Strength of the Infant Industry Argument

In my last blog post I noted that infant industry argument is one of two serious academic attempts at discrediting international trade liberalization. I referred to “race of the bottom” as empirically unfounded as seen by referring to the literature and intuitively makes no sense. The infant industry argument though has some history of being valid policy focus advocated by Alexander Hamilton and John Stuart Mill, especially with the messy analysis of trade policies at the turn of the century. The free trading British Empire and protectionist United States complicate discussion with confusing anecdotal evidence. So, this means to get a good idea you have to look at current evidence and past evidence, specifically look for natural experiments.

To a large extent there is such a thing as an infant industry. Some level of soon to be realized comparative advantage. Take for instance differences in industrialization between European countries and North America between the 20th and 19th. Defining infant industries has always made me hesitant, and in past I considered the argument to a large degree unfounded mostly because I couldn’t picture how a planner could go around determining missed potential in industry. I think the best way to see infant industries besides the traditional definition (industries with costly learning curves compared to established foreign completion) is an industry requiring temporary protection to produce at competitive international prices until economics of scale set in. Further, in academic fields, previously (and in some cases today), using New Trade Theory arguments that increasing-scale-of-returns industries are the preferred industries to protect. Some have used this viewpoint and data from 1800s-1900s to explain two opposing arguments using natural experiments.

I am not an economist (won’t be for a couple of years) and at this point I haven’t take an international trade field elective yet, so this obviously means I am not up to date on New Trade Theory or New New Trade Theory. So, my views will/may change as I continue this blog. So, I guess take this as a disclaimer. My motivation for writing anything is that it is enjoyable. Let’s hope no planner, for the sake of society, reads this.

International trade isn’t as much of a monolith as I thought it was, I think Paul Krugman puts very nicely in his Nobel lecture. Probably as a detriment of macro 101, one assumes that international trade ends with Ricardo. The comparative advantage model explains a lot, specifically why different countries are good at specific things but New Trade Theory (NTT) fills underlying theoretical gaps and shows even in similar countries why different industry grow out. However, sometimes a few people reason from NTT that this means that any country should at any cost recreate Alexander Hamilton’s world and protect firms (preferably new firms) in industries of increasing returns. However, the insight of NTT is that increasing returns in countries of differing capacities or similar capacities is that both exporter and importer benefit. Protection reduces the win-win situation of free trade. There can be protection tools that can assist in the learning of firms in infant industries but as I mentioned in my last post some individuals hold onto infant industry argument as the end all be all of international economics. It is not.

A recent paper by Canadian economists on manufacturing between 1870-1913 uses learning-by-doing models to show that protection increases productivity, market size, output and price reductions. Specifically, ‘National Policy’ during that era, which was Canada’s mirror of the ‘American System’. Critics of national policy are common and certainly are recognized by Harris et al. However, it should be clear that even they noted in a footnote on the second page that tariff protection is not costless. Which again brings me back to insight of NTT. That is that trade liberalization can bring greater positive welfare effects than infant protection, a win-win situation. They do find that using Krueger and Tuncer’s (1982) focus on output of infant-industries, however applied to a large number of industries shows that in fact one can find evidence of infant-industries improving. With increasing returns, some Canadian industrial firms reduced prices. But remember tariffs have a cost. The alternative is cheaper goods from abroad at a lesser cost. And certainly the mixed evidence in other countries, like Turkey in Kreuger and Tuncer (1982), shows that this isn’t a hard and fast rule, one doesn’t have to always and everywhere protect new industries. Plus, Canada’s National Policy wasn’t a huge positive for the general growth of the country in that time period.

Another historical analysis during the turn of the century is by Douglas Irwin, unlike Harris et al (2015) opposes infant Industry argument which is actually quite difficult to do in an absolute sense empirically. Consider the ‘National Policy’ paper. A show in positive output growth that matches a learning-by-doing industrial organization model is a clear result, one has to show an empirical counterfactual to counter the case.

Screen Shot 2015-08-07 at 11.48.36 PM

R. Harris et al. / Explorations in Economic History 56 (2015) 20

Showing the counterfactual is far more difficult empirically. There is a puzzle that can leave one agnostic about infant industries as a whole. Consider that Irwin notes that “Learning by doing had only a modest impact on costs”. He modelled a counterfactual and simulated pricing alternatives. It was clear the tinplate industry didn’t have to create completely new practices. They adopted other practices and technologies, from the far more free trading UK. Irwin’s paper is fairly well cited and it’s findings make the case for protectionist USA as the best USA possible at that time questionable. Learning-by-doing is a large defence of infant-industry but learning isn’t always expensive just because an industry is new.

