased from $ 25.1 million in 1985 to
$ 200.8 million in 1998 and to $ 350.4 million in 2002 (Rashmi Banga,
2005).
(6.) Robust estimator is one that is insensitive to violation of
any of the assumption made about the way in which the data is generated.
Technically speaking when there is presence of outliers, leverage
points, and influential observations we use robust estimators to check
whether the sign and significance of the coefficients of interest
variables are changing are not.
(7.) Results of multicollinearity are not being given here but it
can be obtained from the authors by special request.
AVIRAL KUMAR TIWARI, Management Research Scholar, ICFAI University,
Tripura, E-mail: aviral.eco@gmail.com & aviral.kr.tiwari@gmail.com
M. ARUNA, Faculty Member, IBS IFHE University, Hyderabad, E-mail:
maruna@icfaiuniversity.in
Table 1
Correlation Matrix
Correlation LRNVATGDP LRGFCFTNVA LRWUWTSW RFDITGDP RTTTGDP
LRNVATGDP 1.000 .431 .471 -.121 -.111
LRGFCFTNVA .431 1.000 .659 -.436 -.536
LRWUWTSW .471 .659 1.000 -.781 -.887
RFDITGDP -.121 -.436 -.781 1.000 -.851
RTTTGDP -.111 -.536 -.887 -.851 1.000
Table 2
Total Variance Explained
Initial Eigenualues
Total % of Cumulative
Component Variance %
1 3.243 64.856 64.856
2 1.083 21.667 86.522
3 .466 9.316 95.839
4 .171 3.415 99.254
5 .037 .746 100.000
Extraction Sums of
Squared Loadings
Total % of Cumulative
Component Variance %
1 3.243 64.856 64.856
2 1.083 21.667 86.522
3
4
5
Rotation Sums of
Squared Loadings
Total % of Cumulative
Component Variance %
1 2.789 55.774 55.774
2 1.537 30.749 86.522
3
4
5
Extraction Method: Principal Component Analysis.
Table 3
Rotated Component Matrix
Component
1 2
LRNVATGDP -0.004 0.948
LRGFCFTNVA -0.518 0.635
LRWUWTSW -0.847 0.471
RFDITGDP 0.931 -0.052
RTTTGDP 0.968 -0.104
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
(a). Rotation converged in 3 iterations.
Table 4(1)
Model Summary
Model Summary (b)
Model R R Adjusted Std. Error of Durbin-Watson
Square R Square the Estimate
1 .969 (a) 0.939 0.933 0.01251 1.905
(a). Predictors: (Constant), score of external factor score of
domestic factor
(b). Dependent Variable: SUWW
Table 4(2)
ANOVA
ANOVA (b)
Sum of Mean
Model Squares df Square F Sig.
1 Regression 0.055 2 0.028 176.193 .000 (a)
Residual 0.004 23 0.000
Total 0.059 25
(a). Predictors: (Constant), score of external, factor score of
domestic factor
(b). Dependent Variable: SL'WW
Table 5
Regression Results Coefficients
Regressions Results Coefficients (a)
Model Unstandardized Standardized
Coefficients Coefficients
B Std. Error Beta t Sig.
1 (Constant) .603 .002 245.951 .000
score of
external factor -.041 .003 -.844 -16.360 .000
score for
Domestic factor .023 .003 .475 9.206 .000
Regressions Results
Coefficients (a)
Model Collinearity
Statistics
Tolerance VIF
1 (Constant)
score of
external factor 1.000 1.000
score for
Domestic factor 1.000 1.000
(a). Dependent Variable: SUWW
Table 6
Correlation Matrix
Correlation LRNVATGDP LRGFCFTNVA LRWUWTSW RFDITGDP LRTTDCs
TGDP
LRNVATGDP 1.000 .431 .471 -.121 .335
LRGFCFTNVA .431 1.000 .659 -.436 -.268
LRWUWTSW .471 .659 1.000 -.781 -.568
RFDITGDP -.121 -.436 -.781 1.000 .745
LRTTDCsTGDP -.335 -.268 -.568 .745 1.000
Table 7
Total Variance Explained
Initial Eigenualues
Component Total % of Cumulative
Variance %
1 2.832 56.650 56.856
2 1.473 29.457 86.107
3 .466 9.314 95.421
4 .153 3.068 98.489
5 .076 1.511 100.000
Extraction Sums of
Squared Loadings
Component Total % of Cumulative
Variance %
1 2.832 56.650 56.650
2 1.473 29.457 86.107
3
4
5
Rotation Sums of
Squared Loadings
Component Total % of Cumulative
Variance %
1 2.430 48.605 48.605
2 1.875 37.502 86.102
3
4
5
Extraction Method: Principal Component Analysis.
