Firm Responses to Trade: Winners, Losers,
and Industry Performance
In our numerical example of the auto industry with two countries, we saw how economic
integration led to an increase in competition between firms. Of the 14 firms producing
autos before trade (6 in Home and 8 in Foreign), only 10 firms “survive” after economic
integration; however, each of those firms now produces at a bigger scale (250,000 autos
produced per firm versus either 150,000 for Home firms or 200,000 for Foreign firms before
trade). In that example, the firms were assumed to be symmetric, so exactly which
firms exited and which survived and expanded was inconsequential. In the real world,
however, performance varies widely across firms, so the effects of increased competition
from trade are far from inconsequential. As one would expect, increased competition tends
to hurt the worst-performing firms the hardest, because they are the ones who are forced to
exit. If the increased competition comes from trade (or economic integration), then it is
also associated with sales opportunities in new markets for the surviving firms. Again, as
one would expect, it is the best-performing firms that take greatest advantage of those new
sales opportunities and expand the most.
These composition changes have a crucial consequence at the level of the industry:
When the better-performing firms expand and the worse-performing ones contract or exit,
then overall industry performance improves. This means that trade and economic integration
can have a direct impact on industry performance: It is as if there was technological
growth at the level of the industry. Empirically, these composition changes generate substantial
improvements in industry productivity.
Take the example of Canada’s closer economic integration with the United States (see
the preceding Case Study and the discussion in Chapter 2). We discussed how this integration
led the automobile producers to consolidate production in a smaller number of
Canadian plants, whose production levels rose dramatically. The Canada–U.S. Free Trade
Agreement, which went into effect in 1989, extended the auto pact to most manufacturing
sectors. A similar process of consolidation occurred throughout the affected Canadian
manufacturing sectors. However, this was also associated with a selection process: The
worst-performing producers shut down, while the better-performing ones expanded via
large increases in exports to the U.S. market. Daniel Trefler at the University of Toronto
has studied the effects of this trade agreement in great detail, examining the varied
responses of Canadian firms.9 He found that productivity in the most affected Canadian
industries rose by a dramatic 14 to 15 percent (replicated economy-wide, a 1 percent
increase in productivity translates into a 1 percent increase in GDP, holding employment
constant). On its own, the contraction and exit of the worst-performing firms in response
to increased competition from U.S. firms accounted for half of the 15 percent increase in
those sectors.
Performance Differences Across Producers
We now relax the symmetry assumption that we imposed in our previous development of
the monopolistic competition model so that we can examine how competition from
increased market size affects firms differently. The symmetry assumption meant that all
firms had the same cost curve (8-3) and the same demand curve (8-5). Suppose now that
9See Daniel Trefler, “The Long and Short of the Canada-U.S. Free Trade Agreement,” American Economic
Review 94 (September 2004), pp. 870–895, and the summary of this work in the New York Times: “What
Happened When Two Countries Liberalized Trade? Pain, Then Gain” by Virginia Postel (January 27, 2005).
CHAPTER 8 Firms in the Global Economy 173
firms have different cost curves because they produce with different marginal cost levels .
We assume that all firms still face the same demand curve. Product-quality differences
between firms would lead to very similar predictions for firm performance as the ones we
now derive for cost differences.
Figure 8-6 illustrates the performance differences between firms 1 and 2 when .
In panel (a), we have drawn the common demand curve (8-5) as well as its associated marginal
revenue curve (8-8). Note that both curves have the same intercept on the vertical
axis (plug into (8-8) to obtain ); this intercept is given by the price P from
(8-5) when , which is . The slope of the demand curve is .
As we previously discussed, the marginal revenue curve is steeper than the demand curve.
Firms 1 and 2 choose output levels and , respectively, to maximize their profits. This
occurs where their respective marginal cost curves intersect the common marginal revenue
curve. They set prices and that correspond to those output levels on the common demand
curve. We immediately see that firm 1 will set a lower price and produce a higher
output level than firm 2. Since the marginal revenue curve is steeper than the demand
curve, we also see that firm 1 will set a higher markup over marginal cost than firm 2:
.