But what about areas with strong learning effects, larger than the tinplate industry. Keith Head (1994) discusses steel rail industry and the costs of the protection of the industry in the U.S at the turn of the century. Like Irwin, here is a relatively successful attempt at simulating the counterfactual. In fact, I would argue that the counterfactual of free trade is even important to the pro-infant industry argument even though it is generally lacking. Anyways, Head simulates a free trade policy and finds a net negative effect from the infant-industry protection. In this case the learning by doing effects are existent and yet the policy is no good. He mentions Baldwin’s 1969 paper on the Infant-industry tariff protection. It is very clear that tariffs are a very blunt tool. Baldwin suggests input subsidies as an alternative second best policies that aren’t net negatives on welfare of the public.

I think it is pretty clear the common historical argument that supports early tariffs in the US and Canada is weak, if not false. The counterfactual is rather more rosy about free trade. But tariffs are a blunt instrument, the effect of subsidies I would assume from reading Head and Irwin’s writing are not terribly negative but unnecessary, especially if the learning-by-doing means nothing to the models of NTT. In a blog post, Krugman noted that people in love with the tariff protections mention their arguments as capturing NTT very well. However, he notes that:

[…] that was not at all the point of New Trade Theory, which ended up suggesting that concentration of production due to increasing returns is generally beneficial to importers as well as exporters of increasing-returns goods, that it generally reinforce the case for open trade, rather than undermining it.

I don’t know what I would say considering export subsidies, usually subsidies are a counter argument against the knock down of the ‘American System’ and Canada’s ‘National Policy’. The growth of Asian countries in the last quarter century I believe are due to high savings rates and open markets. I think it is okay to consider in the case of South Korea that industry subsidies not necessarily protected industries that wouldn’t have existed without subsidization but possibly increased their production after the Korean war as part of its own development. Again, what I am trying to emphasize isn’t that there is no such thing as an infant industry, but rather that protecting infant industries through tariffs is a net negative and pointless. Open trade achieves greater gains. Subsidies may as well be industrial policy, with greater domestic effect developmentally than an effect on balance of trade. South Korea captured a level of comparative advantage when it comes to labor for a long time.

I think this also applies to modern developing countries. With increasing foreign direct investment, through free trade many developing countries have increased living standards and through that human capital. I wonder if free trade has reduced the need to protect infant-industries in any way in poor countries, because of the general sharing of technology and resources (specifically inputs).

On other protection methods such as quotas, I’ll refer to Marc Melitz’s paper that discusses a welfare positive model for infant-industry protection through quotas. Melitz is probably one of the most influential international economists out there. Relating to my discussion of subsidies he summarizes to an extent my views on the matter (views that aren’t necessarily my own but rather the consensus of international economists, I believe):

The economics literature has subsequently developed formal models with dynamic learning externalities demonstrating how protection can potentially raise welfare. This literature has also shown that the protection provided by production subsidies is preferable to that provided by tariffs or quotas, as the latter additionally distort consumption. Nevertheless, production subsidies may not be feasible due to government fiscal constraints and distortions associated with raising the needed revenue.

This leads into his research and modelling that concludes with the second best policy of quotas, a quota that depends on the learning curve. It is really interesting and I really don’t have much to say about it. Not that I am knowledgable enough to say much about the others but I found it rather important as I was attempting to gather a perspective on infant industry when no surveys were available online.

There is a lot to infant-industries and I found that it has a lot more to do with industrial organization than a way for developing countries to improve or proof of the incompetence of free trade. Tariffs as I noted are useless, at least according  evidence that considers both output statistics and market analysis (papers that considered welfare implications and costs of tariffs on top of learning benefits). Subsidies I am unsure of, though I conjecture that they don’t explain South Korea entirely (a possible post topic if I get time to do some research) or China’s industrial sectors. Quotas, like subsidies I am not sure of. Something I didn’t really research that I would assume many economists have considered is the market power considerations. I am interested in seeing what the effects of giving market power to infant industries under any artificial conditions are like. Now I only wish this was made easier through some published survey on like this one on infant-industry. Many discussions I found considered infant-industry in parts, not within some generalized trade policy, with a few (Irwin, Head, Baldwin) considering the policy as a whole.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s