Alternative Growth Strategies in Asia: Liberalization, Deregulation, Structural Reform*

Transcription

1 Alternative Growth Strategies in Asia: Liberalization, Deregulation, Structural Reform* Dr Philippa Dee Crawford School of Economics and Government Crawford Building (13) Australian National University Canberra ACT 0200 Australia ph ; fax December 2008 (revised) Abstract : This paper evaluates some of the possible paths to Asian economic integration. It defines three key integration strategies preferential trade agreements, multilateral action through the WTO, and comprehensive domestic regulatory reform. It characterizes the types of policy measures that could be expected to be achieved under each strategy, and compares the economic effects of each set of policy measures. The paper finds that for most of the Asian region, comprehensive domestic regulatory reform dominates as an integration strategy, while PTAs deliver relatively trivial gains. Multilateral action is an intermediate strategy, and is important for delivering trade liberalization in sensitive sectors or in highly protected economies. The results are shown to be relatively robust to the size of the Asian PTA grouping. The paper then considers what forms of regional cooperation could further the domestic regulatory reform agenda in each country. It concludes that the diversity and voluntarist principles of APEC provide a natural home for regional support of domestic regulatory reform, in the same way that the WTO provides a natural home for trade liberalization. JEL Classification: F13, F15 Keywords : economic integration, preferential trade agreements, regulatory barriers to competition, domestic regulatory reform, Asia, institutions, regional cooperation * Background paper prepared for a study by the Asian Development Bank on Emerging Asian Regionalism: Ten Years After the Crisis.

2 IV. Alternative Growth Strategies in Asia: Liberalization, Deregulation, Structural Reform The growth and dynamism of the Asian region was sustained in the twentieth century by a first generation of reforms that gradually opened up the Asian economies through export promotion and trade liberalization. The process was assisted by a first generation of regional cooperation that focused primarily on facilitating and removing barriers to transactions at the border (Soesastro, forthcoming). The result was not only phenomenal growth, but also growing regional integration. Asia s future growth and dynamism will require a second generation of reforms that tackle the behind-the-border barriers to economic performance. But recent initiatives to promote Asian integration reveal deep uncertainties on several fronts. One is about the priority to be given to different types of behind-the-border barriers. Another is about the appropriate nature of the second generation regional cooperation that will best promote a second generation of economic reforms. This chapter deals with these issues. It first clarifies the nature and objectives of economic integration, and then gives a brief history of Asian integration thus far. It then gives a quantitative assessment of three key integration strategies preferential trade agreements, multilateral action through the WTO, and comprehensive domestic regulatory reform by evaluating their impact on economic efficiency and productivity. Finally, it outlines a vision for the type of regional cooperation that will best support a second generation of economic reforms. A. Nature and objectives of economic integration Deep economic integration goes beyond simple trade liberalization. The aim is to reduce the market segmenting effects of behind-the-border, domestic regulatory policies through coordination and cooperation (e.g. Lawrence 1996). This will ensure that domestic prices are no higher than they need to be, relative to foreign prices, thus promoting economic efficiency and maximizing income levels. Some writers (e.g. Cooper 1976, Lloyd and Smith 2004) define deep integration in terms of the law of one price holding in all markets. If all goods and services are relatively homogeneous, and if all are potentially tradable, then an absence of market segmenting barriers will ensure that a single good or service is priced the same in all markets. But this definition of deep integration may be too strong. If goods and services are not the same in all markets, then market segmenting barriers may mean that the prices in one market are higher than they need to be, relative to those elsewhere, but that removal of those market-segmenting barriers may not bring prices fully into equality. It is often assumed (e.g. Commission of the European Communities 1985) that at the core of deep integration is the removal of all border and behind-the-border discrimination against foreign suppliers of goods and services (and sometimes labour and capital). If there are no significant non-discriminatory barriers affecting both domestic and foreign suppliers, then removing all discrimination against foreign suppliers will mean that domestic prices will be no higher than they need to be, relative to foreign prices. But this definition of deep integration may be too weak. If there are significant non-discriminatory 1