Table 8
Rotated Component Matrix (a)
Component
1 2
LRNVATGDP .191 .933
LRGFCFTNVA -.417 .710
LRWUWTSW -.719 .633
RFDITGDP .890 -.255
LRTTDCsTGDP .954 .188
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
(a). Rotation converged in 3 iterations.
Table 9(1)
Model Summary
Model Summary (b)
R Adjusted Std. Error of Durbin-
Model R Square R Square the Estimate Watson
1 .958 (a) .918 .910 .01450 1.811
(a). Predictors: (Constant), score of external, factor score of
domestic factor
(b). Dependent Variable: SUWW
Table 9(2)
ANOVA
ANOVA (b)
Sum of df
Model Squares Mean Square F Sig.
1 Regression .054 2 .027 128.047 .000 (a)
Residual .005 23 .000
Total .059 25
(a). Predictors: (Constant), score of domestic factor, score of
external factor
(b). Dependent Variable: SUWW
Table 10
Regression Results Coefficients (a)
Model Unstandardized Coefficients
Coefficients
B Std. Error Beta t Sig.
1 (Constant) .603 .003 212.073 .000
score of
external factor -.035 .003 -.716 -11.961 .000
score for
domestic factor .031 .003 .636 10.632 .000
Model Collinearity
Statistics
Tolerance VIF
1 (Constant)
score of
external factor 1.000 1.000
score for
domestic factor 1.000 1.000
(A). Dependent Variable: SUWW
Table 11
Correlation Matrix
Correlation LRNVATGDP LRGFCFTNVA LRWUWTSW RFDITGDP
LRNVATGDP 1.000 .431 .471 -.121
LRGFCFTNVA .431 1.000 .659 -.436
LRWUWTSW .471 .659 1.000 -.781
RFDITGDP -.121 -.436 -.781 1.000
LRTTLDCsTGDP -.119 -.573 -.889 .876
Correlation LRTTDCsTGDP
LRNVATGDP -.119
LRGFCFTNVA -.573
LRWUWTSW -.889
RFDITGDP .876
LRTTLDCsTGDP 1.000
Table 12
Total Variance Explained
Component Initial Eigenvalues
Total % of Cumulative
Variance %
1 3.258 65.160 65.160
2 1.085 21.693 86.852
3 .467 9.337 96.189
4 .154 3.083 99.272
5 .036 .728 100.000
Component Extraction Sums of
Squared Loadings
Total % of Cumulative
Variance %
1 3.258 65.160 65.160
2 1.085 21.693 86.852
3
4
5
Component Rotation Sums of
Squared Loadings
Total % of Cumulative
Variance %
1 2.798 55.955 55.955
2 1.545 30.898 86.852
3
4
5
Extraction Method: Principal Component Analysis.
Table 13
Rotated Component Matrix (a)
Component
1 2
LRNVATGDP -0.004 0.946
LRGF CFTNVA -0.512 0.640
LRWUWTSW -0.843 0.475
RFDITGDP 0.939 -0.052
LRTTLDCsTGDP 0.972 -0.111
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
(a). Rotation converged in 3 iterations.
Table 14(1)
Model Summary
Model Summary (b)
Model R R Square Adjusted Std. Error of Durbin-
R Square the Estimate Watson
1 .967 (a) .935 .929 .01287 1.729
(a). Predictors: (Constant), score of external, factor score of
domestic factor
(b). Dependent Variable: SUWW
Table 14(2)
ANOVA (b)
Sum of Mean
Model Squares df Square F Sig.
1 Regression .055 2 .027 165.689 .000 (a)
Residual .004 23 .000
Total .059 25
(a). Predictors: (Constant), score of external, factor score of
domestic factor
(b). Dependent Variable: SUWW
Table 15
Regression Results Coefficients (a)
Model Unstandardized Standardized
Coefficients Coefficients
B Std. Error Beta t Sig.
1 (Constant) .603 .003 238.970 .000
Factor score of
external factor -.041 .003 -.840 -15.807 .000
Factor score for
internal factor .023 .003 .480 9.029 .000
Model Collinearity
Statistics
Tolerance VIF
1 (Constant)
Factor score of
external factor 1.000 1.000
Factor score for
internal factor 1.000 1.000
(a). Dependent Variable: SUWW