The shaded areas represent operating profits for both firms, equal to revenue
minus operating costs (for both firms, and ). Here, we have assumed
that the fixed cost F (assumed to be the same for all firms) cannot be recovered and does not
enter into operating profits (that is, it is a sunk cost). Since operating profits can be rewritten
ci * Qi i = 1 i = 2
Pi * Qi
P1 - c1 7 P2 - c2
P1 P2
Q1 Q2
Q = 0 P + [1/(b * n)] 1/(S * b)
Q = 0 MR = P
c1 6 c2
ci
Cost, C and
Price, P
P2
c
c*
2
c1
P1
Q
D
1 Quantity
MC2
MC1
MR
Intercept = P + [1/(b × n)]
Slope = 1/(S × b)
(P2– c2) × Q2
(P1– c1) × Q1
Q2
(a)
Operating
Profit
c2 Marginal
cost, ci
c1 c*
(b)
Figure 8-6
Performance Differences Across Firms
(a) Demand and cost curves for firms 1 and 2. Firm 1 has a lower marginal cost than firm 2: . Both firms
face the same demand curve and marginal revenue curve. Relative to firm 2, firm 1 sets a lower price and
produces more output. The shaded areas represent operating profits for both firms (before the fixed cost is
deducted). Firm 1 earns higher operating profits than firm 2. (b) Operating profits as a function of a firm’s
marginal cost . Operating profits decrease as the marginal cost increases. Any firm with marginal cost
above c* cannot operate profitably and shuts down.
ci
c1 6 c2
174 PART ONE International Trade Theory
as the product of the markup times the number of output units sold, , we can
determine that firm 1 will earn higher profits than firm 2 (recall that firm 1 sets a higher
markup and produces more output than firm 2). We can thus summarize all the relevant performance
differences based on marginal cost differences across firms. Compared to a firm
with a higher marginal cost, a firm with a lower marginal cost will: (1) set a lower price, but
at a higher markup over marginal cost; (2) produce more output; and (3) earn higher
profits.10
Panel (b) in Figure 8-6 shows how a firm’s operating profits vary with its marginal cost .
As we just mentioned, this will be a decreasing function of marginal cost. Going back to
panel (a), we see that a firm can earn a positive operating profit so long as its marginal cost is
below the intercept of the demand curve on the vertical axis at . Let
denote this cost cutoff. A firm with a marginal cost above this cutoff is effectively
“priced out” of the market and would earn negative operating profits if it were to produce
any output. Such a firm would choose to shut down and not produce (incurring an overall
profit loss equal to the fixed cost F). Why would such a firm enter in the first place?
Clearly, it wouldn’t if it knew about its high cost prior to entering and paying the fixed
cost F.
We assume that entrants face some randomness about their future production cost . This
randomness disappears only after F is paid and is sunk. Thus, some firms will regret their
entry decision if their overall profit (operating profit minus the fixed cost F) is negative. On
the other hand, some firms will discover that their production cost is very low and that they
earn high positive overall profit levels. Entry is driven by a similar process as the one we
described for the case of symmetric firms. In that previous case, firms entered until profits
for all firms were driven to zero. Here, there are profit differences between firms, and entry
occurs until expected profits across all potential cost levels are driven to zero.
The Effects of Increased Market Size
Panel (b) of Figure 8-6 summarizes the industry equilibrium given a market size S. It tells
us which range of firms survive and produce (with cost below ), and how their profits
will vary with their cost levels . What happens when economies integrate into a single
larger market? As was the case with symmetric firms, a larger market can support a larger
number of firms than can a smaller market. This also implies more competition in the
larger market. What are the repercussions for different firms of increased competition?
First, consider the effects of increased competition (higher number of firms n) on the
individual firm-demand curves. Panel (a) of Figure 8-7 shows the effect. Recall that the intercept
on the vertical axis is equal to , which decreases when the number
of firms increases.11 The slope of the demand curve, equal to , decreases from
the direct effect of the increase in the market size S, so the demand curve also becomes
flatter: With increased competition, a producer can gain more market share from a given
price cut. This produces the shift in the demand curve from D to shown in panel (a) of
Figure 8-7. Notice how the demand curve shifts in for the smaller firms (lower-output )
that operate on the top part of the demand curve.
Panel (b) of Figure 8-7 shows the consequences of this demand change for the operating
profits of firms with different cost levels . The decrease in demand for the smaller
firms translates into a new, lower-cost cutoff, : Some firms with the high cost levels
above c*oecannot survive the decrease in demand and are forced to exit. On the other hand,
c*oe
ci
Qi
Doe
1/(S * b)
P + [1/(b * n)]
ci
ci c*
ci
ci
ci
ci
ci
P + [1/(b * n)] c*
ci
(Pi - ci) * Qi
10Recall that we have assumed that all firms face the same nonrecoverable fixed cost F. If a firm earns higher
operating profits, then it also earns higher overall profits (that deduct the fixed cost F).
