A. Introduction

1. The EU underwent a remarkable expansion in 2004, admitting 10 new member states (NMS). Many of these economies have, and continue to undergo, significant changes in their economic structure and relations. Production patterns have often undertaken fundamental realignments, reflecting marked shifts in relative prices of goods and factors of production. International trade patterns have also been sizably altered, with most NMS finding their major trading partners are now other EU members.

2. Although financial markets in some NMS were already relatively well developed, those in most other economies have experienced noteworthy increases in sophistication. While banking systems in Cyprus and Malta were comparatively modern well before EU accession, financial sector evolution in other NMS has been remarkable. Credit has generally grown rapidly, an expected development related to and supporting real economic convergence. In addition to mobilizing domestic savings, financial markets have also facilitated the use of foreign capital to finance high investment ratios. Other domestic reforms, including restructuring of national pension systems, has boosted nonbank financial activity, which for portfolio diversification reasons, has in some cases resulted in increased holdings of foreign assets. In a number of economies, significant foreign ownership of domestic banking and other financial institutions has improved productivity, increased financial sophistication, and spurred international financial integration.

3. This paper examines one aspect of these developments, namely those relating to the economies’ gross and net international investment positions (GIIPs and NIIPs, respectively). Official data on these economies’ GIIPs and NIIPs are limited, with most series beginning in the mid-1990s; figures for Cyprus are available only for 2002-03.² Nevertheless, they show a rich diversity of patterns, both across countries and over time. The influences in these positions are myriad. This paper documents these patterns and analyzes a number of influences that have affected them.

4. The paper is structured as follows. Section B examines a number of stylized facts relating to gross and net IIPs, and some basic influences on their differences across NMS and over time. Section C provides a disaggregated examination of the various determinants of changes in NIIPs of NMS, including the current account, economic growth, and capital gains and losses. Section D looks more closely at the relative influence that component parts have on gross stocks (both by institutional sector and financial instrument). Section E concludes.

B. Gross and Net International Investment Positions and their Influences

5. International financial integration has occurred in all new member states, but its degree varies widely. Malta and Cyprus have significant offshore or nonresident-based banking activity, respectively, and their shares of total external assets and liabilities are 400-600 percent of GDP. The shares of GIIPs of other NMS are more moderate, ranging from over 230 percent of GDP in Estonia, to less than 100 percent of GDP in Poland and Lithuania (Figure 1). Broadly speaking, the level of integration is higher in relatively smaller, Baltic economies (except Lithuania) than in larger, central European NMS (except Hungary). Excluding Malta, the increase in GIIPs since the mid-1990s has ranged from 160 percentage points of GDP in Estonia to only 15 percentage points in Poland. On average, the GIIP share (unweighted, excluding Cyprus and Malta) has increased from 102 percent of GDP in 1997 to 136 percent in 2003. The standard deviation has increased as well over this period, from 28 percentage points of GDP to over 32 percentage points, reflecting some divergence in the degree to which financial integration has occurred.

View Full Size

EU New Member States: Financial and Trade Integration, 1994-2004

Citation: IMF Staff Country Reports 2005, 395; 10.5089/9781451812503.002.A001

Source: IFS.

• Download Figure
• Download figure as PowerPoint slide

6. various factors have been posited as influencing gross international investment positions. Among those suggested by Lane (2000) as potentially influencing GIIPs are trade openness, per capita income and overall economic size. Trade openness is generally thought to be positively related to higher GIIPs because higher trade may expose an economy more to volatility, which increases the desire for consumption smoothing through larger holdings of foreign assets. Trade openness also generates its own parallel financial flows which agents may wish to hedge against. Alternately, foreign direct investment may result in production not only for the domestic market, but also for export. Lane and Milesi-Feretti (2000) suggest that trade openness generally reflects a liberal policy environment that generally stimulates asset trade. Higher per capita incomes may also be associated with higher GIIPs if the formation of international financial linkages involve fixed setup costs, and/or if FDI tends to flow to economies with high levels of human capital. An economy’s size may also influence its GIIP, if small economies “free ride” off of existing deep financial markets in neighboring countries. On the other hand, larger economies may be more attractive to international investors if there are fixed costs in acquiring information about local investment conditions, or if it presents a larger local market for FDI.

