موج خروشان

This paper examines when and why capital sometimes surges to emerging market economies (EMEs). Using data on net capital flows for 56 EMEs over 1980−2011, we find that global factors, including US interest rates and investor risk aversion act as “gatekeepers” that determine when surges of capital to EMEs will occur. Whether a particular EME receives a surge, and the magnitude of that surge, however, depends largely on domestic factors such as its external financing need, capital account openness, and exchange rate regime. Differentiating between surges driven by exceptional behavior of asset flows (repatriation of foreign assets by domestic residents) from those driven by exceptional behavior of liability flows (nonresident investments into the country), shows the latter to be relatively more sensitive to global factors and contagion.

After collapsing during the 2008 global financial crisis, capital flows to emerging market economies (EMEs) surged in late 2009 and 2010, only to recede rapidly in the second half of 2011. A fresh wave of inflows in 2012 then ended in sharp outflows in mid-2013, leaving EMEs grappling with depreciating currencies in their wake.1 While historically capital flows to EMEs have often been episodic (Fig. 1), the magnitude and volatility of recent flows have posed particular macroeconomic policy challenges, and raised financial-stability concerns. A first step in dealing with such large flows is to understand their characteristics and determinants, which is the purpose of this paper. Full-size image (29 K) Fig. 1. Net capital flows to EMEs, 1980–2011 (in USD billions). Figure options The literature on cross-border capital flows has a long tradition of trying to identify global “push” and domestic “pull” factors that influence flows to recipient countries.2 Yet, in equilibrium, capital flows must reflect the confluence of supply and demand, so there must be both push (supply-side) and pull (demand-side) factors, and it is hard to attribute the observed flows to one side or the other. It may therefore be more meaningful (and, from a policy perspective, more important) to consider the determinants of inflows that are abnormally large, which may occur when push factors are different than normal, or when pull factors are particularly strong (or some combination of the two). This is the tack taken here, where we define surges as exceptionally large net capital inflows (positive flows on a net basis) to the country—specifically, flows that are in the top 30th percentile of both the country-specific and of the full sample's distribution of net capital flows, expressed in percent of GDP—and examine the correlates of their occurrence and magnitude.3 It is common to think of inflow surges as being the result of foreigners pouring money into the country (thereby increasing the country's stock of foreign liabilities), but they could equally result from the asset side—residents selling their assets abroad and repatriating the proceeds, or simply not purchasing as many foreign assets as before. Since we define surges as instances of exceptionally large net capital inflows, they must be associated with some exceptional behavior of liability flows or of asset flows (or of both). We therefore classify our surge observations into those that are predominantly liability-flow driven (and likely reflect the investment decisions of nonresidents) and those that are asset-flow driven (and reflect the investment decisions of residents), and also examine their determinants. 4 The distinction is worth making because the properties of these two types of surges may be quite different—for instance, domestic investors may be more responsive to changes in local conditions because of informational advantages, while foreign investors may be more sensitive to global factors—thereby calling for different policy responses. To conduct our empirical analysis, we use annual data for 56 EMEs over 1980–2011. Before looking at surges, however, we first establish (through quantile regressions) that large net flows are indeed distinct phenomena that behave differently from more normal flows, thus justifying our focus on surges. We then apply our definition of surges, which yields 326 surge observations (around one-fourth of the panel)—roughly synchronized in the early 1980s (prior to the onset of the Latin American debt crisis); the early 1990s (as these countries emerged from crisis); and the mid-2000s, as capital flows to EMEs recovered from the Asian crisis and the Russian default, and accelerated in the run-up to the global financial crisis. Next, using conventional probit models, we identify the correlates of the occurrence of surges, and also examine the factors associated with the magnitude of the flow during the surge. Finally, we further classify our 326 net inflow surge observations into those associated with the exceptional behavior of liability flows and those with the exceptional behavior of asset flows, and examine the similarities and differences of both types of surges using regression techniques as well as binary recursive tree analysis to flesh out interactions and threshold effects. The very synchronicity of surge episodes across countries over the decades suggests that global factors might be at play. Indeed, we find that global factors, including US interest rates, and global risk (as captured by the volatility of the S&P 500 index returns)—are key factors associated with large net capital flows to EMEs. At the same time, whether a particular EME experiences a surge also depends on its own attractiveness as an investment destination. Thus, country fundamentals, including external financing need, financial openness and interconnectedness, real economic growth, and institutional quality, are also associated with