![Sample our Area Studies journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5922/TOC-Subject-Sample-2021-Area-Studies.jpg"})
Abstract
This paper applies the generalized purchasing power parity (G-PPP) theory to assess the potential for an optimum currency area (OCA) for the ASEAN5 plus the big three (ASEAN5+Big3) during a period containing significant structural breaks, using the US, Japan and China as base countries. The relevance of considering breaks is demonstrated by utilizing the Johansen et al. (Citation2000) procedure that allows for up to two predetermined breaks. The results provide support for long-run G-PPP and suggest that the larger group ASEAN5+Big3 could form an OCA in one step. Conversely, for smaller groups, such as the Big3, an OCA is not supported. However, adding one or more countries to the Big3 supports an OCA. The impact of the Asian crisis seems to be limited to when the US is the base country and the results suggest a change in the G-PPP relationship between the pre- and postcrisis periods. Moreover, stability tests suggest that G-PPP has been stable for the period analyzed.
Notes
1. Includes Brunei, Burma, Cambodia, Indonesia, Laos, Malaysia, the Philippines, Singapore, Thailand and Vietnam.
4. The New York Times article ‘Asian Finance Ministers Seek a Common Currency’, May 5, 2006.
5. Almost half of total capital inflows to developing countries went to Asia in 1996 – nearly $100 billion (IMF, June 1998).
6. Except China's CPI, which is obtained from the International Labor Organization Geneva Database (http://laborsta.ilo.org/).Only the annual CPI was available hence, frequency conversion from annual to quarterly CPI was deemed necessary for including China in the study.
7. In September 1985, the finance ministers of the USA, the UK, France, West Germany and Japan agreed that the US dollar was overvalued relative to yen. They agreed to depreciate the dollar relative the yen by lowering the interest differential between the two countries. This resulted in a huge appreciation of the yen from an average of 240 yen per US dollar in 1985 to an average of 200 yen in early 1986.
8. Structural breaks are associated with significant economic and political events, such as changes in exchange rates regimes from fixed to managed or free float, financial crises, building up and bursting of bubbles, financial liberalization and external forces, such as oil embargos and wars.
9. Although cointegration analysis is conducted on pairs of countries, it should be noted that the presence of cointegration for a pair of countries does not necessarily imply cointegration for groups of countries.
10. All results are reported at the 5% significance level or better.
11. In some cases, such as the ASEAN5+Big3, evidence of cointegration is found with and without breaks. However, stability tests provided in Section 4.5 suggest significant instability for the results without breaks. On the other hand, once breaks are included, the results suggest sable G-PPP. Hence, finding evidence of cointegration does not necessarily mean that the long-run relationship is stable. In fact, testing the stability of the long-run relationship comes after confirming cointegration. The stability test is conducted in Section 4.5.
12. Asia is used in this paper to stand for the countries under study.
13. It is not our objective in this paper to examine the weights of the three currencies (dollar, yen and yuan) as in a common currency basket, although this would be an interesting issue for further and future research. However, interested readers are referred to Kawasaki and Ogawa (Citation2006) who estimate the weights of the three major currencies (US dollar, euro and yen) in a common currency basket in East Asia and find that a larger weight for the US dollar tends to make the bilateral exchange rates among East Asian economies stable in the long run.
14. In fact, this point is confirmed in the stability tests provided in section 4.5.
15. This means that the first observations from 1973 to 1977 are used for recursion.
16. Due to the large number of cointegration relationships that have been estimated and hence, the large number of graphs, only few graphs are reported. All unreported results are available upon request from the author.