CentrePiece Summer 2010

There has been a proliferation of regional agreements around the world since the early 1990s. In a survey of the latest theoretical and empirical research on regionalism, Caroline Freund and Emanuel Ornelas ask whether we should celebrate or be concerned about this trend. Regional trade agreements: blessing or burden?

n December 2009, a Ministerial lead to a protectionist backlash. Like meeting was concluded at the Regionalism has become riding a bike, the argument goes, if you World Trade Organisation (WTO). and will probably remain don’t move forward, you fall over. The headlines typically read the preferred form of Fears of a retreat to something like ‘Ministers reaffirm reciprocal liberalisation became particularly intense after the Ithe need and their desire to conclude the for most countries onset of the financial crisis – if the road Doha Round by the end of next year’. For gets bumpier, then you really have to many observers, this was a case of déjà and time and again, those expectations push forward to avoid falling. As it turns vu: since the current round of multilateral have been frustrated. out, there has been only a very limited trade negotiations began in 2001, there The main concern with the seemingly retreat towards protectionism, even have been numerous promises that it endless delay in achieving a new though the WTO bike has not been would be concluded in the ‘near term’ agreement is that such a failure could moving forward.

16 CentrePiece Summer 2010

Still, even if the lack of progress in In contrast, under an RTA, tariffs fall multilateral talks does not imply a retreat, on from the other members of many observers remain worried. After all, the agreement, but they need not change if there are gains to be shared with on imports from non-members. As a further liberalisation, we should make result, RTAs imply both trade liberalisation sure we don’t waste the opportunity. and trade discrimination. Whereas there is While this view is correct, it must be a near-consensus among economists that noted that lack of progress in the the former is desirable, that is not true for multilateral arena does not imply that the the latter. world is not becoming more integrated. In Trade liberalisation within a trading fact, governments across the globe are as bloc tends to be beneficial when it active as ever in negotiating new trade promotes a shift of resources from agreements. The difference is that the inefficient domestic suppliers to more agreements are not multilateral but efficient producers within the region. ‘regional’. Economists call this phenomenon In 2009, for example, 25 new regional ‘trade creation’. trade agreements (RTAs) were notified to Conversely, a trading bloc is likely to the WTO. The agreements include be harmful if it generates a shift of developed as well as developing resources from efficient external countries, and involve countries from producers to inefficient producers within most parts of the world. In fact, they are the region. This is a consequence of trade not always regional, increasingly discrimination, which economists call encompassing members from different To date, ‘’. continents. regionalism has In principle, either trade creation or The new agreements bring the total trade diversion can prevail within an RTA. number of RTAs in force to nearly 300. In been more a There are theoretical arguments that the same period, the WTO also received blessing than a support the primacy of each effect under notification of 14 other RTAs that are similar circumstances. Thus, in the currently being negotiated or about to burden for the end, which effect dominates is an enter into force. The WTO estimates that multilateral empirical matter. there are around 100 other agreements Unfortunately, estimating trade currently at this stage (WTO, 2009). trading system creation and trade diversion is no easy Should we celebrate the expansion of task. It requires knowledge of the RTAs, or should we be concerned with counterfactual: what would have this trend, which seems to be only happened to trade if there were no trade accelerating since it took off in the early agreement? As this is unknown, 1990s? In a survey of the theoretical and assumptions must be made. empirical literature on regionalism, we A variety of approaches have been conclude that although countries should employed. While results inevitably vary approach regionalism with care, to date depending on the methodology employed RTAs have been more of a blessing than a (as well as the time period, the trading burden for the multilateral trading system. bloc in question and the level of aggregation in the data), at least two Assessing the impact general messages arise from the large set of RTAs of studies investigating trade creation and To understand the impact of RTAs, it is trade diversion in RTAs around the world: important first to recognise that they are intrinsically different from both First, trade creation tends to be the multilateral liberalisation (under the aegis norm in RTAs – and trade diversion is the of the WTO) and unilateral liberalisation. exception. In both of those cases, countries lower Second, when trade diversion is trade barriers on all sources of imports of observed, its magnitudes are normally a good by exactly the same extent. relatively small.