11The intercept will further decrease because the average price will also decrease.
CHAPTER 8 Firms in the Global Economy 175
Cost, C and
Price, P
D
D′
Quantity
Intercept = P + [1/(b × n)]
Slope = 1/(S × b)
(a)
Operating
Profit
Winners
Marginal
cost, ci
c *
(b)
c *′
Exit
Losers
Figure 8-7
Winners and Losers from Economic Integration
(a) The demand curve for all firms shifts from D to . It is flatter, and has a lower intercept on the
vertical axis. (b) Effects of the shift in demand on the operating profits of firms with different marginal
cost . Firms with marginal cost between the old cutoff, , and the new one, , are forced to exit.
Some firms with the lowest marginal cost levels gain from integration and their profits increase.
ci c* c*¿
D¿
the flatter demand curve is advantageous to some firms with low cost levels: They can
adapt to the increased competition by lowering their markup (and hence their price) and
gain some additional market share.12 This translates into increased profits for some of the
best-performing firms with the lowest cost levels .13
Figure 8-7 illustrates how increased market size generates both winners and losers
among firms in an industry. The low-cost firms thrive and increase their profits and market
shares, while the high-cost firms contract and the highest-cost firms exit. These composition
changes imply that overall productivity in the industry is increasing as production is
concentrated among the more productive (low-cost) firms. This replicates the findings for
Canadian manufacturing following closer integration with U.S. manufacturing, as we previously
described. These effects tend to be most pronounced for smaller countries that
integrate with larger ones, but it is not limited to those small countries. Even for a big
economy such as the United States, increased integration via lower trade costs leads to
important composition effects and productivity gains.14
ci
12Recall that the lower the firm’s marginal cost , the higher its markup over marginal cost . High-cost
firms are already setting low markups and cannot lower their prices to induce positive demand, as this would
mean pricing below their marginal cost of production.
13Another way to deduce that profit increases for some firms is to use the entry condition that drives average
profits to zero: If profit decreases for some of the high-cost firms, then it must increase for some of the low-cost
firms, since the average across all firms must remain equal to zero.
and Industry Performance
In our numerical example of the auto industry with two countries, we saw how economic
integration led to an increase in competition between firms. Of the 14 firms producing
autos before trade (6 in Home and 8 in Foreign), only 10 firms “survive” after economic
integration; however, each of those firms now produces at a bigger scale (250,000 autos
produced per firm versus either 150,000 for Home firms or 200,000 for Foreign firms before
trade). In that example, the firms were assumed to be symmetric, so exactly which
firms exited and which survived and expanded was inconsequential. In the real world,
however, performance varies widely across firms, so the effects of increased competition
from trade are far from inconsequential. As one would expect, increased competition tends
to hurt the worst-performing firms the hardest, because they are the ones who are forced to
exit. If the increased competition comes from trade (or economic integration), then it is
also associated with sales opportunities in new markets for the surviving firms. Again, as
one would expect, it is the best-performing firms that take greatest advantage of those new
sales opportunities and expand the most.
These composition changes have a crucial consequence at the level of the industry:
When the better-performing firms expand and the worse-performing ones contract or exit,
then overall industry performance improves. This means that trade and economic integration
can have a direct impact on industry performance: It is as if there was technological
growth at the level of the industry. Empirically, these composition changes generate substantial
improvements in industry productivity.
Take the example of Canada’s closer economic integration with the United States (see
the preceding Case Study and the discussion in Chapter 2). We discussed how this integration
led the automobile producers to consolidate production in a smaller number of
Canadian plants, whose production levels rose dramatically. The Canada–U.S. Free Trade
Agreement, which went into effect in 1989, extended the auto pact to most manufacturing
sectors. A similar process of consolidation occurred throughout the affected Canadian
manufacturing sectors. However, this was also associated with a selection process: The
worst-performing producers shut down, while the better-performing ones expanded via
large increases in exports to the U.S. market. Daniel Trefler at the University of Toronto
has studied the effects of this trade agreement in great detail, examining the varied
responses of Canadian firms.9 He found that productivity in the most affected Canadian
industries rose by a dramatic 14 to 15 percent (replicated economy-wide, a 1 percent
increase in productivity translates into a 1 percent increase in GDP, holding employment
constant). On its own, the contraction and exit of the worst-performing firms in response
to increased competition from U.S. firms accounted for half of the 15 percent increase in
those sectors.
