Multilateralising Regionalism: How to make regionalism less damaging to the multilateral trade system

Date December 15, 2006
Speaker Richard BALDWIN(Professor of International Economics, Graduate Institute of International Studies, Geneva)
Moderator KAWAMOTO Akira(Director of Research, RIETI)


This presentation will first look at the final steps to global free trade and how a move to free trade is going to look, which I am quite optimistic about while most in the world of trade are pessimistic. In particular I want to look at the political economy logic which will drive it forward. Secondly, I will talk about multilateralising regionalism; to look at how we can make the regionalism that we have less offensive to the multilateral system.

To start, I will give six stylized facts about global trade liberalization, and then come up with a political economy logic to help us understand why it happened.

The first is that the GATT process started when tariffs were very high worldwide. That was due to continuance from the 1930s. Secondly, the rich nations liberalized much more than the poor nations, both in the GATT process with bound rates, and also with regional trade agreements (RTAs). Thirdly, the liberalization focused on industrial goods in which two-way trade in similar goods was prominent. Fourthly, the process took 40 years, so we will need an explanation for gradualism. Fifthly, some sectors were excluded entirely and others experienced much less tariff cutting. The sixth; regional tariff cutting went hand-in-hand with multilateralism. In addition, unilateralism is now rampant.

With reference to the political economy: What are the determinants of the status quo tariffs? Please think about the tariff as a price, clearing the supply and the demand for protection. In microeconomics the supply curve is a marginal cost curve, so we should think of the supply curve as being the marginal welfare damage the tariff does to the national economy. The demand curve is the marginal utility curve, so we should think about the demand for protection as the marginal benefit to special interests in the country. As these two balance out, we will find a tariff which will clear the political economy market.

As we all know, tariffs are chosen politically, so we must ask: What has changed in the political environment to cause the removal of a tariff that it was previously optimal to apply? There are three mechanisms:

The first is for MTN (Multilateral Trade Negotiation) liberalization. I call this the "juggernaut effect." You start from a situation where the political equilibrium tariff balances supply and demand for protection. Now we announce a reciprocal trade talk - a GATT round - which realigns the political economy forces within each participating nation. In particular, this makes the nations' exporters into anti-protectionists. There is a change in the domestic political setting which makes it politically optimal to lower the tariff.

The second cycle of the juggernaut is - as you phase in the tariff cuts - the size of the import competing sectors get smaller, less politically powerful and make less profits, while the export firms get larger, win more profits and become more politically powerful. Anti-trade forces become less strong, pro-trade forces more strong, therefore at the next MTN it is politically optimal to cut the tariffs more. Once you start reciprocal trade talks, it changes things until all the tariffs are gone.

Next is the framework for RTAs. I call this the "domino effect." The first half-cycle starts with an idiosyncratic formation or deepening of a trade bloc that realigns the political economy forces inside non-member nations. FTAs cause trade diversions or trade discrimination for the excluded nations. This diversion becomes a new political-economic force in those nations, making it politically optimal to remove tariffs that were previously optimal to apply, in particular to join FTAs.

If the export sectors are larger than the import competing sectors, the countries will join FTAs. This domino effect has happened in South America (MERCOSUR), North America (NAFTA), Europe (EU), and from 2000 in East Asia too. East Asia did not have a bloc to join, which explains the explosion in bilateral agreements. As each new member joins, the "forces for inclusion" for non-members get stronger.

Next, what I call "race-to-the-bottom (RTB) unilateralism" in tariff cutting by East Asian nations in their race for FDI (foreign direct investment), primarily. This happened from 1985 when the Japanese economy started to hollow out. East Asian nations, one after another, agreed to have unilaterally low tariffs on particular parts and components. The unbundling of manufacturing processes has created forces different from those seen in the past. The fragmentation and vertical integration confuses the difference between imports and exports - most of what is imported is used in production and moved on - so the political consequences of keeping tariffs has changed.

There are three ancillary effects, mainly of academic interest.

First is intra-sectoral special interest politics. Melitz's "New New Trade Model" of two-way trade with heterogeneous firms shows big firms exporting and small firms not, resulting intra-sectoral politics; big firms want reciprocal liberalization, the small firms do not. The fewer large firms are easier to organize politically, so the liberalization gets done.

Secondly, the losers tend to lobby harder to regain their loss than they would lobby to make a profit, showing the importance of trade diversion.

Lastly is the home market magnification effect. Paul Krugman says that industries become "more footloose, not less, as trade barriers fall;" small differences in market access mean more when all the tariffs are low.

