"New" Trade: Product Differentiation & Scale Economies

Last Updated: March 10, 1997

© Copy right 1997 by Keith Head. Do not reuse without permission.

The idea of comparative advantage is that as long as countries differ it is always possible for trade to take advantage of those differences. This suggests that

Countries with very different factor endowments (e.g. Germany and China, or Canada and Japan) should trade a lot with each other.
Flows of goods that use the same factors intensively should just run in one direction.

Facts that FPT has trouble explaining:

Large amount of trade between countries with similar relative factor endowments. (France & Germany)
Large amount of two-way trade in goods whose production involves similar factor proportions.(aircraft example in class)

Product Differentiation

One way of explaining the facts above is that trade consists mainly of exchange of differentiated products. However, product differentiation cannot usually explain trade by itself. Why? Individual firms might develop unique products and patent them so they cannot be imitated by other firms (foreign or domestic). That does not mean they will want to export, however. If there are even very minor transport costs, it will be attractive to produce the good near to customers. As a result, firms would produce their unique varieties abroad, either by licensing to foreign firms or by setting up overseas subsidiaries.

In a few cases, however, the source of differentiation is immobile, e.g. if specialized inputs (labour or land) are not internationally mobile or techniques of production are impossible to transfer internationally. In that case trade could arise through product differention alone. We will refer to this as Armington trade.

Examples of Armington (nation-of-origin) product differentiation:

Japanese defect-free cars (???)
Resort destinations (except Club Med).
Wine (Bordeaux) and coffee (Juan Valdez).

Acquired Comparative Advantage

Consider the following example of specialization without (ex ante) comparative advantage. We will assume that a traditional method for making cars exists where each car requires 50 worker-hours. The traditional method works equally well for any type of motor vehicle and for workers in any country. Hence, the productivity matrix looks like:

Vehicles per 100 worker-hours

(traditional methods)

	Sedans	Minivans
Canada	2	2
USA	2	2

Neither country has comparative advantage so if the good is at all expensive to transport, then each country would produce both types of vehicles and no trade would take place.

Now suppose it takes 1000 hours to set up a product-specific assembly line but then it only takes 25 hours per car (productivity = 4 cars per 100 worker-hours). If both countries set up assembly lines for both products there would still be no comparative advantage. But if one country (say Canada) built a minivan assembly line, it would acquire a comparative advantage in minivans. The other would necessarily acquire a comparative advantage in sedans. The comparative advantage would be further reinforced if it also built a sedan assembly line.

Upfront investments scale economies gains from specialization.

Scale Effects:

Static: Fixed Up-front Investment in Product Development or Plant Capacity.
Dynamic: Learning-by-Doing
unit costs decline with cumulative output.

Implications

First-mover Advantages: Location of production driven by history (government, chance) rather than "fundamentals."
Home-market-size effects: Big markets attract firms because they would rather produce close to large numbers of consumers and only incur trade costs on their flow of exports to the small markets.

NOTE:

Factor Abundance is still important in a world with scale economies and product differentiation. How? Investments in big plants or product development will be lower cost in countries with abundant capital (i.e. high savings rates). Thus it is not surprising that Japan and Germany are net exporters of many capital-intensive goods.

Previous Note
Next Note
Return to Home