Appendix 2A: Trade
Balance
Table 1
Value in millions of dollars: 1960 thru 2010
|
|||
Year
|
Total
|
Goods
|
Services
|
1960
|
3,508
|
4,892
|
-1,384
|
1961
|
4,195
|
5,571
|
-1,376
|
1962
|
3,370
|
4,521
|
-1,151
|
1963
|
4,210
|
5,224
|
-1,014
|
1964
|
6,022
|
6,801
|
-779
|
1965
|
4,664
|
4,951
|
-287
|
1966
|
2,939
|
3,817
|
-878
|
1967
|
2,604
|
3,800
|
-1,196
|
1968
|
250
|
635
|
-385
|
1969
|
91
|
607
|
-516
|
1970
|
2,254
|
2,603
|
-349
|
1971
|
-1,302
|
-2,260
|
958
|
1972
|
-5,443
|
-6,416
|
973
|
1973
|
1,900
|
911
|
989
|
1974
|
-4,293
|
-5,505
|
1,212
|
1975
|
12,404
|
8,903
|
3,501
|
1976
|
-6,082
|
-9,483
|
3,401
|
1977
|
-27,246
|
-31,091
|
3,845
|
1978
|
-29,763
|
-33,927
|
4,164
|
1979
|
-24,565
|
-27,568
|
3,003
|
1980
|
-19,407
|
-25,500
|
6,093
|
1981
|
-16,172
|
-28,023
|
11,851
|
1982
|
-24,156
|
-36,485
|
12,329
|
1983
|
-57,767
|
-67,102
|
9,335
|
1984
|
-109,072
|
-112,492
|
3,420
|
1985
|
-121,880
|
-122,173
|
294
|
1986
|
-138,538
|
-145,081
|
6,543
|
1987
|
-151,684
|
-159,557
|
7,874
|
1988
|
-114,566
|
-126,959
|
12,393
|
1989
|
-93,141
|
-117,749
|
24,607
|
1990
|
-80,864
|
-111,037
|
30,173
|
1991
|
-31,135
|
-76,937
|
45,802
|
1992
|
-39,212
|
-96,897
|
57,685
|
1993
|
-70,311
|
-132,451
|
62,141
|
1994
|
-98,493
|
-165,831
|
67,338
|
1995
|
-96,384
|
-174,170
|
77,786
|
1996
|
-104,065
|
-191,000
|
86,935
|
1997
|
-108,273
|
-198,428
|
90,155
|
1998
|
-166,140
|
-248,221
|
82,081
|
1999
|
-263,160
|
-336,171
|
73,011
|
2000
|
-376,749
|
-445,787
|
69,038
|
2001
|
-361,771
|
-421,276
|
59,505
|
2002
|
-417,432
|
-474,491
|
57,059
|
2003
|
-490,984
|
-540,409
|
49,425
|
2004
|
-605,357
|
-663,507
|
58,150
|
2005
|
-708,624
|
-780,730
|
72,106
|
2006
|
-753,288
|
-835,689
|
82,401
|
2007
|
-696,728
|
-818,886
|
122,158
|
2008
|
-698,338
|
-830,109
|
131,770
|
2009
|
-381,272
|
-505,910
|
124,637
|
2010
|
-500,027
|
-645,857
|
145,830
|
Data presented on a
Balance of Payment (BOP) basis. Information on data sources and methodology
are available at www.census.gov/foreign-trade/www/press.html.
