Teaching comparative advantage: barter vs money (bleg)

When we teach comparative advantage in Intro Econ we assume barter exchange. We swap Canadian apples for US bananas. But the students, quite naturally, are thinking of monetary exchange. What happens if Canadian apples and bananas cost more dollars [to produce] than US apples and bananas? That's possible, isn't it? Wouldn't that mean [US prices for apples and bananas are lower than Canadian production costs of apples and bananas, so] there's a shortage of demand for all Canadian-produced goods, which causes unemployed resources in Canada?

The students are right, of course. That might happen. We need to talk about money. Unless exchange rates adjust, or resource prices (wages, rents) adjust in terms of money, it could happen.

But I can't think up a nice simple clean example where it does happen. All I've got is two nice simple clean examples where it doesn't happen. Hence the bleg.

Start with a very simple standard model:

Canada can produce one apple per hectare, or one banana per hectare. The US can produce two apples per hectare, or four bananas per hectare. Because the Americans have a better climate, or better land, or are just smarter than us. So the US has an absolute advantage over Canada in both apples and bananas.

Under autarky (no trade) the opportunity cost and price of an apple will be one banana in Canada, and two bananas in the US, so Canada has a comparative advantage in apples, because Canada has the lower opportunity cost of producing apples. And the US has a comparative advantage in bananas, because the US has a lower opportunity cost of producing bananas.

Traders buy where prices are low and sell where prices are high. When Canada discovers America, traders will take Canadian apples and swap them for bananas in the US, then return to Canada to swap those bananas for apples, and repeat, so Canada will export apples and import bananas, and this arbitrage will equalise Canadian and US prices.

If both countries specialise under free trade, the free trade price of apples will be somewhere between one and two bananas (and equal to one banana if only the US specialises, and equal to two bananas if only Canada specialises). If both specialise, both countries are better off, because their budget lines are now outside their Production Possibility Frontiers. If only one country specialises, that country is better off, and the other country is neither better off nor worse off.

Suppose both specialise and the price of apples is p bananas per apple, where 1 < p < 2.

What happens if we have monetary exchange instead of barter?

Suppose that both countries adopt the apple standard. One Canadian dollar equals one apple, and one US dollar equals one apple, so one Canadian dollar equals one US dollar. Under autarky, Canadian land will rent for $1 per hectare, and US land will rent for $2 per hectare.  Under free trade, Canadian land will all be growing apples, and will still be earning $1 per hectare, but those dollars will now buy more bananas than before. Under free trade, US land will all be growing bananas, and will be earning $4/p per hectare. Since p < 2, US land rents will rise.

Now suppose instead that both countries adopt the banana standard. One Canadian dollar equals one banana, and one US dollar equals one banana, so one Canadian dollar equals one US dollar. Just like before. Under autarky, Canadian land will rent for $1 per hectare, and US land will rent for $4 per hectare. Under free trade, Canadian land will all be growing apples, and will now be earning $p per hectare. Since p > 1, Canadian land rents will rise. Under free trade, US land will all be growing bananas, and will still be earning $4 per hectare, but those dollars will buy more apples than before.

In both my examples, equilibrium land rents measured in dollars either rise or stay the same. So opening up for trade won't cause unemployed land even if exchange rates are fixed and land rents are sticky downwards.

A two factor model could give me an example where either rents or wages might need to fall. But I want to keep it simple.

Any ideas?

(And when you teach first teach comparative advantage in Intro Econ, do you talk about money and exchange rates at all? I don't feel comfortable saying nothing, and telling students to wait till we do macro.)

Comments

[Tom Brown]

Canadian bananas?... Isn't that the same price as Ecuadorian Maple syrup?

[Nick Rowe]

But if we had no international trade, I think Canada would be growing bananas, under glass somewhere, very expensively. The whole point of trade is to destroy the Canadian banana industry.

(BTW, off-topic, in response to your question elsewhere: I think you understand my views on money and banks very well. Sometimes I forget them myself, and am surprised when I read old stuff I have written.)

