Purchasing Power Parity of ASEAN-5 Countries Revisited: Heterogeneity, Structural Breaks and Cross-sectional Dependence

Abstract

This paper tests the purchasing power parity (PPP) hypothesis for a collection of ASEAN-5 countries using monthly data spanning the period 1968:1–2009:11. For this purpose, a number of recently developed more powerful panel unit root tests that permit for dependence among the individual countries are employed. In addition to this, we utilize the Lagrange multiplier (LM) cointegration test developed by Westerlund, which is flexible enough to accommodate a large degree of country specific heterogeneity, cross-country dependence as well as multiple structural breaks. The main results derived from this study are: first, our findings from panel unit root tests which do not control for cross-sectional dependence appear to be clearly showing evidence against PPP. Second, the evidence from panel tests controlling for cross-sectional dependence is against PPP over the whole and 1997 pre-financial crisis periods. On the other hand, we find sufficient evidence to support PPP for ASEAN-5 countries over the post-financial crisis period. Third, in stark contrast stand the results obtained from the application of the panel cointegration test provide strong evidence of panel cointegration in whole and sub-periods, providing evidence for PPP; however, these findings have become apparent after allowing for multiple structural breaks as well as for general forms of cross-sectional dependence through bootstrap methods. We provide a detailed description of the breaks identified in the analysis, which appear to be closely associated with some macroeconomic shocks and institutional arrangements. The findings of this study offer important policy implications.

Key Words:

• Purchasing power parity
• panel data
• cross-sectional dependence
• mean-reversion
• structural break
• ASEAN economies

Jel Classification:

• C12
• C33
• F31

Acknowledgements

An earlier version of this paper was presented at the 6th International Borneo Business Conference, Universiti Malaysia Sarawak (Malaysia), 20th–21st August, 2014. The authors are grateful to the participants for their comments and suggestions on the earlier draft of this paper. This paper has also benefited from the constructive comments and suggestions of two anonymous referees. These all make this article more valuable and readable. All remaining errors are our own.

Notes

1. Panel methods have become more prominent in recent years since several authors have documented that even for long-run data the available time series suffer from severe size distortion and low power. It is well known that the power of unit root tests for a given sample size can be increased by exploiting cross-sectional information (Levin & Lin, 1993). As such, panel unit root tests have found wide application in testing PPP. For some application of the various panel unit root tests, see Taylor and Sarno (1998), Wu (1996) and O'Connell (1998). Some serious drawbacks of these panel tests were also investigated in O'Connell (1998), Taylor and Sarno (1998) and Breuer et al. (2002).

2. In order to directly compare our results with those (Fujii, 2002; Baharumshah et al., 2008 and Nusair, 2008) among others) we follow these studies and select similar pre-crisis and post-crisis periods which used in their analysis. Furthermore many authors documented that the pre-financial crises period that coincides with the era of financial deregulation and the fast growing phase of the East Asian countries and, the 1998 to 2009 period that incorporates the post-financial crisis years along with the major structural and financial reforms undertaken by the crisis-affected East Asian countries (Zurbruegg & Allsopp, 2004; Kim et al., 2009).

3. Authors points out that regime shifts in these real rates took place due to both the Plaza Accord in 1985 and the Asian currency crisis in 1997. The persistent real appreciation was followed by the Plaza Accord except Indonesia and Malaysia. The outbreak of southeast currency crisis in 1997 caused a sharp real devaluation of Southeast Asian currencies.

4. Another explanation provided by the Wu and Lee (2009) for the failure of rejecting the stationarity could be due to the recent argument that the real exchange rate process is likely to be non-linear due to the existence of transaction costs; hence, the power of these three tests is poor in this situation.