3 regulatory barriers affecting both domestic and foreign suppliers, then domestic prices may still be higher than they need to be, relative to foreign prices, even if all discrimination against foreign suppliers is removed. To restate, the aim of deep integration is to remove the market segmenting effects of behind-the-border barriers so that domestic prices are no higher than they need to be, relative to foreign prices. The relevant barriers need not be explicitly discriminatory against foreign suppliers, but could affect domestic and foreign suppliers equally. Removing these market segmenting barriers should bring domestic and foreign prices closer together, although it may not bring them into full equality. This definition raises broad questions about the appropriate policies to achieve economic integration. What relative priorities should be given to removing explicit discrimination against foreign suppliers, or removing the non-discriminatory regulatory restrictions that affect domestic and foreign suppliers equally? It also raises questions about the appropriate policy forums in which economic integration should be pursued. Can the important policies to achieve economic integration be pursued within the confines of a formal free trade agreement? What is the relative importance to be placed on multilateral trade forums, or other, more voluntarist vehicles for regional cooperation such as APEC, in achieving the desired goals of deep economic integration? These are the questions to be addressed in this paper. B. Process of Asian integration thus far From the 1980s, Asia developed high levels of intra-regional trade, but this was largely in response to market forces. The reasons included rising incomes, falling information and transport costs, and unilateral tariff cuts (either generally or through duty drawbacks and duty-free export processing zones). These forces allowed the fragmentation of production chains and encouraged the relocation of the labour-intensive stages of production to lower-wage countries in East Asia. Over 50 per cent of East Asia s trade is now intra-regional trade. 1 Japan and the newly industrialized economies (NIEs) also account for a rising proportion of foreign direct investment in ASEAN and China. During the 1990s this integration occurred largely without preferential trade agreements (PTAs). Over that period the ASEAN Free Trade Agreement (AFTA) and the SAARC Preferential Trading Agreement among South Asian economies were signed, but both were limited in coverage. AFTA did not create strong preferences, with less than 5 per cent of trade within the region taking place at the preferential rate (WTO, 2003), and neither agreement has significant commitments on services or investment. The Asia- Pacific Economic Cooperation (APEC) agreement, also signed in the 1990s, was designed to be non-preferential and fully consistent with the most favoured nation principle. However, the recent spate of bilateral agreements are establishing the legal basis for significant preferences within the region. Baldwin (2006) has argued that the 2003 China-ASEAN agreement in particular appears to have triggered a domino effect within the region, encouraging Japan and Korea to consider agreements with each other and 1 Calculated for Japan, the Asian NIEs (Hong Kong, Korea, Singapore and Taiwan), the remaining ASEAN economies (Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, Philippines, Thailand, Vietnam) and China. 2

4 with the ASEANs. Many East Asian economies have been also signing bilateral agreements with partners outside the region. In part, this has been a defensive response to the proliferation of preferential trade agreements elsewhere. But it is also rooted in a desire to boost productivity and international competitiveness, as well as to promote deeper integration and institution building at the regional level (Kawai, 2005). A further move to promote regional integration was the commitment in 2004 to establish an East Asian Community among the ASEAN+3 (China, Japan and Korea). This aims to promote economic and financial cooperation, political and security cooperation, environmental cooperation, social and cultural cooperation, and institutional cooperation. Economic cooperation is to include the establishment of an East Asian Free Trade Area and the expansion of the Framework Agreement on an ASEAN Investment Area to all of East Asia. By extrapolation from the AFTA and China-ASEAN agreements, the new initiative will include ongoing negotiations on services and investment, as well as commitments on goods trade. Yet another move has been the establishment of an East Asian Summit, with the first meeting held in Kuala Lumpur in December Membership extends beyond ASEAN+3 to include India, Australia and New Zealand. The extended membership was seen as ensuring that ASEAN would remain at the geographic centre of the emerging East Asian Community, thus diluting China s influence, and annual meetings occur after the annual ASEAN leaders meetings. The members agreed to study a proposal for a Comprehensive Economic Partnership in East Asia, but there are different views among the membership about whether ASEAN+3 or ASEAN+6 will be the real focal point for regional community building in East Asia. These recent attempts to promote Asian economic integration reveal deep uncertainties on several fronts. The first element of uncertainty is whether Asian integration needs to encompass a formal plurilateral free trade agreement. Some see the East Asian Summit as inheriting this element from the ASEAN+3 proposal; others see it only as a high-level forum for discussion on strategic issues. Some see a natural path for APEC to progress to a Free Trade Agreement of the Asia Pacific (FTAAP); others do not see APEC as having a role in trade negotiations. A second element of uncertainty is about the nature of the behind-the-border economic cooperation required to progress Asian integration. When a specific crisis has arisen (e.g. the Asian financial crisis, SARS), specific forms of Asian regional cooperation have been developed to cope. More general proposals for economic cooperation remain vague, and when Asian PTAs have encompassed behind-the-border regulatory issues in chapters on services or investment, the results have tended to be talk rather than action. These two issues interact with each other. So the current uncertainties include the following. Will an Asian integration strategy include a conventional PTA that liberalizes goods trade? Will it also include behind-the-border economic cooperation? If so, will this cooperation be of the type that would normally take place in a PTA? Or will it be of another sort, perhaps requiring another forum (e.g. WTO, APEC or unilaterally). C. Recent assessments of Asian integration strategies 3

