Trade protection and growth

Ugh, yet again!  But the question of whether trade liberalization/protection promotes or retards economic growth is one of those venerable topic of discussion in economics that simply refuses to go away.  See for example Ha-Joon Chang's recent op-ed in the FT.  One reason is that much of the discussion is driven by pre-conceived ideas (on the efficacy of markets versus governments) instead of actual evidence.

Two recent papers represent a significant advance.  One, by Lehmann and O'Rourke focuses on the late 19th century, while the other, by Estevadeordal and Taylor, looks at the last thirty years.  The chief contribution of these two papers is that they actually differentiate between different types of tariffs.  Lehmann and O'Rourke distinguish between tariffs that protected industry, tariffs that protected agriculture, and tariffs intended to simply raise revenue.  Their conclusion:

Industrial tariffs were positively correlated with growth. Agricultural tariffs were negatively correlated with growth, although the relationship was often statistically insignificant at conventional levels. There was no relationship between revenue tariffs and growth.

The sample of countries is a group of mostly developed countries over the period of 1875-1913. 

Estevadeordal and Taylor, meanwhile, distinguish among tariffs on capital, intermediate, and consumer goods (plus they use an imaginative identification strategy to alleviate reverse-causation concerns).  They find that it is mainly tariffs on capital and intermediate goods that retard growth, while tariffs on consumer goods have a much weaker effect.

Now, the two sets of results are somewhat in tension with each other, and it is not clear whether the differences are due to differences in statistical methods, or to the fact that the late 19th and late 20th centuries were inherently different, with the former being a period in which protection of industrial goods was good for growth while the latter was one where, at best, it was not too damaging.

Regardless of reconciliation, the bottom line is this: what matters is the structure of protection (what is being protected).  The answer to the age-old question is one that economists should be accustomed to giving: it depends.

And of course one thing that it depends on is the overall state of the global macro-economy.  At a time when the world is digging deeper into recession, exporting your problems through trade protection is the last thing that any responsible country should be doing.

Comments

[lark]

Of course, Chinese export subsidies and currency manipulation are a form of trade protection.

[James]

"...one thing that it depends on is the overall state of the global macro-economy."

Why's that? If certain kinds of protectionism work then why should they only be used when the world economy is OK?

[the dude]

The reason all Protectionism is wrong is that you can't raise tariffs on one group of products without risking that your trade partners will raise trade barriers on a different product that you sell. Then you have a vicious downward cycle. PLEASE, PLEASE, PLEASE don't ruin the half century of progress by promoting protectionism, or else dark day are indeed upon us.

[kirsten duntz]

by "imaginative" identification strategy, did u mean "no" identification strategy? Diff-n'-Diff won't solve an endogeneity problem...

[Thomas]

The important issue is the policy alternative. Is placing tariffs on consumer goods better than taxing imported and domestically produced consumer good equally? Are tarifis on specific goods better than producting subsidies to those goods? The argument against protection is generally taht for any given policy objective you can usually find a better instrument.

[save_the_rustbelt]

In the "short run" American blue collar workers are getting the stuffing beat out of them.

Standard of living - gone. Pensions - gone. Job security - gone. Need I go on?

So will it really be better in the "long run?"

[save_the_rustbelt]

In the "short run" American blue collar workers are getting the stuffing beat out of them.

Standard of living - gone. Pensions - gone. Job security - gone. Need I go on?

So will it really be better in the "long run?"

[Tord Steiro]

Perhaps some of the differences in results can be attributed to the effect of financial liberalization in combination with tariff-slashing?

Imagine this:
A previously poorly governed country, like, say former Communist country, experiences a political revolution. The government is democratized, and the economy is turned into a capitalist one, with economic (perhaps especially import) liberalization as well as financial liberalization.

Common sense, as well as most economic theory, suggest that there should be plenty of profitable investment opportunities in the country's inefficient firms. New capital and management skills will increase profitability en masse, and democratization of weak, authoritarian institutions should give the country a second boost.

However, that will not happen. As capital flow into the country, through loans and FDI, the currency appreciates. And as the currency appreciates at the same time as tariffs are removed, domestic industry is hit by a double whammy, and lose competitiveness far faster than institutions can be reformed and investments can pay off.

Hence, the incentive for the country's citizens is not one investing in export or import-competing production, but one of investing in shielded production - services - that is not hit by the problem of currency appreciation and tariff slashing.

And, since no exports means there is no foreign currency to pay for imports... Wait! Since loans and FDI still flows into the country for a while after most investors stopped investing in production, there is a boom in import consumption financed with money borrowed abroad. And since the governing institutions are weak, it takes a considerable amount of time before foreign lenders turn on the red lights.

This in not far from what is happening in the US either - a consumption boom financed by borrowed money rather than exports - and I see it as very difficult how a country can embark on a situation of robust growth under such conditions.

Hence, the effect of import protection and financial liberalization should perhaps be evaluated together?

I have not read the linked papers yet, but my guess would be that capital inflows plays a role in how import tariffs (and export subsidies) perform in regard to growth.

Just a thought. Comments highly appreciated!

[fatbrick]

What about export restriction? If one country does not allow its companies to export, is that a trade barrier?

