Gains and losses

from trade in the

specific-factors
model
Chapter 3
Look for the answers to these questions:

01 02 03
Do you personally Besides you, who What government
gain from gains and who loses policies can help
inexpensive imported from trade? firms and workers
goods? that lose from trade?
Introduction
• Opening a country to trade generates winners and losers.
• Determining who gains and who loses answers many questions
about trade politics.
• Specific-Factors model helps explain who gains and who loses.
• Short Run Specific-Factor model offers new insights beyond
the Ricardian model.
Specific-Factors Model
• How does trade affect the earnings of capital, labor, and
land?

• From the Ricardian model, free trade leads to:

• Rising relative prices in the export sector
• Falling relative prices in the import sector

• So what we really want to know is how changes in

relative prices affect the earnings of factors
Specific-Factors Model
• Why are we concerned with relative prices?

The earnings of specific or fixed factors (such as capital and land) go up or down the
most with changes in relative prices because they are “stuck” in a sector and cannot
be employed elsewhere.

Mobile factors (such as labor) can offset losses from changes in relative prices by
seeking employment in other sectors.
Specific-Factors Model
• Will continue to use two countries: Home and Foreign.

• Home Country
• Manufacturing uses labor and capital.
• Agriculture uses labor and land.
• Diminishing returns to labor—decreasing MPLM and MPL A (see Figure 3.1).

• This is a two good, three factor model.

Specific-Factors Model
Diminishing Marginal Product of Labor

Figure 3.1
❖ Manufacturing uses labor and capital, whereas agriculture uses labor and land. In each
industry, increases in the amount of labor used are subject to diminishing returns; that is, the
marginal product of labor declines as the amount of labor used in the industry increases.
Specific-Factors Model
• Each country faces a standard Production Possibilities Frontier.
• Concave (shapes that curve inward) to the origin because of diminishing returns
to labor in both industries.

• The slope of the PPF is the negative of the ratio of the marginal products (see
figure 3.2).
• The slope is the opportunity cost of producing one unit of manufacturing.

• If L continues to move to manufacturing, MPL A rises and MPLM falls so the slope
of the PPF gets steeper.
Specific-Factors Model

• Production
Possibilities Frontier
Specific-Factors Model
• Opportunity Cost and Prices

• As in the Ricardian model, the slope of the PPF equals the opportunity cost or
relative price of the good on the horizontal axis: here it is manufacturing.

• Firms hire labor up to the point where the cost of one more hour of labor (the
wage) equals the value of one more hour of labor in production.

W = PM  MPLM
W = PA  MPL A
Specific-Factors Model
• Since we assume that labor is mobile, the wages in the two industries
must be equal.

• Relative price of manufacturing equals the opportunity cost of

manufacturing (slope of PPF).
• PM*MPLM = PA*MPL A → PM/PA = MPL A/MPLM

• The no-trade equilibrium for Home is shown on the next slide at point A
• PM/PA = −(slope of PPF) = −(slope of indifference curve)
• The indifference curve is tangent to PPF
Specific-Factors Model
Specific-Factors Model
• The Foreign Country
• Assume the no-trade price in the foreign country (P M*/PA*) is
higher than that in the Home Country (P M/PA).

• For now we will ignore reasons for price differences.

• Home country has comparative advantage in manufacturing (can

produce at lower opportunity cost than Foreign country).
Specific-Factors Model
• Overall Gains from Trade

• When trade opens the world price will end up between

the no-trade prices of the Home and Foreign countries.

• After trade
• Relative Home price of manufacturing will rise
• Relative Foreign price of manufacturing will fall
• Total gains from trade can be measured by the increased utility
of the higher indifference curve
 Specific-Factors Model
Home Country with Trade Once trade is opened and co
Agriculture O nsumers face the new world
utput, QA
price, they are able to move t
o a higher indifference curve
Slope = –(PM/PA)W
(U2)
C
The gains from trade can be
measured by the rise in utility
from U1 to U2.
A U2
Gains from trade
U1
Slope = –(PM/PA) Trade makes prices for
B manufacturing in Home
rise as seen from new
PPF
price line
Manufacturing Outp
ut, QM
 Specific-Factors Model
• What has happened at Home?
• The relatively higher price in manufacturing attracts more workers to that
industry—production now at point B (instead of A).