7. Greater economic openness among new member states is associated with larger gross international investment positions. Data limitations preclude a detailed statistical analysis for the NMS to test these possibilities. However, the Spearman rank correlation of the level of goods and services “openness,” measured by the ratio of exports and imports of goods and services to GDP, and international financial integration (the GIIP share) for 2003-04 is 0.38, (on a minus one to plus one scale), suggesting a moderate positive correlation (Table 1).³ Estonia and Malta both exhibit high trade and financial integration, while Hungary and Slovakia display middling integration, while relatively large Poland has comparatively low levels of external trade and financial linkages. In contrast to this pattern, Cyprus and Latvia both have large gross financial stocks (from offshore banking and nonresident deposits in the domestic banking system, respectively), while their trade shares are comparatively low. Excluding Cyprus and Malta, the result of their outlier status given their disproportionate banking activities, the coefficient rises to 0.47. Thus, broadly speaking, economies with greater international integration of markets for goods and services appear to also have greater financial integration. These correlations are broadly unchanged when examining foreign assets and foreign liabilities separately.⁴

Table 1.

EU New Member States: GIIPs and their Influences, 2003-04

Sources: IFS; and staff calculations.

Table 1.

EU New Member States: GIIPs and their Influences, 2003-04

		Foreign	Foreign
		GIIP	Assets	Liabilities	NIIP
		Spearman Rank Correlation Coefficients
Trade openness
	All NMS	0.38	0.38	0.41	−0.01
	Excluding Cyprus and Malta	0.47	0.52	0.54	0.05
Per capita income
	All NMS	0.39	0.51	0.45	0.71
	Excluding Cyprus and Malta	−0.01	0.15	0.11	0.51
Economic size
	All NMS	−0.72	−0.87	−0.67	−0.36
	Excluding Cyprus and Malta	−0.46	−0.76	−0.36	0.25

Sources: IFS; and staff calculations.

View Table

Table 1.

EU New Member States: GIIPs and their Influences, 2003-04

		Foreign	Foreign
		GIIP	Assets	Liabilities	NIIP
		Spearman Rank Correlation Coefficients
Trade openness
	All NMS	0.38	0.38	0.41	−0.01
	Excluding Cyprus and Malta	0.47	0.52	0.54	0.05
Per capita income
	All NMS	0.39	0.51	0.45	0.71
	Excluding Cyprus and Malta	−0.01	0.15	0.11	0.51
Economic size
	All NMS	−0.72	−0.87	−0.67	−0.36
	Excluding Cyprus and Malta	−0.46	−0.76	−0.36	0.25

Sources: IFS; and staff calculations.

8. However, the influence of levels of economic development on gross international investment positions is mixed, and economic size appears to be negatively related. The rank correlation coefficient for GIIPs and economic development (proxied by Eurostat’s purchasing power parity adjusted per capita income for 2003-04) was 0.39, suggesting a weak positive correlation. However, excluding Cyprus and Malta, the first and third richest NMS and with massive GIIPs, the correlation is eliminated. Slovenia and Poland, the second and ninth richest, both have relatively low GIIPs, while the Czech Republic and Estonia (the fourth and seventh richest, respectively), both had relatively high GIIPs. Economic size itself (using Eurostat’s purchasing power parity adjusted GDPs) was moderately to strongly negatively correlated with GIIPs. A number of small economies (Cyprus, Malta, Estonia, Latvia) all have large GIIPs, while Poland’s gross external position is relatively low. Again, similar correlations are obtained when looking at foreign assets and liabilities separately.⁵

9. Increased foreign goods and services integration within an economy over time does not appear to have been systematically associated with increased foreign financial integration among the new member states. Over the last decade, increases in financial integration has far exceeded the increase in trade integration. The former development reflects a growing sophistication of capital market transactions, while the latter masks a dramatic shift in trade patterns to market-based, and increasingly EU-centered, trade flows (although much of this occurred in the early 1990s, a period for which external financial stock data are not available). While GIIPs increased by 90 percentage points of GDP on (an unweighted) average since the mid-1990s (and by 60 percentage points excluding Cyprus and Malta), the trade ratio increased by only 12 percentage points over the same period (by 21 percentage points, again excluding Cyprus and Malta). As already discussed, one might expect a positive correlation within those economies experiencing a large increase in international integration of both financial sectors and goods and services markets. However, somewhat surprisingly, this has not been the case for these economies over the last decade. In fact, the rank correlation coefficient between those economies experiencing the largest increase in financial stocks and those with the largest increase in goods and services flows is in fact strongly negative (-0.78 excluding Cyprus, for whom only one observation is available, and -0.69 excluding Cyprus and Malta). In general, the Baltic economies have experienced disproportionately large increases in financial integration relative to trade integration, while the Central European economies have experienced the opposite. This may reflect the widespread foreign ownership of the banking systems in the former, resulting in large foreign direct investment and other capital flows. Thus, it would appear that factors other than changes in economies’ gross goods and services flows have had more important influences on developments in external financial stocks.

