Financial globalization (FG), understood as the deepening of cross-border capital flows and asset holdings, has become increasingly relevant for the developing world for a number of reasons, including the consequences of its changing composition on countries’ balance sheets, its role in the transmission of global financial shocks, its benefits in terms of financial development, international risk, and business cycle smoothing, and the implication of all of the above for macroeconomic and prudential policies. In this paper, we focus on these issues from an empirical perspective, building on, updating, and refocusing the existing literature to characterize the evolution and implications of financial globalization in emerging economies.
As conventional wisdom has it, the globalization process has been growing steadily since the mid-1980s, particularly in developing countries (Kose and others 2010) and has accelerated since the turn of this century, with a dramatic increase in cross-border portfolio flows as a fraction of global wealth (Gagnon and Karolyi 2010). However, this pattern depends on the measure of FG, usually proxied in the literature by the average of cross-border assets and liabilities over GDP (FG-to-GDP ratios). As we show in the first part of the paper, a more natural normalization of foreign holdings by host market size (to control for financial market deepening and spurious relative price [End Page 91] effects) reveals a more stable FG pattern over this period.1 In turn, normalizing foreign portfolio asset holdings by total portfolio holdings by residents shows that, despite the growing financial globalization ratios, international portfolio diversification in the emerging world is still remarkably low and has remained stable or declined.
The second part of the paper is devoted to the costs and benefits of FG in emerging economies, an elusive subject that has produced conflicting results in the literature. Financial globalization has been associated with the deepening of local markets (in terms of credit to the private sector and equity market capitalization) with varied success: the literature has found a positive influence from market depth to FG (Lane and Milesi-Ferretti, 2008; Kose and others 2010) and vice versa (Baltagi, Demetriades, and Law 2009). Identification of causality is further complicated by the choice of the time window: as Mishkin (2007) notes, while entry of foreign capital and institutions may improve domestic financial markets’ conditions through greater competition and liquidity, financial crises could end up blurring this link. We revisit the existing evidence and analyze it through the lens of new proposed metrics that, in our view, are better suited to analyze the question. We find that there is indeed a positive effect that works through market-specific channels (for example, foreign equity liabilities, associated with foreign participation, help deepen local equity markets rather than local financial markets as a whole).
In turn, empirical evidence on the link between financial globalization and consumption smoothing has shown mixed results at best. On one hand, Giannone and Reichlin (2006) report an increase in risk sharing for European countries in the early 1990s, when FG advanced significantly (although their result may be dependent on the specific subsamples used), and Artis and Hoffmann (2006) argue that financial globalization improves risk sharing in the long term. On the other hand, Bai and Zhang (2012) analyze a two-period sample, 1973–85 and 1986–98, for advanced and developing economies and show that although according to their measure financial globalization doubles from period to period, there is no substantial improvement in international [End Page 92] risk sharing. In the same vein, Kose, Prasad, and Terrones (2007) discuss the theoretical advantages of financial globalization in terms of international risk sharing as a way to hedge consumption against domestic income shocks, but they find that only advanced economies have reaped those benefits so far.
We examine the risk-sharing benefits of FG from a critical perspective. We test the evolution of risk sharing, based on the output sensitivity of consumption in emerging markets (“consumption betas,” where both output and consumption are computed relative to the world’s) and find neither improvement in nor link with conventional FG-to-GDP ratios. We argue that this negative result can be attributed to two...