5 When there is such uncertainty about the content of the strategies to be assessed, it is little wonder that there are few comparative assessments of the different integration strategies. One recent assessment is by Dee (2007). This paper is an attempt to distinguish the content of various integration strategies, as well as to assess their relative economic effects. In particular, the paper characterizes the type of behind-the-border economic cooperation that appears to take place in PTAs, and distinguishes it from behind-theborder cooperation that might require another forum. 1. Definition of integration strategies The paper notes that while those PTAs that have addressed behind-the-border issues in their chapters on services and investment have generally gone further than the GATS (see also Roy, Marchetti and Lim 2006), they have tended to be selective in two important ways: they have tended to be preferential, even in the provisions that go beyond goods trade; and they have tended to target only those provisions that explicitly discriminate against foreigners. There are strong political economy explanations for both of these outcomes. One reason for concessions in PTAs to be preferential, even when logic suggests they could sensibly be made non-preferentially, is to protect offensive interests in the Doha Round by not giving away negotiating coin in advance of Doha settlement. This reason for selectivity need not apply in the WTO itself. But other reasons for selectivity may apply in both forums. If PTA commitments are to be preferential, this often dictates that they target only those provisions that explicitly discriminate against foreigners in many cases, these are the only provisions that can feasibly be liberalized on a preferential basis. 2 Another potential reason for reluctance to make deep non-discriminatory commitments within PTAs is that this can be seen by many countries, both developed and developing, as too much of a threat to the right to regulate. Negotiating modalities also contribute a focus on provisions that explicitly discriminate against foreigners. The request-and-offer approach has been used in the Doha Round, and is also the means by which many PTAs are negotiated. This asks countries to contemplate reforms that are in their trading partners best interests. It will generally be in a trading partner s best interests to target only those provisions that explicitly discriminate against foreigners, since this maximizes their market share. Foreign producers generally do not want to unleash competition from potential domestic new 2 The converse does not hold. Because some provisions do discriminate against foreigners, it does not mean that they can be liberalized on a preferential basis. For example, when countries liberalize restrictions on foreign ownership, it may be very difficult to ensure that the new foreign owners are only from selected partner countries. 4

6 entrants. And if domestic competition is sufficiently constrained to begin with, the trade commitment may simply allow foreign producers to join or supplant an existing cartel, thus handing monopoly rents to foreigners (see also Francois and Wooton, 2001). Indeed, this is partly the basis of the East Asian desire to have safeguard provisions in services negotiations. Visibility is also a factor. Regulatory regimes are typically both complex and nontransparent. The regulations most visible to potential foreign entrants may be those that discriminate against foreigners. Finally, WTO disciplines only require PTAs to remove substantially all discrimination against foreigners. They do not require them to address regulatory restrictions that affect domestic and foreign suppliers equally. Thus the paper distinguishes two forms of behind-the-border economic cooperation. The first is granting national treatment, which is GATS terminology for removing discrimination against foreigners either in the terms on which they enter the market, or the conditions under which they operate once they have entered. Granting national treatment is what tends to occur in PTAs or in the WTO. The other strategy is comprehensive domestic regulatory reform, defined as removing all the nondiscriminatory regulatory restrictions that affect domestic and foreign entrants equally. This strategy tends not to occur in either PTAs or in the WTO (except through specific agreements such as the GATS Reference Paper on Telecommunications, which is largely a template for domestic regulatory reform). 2. Evaluation of economic effects The paper then goes on to assess the economic effects of three alternative scenarios: a PTA among ASEAN+3 members, involving elimination of tariffs on manufactures on a preferential basis, and the granting of national treatment to other ASEAN+3 members in various services sectors for which estimates of behind-the-border regulatory barriers were available; an indicative Doha Round settlement, involving a 25 per cent cut in agricultural protection (tariffs, export subsidies and domestic support), a 25 per cent cut in tariffs on manufactures, and the granting of national treatment to all other WTO members in the same services sectors; and comprehensive domestic regulatory reform in each of the ASEAN+3 members. It was thus judged that significant progress on agricultural trade reform would not occur in an ASEAN+3 PTA, but would require multilateral action in the Doha Round. The paper compares the effects of these three scenarios on the economic well-being of the ASEAN+3 countries as a group. To understand the economic effects of the various scenarios, it helps to understand the sources of economic gain from behind-the-border regulatory reforms. Some regulatory trade restrictions, particularly quantitative restrictions, create artificial scarcity. The prices of services are inflated, not because the real resource cost of 5