10. with few exceptions, the new member states’ net international investment positions are negative and have declined in recent years. While foreign banking-centered Malta’s and Cyprus’ NIIP positions are positive or close to balance, those for the other NMS have been negative in recent years (Figure 2). Moreover, all NIIP/GDP positions have declined over the past decade, on average by 32 percentage points of GDP (excluding Cyprus), ranging from only 5 percentage points in Malta to an notable 78 percentage points in Estonia. These developments are not unexpected in light of the investment needs associated with real convergence and households’ attempts in rapidly growing economies with newly available credit opportunities to "smooth" consumption through borrowing in anticipation of higher future incomes ( see Stavrev, 2003). Net saving shortfalls for both investment and increased consumption might be met through foreign capital inflows, thereby increasing gross external liabilities.

View Full Size

EU New Member States: Financial Integration and Net International Positions, 1994-2004

Citation: IMF Staff Country Reports 2005, 395; 10.5089/9781451812503.002.A001

Source: IFS.

• Download Figure
• Download figure as PowerPoint slide

11. while the level of development appears to be strongly correlated with the level of net external claims, the latter appears to be little correlated with trade openness and economic size. Even excluding relatively rich Cyprus and Malta, the NIIP/per capita income correlation still exceeds 0.5 However, relatively rich Slovenia and relatively poor Estonia are both ardent traders; thus, there is almost no correlation between NIIPs and openness.⁶. As regards economic size, tiny Cyprus and Malta are net creditors, while tiny Estonia and Latvia have large negative NIIPs. Excluding the island economies, the correlation becomes weakly positive.⁷

12. Levels and changes in gross international investment positions also appear to be related to levels and changes in net positions. Simply from an accounting standpoint, it is possible that large swings in net positions, possibly the result of cumulative external deficits for example, may overwhelm underlying trend increases in international financial integration, and thus be correlated with levels and changes in gross positions. However, the Spearman rank correlation between levels of GIIPs and the absolute values of NIIP positions is only slightly positive, at 0.27, for all NMS, but rises to a strong 0.64 when Cyprus and Malta are excluded. In addition, the rank correlation for changes is about 0.35 for changes in gross positions and the absolute value of net positions (regardless of including or excluding Malta and Cyprus). Thus, it would appear that despite increasing financial integration of both assets and liabilities, large levels and changes in net positions are associated with large levels and changes in gross positions.

13. An alternative measure of gross/net foreign asset linkages supports these patterns as well. ( Obstfeld (2004) and Lane and Milesi-Ferretti (2005a) have suggested a modified Grubel-Lloyd (G-L) index relating the magnitude of the absolute value of net international positions to the size of gross asset stocks, given by:

1 - |A - L|/(A + L)

where A is gross external assets and L is gross external liabilities. The index varies between 1 when the net position is zero—consistent with a balanced NIIP position and only gross cross-border trade taking place—to 0 if changes in NIIP reflect only net financing changes. Thus, the index depends both on the net cumulative stock of external imbalances and gross asset integration. As seen in Figure 3, there is a moderately positive correlation over the period considered among the NMS between the levels of G-L indices and gross asset integration.⁸ The two economies with sizable offshore/foreign banking activities, Cyprus and Malta, exhibited the highest indices; this also reflected for Cyprus its almost balanced NIIP position. Other economies with large GIIPs, Estonia, Latvia and Hungary, had lower G-L indices, reflecting sizable negative NIIP positions. In contrast, somewhat less financially integrated economies as measured by GIIPs (Slovenia, the Czech Republic, and Slovakia) had relatively higher G-L indices because of more balanced NIIP positions. Lithuania and Poland exhibited both relatively less integrated capital markets and middling NIIPs. Also, with the exception of Cyprus and Malta, G-L indices have generally declined over the past decade, suggesting that gross capital flows may increasingly reflect the financing of current account positions, rather than growing financial sector integration.

