The external current account in Hungary has been volatile during the transition period since 1989: small surpluses in the early years were followed by current account deficits of about 10 percent of GDP in 1993–94, and a subsequent strengthening with deficits averaging 3¼percent of GDP during 1996–97 (Table 2.3 in Chapter II). From a saving-investment perspective, the relatively strong current account position in the earlier years reflected robust household saving but also weak investment demand in the wake of initial uncertainties surrounding the transition process (Table 2.2 in Chapter II). But with a considerable weakening in public and private sector savings, domestic saving was inadequate to finance the subsequent rebound in investment, leading to a substantial reliance on foreign saving. More recently, following the introduction of a policy adjustment package in early 1995, the current account deficit has dropped sharply: the 1997 deficit-to-GDP ratio is projected to be less than one-third of its 1994 level (Figure 3.1).
External Current Account Deficit and External Debt
Sources: International Monetary Fund, World Economic Outlook, 1997; and IMF staff calculations and projections.1 The figure reports the ratio between gross external debt (minus international reserves) and GDP in Hungary over the simple average of the same ratio computed for the countries classified in the World Economic Outlook as “market borrowers” or “countries without recent debt service difficulties.” The figure for 1997 assumes that the ratio for the other countries is unchanged with respect to 1996.2 The figure reports the ratio between gross external debt (minus international reserves) and exports of goods and services in Hungary over the simple average of the same ratio computed for the countries classified in the World Economic Outlook as “market borrowers” or “countries without recent debt service difficulties.”Developments in the current account have been mirrored to some extent by movements in real exchange rates, in particular, those based on unit labor cost (Figure 2.2 in Chapter II). The appreciation of the forint contributed, with some lag, to the deterioration of the current account in 1993/94 (although several other factors, including the impact of new bankruptcy regulations, were also important).1 With product-based gross real wages declining by about 6½ percent during 1995–96, and concurrent substantial productivity gains, the gains in competitiveness were quickly reflected in a turnaround of the current account. These gains in competitiveness were also evidenced in strong gains in export market share that continued into 1997.
In parallel with the swings in the external current account, Hungary also witnessed relatively large movements in its foreign asset and debt position over the past seven years (Figure 3.1). The strong current account position at the beginning of the decade, as well as foreign direct investment inflows, contributed to a decline in the external debt-to-GDP ratio, which, however, by the end of 1992 remained well above levels in many other emerging markets—a legacy of the past. During 1993–94, sizable new net borrowing was needed to finance the current account deficit, and it was only in 1995 that the net debt ratio began to decline again, helped by the narrowing of the current account deficit and exceptionally large inflows of foreign direct investment, including privatization receipts. Even with its large external debt and debt service burden, Hungary has maintained its impeccable debt service record and, following policy adjustments since 1995, yield spreads have declined rapidly.
The volatility of the external account balance during the 1990s raises the issue of what level of this variable is broadly appropriate as a longer-term equilibrium. To address this issue, this chapter draws mainly on theoretical models that analyze the current account from an intertemporal saving-investment perspective. These models, which build on earlier work in the context of closed economies (for example, Lucas, 1981), highlight the role of the current account as a buffer against temporary shocks (such as a relative cyclical weakening or temporary negative supply shocks). But they also underscore the role of the current account in allocating resources in a more sustained way over time, for example, as a way of building up a country’s foreign assets ahead of a relative aging of its population.
While the main focus of this approach is on flow equilibria, as reflected in the external current account, the approach also has implications for the cumulated sum of a country’s historical current accounts, as reflected in the stock of external debt and foreign assets. Accordingly, in addition to analyzing the current account position, this chapter will also attempt to examine if, from cross-country comparisons, lessons can be drawn about Hungary’s current net foreign asset position and its projected future path.
The chapter draws on ongoing analysis in the IMF on longer-run equilibrium current account and exchange rate paths, and, in particular, on a recent study by Debelle and Faruqee (1996). Based on these results, simulations are presented that may provide a reference point for a current account or net foreign asset position.
Determinants of the External Current Account and Net Foreign Assets: Theoretical Aspects
Intertemporal Models of the Current Account as Saving-Investment Balance
External capital flows enable a country to import for some time more goods and services than it exports. Intertemporal solvency implies that these debts should eventually be repaid through surpluses on the trade and services accounts. In this way, the external current account is fundamentally a venue of intertemporal resource allocation between countries. This aspect is at the center of the approach that analyzes the current account within an intertemporal saving-investment model, and identifies the factors that may make it optimal for countries to run current account imbalances for some periods.