7 producing them has gone up, but because incumbent firms are able to earn economic rents akin to a tax, but with the revenue flowing to the incumbent rather than to government. Liberalization of these barriers would yield relatively small gains associated with better resource allocation, but also have redistributive effects associated with the elimination of rents to incumbents. Such rent-creating restrictions are tariff-like, with the redistribution of rent having effects similar to the redistribution of tariff revenue. Alternatively, services trade restrictions could increase the real resource cost of doing business. An example would be a requirement for foreign service professionals to retrain in a new economy, rather than to pass an accreditation process. Liberalization would be equivalent to a productivity improvement (saving in real resources), and yield relatively large gains. This could increase returns for the incumbent service providers, as well as lowering costs for users elsewhere in the economy. This distinction has two important implications. First, the gains from liberalizing costescalating barriers is likely to exceed the gains from liberalizing rent-creating barriers by a significant margin (compare figures 1a and 1b). Secondly, in the context of PTAs, the danger of net welfare losses from net trade diversion arises if the relevant barriers are rent-creating, since rent distribution can have the same effects as tariff redistribution (see also Pomfret 1997). So a key question for establishing the economic significance of any behind-the-border regional cooperation is whether it targets trade barriers that create rents or raise costs. The limited empirical evidence suggests that in banking and telecommunications, where explicit barriers to entry are rife, barriers appear to create rents. In distribution services, where indirect trade restrictions also apply, barriers appear to increase costs. In air passenger transport and the professions, barriers appear to have both effects. In particular, discriminatory barriers in the professions appear to create rents, while the non-discriminatory restrictions (such as restrictions that require partnerships, and require both the investors and managers of professional firms to themselves be licensed professionals) increase costs. 3 And theoretical arguments suggest that barriers in maritime and electricity generation primarily affect costs. 4 Dee (2007) shows that if an ASEAN+3 PTA managed to eliminate all discrimination against foreigners in these sectors where empirical evidence is available, the gains would be small compared to a successful completion of the Doha Round. And they would be trivial compared to a comprehensive program of unilateral regulatory reform, one that instead targeted non-discriminatory behind-the-border restrictions that affected domestic and foreign suppliers equally. The results from that exercise are reproduced in Table 1. A key reason for the findings in that paper is that there appears to be a reasonably strong correlation in practice between measures that discriminate against foreigners and measures that create rents. In a broad sense, this is understandable. One of the easiest ways to discriminate against foreign suppliers is to create explicit artificial barriers to their entry, either cross-border or via direct investment, and these barriers tend to create rents rather than increase costs. 3 Gregan and Johnson (1999), Kalirajan et al. (2000), Kalirajan (2000), Nguyen-Hong (2000), OECD (2005), Copenhagen Economics (2005). 4 Steiner (2000), Clark, Dollar and Micco (2004). 6

8 D. An assessment of Asian integration through the East Asian Summit The current paper extends the analysis of Dee (2007) to consider the economic effects of Asian integration pursued through the East Asian Summit (or ASEAN+6) grouping. This allows an assessment of whether the size of the group influences the relative attractiveness of the formal PTA route to Asian integration, relative to a multilateral route through the WTO, or through comprehensive domestic reform. Accordingly, the analysis considers five possible scenarios: a PTA among ASEAN+3 members; a PTA among ASEAN+6 members; a possible Doha Round settlement; comprehensive domestic regulatory reform in each of the ASEAN+3 members; and comprehensive domestic regulatory reform in each of the ASEAN+6 members. The analysis compares the effects of these three scenarios on the economic well-being of the ASEAN+3 countries as a group, the additional three East Asian Summit members as a group, and the entire ASEAN+6 grouping. So the analysis can show the effects of integration strategies involving ASEAN+3 on those Asian countries such as India that would be left out of the group. It can also show the effects on the smaller ASEAN+3 group of extending the membership of the group toasean+6. In short, the results can show the full payoff matrix from extending the group membership. The content of each scenario is kept qualitatively the same as in Dee (2007). Thus, the PTA scenarios involve the elimination of manufacturing tariffs among group members, and the granting of national treatment to other group members in six services sectors banking, distribution, electricity generation, maritime, the professions, and telecommunications. The Doha Round scenario involves a 25 per cent cut in both manufacturing tariffs and agricultural protection, along with the global granting of national treatment in the same six services sectors. Comprehensive domestic regulatory reform involves the elimination of all non-discriminatory regulatory restrictions that affect domestic and foreign suppliers equally in the same six services sectors. Quantitatively, the scenarios differ from those in Dee (2007) in several respects. First, the database for the modelling framework has been updated from 1997 to The model s database is now drawn from version 6 of the GTAP model database (Hertel 1997). The model s theoretical framework is an extension of the GTAP model framework, which accounts for services delivered via foreign direct investment. It is described briefly in Box 1. The model gives a long run snapshot of how economic outcomes would differ from a business as usual scenario, about 10 years after all policy changes have taken place. The regional and sectoral aggregation used for the current exercise is more detailed than in Dee (2007), and is shown in Table 2. There are two reasons for retaining as 7