Next is the historical narrative. The 1934 Trade Act allowed the U.S. government to reciprocally negotiate down tariffs, but only product by product and partner by partner, meaning that reciprocity existed but it was not very systemic. The EEC could not negotiate that way, so President Kennedy got Congress to grant the authority to negotiate across the board, leading to further tariff drops.

The EC formation happened at the same time as the Kennedy Round. The U.S. was also negotiating an Auto Pact, and both the U.S. and the EC were implementing unilateral preferences for developing countries. So, during this time, multilateral, regional and unilateral liberalization all went hand-in-hand. Hence, liberalization begets liberalization.

In Europe, when the USSR dissolved, a "spaghetti bowl" of FTAs occurred; in the mid-1990s there were around 80 FTAs in Europe. The U.S.-Mexico FTA also triggered a domino effect resulting in a spaghetti bowl.

Considering Asia, RTB unilateralism in East Asia started well before the Asian crisis. China's approach to ASEAN has triggered a domino effect, leading to the "noodle bowl" of FTAs. Also, ASEAN is not a single agreement, but a series of separate bilaterals, unlike the EU.

Projecting forward: East Asia will undergo big changes, as 2010 is the deadline for most of the agreements. Now there is little trade preference discrimination as it has been driven by unilateralism, but if the agreements which have been signed get implemented there will be a lot of discrimination appearing. This will lead to a world with three "fuzzy," "leaky" trade blocs: North America, Europe, and East Asia.

Fuzzy because members of the blocs have FTAs with many other countries, and leaky because the trade blocs do not exist in isolation; there are channels of free trade linking all the blocs. Moreover, the blocs have fuzzy, leaky sub-trade blocs like fractals, so I predict that applied tariffs will be near zero for all the world's major trade flows by 2010. The point of all of this is that if you can find the solution to tame one of these blocs or sub blocs, the same solution would apply to the world.

I will give two examples of how regionalism has been multilateralised. First is the Pan-European Cumulation System (PECS) in 1997. Spaghetti bowls have two problems: different Rules of Origin (ROOs), and bilateral cumulation. PECS imposed a common set of ROOs on the EU, the European Free Trade Association (EFTA), and the Central and East European countries (CEECs), and it imposed diagonal cumulation. So, if a product entered into PECS it could not lose origin status, making it easy to do outsourcing. PECS is essentially a customs union, but of ROOs, not tariffs.

Why did this get done? EU companies that had originally asked for and benefited from the complex ROOs - as China emerged and companies were offshored to Eastern Europe - now found themselves in the spokes and being harmed. They asked for this to be fixed. I call this "spaghetti bowls as building blocks." The complexity, together with fragmentation and offshoring leads to a political economy force from companies for governments to tame the tangle.

As global unbundling continues, the "spaghetti bowls as building blocks" pressures will mount. It will be impractical to do a PECS worldwide because the three major ROOs - PECS, NAFTA, and ASEAN - are not compatible and never will be. An alternative is the 1996 Information Technology Agreement (ITA), which is my second example of multilateralising regionalism, this time on a sectoral basis but globally; rather than all the sectors but regionally.

By the mid-1990s, tariffs on IT products worldwide were approximately zero, based on an assortment of unilateral, regional, and multilateral commitments, and there were ROOs. Also, this industry was particularly highly fragmented, so the industry firms asked the governments to tame the tangle. Since it was impossible to negotiate ROOs between Asia, the U.S., and Europe, the only solution was to bind the MFN tariffs to zero, thereby making the noodle bowl problem go away. Again, the companies that had originally asked for ROOs asked for it to be fixed. It started with 80% of world production and is expanding.

Given the current situation, what should the WTO do? Here is an optimistic view of how we are going to worldwide free trade through a combination of regionalism and pressure to tame the tangle. The following reasoning is based on two assertions: regionalism is here to stay, and regionalism has happened with the WTO as a bystander.

The WTO has weak rules and has not got into the business of disciplining, and all this business of noodle bowls has happened with the WTO on the sidelines. It risks an erosion of its relevance if it does nothing. Barfield says that the only things that are working are the WTO's dispute settlement panels, but they should not be used too much because if rulings continue to go against major players such as the U.S. and the EU, they may be tempted to ignore or try to get around the WTO in future. Also, bilateralism is exploding. The countries that are coming into the trading system for the first time (China in particular, but also India and ASEAN) are learning that trade liberalization means bilateralism. To become more relevant, the WTO has to get back into the regionalism game.