In addressing these issues, it is helpful to recall that net exports (X–M) have to equal ex post private sector saving (SP) minus the sum of the government deficit (DEF) and private investment (I):2
Net exports is the national accounts equivalent to the external current account (CA), excluding transfers; for simplicity, it is assumed in this section that these transfers are zero so that the current account is equal to net exports.
The main focus of the intertemporal saving-investment approach is to identify the determinants of the left-hand side of equation (1), taking into account the intertemporal solvency condition of a country:
whereby the present value of all future current account balances, PV(CA), must equal (the negative value of) a country’s initial net foreign asset position. This allows a country with an initially positive net foreign asset position to “eat up” the assets over time by running current account deficits, and vice versa.
The intertemporal model highlights the role of the current account as buffer for transitory shocks. This essentially extends the consumption smoothing model to an open economy: among other things, it suggests a relatively strong current account position for a country that is in a weaker cyclical position than its trading partners; also, in the case of temporary real exchange rate, terms of trade, or supply shocks, demand smoothing would result in temporary current account imbalances. For example, an unusually bad harvest, as Hungary witnessed for several years in the early 1990s, may have only a limited effect on domestic demand and result in a concurrent temporary deterioration of the external current account.3 Temporary shocks that affect the current account can also be policy induced, including monetary policy shocks and temporary changes in the fiscal policy stance.
The intertemporal approach also identifies factors that may result in more sustained periods of current account imbalances. A potentially important factor in this regard is a country’s relative stage of development: a country that starts from a less-developed position, and is expected to narrow its relative income gap, could be expected to run a current account deficit during part of the catch-up period. While this would allow some intertemporal smoothing of consumption, it should also be reflected in a relatively large share of capital goods imports. As a result, the composition of a current account imbalance may be an important indicator of its sustainability. To some extent, these factors have an analogue on the capital account side. In particular, FDI-based financing may be able to sustain larger current account deficits or deficits for longer time periods.
Demographic factors may also result in sustained periods of current account imbalances. Theories in this respect mostly build on aggregate life-cycle models. In particular, as a country’s population ages relative to its partner countries, its dependency ratio tends to rise; and if, as the life cycle hypothesis suggests, older dependent people have lower propensities to save, this would tend to lower a country’s saving rate over time.4 In order for the intertemporal solvency condition (2) to hold, this country would have to save relatively more in the years prior to the relative rise in its dependency ratio, that is, ceteris paribus, a country would have to record a stronger current account position in the earlier years ahead of the relative aging of its population. Also, demographic developments can affect the fiscal position, for example, through their effect on the pension system. Let us, therefore, turn to the impact of fiscal policy on the current account.
Fiscal policy can affect the current account in these models through several channels (Frenkel and Razin, 1996). The fiscal deficit may affect a country’s current account if Ricardian equivalence does not (fully) hold, so that changes in public sector saving are not fully offset by changes in private sector saving (see equation (1)). Fiscal policy also exerts direct demand effects (or indirect effects via taxation and transfers), and these have repercussions on the external current account. For example, to the extent that the private sector has a different (generally lower) marginal propensity to consume nontradable goods than the public sector, a temporary increase in government demand would affect the current account even if Ricardian equivalence held (Ostry, 1988).
The real exchange rate may also affect saving and investment directly and thereby the left-hand side of equation (1). While empirical evidence in this regard is at best mixed, the real exchange rate has generally a strong impact on external trade flows—that is, the right-hand side of equation (1)—as is captured in standard estimates of relative price elasticities for imports and exports. Therefore, in the empirical model employed in the next section, the real exchange rate is the key endogenous variable that ensures over the medium term ex ante balance between saving minus investment and the external current account, as captured by the two sides of equation (1).5
Finally, intertemporal models of the current account ascribe an important role to interest rates and, more generally, rates of return on assets. First, a rise in real interest rates has a wealth effect. It is negative in countries that are net borrowers, like Hungary, as it increases their debt service burden; as a result, Hungary would have to run larger current account (net of interest payments) surpluses over time if there were a sustained rise in real interest rates. Second, as an intertemporal price, the real interest rate also affects the intertemporal demand and supply pattern, and thereby the current account.