• Manufactured goods are exported and Agricultural goods are imported.

• Consumption changes — moving individuals to a higher indifference

curve allowing them to now consume at C (instead of A).

• Good whose price rises (falls) becomes the exported (imported) good.
Specific-Factors Model
Old production = A
New production = B
Old consumption = A
New consumption = C
Earnings of Labor
• Although a country as a whole is better off from trade, that does
not mean that every individual is better off.
• How are earnings of labor affected in importing and exporting
industries after trade?
• Determination of Wages
• We can show the amount of labor used in each industry on one graph.

LM + LA = L
• Labor used in manufacturing is measured from the left axis.
• Labor used in agriculture is measured from the right axis.
• See Figure 3.4
 Earnings of Labor
• PM*MPLM is drawn from left to right
Labor mar • PA*MPL A is drawn from right to left
ket equilibr
Wage Wage
ium
PAMPLA • Labor Market Equilibrium is where the two
Value of margin
al product of ag curves cross
riculture
• Allocation of Labor Between
A
W Manufacturing and Agriculture: The
Value of margin amount of labor used in manufacturing is
PM MPLM
measured from left to right along the
al product of m
anufacturing
horizontal axis, and the amount of labor
0M L → L L 0A
MA
used in agriculture is measured from right
L to left. Labor market equilibrium is at point
Manufacturing la
bor
Agriculture la
bor
A. At the equilibrium wage of W,
manufacturing uses 0ML units of labor and
Total labor su
pply
agriculture uses 0AL units.
Earnings of Labor
• Change in Relative Price of Manufactures
• Assume the relative price of manufactures rises (because of the foreign demand for
them).

• A rise in relative price of manufacturing can be caused by an

increase in PM or a decrease in PA.

• Effect on real wage (W) is the same.

• Both these price movements will have the same effect on the real wage; that is, on the amount of
manufactures and food that a worker can afford to buy.

• Assume PM rises
• PM*MPLM curve shifts up by Δ PM ∙ MPLM
• New equilibrium at higher wage
• LM has increased and LA has decreased
 Earnings of Labor
Increase in the Price of Manuf
actured Goods • PM*MPLM shifts up creating a new equilibrium.
• The vertical distance between the old and new
Wage Vertical distance Wage
curves is greater than the increase in wages.
= PM (MPLM)
• Increase in the Price of Manufactured Goods: With
PAMPLA an increase in the price of the manufactured good,
the curve PM · MPLM shifts up to P′M · MPLM and the
W’
B
equilibrium shifts from point A to point B. The
ΔW W PM'MPLM amount of labor used in manufacturing rises from
A 0ML to 0ML′, and ′. The wage increases the amount
PMMPLM of labor used in agriculture falls from 0AL to
0AL’from W to W′ (equilibrium wage has risen from
0M L → L L L 0A W to W′), but this increase is less than the upward
M
A
L shift ΔPM · MPLM.
Earnings of Labor
• Effect on Real Wages
• Do higher wages translate into higher real wages?
• The fact that the wage has risen does not really tell us whether workers are better off or worse off in
terms of the amount of food and manufactured goods they can buy. To determine this, we have to
take into account any change in the prices of these goods.

• We assumed PA did not change so W/PA has increased—workers can buy

more food.
• Because W has increased from W to W′ and we have assumed that P A has not
changed, workers can afford to buy more food. In other words, the real wage has
increased in terms of food.
• the amount of food that a worker can afford to buy (buying power) with his or her
hourly wage is W/P A.
Earnings of Labor
• Effect on Real Wages
• We assume that PM increased, as did W; what is the net effect on W/P M?
• The amount of the manufactured good that a worker can buy is measured by W/P M.
• While W has increased, P M has also increased, so at first glance we do not know
whether W/PM has increased or decreased.