9 much commodity detail as computer capacity will allow, given the regional breakdown. The first is that it is only with sufficient commodity detail that a model such as this can accurately capture tariff peaks, which account for any trade diversion outcomes in PTA scenarios. Secondly, it is only with sufficient commodity detail that a model such as this can accurately capture the network nature of existing trade and investment relationships among individual Asian economies. These network relationships can also have a powerful influence over the outcomes of any PTAs (see also Dee 2008). The estimates of the direct price impacts of current regulatory restrictions are drawn from a variety of sources, which are summarized in detail in Dee (2005). For the current exercise, the estimates have been updated wherever possible to reflect current regulatory policy settings. For example, the estimates for China and Vietnam take into account their WTO accession, and the estimates for Vietnam have been incorporated into the average for the ASEAN group. Tables 3-9 show the resulting estimates of the tax or productivity equivalents of the restrictions currently found in each Asian region for five services sectors banking, distribution, electricity generation, the professions, and telecommunications. Where the empirical evidence suggests that barriers create rents, they are modelled as tax equivalents, and this is indicated in the tables as price impacts affecting markups. Where the empirical evidence suggests that barriers instead raise costs, they are modelled as productivity equivalents, and this is indicated in the tables as price impacts affecting costs. These tables indicate that although there are some significant restrictions in Asia that discriminate explicitly against foreigners, especially in banking, telecommunications and the professions, they tend to be of the sort to increase markups of prices over costs. Indeed, the restrictions that are prevalent in these sectors are regulatory barriers to entry that tend to create rents for incumbent service providers. By contrast, regulatory impediments that raise real resource costs are also prevalent in the region, in sectors such as distribution and electricity generation, but these impediments tend to affect both domestic and foreign providers. 5 Where national treatment is granted in the PTA scenarios, in most cases this means that the tax or productivity equivalents of the regulations affecting foreign providers (shown in tables 3-9) are reduced so as to be the same as those affecting domestic providers. 6 Where the scenarios involve comprehensive domestic regulatory reform, in most cases this means that the tax or productivity equivalents of the regulations affecting domestic providers (shown in tables 3-9) are reduced to zero, while the tax or productivity equivalents of the regulations affecting foreign providers are reduced by the same absolute amount. With only non-discriminatory regulations being reformed, any additional margin of discrimination against foreign providers is retained. 5 The estimate of the price impact of regulatory restrictions in the distribution sector in India may appear small, but this is because many of the significant barriers remaining in this sector affect the ongoing operations of distribution firms, while the econometric evidence is that the barriers that have a significant impact on costs are those that affect entry (see also Kalirajan, 2000). India has liberalized its entry restrictions in recent years, although some remain. 6 In two instances (distribution and the professions), full national treatment implies tax or productivity equivalents that are lower than otherwise, but still not equal to those affecting domestic producers, because the underlying econometric analysis in Kalirajan (2000) and Nguyen-Hong (2000) suggested that the same regulations have different effects on domestic and foreign service providers. 8

10 The treatment of regulatory restrictions in maritime differs from that in Dee (2007). The earlier paper was restricted to regulations affecting port services, as quantified in Clark, Dollar and Micco (2004), and the regulations were modelled via a productivity shifter affecting the domestic output of water transport services. The current treatment uses information on regulations affecting both port services and maritime shipping services, as collected by McGuire, Schuele and Smith (2000), and the subsequent quantification of their direct impact on maritime shipping charges by Kang (2000). Kang shows that the shipping charges for exporting from country A to country B are affected relatively heavily by regulatory restrictions in the exporting country A, and are somewhat affected by regulations in the importing country B. Thus there is a bilateral dimension to the direct price effects of current maritime restrictions, as shown in table 10. These price impacts are modelled as productivity shifters affecting the maritime shipping margins on shipping goods and services between countries (the so-called iceberg treatment). Significantly, this means that maritime is a sector where the removal of discrimination against foreigners is now modelled as having productivity effects. The granting of national treatment in maritime involves reducing the productivity effects shown in table 10 to a level that would apply if foreigners were regulated the same as domestic operators, information that is available in McGuire, Schuele and Smith (2000). For most of the Asian economies, it means that the price impacts are roughly halved. This alternative treatment of maritime means that the current analysis is more likely to find in favour of the PTA route to integration, relative to the analysis in Dee (2007). Finally, as in Dee (2007), none of the scenarios involve any liberalization of air passenger transport, although the price impacts of the restrictions currently embodied in bilateral air service agreements are recognized in the model as a pre-existing distortion (see table 11, which is based on methodology described in Dee, 2005, but updated with more recent information on the content of air service agreements from ICAO, 2004). Air passenger transport is carved out of the GATS, and is also excluded from most PTAs. The focus of the domestic reform scenarios is on measures that can be taken unilaterally, whereas liberalization of air services agreements requires bilateral action. 1. Evaluation of economic effects The effects of each of the five integration scenarios on the combined welfare of the ASEAN+3 group of countries is shown in table 12. The effects of the same scenarios on the combined welfare of India, Australia and New Zealand is shown in Table 12. For convenience, the effects of the scenarios on the combined welfare of the ASEAN+6 group of countries is shown in Table 14, although the numbers there are simply the sum of the numbers in the previous two tables. In each case, the results are a long run snapshot, showing the extent of deviation from baseline in economic well-being about ten years after the full implementation of the integration package. The results in table 12 show that despite an alternative treatment of maritime regulation that is biased in favour of finding bigger relative gains from a PTA route to integration, the relative benefits of the alternative paths for the ASEAN+3 group of countries are about the same as in Dee (2007). Interestingly, expanding the size of the PTA group from ASEAN+3 to ASEAN+6 does not greatly increase the benefits of the PTA route to the ASEAN+3 group. The PTA route then provides gains that are still only two-thirds as 9