the likelihood of experiencing an inflow surge. Conditional on the surge occurring, moreover, domestic pull factors, including the country's external financing need, financial openness, and exchange rate regime are strongly related to its magnitude. Broadly speaking, therefore, global push factors act as “gatekeepers,” controlling whether capital will surge toward EMEs at all, but domestic pull factors determine where—and in what magnitude—they will end up. This explains why inflow surges tend to be synchronized, but also why not all countries experience a surge when, in aggregate, capital flows toward EMEs. Our analysis also shows that most (two-thirds) of the surges in EMEs are associated with the exceptional behavior of liability rather than of asset flows. The correlates of the two types of surges turn out to be quite similar: global factors matter for both, with lower US interest rates and greater global risk appetite encouraging both foreigners to invest more in EMEs, and domestic residents to invest less abroad. Yet some differences are discernible. Foreign investors are equally attuned to local conditions as domestic investors, but tend to be more sensitive to changes in global conditions, and are also more subject to regional contagion than residents. These conclusions are reaffirmed by the binary recursive tree analysis, which shows that global factors are relatively more important for liability-flow surges than for asset-flow surges. These findings—which are robust to different surge definitions, estimation methodologies and specifications, as well as to the potential endogeneity of domestic macroeconomic factors to the inflow surge—hold important policy implications. First, inasmuch as surges reflect exogenous supply-side factors that could reverse abruptly, or are driven by contagion rather than by fundamentals, the case for imposing capital controls on inflow surges that may cause economic or financial disruption is correspondingly stronger. Second, to the extent that advanced economy interest rates are a key determinant of capital flow surges to EMEs, there may be a need for multilateral surveillance to ensure that spillovers are taken into account. And third, if the aggregate volume of capital flows to EMEs is largely determined by supply-side factors, but the allocation of flows across countries depends on local factors (including capital account openness), there may also be a need for coordination among recipient countries to ensure that they do not pursue beggar-thy-neighbor policies in an effort to deflect unwanted surges to each other. Our findings complement those of previous studies. Earlier work on large capital flows by Reinhart and Reinhart (2008) and Cardarelli et al. (2009) mainly catalogs stylized facts surrounding capital inflow “bonanzas” but does not undertake formal analysis of their determinants. A recent paper by Forbes and Warnock (2012) is the closest to our study, but there are some important differences in focus, methodology, findings, and policy implications. Forbes and Warnock identify surges on the basis of gross flows of assets or liabilities, rather than on the basis of net flows. 5 Many of their identified surges in EMEs, therefore, do not necessarily correspond to periods of exceptionally large net flows. 6 While gross flows matter for some purposes, it is the net surge that matters for competitiveness, macroeconomic management, and the economy's aggregate foreign currency exposure that are of key concern to EMEs. Forbes and Warnock also pool advanced economies and EMEs in their analysis, yet capital flow dynamics for the two groups may be quite different (e.g., unlike advanced economies, EMEs typically borrow in foreign currency and are much more susceptible to sudden stops). This may account for their result (in contrast to ours) that advanced economy interest rates are not important for determining surges to EMEs, with the corresponding policy implication that advanced economy monetary policy has no spillovers. 7 Finally, Forbes and Warnock only study the occurrence of a surge, whereas we also look at why the magnitude of the flow varies across surges, and find that global factors act as “gatekeepers” but local factors determine where, and in what magnitude, flows end up. Our contribution to the existing literature is thus three-fold. First, we focus on surges of net capital flows, and establish that they behave differently from more normal levels of net flows. Second, we examine the determinants of both the occurrence of surges, and their magnitude. To this end, we systematically account for the plausible drivers of surges—including a range of global push and domestic pull factors—and exploit a unique database of IMF desk projections to obtain instruments for potentially endogenous domestic determinants of capital inflows (such as real GDP growth). Third, we classify our net flow surges according to whether they are predominantly driven by the exceptional behavior of asset flows or of liability flows, and examine the association of both types of surges with global and local factors using regression techniques as well as binary recursive tree analysis. The rest of the paper is organized as follows. Section 2 establishes that large capital inflows behave qualitatively differently from more normal flows, thus meriting distinct treatment. Section 3 turns to identifying large capital inflow—surge—observations, documents their key features, and presents the main empirical results on the determinants of their occurrence and magnitude, along with an extensive sensitivity analysis of the results. Section 4 describes how surges may be classified as asset-flow and liability-flow driven, and explores the similarities and differences of the two types of surges. Section 5 concludes.