• We showed in figure 3.5 that ΔW < ΔPM·MPLM

• If we divide both sides by W we get:

W PM  MPLM PM
 =
W PM  MPLM PM
Earnings of Labor
• Effects on Real Wages

• the term ΔW/W is the percentage change in wages.

• the term ΔPM/PM is the percentage change in the price of
manufactured goods

• Since ΔW/W < ΔPM/PM, the amount of manufactured goods that

can be purchased with the money wage has fallen.

• The real wage in terms of manufactured goods has decreased.

Earnings of Labor
• Is labor better or worse off after the price increase?
• A person who spends more of his or her income on agricultural goods is better off.
• But a person who spends more of his or her income on manufactured goods is worse
off.
• The overall effect on the well-being of workers is thus ambiguous.
• Although ambiguous, this conclusion is important.
• The result is different than what was found in the Ricardian model, where labor
unambiguously earned a higher real wage.
• This warns us that one cannot make unqualified statements about the effects of trade
on workers.
• The effect of trade on real wages can be complex.
Earnings of Labor
• Unemployment in Specific Factors Model
• Total labor is always LM + L A so no unemployment.
• Why do we ignore unemployment?
• Unemployment is usually considered a macro phenomenon affected by
business cycles.
• Many people laid off due to trade often find new jobs within a reasonable
amount of time, often with higher wages.

• Even if we were to consider the spells of unemployment due to

trade, we see that workers can find new jobs typically in the
expanding exporting industry.
• Even after we take into account that workers eventually find new
jobs, we still cannot conclude whether trade is necessarily good
or bad for workers
Manufacturing and Services in the US
• Manufacturing Services in the U.S.: Employment and Wages Across Sectors.
• Figure 3.6 shows employment in U.S. manufacturing industry over time.
• Figure 3.7 shows real wages earned by production workers in manufacturing, all
private services, and in information services.
Manufacturing and Services in the US
Manufacturing and Services in the US
• Conclusions
1. Wages differ across different sectors in the economy, so the
assumption that wages are the same in both industries is a
simplification.
2. Many workers that are displaced every year for various reasons
must find jobs elsewhere.
• Some are laid off because of import competition, but there are many
other reasons
3. The majority of workers find new jobs within 2–3 years, but not
necessarily at the same wage
4. Real wages for all production workers fell in most years between
1972–95, but have since risen.
Manufacturing and Services in the US
Job Losses in Manufacturing and Service Industries, 2011–2013

• Losses in Manufacturing and Service Industries, 2011–2013: This table shows the number of
displaced workers in manufacturing and service industries from 2011 to 2013. A total of 61% of
the manufacturing workers displaced from 2011 to 2013 were reemployed by January 2014,
with 57% earning less in their new jobs in manufacturing and 43% earning the same or more.
But in service industries, 72% of the workers who were reemployed earned less in their new
jobs with the other 28% earning the same or more.
Earnings of Capital and Land
• Although there are “overall” gains from trade for the country, we
have found that labor (the mobile factor) does not necessarily gain

• What about the earnings of the other factors of production, capital

and land, which cannot switch between industries?

• Determining the Payment to Capital and Land

• Capital and Land earn what is left over from sales revenue after labor is paid.
Earnings of Capital and Land
• Determining the Payment to Capital and Land
• Payments to capital = P M·QM – W·LM
• Payments to land = PA·QA – W·L A

• Let T denote land and K denote capital.

• We can now determine the earnings of capital, R K, and land, RT

Payments to capital PM QM − WL M
RK = =
K K
Payments to land PAQA − WL A
RT = =
T T
Earnings of Capital and Land
• Determining the Payment to Capital and Land
• RK and RT reflect what these factors earn during a given period when used in
these industries.

• Also, the amount the factors could earn if rented to someone else over the
same time.

• Alternatively, we can calculate the rental on capital and land via the value of the
additional output we get from hiring those factors.

