NN IP: Free trade has benefits and costs

Even though the performance of the global economy remains pretty good, the downside risks have arguably increased somewhat over the past few weeks due to protectionist actions of President Trump. It is important to bear in mind that Trump and his supporters actually do have a point in the sense that it is not at all clear that the world should take further big steps in the direction of more free trade. However, the problem is that populists like Trump completely misdiagnose the root causes of the problem. What’s more, they propose a nuclear solution which will not solve the problem but rather impose substantial costs on the rest of the economy to provide a benefit for a few sectors. A move towards more free trade involves both benefits and costs. This suggests there is an “optimal point of free trade” which occurs where the marginal benefits are equal to the marginal costs. Public mainly emphasises the benefits which creates the false impression that more trade openness is always better. In general, the benefit of more trade openness is that it allows a country to increase the consumption it obtains out of a given set of inputs/production factors. In this sense, more trade openness has an effect which is similar to technological progress.  

One source of gains from trade comes from specialization according to comparative advantage, which roughly means doing what you do best in a relative sense. Even if a CEO is more efficient than his secretary in both managerial and supporting tasks in an absolute sense, it will still be better if he focuses entirely on managerial tasks and leaves the supporting tasks to his secretary. The ultimate reason for this is that the CEO is time-constrained. Another reason for trade gains arises when countries have different production factor endowments. Free trade then allows all countries involved to use the total amount of production factors available to them more efficiently. For instance, if there is no trade a country with a large labour to capital ratio will have to produce capital intensive goods itself while free trade allows the  country to import them at lower cost. A final free trade benefit resides in the presences of increasing returns to scale, e.g. due to the existence of a large lump sum start-up cost. The essence of the benefit then arises because free trade enlarges the market which will reduce the average cost of products. 

The costs of more trade openness stem from two sources. First of all, more free trade will change the income distribution within a country. For instance, unskilled labour in a capital rich country (the US) will lose out if that country starts to trade more with a country where unskilled labour is abundant (China). Economic theory recognizes this very well but implicitly assumes that the gains from trade (lower import prices and more profits for the capital intensive sector) are used to compensate the losers. In practice this not always happens to a sufficient extent. The second cost from free trade is generally not recognized in the economics textbooks and stems from the fact that formal institutions as well as informal customs and norms can differ considerably between countries. A move to more trade openness invariably implies that there has to be greater degree of international harmonization of the “rules of the game” embodied in these formal and informal institutions. This will reduce the welfare of domestic voters who are now faced with institutions that reflect their preferences more poorly. If driven too far, both these sources of costs could lead to voter revolt against free trade. 

The socially optimal degree of trade openness will thus generally not be the maximum degree of openness possible. In fact, voter protest in the US, the UK and parts of the Eurozone suggests that the world may already have moved beyond the optimal point of free trade. The solution proposed by the populists is then basically to turn back the clock and erect trade barriers. It is important to understand that this “solution” does NOT stem from a careful economic analysis which aims to raise national welfare. The drivers are usually a gain in real income for the small group in society that benefits from the redistribution of income on the back of protectionist barriers. Alternatively, such policies can be inspired by a desire to win more votes. This can be very effective, as for many voters protectionism seems like a logical solution to the problems of a skewed income distribution and/or voter dissatisfaction with the institutions required for free trade. Meanwhile, the protectionists never mention the cost of their proposed actions which amount to a widespread reduction in purchasing power for the entire population due to higher import prices. Protectionism is also a bit like trying to unscramble scrambled eggs: An increase in trade openness usually significantly alters the structure of the economy. Production processes in various sectors become much more integrated internationally, especially in the presence of a large decline in transportation and communication costs. As such, protectionism can be very disruptive. Hence, a good solution to the problem of “too much free trade” is never a simple and crude one, but rather a tailor-made and piecemeal one. It should entail mechanisms to compensate the losers of free trade as well as a trial-and-error process which allows trade institutions to evolve in such a way as to have a greater respect for different national preferences. The latter could well mean that trade deals will have to be renegotiated but not in the way Trump wants to do this. It should be done in a way that benefits all parties involved. 

Trump’s trade fallacies

The basic source of Trump’s protectionist agenda is that he views the trade balance as a profit and loss statement for the US economy. This mercantilist fallacy is quite common but economic theory tells a very different story: The sole benefit of exports is that they allow you to import the goods and services you want to consume but cannot produce yourself or only produce at a higher cost. This consumption can either take place in the present or in the future. In the latter case you acquire a foreign asset in exchange for the goods you export. Of course you want this exchange to take place at as favourable as possible terms. This means you want to buy the maximum amount of imports from a given amount of exports you give up. Another way of saying this is that, all else equal, you will be better off if the relative price of exports in terms of imports (the terms of trade) rises. One factor which will bring this about is actually a stronger currency. Also, if you acquire foreign assets you want the risk/return profile to be as favourable as possible. Conversely, if you are living above your means you want the yield you have to pay on your liabilities to be as low as possible.