Foreign Asset Equilibria
The cumulated sum of a country’s historical current accounts, adjusted for transfer payments and valuation gains, determines its net foreign asset position at any particular point in time. An alternative way to apply the intertemporal saving-investment model is, therefore, to focus on a country’s net foreign asset position (i.e., a stock equilibrium) rather than the current account (i.e., a flow equilibrium). This may provide some additional insights, in particular in empirical work. For example, by analyzing the net foreign asset position, one can draw on the empirical observation that some, notably developing, countries have encountered (negative) net foreign asset positions that turned out to be ultimately unsustainable. The outstanding stock of net foreign assets may also provide an important indicator of a country’s access to financial markets, an aspect that may not be adequately captured by an exclusive focus on current account flow developments.
With net foreign assets essentially representing the stock equivalent to the current account, the theoretical determinants of a country’s optimal foreign asset position are closely related to the factors identified above in the intertemporal saving-investment model. In turn, net foreign assets have two important effects on a country’s current account. First, the return on these assets directly affects the current account in the form of net interest and dividend payments as well as profit remittances. Second, net foreign asset positions should, over longer time periods, be reversed as indicated by the solvency condition (2) above. It has proven difficult to disentangle these two effects in empirical work on the saving-investment model. As a result, analyzing a country’s foreign asset equilibrium directly may provide useful additional insights into a country’s external position.
Model-Based Results and Cross-Country Comparisons
The theoretical framework of the previous section can now be applied to analyze the external equilibrium for Hungary. However, in view of the fairly short experience with a relatively liberalized balance of payments, the analysis to identify the longer-run importance of different variables has to rely largely on historical evidence and parameter estimates for other countries. These parameter estimates are applied in this section to Hungary and provide a benchmark against which one can assess Hungary’s developments and prospects.
The analysis in this section draws extensively on a recent study by Debelle and Faruqee (1996). Covering 1971–93, they provide an empirical investigation of the external current account and net foreign asset determinants for a large set of countries, including 21 industrial countries and 34 other, mainly developing, countries. Their panel data study, which does not include Hungary, uses cross-section analysis—where each country is treated as a single observation—as well as time series panel data—where also the behavior over time in the different countries is analyzed.
Intertemporal Saving-Investment Model
Based on the econometric relations estimated by Debelle and Faruqee (1996), the following equation can be used to derive a benchmark value for the medium-term equilibrium value of the external current account in Hungary:6
Here, the current account depends on three factors: the general government deficit-to-GDP ratio (DEF), the dependency ratio (DEP), and a country’s relative stage of development (DEVREL). The parameter estimates in equation (3) suggest that a decline in the general government deficit-to-GDP ratio by 1 percentage point would tend to raise the current account balance by 0.13 percentage point of GDP, an estimate that indicates a small deviation from Ricardian equivalence.7 One issue that needs to be solved in deriving estimates for the equilibrium current account is which value of the fiscal deficit should be used in equation (3). The solution used here is to use the maximum deficit that would be consistent with a “sustainable” path of the public-debt-to-GDP ratio, that is, with keeping the debt ratio constant at a level of about 60 percent of GDP (not far from the average level at which the Hungarian authorities intend to keep the public debt ratio in the next five years; see Ministry of Finance and others, 1997). This approach has the advantage of focusing on values of the fiscal and external deficits that are consistent with both the intertemporal budget constraint of the country and with the intertemporal budget constraint of the government.8
The relative dependency ratio of a country is also estimated to have a significant effect on the current account. As predicted by the model discussed in the previous section, countries with a relatively small dependency ratio tend to have stronger current account positions, with a 1 percentage point rise in the dependency ratio estimated to result in a 0.04 percentage point decline in the current account-to-GDP ratio. At present, Hungary’s dependency ratio is below the developing country average underlying the parameter estimates in equation (3)—reflecting the large proportion of underaged youth in developing countries—with some relative increase expected over the next five years.9
The empirical model of equation (3) also indicates that the relative stage of development has an important effect on the current account. In the estimates of Debelle and Faruqee (1996), this variable is proxied by a country’s relative income and relative capital stock position.10 This effect tends to strengthen the estimated current account balance in the case of Hungary, especially in view of its larger estimated capital stock compared with many developing countries.