11 big as those from a successful Doha Round, and about a quarter of those from comprehensive unilateral regulatory reform. The results in table 13 confirm that India, Australia and New Zealand would be worse off that they otherwise would be if ASEAN+3 were to form a PTA without them. But interestingly, they do not gain much as a group from being included in the PTA either. A key driver of this latter finding is that India is projected to be slightly worse off than otherwise from its inclusion in an ASEAN+6 PTA. According to the levels of tariff protection embodied in version 6 of the GTAP database (the source of data on levels of protection in merchandise trade in the current exercise), India s tariffs on manufacturing are higher on average than elsewhere in the ASEAN+6 grouping, despite its recent trade reforms. India is thus susceptible to terms of trade losses from joining a PTA with more open trading partners. This has been highlighted as a theoretical possibility by Panagariya (1999, 2000). It is reflected in the projections shown in table 13. China is projected to suffer from the same phenomenon when the PTA is restricted to ASEAN+3, and for the same reason. Both India and China are therefore likely to be much better off when tariff liberalization takes place multilaterally, with many more trading partners. The terms of trade effects are much more likely to be benign in these circumstances than when they join a PTA with a few, more open trading partners. So while the Doha route to integration looks only mildly preferable to the PTA route for the groups to which India and China belong, the Doha route is strongly preferable to them as individual countries. In fact, for India, Australia and New Zealand as a group, a successful Doha Round settlement is projected to be almost as worthwhile as comprehensive domestic regulatory reform to which they were a party. Elsewhere in the Asian region, comprehensive domestic regulatory reform clearly dominates both the multilateral and PTA strategies as a way of raising real incomes. The analysis so far suggests that for most of the Asian region, comprehensive domestic regulatory reform dominates as an integration strategy, while PTAs deliver relatively trivial gains. 7 Multilateral action is an intermediate strategy, and is important for delivering trade liberalization in sensitive sectors or in highly protected economies. The results are shown to be relatively robust to the size of the Asian PTA grouping. E. Furthering domestic regulatory reform Comprehensive domestic regulatory reform of non-discriminatory restrictions can deliver large gains to much of the Asian region. But to date this agenda has not been greatly fostered by PTAs or by the WTO, both of which tend to concentrate on national treatment and discrimination against foreigners. So what alternative institution for regional cooperation could best support this policy agenda in the individual countries of the Asian region? To understand the ways in which regional cooperation could support a domestic regulatory reform agenda, it is important to understand the impediments to domestic regulatory reform. Since the economic focus of domestic regulatory reform is on behindthe-border measures, the political economy of structural reform is primarily domestic. 7 The modeling results do not presuppose that all countries remove their non-discriminatory restrictions at the same time, or at the same rate. 10

12 The core of this type of structural reform is to remove the artificial barriers that protect incumbent firms from any competition, be it from domestic or foreign new entrants. And the real art is to ensure that policy settings promote competition, not just particular competitors. This means that domestic regulatory reform is not easy, because of the diversity of economic interests involved. These include incumbent producers, potential new entrants (domestic and foreign), upstream industries that supply inputs to the reforming sector, downstream industries that buy inputs from the reforming sector, consumers, and governments. If institutions are to support the domestic reform process, they need to support both the economics and the political economy of domestic reform. In particular, they need to mediate among the diverse range of economic interests involved and help build a coalition for reform Impediments to structural reform There are at least three possible reasons why good domestic regulatory reforms are not adopted, each with different implications for the type of regional cooperation that could help support the domestic reform process. Governments do not know what is policy best or better practice, or lack the capacity to implement it. Governments know what is better practice, but face political resistance from vested interests. Governments do not want better policy, because they rely on the rents from current policies for political funding purposes. In all three cases, there is a useful role for policy reviews, and hence for institutions that can undertake them. If the problem is one of identifying better policies, then the policy reviews provide a technical solution they review current policy settings and identify better options. But if the problem is managing vested interests (which may include those within government), then the policy reviews are a strategy rather than a technical solution. They can provide ammunition with which to manage vested interests and build a coalition in favour of reform, but there is no guarantee that the reviews will be decisive in any particular instance. Over time, however, they can help influence the terms of the debate. Each of these roles of policy review institutions is now discussed in turn. a. Identifying better policy options For a policy review to make a useful contribution to identifying better policy options, it needs to be done early in the policy development process, before positions become locked in. A policy review involves a sequential process of articulating the policy problem and the desired objectives, assessing a range of options, evaluating the effects of each option on all stakeholders, recommending the best option, or else explaining why some other option is preferred, and outlining a strategy to implement the preferred option and then review its operation, thus providing an ex post check on policy performance. 8 This section summarizes analysis and case studies from Dee (forthcoming). 11

