The Impact of Offshore Production on Within-Industry Reallocations and
Aggregate Industry Productivity
Lo, Chu-Ping
Abstract
This paper extends Melitz’s (2003) dynamic industry model by incorporating Antràs and Helpman’s (2004) outsourcing structure to show that a within-industry selection of
heterogeneous firms can be driven by offshoring. In a North-South world, offshoring makes relatively more productive firms in the North able to leverage foreign resources to expand their market shares at the expense of the relatively less productive firms. This endogenous selection enables an economy to transit to high-productivity varieties, thereby raising the aggregate industry productivity. In particular, offshore production increases the number of product varieties in offshore-intensive industries but reduces that in less offshore-intensive industries.
Keywords: offshore production, firm heterogeneity, selection. JEL Classification: F12, F14, F23
1. Introduction
In recent decades, due to significant developments in communications, transportation,
* Contact information: Chu-Ping Lo. E-Mail: cplo@ntu.edu.tw. Phone: (886)2-3366-2653. Department of Agricultural Economics, National Taiwan University, Taipei, Taiwan.
and computer technology, there has been an acceleration in offshore production. By
segmenting production into many stages, a firm can outsource some of these stages to regions with comparative advantages. As a result, a new pattern of trade, the so-called vertical
specialization trade or international outsourcing trade has emerged from the offshore production (Hummels, Ishii, and Yi, 2001).1
While it has been well documented that open trade improves firm productivity (e.g., Head and Ries 2003, Bernard et al. 2003, and Melitz 2003),2 recent empirical research has
shown that offshore production has also a significant positive effect on the measured
productivity of firms. For example, Amiti and Wei (2009) examined firm-level data for U.S. manufacturing industries between 1992 and 2000,3 Görg et al. (2008) examined panel-level
data for Irish manufacturing over the period 1990-98, and Egger and Egger (2006) examined the productivity of unskilled workers in EU manufacturing over the period 1992-97. All these studies provide some support for this assertion. In particular, Mann (2003) reported that globalized production and trade made information technology hardware 10 to 30 percent cheaper than it would have been otherwise, and the price decrease translated into higher productivity growth in the United States. Theoretically, Jones and Kierkowski (2001) developed a Hecksher-Ohlin framework with two industries and two factors in which the 1 Using Input-Output tables for OECD as well as some emerging market countries, Hummels, Ishii, and Yi (2001) documented that vertical specialization (through offshore production) accounted for approximately 30% of world exports, growing by more than 30% between 1970 and 1990.
2 In particular, using a panel of half a million firms in 1996-2007, Bloom et al. (2010) founded that an increase in import competition with China due to an open trade policy has caused a significant technological upgrading in European firms through both faster diffusion and innovation, leading to welfare gains.
3 Among others, Feenstra and Hanson (1999), Siegel and Griliches (1992) and Slaughter (2000), who examined the impact of outsourcing on U.S. labor productivity, all found evidence of a positive effect.
production process can be fragmented into two intermediate goods and then performed in different locations. They argued that the possibility of performing these intermediate goods in different locations (i.e., offshore production) can improve the measured productivity of firms.
Another strand of the literature shows that performing the stages in different locations also has a significant impact on the identity of the industry. Antràs and Rossi-Hansberg (2009) have presented a good survey on how offshore production can not only lower factor prices due to trade liberalization or reductions in trade costs (Grossman and Rossi-Hansberg 2008, Rodríguez-Clare 2010) and then improve the productivity of firms (Jones and
Kierkowski, 2001), but can also raise the aggregate productivity of industry (Dixit and Grossman 1982, Baldwin and Robert-Nicoud 2007, Grossman and Rossi-Hansberg 2008).
Specifically, Grossman and Rossi-Hansberg (2008) have introduced offshore production into a Heckscher-Ohlin model, where there are two industries in perfect competition and each industry produces one good. In their model, each good is assembled by a continuum of high and low-skilled tasks where the low-skilled tasks can be performed abroad through offshore production. One industry is labor-intensive while the other is skill-intensive if the latter involves a higher intensity of high-skilled tasks. Their model presents a positive productivity effect of offshoring that benefits the low-skilled labor only, so that the labor-intensive industry enjoys a greater increase in productivity because it is more easily moved offshore. However, their model mostly focuses on the effects of offshoring on factor prices while the
within-industry effect is not their concern. They also place less emphasis on the welfare gains while the product variety is fixed (with two varieties) in their model.
