Trade and External Economies of Scale - YouTube

Today is the first of a series of lectures on trade and economies of scale.

In this one, we are going to look at trade and external economies of scale.

We've already looked at a number of theories which implicitly made predictions

about which countries should trade the most.

So comparative advantage says the trade ought to occur most

between countries which have different levels of productivity.

The Hecksher-Ohlin theory says that trade ought to occur the most

between countries with different factor shares.

So both of this theories imply that countries should trade the most

with countries from which they are the most different.

However, a large fraction of trade

actually occurs between countries which are quite similar

in productivity and in factor shares.

For example, there is a huge amount of trade between US and Canada,

and, more generally, there is a lot of trade between

the United States, Canada, Western Europe, and Japan,

the developed economies.

Now, this doesn't mean that

comparative advantages and Hecksher-Ohlin theory are wrong,

but it does certainly mean that they are incomplete,

and we want to look at some other theories

which may help us to explain some of the facts about trade.

In particular, we are going to be focusing on economies of scale,

which come in two types.

First, external economies of scale.

These are at the industry level, and I'll explain this more as I go along.

Then, second, internal economies of scale, at the firm level.

It turns out that the second is more

difficult to analyze analytically

because it requires us to understand

monopolistic competition;

we can't have straightforward competition

with internal economies of scale at the firm level.

So we'll talk more about that in a future lecture.

Today, we are going to focus on external economies of scale.

Let's take a look.

An important fact about many industries

is they are not spread randomly around the country or around the world.

Take carpet for example.

For over 100 years, the city of Dalton has been known

as the carpet capital of the world.

Over 100 firms produce carpet in Dalton, Georgia,

and they, combined,

produce over 70% of the world's entire

production of tufted carpet.

Now, not all industries cluster.

So which ones cluster? And why do they cluster?

Why did Dalton, Georgia become the carpet capital of the world?

Industrial clusters are very common.

There's a small town in China

with a population of just 36,000

that makes over 3 billion toothbrushes a year.

Hangji, China, produces 80% of China's toothbrush production.

There are over 990 manufacturers in the Hangji cluster,

and toothbrushes have been made there since 1827.

That's another interesting fact about this clusters.

Once they start,

they tend to remain in place for a long time.

They're very stable.

Here is another cluster with which you are very familiar, Hollywood.

Why is it that so many movies,

not just in the United States, but for the world, are made in Hollywood?

And if they're not made in Hollywood,

they're probably made in another industrial cluster,

Bollywood, in India.

In order to understand why some industries cluster,

it's useful to think about how a cluster evolves.

And these principles have repeated themselves over and over again

in many different types of industries.

So typically, first, one firm enters

and it might enter at a particular location

due, let's say, to some geographic factors

which lower its costs modestly in that location.

So for example, California is good for filming all year round,

so it made some sense for an early movie studio

to begin in California.

The first entrant doesn't have to have cost advantages, however.

Dalton, Georgia got its start

when a young girl hand-crafted a bedspread

which proved particularly popular.

She got her friends making similar bedspreads

and this gave rise to an entire industry.

Now, think about a second firm.

It looks around and asks where should it locate,

and it notices that by locating near the first firm,

it can take advantage of some specialized labor, suppliers, of shippers,

of some ready-made infrastructure.

So the second firm enters at the same location.

The third firm finds cost advantages of locating near the first two

even greater,

and so a cluster begins to develop.

Until pretty soon,

this one spot happens to be the lowest-cost place

to producing that particular type of good.

Let's think about Hollywood, for example.

In order to make a movie, you need very quickly

to assemble a large team of --

specialized, talented labor.

If you're making a movie in Dubuque, Iowa,

and your cameraman gets sick and, heaven forbid, dies,

it's going to be really hard to find another cameraman.

If you are making the same movie in the LA area,

you can have another cameraman there in hours, --

and suppliers.

If you need, in your movie, --

suddenly you need a rewrite and you need to have an alien costume

and a Roman temple,

you can have all those goods delivered to you.

There are prop shops built, --

large prop shops which service the entire industry.

So, at low-cost, you can find things that you want very quickly.

The same thing is true for carpets,

which are created through specialized machinery,

which requires informed, educated, labor.

So it's no surprise that --

the community colleges in Dalton, Georgia

also specialize in teaching people the skills used to make carpets.