If we translate this to the US situation we can draw the following conclusions. First of all, a dollar appreciation and a concomitant increase in the trade deficit need not to be bad at all for US welfare. After all, the stronger dollar implies a terms of trade improvement which means that the US will have to give up less exports to buy a given amount of imports. The rising trade deficit could still be detrimental to US welfare though if the costs of financing it become very high. We sometimes see this in EM space where large and persistent current account deficits can be a trigger for a sharp rise in domestic yields. However, as the US is the chief global issuer of safe assets, it has been able to persistently live beyond its means at a very low cost. An obvious example is the rise in the US trade deficit between 2003 and 2007. The dollar remained strong against Asian currencies while safe Treasury yields stayed at relatively low levels. The essential reason for this was that China wanted to prevent its currency from appreciating too much and the only way to do that was to essentially provide cheap funding for the US trade deficit. The fact that the US net foreign debt position as a % of GDP is below its cumulative current account deficit as a % of GDP suggests that the US has “paid” a negative yield on its net debt on average. Therefore one cannot argue that trade deficits have been bad for US welfare. Actually it is quite the opposite!

The second fallacy of Trump’s view on trade is that other countries are to blame for a rising US trade deficit. Actually, to a first approximation it is the US itself which is to blame. We live in an era of high capital mobility which, from a balance of payments point of view, means that the capital account “rules” the current account. In a sense, each country comes to the global capital market with some value of desired national net savings (savings minus investment). Of course these ex-ante desires will not be consistent with each other. If you add all national net savings up, the result must be zero. Hence, this set of desired net savings then alters financial market prices as well as national production to make them consistent ex-post. The current account balance is just another name for ex-post net national savings (a balance of payments balances if the trade balance and the capital account balance (including public capital flows) add up to zero). Because the US is a big country in both the real economy and the financial sphere, it usually gets close to what it wants in this game. In other words, if the US “wants” a bigger trade deficit because its desired national spending rises relative to its income, the rest of the world usually cannot or does not stop it from achieving this by triggering a sharp tightening of US financial conditions.

Assessing protectionist risks

With this in mind, Trump can be accused of pursuing an inconsistent policy mix. On the one hand, his administration has just announced a fiscal easing package which implies a substantial increase in the structural deficit. Provided this is not offset by an increase in the private sector’s desired savings balance (which is unlikely given solid confidence), this will automatically lead to an increase in the trade deficit as a matter of simple accounting. On the other hand, Trump continues to fret about the “unfair” trade deficit (which he helps to enlarge himself) and tends to blame other nations for the perceived problem. The rationale given for the steel and aluminium tariffs is that they are a matter of national security. One can understand why the US took this avenue because the WTO actually allows such action provided it can be justified. If other countries do bring the case to the WTO, chances are that it will rule this is not the case for the steel and aluminium industries. The majority of US steel imports come from allies such as the EU and Japan, while steel output has been roughly stable in level terms for the past 25 years, i.e. there is no obvious trigger for protecting the industry now.

The big question is what will happen going forward. It seems that the tariffs will be applied in a more tailor-made way. In theory this could be a good thing as it would allow Trump to claim the headlines for his voter base but at the same time reduce the retaliation backlash from other countries. Mexico and Canada have already been exempted as long as NAFTA renegotiations continue while Australia also seems to be off the hook. Meanwhile, it is uncertain to what extent the EU will get an exemption. If not, the EU is almost sure to retaliate in which case Trump has threatened tariffs on other EU products as well. Whether or not China will retaliate big time remains to be seen, as China only accounts for less than 1 % of US steel imports.

The next question is then to what extent the possible damage is likely to be contained. Various Republicans have made very critical comments about the measures but it is unlikely that Congress will be able to overturn them, as this would take a 2/3 majority in both Chambers. A section of the GOP adheres to the national security arguments while some Democrats also have protectionist leanings. The best one can hope for is that Congress will increase the degree to which the tariffs will be applied in a tailor-made way. Beyond all this there is a risk that Trump will get a taste for such measures especially if his approval ratings go up. In that case, industries such as vehicles, aircraft, shipbuilding and semiconductors could also be on the list for protectionist measures. If so, the risk of large scale retaliation will only increase. Whether or not Trump will move in that direction is unclear. In the past Trump has said he was positive about the WTO but that is no guarantee. A somewhat worrying development is the departure of Gary Cohn from the National Economic Council which, together with the demotion of Kushner, leaves Mnuchin as the only free trader against the protectionist wing in Trump’s team. If Cohn is replaced by a protectionist, this would clearly be a reason to become more worried. Perhaps the most worrying development for the long term is that the rules-based international trade order could be severely damaged because of all this. This order is the result of a decade-long process of negotiation. Even though there is certainly room for evolutionary improvement in this order, the maintenance of its essential rules is absolutely vital for a healthy global economy.