Melitz (2003) introduces heterogeneous firms to Krugman’s (1979) monopolistic
competition model (with a representative firm) and allows a Hopenhayn (1992a, 1992b) type of endogenous selection of heterogeneous firms within an industry. In Melitz’s model, only the more productive firms are able to export and then to expand their market shares, while the least productive firms are forced to exit the market after exposure to trade. As a result, the consequent intra-industry trade leads to intra-firm reallocations within an industry, and the selection favors the high-productivity firms and then raises the average industry productivity.
Alternatively, this current paper provides an in-depth discussion of how the endogenous selection of heterogeneous firms within an industry may be driven by offshoring. Drawing heavily from Melitz’s (2003) model, this paper tries to build up a different channel through which offshore production increases productivity and product variety in a monopolistic competition model. While a foreign country has a comparative advantage in some stages of production, offshoring these stages of production enables the domestic high productive firms to free up resources to expand their market shares and revenues. On the other hand, the expanding market shares and increasing revenues of the highly-productive firms may induce the creation of more highly-productive varieties. As a result, offshore production can generate a within-industry reallocation of resources favoring the firms with high productivity because
only such firms are able to overcome the cost of offshoring to leverage the foreign resources. It turns out that the within-industry selection, which drives the disproportionate increase in the highly-productive firms, leads to a productivity gain while the average industry productivity improves with the expansion of the highly-productive firms.
Although, being able to leverage foreign resource, the offshore production incurs a greater organizational cost than domestic production. The advantage of leveraging foreign resources increases with the extent of the offshore production that has been undertaken. As a result, for the relatively offshore-intensive industries, the advantage of leveraging foreign resources exceeds the disadvantage of higher organizational cost, leading to more product varieties in these industries. On the contrary, for the relatively less offshore-intensive
industries, the disadvantage of higher organizational cost exceeds the advantage of leveraging foreign resources, leading to less product varieties.
The remainder of this paper is organized as follows. In Section 2, a simple model is presented, illustrating how the endogenous selection of heterogeneous firms within an industry may be driven by offshore production. Section 3 concludes.
2. The Model
Assume that in the North-South world, there are J+1 industries which produce a homogeneous good and J continua of differentiated products. Labor is the only production factor in the world. Consumers in the North and South have identical and homothetic preferences, which can be expressed as follows:
0 1 1 J u j j U q Q u
, 0 u 1, (1)where q represents the demand for the homogeneous good, 0 Qj is an index of aggregate
demand in sector j , and u is a parameter representing the degree of substitution across differentiated-product industries. The homogeneous good is used as a numeraire, which can be freely produced and traded in both countries. The freely-traded homogeneous good pins
down the relative wage between the North and the South. Let w and N w denote the wage S
rate of the North and South, respectively. Assume that the North is more productive in
producing the homogeneous goods q and that the labor supply is large enough in each 0
country so that the homogeneous goods q remain active in both countries after free trade. 0
We can then justify a general equilibrium in which w and N w are fixed. S
The aggregate demand is a constant elasticity of substitution function of the demand for different varieties ( )q i :j [ ( )] ]1/ j i j Q q i di
, 0 . (2) 1The elasticity of substitution between any two differentiated goods is given by 1 1 1
Assume that u such that varieties are more substitutable within an industry than across industries (Antràs and Helpman, 2004).
The aggregate demand Qj in (2) is associated with the aggregate price index:
( ( ) 1 ) 1 j i j P p i di
. (3)As implied in the utility function (1), the consumers will spend u1 u j
P on the differentiated goods of the industry j. As for individual variety i in industry j, the optimal consumption and expenditure decisions are respectively given by
j( ) j( j( )) j p i q i Q P and ( ) ( j( ))1 j j j p i r i R P , (4)
where Rj represents the aggregate revenue of industry j.
Each variety is exclusively produced by one firm. However, the production of any final-good variety requires a costless combination of two variety-specific inputs, ( )h i and ( )m i , which are associated with high-tech and low-tech inputs, respectively. The production
function of firm i with a productivity level is given by i
( ) ( ( ))1 ( ( )) 1 j j z z j j j i j j h i m i q i z z , (5)
relatively offshore-intensive while a smaller zj is relatively less offshore-intensive. The
high-tech inputs must be provided by the Northern firms while the low-high-tech inputs can be produced in either the North or the South. The labor requirement of producing one unit low-tech input
( )
j
m i is normalized as one in both the North and the South. For the purpose of simplicity, we drop the industry index j from all of the variables hereafter.