Using the parameters in equation (3) yields a benchmark current account deficit for 1996 of 3¾ percent of GDP. This value is close to the actual deficit registered that year and somewhat larger than the one expected in 1997 (Figure 3.1). Thus, the authorities’ adjustment policies after March 1995 succeeded in lowering the external deficit to a level that can be regarded as broadly consistent with the structural features of the Hungarian economy, based on cross-country comparisons.
The above results should, of course, be taken with caution, for two reasons. First, the state of the art of empirical work in this area does not generate econometric estimates that can be put into narrow confidence bands. Indeed, some other parameter estimates presented by Debelle and Faruqee (1996) would yield a somewhat lower benchmark current account deficit for Hungary (Krueger, 1996). Second, the model estimated by Debelle and Faruqee (1996) does not take into account some factors that may be important for Hungary, as well as for other transition economies. For example, the capital requirement of transition economies may be temporarily higher than in other economies, because of the need to rebuild the capital stock (Krueger, 1996). Another aspect concerns the composition of the financing of the current account deficit. As noted above, foreign direct investment may allow for sustaining larger external imbalances for longer periods of time, as they typically tend to be less volatile than other forms of external financing.11 Of course, this conclusion should not be drawn to the extreme of ignoring altogether the long-run implications of FDI, for example, for the service component of the external current account.12
The above-described econometric estimates also provide some indications of how the external current account should move in time, as the determinants of the current account in equation (3) are not fixed. Based on projections for these determinants,13 the benchmark external current account would be expected to decline over time (on average by about ¼ of a percentage point of GDP a year in the six years starting in 1996; as noted, the improvement expected in 1997 is larger than this amount). Such a decline reflects the relatively rapid increase in per capita income and the capital stock projected by the authorities (per capita income is expected to rise by almost 4 percent a year during 1997–2002). This would be only partially offset by a projected concurrent deterioration in the dependency ratio relative to developing countries.
Stock of Net Foreign Asset
Hungary’s net external debt position has improved rapidly since 1994. Net external debt is projected to be slightly above 29 percent of GDP at the end of 1997, against 45 percent of GDP at the end of 1994.14 Moreover, reflecting the intense activity of foreign exchange intervention since March 1995 (Chapter II), net public external debt has declined even more rapidly.
There has been a strong improvement also relative to other countries during this period. Cross country comparisons of external debt are fraught with statistical difficulties, as in many countries the information on external assets and liabilities of the private sector is incomplete. With this caveat, and based on available data in the World Economic Outlook database, Figure 3.1 shows the gross debt (net of official reserves)-to-GDP ratio in Hungary relative to a large group of emerging economies. In the early 1990s and through 1994, the external debt ratio in Hungary was about twice as large as in the emerging market group. This reflected not only the large external imbalances that had characterized the 1980s but also the fact that Hungary had continued to service its debt impeccably, while other countries in the group had undertaken debt renegotiations. Hungary’s external debt-to-GDP ratio dropped much faster than in the emerging market group during 1995–97, although at the end of 1997, it is projected to remain well above the average. This fact is, however, mitigated by several factors, including Hungary’s fairly small share of short-term debt (about 12 percent of total debt at the end of 1996). Furthermore, Hungary’s external position ranks better based on other relative debt indicators. In particular, reflecting the relative openness of its economy, Hungary’s ratio of external debt to exports had fallen well below the ratio for the emerging market group at the end of 1996 (and probably further below the average in 1997).15
What are the implications for the external debt-to-GDP ratio of the improvement in the external current account observed through 1997? In spite of the expected fall in privatization receipts from abroad after 1997, it is reasonable to assume that in the medium term, FDI could continue to offset external current account deficit of the order of magnitude of the one observed for 1997. In this case, external debt would stabilize in dollar terms and would fall in relation to GDP. Assuming an average growth rate of 5½ percent (see Chapter VI), the external debt-to-GDP ratio would drop to 22½ percent by 2002.