I am not sure what the WTO should do, but it must do something. For example, study the spillover of things like PECS more closely. There is evidence that shows the formation of PECS harmed the exports of third countries; their exports went down. In principle the WTO was set up to discipline these kinds of spillovers, so harmonization of ROOs at a regional level might be a subject for WTO discussion. Also, there is "standing" in FTAs. There is a presumption now that the only people who should be at the table are those involved, but we know there are spillovers. Maybe there could be a transparency mechanism where intermediate studies are shared through the WTO with affected parties. Another could be negotiating international standardization of ROOs and Rules of Cumulation (ROCs). One way forward might be to have limited mutual recognition of origin certificates, instead of deciding common ROOs. I also suggest that if industry was left alone to talk about ROOs and they were voluntary, there would be a pull of governments into those ROOs. Another idea is a sectoral initiative, like the ITA.

Lastly, all the things I have talked about are good for rich northern exporting nations, but for the poorer southern countries perhaps the WTO should set up something like the Legal Advisory Centre (which was set up to help poorer countries in the dispute settlement mechanism), but for negotiating FTAs. The idea would be to get them to make the FTAs less offensive to the multilateral system.

Questions and Answers

Q: What are the implications of what is happening in East Asia now?

A: I wrote a paper on this subject titled "Managing the Noodle Bowl: The Fragility of East Asian Regionalism." The proliferation of FTAs caused a domino effect, and China's approach to ASEAN caused many reactions. The trouble with East Asia now is that it has a "vision deadlock," which means that it will be stuck with a noodle bowl of FTAs for a long time. Instead of being concerned with the "vision for East Asia" it should be concerned with management.

I make the analogy with the Asian monetary world, which has a management committee to deal with minor problems now without losing the vision of a possible monetary union in the long, long run. That does not exist on the trade side. There is a pressing need for a management committee, and in the paper I have suggested that that should be ASEAN Plus Three, because that is up and running. I am not proposing an FTA with ASEAN Plus Three, I am proposing a management committee for when trade disputes get out of hand.

I do not think that East Asian regionalism can be multilateralised very quickly, but with this idea of "spaghetti bowls as building blocks," hopefully the multinational corporations will soon put the tariffs to zero or harmonize the ROOs. That pressure will eventually appear, as it already has done in electrical machinery and general machinery where the tariffs are almost zero, so the ROOs do not matter. But, after the agreements get done there will be a lot of discrimination so they will either have to harmonize the rules of origin and allow diagonal cumulation or they have to put all the MFN tariffs to zero.

As for the WTO; we need an extension of fast-track to do the Doha Round. The U.S. Congress will have to pass at least a temporary extension. We will find out in January 2007 what their legislative initiative is. If it is there, there is one more chance; if not, it is over. If it does not get done, I think 2008/2009 is a chance to come back, but there is a real problem of what the WTO is going to do in the meantime.

Q: Is it correct to say that lobbying by the export industry is a key factor in your domino theory?

A: Absolutely. Some people call it defensive regionalism; at least the first phase of the domino is defensive. When two countries have preferences, it harms third country exporters, even if they do not raise the tariffs. What matters is the relative price. It is the pressure of the exporters to redress the discrimination which leads those countries to join the first time. The first turn of the crank on the domino theory is that of exporters in excluded nations pushing their governments to redress the discrimination by signing another FTA. The second part is that if they actually do sign a FTA, the problem is even bigger for those who are still outside. The pressure mounts on outsiders to join.

Q: How much does regionalism actually matter? Are the current forums for regionalism in East Asia (APEC, East Asia Summit, ASEAN) viable options for future liberalization in this area?

A: It is important to distinguish between what has happened up to now and what is promised in the future. Many people look at what has happened in East Asia up to now and say that regionalism is not a problem, but that has not involved discriminatory, preferential tariffs; it has involved unilateral tariff cutting. It is the discriminatory aspect of FTAs which creates the conflicts. Up to now AFTA and the others have not included preferences, but if they do what they say they are going to do between now and 2010, there will be a serious amount of discrimination that appears. The very favorable experience up to now is not a predictor of what will happen if people do what they have promised to do in the next five years.

Secondly, APEC is useless for negotiating FTAs. APEC has too many members with too many diverse interests. The East Asian Summit; that is basically ASEAN Plus Three Plus Three. I prefer ASEAN Plus Three because it has a history, it has a track record of some success and it has political legitimacy. Above all, you would not have to decide who was a member; that has already been historically decided. That is why I think ASEAN Plus Three is a good vehicle for it.

Q: How do you deal with the difference between developing countries and developed countries, especially in your domino theory?