2.1 Autarky
There is a large pool of prospective firms that enter the industry. The firms, which are identical before entry, must first make an initial investment, f (measured in units of the 0
northern labor). Each differentiated firm incurs the sunk entry cost f to perform research and development and marketing as well in order to create a firm brand and a blueprint for one variety of product (Bernard et al., 2010), thereby differentiating the firm from the others.
Next, firms draw their productivity from a common distribution ( )g , where ( ) (0, )
g . This product distribution has a cumulative distribution ( )G . Upon observing its productivity level, each firm decides to produce or exit the market. In monopolistic competition, each firm faces a demand curve with a constant elasticity of demand, and then the profit maximization leads to the price of a representative variety: ( )p wN
The revenue of a firm with productivity is then given by 1 1
( ) ( N )
r R P w . (7) We can derive a firm’s profit as
( ) r( ) w h w m w fN N N d r( ) w fN d
, (8)
where fd represents a fixed per period organization cost occurring in domestic production.
If choosing to produce, a firm may come across a spot of bad luck and then exit the market. In every time period, the possibility of encountering bad luck is , where 1 0. For simplicity, let the probability of encountering bad luck is constant across time and productivity as well (Melitz, 2003).4
The present value of the profit flows is given by
0 ( ) ( ) (1 )t N d t w f r v
for afirm with productivity . The net present value of entry for this firm is given by
( ) ( ) N d N N w f r v w f w f
. There exists a productivity cutoff to satisfy a zero net *
profit as *
( ) N d N
r w f w f . On the other hand, the expected net value for an entry is
ve 1 G( ) * ( ) w fN , (9) where * * ( ) ( ) ( ) 1 ( ) g d G
represents the weighted average profit. There exists a free 4 It is likely that the probability of encountering the bad luck is a country-specific parameter and is highly determined by an economy’s business environment. In general, a better business environment implies a lower probability of bad luck.entry condition such that 1 ( )* ( )* N G w f
. The productivity cutoff condition coincides
with the free entry condition to determine the equilibrium cutoff .* 5
Obviously, a larger
entry cost f leads to a higher productivity cutoff . Firms with productivity below this *
cutoff exit the industry and firms with productivity greater than * generate positive *
profits.
By letting M denote the mass of producing firms in industry j in a closed economy, we
then obtain the average profit as ( ) r( ) w fN d R w fN d M
. By combining the free
entry condition in (9), we obtain a stationary equilibrium ( ) *
1 ( ) N N d w f R w f G M .
Then, by rewriting the equilibrium, the mass of producing firms is given by
* * [1 ( )] [ (1 ( )) ] N d G R M w f G f . (10)
The mass of producing firms M in (10) also represents the aggregate variety of differentiated
5 Free entry requires that 0
e
v since the mass of prospective entrants is unbounded. Furthermore, no firm would want to enter the market when the expected net value is negative (Hopenhayn, 1992a, 1992b). In equilibrium, we reach ve , implying 0 ( )* *
1 ( ) N w f G
. As shown in Melitz (2003, Figure 1), the free entry condition and zero net profit condition determine the equilibrium .*
goods in a specific industry. It is interesting to note that the mass of producing firms, ceteris paribus, in any period of time is increasing with a lower possibility of bad luck. Apparently, an economy that is endowed with a better business environment is able to reduce the
possibility of bad luck, and then in turn to cause the mass of producing firms to increase. In the following, let’s rewrite the aggregate price index in (4) as
* 1 1 1 * ( ) [ ( ) ] 1 ( ) g P p M d G
. (11)Melitz (2003) defines a weighted average of the firm’s productivity level as * 1 1 1 * ( ) [ ] 1 ( ) g d G
. (12)The average industry productivity level summarizes the effects of the distribution of the heterogeneous firms’ productivity levels on the aggregate outcome. Therefore, (12) also represents the aggregate industry productivity. We can then rewrite (11) by incorporating (6)
and (12) as 1 1 1 ( ) 1 wN P M p M
. The industry aggregate price declines with aggregate product variety and aggregate industry productivity as well.
