A widely shared view among academics and policymakers is that foreign direct investment (FDI) is a stabilizing factor during episodes of financial crisis in emerging market countries. Underlying this view is the notion that FDI is driven by positive longer-term sentiment about the recipient country and, to the extent that it entails physical investment in plant and equipment, is more difficult to reverse than other capital flows. On this basis, it is argued that policymakers should encourage such flows to “insure” against sudden reversals of capital flows.
Empirical evidence broadly supports the hypothesis that FDI flows are more stable than all other forms of capital. However, one should not be too quick to infer from aggregate data that the behavior of direct investment enterprises during crises is necessarily stabilizing from the perspective of the balance of payments. In particular, a narrow look at FDI flows risks providing only a partial perspective on the financing decisions of FDI enterprises. Further improvements in the scope and coverage of FDI data and more analysis at the firm level are needed to better understand the behavior of direct investors and their host country affiliates in the context of a financial crisis.
What the FDI data tell us
Net FDI flows to emerging market countries increased dramatically in the 1990s, making direct investment the predominant source of private external financing for this country group. Moreover, the positive trend prevailed, notwithstanding several financial crises in emerging markets during 1990–2005 (see chart). Indeed, annual net FDI flows to emerging market countries—both in U.S. dollar terms and as a share of recipient countries’ GDP—have remained consistently positive throughout the period and increased steadily, except in 2002–03, when flows to Latin America declined temporarily in the wake of the crisis in Argentina. In contrast, net portfolio investment and other investment registered substantial net outflows from emerging markets during some of the years under review.
A closer look at selected episodes of financial distress at the country level generally reinforces the view that FDI flows are stabilizing. In particular, FDI is typically more stable than portfolio and other investment flows when the volatility of the respective flows is scaled by their mean size (see table). The data also reveal that in some countries—including Argentina, Indonesia, Brazil, Russia, and Thailand—FDI became the only channel for net private capital inflows immediately after the crisis, a period during which portfolio and other private capital flows exited the country.
The issue: Are FDI flows a stabilizing factor during a financial crisis?
The bottom line: Yes, FDI flows are generally more stable than portfolio and other investment flows.
What’s the problem?: A narrow look at FDI flows risks providing only a partial perspective on the behavior of direct investment enterprises during crises. Moreover, further improvements in the quality and coverage of FDI data are needed to better assess the stabilizing nature of FDI flows.
Despite these findings, capital flows associated with FDI activity are not necessarily more stabilizing than other cross-border capital flows. This is because official FDI data may not adequately capture all the transactions that fall under the definition of FDI. Specifically, while the relevant data compilation practices tend to do a good job of capturing the initial transaction establishing an FDI relationship (the acquisition of at least 10 percent of the ordinary shares or voting power of an enterprise abroad), they are often less accurate at recording subsequent capital transactions between the direct investor in the source country and the direct investment enterprise in the host country. These subsequent flows include fresh injections of equity capital; a reinvestment of earnings; and other flows, including intercompany loans and suppliers credits.
Source: IMF staff estimates.
Note: Includes countries in the Emerging Markets Bond Index, India, and Czech Republic. Darker bars indicate the years of financial crisis: 1995-Mexico; 1997-Korea and Thailand; 1998-Brazil, Indonesia, Philippines, Russia, and Ukraine; 1999-Ecuador; 2001-Argentina and Turkey; 2002-Uruguay.
The information content of FDI data may also be undermined in cases where direct investment enterprises use sophisticated financial engineering to manipulate cross-border financial flows with a view to circumventing capital controls or reducing tax liabilities.
More fundamentally, the accounting of FDI needs to be distinguished clearly from the analysis of financing sources of the direct investment enterprise. For example, to the extent that the parent FDI firm finances its investments by borrowing from host country capital markets, the stabilizing effect of FDI would be partially offset by a potentially destabilizing outward portfolio flow. Conversely, the exposure of the parent firm to the domestic affiliate could be underestimated to the extent that the parent firm provides guarantees for the affiliate’s borrowing from unconnected foreign sources. Such borrowing would show up as a portfolio inflow, not as direct investment, even though the investment risk is borne by the parent.
Furthermore, surveys of FDI investors reveal that in the context of financial crises, multinational firms encourage their foreign affiliates to borrow more in domestic capital markets based on their own capacity. The scope for such transactions is, of course, critically related to the strength of the affiliate’s balance sheet and depends on the continued willingness of banks and capital markets to provide financing during periods of financial distress.
Recent studies indicate that this disconnect between reported FDI flows and the actual financing of FDI enterprises is not just conceptual. For example, economist Alexander Lehmann found in a 2004 IMF Working Paper (WP/04/107) that once all financing from host country and other external sources is included, the balance sheets of U.S. foreign affiliates with majority U.S. ownership are about three times as large as suggested by the FDI data, which cover only the debt and equity provided by the parent company. As financing of FDI investments is increasingly sourced in local markets, this gap is likely to grow in the future.
|Coefficient of variation|
|Direct investment||Portfolio and other|
Room for improvement
Despite these potential complications, evidence from a limited number of case studies suggests that the financial decisions of direct investment enterprises in emerging markets tend to contribute to current and capital account stability during periods of financial distress. In particular, local affiliates of multinational firms were found to shift quickly from host country sales to export sales; reduce dividend outflows with a view to strengthening their capital base; and expand economic activity at a time when locally owned firms facing more severe financing constraints were unable to do the same.
Additional firm-level research is needed to build a more solid body of empirical evidence on whether or not the behavior of direct investment enterprises in emerging economies helps avoid or mitigates balance of payments crises. Moreover, weaknesses in existing FDI data point to the need for better statistics. In this context, policymakers can be expected to benefit from work in the area of foreign affiliate trade statistics, as well as from ongoing initiatives aimed at improving the collection of comprehensive and harmonized information on the stock of inward and outward FDI—possibly including the Coordinated Direct Investment Survey conducted by IMF staff in collaboration with other international agencies.
Patricia Brukoff and Bjorn Rother
IMF Policy Development and Review Department