Notes on Trade Part 1

Economics 4999-005 Global Issues

Fall, 2000 Keith Maskus

INTERNATIONAL ECONOMICS NOTES

PART ONE: PATTERNS OF INTERNATIONAL TRADE AND THE GAINS FROM TRADE

We begin our overview study of international economics by answering the following questions:

1. What factors determine the products that countries export and import?

2. What are the gains from engaging in international trade?

3. How are those gains distributed among people within an economy?

1. THE DETERMINANTS OF COMPARATIVE ADVANTAGE

It is evident that (but actually false) that if countries were exactly alike they would not need to trade with each other. So the main question is "in what important ways are countries different?" Important possibilities include:

a. differences in technology or productivity;

b. differences in factor endowments;

c. differences in market size;

d. differences in consumer preferences (arising, say, from culture);

e. differences in government policies affecting markets;

All of these (and other) differences can give rise to trade. They do so by generating differences in relative costs or relative prices, which we associate with comparative advantage:

Consider 2 countries, H and F, and 2 goods, X and Y. Country H has a comparative advantage (CA) in good X if its relative price in autarky (no-trade) is lower than in F: (p_x/p_y)^H < (p_x/p_y)^F.

Note that if H has CA in X then F has CA in Y. More generally, any country has a CA in some goods, which is central to the claim that they gain from trade.

Important note: unless there is imperfect competition (eg, monopolies), commodity prices will equal marginal costs of production. So it is equally valid to define CA in terms of relative marginal production costs:

Consider 2 countries, H and F, and 2 goods, X and Y. Country H has a comparative advantage (CA) in good X if its relative marginal cost in autarky (no-trade) is lower than in F: (c_x/c_y)^H < (c_x/c_y)^F.

In fact, this is a better definition because it focuses attention on the determinants of marginal costs. For our purposes it is sufficient to consider a series of simple models.

MODEL ONE: DIFFERENCES IN LABOR PRODUCTIVITY

Consider a world with 2 countries, H and F, but only one factor of production, L. Countries differ in the inherent labor-productivity coefficients in the 2 goods, X and Y. These coefficients are fixed no matter the level of production. Here is an example: let H be the US, F be Mexico, and good X be software (S) and Y be radios (R).

Output per unit of labor (marginal products of labor):

S R

US 10 5

M 1 3

Note: US has an absolute advantage in both S and R. But consider relative opportunity costs. In the US, S is half as expensive as R in terms of labor content (that is, it takes 2 workers to produce 10 R but only one worker to produce 10 S; put another way it takes 0.1 workers per S and 0.2 workers per R). In Mexico, S is 3 times as expensive as R (that is, it takes 3 workers to produce 3 S but only one worker to produce 3 R; it takes 1 worker per S and 0.33 workers per R).

Now marginal costs would be wage rate times workers per unit of output, or

US: p_s = c_s = w_us*(0.1) p_r = c_r = w_us*(0.2) Þ (p_s/p_r)^us = 1/2

Mexico: p_s = c_s = w_m*(1) p_r = c_r = w_m*(0.33) Þ (p_s/p_r)^m = 3

So software is comparatively cheap in US, radios are cheap in Mexico. This provides room for mutually beneficial trade and gains from trade (GFT).

Specifically, if the free-trade price ratio lies between these autarky price ratios, both countries will be better off:

1/2 < (p_s/p_r)^* < 3

For example, let (p_s/p_r)^* = 1. For the US, a unit of S in trade is worth 1 R, whereas in autarky it was worth only 1/2 R. The US would choose to specialize its labor force completely in S and export S in return for its desired R. For Mexico, a unit of R is worth 1 S, whereas in autarky it was worth only 1/3 S. Mexico would specialize its labor force completely in R and export R in return for its desired S. Note that both countries enjoy a higher relative price for their export good in free trade than in autarky.

Definition: a country's terms of trade is the ratio of the price of its export good to the price of its import good.

Note that a rise in the terms of trade is beneficial; a fall in the terms of trade is harmful.

INTERPRETATION: free trade offers a better technological opportunity than autarky does as long as the autarky price ratios are different. Specialization by CA raises the productivity of workers (and their real wages) by permitting them to produce what they are relatively best at.

Example: let labor force in US be 100 workers and labor force in Mexico be 300 workers. Their PPFs would look like the following (we'll fill in some numbers in class):

R R

p^us p^m p*

US S Mexico S

Suppose that both US and Mexico split their labor in half between S and R in autarky. Then we have the following equilibrium points in autarky:

Production Consumption Trade

US S R S R S R

M S R S R S R

Now with both countries specialized and the free-trade price ratio of 1S = 1R, suppose that the US exports 300 S and imports 300 R. We have the following equilibrium points in free trade:

Production Consumption Trade

US S R S R S R

M S R S R S R

We calculate gains from trade by comparing consumption levels in autarky and free trade:

US GFT = S, R

M GFT = S, R

Another way of showing these gains is to draw the free-trade price line, which is also the trade possibilities frontier, in the diagrams above. Note each country is permitted by trade to consume outside its PPF.

