Prospects for enhanced exchange rate cooperation in East Asia: some preliminary findings from generalized PPP theory

Abstract

The Asian financial crisis increased economic disparities in the East Asian (EA) region, thus making monetary integration more difficult, but rekindled political interest in Asian monetary and exchange rate cooperation. This article applies the theory of Generalized Purchasing Power Parity (G-PPP), which looks at the behaviour of long-run real exchange rates, to assess the potential for an optimum currency area (OCA) among a subset of EA countries based on five of the more advanced members of the Association of Southeast Asian Nations (ASEAN5). Our findings suggest little support for an OCA for ASEAN5 as a bloc prior to the Asian financial crisis and mixed results in the post-crisis period. In particular, asymmetries in the way countries adjust to shocks and low or insignificant speeds of adjustment were found. Thus, although the application of single OCA criteria is notoriously demanding and our tests apply to only one of the many criteria for the successful formation of an OCA, we cannot find persuasive evidence that ASEAN5 as a group constitute a potential currency area with either the USA or Japan, even when the ‘noisy’ period of the Asian financial crisis is omitted.

¹ A first draft of this article was presented at the Western Economic Association International Pacific Rim Conference, Taipei in January 2003. We would also like to thank the Staff of the Regional Economic Monitoring Unit (REMU) at the Asian Development Bank (ADB) in Manila for their helpful comments on an earlier draft.

Notes

¹ A first draft of this article was presented at the Western Economic Association International Pacific Rim Conference, Taipei in January 2003. We would also like to thank the Staff of the Regional Economic Monitoring Unit (REMU) at the Asian Development Bank (ADB) in Manila for their helpful comments on an earlier draft.

² ASEAN5 consists of Singapore, Malaysia, Indonesia, Thailand and the Philippines.

³ For example, the ‘OCA index’ produced by Eichengreen and Bayoumi (1998).

⁴ As in Bayoumi and Eichengreen (1994) and Chow and Kim (2000).

⁵ For more views on the progress of Asian monetary and exchange rate cooperation (see Kwack 2004, 2005; Stevensen, 2004; Kohsaka, 2004; Nasution, 2005; Pomfret, 2005; Wilson, 2005a, b).

⁶ ASEAN plus Korea, China and Japan.

⁷ Both REMU and ARIC can be found at http://aric.adb.org. For an update on the commitment to monetary integration, see ADB.org

⁸ Although Mundell worried in 1961 that countries which were dissimilar would not be able to have a common monetary policy in the face of asymmetric shocks, he later (Mundell, 1973) realized that this doesn’t make sense for highly diversified industrial countries producing a wide range of goods. Heterogeneous economies could share the risks from asymmetric shocks better within a common currency area as long as there is international portfolio diversification in capital markets, so a country which suffered an adverse shock could easily borrow from other countries in the currency area and so share the risks of asymmetric shocks. Mundell's own views on the prospects for an Asian currency area can be found in Mundell (2003).

⁹ Bayoumi (1994) has neatly incorporated Mundell, McKinnon and Kenen criteria into a general equilibrium model of the OCA adding the key insight that the bloc's common external exchange rate is, by definition, some average of what would have been the equilibrium exchange rates of the members had they not been in a currency union. So if some members experience asymmetric shocks, the external exchange rate cannot move to satisfy them all. However, the net benefits of membership increase for a given country facing a negative asymmetric shock with the degree of openness representing a high level of ‘cross’ or diversified consumption (McKinnon), the degree of international labour mobility (Mundell), and the degree of industrial diversification which reduces the size of aggregate shocks (Kenen).

¹⁰ For a discussion of this issue, see Chow and Kim (2000).

¹¹ Artis and Zhang (2002), for example, found quite distinct ‘core’ and ‘peripheral’ groups in the European Union using fuzzy clustering techniques and Park (2002) finds clear evidence that Europe became a ‘convergence club’ between 1960 and 2000.

¹² Country-specific shocks might include changes in monetary or fiscal policy, in productivity, or in the terms of trade, while regional shocks from the EA point of view could arise from changes in the yen-dollar rate or China's accession to the World Trade Organization. Good examples of global shocks would be an oil price hike, cycles in global electronics demand or supply and the terrorist attacks in the USA in September 2001.

¹³ Since the establishment of the ASEAN Free Trade Agreement in 1992 some progress has been made in tariff reductions, but significant institutional and nontariff barriers remain. Nonetheless in November 2002 ASEAN leaders agreed to work towards an ASEAN Economic Community. For some background on ASEAN trade integration, see Tongzon (2003).

¹⁴ Although Brunei is one of the older members of the bloc and is relatively advanced in terms of income per capita, this is largely a result of its oil and gas resources. There are also data problems since it is not a member of the International Monetary Fund.

¹⁵ McAleer and Lee (2004) also found little evidence of income convergence between pairs of ASEAN5 countries using time series tests.

¹⁶ The data on regional trade integration is taken from the Monetary Authority of Singapore (2005).

¹⁷ We are grateful to an anonymous referee for pointing out the significance of this point.

¹⁸ Australia, Korea, Philippines, Thailand, Singapore.

¹⁹ The CMA was formed in 1974 and by 1990 consisted of Lesotho, Namibia and Swaziland, with Botswana as a de facto member.

²⁰ Although we have taken some care to confirm the nonstationarity of the RER series, since this is fundamental to the G-PPP theory, there remains the possibility that the series may be nonstationary locally but globally mean reverting. A number of recent studies, such as Sarno (2000) and Taylor and Peel (2000), have suggested that the behaviour of the exchange rate for industrial countries is indeed nonlinear in nature, further reducing the power of the standard ADF test. We are grateful to an anonymous referee for pointing this out. For the Asian context, see Chortareas et al. (2002) and Liew et al . (2004).

²¹ It appears to be quite easy to find significant cointegration when testing G-PPP, but difficult to select meaningful cases on the basis of standard robustness tests (Ogawa and Kawasaki, 2003).

²² This implies that the relevant exchange rate is weakly exogenous i.e. the cointegration relationship does not enter into the equation.

²³ See, for example, Peebles and Wilson (2002).