The value of the WTO: An economic viewpoint in six instalments Part I: Trade policy and international spillovers

Robert Staiger

What is the economic value of the multilateral trading system embodied in the General Agreement on Tariffs and Trade (GATT) and its successor, the World Trade Organization (WTO)? As the trading system faces growing pressure from global challenges, this question has become increasingly important.

As outlined in my introductory blog on this topic, this post presents one way of understanding the world trading system based on basic economic principles. It is the first in a series that explores how the core principles of the GATT/WTO create value for its members, and how changes in the global economy have created new challenges for the trading system over time.  

For clarity, the discussion focuses primarily on trade in goods, where the key mechanisms are easiest to see. However, the same logic applies more broadly to trade in services and other areas of trade policy, which will be discussed in a later post. The Agreement on Trade-related Aspects of Intellectual Property Rights (TRIPS) is set aside here because it requires a different economic interpretation.

This post develops five main ideas. First, unilateral trade policy can create cross-border effects on trading partners. Second, the GATT/WTO provides a forum where governments can negotiate and take these effects into account, to their mutual benefit. Third, market power plays an important role in shaping unilateral tariff choices. Fourth, WTO accession reduces the use of this market power regarding tariffs. Finally, the apparent "mercantilist" focus of negotiating market access for exporters has a coherent economic logic, and that logic is not built on the case for free trade.

The GATT/WTO and cross-border spillovers

A useful way to understand the GATT/WTO is to start from a simple observation: when governments set trade policies independently, they focus on domestic effects - such as impact on local firms, workers and consumers - but often overlook the costs imposed on trading partners.

These cross-border effects are sometimes called "spillovers" or, in economic terms, "externalities." When they are not taken into account, they can lead to inefficiently high levels of protection. 

From this perspective, the GATT/WTO provides a forum where governments negotiate with each other and take these cross-border effects into account.  By doing so, they can reach outcomes that are mutually beneficial.

This is not the only way to interpret the trading system. The "commitment approach," for example, emphasizes how international agreements can help governments establish the credibility of their policies and reassure investors or manage domestic political-economy pressures. The WTO's World Trade Report 2007 provides a comprehensive overview of this and other alternative frameworks.

However, the "international spillovers" perspective is particularly useful for understanding how the GATT/WTO's core principles work. 

The GATT/WTO as a bargaining institution

The idea that negotiations can address cross-border spillovers is closely linked to the work of British economist and Nobel Prize laureate Ronald Coase. His key insight (often referred to as the Coase Theorem) was simple: if parties can negotiate easily and the rules are clear, they can often reach mutually beneficial outcomes - even when their actions affect one another.

For this to work, two conditions are especially important: the rules about what is being bargained must be clear and reliable; and the process of negotiating must not be too costly or complicated (for example, not limited by lack of information or high transaction costs).

The GATT/WTO can be seen as an institutional attempt to create these conditions at the international level.

Some of its core features reflect this directly. For example, tariff bindings (legally agreed maximum duty rates) make commitments more secure, while transparency reduces uncertainty and facilitates negotiations. Other core principles of the GATT/WTO - such as non-discrimination embodied in most-favoured-nation and national treatment, reciprocity, disciplines on domestic subsidies - also help support a stable bargaining environment. These will be discussed in detail in the next posts.

Together, these features make it easier for governments to negotiate and sustain cooperative outcomes. 

Trade policy and price effects

One important type of cross-border effect arises through prices. [In economic terms, this is sometimes referred to as a "pecuniary externality."]

When a government raises a tariff, it can affect not only domestic prices and the quantity of imports but also the prices received by foreign exporters. Economists refer to this as a "terms-of-trade" effect.

The importance of this effect in driving the need for trade agreements was already clearly understood at the time the GATT was created. The British economist and Nobel laureate James Meade - often credited with writing the first draft of the GATT - noted that governments may hesitate to lower trade barriers unilaterally because doing so could worsen the prices at which they trade internationally (their terms of trade):

"One cannot hope to see the abandonment of protective devices except in the framework of all-round international agreement, since unilateral action is quite likely to cause the free-trade country to lose more from a deterioration in its terms of trade than it gains from the expansion of trade."
"The Theory of International Economic Policy, Volume Two: Trade and Welfare", by J. E. Meade. Oxford University Press, London: 1955, p 570

In modern terms, restricting market access affects foreign exporters in two ways: it reduces the quantity they can sell but also lowers the prices they receive. These are precisely the kinds of cross-border effects that the GATT/WTO is meant to address. 

Why negotiations focus on market access

Seen in this light, the apparently mercantilist logic of trade negotiations, where imports are regarded as concessions and exports as gains, becomes easier to understand.

Governments seek better access for their exporters not because imports are undesirable, but because exporters are often the domestic actors most directly harmed by foreign trade barriers. Exporters can therefore help identify the foreign measures that create the most important claims of cross-border spillovers. The WTO's role is to channel these claims into mutually beneficial bargains that liberalize market access, instead of leaving them to be addressed through unilateral retaliation.  This represents an application of Coase's theory in real life.

Notably absent from this role for the WTO is any reliance on the economist's case for free trade. The logic implies that governments would be expected to negotiate a "reduction in tariffs and other barriers to trade" as stated in the Preamble of the Marrakesh Agreement that established the WTO; but whether a government chooses to negotiate all the way to free trade depends on its individual circumstances and objectives.

Meade also anticipated challenges to these negotiations on market access that have only grown more pressing since his time. As he saw it, once you discipline trade barriers, you may need to grapple with the full range of domestic policies - regulations, standards, subsidies - that might be used to achieve similar protectionist effects.⁽¹⁾  This will be covered in the posts that follow.

Evidence

What evidence is there for or against this view of the trading system?  There are a number of ways that one might attempt to answer this question. 