Questions:

1. Is each person in the US and Mexico better off? Why?

2. Who got the larger gains from trade, the US or Mexico? Why?

Essential points of this analysis:

a. Trade permits specialization according to CA.

b. Countries specialize resources in products with highest relative productivity levels, raising real wages compared to situations with less trade.

c. Nominal costs (wages, exchange rates) must adjust to reflect this structure of CA. This can be painful: consider US textile workers, steelworkers, etc. where productivity was not enough to support high wages. But also enjoyable: consider the wage gains to workers in software, services where productivity is very high.

PROBLEMS WITH MODEL ONE: one factor of production, fixed productivity coefficients, all workers gain from specialization. Unrealistic.

MODEL TWO: DIFFERENCES IN RELATIVE FACTOR ENDOWMENTS

Consider a world with 2 countries, H and F, and 2 factors, K and L. K and L are identical in H and F, which share the same technologies for producing X and Y. The only difference between H and F is relative endowments of labor while X and Y have different production functions with constant returns to scale.

Let H be capital-abundant: (K*/L*)^H > (K*/L*)^F where endowments are fixed;

Let X be capital-intensive (K/L)_x > (K/L)_y for any set of factor prices (w,r).

Note this means F is labor-abundant and Y is labor-intensive.

Then we can specify the Heckscher-Ohlin theorem ("factor proportions" theorem):

Each country will export the good that is intensive in its relatively abundant factor and import the good that is intensive in its relatively scarce factor.

The technical proof of the theorem is complicated (if you like microeconomic theory, give it a try). But it boils down to this: because H (F) is capital-abundant (labor-abundant), H will have a high ratio of wage costs to capital costs in autarky and F will have a low ratio. That is:

(w/r)^H > (w/r)^H .

This means that H (F) will have a relatively low cost of producing good X (good Y) so that:

(p_x/p_y)^H < (p_x/p_y)^F .

So the structure is that H has a CA in X and F has a CA in Y. In free trade, an intermediate price must be established. Thus, for H the relative price of X rises (of Y falls), while for F the relative price of Y rises (of X falls). Again, trade improves each nations' terms of trade and is a technological improvement.

Here is an important difference between models one and two. In the HO model, as relative goods prices change, so do relative factor prices. Where free trade generates a common goods price ratio, it will also achieve a common factor price ratio in H and F (assuming H and F both remain incompletely specialized). That is:

(w/r)^H > (w/r)* > (w/r)^H .

So an outcome of this model is factor-price equalization (FPE), in which the high-priced scarce factors see their prices fall and the low-priced abundant factors see their prices rise in the movement from autarky to free trade.

EXAMPLE: Let H be US, F be China, X be machinery and Y be textiles. With 2 factors and CRS, the PPFs take on the usual concave shapes:

T T p^c p*

p^us

US M China M

Notes on the free-trade equilibrium:

a. There is incomplete specialization in production (points B, B').

b. Trade permits consumption outside the PPF, so there are gains from trade.

c. Because technologies are identical, incomplete specialization implies FPE.

Explanation of FPE:

Trade encourages exports of good intensive in abundant (low-priced) factor and imports of good intensive in scarce (high-priced) factor. The adjustment implies more production of export good and less production of import good. Thus, trade raises the demand for abundant factors and reduces the demand for scarce factors.

Implicitly, trade increases competition for scarce factors by importing their services from where they are relatively abundant. THAT IS, IN THE HO MODEL TRADE IN GOODS IS IMPLICITLY TRADE IN FACTORS; TRADE ECONOMIZES ON SCARCE FACTORS. IN DOING SO IT CHANGES RELATIVE FACTOR PRICES.

How do we know that abundant factors are actually better off and scarce factors are actually worse off? Consider the US-China example above. Partial specialization of labor and capital in the US into machinery raises the marginal productivity of capital but lowers the marginal productivity of labor because machinery is capital-intensive. In China the marginal productivity of labor rises and that of capital falls as both move into textiles. But in efficient factor markets, production factors are paid real wages that equal their marginal productivities. So capital enjoys a higher real return and labor a lower real wage in the US. The opposite holds in China.

An alternative mechanism would be trade in factors directly.

We will study trade and wage inequality later.

What if there are many factors? The model is changed somewhat but it still follows that exports are intensive overall in abundant factors and imports are intensive overall in scarce factors. (US: human capital, land; Japan: human capital, physical capital; Canada: natural resources; Mexico: partially skilled labor; China: unskilled labor, etc.)