Even producing toothbrushes can require a lot of specialized machinery,

and it helps for a firm to be located in the Hangji area

because it can easily find the labor and the expertise

that it needs in order to build toothbrushes.

So this is how an industry evolves to become a cluster,

and when it is clustered,

it becomes the lowest-cost place

to produce these particular types of goods.

Clusters can also help to create

new innovations, and inventions, and improvements

which lower the cost curve even further at any different quantity.

So the great economist Alfred Marshall said,

This is definitely the case in Dalton, Georgia, and in Hanji, China.

So again, once a cluster becomes established,

the costs fall quickly in that cluster, first in that cluster.

They fall there before they fall anywhere else,

and, again, this is the reason why a firm wants to be located

in a cluster with other similar firms.

External economies of scale imply that as the industry gets larger,

the costs of each firm in that industry begin to fall,

and this means that the supply curve for the industry

can actually be downward sloping,

at least over a part of its range.

So the supplier industry average cost curve,

in contrast to the usual supply curve, which is upward sloping,

can actually slope downwards particularly

as the cluster is being established.

So, when Dalton, Georgia is small,

the carpet industry has high costs.

But, as the industry in Dalton grows larger,

the costs to each firm in that industry begin to fall.

Notice that what this means is that, when an industry is large,

it's going to be very difficult to start an industry anywhere else

because the costs to production are lowest in Dalton, Georgia,

where it already has economies of scale,

where the economies of scale have already been built up.

Let's now see what implications there are for trade

when we have external economies of scale.

So let's suppose there are two industries, toothbrushes and carpets,

two countries, the United States and China,

or Canada, if you like, it doesn't matter which,

and let's suppose that we have no trade.

So, in the initial equilibrium,

let's suppose that the US demand is in blue,

and the Chinese demand, we'll say, is in green.

Same thing over here. US demand for carpets is in blue,

Chinese demand for carpets is in green.

So the demand is the size of the market in these two countries.

Given the size of the market, given the demand in the United States,

the price of toothbrushes in the United States will be here,

and the quantity of toothbrushes sold will be here.

Same thing for carpets.

The price of US carpets is given here and the quantity here.

China has got a, let's say, slightly bigger market than the US market.

So, because of economies of scale, it has a lower price of toothbrushes

and a bigger quantity, and same here,

a lower price of carpets and a bigger quantity.

We can mix and match these however we want,

it doesn't really matter.

Now, here's the point.

Suppose that we now introduce trade.

Well, what will happen?

Well, one thing which could happen --

is that these industries cluster --

in just such a way that we get more of both goods.

So for example, let's look at the total world demand for toothbrushes,

which is simply just the US demand plus the Chinese demand.

So this is the total world demand for toothbrushes.

This is the total world demand for carpet,

again, just the US demand plus the Chinese demand.

Now, if now we have a cluster

such that all the toothbrushes are made in China,

and all the carpets are made in the United States,

then the cost of producing toothbrushes falls,

the cost of producing carpets falls.

So, now, China produces a lot more toothbrushes;

their quantity of production increases.

The US produces a lot more carpet,

but because the cost of both of these goods has fallen,

due to taking advantage of these external economies of scale.

China and the United States can trade carpet for toothbrushes,

and both countries can be better off.

So, by taking advantage --

of larger markets,

larger markets allow you to extract everything possible

from these external economies of scale.

By building the toothbrush industry bigger in one location

and building the carpet industry in Dalton, Georgia,

by making that larger,

we can take advantage of all the possible external economies of scale,

push our cost down to the minimum level, and then trade.

And since cost for both products have gone down,

both countries will be better off.

Let's look at two interesting facts in the presence of

external economies of scale.

First I'm going to call this "the big get bigger" effect.

Let's take a look.

Suppose there are two countries.

We're going to call them <i>Big</i> and <i>Small</i>.

The <i>Big</i> country is not necessarily big in size,

but it has a big market for this particular product.

So here is the demand curve for <i>Big</i> here is the one for <i>Small</i>.

Also, we are going to suppose that --

the average cost curve, or the supply curve, in the <i>Big</i> market

does not fall as far as in the <i>Small</i> market.

In other words, at any particular quantity,

the cost of producing that quantity of good

is lower in the <i>Small</i> market than they are in the <i>Big</i> market.

However, the <i>Big</i> market starts off big,

so the <i>Big</i> market starts off with a cost advantage

due to the fact that it can take advantage of these economies of scale.

So, now, notice what happens if we now introduce free trade.