|
|||
Table 2
Merchandise
trade balance
|
In
millions of dollars
|
||||||||||||
Total
|
Nether
lands
|
||||||||||||
2000
|
-436,104
|
-51,897
|
-83,833
|
-9,439
|
-29,064
|
-13,982
|
-81,555
|
-12,478
|
-24,577
|
12,165
|
-16,097
|
-1,775
|
|
2001
|
-411,899
|
-52,844
|
-83,096
|
-10,544
|
-29,081
|
-13,874
|
-69,022
|
-13,001
|
-30,041
|
9,969
|
-15,253
|
-655
|
|
2002
|
-468,263
|
-48,165
|
-103,065
|
-9,224
|
-35,876
|
-14,164
|
-69,979
|
-12,996
|
-37,146
|
8,462
|
-13,766
|
-7,540
|
|
2003
|
-532,350
|
-51,671
|
-124,068
|
-12,166
|
-39,281
|
-14,854
|
-66,032
|
-13,157
|
-40,648
|
9,742
|
-14,152
|
-8,967
|
|
2004
|
-654,830
|
-66,480
|
-162,254
|
-10,688
|
-45,850
|
-17,413
|
-76,237
|
-19,981
|
-45,170
|
11,689
|
-13,038
|
-10,372
|
|
2005
|
-772,373
|
-78,486
|
-202,278
|
-11,583
|
-50,567
|
-19,485
|
-83,323
|
-16,210
|
-49,861
|
11,606
|
-13,211
|
-12,465
|
|
2006
|
-827,971
|
-71,782
|
-234,101
|
-13,528
|
-47,923
|
-20,109
|
-89,722
|
-13,584
|
-64,531
|
13,617
|
-15,502
|
-8,103
|
|
2007
|
-808,763
|
-68,169
|
-258,506
|
-14,877
|
-44,744
|
-20,878
|
-84,304
|
-13,161
|
-74,796
|
14,434
|
-12,449
|
-6,876
|
|
2008
|
-816,199
|
-78,342
|
-268,040
|
-15,209
|
-42,991
|
-20,674
|
-74,120
|
-13,400
|
-64,722
|
18,597
|
-11,400
|
-4,988
|
|
2009
|
-503,582
|
-21,590
|
-226,877
|
-7,743
|
-28,192
|
-14,162
|
-44,669
|
-10,604
|
-47,762
|
16,143
|
-9,877
|
-1,776
|
|
Source:
|
|||||||||||||
Appendix 2B: The
Standard Theory
Comparative
Advantage: Ricardo Model and Gains from Trade
There are two countries, Industria and Bucolia. Industria
produces the manufactured good widgets; Bucolia grows cycads. For Industria it
takes one man-hour to produce one widget and 2 man-hours to produce one cycad.
Bucolia must devote more man-hours to produce both goods; 10 for a widget and 5
for a cycad as shown in the following table:
Unit
Labor Requirement
Product
Widgets Cycads
Country Industria 1 2
Bucolia 10 5
Without trade the relative price in man-hours of widgets
with respect to cycads (Pw/Pc) for each country is calculated as:
W/C
Industria 0.5
Bucolia 2.0
The relative world price must lie between these country
extremes; assume that it is Pw/Pc = 1.
If Industria desires 1 cycad what is the best way to get
it? It can either produce it itself or it can trade for it. Producing it
requires 2 units (man-hours) of labor with the following results:
Labor
Widgets Cycads
used produced produced
2 0 1
These 2 labor units produce 1 cycad and zero widgets.
Industria can, however, devote these same 2 units of labor to produce widgets
instead of cycads:
Labor Widgets Cycads
used produced produced
2 2 0
Industria uses two units of labor to produce 2 widgets.
At the world price it trades one widget for one cycad with the following
result:
Labor Widgets Cycads
After trade 2
1 1
For the same amount of labor Industria now has a widget
and a cycad.
Bucolia also benefits from trade. If Bucolia desires 1
widget what is the best way to get it? It can either produce it itself or it
can trade for it. Producing it requires 10 units (man-hours) of labor with the
following results:
Labor
Widgets Cycads
used produced produced
10 1 0
Bucolia can, however, devote these same 10 units of labor
to produce cycads instead of widgets:
Labor
Widgets Cycads
used produced produced
10 0 2
Trading a cycad for a widget gives the following:
Labor Widgets Cycads
After trade 10
1 1
For the same amount of labor Bucolia now has a cycad and
a widget. Thus each country by specializing in one good increases its total
consumption.
