tariffs and world trade

If country A enacts barriers to trade, lets say tariffs on steel for example, is it in country B’s best interest to enact similar barriers to trade, to ensure there is “fairness”? Or is Country B always better off economically without trade barriers, regardless of any sort of barriers that Country A may enact?

Put another way, is it still economically beneficial for a country to have little trade barriers regardless of what any other country may do regarding free trade?

And the obligatory, why is that?

According to my economics textbook, it makes sense to abolish your tariffs regardless of the level of tariffs set by your trading partner.

The follows from the Theory of Comparative Advantage which is, at bottom, a mathematical argument.

Whenever a country raises its tariffs, some of its citizens will gain income while others will lose. However, the gains by the winners will outweigh the losses by those who lose income. This result obtains regardless of the level of tariffs set by the other trading partner.

Caveats:

We are ignoring costs of adjusting from one productive endevour to another.

Also, we are ignoring the possibility that a threat to raise one’s own tariffs may discourage your trading partners from raising their tariffs.

Here’s a thought experiment that might clarify the issue.

Let’s say a scientist invented a magical machine that allowed you to feed grain in one end, and fully built cars came out the other end. The price of the grain fed into the machine was less than what it would cost you to make cars any other way. Is there any doubt that the economy would be better off with this machine? Sure, car makers in the old, more inefficient industry would be hurt, and there would be dislocations as they retrained for other, more productive work. But in the end, your magical car machine would be a boon for the economy. More goods being produced for the same investment.

Now call that machine “Japan”. Put grain or other goods they want onto ships, and sail them over the horizon. Wait a while, and the ships come back loaded with cars.

Both replies are excellent, but there is a wrinkle: the optimal tariff is not in general zero. This is because countries do not really face infinitely elastic demand curves for their exports. This means that for pretty low tariffs (say 10% in the case of Australian textiles) the benefits of trade expansion are largely or more than offset by a deterioration in the terms of trade (the price of exports relative to imports). A non zero tariff (or equivalent export tax) will be welfare improving.

Caveats to this are (i) it ignores adminstrative costs; (ii) it ignores the fact that having a tariff regime makes it easier for protectionists to get increases in tariffs; and (iii) it assumes that economists have the talent and the data to calculate the correct value.

But it’s important for two reasons: (i) in the neighborhood of zero, there are better things to do than worry about tariffs; and (ii) it partly explains why we’re seeing so much action in the “free trade agreement” area at the moment. At least in most developed countries, tariffs are now fairly low. You just don’t see the rates of 50 - 100% that you used to. This means that the unilateral gains you can get from tariff cuts - which are as has been mentioned subject to subtantial adjustment costs as well as great political pain - are pretty much exhausted. At rates of 5-10% you’ve reached the point where demanding reciprocity in lowering trade barriers makes sense.

Sam, et al,

I’m not understanding your example very well. Can you dumb it down a bit for me?

Does enacting protectionist measures in response to a specific country’s trade barriers hurt the original country that is responding, or does it hurt the world economy in general? I thought it was a “diffused cost” problem, where enacting retaliatory trade tariffs helped the country doing so in the short term, but hurt the world economy in general.

---- I thought it was a “diffused cost” problem, where enacting retaliatory trade tariffs helped the country doing so in the short term, but hurt the world economy in general.

One would get that impression from the media. It is, however, mistaken. Small countries, who cannot affect the tariffs of their trading partners, should set their tarrifs at zero. <<Hawthorne, I’m ignoring your point at present.>>

You see, most have absorbed a prejudice towards free trade, without understanding (or even being exposed to) the underlying argument.

Sam’s magic machine is a decent way of thinking about it. You can think of the ability to trade cheap wheat (which the US grows efficiently) for expensive coffee or as an amazing technological innovation. Yet that is what free trade permits: it permits each country to produce what it is best at, relatively speaking.

<<Alas, this machine argument invokes absolute advantage, where in fact trade follows comparative advantage.>>

For a college-level presentation, see the following:
http://internationalecon.com/v1.0/ch40/40c000.html

Hawthorne —

From what I can tell from my international trade text, the terms of trade argument hinges on the tarrif-imposing country being able to drive down foreign export prices.

(eg, the US slaps a 10% tariff on textiles, domestic prices go up, but this is offset by a shift along the exporter’s supply curve resulting in lower pre-tax textile prices.)

If the importer is small, or the market that we’re discussing is reasonably competitive (as in textiles), I would argue that the optimal tarrif would be close to zero. That is, I am claiming that the relevant demand elasticities are indeed close to zero for products not produced by oligopolies.*

(There’s another argument: when the tariff is low, the distortion is also likely to be low and may not be worth the political effort of offending the relevant constituency.)

  • In the oligopoly context, it becomes an empirical question. Catchword: Strategic Trade Theory.

Hilarious. I mean I think the demand elasticities are close to infinite.

The question of what these trade elasticities are is still pretty controversial, Measure for Measure - partly because they’re not at all easy to observe. But most applied economists don’t think they’re very high any more. The folks over at GTAP work with -5 IIRC, whereas say 20 years ago people thought they were more like -20 (which is close to - infinity for a parameter like that).

If export elasticities were really high we would (obviously) expect to see large changes in exports for small changes in prices, but mostly we don’t. And on the flip side we’d expect to see small changes in tariffs pretty much eliminate domestic industries - and yet despite large changes in protection in industrialised countries, they still have textile industries. The reason seems to be that “textiles” isn’t a good, it’s a large bundle of goods that consumers (and firms) seem to think are slightly different. Once we’re in a world of differentiated products we’re in a world where every producer faces a downward sloping demand curve for their product. Sometimes it’s obvious on reflection why the products are differentiated, sometimes it’s a mystery.

If the small country assumption were really right for export markets, no export industry would talk of “looking for new markets” or having “traditional markets” or lobbying for trade promotion in foreign countries - they’d be happy with the amount they were selling given the fixed world price.

So the empirical reason that economists revised their thinking on this is that we couldn’t get our models to fit with what actually occurred when tariffs changed. The sociology of science reason is that monopolistic competition models became fashionable again.

A tariff costs more than it saves. For example, “to save one American autoworker’s $50,000 job through tariffs or import quotas, car buyers collectively pay an extra $150,000 a year through higher prices.”

To put it another way, tariffs create distortions, and distortions are necessarily costly because we have to use resources to adjust for the distortion as well as suffering the cost of the intervention. They’re costly no matter what the other guy is doing.

I think that in practical terms, Sam hit the nail on the head: Responding to your tariff with a tariff of my own is costly to me, but it may be worth it as a way to get you to work with me. That’s the only logic that I can think of that really works for me.

Hawthorne’s remarks notwithstanding, of course. However, it is not obvious to me that the optimal tariff should change in response to a change in someone else’s tariffs. In other words, if a 10% tariff on steel is optimal, why would that optimum change because another country slaps a large tariff on our wheat?