[assman]

Correlation does not equal causation

[camilo ferreira]

Some ideas:
1) It is possible that some kinds of protection are beneficial for a country meanwhile there are others that damage economic growth. There could be situations in wich the infant industry argumen works. But why to think that the state will be better to detect those sectors than a private firm? If to create an industry creates value why the private sector doesnt take the oportunity?
2)Trade protection effects over growth could be sthocastic with the actual knowledge we have, it could have a negative expected value, but even so it could be possible to find successful cases.
2)There could be a biased sample in the study you refer because those countries are mostly developed, so the ones that failed with their protection strategies are not taken in count.
3) It could be even possible that some easy to identify kind of trade protection are good for the economic growth of one country but even so it reduces the growth of world GDP. So free trade the best for the world as a whole.
If i made any mistake is because english is not my first language.

[Dave Hart]

Please let's not blame this all on the Chinese. A massive financial injection into a failing (auto) industry is a relatively straightforward method of protectionism. Dani is right, this is no time for beggar-thy-neighbour policies - particularly coming on the heels of last weekend's G20.

[wjd123]

"Mr Obama should use protectionism in a similarly forward-looking way. Industries that can be revived through re-tooling of its factories and re-training of its workers should be given protection, but only if they fulfill certain conditions regarding investment and training. Industries that have no future should be given strictly temporary protection to ease phasing-out through orderly liquidation and redundancy."--Ha-Joon Chang

Automakers do have a future, a future where their products fit into a greener world. The greener world that policy makers will help to make. Those who have the ear of policy makers should control the boards of our automakers.

With policy makers controlling the boards of American automakers taxpayer dollars would be a vote to where society wants to go. Where automakers want to go would be the past.

Social interest should be placed before automakers interests. That today's imperative.

Automakers' future is in building cars the world needs and not ones that lobbyist advocate for. To build these cars quickly, we will need a bailout; to see that they are used, we will need protectionism.

We'll have these cars all the faster if American automakers are directed toward a single minded policy. No more "it can't be done," or "we can't do it until our customers want it." It will get done and it will get done all the sooner because there will be fewer tensions between what ought to be done and what can be done.

Think of a global NASA program for green technology. This time not to go to the moon by the end of the decade but to save the earth from climate change.

The world needs a green car. Especially China and India as more people in those societies buy automobiles. The world will need ways punish them if they undo the gains from green technology.

We all need to clime abord the green technology bandwagon. We are in a race against time. And when the world economy gets going again a green car can no longer be an option but the only option.

One way to ensure that our efforts to make the world greener isn't offset by countries which see a competitive advantage in using cheaper oil to maintain a competitive advantage rather than using green technology is the judicious and ready imposition of tariffs.

In a world without tariffs a greener world can't work, particularly if green technology ends up decreasing the price of oil.

We could just let Congress mandate standards that fit with our social needs and give all automakers who want to sell into our market a time limit in which to meet those standards. That's the slow way, the way with the most resistance. And letting American automakers fail certainly wouldn't end up giving us the redundancy we would need for competition.

Rescue our American automakers, give them directors with a mandate and we will answer our social needs and the world's that much faster. That's worth a bailout

Driving the very people who have the expertize to keep our green project on tract out of business through the use of bankruptcy proceedings makes no sense if speed is needed for a greener world.

Even the economists' arguments that allowing the automakers to go into bankurptcy in the middle of a recession, in the middle of a credit crunch, would make things worse, while true, miss the point: this is an argument we don't have time for.

It's the old thinking of the 20th Century. It's the single mindedness of the NASA thinking of that century and not the orderly liquidation through bankruptcy thinking that we want to bring to the 21st Century. That's where our future is; that's where the worlds future is.

[satish]

us does not need protectionism . what we all need is free floating currencies.

[Anarcho]

"PLEASE, PLEASE, PLEASE don't ruin the half century of progress by promoting protectionism, or else dark day are indeed upon us."

In other words: PLEASE, PLEASE, PLEASE don't destroy the wonderful assumptions of neo-classical economics with such trivial things as empirical evidence or else dark days are indeed upon us!

If quoting Marx is not too out of place, tariffs are introduced with "the objective of manufacturing capitalists artificially" for the "system of protection was an artificial means of manufacturing manufacturers, or expropriating independent workers, of capitalising the national means of production and subsistence, and of forcibly cutting short the transition . . . to the modern mode of production." [Capital, vol. 1, p. 932 and pp. 921-2]

So protectionism in the 19th century can be seen as a part of primitive accumulation, as the state aiding the creation of capitalism. In the 20th century, when capitalism has been created, then its impact may be different.

Perhaps that could explain part of the differences?

[Steve]

I wonder if agriculture tariffs for a developing country would cause food shortage problems, especially in the 19th century. Most poor nations would want to make food purchases as cheap as possible, not only to enlarge the industrial sector but also to have enough grain to keep city-dwelling factory workers well fed.

[Richard H. Serlin]

"Industrial tariffs were positively correlated with growth."

Note they find correlation, not necessary causation. Given the other evidence, theoretical and empirical, I think industrial tariffs usually hurt growth more than they help it. We see the correlation probably because countries that impose industrial tariffs are often making earnest efforts to grow and are likely doing other things at the same time which promote growth like investment in infrastructure and education.