2.2 The Offshore Production
Suppose there are no trading costs involved in the exchange of the intermediate and final
goods in the North-South world. Nevertheless, it is well known that incomplete contract distortion discourages offshore production (Antràs and Helpman, 2004). Technological
barriers may also prohibit offshore production from the North to the South.6 Suppose that
there exists substantial incomplete contract distortion or technological barriers such that offshore production is prohibited in the first stage. Suppose further that there is significant improvement in the contract enforcement of the South and also innovative advances in communications, transportation, and computation technology, making offshore production plausible. In such circumstances, the North starts to carry out offshore production in the South and export final goods to the South in exchange for the intermediate goods produced in the South. The freely-traded homogeneous goods balanced the trade.
As in Melitz’s (2003) model, a firm decides on whether to offshore production or not after the firm’s productivity is observed. Either way, firms face the same ex ante distribution of productivity levels ( )g and probability of encountering bad luck. Although no trading costs are involved in the exchange of the intermediate and final goods, an organizational cost
of offshoring, fo , is periodically incurred by offshore production and is denominated in 0
terms of the northern labor. The subscript o denotes offshore production. We have fo fd
while the organizational costs are often higher when organizing production across national boundaries than within a country’s boundary (Antràs and Helpman, 2004).
6 Grossman and Rossi-Hansberg (2008) presume the existence of technological barriers against the offshore production, and the cost of offshoring is reduced with technological improvement. An innovation in technology (e.g., Internet) may effectively reduce the cost of offshoring from infinity to virtually nothing.
The profit function for the firms that engage in offshoring is given by ( ) o( ) o N o r w f . (13)
There exists a productivity cutoff to satisfy o ro( )o w fN o. With fo fd, we obtain
*
o
. Here, 1 1
( ) ( z)
o o o N
r R Pw . In comparison with autarky, the aggregate outcome
after exposure to offshore production are R , o Q , and o P , respectively, and o Ro P Qo o.
While some firms with relatively higher productivity find it beneficial to relocate their low-tech inputs to the South, the prices of their products become
1 z z N S N o z w w w p , (14)
where p zo( ) 0 . Here, let’s presume that the wage gap between the North and South
1
N S
w w
is sufficiently large in order to motivate the offshore production. Having 1, firms that are engaged in offshore production charge a lower price than otherwise as
1
z o
p
p , where 0 z 1. This implies in (3) and (14) that the ratio of the aggregate
industry price P zo( )
and P zo( )P.
Some incumbent low-productive firms may not engage in offshore production because of the prohibitive cost of offshoring. These firms will therefore produce all their inputs
domestically even when the offshore production is an option. For these firms, as implied in
(7) and (8), there exists a cut-off productivity *
d
to satisfy * * 1 1
( ) ( )
o d o o d N N d
r R P w w f
. We can calculate the cut-off productivity as
1 1 1 * N ( ) 1 1( ) 1 d N d o o w w f P R . In (7), the
cut-off productivity in autarky is given by
1 1 1 * N ( ) 1 1( ) 1 N d w w f P R
. Let’s compare the
productivity cutoffs and then obtain a ratio as
1 * 1 ( 1)( 1) 1 * ( ) ( ) u u d o o o P R P P R P . With 1 1
, we can rearrange the ratio of productivity cut-offs as * ( 1) * ( ) 1 u u d Po P , (15)
where we already have 1 u 0 and P zo( )P. In (15), we obtain
* *
d
when
exposure to offshore production (i.e., z0). This implies that the least productive firms with
productivity along * *
[ , ) are forced out of the market after exposure to offshore production.d
The reason for this is because that the relatively more productive firms gain with price competition due to leveraging the foreign resources, and then drive the least productive firms
out of the market. While the least productive firms with productivity along * *
[ , ) exit the d
market, the relatively less productive firms with productivity levels along * 0
[ , ) are d
domestic production firms that engage in no offshore production, and the relatively more
productive firms with productivity along [ , ) carry out offshore production.o
2.3 The Product Variety
Suppose the equilibrium mass of incumbent firms in an industry after exposure to
industry undertaking offshore production is given by 1 ( )* 1 ( ) o d G G . We can assign * 1 ( ) 1 ( ) o o t d G M M G
to represent the mass of offshore production firms while using
* * ( ) ( ) 1 ( ) d o d t d G G M M G
to represent the mass of domestic production firms (without offshore
production) in the industry. To sum up, we have Mt MoMd .