The Geometry of
Tariffs
The above figure represents the simple geometry of
tariffs assuming a home country on one side and the rest of the world as the
other party.
S: supply curve in home country; this is less than the
world supply curve; It represents domestic production exclusive of imports of
foreign production.
D: demand curve in home country; this is less than the
world demand curve.
Pt: home price with tariff – higher due to reduction in
imports. Foreign producers receive a lower price and are therefore only willing
to supply less.
PFt: price in foreign countries with tariff – lower
because some foreign production that would have been sold to “home” is now kept
in “foreign”.
S1: amount supplied by home producers at pre-tariff
price.
S2: amount supplied by home producers at higher
post-tariff price.
D1: amount demanded by home consumers at pre-tariff
price.
D2: amount demanded by home consumers at higher
post-tariff price.
Consumer surplus:
w+x+y+z = loss in consumer surplus due to rise in price.
Producer surplus: w = gain in home producer surplus due
to rise in price.
Tariff revenue: y+g = amount of tariff (Pt – PFt) times
quantity imported (D2-S2).
Assuming equal weights for each party the net cost is loss in consumer surplus – gain
in producer surplus – government tariff revenue: (w+x+y+z)-w-(y+g) = x+z-g.
Efficiency of free trade: triangle x represents the
production distortion cost resulting from domestic producers producing an
excess quantity of the good, triangle z represents the consumption distortion
cost resulting from consumers consuming less of the good.
Appendix 2C: Multiple
Equilibrium
Gomory and Baumol provide a brief outline of their
multiple equilibrium model: 1) The model deviates from classical trade models
only in the assumption of economies of scale. 2) It consists of a set of demand
and production functions from each commodity in each country. 3) Supply and
demand for each commodity are equal; there is one variable input, labor, where
total revenue equals labor cost for each industry (zero economic profit); labor
in each country is fully employed. The model outputs are: quantities of each
good produced and consumed in each of the two countries, share of total output
of each good produced by each country, quantity of labor in each country for
the production of each good, price of each good, wage rate in each country and
total income in each country. They also assume that labor is the only input to production; there is a fixed
quantity of labor available in each country; and demand functions in each
country combine into a Cobb-Douglas type national utility function. This
implies that the amount spent on each product is fixed, i.e. demands are unit
elastic over their full range.[1]
Assumptions similar to these are employed in many
standard economic models to bring out the underlying factors; these are not at
all eccentric or peculiar. Cobb-Douglas functions, i.e. functions of the form Q
= AKxL1-x are generally used by economists owing to their
convenient and easy to interpret economic properties.
It is the case that scale economies tend to natural
monopoly or perfect specialization in each industry. The authors sketch a proof
that under the assumptions of their model each such perfect specialization
outcome is also an equilibrium outcome. [2]
They also show that with n goods and assuming that each of the countries must
have at least one industry the number of possible industry assignments to both
countries total 2n – 2 specialized assignments. Thus there can be an
enormous number of equilibrium points.
How do they calculate the points on the upper and lower
boundaries? To find the upper boundary they formulate a mathematical program
which maximizes national income (or utility) subject to the following three
constraints: full employment in each country, quantity supplied equals quantity
demanded for each product; each industry has zero (economic) profit. The lower
boundary curve is calculated by minimizing national income (utility) subject to
the three constraints. In practice there are a number of simplifications which
render solving the problem more mathematically tractable.[3]
In the event of “nonspecialized” equilibrium points, i.e.