Performance Differences Across Producers
We now relax the symmetry assumption that we imposed in our previous development of
the monopolistic competition model so that we can examine how competition from
increased market size affects firms differently. The symmetry assumption meant that all
firms had the same cost curve (8-3) and the same demand curve (8-5). Suppose now that
9See Daniel Trefler, “The Long and Short of the Canada-U.S. Free Trade Agreement,” American Economic
Review 94 (September 2004), pp. 870–895, and the summary of this work in the New York Times: “What
Happened When Two Countries Liberalized Trade? Pain, Then Gain” by Virginia Postel (January 27, 2005).
CHAPTER 8 Firms in the Global Economy 173
firms have different cost curves because they produce with different marginal cost levels .
We assume that all firms still face the same demand curve. Product-quality differences
between firms would lead to very similar predictions for firm performance as the ones we
now derive for cost differences.
Figure 8-6 illustrates the performance differences between firms 1 and 2 when .
In panel (a), we have drawn the common demand curve (8-5) as well as its associated marginal
revenue curve (8-8). Note that both curves have the same intercept on the vertical
axis (plug into (8-8) to obtain ); this intercept is given by the price P from
(8-5) when , which is . The slope of the demand curve is .
As we previously discussed, the marginal revenue curve is steeper than the demand curve.
Firms 1 and 2 choose output levels and , respectively, to maximize their profits. This
occurs where their respective marginal cost curves intersect the common marginal revenue
curve. They set prices and that correspond to those output levels on the common demand
curve. We immediately see that firm 1 will set a lower price and produce a higher
output level than firm 2. Since the marginal revenue curve is steeper than the demand
curve, we also see that firm 1 will set a higher markup over marginal cost than firm 2:
.
The shaded areas represent operating profits for both firms, equal to revenue
minus operating costs (for both firms, and ). Here, we have assumed
that the fixed cost F (assumed to be the same for all firms) cannot be recovered and does not
enter into operating profits (that is, it is a sunk cost). Since operating profits can be rewritten
ci * Qi i = 1 i = 2
Pi * Qi
P1 - c1 7 P2 - c2
P1 P2
Q1 Q2
Q = 0 P + [1/(b * n)] 1/(S * b)
Q = 0 MR = P
c1 6 c2
ci
Cost, C and
Price, P
P2
c
c*
2
c1
P1
Q
D
1 Quantity
MC2
MC1
MR
Intercept = P + [1/(b × n)]
Slope = 1/(S × b)
(P2– c2) × Q2
(P1– c1) × Q1
Q2
(a)
Operating
Profit
c2 Marginal
cost, ci
c1 c*
(b)
Figure 8-6
Performance Differences Across Firms
(a) Demand and cost curves for firms 1 and 2. Firm 1 has a lower marginal cost than firm 2: . Both firms
face the same demand curve and marginal revenue curve. Relative to firm 2, firm 1 sets a lower price and
produces more output. The shaded areas represent operating profits for both firms (before the fixed cost is
deducted). Firm 1 earns higher operating profits than firm 2. (b) Operating profits as a function of a firm’s
marginal cost . Operating profits decrease as the marginal cost increases. Any firm with marginal cost
above c* cannot operate profitably and shuts down.
ci
c1 6 c2
174 PART ONE International Trade Theory
as the product of the markup times the number of output units sold, , we can
determine that firm 1 will earn higher profits than firm 2 (recall that firm 1 sets a higher
markup and produces more output than firm 2). We can thus summarize all the relevant performance
differences based on marginal cost differences across firms. Compared to a firm
with a higher marginal cost, a firm with a lower marginal cost will: (1) set a lower price, but
at a higher markup over marginal cost; (2) produce more output; and (3) earn higher
profits.10
Panel (b) in Figure 8-6 shows how a firm’s operating profits vary with its marginal cost .
As we just mentioned, this will be a decreasing function of marginal cost. Going back to
panel (a), we see that a firm can earn a positive operating profit so long as its marginal cost is
below the intercept of the demand curve on the vertical axis at . Let
denote this cost cutoff. A firm with a marginal cost above this cutoff is effectively
“priced out” of the market and would earn negative operating profits if it were to produce
any output. Such a firm would choose to shut down and not produce (incurring an overall
profit loss equal to the fixed cost F). Why would such a firm enter in the first place?