The new aggregate price in (11) can be rewritten as
* 1 1 1 1 * ( ) ( ) [ ( ) ( ) ] ( ) ( ) 1 ( ) o d o o d o o d o o g g P p M d p M d G G G
1 1 ( N ) t z t w M , (17) where [ z( 1) 1 (1 z( 1)) 1] 11 t d o is defined as the aggregate industry
productivity after exposure to offshore production. Here, let
1 1 ( ) 1 [ ] 1 ( ) o o o g d G
and* 1 1 1 * ( ) [ ] 1 ( ) d d d g d G
denote the aggregate productivity of domestic firms and
offshore firms in the industry when exposure to offshore production, implying o d as
*
o d
. Therefore, t is a combined average industry of o and d. With o d, we obtain
o t d as * * o t d . With * * d , we then have o t d .
The aggregate productivity of not only the offshore production firms but also the
domestic production firms improves with offshoring as represented by t and d ,
respectively. That is, in this current model, the offshore production plays a role in driving a within-industry selection and then raising the aggregate industry productivity for both the
offshoring and domestic firms. Furthermore, we can take the derivative of t with respect to
z to get t 0 z
.7 In line with Grossman and Rossi-Hansberg (2008), this current model
implies that the offshore-intensive industries enjoy a greater increase in aggregate industry
7 We have 1 ( 1) 1 ( 1) ( 1) 1 ( 1) 1 1 1 1 ( 1) (ln ) ( ) ( ) 0 z z z z t t d o d o z z z because of 1 and o d.
productivity because they leverage more offshore production.
After exposure to offshore production, the expected net value for an entry in (9) can be rewritten as ˆ ( )d* ( )o ( ) 1 ( )o ( ) e d o N G G G v w f . (18)
Here, s is assumed to denote the share of total revenue that is generated by the domestic firms while 1 s is the share of revenue that is generated by the offshore production firms,
and 1 s 0. Therefore, we obtain * *
( ) ( ) ( ) 1 ( ) d o d N d d d sR g d w f G M
and (1 ) ( ) ( ) ( ) 1 ( ) o o o N o o o s R g d w f G M
. Free entry leads to ˆve . Rearranging (18) 0by plugging 1 ( )* 1 ( ) o o t d G M M G and * * ( ) ( ) 1 ( ) d o d t d G G M M G
, the aggregate mass of producing firms becomes
* * (1 ( )) ( ) [ ( ( ) ( )) (1 ( )) ] d o t N d d o o o G R z M w f G G f G f . (19)
Dividing (10) by (19) and incorporating (15), we obtain the ratio of product variety with and without offshore production as
* * 1 * * ( ) 1 ( ) ( ( ) ( )) (1 ( )) ( )( ) 1 ( ) ( ( ) ( )) (1 ( )) ( ) u t d o d o d u o d d o o o M z G G G f G f f P M G G G f G f f P z ,8 (20)
where 1 u 0. The ratio of product variety Mt
M is a function of the productivity cutoff
*
d
and and the offshore intensity * z
as well. Due to * *
d
, the first term on the right-hand side of (20) is less than one. The second term on the right-hand side of (20) is also smaller
than one because of * *
d
and fo fd. However, the third term is always greater than one
because of P P z o( ) and P zo( ) 0 as argued above.
We can then illustrate M zt( )
M with respect to z as in Figure 1, where 0 z 1. When
0
z , we have P P z o( 0)and fo fd. This implies that
( ) 1
t
M z
M because both of the first and second terms on the right-hand side of (20) are less than one and the third term is
8 Supposed that after exposure to offshore production, there exists an industry j with 0
j
z , such that the
advantages of offshoring are counterbalanced by the disadvantages of higher organizational cost and thereby no firms carry out offshore production in the industry j. On the contrary, in the other industry k with zk 0, offshore production is undertaken by the relatively-high productive firms in this industry. As implied in (1), we should obtain 1 1 u k k u j j Q P P Q
in equilibrium, which implies that ( ) 1
u k k k k u
j j j j
R P Q P
R P Q P . Here, the industry j is
analogous to the industry in autarky while the industry k offshoring, so that we obtain ( )1
u
o o o u
o
R P Q P
equal to one. When z1, we obtain
( 1)
o
P
P z . While having already presumed that the
wage gap between the North and South is sufficiently large, we have M zt( 1) 1 M
. Putting them together, there exists a cut-off of offshore intensity z*. For the industries with
offshore intensity 1 z z *, product variety increases with offshore production. However, for
the industries with offshore intensity z* z 0, product variety declines with offshore
production.
