Capital flows often come in waves. An extensive literature has documented 'surges' and 'bonanzas' in capital flows (e.g. Forbes and Warnock 2012, Reinhart and Reinhart 2009, Ghosh et al. 2012). While capital flows can bring many benefits the literature has also documented the risks associated with this cyclical nature. Indeed, they can contribute to amplifying economic cycles, fuel credit booms, appreciate the real exchange rate, and can be subject to sudden reversals (Calvo et al. 2008).
The perceived wisdom is that there is a pecking order among capital flows, with foreign direct investment (FDI) perceived as ‘good’ as it promotes growth, while portfolio investment is seen as ‘bad’ as it is more volatile and can lead to excessive business-cycle fluctuations. Yet evidence from the latest financial crisis suggests that large FDI flows in the financial sector may be related to macroeconomic instability in the receiving countries (Ostry et al. 2010). Isn’t any type of FDI supposed to be good? Maybe heterogeneity across sectors is an aspect so far neglected in the literature.
In new research (Dell’Erba and Reinhardt 2013) we examine episodes of large gross FDI inflows (surges) at the sectoral level. We focus on emerging market economies during the period 1994-2009, employing a newly constructed data set for gross sector-level FDI inflows.
As in Forbes and Warnock (2012) and Rothenberg and Warnock (2011), we focus on gross large capital inflows as net flows can mask dramatic changes in gross flows. We focus on the split between financial-sector FDI and non-financial sector FDI which is available for 56 emerging markets.1
Measuring surges in sectoral FDI inflows
To identify 'surges', we follow a methodology similar to that proposed by Cardarelli et al. (2010). We classify a sectoral inflow as large if (i) it exceeds a country-specific historical trend or (ii) exceeds, by a pre-specified threshold, the historical distribution of the sectoral FDI inflows within a region. In both cases, we do not classify an inflow as a surge if it is smaller than 0.3 % of GDP.
Characteristics and macroeconomic outcomes of surges in FDI
The shares in countries experiencing surges in FDI vary over time and sector (Figure 1). The first wave of large FDI inflows started in 1996 and is characterised by a high share of surges in the manufacturing sector, which subsequently retrench after the Asian crisis (1997-98); the second one started in 2005 and is associated mostly with surges in the financial sector. Across sectors, we observe similarity with respect to the size of the sectoral flows scaled by GDP around surges (Figure 2).
Figure 1. Share of countries experiencing surges (in %)
Figure 2. Size of sectoral FDI surges
We identify three main facts about sectoral FDI surges:
- FDI surges come in cycles.
- FDI in services (including the financial sector) show a bigger cyclical component.
- The size of FDI inflows during surges in the primary and manufacturing sector is more stable across time.
We find also heterogeneity in the response of macroeconomic indicators around surges in FDI across sectors. In particular, the volatility of real-GDP growth and sectoral value added is much higher around surges in the financial intermediation sector (Figure 3). The results are not driven by the size of flows or the size of the financial sector (on average it is the smallest sector of the economy in terms of its share in total value-added).
Figure 3. Event study: Sectoral FDI surges and macroeconomic outcomes two years before, during booms, and two years after the surges
What can explain the observed pattern in GDP growth during surges in financial-sector FDI? Our analysis suggests the level of domestic credit as the most promising factor (Figure 4). Foreign banks play an important role in the provision of credit to the domestic economy in many emerging markets. We see that domestic credit generally increases during episodes of surges in financial FDI, but the patterns are more pronounced when we look specifically at credit denominated in foreign exchange: we observe a significant increase in the change of the foreign-exchange credit to GDP ratio during surges as well as a significant decline in the share of foreign-exchange credit in total credit after the surge. This indicates that expansions and subsequent reductions in foreign exchange credit may be one factor behind the observed volatility in GDP. As the theoretical literature has shown, the presence of small financial frictions may amplify the transmission of external shocks to the domestic economy when the level of private borrowing increases (Mendoza 2010).
Figure 4. Event Study: Domestic Credit. The ﬁgure reports the mean values of the respective credit variables 2 years before, during booms, and 2 years after the identiﬁed episodes.