RK = PM  MPK M
RT = PA  MPTA
 Earnings of Capital and Land
• Change in the Real Rental on Capital
• Assume PM increases as before, PA constant.
• We saw before that wages rise and labor shifts from agriculture to
manufacturing.
• As more labor is used in manufacturing, the marginal product of capital will
rise.
• As more labor leaves agriculture, the marginal product of land will fall.

• General Conclusion:
An increase in the quantity of labor used in an industry will raise the
marginal product of the factor specific to that industry, and a decrease
in labor will lower the marginal product of the specific factor.
Earnings of Capital and Land
• We can get more from this conclusion:

• Remember MPKM = RK/PM

• PM is rising and we know MPK M is rising.
• RK must be increasing by more than P M is increasing

• Also remember RK/PA is amount of food that can be purchased by capital owners.
• RK increased, PA fixed, so RK/PA increases.

• The owners of capital are clearly better off under trade than in the no-trade case.
 Earnings of Capital and Land
• Change in the Real Rental Rate on Land
• Labor leaves agriculture, causing MPTA to fall.
• Since MPTA = RT/PA, RT/PA must fall, meaning RT falls
• PA has not changed and PM has increased, so landowners cannot buy as much of either
good.
• Landowners are clearly worse off with trade.

• Summary
• An increase in the relative price of an industry’s output will increase the real rental earned
by the factor specific to that industry, but will decrease the real rental of factors specific
to other industries.

• Generally, specific factors in export industries gain, and specific factors in importing
industries lose.
Numerical Example
• Manufacturing:
• Sales Revenue = PMQM = $100
• Payments to Labor = WL M = $60
• Payments of Capital = R KK = $40

• Agriculture
• Sales Revenue = PAQA = $100
• Payments to Labor = WL A = $50
• Payments to Land = RTT = $50

• Assume PM increases by 10% and PA remains unchanged. Percentage

change in labors wages is 5%
• ΔPM/PM = 10%
• ΔPA/PA = 0%
• ΔW/W = 5%
• Remember the percent change in wages will be between the percent change in
the two industry price changes.
Numerical Example
• Change in the Rental Rate on Capital
Payments to Capital PM QM − WL M
RK = =
K K
PM QM − WL M
RK =
K
Rewriting using percentage changes gives :
RK (PM PM )PM QM − (W W )WL M
=
RK RK K
Numerical Example
• Change in the Rental Rate on Capital
• We want to rewrite this equation (change in the rental) using percentage changes, like ΔP M/PM,
ΔW/W, and ΔRK/RK. To achieve this, divide both sides by RK and rewrite the equation (change
in the rental) as

You can cancel terms i

n this equation to chec
k that it is the same as
before.
Numerical Example
• Change in the Rental Rate on Capital
• ΔRK/R = (10%*100-5%*60)/40 = 17.5%

• The percentage increase in rental on capital is greater than the

percentage increase in the relative price of manufacturing, 10%

• This holds no matter what, given that the percentage increase in

the wage is less than the percentage increase in the price of the
manufactured good.
Numerical Example
• Change in Rental Rate on Land
0(QA ) − WL A
RT =
T
We know that wages are increasing which means the rental on land is falling. We
can calculate how much.
• Land rent falls by same percentage as wage increases.
• This occurs because we assumed labor and land received the same share of
sales revenue

• dRT/RT = -5%*(50/50) = -5%

Earnings of Capital and Land
• General Equation for the Change in Factor Prices

• All changes in factor and industry prices are related.

• Assume PM increases and PA does not change, then:

RT W PM RK

0  
RT W PM RK
• We get what is called factor price magnification.
Earnings of Capital and Land
• General Equation for the Change in Factor Prices
• The opposite is true if P M falls.
• Wages fall by less than percent change in the manufactured good, rental on
capital falls by more than the manufacturing price, and rental on land rises .
• What happens if PA increases?

RK W PA RT

0  
RK W PA RT
What It All Means
• The earnings of specific factors change the most from relative price
changes due to international trade.
• This is because these factors (land and capital) cannot move between
industries.
• The earnings changes are in opposite directions and so the interests
of T and K are opposed to each other.
• Changes in wages paid to labor are less extreme.