[Tom Brown]

OK, this post is out of my league... I'm struggling to even understand this sentence:

"What happens if Canadian apples and bananas cost more dollars than US apples and bananas? That's possible, isn't it? Wouldn't that mean there's a shortage of demand for all Canadian-produced goods, which causes unemployed resources in Canada?"

Why does a higher cost for apples and bananas translate into a shortage of demand for all Canadian-produced goods?

[Tom Brown]

Wow, thanks!

[greg]

Are you assuming total production increases? It seems to me with trade the production of apples and/or bananas is going to increase, and so too the quantity of money pegged to which ever fruit.

Totally beside the point, I've always wondered about specialization and the law of diminishing returns. The production frontier of a two good producing economy bows outward, implying that a greater value of goods is produced in a mixed economy, than when that country specializes in one good. This is because some resources which can be efficiently applied to the production of one good cannot be applied as efficiently to the other good, and vice versa. So much of what a country supposedly gains through trade, it would seem to lose through diminishing returns. (This would seem to apply to monoculture, too.)

[Michael Trepas]

Greg

The post assumes - a la Ricardo - stable returns per hectare, that is, a straight Production Possibility Frontier. Hence the total specialization (each country produces one good), without deadweight losses. If we assume diminishing returns, we get only partial specialization (each country produces both goods).

Very interesting post and comments, btw.

[J.V. Dubois]

I think you cannot create an example with unemployed land until you add some disutility of ineffective Canadian production into the model - or alternatively as you say without two factor model. Then you may end up with Scott's model (he used coconuts instead of bananas) where wages frozen above productivity [not high enough to offset disutility] cause "unemployment".

If the model does not assume any of the above, meaning that Canadians have to just pick up bananas/apples for free it is hard to explain why should they not do that and why should they leave the lend "unemployed" instead.

[Kevin Donoghue]

"What happens if Canadian apples and bananas cost more dollars than US apples and bananas? That's possible, isn't it?"

It's not possible, or at least it's hard to imagine, if Canadian and US bananas are perfect substitutes. I'm not really sure what you're after here but I think the simplest way to introduce exchange rates is via the Salter diagram, with a small open economy producing a tradeable good and a non-tradeable service.

[Kevin Donoghue]

Sorry about my spelling. Here's a fairly simple example of the tradable & non-tradable approach, though it might need to be simplified further for complete beginners:

http://www.treasury.govt.nz/publications/research-policy/wp/2009/09-02/01.htm

[Nick Rowe]

Tom: I have edited that bit of my post to (I hope) make it clearer.

greg: Aha! Real output and real incomes will increase if all land is fully employed. So if the quantity of money stayed the same, dollar prices would need to fall. But if they can't fall, we get unemployed land. Yes, that could work. Now can I make it simple and clear enough? Hmmm.

See Michael's response to your other question. If land varied, from being good apple land in the north to good banana land in the south, we would get diminishing returns as the margin of production moved north or south, and a curved PPF, and free trade would move us along that PPF, but probably not all the way to complete specialisation.

Michael: thanks.

JV: OK, but that's like saying land rents would immediately drop to zero if there was unemployed land.

Kevin. The idea I'm trying to capture simply is that if the dollar cost of producing both apples and bananas is lower in the US than in Canada (at a given fixed exchange rate) then Canadian producers "won't be able to compete" in either good. Which is something that can obviously happen if land rents are sticky down, and if the exchange rate is fixed too high. But could it happen simply as a result of opening up for free trade?

[Frank Restly]

Nick,

"In both my examples, equilibrium land rents measured in dollars either rise or stay the same. So opening up for trade won't cause unemployed land even if exchange rates are fixed and land rents are sticky downwards."

Sticky production choices? My father has always been an apple grower. My father's father has always been an apple grower. Even though I could get more money per hectare as a banana grower, I am and will always be an apple grower.

