Modern Economy, 2011, 2, 340-343
doi:10.4236/me.2011.23037 Published Online July 2011 (http://www.SciRP.org/journal/me)
A Note on Wage Inequality, Technology, and Trade
Chu-Ping Lo
Department of Agricultural Economics, National Ta iwan University, Chinese Taibei
Received February 24, 2011; revised April 5, 2011; accepted April 25, 2011
Abstract
Zeira (2007) presents a two-country model of endogenous technology and trade, illustrating that trade liber-
alization reduces wage inequality in developing countries. The result contrasts the current outsourcing trade
literature; the conflict is due to the critical assumption made in his model that “the most rewarding technolo-
gies are invested first.” If we relax this assumption, or allow the technology frontier to foster labor gains in
all existing industries, then Zeira’s model is, in fact, consistent with the current outsourcing trade literature.
Keywords: Wage Inequality, International Outsourcing Trade
1. Introduction
In a two-country model, Zeira (2007) shows that trade
liberalization reduces wage inequality in a developing
country bur increases wage inequality in a developed
country. The technological progress that occurs in the
developed country generates a skill-biased technology
change in the developed country, but no impact on the
labor market in the developing country. The trade liber-
alization mentioned in Zeira’s model can be referred to
as international outsourcing trade, because an increase in
trade liberalization leads to greater trade in intermediate
goods in his model. However, in the literature on inter-
national outsourcing trade1, an increase in international
outsourcing trade or technology development leads to a
widening wage inequality not only in the developed
countries but also in the less deve loped countries.
I argue that this conflict between Zeira’s (2007) model
and the current outsourcing trade literature in terms of
the labor market in the less developed countries, may be
derived from the critical assumption in Zeira’s (2007)
model: “the most rewarding technologies are invested
first” and that the technology frontier is irrelevant to the
developing country. If we allow “the most rewarding
technologies are inv ested lately” or allow the technolog y
frontier to foster labor gains in all existing industries,
then the wage inequality o f the less developed country is
likely to increase with trade ex pansion.
The next section provides an introduction to Zeira’s
(2007) model. However, I redefin e the relative labor gain
in his model and reach a new equilibrium in Section 3.
The final section concludes.
2. The Zeira Model
Zeira (2007) assumes that there is one final good pro-
duced for consumption that is not tradable. This final
good is produced by a continuum of tradable intermedi-
ate goods. The tradable intermediate goods are produ ced
using two alternative technologies: either unskilled labor
or skilled labor. No intermediate good is prod uced with a
combination of skilled and unskilled labor. The devel-
oped and less developed countries trade the intermediate
goods with each other based on their comparative ad-
vantages.
More specifically, the set of traded intermediate goods
is t
M
, which is distributed uniformly over t
M
, meas-
ured for trade openness as t
tM. Technology
adoption, human capital acquisition, and trade patterns
are determined endogenously. The production of one unit
of intermediate good using skilled technologies re-
quires
dmi
i


s
ia units of skilled labor in the developed
country and ()
s
i in the less developed country.
Zeira (2007) also specifies a variable
f
to represent
the technology frontier, which measures the level of
technological progress available to a range of intermedi-
ate goods


0,
f
. He assumes that the populations of the
developed and less developed countries are given exo-
genously by and
A
L
B
L, respectively. The developed
1See Feenstra and Hanson (196), Antràs et al. (2006), and Ethier (2005)
for theoretical analysis, and Epifani and Gancia (2008), Attanasio et al.
(2004), Beyer et al. (1999), and Feenstra and Hanson (1997) for em-
p
irical analysis. All of these studies found that openness exacerbates the
wage inequality between skilled and unskilled labor in less developed
countries.

C.-P LO341
country has a larger share of skilled labor in the popula-
tion than the less developed country as AB
. Using
primitive technologies, the production of one unit of in-
termediate good , requires
hh
i()ni a unit of skilled la-
bor in the developed country and in the less de-
veloped country. The relative ga in of adopting the skilled
()ni


technology is given by

1
ni
gi si , but at the cost


of paying a skill premium for the skilled labor. A key
assumption in Zeira’s model is that the most rewarding
technologies are invented first. Here, the technologies
with higher relative gains are researched and invented
first, such that . This critical assumption im-
plies that the frontier technology has the smallest relative
labor gain in replacing unskilled with skilled labor,
namely,