Determinants of sectoral FDI surges
When we turn to the empirical determinants of the probability of surges in financial and non-financial FDI, we follow the literature and distinguish across global, domestic and contagion variables. Global growth is the dominant variable in explaining surges in sector-level FDI, with a particularly strong impact on FDI surges in the financial sector. The odds of a country experiencing a surge in financial-sector FDI are 46.7% higher when global growth increases by one percentage point; the figure is 27.3% for the non-financial sector. For regional contagion, we find that it is only significant in explaining FDI surges in financial FDI. The odds of a country experiencing a surge in financial-sector FDI are 33.2% higher if the share of countries that experienced a surge in the preceding year increases by ten percentage points.
With regard to domestic variables, we find - remarkably - that financial-sector FDI surges are more likely in economies that are less financially open, i.e. economies that have more stringent capital controls. Disaggregating capital controls, we find that controls on instruments which may constitute alternative sources of funding for subsidiaries of foreign banks (such as bonds, portfolio equity and money market instruments) tend to increase the likelihood of FDI surges. The results are strongest for controls on bond inflows.
An aggregate measure of the quality of financial regulation (Abiad et al. 2008) is associated negatively, however not significantly, with the probability of experiencing surges in financial-sector FDI. A more specific measure on restrictions on the financial sector's use of foreign exchange (including forex lending), taken from Ostry et al. (2011), impacts the probability of surges in FDI negatively.
The evidence in our paper suggests that FDI surges occur across all sectors but only surges in FDI in the financial sector are accompanied by a boom-bust cycle in GDP growth. These surges in financial-sector FDI are driven by global and contagion factors to a greater extent than FDI surges in other sectors. The results are suggestive of leakages: financial-sector FDI may have been used as a substitute to debt inflows when a country implemented bond-inflow controls. To the extent that financial-sector FDI is a less safe capital flow than other types of FDI, this puts an interesting twist on the implication of other papers studying the impact of capital controls (see Magud et al. (2011) for a review), specifically, that a shift in the composition towards FDI has been beneficial with respect to the riskiness of a country’s external balance sheet. More generally, the results on financial-sector FDI caution against a simplistic view on what constitutes ‘good’ capital flows and suggest that any surveillance on capital inflows may need to be more granular.
Finally, the tentative evidence we provided on regulations restricting lending and borrowing in foreign currencies reducing the probability of surges in financial-sector FDI may have implications for the design of future prudential regulation policies.
Abiad A, E Detragiache and T Tressel (2008), "A new database of financial reforms", IMF Working Paper 08/266.
Calvo G, A Izquierdo and L-F Mejia (2008), "Systemic sudden stops: the relevance of balance-sheet effects and financial integration", NBER Working Paper 14026.
Cardarelli R, S Elekdag and M A Kose (2010), "Capital inflows: macroeconomic implications and policy responses", Economic System 34 (4), 333-356.
Dell’Erba, S and D Reinhardt (2013), “Not all capital waves are alike: A sector-level examination of surges in FDI inflows “, Bank of England Working Paper No. 474.
Forbes, K and F E Warnock (2012a), Capital flow waves: Surges, stops, flight and retrenchment. Journal of International Economics 88, 235-51.
Ghosh A R, J I Kim, M S Qureshi and J Zalduendo (2012), "Surges", IMF Working Paper 12/22.
Magud N E, C Reinhart and K S Rogoff (2011), "Capital controls: myth and reality – a portfolio balance approach", NBER Working Paper 16805.
Mendoza, E G (2010), "Sudden stop, financial crises and leverage", The American Economic Review 100 (5), 1941-66.
Ostry J, A R Ghosh, K F Habermeier, M Chamon, M S Qureshi and D Reinhardt (2010), "Capital inflows: the role of controls", IMF Staff Position Note 2010/04.
Ostry J D, A R Ghosh, K Habermeier, L Laeven, M Chamon, M S Qureshi and A Kokenyne (2011), "Managing capital inflows: what tools to use?" IMF Staff Discussion Note.
Reinhardt, D (2010), "Into the allocation puzzle – A sectoral analysis" Study Center Gerzensee Working Paper Series No. 10.02.
Reinhart, C and V Reinhart (2009), "Capital flow bonanzas: an encompassing view of the past and present", in NBER International Seminar on Macroeconomics (2008), 9-62.
Rothenberg, A and F Warnock (2011), "Sudden flight and true sudden stops", Review of International Economics 19 (3), 509-524.
1 FDI into the aggregate `non-financial sector' is calculated by subtracting financial intermediation sector FDI from total aggregate FDI.