ادغام مالی آسیا: جهانی یا منطقه ای؟ شواهد از بازار پول و اوراق قرضه  Asian financial integration: Global or regional? Evidence from money and bond markets

1. Introduction Since the financial crisis in the late 1990s, financial cooperation within the Asian region is primarily centred around regional financial forums, including the Association of Southeast Asian National Plus Three (ASEAN + 3)1 and the Executive Meeting of East Asia-Pacific Central Banks (EMEAP). Several regional initiatives, including the Chiang Mai initiative (CMI) and the Asian Bond Market Initiative (ABMI) were put in place in 2000 and 2003 respectively to strengthen regional cooperation and integration. Under the CMI, a regime of regional emergency liquidity provision through bilateral swap arrangements was established whilst since the endorsement of the ABMI, local currency-dominated bond markets in the region have achieved remarkable growth in terms of size and diversity of issuers. The ASEAN + 3 Finance Ministers’ meeting in May 2008 agreed on a new ABMI Roadmap setting out tasks to further develop the regional bond market in Southeast Asia. At the same time, cross-border trade and investment within the Asian region have developed at an astonishing pace since the 1990s, which, in turn, have spurred on cross-border financial activities. However, despite the efforts of the regional financial forums and a favourable economic environment, researchers have reached varying conclusions in terms of the degree of financial integration within Asia. The debate rests on whether Asia countries are, in fact, more integrated with the world rather than with each other. Given that financial integration is a multi-faceted and dynamic issue, existing studies have employed various indices to measure the extent of financial integration. Among existing literature that examining financial integration within Asia, price-based measures have been widely employed, although many alternative measures have also been examined.2 A review of existing literature employing price-based measures reveals the following five issues that we intend to address in our study. First, very few studies employ the covered interest parity (CIP) condition, despite the fact that it requires fewer underlying assumptions than uncovered interest parity (UIP) and real interest parity (RIP). Specifically, as suggested by Kim and Lee (2012), the CIP-based measure of financial integration allows one to examine whether the rate of return of comparable financial assets across countries are equalised after factoring in exchange rate. As such, it is generally a more appropriate measure of financial integration. A second worth noting fact is that, although the Asian bond market has grown substantially in the past decade (see Fig. 1) and it is an important part of Asia’s integration process, very few empirical studies have been conducted on this market to analyse the issue of financial integration in Southeast Asia. Third, we find that there is lack of comparison on the level of financial integration among markets where interest rates are of different maturities. Often the short-term interest rates are more responsive to changes in liquidity whilst longer-term interest rates are more responsive to economic fundamentals. Therefore, it would be interesting and informative to examine whether maturity duration affects the level of financial integration. Fourth, the 2008 global financial crisis and recent regional bond initiatives are likely to affect the level of liquidity in the money market and promote the development of Asian bond market, respectively. However, there is no empirical paper that analyses the impact of these two events on the process of Asian financial integration in the context of global versus regional integration. Last but not least, from a methodological perspective, we find that existing studies often employ methods that do not fully capture the dynamics of the integration process in the sense that it is unknown whether countries within the Asian group are forming sub-groups that are integrating at different speeds, irrespective of whether integration exists for all group members as a whole.