The difference in the number of product varieties after exposure to offshore production is
given by * 1*
0 ( ( )) ( ( ) )
z
t z t
M M z dz M z M dz
, which declines with a reduction in z*.Obviously, as implied in (20), the cut-off of offshore intensity z* decreases with , as
illustrated by the dash line in Figure 1. This implies that the aggregate product variety in an economy increases with the wage gap between the North and South.
Figure 1
The offshore production makes the relatively-high productive firms in the North be able to leverage the low wage costs in the South, such that more resources in the North are freed up to generate product varieties. However, the offshore production incurs a greater
organizational cost than domestic production, crowding out resources to generate product varieties. The advantage of leveraging foreign resources increases according to the extent of offshore production undertaken. Overall, for the offshore-intensive industries (i.e., 1 z z *),
the advantage of leveraging foreign resources dominates the disadvantage of higher
organizational cost, leading to more product varieties in these industries. By contrast, for the less offshore-intensive industries (i.e., 0 z z *), the disadvantage of higher organizational
cost dominates the advantage of leveraging foreign resources, leading to less product varieties in these industries. This is the main implication of this model.
2.4 The Reallocation of Resources
Finally, let’s examine the effect of offshore production on firms with different levels of
productivity. The market share of a firm in autarky is given by 1 1
( ) ( N )
r R Pw . When
exposed to offshore production, the least productive firms with productivity along * *
[ , ) d
exit the market. The firms with productive levels along *
[ , ) continue to produce d o
domestically. The market share of a firm within this group is as 1 1
( ) ( )
d o o N
r R P w ,
where *
[ , ]d o
. On the other hand, the firms with productivity levels along [ , )o engage
in offshore production, having market shares 1 1
( ) ( z)
o o o N
r R P w , where [ , )o .
For the firms with the lower productive levels along *
[ , ) , due to d o Po , we haveP 1 ( ) ( ) 1 ( ) d o o r R P r R P
. As implied in (7) and (13), we also have ( ) ( )(1 )(1 ) 1
( ) u u d o r P r P
market shares but also revenues in comparison to the situation where autarky prevails. On the
contrary, for the firms with higher productivity levels along [ , ]o , we have
1 ( ) ( ) 1 ( ) z o o o r R P r R P
when is sufficiently large.9 As implied in (7) and (13), we also
observe ( ) ( )(1 )(1 ) 1 ( )1 ( ) 1 ( ) u z z u u o o o o r P P P r P P P because (Po z) 1 1 P . This implies
that while the wage gap between the North and South is sufficiently large offshore production make the relatively more productive firms gain not only a larger market share but also a greater revenue due to their increased price competition after leveraging foreign resources.
3. Conclusions
As highlighted by Bhagwatti et al. (2004), the public debate on offshore production is muddled by confusions regarding whether offshoring is welfare and productivity improving as clarified by the conventional principles of international trade. In this paper, I have
presented a simple model to show that offshore production drives an endogenous selection of heterogeneous firms within an industry. The relatively more productive firms are able to
9 As implied in (3) and (6), the aggregate price
o
P is a weighted sum of ( )p wN
for the firms with productivity levels along [ ,* ]
d o plus ( ) N o z w p
for the firms with productivity levels along [ , ]
o
. We
have po( ) p( ) if is sufficiently large, which overshadows the impact of d**, such that z 1 o
P P .
leverage offshoring to expand their revenues, market shares, and product varieties, while the relatively less productive firms are deterred by the cost of offshoring. As a result, the
disproportionate increase in highly-productive firms gives a rise to the aggregate industry productivity. Furthermore, in line with Grossman and Rossi-Hansberg (2008), the offshore-intensive industries enjoy a greater increase in aggregate industry productivity because they leverage more offshore production.
In Melitz’s (2003) model, the impact of exposure to trade in the total number of product varieties is ambiguous. However, this paper shows that offshore production increases product varieties in offshore-intensive industries, but reduces them in less offshore-intensive
industries. Overall, the resource allocation due to exposure to offshoring gives rise to aggregate product varieties at lower prices if the wage gap between the North and South is sufficiently large, thereby improving the welfare of the economy.