[Nick Rowe]

Frank: any Canadian farmer who insisted on growing bananas and selling some of those bananas for apples would be worse off as a result of free trade, even in a barter economy. But that's not the nightmare scenario opponents of trade have in mind. They fear that free trade will make Canadian farmers unable to compete with US farmers in either good.

[Frank Restly]

Nick,

If the US has unemployed (idle) land that is not being used toward the production of either apples or bananas then it is probably a valid concern.

Ever notice how trade frictions become elevated during recessions and fade during economic booms? Add a third good to your model (oranges) where neither country has a comparative advantage and then take away the local demand for that good.

[Min]

I understand your desire to make things simple for your students. :) However, may I suggest that this is a good example for a computer simulation? Students could see graphically the effect of different conditions and assumptions. :)

[Tom Brown]

Nick, yes, much more clear. Thanks.

[Jacques René Giguère]

I don't really talk about exchange rates. I tell them they would adjust and somehow take care of some of the adjustments. They trust me that those interested will see it later at university? Maybe the Québec system of separate Cegeps and university help in keeping the subjects comfortably sequential...

[Jacques René Giguère]

Does keeping money out of it helps or hinder understanding that if we buy French wine but they don't buy anything from us, we are in the first stage winner as we get something for free? My father doesn't buy french wine because "They don't buy much from us." Does the introduction of money helps understand that the tar baby is shifted and that in the end, let's say Croatia will buy planes from Bombardier?

[greg]

Nick: "But that's not the nightmare scenario opponents of trade have in mind. They fear that free trade will make Canadian farmers unable to compete with US farmers in either good."

I don't know if you can make the case with 'free' trade. I doubt also that you can make a case in a simple model that assumes any increase in production is still going to be consumed. I did make a case for idled resources in a country with a trade deficit. The argument is based on the idea of the Aggregate Supply Curve for a deficit country shifting to the right, since it is now 'supplied' with what it produces and its excess imports. This lowers its price level along the Aggregate Demand curve, and reduces the revenue of its producers (though along the original AS curve.) This causes the deficit nation's entire economy to contract. More detail at: http://anamecon.blogspot.com/2010/04/effects-of-unbalanced-trade.html

I think as long as you assume that US exporters (have to) spend their Canadian dollars on Canadian fruit, Canadians will still be employed. If the US doesn't spend their Canadian dollars, then Canadian producers have a problem.

[Lorenzo from Oz]

Tell students that Ricardo developed comparative advantage under a gold standard where the price level was set by the value of gold and they can wait until later when you introduce money (i.e. where currency is the medium of account and sets the price level).

If you want an actual barter example of comparative advantage--hunter-gatherer societies. In all such societies studied, men hunt and women gather; a clear operation of comparative advantage. There is one apparent exception, but it was a pure hunter society which traded meat with the local farmers for edible plant products. Another example of comparative advantage. Citations in The Origins of Virtue by Matt Ridley.

Perhaps you should also assume that gains from trade are more basic than comparative advantage?

[Peter N]

Where does the exchange rate come from which makes both goods more expensive in Canada? If apples and bananas are the only traded goods or services this should be impossible. To get this you would need another commodity, commodity money (like gold) or Americans investing in Canada. With another source of US dollars, sellers holding US dollars have nothing to buy with them but US apples and bananas, Canadians will buy US bananas and sell Canadian apples as long as there is an arbitrage profit from circular trade.

Your model definition is partially unspecified. You can always generate exogenous factors that affect a model. Its your model, after all.

A lot of blog controversy comes from the two sides assuming different exogenous influences,

[Min]

FWIW, Burson and Repede's TradeSim software may be of interest. :) http://www.na-businesspress.com/bursonweb.pdf

It probably would have to be modified to include money, but they may be interested.

AFAICT, it is not simulation of economic actors who approach an equilibrium over time. (Or not! ;) )

[Nick Rowe]

Min: I would never be able to understand that computer simulation myself, let alone explain its results to the students! And in any case, I would first need to figure out what model to tell it to solve.