17 CentrePiece Summer 2010

Trade creation versus benefit (and support the agreement) trade diversion because of the better access to foreign Why is there such a dominance of trade markets, whereas purely domestic firms creation over trade diversion? The answer suffer from the tougher competition from is in two parts. First, governments seem to the RTA partners. be choosing their RTA partners well. For But this also weakens the domestic example, variables that suggest greater firms’ stance on protection against non- gains from a bilateral RTA (such as members. The reason is that the free proximity between the members, a access to the domestic market enjoyed by similarity in their GDPs and a large the partners’ exporters under the RTA difference in their factor endowments) are lowers the market share of the domestic also sharp predictors of whether the two industry. As a result, the RTA makes any countries actually have a common RTA price increase generated by a higher (Baier and Bergstrand, 2004). less valuable for the domestic industry: Second, when countries form an RTA, now whenever the government attempts their governments not only lower tariffs to help domestic producers through higher vis-à-vis their RTA partners, as they are external tariffs, the partners’ producers supposed to do; they also bring down the absorb part of that surplus. trade barriers on imports from countries In other words, the RTA creates outside the bloc. This is not part of the ‘leakage’ in the trade policy redistributive agreement, so governments liberalise channel. External protection also becomes externally because they choose to do so – more costly, because of the trade diversion and without any type of reciprocity from Regional trade that accompanies the RTA. As a result, the favoured non-members of the bloc. agreements external tariffs tend to fall after the There is increasing evidence of formation of an RTA both because the external trade liberalisation following an generally seem economic marginal cost of external RTA, especially in developing countries. to amplify trade protection rises and because the political- (See, for example, Estevadeordal et al, marginal gain from external 2008, for an analysis of how Latin creation and protection falls. American countries changed their limit trade For developing countries, empirical tariffs on non-members after forming or research supports this rationale because expanding their trade ties during the diversion trade preferences tend to lead to lower 1990s.) The lower external tariffs external tariffs. Results for the United provide a double blessing: they imply that States and the European Union (EU), RTAs are responsible for more trade however, indicate that they are less likely liberalisation than they mandate to reduce external tariffs on goods where (amplifying trade creation) and for less preferences are offered (see Limao, 2006). trade discrimination than might be But since the tariffs of both the United expected (limiting trade diversion). States and the EU are very low to start At first, it may seem odd that with, and cannot be raised because of governments, pressured by special their WTO commitments, there is not interests as they are virtually everywhere, much room for change anyway. would voluntarily lower their external tariffs without any compensation Are RTAs welcome? from the favoured countries. But it does The benign view that trade creation make sense. dominates trade diversion does not imply Suppose that domestic special interest that RTAs are necessarily welcome. In fact, groups pressure the government and some commentators argue that the induce it to set relatively high tariffs, multilateral negotiations at the WTO are which allow the domestic industry to stuck because RTAs are spreading. Their maintain high prices and enjoy a large basic argument suggests that market share. If subsequently the country governments’ resources are scarce, so if enters into a RTA, -oriented firms officials are busy negotiating bilateral

18 CentrePiece Summer 2010

This article summarises ‘Regional Trade Agreements’ by Caroline Freund and Emanuel Ornelas, CEP Discussion Paper No. 961 (http://cep.lse.ac.uk/pubs/download/ dp0961.pdf) and forthcoming in the Annual Review of Economics 2: 139-67 agreements, they will be unable to focus relevant) from bad (empirically (September 2010). on more evolving multilateral negotiations. inconsequential) theories. Others put forward more elaborate Hence, to the extent that we can Caroline Freund is at the World Bank. arguments. For example, it is possible that measure, the increasing wave of Emanuel Ornelas is a reader in LSE’s RTAs may create rents to some groups. regionalism has been largely beneficial to management department and a research Once well entrenched, these groups may the world trading system. Most empirical associate in CEP’s globalisation programme. be able to block further liberalisation analyses indicate that trade creation, not initiatives that would destroy such rents. trade diversion, is the norm, both because On the other hand, there are governments choose well when forming With regionalism here to arguments indicating that the opposite RTAs and because they adjust other stay, economists should may be true. A simple one is that trade policies to moderate the distortions focus on ways of negotiating RTAs helps officials to develop from discrimination. integrating it more the expertise and the frameworks to Although it is possible that regionalism effectively with implement agreements, could endanger multilateralism, multilateralism and these could be useful at subsequent at the moment we just do not know. WTO negotiations. Moreover, RTAs also destroy rents in parts of

Further reading the economy. If the rent- Scott Baier and Jeffrey Bergstrand holders who lose with RTAs were the (2004) ‘Economic Determinants of Free ones slowing down multilateral talks, Trade Agreements’, Journal of then RTAs can actually provide a boost to 64: 29-63. such negotiations. These and related arguments Antoni Estevadeordal, Caroline Freund and constitute an intellectually engaging Since regionalism has become and will Emanuel Ornelas (2008) ‘Does Regionalism debate, but which of them dominate? probably remain the preferred form of Affect Trade Liberalization towards When faced with opposing theoretical reciprocal liberalisation for most countries, Non-members?’, Quarterly Journal of results, the solution is typically to scrutinise no matter what we economists say, Economics 123: 1531-75. the divergent predictions empirically. The henceforth we should focus on ways to problem here is that the nature of the integrate regionalism with multilateralism Nuno Limao (2006) ‘Preferential Trade question – whether regionalism helps or more effectively. Agreements as Stumbling Blocks for hinders multilateralism – does not lend Multilateral Trade Liberalization: Evidence itself easily to testing. for the US’, American Economic Review 96: Simply put, at any point in time we 896-914. observe a single realisation of WTO negotiations. Would they have been any WTO, World Trade Organisation (2009) faster, or easier, had there been fewer (or ‘Overview of Developments in the more) RTAs? This is a very difficult International Trading Environment,’ question. Consequently, to date empirical document WT/TPR/OV/12 scrutiny has not been able to help us (http://www.wto.org/english/news_e/ distinguish good (that is, empirically news09_e/wt_tpr_ov_12_a_e.doc).

19