Well, if we have free trade, and here is our world demand curve,

we made end up in what I call the <i>historical equilibrium</i>,

that is, something plausible which happens is that the <i>Big</i> market gets bigger.

Because it begins with a cost advantage,

it's able to outcompete the suppliers in the small market

so that we end up with the big market taking the dominant share,

taking the world market.

That historical equilibrium is opposed to the <i>ideal equilibrium</i>.

If everything was ideal,

the location of the cluster would be in the <i>Small</i> market

because that's where you reduce cost the most.

But because the <i>Small</i> market starts off small,

it ends up even smaller.

The big get bigger, the small get smaller.

So, in theory,

what this means is that we don't always end up at the ideal equilibrium.

This is a little bit sad, probably not too much to worry about

because what we really want to do is make sure that we take advantage

of the trade, to take advantage of these economies of scale.

But it does mean that we don't necessarily get to the ideal equilibrium.

The chains of history bind us

so that we may end up at the historical equilibrium.

The second effect,

the &quot;there can only be one&quot; effect is closely related.

There can only be one Dalton, Georgia.

There can only be one Seattle, Washington, or Silicon Valley, or Hollywood.

I mean, literally, this is not true but there certainly aren't going to be many.

So many some people argue

that, in order to be the one, we should have strategic trade policy.

That is, there are some benefits to be in Dalton, Georgia,

or to be in Seattle, Washington, or Silicon Valley,

and if we could we would like to be the ones who eat up those benefits.

So what these people suggest is that

we ought to, for example, protect our industries.

We ought to throw up tariffs and barriers to trade,

so that our small industries become big.

We allow them to grow

until they get the cost advantages of economies of scale,

and, only then, do we open them up to world trade.

And because we only open up to world trade;

once our industries have got these economies of scale,

we're then be able to dominate the world

and take over these industries.

We may end up at the historical equilibrium or, if we are lucky,

we may end up at the ideal equilibrium

if we happen to start off small and then we grew our industry.

But, either way, the point of strategic trade policy

is that we end up being the one.

Now, again, this idea works in theory.

In practice, it can be a little bit dangerous.

If every country tries to be the one,

then every country may end up trying to protect its industries,

trying to grow its industries behind these tariff walls,

and we may never get the advantages of the economies of scale

which free trade can produce.

So, for the world as a whole is probably more important

that industries cluster around the world,

and then take advantage of clustering, and then trade

than it is that we are the ones who end up with the cluster.

It would be nice if we did,

but we probably can't tip the balance that often.

It's going to be very tricky to be able to tip the balance in our favor

and the times in which we are able to tip the balance,

probably do not justify the number of times where,

because of the tariffs and the quotas,

we don't get all of the economies of scale that we can get.

Anyway, that is my opinion from evaluating this literature,

other people may differ

but strategic policy is something that can work in theory.

One more point before we conclude.

External economies of scale and dynamic economies of scale

can also occur across industries.

That is, the ideas in one industry can cross-pollinate or fertilize

ideas in another industry.

It can lower cost in another industry.

Having a lot of smart people together

in an area where they can communicate with one another,

you never know what idea is going to be particularly fruitful,

be particularly profitable

in a place where the first person might never have thought to apply that idea,

but the second person can apply that idea.

These are also called the agglomeration economies

or synergies

and this most fundamentally explains why we have cities.

After all, there are a lot of disadvantages to cities.

There are congestions, it's more expensive to live in a city,

costs are higher in a city.

And, yet, people are more productive in cities.

It's not just that more productive people

go to live in cities, although that happens too,

it's that the same people are more productive in a city.

They generate more ideas when they are around other people

who generate ideas.

So, this explains why --

the external economies of scale and international trade

also have implications for economic geography,

and it's why Paul Krugman, for example, made contributions to both fields.

Anyway, perhaps we'll say more about that another day,

but it's a little bit far afield.

So we'll turn now to further references.

So Paul Krugman's 1996 paper does a good job explaining increasing returns

and how they apply to trade and also to clusters.

Clusters are also discussed in Michael Porter's book

from a business point of view.

This is a more empirical point of view, &quot;The Competitive Advantage of Nations&quot;.

The World Bank, their 2009 report has got a lot of information

on clusters from all over the world.

And Paul Krugman's 1991 book &quot;Geography and Trade&quot;,

explains the link in greater detail between international trade,

the ideas of international trade, and the ideas of economic geography.

Thanks!