0

gi


g
fgi for all 0. if
zv

In equilibrium with full specialization, there is a trade
threshold . We always have B to indi-
cate that the less developed country cannot access the
frontier technology. The developed country is exporting
the set of intermediate goods
v1f



0, 0,
M
fv, all of
which are produced by the skilled labor. The less devel-
oped country is expor ting intermediate goods


,1
M
v,
which are produced by unskilled labor as shown in Fig-
ure 1, which illustrates the equilibrium conditions for
labor markets while ,,SA SB
wa. Note that the less
developed country produces a set of nontradable inter-
mediate goods
w


0,
C
B
M
z by applying skilled labor
for domest i c use.
3. Equilibrium
Zeira (2007) assumes that the developed country is suffi-
ciently more skill abundant and has a lower wage ine-
quality than the less developed country. As illustrated in
Figure 1. Equilibrium with Full Specialization. Note: A
denotes the developed country and B denotes the less de-
veloped country.
Figure 1, the equilibrium conditions for skilled and un-
skilled labor in country A (i.e., the developed country) is
given by

 

 

1
0,
1
0,
d
d
C
AAA B
vM
A
fM
Lha siXiXii
asiXi i









and



1
,1
1d
C AA A
fM
Lh aniXi





i
,
respectively. Similarly, the equilibrium in the market for
skilled and unskilled labor in country B (i.e., developing
country) are g iven by
 

0, d
C
B
BB B
zM
Lhsi Xii


and



 

 

,1
,1
1d
d
C
B
BBB A
vM
B
zM
LhniXi Xii
niXi i


 



,
respectively. From the equilibrium conditions of these
labor markets, we can derive the wage inequality in the
two countries2. Therefore, in the benchmark case of full
specialization in Zeira’s model, the wage inequality in
the developed country is

 

11 1
1
11
A
AA
mf
h
Whmf
 


, (1)
which increases with an increase in trade liberalization
. The wage inequality in the less developed country is
then determined by
m
 

1
1
11
B
B
BBBB
zm
h
Wgz hzm


 
. (2)
It is the developed country that determines the tech-
nology frontier, and the creation of new technology is
without cost.
The key assumption in reducing relative labor gains
(i.e. () 0gi


) in Zeira’s (2 007) model is represented by
the downward sloping curve G in Figure 2, while A
and W
B
Wm are upward sloping. As in (1) and (2), an in-
crease in shifts A
W curve upward and
B
W curve
downward, leading to a higher wage inequality in coun-
try A but a lower wage inequality in country B. Specifi-
cally,
B
W must reduce with while m()gi0


. It
also shows that the development of the frontier technol-
ogy increases the wage inequality in the developed coun-
try, but has no impact on the wage inequality in the de-
veloping country.
However, if we relax this critical assumption of “the
2See Appendix 1 in Zeira (2007) for more details.

C.-P. LO 342
Figure 2. Equilibrium in Zeira’s (2007) model.
most rewarding technologies are invested first” and al-
low the technology with higher relative gains to be in-
vested into later, then the G curve becomes an upward
sloping curve while for . This al-
ternative assumption allows Zeira’s (2007) model to
correspond with the current literature on outsourcing
trade; specifically, that the wage inequality in both the
developed and developing country increases with trade
liberalization as shown in Figure 3. Nevertheless, tech-
nology development is irrelevant to the labor market in
the developing country.
() 0gi
0if
E ven if it stands to reason that “the most rewarding
technologies are invested first,” I argue that Zeira’s (2007)
model may still lead to an increase in wage inequality in
the less developed country if the trade expansion is due
to technology development. Throughout human history,
the development of frontier technology usually leads to
Figure 3 Equilibrium when
skill-biase proves productivity
began with
th
of the
In
()0gi.
d technology changes and im
in all other industries. Evidence for this exists in the
following historic events: the Industrial Revolution that
occurred during the late 18th and early 19th centuries
and the evolution of the Internet and personal computers
that took place during the late 20th century.
The Industrial Revolution in Great Britain
e mechanization of the textile industry through the
utilization of steam power, which transferred the primarily
manual labor and draft animal-based economy towards
machine-based manufacturing. This resulted in a dramatic
increase in productivity capacity and spurred the
manufacture of increasingly productive machinery for
use in other industries (Meier and Rauch, 2000).
Furthermore, the revolutionary development
ternet and personal computers pushed machine-based
manufacturing towards computation-based manufacturing,
thus generating more skill-biased technology changes
and stimulating productivity improvements in all other
industries. While the relative labor gains in the existing
industries should increase with the development of the
technology frontier, I redefine Zeira’s (2007) relative labor
gain as