13 Ideally, the policy review process also involves a great deal of consultation. There can be at least two rounds of consultation with all relevant stakeholders one at the beginning, as the review panel or agency is starting to develop its ideas, and another after the preparation of its draft report that outlines the full analysis and possible policy solutions (sometimes, but not always, including a preferred solution at the draft stage). This consultation helps the review agency to identify all the costs and benefits to each stakeholder. It also provides an opportunity for the review panel to explain its preferred solution to stakeholders. Sometimes the policy development process does not work in an orderly fashion. Instead, policy-making results from political imperatives for quick action, or deals with particular interest groups, or bargaining between political parties. Proposals may be put to government without any prior policy review or consultation. There may be very little scope for other stakeholders to have input into the decision-making process. And there may be little public scrutiny, if at all, until after the decision is made. No country in the Asia-Pacific region always has orderly policy development. But many countries have at least some of the institutional elements that allow for prior policy review and consultation, as will be seen later. b. Managing vested interests The opportunity provided by policy reviews for transparent consultation is the key to their strategic role in managing vested interests. Policy reviews can help deal with vested interests in a number of ways. First, policy reviews can help set the agenda policy change will not happen if nobody talks about it. At this early stage, the aim of the review is to highlight who is losing from current policy settings, perhaps to quantify the size of those costs, and to show how the losers might be better off under policy alternatives. This can help alert the losers, who may be aware of the direct costs they face, but may be less aware of the indirect costs. Policy reviews can galvanize them to start pressuring for policy change, and provide ammunition for such a campaign. Second, reviews can also set the parameters of the debate. They can establish analytical frameworks within which to examine the costs and benefits of current policy settings to all stakeholders. While the analytical frameworks may themselves become the targets of debate, they nevertheless make it much harder to ignore the interests of particular stakeholders altogether, particularly those, such as consumers, who may not be well-organized or are less able to represent themselves. Third, a policy review process can help to depoliticize a debate. And most governments will find it in their own interests on occasion to take the heat off an issue by referring it for independent, objective study. Fourth, policy reviews can name and shame the recipients of special deals. Reviews can help to clarify the gainers from current policy settings, as well as the losers. This in turn can also start to put pressure on the political processes themselves. Fifth, a policy review process can marshal countervailing interests against vested interests, and help to build a coalition in favour of reform. It may not always be obvious 12

14 to governments in advance who the champions of structural reform might be. A policy review process that invites the participation of all interested parties can allow the champions to self-select. It also provides a forum for groups of champions to identify their common interests, and to agree to cooperate in pro-reform strategies. Sixth, policy reviews with a sufficiently broad purview can help to identify policy combinations that lead to so-called Pareto improvements, where at least some stakeholders are better off and none is worse off. Policy reviews can also identify options, such as phased implementation or adjustment assistance, to ease the burden on those that would otherwise lose from structural reform. Both the policy reviews and the review institutions can thereby achieve greater acceptance from particular vested interests, and the reviews can thus help to build a grand coalition in favour of reform. c. Assisting policy coordination within government Ministers with an economic reform agenda also have to build consensus within government. Most government structures have at least some central agencies with economy-wide responsibilities. These tend to take a broad, horizontal view of economic issues. There are also line government departments in charge of overseeing sectional policies, staffed with skilled specialists who are responsible for policy implementation. The interests of such latter ministries are often closely aligned with the interests of the producers in their respective sectors. Ministers with an economic reform agenda need to be able to harness the economywide views of the central agencies in the policy development process, while using the specialist skills and expertise of the sectoral agencies for both policy development and implementation. Coordination problems can arise when the narrower perspective of the sectoral line agencies means that they are against structural reform. Policy reviews can help with such coordination problems in three ways. A policy review agency can consult with the various central and line government departments, just as it consults with all other potential stakeholders. This can alert the sectoral departments to broader economic considerations, just as it does to other stakeholders. Where central agencies are assigned an active coordinating role, they can use independent policy reviews to strengthen the public interest during the policy coordination process. When sectoral departments need to abide by coordinated decisions and implement the structural reforms as intended, ex post policy reviews can expose any nonperformance. 2. The characteristics of effective institutions to support the policy process Given these various uses of policy reviews, there are three key attributes that help to ensure the effectiveness of the policy review institutions: 13