A: I have never made a big difference between the two. I have felt that it has applied relatively equally, for instance the domino effect going on around MERCOSUR; those are all southern countries. One thing that is quite different about developing countries is that their FTAs are "enabling clause," not Article 24. Enabling clause disables all the GATT disciplines, in particular it disables Article 24 which says that a FTA has to be substantially all trade, has to go to zero in a reasonable period of time and you cannot raise your barriers against anybody else. So, I think the domino effect works even faster with developing countries because they can write FTAs that are not substantial, which just redress the discrimination in the industries their exporters are most interested in.

Q: How does your spaghetti bowl theory apply to new areas like services, investment and intellectual property rights, with the proliferation of FTAs?

A: I explicitly left out those issues because I do not really understand them. In services, we can say that GATT has been a disappointment; it has not led to much liberalization. The nature of barriers in services is fundamentally different and not affected by reciprocity; for example, banking. Often, the banking protection protects a given set of firms in the country both from foreign and domestic potential entrants. In that kind of world, multilateral liberalization gives up the advantage against both foreign and domestic firms, only giving you better access to the foreign market in exchange, which is also diluting the value of the access because they are liberalizing. So, the liberalization begets liberalization logic does not work in GATTs, because it is protecting against domestic competition too, so the exchange of market access works in a different way.

If that is true, GATT-Plus, which leads to a partial liberalization, tends to reduce the resistance to multilateral liberalization. So, if the monopoly position of the handful of banks has been loosened against domestic or regional partners, they have less to lose then by going further. I am hopeful that GATT-Plus will make it easier to do GATTs in the MTNs, which will just be multilateralising the trade agreements that have been done regionally.

Very often the protection of intellectual property rights is protecting one firm against domestic competitors as well as foreigners, so the exchange of market access does not work. I, like Mandelson, believe that deep agreements on these new areas are good for the world trading system.

Investments, I have no idea about. I do not understand why countries are so interested in bilateral investment treaties but are not willing to multilateralise it. But on GATTs, on services and intellectual property rights, what is good in regionalism will eventually be good for the WTO.

Q: What is the overall effect of small-volume independent elements?

A: When I am talking about favorable development and moving toward free trade I am leaving off the small volume, high protection elements. For example, agriculture in Japan; it is a relatively small amount of trade if you looked at it as a percentage of world trade. It is the same with sugar in the U.S. or dairy in Europe. So, in this vision going forward that there will be almost duty-free trade by 2010, I have left off these "hard nuts" of protection. I do not have any good solution for them, but I would assert that if we are worried about the global trading system, they are not that important. Whether they stay protected or not is not going to affect the WTO, so I have left those aside in my discussion. In Ireland for example, the farmers have been bought out and that land is now liberalized, so there could be a creative solution like that. When talking about the global trading system, these are very small amounts of trade. Cars, chemicals, steel, textiles, apparel, electronic machinery; those big trade volumes are what I was talking about.

Q: What is the role of agriculture?

A: When the GATT was set up it was a small group of countries. A number of times there was a decision on the agenda to include agriculture, but when the talks got a little difficult, the "quad" (the U.S., Canada, Japan, and the EEC) decided to leave agriculture off, so it was left off the majority of the trade talks up until the Uruguay Round, so the Juggernaut did not work. There are no Japanese exporters lobbying the government to liberalize agriculture because farmers in the U.S., Japan, and the EU together said they wanted it left off the table. The dynamo of liberalization did not work in agriculture. Finally, in the Uruguay Round there was a group of countries that said they would not sign unless agriculture gets included. Then we saw, for example in Europe, that governments were pressed by their exporters to eventually liberalize agriculture. I think that will go forward as the gradual liberalization in the MFN of agriculture reduces resistance, and agriculture will get liberalized. It took 40 years in industrial products; maybe it will take another 40 years in agriculture, but it only started in 1994.

Q: How do evaluate the moves by the European Commission to sign FTAs with Asian countries? What will happen in the near future?

A: I think the EU will sign FTAs, and they will look a lot like the EFTA-Korea FTA; they will exclude agriculture entirely and have a lot of exceptions, but still be Article 24. The EU will rapidly be able to sign FTAs with the major ASEAN countries, not with China but certainly with Korea. That will cause a big problem both for the U.S. and for Japan. That is what starts the domino effect. Then, the U.S. and Japan will have to come up with a way to redress that discrimination if it happens; the classic way would be to join.

But, clearly, the political climate now in Japan is not the one that would exist if Europe signs FTAs with those countries. I believe it could lead to Japan react and start signing FTAs with the EU, U.S., and Korea, and therefore spark the completion of worldwide FTAs; but I am an optimist. It could of course instead lead to trade warfare, but I do not believe that.

*This summary was compiled by RIETI Editorial staff.