Clearly, it wouldn’t if it knew about its high cost prior to entering and paying the fixed
cost F.
We assume that entrants face some randomness about their future production cost . This
randomness disappears only after F is paid and is sunk. Thus, some firms will regret their
entry decision if their overall profit (operating profit minus the fixed cost F) is negative. On
the other hand, some firms will discover that their production cost is very low and that they
earn high positive overall profit levels. Entry is driven by a similar process as the one we
described for the case of symmetric firms. In that previous case, firms entered until profits
for all firms were driven to zero. Here, there are profit differences between firms, and entry
occurs until expected profits across all potential cost levels are driven to zero.
The Effects of Increased Market Size
Panel (b) of Figure 8-6 summarizes the industry equilibrium given a market size S. It tells
us which range of firms survive and produce (with cost below ), and how their profits
will vary with their cost levels . What happens when economies integrate into a single
larger market? As was the case with symmetric firms, a larger market can support a larger
number of firms than can a smaller market. This also implies more competition in the
larger market. What are the repercussions for different firms of increased competition?
First, consider the effects of increased competition (higher number of firms n) on the
individual firm-demand curves. Panel (a) of Figure 8-7 shows the effect. Recall that the intercept
on the vertical axis is equal to , which decreases when the number
of firms increases.11 The slope of the demand curve, equal to , decreases from
the direct effect of the increase in the market size S, so the demand curve also becomes
flatter: With increased competition, a producer can gain more market share from a given
price cut. This produces the shift in the demand curve from D to shown in panel (a) of
Figure 8-7. Notice how the demand curve shifts in for the smaller firms (lower-output )
that operate on the top part of the demand curve.
Panel (b) of Figure 8-7 shows the consequences of this demand change for the operating
profits of firms with different cost levels . The decrease in demand for the smaller
firms translates into a new, lower-cost cutoff, : Some firms with the high cost levels
above c*oecannot survive the decrease in demand and are forced to exit. On the other hand,
c*oe
ci
Qi
Doe
1/(S * b)
P + [1/(b * n)]
ci
ci c*
ci
ci
ci
ci
ci
P + [1/(b * n)] c*
ci
(Pi - ci) * Qi
10Recall that we have assumed that all firms face the same nonrecoverable fixed cost F. If a firm earns higher
operating profits, then it also earns higher overall profits (that deduct the fixed cost F).
11The intercept will further decrease because the average price will also decrease.
CHAPTER 8 Firms in the Global Economy 175
Cost, C and
Price, P
D
D′
Quantity
Intercept = P + [1/(b × n)]
Slope = 1/(S × b)
(a)
Operating
Profit
Winners
Marginal
cost, ci
c *
(b)
c *′
Exit
Losers
Figure 8-7
Winners and Losers from Economic Integration
(a) The demand curve for all firms shifts from D to . It is flatter, and has a lower intercept on the
vertical axis. (b) Effects of the shift in demand on the operating profits of firms with different marginal
cost . Firms with marginal cost between the old cutoff, , and the new one, , are forced to exit.
Some firms with the lowest marginal cost levels gain from integration and their profits increase.
ci c* c*¿
D¿
the flatter demand curve is advantageous to some firms with low cost levels: They can
adapt to the increased competition by lowering their markup (and hence their price) and
gain some additional market share.12 This translates into increased profits for some of the
best-performing firms with the lowest cost levels .13
Figure 8-7 illustrates how increased market size generates both winners and losers
among firms in an industry. The low-cost firms thrive and increase their profits and market
shares, while the high-cost firms contract and the highest-cost firms exit. These composition
changes imply that overall productivity in the industry is increasing as production is
concentrated among the more productive (low-cost) firms. This replicates the findings for
Canadian manufacturing following closer integration with U.S. manufacturing, as we previously
described. These effects tend to be most pronounced for smaller countries that
integrate with larger ones, but it is not limited to those small countries. Even for a big
economy such as the United States, increased integration via lower trade costs leads to
important composition effects and productivity gains.14
ci
12Recall that the lower the firm’s marginal cost , the higher its markup over marginal cost . High-cost
firms are already setting low markups and cannot lower their prices to induce positive demand, as this would
mean pricing below their marginal cost of production.
13Another way to deduce that profit increases for some firms is to use the entry condition that drives average
profits to zero: If profit decreases for some of the high-cost firms, then it must increase for some of the low-cost
firms, since the average across all firms must remain equal to zero.
No comments:
Post a Comment