Ebook A Cross-Country Empirical Analysis of International Reserves
The recent Asian financial crisis has rekindled considerable interest in issues related to international reserves. Although numerous studies have attempted to unravel the fundamental rationale for the reserve hoarding behavior – ranging from the transaction demand, precautionary motives, collateral asset argument, and mercantilist behavior, the debate on the determinants of international reserves is far from settled. The difficulty of explicating the reserve holding behavior is consistent with the anecdotal view that the role and functionality of international reserves has evolved along with developments in the global financial markets. Recent financial globalization and tremendous advancement in international capital markets has made reserve holding behavior increasingly susceptible to capital account transactions while recent financial crises have also increased the importance of the role of expectations, policy credibility, and institutional structures.
Each wave of balance of payments crises in the last four decades has brought new insights on the causes and consequences of crises. The recent Asian financial crisis in 1997-98 has revealed several new features that are unseen in the previous crises. The so-called “Third Generation” crisis models have highlighted balance sheet factors and the role of financial sector weaknesses as new determinants of currency crises, that were not explicitly incorporated in the previous two generations of crisis models.
Unarguably, the most unique feature of the East Asian financial crisis is that in the aftermath of the event, the economies in the region, mostly notably China, accumulated international reserves rapidly and intensively. The first few years of the 21 st century has witnessed an unprecedented growth of global reserves – over 76.2% between 2000 and 2004, but driven by a handful of economies. During the period, China, Japan, Korea, Malaysia and Taiwan have increased their reserve holdings by 268%, 137%, 107%, 103% and 126%, respectively.
The phenomenal reserve build-up has revived research interest in the determinants of international reserves in the literature. One strain of studies elaborates on the notion of buffer-stock and precautionary demand motivation and incorporates the crisis-induced costs of output and investment contractions (Aizenman et al. 2004, Lee, 2004). In a well-cited paper, Dooley et al. (2003, 2004a, b) resurrect the mercantilist view and suggest that reserve accumulation in East Asia is a consequence of export-oriented growth strategy and the absence of a well-functioning financial system in the region. Aizenman and Lee (2005), on the other hand, empirically confirm the mercantilist motivation, but find that, compared to precautionary demand, it accounts for a relative small amount of reserve hoarding. Other studies also highlight short-term external debts, financial development, and political and institution variables as “new” determinants of reserve adequacy.
This new trend in the literature has raised two questions to our attention: To what extent do these new factors help us understand the observed reserve holdings? Have the determinants of reserve holding changed over time? Answering these two questions should also shed light on the question of what is the optimal level of international reserve holding that would reduce balance of payments vulnerability and help avert possible balance of payments crisis.
To explore these questions while uncovering the determinants of reserve holdings, we conduct an extensive empirical analysis using the data for more than 100 economies over the period of 1975 to 2004. In designing the empirical architecture, we account for some known results in the literature. For instance, previous studies have documented that the developed and developing economies display different demand for international reserves (Frenkel 1974). Other have evidenced that the nature of the demand for international reserves has changed in the presence of significant historical events such as the breakdown of the Bretton Woods system and oil crises (Bahmani-Oskooee, 1988; Frenkel, 1980; Lizondo and Mathieson 1987). Most recently, Aizenman et al. (2004) also identify structural changes in Korean reserve holding behavior after the East Asian crisis. Thus, in this paper, we sort the economies into two groups, the developed and developing economies, and investigate the determinants of the demand for international reserves in separate sample periods partitioned by crisis episodes.
In our regression results, we confirm that the demand for reserves of developed economies is different from that of developing economies. The (significant) explanatory variables are found to be quite different across different sample periods. We find statistically significant evidence that both the debt crisis of 1982 and the two crisis episodes in the 1990s the Tequila crisis and the Asian crisis – changed the determination of reserves holding for developing economies, while only the two 1990s crises affected developed economies. Among the determinants of reserves holding, we find import propensity, a proxy to trade openness, as the only determinant that is significant throughout the entire sample period. However, its contribution in terms of the goodness of fit of the models has been declining over years for both developed and developing economies. Financial variables, especially those on external finances, are increasing their importance in explaining the behavior of reserves holding for both groups of economies. Developed economies seem to retain premium in reserves holding compared to developing ones since the early 1980s, that allows the former economies to hold lower volumes of reserves, while developing economies must prepare more insurance to maintain their exchange rate regimes, to recover from a crisis, and to keep providing external finances. Lastly, we find little evidence that developing economies, especially emerging market economies in East Asia such as China, Hong Kong, Singapore, and Korea, are holding excessive reserves from historical perspectives.
The next session introduces the determinants tested in this study. The empirical framework is presented in Section 3, in which careful analysis on each sub-period and different groups of economies is provided. Section 4 compares the reserve holding behavior between developed and developing economies whereas Section 5 examines the optimality of reserves holding during the 1999 – 2004 period. Section 6 presents some concluding remarks.
Download
PDF Ebook A Cross-Country Empirical Analysis of International Reserves
Posted in :