Peter N: if I had gold in the model, as a third (non-produced) good, I could assume that both Canada and the US used gold as money. Then if I assumed that Canada in autarky had a lot more gold than the US (relative to autarky production of apples and bananas) Canadian prices in gold could be higher than US prices for both apples and bananas. That could work. Then gold would flow from Canada to the US, until Canadian prices fell and US prices rose. David Hume's price-specie flow mechanism.

Hmmm. I rather like that. A little more complicated than I wanted, if I wanted to put actual numbers on it. But the story would be simple enough to tell.

greg's approach would also be a good one, that could be used with flexible exchange rates. Canada has a stock of gold, and the US has a stock of silver, and we use gold as money and they use silver, with flexible exchange rates between gold and silver. But opening up for trade increases real income in both countries, which increases the demand for money, and causes a recession until prices fall to the new equilbrium.

Hmmm.

Lorenzo: "Perhaps you should also assume that gains from trade are more basic than comparative advantage?"

The way I teach it, all gains from trade come either from differences between people, like comparative advantage, or else from economies of scale. So I think gains from trade is not more basic than comparative advantage. Gains from trade are a result of (or the same thing as) comparative advantage.

Jacques Rene: I suspect that most people who teach it keep money out of it. Perhaps because most who teach it are microeconomists, and (no offence) microeconomists are usually less comfortable with basic macro than macroeconomists are with basic micro. And maybe that's the right approach.

[Peter N]

I think that remittances may be better than gold. The idea of one trade party having to realize value on another's currency is an important one to illustrate. That and possibilities for arbitrage supply all the mechanism you need.

Regardless of whether both apples and bananas are more expensive in Canada or less expensive, you can still make a profit selling apples to and buying bananas from the US. There are a number of ways this can end, but it has to end, unless it is protected. An example of protection would be the difference between official and import use exchange rates of the dollar and the Bolivar, which causes flights out of Venezuela to be fully booked, but fly half empty. You have to show an airline ticket to get the better exchange rate. This is a peculiar and wasteful form of arbitrage.

[Jacques René Giguère]

Nick: "Perhaps because most who teach it are microeconomists, and (no offence) microeconomists are usually less comfortable with basic macro than macroeconomists are with basic micro." True. That's why John Cochrane, among others, should not comment on macro (sigh). But trade is first and foremost a micro thing, an efficiency thing.Nothing macro here.
It become macro only when you make it large scale and introduce money. And then,of course,great fun is had by all.

[Min]

If you don't mind, I have a question.

I took another look at Krugman on the difficulties of the concept of comparative advantage ( http://web.mit.edu/krugman/www/ricardo.htm ) in which he discusses assumptions and conditions for comparative advantage. At one point he states:

"The basic Ricardian model envisages a single factor, labor, which can move freely between industries. When one tries to talk about trade with laymen, however, one at least sometimes realizes that they do not think about things that way at all. They think about steelworkers, textile workers, and so on; there is no such thing as a national labor market. . . . First, unless it is carefully explained, the standard demonstration of the gains from trade in a Ricardian model -- workers can earn more by moving into the industries in which you [the country] have a comparative advantage -- simply fails to register with lay intellectuals. Their picture is of aircraft workers gaining and textile workers losing, and the idea that it is useful even for the sake of argument to imagine that workers can move from one industry to the other is foreign to them."

I get Krugman's point. :) However, it seems to me that there is a conflict of levels here, between comparative advantage and specialization at the national level and comparative advantage and specialization at the personal level. Fully cashing in on comparative advantage at the national level seems to require overriding comparative advantage at the personal level. Workers cannot specialize at what they do best. (Or they have to move to another country.) Does that make sense? Thanks. :)

[Jacques René Giguère]

Min: that's where money comes in by adjusting exchange rates...

[Min]

Thanks, Jacques. :)

But workers are still required to switch jobs, right? Regardless of any personal comparative advantage, right?