industries in which production is shared by the two countries, the authors show
that these are bounded by the upper frontier. However, these may fall beneath
the lower specialized frontier. They show that these lower equilibrium points
are unstable and conclude that the space between the boundary curves closely
approximate the region of stable equilibrium points.[4]
They prove that under scale economies a locally stable
equilibrium point can exist that does not attain productive efficiency i.e. the
largest output of one commodity attainable that does not reduce the output of
any of the other commodities. They conclude that the “invisible hand can
indeed, by the happenstance of history find itself stuck at an equilibrium that
is locally optimal but globally far inferior to others, even inferior to the
autarky [that is one country monopolizes production] equilibrium for at least
one of the trading countries.” In other words “some of the locally stable
equilibria will keep the absolute incomes of one of the two countries, and in
many cases those of both countries, below their maximal attainable levels.”[5]
They examine the case of linear models, i.e. those that
assume constant returns to scale for all factors of production except that of
land which is, of course subject to diminishing returns. They find that 1)
there also exist upper and lower boundaries of the equilibria as in the case of
scale economies, 2) if one country’s equilibria is at a maximum productivity
for each commodity it produces, then increases in its trading partner’s
productivity in one of those goods may result in the industry being lost to
that country, 3) it is possible to determine the characteristics of an ideal
trading partner, these being that the partner is impoverished with wages less
than one-third the level of the more affluent trading country, 4) in case 3 it
would benefit the more industrial country to help its low wage partner improve
its trading position; however if the poorer partner’s wage is higher than a low
threshold level then the richer country would lose out by that strategy. Again
this is like the scale economies model.[6]
If the poorer country is an aggressive potential threat then it might in the
long run never be beneficial for the industrial country to help out. But of
course this is beyond the purview of the economics profession.
These are the conditions for a country to constitute an
ideal trading partner under the simulations of the linear model. “It must be a
producer of a modest share of the traded commodities, leaving it with low
relative wages and a small share of world income; it must be a maximally
efficient producer of just those goods that it does supply; and it must be an
inefficient producer of all the remaining commodities, so that it constitutes
no competitive threat in those industries.” [7]
There is one additional conclusion to be drawn from these
series of simulations. Rising productivity in industry i in the minor trading
partner, country 2, benefits country 2 through a greater abundance in good i and
also by a rise in the wage rate which more than compensates for the rise in
prices. But the major partner, country 1, is harmed because the rise in the
prices of country 2’s goods increases the cost of country 1’s imports without
any increase in its purchasing power.[8]
Gomory and Baumol look at a number of real world
conditions that add complications to their model. 1) There are in reality more
than two countries. While some new features arise in the multi-country case the
fundamental finding remains that there is a zone of conflict in the interests
of countries similar in wealth. 2) In reality goods may be subject to
diminishing returns to scale or in which after an interval of increasing
returns, diminishing returns to scale sets in. These lead to the likelihood
that perfectly specialized outcomes give way to multi-country production of the
commodities. They extend the model to deal with the case where equilibrium
points may include multi-country production. 3) They also extend the model for
the case where some goods are not tradable internationally. 4) They include in
their analysis the case where some industries have economies of scale while
others do not. None of these cases fundamentally change the conclusions that
there are inherent conflicts as well as regions of mutual gain in international
trade.[9]
Appendix 2D: Stolper-Samuelson
Theorem
The Stolper–Samuelson
theorem is a basic theorem in Heckscher–Ohlin type trade theory. It describes
a relation between real wages and real returns to capital. Under
the economic
assumptions of constant returns to scale, perfect competition, and an equality
of the number of factors to the number of products a rise in the relative price
of a good will lead to a rise in the return to that factor which is used most
intensively in the production of the good. A fall in the relative price means a
fall in the return to the other factor.
In an economy that produces only wheat and cloth, with
labor and land as the factors of production, wheat is a land-intensive industry
and cloth is labor-intensive. With the
usual assumption of microeconomics that the price of each product equals its
marginal cost the price of cloth should be:
1)
Pcloth = ar + bw
2)
Pwheat = cr + dw
where r is rent, w is the wage; a and b are the amounts
of each factor used in cloth production; c and d are the amounts of each factor
used in wheat production.
When cloth experiences a rise in its price, at least one
of its factors must also become more expensive, for equation 1 to hold true. It
can be assumed that labor, the intensively used factor in the production of
cloth, is the one that would rise. Similarly, when the wage rises, rent must
fall, in order for equation 2 to hold true. But a fall in rent also affects equation
1. For it to still hold true, then, the rise in wages must be more than
proportional to the rise in cloth prices.