REFERENCES
Amiti, M. and S-J. Wei. 2009. “Service Offshoring and Productivity: Evidence from the US.” The World Economy, Blackwell Publishing, 32(2): 203–220.
Antràs, P. and E. Helpman. 2004. “Global Sourcing.” Journal of Political Economy, 112: 552–580.
Antràs, P. and E. Rossi-Hansberg. 2009, “Organizations and Trade.” Annual Review of Economics, 1: 43–64.
Baldwin, R and F. Robert-Nicoud. 2007. “Offshoring: General Equilibrium Effects on Wages, Production and Trade.” NBER Working paper 12991.
Bernard, A. B., J. Eaton, and S. Kortum. 2003. “Plants and Productivity in International Trade.” American Economic Review, 93(4): 1268–1290.
Bernard, A. B., S. J. Redding, and P. K. Schott. 2010. “Multiple-Product Firms and Product Switching.” American Economic Review, 100(1): 70–97.
Bhagwati, J., A. Panagariya, and T. N. Srinivasan. 2004. “The Muddles over Outsourcing.” The Journal of Economic Perspectives, 18: 93–114.
Bloom, N., M. Draca, V. van Reenen. 2010. “Trade Induced Technical Change? The Impact of Chinese Imports on Innovation, IT and Productivity.” CEP Discussion Paper No 1000. Dixit, AK and G. M. Grossman. 1982. “Trade and Protection with Multistage Production.”
Review of Economic Studies, 49: 583–594.
Egger, H. and P. Egger. 2006. “International Outsourcing and the Productivity of Low-Skilled Labor in the EU.” Economic Inquiry, 44(1): 98–108.
Feenstra, R. C. and G. H. Hanson. 1996. “Foreign Investment, Outsourcing and Relative Wages.” The Political Economics of Trade Policy: Papers in Honor of Jagdish Bhagwati, MIT Press, 89–127.
Feenstra, R. C. and G. H. Hanson. 1999. “The Impact of Outsourcing and High-technology Capital on Wages: Estimates for the United States, 1979-1990.” Quarterly Journal of Economics, 114: 907–40.
Grossman, G. and E. Rossi-Hansberg. 2008. “Trading Tasks: A Simple Theory of Offshoring.” American Economic Review, 98 (5): 1978–1997.
Görg, H., A. Hanley and E. A. Strobl. 2008. “Productivity Effects of International
Outsourcing: Evidence from Plant Level Data.” Canadian Journal of Economics, 41 (2):
Head, K. and J. Ries. 2003. “Heterogeneity and the FDI versus Export Decision of Japanese Manufacturers.” Journal of the Japanese and International Economies, 17: 448–467. Hopenhayn, H. A. 1992a. “Entry, Exit, and Firm Dynamics in Long Run Equilibrium.”
Econometrica , 60 (5): 1127–1150.
Hopenhayn, H. A. 1992b. “Exit, Selection, and the Value of Firms.” Journal of Economic Dynamics and Control, 16 (3-4): 621–653.
Hummels, D., J. Ishii, and K. Yi. 2001. “The Nature and Growth of Vertical Specialization in World Trade.” Journal of International Economics, 54: 75–96.
Jones, Ronald W. and H. K. Kierzkowski. 2001. “Globalization and the Consequences of International Fragmentation.” In Money, Capital Mobility, and Trade: Essays in Honor of Robert A. Mundell,ed. Guillermo A. Calvo, Rudiger Dornbusch, and Maurice
Obstfeld, 365–83. Cambridge, MA: MIT Press.
Krugman, P. 1979. “Increasing Returns, Monopolistic Competition, and International Trade.” Journal of International Economics, 9: 469–479.
Mann, C. L. 2003. “Globalization of IT Service and White Collar Jobs: The Next Wave of Productivity Growth.” International Economics Policy Briefs, PB, 03–11.
Melitz, M.J. 2003. “The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity.” Econometrica, 71(6): 1695–1725.
Rodríguez-Clare, A. 2010. “Offshoring in a Ricardian World. ” American Economic Journal: Macroeconomics, American Economic Association, 2(2): 227–58.
Siegel, D. and Z. Griliches. 1992. “Purchased Services, Outsourcing, Computers, and Productivity in Manufacturing.” Output Measurement in the Service Sector, University of Chicago Press. Also as (1991) NBER Working Paper, n. 3678.