In Nick's example, suppose that instead of growing apples vs. bananas, the two options are growing apples vs. making bicycles. Not so easy to switch, eh?

[Nick Rowe]

Min: if the PPF is rectangular (reverse-L shaped) because labour is totally immobile between sectors, there are no gains from trade due to comparative advantage. (There will still be gains from trade by changing the consumption mix.)

Jacques Rene: but if there are no gains from trade, even in principle, flexible exchange rates won't help.

[Jacques René Giguère]

Nick : Of course. But there would be no trade anyway, except maybe some random accidents...

[Min]

Nick Rowe: "if the PPF is rectangular (reverse-L shaped) because labour is totally immobile between sectors, there are no gains from trade due to comparative advantage. (There will still be gains from trade by changing the consumption mix.)"

Thanks, Nick. :)

My question was not about the immobility of labor between sectors, per se, but about specialization. IIUC, comparative advantage is applied to individuals as well as nations. The arguments I have seen apply to nations as players. They should equally apply to individuals as players. And in fact that is done, right? But one of the assumptions of the value of comparative advantage specialization at the national level seems in general to mean that individuals cannot apply it to themselves. If they already specialize in what they did best, then to change jobs in order to enable comparative advantage at the national level, they would have to do what they do less well. (Or they could emigrate to where they could still specialize in what they did best.)

[Lorenzo from Oz]

I would have thought that comparative advantage is a counter-intuitive consequence of gains from trade + opportunity cost. Thinking of the basic history of (long distance) trade, such trade started between people who had access to completely different goods. Comparative advantage is not a good way to think about such trade, because each was trading things they produced for things they could not produce. (Except in the sense of choosing between producing the tradable good and non-tradable goods -- i.e. opportunity costs + gains from trade.)

So yes, gains from such trade came from differences, but not comparative advantage differences in the "each side can produce the same tradable goods" sense. Comparative advantage in that sense can only apply where both could produce the same goods, but have differences in opportunity costs between producing those goods. But it can also only apply where there are gains from trade to be had. So, setting up an example where there are no gains from trade to be had for one of the countries is not a problem unless one thinks that trade is somehow compulsory. The trade has to be motivated in the first place. After all, if it is a barter economy Say's Law applies.

Introduce money and suddenly exchanges are not time-bounded in the way that Say's Law presumes. So, yes, one side can run down assets in trade, but that also has a limit.

[Lorenzo from Oz]

To put it another way, gains from trade come from capacities + preferences. Comparative advantage applies to a particular (if every wide) conditions of capacities + preferences.

[Lorenzo from Oz]

That should be "very wide"

[Michael Trepas]

Nick

For the record: If I was a student of yours, the price-specie mechanism would suit me fine.

One real-world example I can think of is the intra-Eurozone trade:

it is literally free (no barriers or tariffs)
there is a "fixed" exchange rate (the common currency)
there is no meaningful labour mobility

Long story gone short, the South ended up with a rising price level relative to the price level of the North and, eventually, with idle resources.

Not exactly the topic we are discussing here, but I think the students tend to value vivid examples and discussion.

[Brian Romanchuk]

All I can see is the need for time dynamics. For example, you could get unemployed resources from a transition to free trade if you have producers that were previously profitable go bankrupt under the new free trade regime. But those resources would be eventually returned from bankruptcy to production in the new equilibrium. If my interpretation of this is correct, why not quickly explain this issue, and that a more complete response will be covered in a more advanced course?

[kevin quinn]

Nick: I'm late to the party and perhaps I don't get it. Under fixed exchange rates, if Canadian prices are higher for both goods, won't the Canadian money supply fall via capital outflow until Canada is competitive in one of the goods - the one in which it has a comparative advantage?

[Nick Rowe]

kevin: yes. That's David Hume's price-specie flow mechanism. But if prices are slow to adjust, there could be unemployed resources until they do adjust.