Thus, a rise in the price of a product will more than
proportionally raise the return to the most intensively used factor, and a fall
on the return to the less intensively used factor.
Appendix 2E:
Multiplier Effect of Imports
The following illustrates the simple multiplier effect of
trade:
Assume that the Marginal
Propensity to Consume is
dC/dY = 0.6
Therefore the Marginal Propensity to Save is
dS/dY = 0.4
The multiplier is calculated as
1/(dS/dY) = 2.5
But if dC/dY is divided into parts:
1) Marginal propensity to consume domestic goods
dCd/dY = 0.5
2) Marginal propensity to consume imported goods
dCm/dY = 0.1
The multiplier is then
1/[(dS/dY) + (dCm/dY)] = 2
With a stimulus injection of $10 billion and no marginal
propensity to import:
Income increases by 25 billion
With an injection of $10 billion and with a marginal
propensity to import:
Income increases by only 20 billion
This simple analysis ignores the effect of increased
imports increasing foreign incomes and hence inducing some increase in exports.
Appendix 2F: Trade
Warfare
Problem of trade warfare
Japan Japan
Free trade Protection
Free 10 20
U.S. Trade
10 -10
Protec- -10 -5
U.S. tion
20 -5
The table shows the free trade trap as an example of a
payoff matrix from the classic game-theoretic prisoner’s dilemma. Assume the
payoffs from two trade policies as given in the payoff matrix. If both
countries agree to follow free trade strategies the payoff to each is 10. If
one chooses free trade and the other chooses protection the free trader loses
10, the protector gains 20. Minimizing the maximum loss results in both opting
for protection with a loss of 5 for each. A cooperative policy through a free
trade agreement improves the outcomes for both with each gaining 10.
Appendix 2G: Brander-Spencer
Analysis
Krugman and Obstfeld present the following analysis by Brander
and Spencer.[11]
The latter have proposed an argument for industrial policy as illustrated in
the following payoff matrices. In some industries with very large economies of
scale there will be a small number of firms with excess returns that impel
great international competition. A “subsidy to domestic firms, by deterring
investment and production by foreign competitors, can raise the profits of
domestic firms by more than the amount of the subsidy.” Assume that an American
firm, Boeing and a European firm, Airbus are competing in the production of a new
type of aircraft; the following are the payoffs resulting from the two
strategies: produce or don’t produce.
Brander-Spencer Analysis
Airbus
Produce
Don't Produce
Produce -5 0
Boeing -5 100
Don't produce
100 0
0 0
If both firms try to produce at the same time each one loses
– upper left. If Boeing obtains a small head start it will produce – upper
right (Airbus would have the same advantage). However Europe
can reverse that advantage with a subsidy of 25 resulting in the following.
Brander-Spencer Analysis
Airbus
Produce
Don't Produce
Produce 20 0
Boeing -5 100
Don't produce
125 0
0 0
Equilibrium shifts from the upper right to the lower
left. Airbus profits while Boeing is deterred from entering or leaves the
industry. Airbus profit (125) far exceeds the subsidy (25).
On the other hand, a slight difference can change the outcome.
Airbus
Produce Don't Produce
Produce -20 0
Boeing 5 125
Don't produce 100 0
0 0
If
Boeing has some advantage sufficient to keep it in the industry while Airbus is
at a relative disadvantage then the latter cannot produce profitably;
equilibrium is in the upper right corner. Now a subsidy of 25 would still
induce Airbus to enter but profits would be less than the subsidy.
Airbus
Produce Don't Produce
Produce 5 0
Boeing 5 125
Don't produce 125 0
0 0
Thus the total gain to Europe through Airbus’ profits is
less than the cost to Europe of providing the
subsidy. The difference is because the subsidy to Airbus is not a deterrent to the
entry of Boeing. Also such subsidies
face the prospect of retaliation risking a trade war that leaves all parties
worse off.
No comments:
Post a Comment