One way is to ask who ultimately bears the cost of a tariff.  If each economy bears all the costs of each of its tariff choices, then no economy's unilateral tariff choices would impose any pecuniary externalities on any other economy, and there would be no reason for the GATT/WTO to exist according to this view of the trading system. There is a wide range of estimates in the literature, with new estimates coming out all the time. However, the majority of these estimates conclude that, while much of the cost of a tariff is borne by the tariff-imposing economy, somewhere between 5% and 40% and possibly as high as 60-70% of the cost of tariffs is borne by other economies, depending on the economy/product under consideration and the timeframe (short run vs long run) of the analysis. This range of estimates provides ample room for the possibility of substantial pecuniary externalities arising from unilateral tariff choices. Consistent with this observation, standard quantitative trade models used by economists to study the impacts of trade policy typically predict that at least some of the cost of a tariff will be borne by trading partners.      

Another way to assess the evidence for or against this view of the trading system is to ask how governments set tariffs in practice - both when they act unilaterally, and in negotiated tariff commitments agreed when they joined the GATT/WTO. Two key points emerge from this research. First, many economies - even relatively small ones - have some degree of market power (monopsony) in specific industries. In other words, they are large enough in certain markets to influence the prices paid to foreign suppliers. They also tend to use this market power when setting their tariffs outside trade agreements (i.e. unilaterally): higher market power is associated with higher unilateral tariffs. Second, those same economies tend to give up the use of their market power when they join the WTO.

A large empirical literature supports these key points. Two studies are highlighted here.

Broda, Limao and Weinstein (2008) provides evidence on the first point. The study examines the unilateral tariff choices of 15 economies of different sizes - ranging from Lebanon, Latvia and Lithuania to Russia and China - that were not WTO members at the time. Using detailed data, they estimate the degree of market power that each economy had over foreign suppliers in different products. In simple terms, they ask whether economies are large enough in particular markets to influence the prices paid to foreign exporters. Their findings show that even seemingly "small" economies have significant market power in at least some industries. They also show that economies indeed use their market power when setting tariffs outside a trade agreement.

This is illustrated in Figure 1. The horizontal axis measures market power (increasing moving from left to right), using a standard indicator based on how sensitive foreign export supply is to price changes. The vertical axis plots the median ad-valorem tariff chosen unilaterally by each economy. Each dot represents an economy.

Figure 1: Unilateral tariffs and market power

Note: Constructed from the data underlying Figure 3 of Broda, Limao and Weinstein (2008).

Figure 1 highlights two features of the data. First, economies differ widely in the market power they have over foreign suppliers, and this market power is often substantial. Second, there is a clear positive relationship between market power and tariffs: economies with greater market power tend to set higher unilateral tariffs. This relationship stands even after accounting for other factors.                        

The study by Bagwell and Staiger (2011) looks at what happens when economies join the WTO, and in doing so provides evidence on the second point raised above. The authors consider the tariff concessions agreed by 16 economies as part of their accession protocols when joining the Organization after its creation in 1995. The key question is whether these economies made greater tariff concessions on products where they had more market power - and therefore where they were imposing larger costs on their trading partners when setting tariffs unilaterally.

Figure 2, reproduced from Bagwell and Staiger (2011), shows that this is indeed the case. The horizontal axis measures market power (as the ratio of pre-negotiation import volume to the world price), split into ten deciles from low to high. The vertical axis shows how large tariff concessions are, relative to the mean.

Figure 2: Cooperative tariff cuts and market power

Note: Reproduced from Figure 1 of Bagwell and Staiger (2011).  Copyright American Economic Association; reproduced with permission of the American Economic Review.

The figure reveals a strong positive relationship between the product level market power that an economy possesses going into its WTO accession negotiations and the size of the product level tariff cuts that it agrees to in its WTO accession protocol. This suggests that the use of market power as a contributing factor in determining an economy's tariff levels is indeed removed or at least substantially diminished when it joins the GATT/WTO.

In isolation, giving up market power may be viewed as a cost to the economy, but the benefits are clear once it is remembered that similar behaviour is expected of all GATT/WTO members. When all members choose their tariffs in this way, the result is mutually beneficial. This is how Coase's solution to the externality problem is supposed to work. 

Main takeaway

Taken together, theory and evidence suggest that the value of the WTO lies not only in promoting trade expansion per se, but in providing an institutional solution to a specific and pervasive coordination problem in international economic policy. 

In the posts that follow, this perspective will be used to examine how the GATT/WTO core principles create value, and how evolving economic conditions are challenging their operation. 

References

Bagwell, Kyle, and Robert W. Staiger (2011). "What Do Trade Negotiators Negotiate About? Empirical Evidence from the World Trade Organization." American Economic Review 101(4): 1238-1273.

Broda, Christian, Nuno Limão, and David E. Weinstein (2008). "Optimal Tariffs and Market Power: The Evidence." American Economic Review 98(5): 2032-2065.

---

Footnotes

1. "... But as soon as any attempt is made to limit protective devices, a whole host of domestic economic arrangements must be brought under examination. Tariffs and quantitative import restrictions are not the only means for protecting domestic industries. Subsidies, domestic taxes, domestic price and quantity controls, nationalization schemes - all can be used for similar purposes. Yet all of these instruments may be perfectly legitimate instruments of policy for the attainment of certain other perfectly legitimate objectives. ... If a more liberal international economy is to be established by international agreement, one must search for a working compromise between the need effectively to curb protective devices and the need to give national governments freedom to adopt effective domestic economic policies for the attainment of legitimate domestic objectives. "

   "The Theory of International Economic Policy, Volume Two: Trade and Welfare", by J. E. Meade. Oxford University Press, London: 1955, p 570 Back to text