Share this:

Members have the opportunity to set their tariffs in a mutually beneficial manner, but little cooperation in tariff setting is actually observed, write Alessandro Nicita, Marcelo Olarreaga, and Peri Silva.

The ongoing international trade tensions between the US and a host of countries serves as a reminder of the importance of strong international institutions in helping avoid the escalation of trade disagreements into trade wars. The WTO and its predecessor the GATT are often seen as not only a strong force behind multilateral trade liberalisation, but also a place where governments can cooperate in order to set tariffs and resolve disputes in a mutually beneficial manner. But exactly how much do countries cooperate within the WTO framework?

We address this question in a new study where we examine the extent to which tariffs set by WTO members (before the recent trade frictions) reflect cooperative behaviour. Tariff negotiations are considered the bread and butter of the WTO and, therefore, if there is one area in which cooperation should be observed, it is definitely in the area of tariffs. Surprisingly, we found that more than three-quarters of tariffs applied by WTO members do not reflect any type of cooperation. This matters because we estimate that a move from non-cooperation to cooperation leads on average to a 35 percentage points reduction in tariffs for products in which importers have significant market power in world markets. Note that 35 percentage points is roughly the average reduction in tariffs observed among high-income countries since the creation of the GATT. This is consistent with our finding that most high-income countries tend to set their tariffs cooperatively within the WTO system.

Much less cooperation is observed among middle- and low-income countries. Most of their tariffs are set non-cooperatively. One may think that this is a minor problem. Indeed, most middle- and low-income countries tend to be small, and therefore have little market power in world markets. There is no need to negotiate with these countries because they have little impact on world markets. This reasoning is incorrect for at least two reasons. First, if individual countries have little market power, collectively they obviously matter and can significantly affect world markets. Second, our estimates suggest that many middle- and low-income countries have strong market power in world markets by themselves. Brazil and China are obvious examples. But even countries such as Uruguay or Burundi can have strong market power in a few items. Among the 21,000 tariff lines in the top 10 per cent of the distribution of market power among WTO members, Uruguay appears 100 times, and Burundi 7 times. And this matters because according to our estimates, without cooperation in tariff setting, importers could be setting tariffs at levels higher than 125 per cent if they are at the top 10 per cent of the market power distribution.

Thus, despite the success that is often attributed to the GATT and WTO negotiations in terms of tariff reductions, it seems that there is room for further cooperation in tariff setting among WTO members, particularly in the case of middle- and low-income countries. There are some significant gains from tariff cooperation that are currently being left on the negotiating table.

Does this limited degree of cooperation mean that the WTO and GATT have not been very useful? Does it imply that the collapse of the WTO would have little effect on international trade? The answer to these two questions is a clear NO. Even if less than a quarter of WTO member tariff lines are set cooperatively, these are tariff lines imposed by relatively large countries covering a large share of world trade. Our estimates suggests that if all tariffs lines that are currently set cooperatively were to be set non-cooperatively, as might be observed during a full-blown trade war, then the simple average tariff in the United States would increase by 39 percentage points, and in the EU by 51 percentage points.

Depending on the export bundle of each country, as well as its trading partners, such a move towards a full-blown trade war can lead to very large increases in the tariffs exporters face in world markets. We estimate an increase in tariffs faced by the average world exporters of 32 percentage points. Today the average tariff in the world is 5 per cent. Getting rid of the WTO would therefore imply a six-fold increase in the average tariff faced by exporters. And some countries could face much larger increases. Haiti exporters, for example, would experience an increase in the average tariff they face in world markets of around 85 percentage points. Honduras and Mexico exporters would experience average tariff increases abroad in the order of 60 percentage points. The economic and social consequences of such increases in tariffs faced abroad can be very important.

A well-functioning WTO helps prevent such scenarios. There is clearly comfort in what has been achieved so far, but the WTO also needs to be looking into what remains to be done. To move forward there is no need to move into nontraditional WTO areas such as labour, environment, investment or competition. There is still a lot to be done in terms of simple tariff cooperation in the case of low- and middle-income countries. Reducing negotiating costs faced by those WTO members at the multilateral level would definitely help to bring them to the negotiating table. Less discretion in negotiation and more negotiating rules, even if sophisticated, can be one avenue to explore.

Please read our comments policy before commenting 

Note:  This article gives the views of the author, and not the position of USAPP– American Politics and Policy, nor of the London School of Economics.  

Shortened URL for this post:

About the authors

Alessandro Nicita – United Nations Conference for Trade and Development
Alessandro Nicita is an economist at the United Nations Conference for Trade and Development.



Marcelo Olarreaga – CEPR
Marcelo Olarreaga is professor of economics at the University of Geneva and research fellow at the CEPR.



Peri Silva – Ld’A
Peri Silva is professor of economics at Kansas State University and research fellow at the Ld’A in Milan and at the GEP in Nottingham.



Print Friendly