15 statutory independence; an economy-wide view; and transparent processes. Independence is needed to ensure that the review agency is not bound by current government policy, as line government departments often are. In addition, the review agency should not have an implicit stake in the status quo, as regulatory agencies in charge of implementing current economic policy often do. The review agency need not be independent of government funding secure tenure arrangements for the individuals responsible for the reviews may be sufficient. The critical point is that the review agency should be able to critique current policy without constraint. The review agency needs to look beyond narrow sectional interests, and to consider net gains to the economy as a whole. Having a review agency required to take an economywide view ensures that consumers interests are taken into account, as well as those of producers and upstream and downstream industries. Ensuring that the review agencies has access to analytical resources, including skills in partial and general equilibrium modelling and cost benefit analysis, can help it to evaluate impacts on consumers, as well as other stakeholders, even when consumer groups do not participate in the review process. The review agency needs to ensure the transparency of the arguments and analysis put to it, by holding public hearings or publishing summaries of submissions, for example. The agency s reports to government should also be made public, ensuring the transparency of its own advice to government. Transparent processes bolster the ability of at least some countervailing interests to marshal against particular vested interests. This in turn helps to ensure that an economy-wide view is taken by policy makers. Transparent processes can also relieve the government from having to marshal those countervailing interests itself. A review agency cannot possibly undertake the policy development for every single policy proposal. It is particularly useful, however, when there are major potential efficiency or other payoffs from reform, but where existing entitlements create resistance to that reform. The government need not be bound by the recommendations of a review agency. Indeed, the independence of the agency may be bolstered by removing it from any responsibility for downstream policy implementation. Nevertheless, a purely advisory review agency can still have a major impact over the longer term, even if it is sometimes ignored, because both ideas and transparent processes have influence. A policy review institution faces inevitable attack from vested interests, and will need to protect its credibility in the face of such attacks. Credibility is enhanced if its analysis is high quality, and if the agency is resourced accordingly. Credibility may also be enhanced if the agency maintains the status of an honest broker, mediating among the special interests and making recommendations, but not becoming involved in the politics of the subsequent decision-making. But to have influence, while remaining above the fray, its ideas will still need to be championed, or at least debated, by others. Thus an educated and literate commentariat may also bolster both the credibility and the 14

16 influence of the review agency. But by the same token, a review agency can help create an educated and literate commentariat, through its own agenda-setting. Finally, credibility can be built over time as each vested interest becomes the beneficiary of some reforms, even though it loses from others. Thus, even vested interests can come to recognize that they have a stake in the long-term survival of a policy review institution. A statutorily independent review body is not the only type of organization that could carry out the review and consultation phases of policy development. Government departments can develop their own consultation mechanisms, as input to their own policy development processes. Inter-departmental committee processes convened by a central coordinating agency can also bring a number of stakeholder interests to bear by proxy, through the representative line departments. However, neither of these mechanisms necessarily ensure that consumer interests are taken into account. Further, the public consultation tends to be limited to an early stage, because rarely do government departments or inter-departmental committees circulate their own reform proposals for public comment once they have been formulated. Another type of review mechanism is to convene a review panel of eminent persons on a once-off basis to consider a particular issue. Such panels often rely on the integrity of individual appointees for their independence and impartiality. Such panels also depend for their effectiveness on their terms of reference those that are tailor-made to a particular issue are more susceptible to manipulation than policy guidelines that need to be applied across a whole range of issues. 3. Regional examples of institutions to support structural reform Regional examples of policy review institutions that support the political economy of structural reform vary widely. However, they all bear at least some of the hallmarks of effective policy review institutions. The Council on Economic and Fiscal Policy in Japan has been instrumental in achieving structural reforms of pensions and medical care, among others. One key attribute is its transparency summary minutes of Council meetings are published within three days of Council meetings. This has helped to expose, and thereby neutralize, the special pleading of business vested interests. Another key attribute is the representation on the Council of government departments with broad portfolio responsibility, as well as two representatives each from academia and the business community. This has ensured an integrated approach to economic policy-making and has helped to neutralize vested interests within the bureaucracy. Another example is the Australian Productivity Commission. It is a purely advisory body, but the three key characteristics of an effective policy review organization are enshrined in its legislation. It was recently commended by the Australian Treasurer as a vital source of public information and advice to government on policy reforms needed to underpin Australia s long-term prosperity. It has a long track record of influencing policy outcomes and steering economic debate. The National Economic and Development Authority (NEDA) in the Philippines serves as the national and regional development plan and program coordinator among the various 15