MODEL THREE: INTRA-INDUSTRY TRADE (IIT)

One obvious difficulty with above models is they predict that trade is inter-industry in nature, eg, machinery traded for textiles. There certainly is much of this trade in the world, especially between countries at different levels of economic development and factor costs. Yet we observe that perhaps 65% of global trade is intra-industry in nature, e.g., types of machinery traded for other types of machinery. The vast bulk of this trade is among the developed economies. Some products in which IIT is common include chemicals, industrial machinery, autos, wine, beer, designer clothing, and many others.

What factors explain IIT?

1. Seasonal growing variations.

2. Natural market areas that contain national borders (eg, trade between Vancouver and Seattle as opposed to trade between Seattle and Detroit).

3. Consumer preferences for variety, or product differentiation. Products may be differentiated by style ("horizontal differentiation") and quality ("vertical differentiation"). True for both consumer goods and producer inputs.

4. Increasing returns to scale in producing particular varieties. For example, consider the automobile industry and its specialization by country.

Elements 3 and 4 combine to explain most IIT. Note that producers have an incentive to innovate new varieties in order to achieve some market power through product differentiation. This incentive to engage in R&D is an important source of economic growth in the world.

How does trade enter? Make a simple comparison of 2 countries in autarky with the countries integrated in free trade. Suppose H and F are about the same size, have the same preferences, and the same technologies.

Autarky: both H and F have isolated markets, which can support only a small number of varieties, operating at low scale, which is high cost. For example, consider that H and F could produce and consume 10 varieties each of good X, at a cost of $50 per unit. Now in free trade the market size is doubled for both countries even though their own factor supplies have not changed. What effects would come from this?

a. Cost effect. Because the market is larger, firms are able to produce at higher scale, achieving lower costs per unit, say $40 per unit. Because scale economies are achieved, production is more efficient.

b. Rationalization effect. The difficulty is that not all firms can expand production, given fixed supplies of labor and capital. Some firms are driven out of business. Those firms that close down are the highest-cost firms, which is beneficial in welfare terms even if it is a hardship for those laid off.

c. Product variety effect. With some firms shutting down, the number of goods produced in each of H and F would fall. However, the combined number of goods available to consumers would rise. For example, if H and F each produce 8 varieties in free trade, consumers can choose from 16 types. This increases consumer welfare.

In general, we'd expect free trade to encourage both inter-industry trade and IIT. The former trade is associated with factor-price changes and possibly significant adjustment costs. The latter is less difficult in terms of factor-demand changes and needs for adjustment. It helps explain why tariffs are higher against products from low-wage countries than from high-wage countries.

MODEL FOUR: THE PRODUCT-CYCLE MODEL OF DYNAMIC TECHNOLOGICAL CHANGE

A final observation is that technologies are not the same across countries (WHY? IT'S NOT AN OBVIOUS ANSWER.). Also, there is continuous technological change as firms engage in R&D to produce new goods and new production processes. One basic model that explains this is the product-cycle theory.

Suppose there are innovative countries, called the North, and countries that are not innovative, called the South. New products are developed in the North and go through a cycle of adjustment that permits their production to be transferred to the South.

Innovation stage: innovation is in some location in the North to meet market demand, save on labor, etc. Production is there initially as well though there may be some transfer through FDI to other advanced countries. Production is intensive in capital and engineering. Exports from innovator to other North countries.

Maturing stage: product is more easily transferred through blueprints, etc. and tends to move toward more advantageous locations, including other North countries. Production is losing its capital intensity. Exports from new locations to South and perhaps back to original innovator.

Standardized stage: product is easily learned due to stable technology and becomes labor-intensive. Production is transferred to South to take advantage of low wages. Transfer is through FDI, licensing, or simple imitation. Exports from South to North.

Implicitly there are two essential processes here: innovation and technology transfer. The higher is the innovation rate, the greater will be productivity and wages in the North. The higher is the tech transfer rate, the greater will be productivity and wages in the South. Thus, there is a tension between these processes, which could be resolved through trade policy, patents, etc. The importance of continuous innovation in North is obvious. The model is driven by preferences for new goods and North-South wage differences.

How does a country move up the "technology ladder"? Note that increasing wages in Korea, etc. reflect rising use and adaptation of technology in production, moving into a stage of becoming an innovative country itself.

Multinational firms wish to have their choices unimpeded about location of R&D and of production.

A FINAL CONCEPT: "OUTSOURCING" OR "FRAGMENTATION"

Production processes may be broken up into stages, such as R&D, marketing, production of parts, assembly, and distribution. Market integration permits greater outsourcing of this production process into component parts across countries. Such fragmentation is driven by differences in wages and productivity. It is interesting that this process seems to concentrate high-end services and design in developed countries. But it tends to shift production activities that are low-skilled-intensive in the developed countries to the developing countries, where they are actually intensive in skills that are somewhat higher than average. The result can be a rising demand for skilled workers in both North and South and a declining demand for unskilled workers in both North and South.