A further decline in the external debt ratio is consistent with the expected trends in Hungary’s fundamentals, based on parameter estimates in Debelle and Faruqee (1996). Their estimates suggest the following long-run determinants of a country’s net external debt position:
where net debt is determined by the ratio of government debt-to-GDP (GDEBT) as well as a country’s dependency ratio (DEP) and relative per capita income (YREP). For example, a decline in the government debt ratio of 10 percentage points is estimated to be associated with a reduction in net debt of 5.7 percent of GDP; this relationship between the stock of foreign debt and government debt corresponds to the earlier discussed relationship between the current account and government deficits.16
These estimates would imply that a further decline in Hungary’s external debt should be expected, as a result of changes in all three right-hand-side variables in equation (4): a sustained decline of the public-debt-to-GDP ratio, reflecting a gradual reduction in the budget deficit and robust GDP growth targeted by the authorities; the targeted narrowing in the income gap; and a (temporary) decline in Hungary’s dependency ratio relative to industrial countries.
References
Artus, Jacques R., 1978, “Methods of Assessing the Long-Run Equilibrium Value of an Exchange Rate,” Journal of International Economics, Vol. 8, pp. 277–99.
Debelle, Guy, and Hamid Faruqee, 1996, “What Determines the Current Account? A Cross-Sectional and Panel Approach,” IMF Working Paper No. 96/58 (Washington: International Monetary Fund).
Edwards, Sebastian, 1989, Real Exchange Rates, Devaluation, and Adjustment: Exchange Rate Policy in Developing Countries (Cambridge, Massachusetts: MIT Press).
Frankel, Jeffrey A., and Andrew K. Rose, 1996, “Currency Crashes in Emerging Markets: An Empirical Treatment,” Journal of International Economics, Vol. 41, pp. 351–66.
Frenkel, Jacob A., and Assaf Razin, 1996, Fiscal Policies and the World Economy (Cambridge, Massachusetts: MIT Press, 3rd ed.).
Harberger, Arnold C., 1950, “Currency Depreciation, Income and the Balance of Trade,” Journal of Political Economy, Vol. 58 (February), pp. 47–60.
International Monetary Fund, 1997, World Economic Outlook (Washington: International Monetary Fund).
Isard, Peter, 1995, Exchange Rate Economics (Cambridge, England; New York, N.Y.: Cambridge University Press).
Krueger, Thomas, 1996, “The External Current Account and Net Foreign Assets in Hungary: Longer-Run Equilibrium Perspectives,” in Hungary—Selected Issues, IMF Staff Country Report No. 96/109 (Washington: International Monetary Fund).
Laursen, Svend, and Lloyd A. Metzler, 1950, “Flexible Exchange Rates and the Theory of Employment,” Review of Economics and Statistics, Vol. 32, pp. 281–99.
Lipschitz, Leslie, and Donogh McDonald, 1992, “Real Exchange Rates and Competitiveness: A Clarification of Concepts and Some Measurement for Europe,” Empirica, Vol. 19, pp. 37–69.
Lucas, Robert E. Jr., 1981, Studies in Business-Cycle Theory (Cambridge, Massachusetts: MIT Press).
Manzocchi, Stefano, 1997, “External Finance and Foreign Debt in Central and Eastern European Countries,” IMF Working Paper No. 97/134 (Washington: International Monetary Fund).
Masson, Paul R., Tamim Bayoumi, and Hossein Samiei, 1995, “International Evidence on the Determinants of Private Saving,” IMF Working Paper No. 95/51 (Washington: International Monetary Fund).
Meade, James E., 1951, The Theory of International Economic Policy: The Balance of Payments, Vol. 1 (London; New York: Oxford University Press).
Ministry of Finance and Directorate-General for Economic and Financial Affairs of the European Commission, June 1997, Joint Assessment of Hungary’s Medium-Term Economic Policy Priorities (Budapest).
Ostry, Jonathan D., 1988, “The Balance of Trade, Terms of Trade, and Real Exchange Rate,” Staff Papers, International Monetary Fund, Vol. 35 (December), pp. 541–73.
Poterba, James M. (ed.), 1994, International Comparisons of Household Saving, NBER Report Series (Chicago: University of Chicago).
Rogoff, Kenneth, 1996, “The Purchasing Power Parity Puzzle,” Journal of Economic Literature, Vol. 34, No. 2 (June), pp. 647–68.
Svensson, Lars E. O., and Assaf Razin, 1983, “The Terms of Trade and the Current Account: The Harberger-Laursen-Metzler Effect,” Journal of Political Economy, Vol. 91 (February), pp. 91–125.
Williamson, John, 1985, The Exchange Rate System (Washington: Institute for International Economics, 2nd ed.).