,
g
if , where

,0
i
gif but


,0
f
gif.
In this way, te wage inthe le
country is then determined by

h equality in ss developed


1
1B h
,11
B
BB BB
zm
Wgzf hzm



, (3)
where


,0
B
zB
gzf

but .
assume that te barriers are ced by ei-
th
nsion is improved by the reduction of po-
lit

,0
fB
gzf
Let’s he tradredu
er the removal of political barriers or the improvement
of technologies. Using the Industrial Revolution as an
example again, the introduction of steam power, fuelled
primarily by coal, expanded world trade enormously by
providing a quick and easy way to transport goods, as
well as an easy way to transport mail and information
through the wide utilization of steam-powered trains and
ships (Meier and Rauch, 2000). The innovation prompted
by the personal computer and the Internet are fostering
more efficient ways of trade and communication as well.
Thus, the development of the technology frontier ex-
pands trade.
If trade expa
ical barriers, as in Zeira’s (2007) model, the wage ine-
quality in the developing country decreases with trade as
represented by the equilibrium b in Figure 4. However,
in contrast to Zeira’s model, if trade expansion is in-
duced by the development of technology frontier, I argue
that wage inequality in the developing country may in-
crease with trade if the skill-biased technology change is
sufficiently large. As shown in Figure 4, an increase in
f
, which shifts the G curve upward, also induces trade
expansion, thus shifting the
B
W curve downward. The

C.-P LO 343
5. Reference
[1] P. Antràs, L. Garicano and E. Rossi-Hansberg, “Offshor-
ing in a Knowledge Economy,” Quarterly Journal of
Economics, Vol. 121, No. 1, 2006, pp. 31-77.
[2] O. Attanasio, P. K. Goldberg and N. Pavcnik, “Trade
Reforms and Wage Inequality in Colombia,” Journal of
Development Economics, Vol. 74, No. 2, 2004, pp. 331-
366. doi:10.1016/j.jdeveco.2003.07.001
[3] H. Beyer, P. Rojas and R. Vergara, “Trade Liberalization
and Wage Inequality ,” Journal of Development E conomics,
Vol. 59, No. 1, 1999, pp. 103-123.
doi:10.1016/S0304-3878(99)00007-3
[4] P. Epifani and G. Gino, “The Skill Bias of World Trade,”
Economic Journal, Vol. 118, No. 530, 2008, pp. 927-960.
doi:10.1111/j.1468-0297.2008.02156.x
[5] W. J. Ethier, “Globalisation: Trade, Technology, and
Wages,” International Review of Economics and Finance,
Vol. 14, No. 3, 2005, pp. 237-258.
doi:10.1016/j.iref.2004.12.001
Figure 4. New Equilibrium.
net effect may, which indi-
ates a higher wage inequality for the developing country
ue that Zeira’s (2007) model corre-
sponds with the current literature of outsourcing trade
lead to a new equilibrium c
c [6] R. C. Feenstra and G. H. Hanson, “Foreign Investment,
Outsourcing and Relative Wages,” The Political Eco-
nomics of Trade Policy: Papers in Honor of Jagdish
Bhagwati, MIT P ress, Cam bridg e, 1996, pp. 89-127.
than in the initial equilibrium a.
4. Conclusions [7] R. C. Feenstra and G. H. Hanson, “Foreign Direct In-
vestment and Relative Wages: Evidence from Mexico’s
Maquiladoras,” Journal of International Economics, Vol.
42, No. 3-4, 1997, pp. 371-393.
In this paper, I arg
(e.g., Feenstra and Hanson, 1996) if we allow the tech-
nology with higher relative gains is to be invested into
later. This paper also shows that development in the
technology frontier in the developed country induces
skill-biased technology change in the labor markets in
the developing country through trade.
[8] G. M. Meier and J. E. Rauch, “Leading Issues in Eco-
nomic Development,” Oxford University Press, New
York and Oxford, 2000.
[9] J. Zeira, “Wage Inequality, Technology, and Trade,”
Journal of Economic Theory, Vol. 137, No. 1, 2007, pp.
79-103. doi:10.1016/j.jet.2006.03.011