Are Portfolio Flows to Emerging Markets Complementary or Competitive

Abstract

Increasing portfolio investment flows to emerging markets in the past few years have led to fears of a sudden reversal of these flows and possible portfolio switching (from one emerging market to another) among foreign investors. To assess the sustainability of such portfolio flows, the author examines econometrically whether portfolio investment flows to one region in the developing world are significantly related to those going to another region. This question has important policy implications for policymakers in developing countries who, in considering domestic policy reforms to attract foreign portfolio investment, want to ascertain whether financial flows from abroad are coming from an increasing pool of investible resources in the industrial world or whether they represent the same funds chasing different high-yield securities as emerging markets change. In other words, does a sort of"adding-up"constraint apply to these flows - do they function as substitutes or not? Or could these flows be complementary? The author analyzes new quarterly World Bank data on gross portfolio investment flows for eight emerging markets (India, Indonesia, Korea, Thailand, Argentina, Brazil, Chile and Mexico) for the period from the first quarter of 1989 to the second quarter of 1993. Results indicate an inverse relationship between total portfolio flows to emerging Asian stock markets and those to Latin America. This negative relationship holds for both debt portfolio flows and equity portfolio flows. There has been a surge of portfolio flows to developing countries in the 1990s, but developing countries must compete for those flows. In the long term, portfolio flows to well-performing countries will be sustained because of improved creditworthiness and proportionately greater investor interest. Increasing the pace of reform in an emerging stock market is essential for sustaining portfolio flows.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 37 citations exist upstream.