This paper examines the drivers of exceptionally large net capital flows—surges—to EMEs, distinguishing between their occurrence and magnitude, and differentiating between those that reflect exceptional behavior of asset flows from those that reflect exceptional behavior of liability flows. We use a simple algorithm—based on a threshold approach—to identify surges, which correspond to most “well-known” episodes of large capital flows to EMEs: prior to the 1980s debt crisis, the post-debt crisis recovery and run-up to the Asian crisis, and the post-Asian crisis recovery and run-up to the global financial crisis. Our descriptive analysis based on a sample of 56 EMEs over 1980–2011 indicates that surges are synchronized internationally, and have become more frequent in recent years. Nevertheless, even when surges occur globally, they are relatively concentrated: no more than one-half of the EMEs in the sample experience a surge at the same time, with some countries experiencing them repeatedly. The amount of capital received in a surge also varies considerably across countries. Further, classifying surges based on whether they are driven by asset flows or liability flows, we find that most (over two-thirds) of the net flow surges to EMEs are driven by the latter. The picture that emerges from our regression analysis is one in which global push factors, notably, the US real interest rate and global market uncertainty, determine whether there will be a surge of capital flows towards EMEs generally—which explains why surges are synchronized internationally and why they recur. Even during such episodes, however, a country that has no need for capital or that is an unattractive destination for investors is unlikely to experience an inflow surge. Hence pull factors, including external financing needs, growth performance, the exchange rate regime, financial openness, and institutional quality help determine which countries experience an inflow surge. Conditional on a surge occurring, moreover, it is mainly the domestic pull factors that determine its magnitude. Our results also indicate that domestic and foreign investors respond to both global and local factors such that lower US interest rates encourage capital to flow to EMEs, while increased global market uncertainty drives capital towards traditional safe havens. Foreign investors, however, appear to be more sensitive to global conditions than their domestic counterparts, with a change in the US interest rate and global uncertainty raising the predicted likelihood of liability-flow surges by more than the likelihood of asset-flow surges. A binary recursive tree analysis echoes these results with global factors being most important in explaining liability-flow surges but domestic pull factors dominating in asset-flow surges. Overall, our findings provide a better understanding of large upward swings in capital flows to EMEs. From the policy perspective, they suggest that inasmuch as surges reflect exogenous supply-side factors that could reverse abruptly, the case for EMEs to take preventive measures, including macroprudential measures or capital controls, is correspondingly stronger. To the extent that advanced economy monetary policies are key drivers of inflow surges to EMEs, there may be a need for multilateral surveillance over such policies to ensure that spillovers are taken into account (Ostry et al., 2012). Further, since local factors (including capital account openness) also play a role in determining where the capital ends up, there may be a need for greater coordination between EMEs to ensure that they do not inadvertently pursue beggar-thy-neighbor policies. While the drivers of asset- and liability-flow surges tend to be largely similar, policy responses may need to be tailored to the type of surge—for example, prudential measures might be more important for dealing with financial-stability risks caused by asset-flow surges, but capital controls on inflows may be a viable option for liability-flow surges.