As discussed in Chapters II and VI, these regulations forced many firms, including those in the export sector, to close down.
Recall the national accounts identity that relates national income (Y) to the different demand components: Y = C + I + G + X –M. Equation (1) is derived directly from this identity, defining the government deficit (DEF)—on a national accounts basis—as DEF = G-T and private sector saving as SP = (Y - C - T), where T denotes taxes net of transfers. X–M includes (excludes) net factor payments if Y is based on gross national (gross domestic) income.
However, to the extent that temporary shocks affect a country’s net foreign asset position—for example, a temporary deteri-oration in the current account raises the external debt level and increases the future financing need—temporary shocks may also have longer-run implications for saving and investment and thereby the current account.
Cross-section and cross-country evidence on the importance of these factors is mixed; see, for example, Poterba (1994), but also Masson, Bayoumi, and Samiei (1995).
This role of the real exchange rate is a cornerstone of the macroeconomic balance approach, which goes back at least to Laursen and Metzler (1950), Harberger (1950), and Meade (1951). More recent analyses of the effect of real exchange rates on saving include Svensson and Razin (1983).
Debelle and Faruqee (1996) report several estimates for industrial as well as for developing countries. Equation (3) is based on one of their estimates for developing countries (see column (3) of Table A2 in their paper); using other specifications yields in most cases quite similar results for Hungary’s benchmark current account level; also see Krueger (1996).
Other specifications in Debelle and Faruqee (1996) yield sometimes larger deviations from Ricardian equivalence. Also, note that, in contrast to the fiscal deficit, government expenditures were not found to be a statistically significant determinant of the current account.
This yields a fiscal deficit of about 4 percent of GDP, after adjusting for inflation. Note that the operational deficit implicit in the Hungarian authorities’ medium-term scenario is lower, as the authorities target a decline in the public debt ratio. Thus, using the actual fiscal targets would yield lower values for the benchmark current account deficit. Finally, note that the inclusion in equation (3) of the fiscal deficit by no means implies that any increase in the fiscal deficit would raise the level of the sustainable normal external deficit. A fiscal deficit level that was inconsistent with the government’s intertemporal budget constraint could lead to an unsustainable deterioration of the external accounts.
The dependency ratio is defined here as the sum of the number of people under 19 and above 65, divided by the rest of the population. This understates the ratio in cases where the mandatory retirement age is below 65 or a portion of the working age population is not employed. Particularly in comparison with some industrial countries, these factors may be relatively important in Hungary, although the recently introduced pension reform could be expected to alleviate some of these effects; see also the discussion in Chapter V. It should also be noted that the dependency ratio is only a rough approximation of the intertemporal demographic factors suggested by theoretical economic models, as discussed earlier.
Their specification included level and square terms for both variables. When only the relative income level is included in the regression, the coefficient a is positive, as expected in the theoretical model of the previous section.
For example, results reported in Frankel and Rose (1996) suggest that the share of FDI in total debt tends to lower the probability of a currency crash in emerging markets.
FDI, like any other external liability, entails profit remittances. However, in most countries, reinvested profits from FDI are not fully captured in the external current account. This lowers the measured current account deficit in countries, like Hungary, that have net FDI liabilities (and it correspondingly underestimates FDI). Note that, as in other countries, the Hungarian official statistics include the cumulated sum of FDI in the country’s external liabilities.
All projections and targets for Hungary are based on Ministry of Finance and others (1997).
Net external debt would be about 3½ percentage points of GDP lower, excluding intercompany loans.
Recent empirical research (Manzocchi, 1997) suggests that the export-to-GDP ratio is a key factor in determining a country’s access to international capital markets. Another traditional indicator of a country’s external position is the debt service ratio. However, in countries with an active debt management policy, such as Hungary, this indicator has to be interpreted with caution, as it is highly sensitive to changes in the schedule of debt amortization related, inter alia, to early repayments. In particular, the debt service ratio for Hungary has remained high in 1995 and 1996 because of early repayments or refinancing of outstanding loans.
Note that equation (4) is estimated by Debelle and Faruqee (1996) for all net external liabilities, including those arising from FDI. At the end of 1996, the stock of foreign direct investment in Hungary was estimated at about 30 percent of GDP. However, as discussed above, there are good reasons for a separate treatment of FDI stocks.