View Full Size

EU New Member States: Financial Integration and Grubel-Lloyd Indices, 1994-2004

Citation: IMF Staff Country Reports 2005, 395; 10.5089/9781451812503.002.A001

Source: IFS; and Fund staff calculations.

• Download Figure
• Download figure as PowerPoint slide

C. Disaggregated Influences on Net International Investment Position Developments

14. while net international investment positions have broadly declined among new member states, the reasons for this are myriad. Although declining NIIPs are generally a concomitant outcome of real economic convergence, the diversity of patterns experienced among new member states is notable. Table 2 documents the change in NIIPs generally since the mid-1990s. As discussed earlier, the variance is vast, from only 5 to almost 80 percentage points of GDP. These changes can be disaggregated into their major components through the following accounting identities:

B_t − B_t-1 = CA_t + KG_t + E_t

Table 2.

EU New Member States:Changes in NIIPs and Component Contributions, 1994-2004

(percent of GDP)

Sources: IFS; national sources; and staff calculations.

^1/2002

^2/1996

^3/1997

Table 2.

EU New Member States:Changes in NIIPs and Component Contributions, 1994-2004

(percent of GDP)

					Cumulative current account and growth factors				Cumulative other factors
	Period covered	Initial NIIP position		Change in NIIP 1994	Total	Cumulative trade and transfer balance	Cumulative factor payments	Growth effect	Total	Capital account and errors & omissions	Capital gains/losses
Cyprus	2002-2003	9.2	1/	−8.1	−3.9	−0.4	−2.9	−0.6	−4.1	0.5	−4.6
Czech Republic	1994-2004	5.7		−35.6	−44.7	−20.3	−27.6	3.2	9.0	1.5	7.5
Estonia	1996-2004	−14.0	2/	−78.4	−33.6	−39.8	−31.6	37.8	−44.8	1.2	−46.0
Hungary	1997-2004	−66.8	3/	−17.0	2.9	−14.5	−40.1	57.5	−19.8	1.0	−20.9
Latvia	1996-2004	−6.3	2/	−48.4	−42.3	−64.7	0.7	21.7	−6.1	5.0	−11.1
Lithuania	1994-2004	−2.8		−32.2	−62.8	−63.6	−17.5	18.4	30.6	15.8	14.8
Malta	1994-2003	48.4		−5.2	−76.2	−60.6	−0.6	−15.0	71.0	24.7	46.3
Poland	1994-2003	−32.0		−10.3	−0.2	−22.0	−8.6	30.4	−10.1	3.5	−13.6
Slovakia	1994-2004	14.4		−34.6	−36.9	−39.5	−9.2	11.7	8.0	4.7	3.2
Slovenia	1994-2003	5.0		−20.4	−2.2	−6.9	1.7	3.0	−18.2	−1.8	−16.4

Sources: IFS; national sources; and staff calculations.

^1/2002

^2/1996

^3/1997

View Table

Table 2.

EU New Member States:Changes in NIIPs and Component Contributions, 1994-2004

(percent of GDP)

					Cumulative current account and growth factors				Cumulative other factors
	Period covered	Initial NIIP position		Change in NIIP 1994	Total	Cumulative trade and transfer balance	Cumulative factor payments	Growth effect	Total	Capital account and errors & omissions	Capital gains/losses
Cyprus	2002-2003	9.2	1/	−8.1	−3.9	−0.4	−2.9	−0.6	−4.1	0.5	−4.6
Czech Republic	1994-2004	5.7		−35.6	−44.7	−20.3	−27.6	3.2	9.0	1.5	7.5
Estonia	1996-2004	−14.0	2/	−78.4	−33.6	−39.8	−31.6	37.8	−44.8	1.2	−46.0
Hungary	1997-2004	−66.8	3/	−17.0	2.9	−14.5	−40.1	57.5	−19.8	1.0	−20.9
Latvia	1996-2004	−6.3	2/	−48.4	−42.3	−64.7	0.7	21.7	−6.1	5.0	−11.1
Lithuania	1994-2004	−2.8		−32.2	−62.8	−63.6	−17.5	18.4	30.6	15.8	14.8
Malta	1994-2003	48.4		−5.2	−76.2	−60.6	−0.6	−15.0	71.0	24.7	46.3
Poland	1994-2003	−32.0		−10.3	−0.2	−22.0	−8.6	30.4	−10.1	3.5	−13.6
Slovakia	1994-2004	14.4		−34.6	−36.9	−39.5	−9.2	11.7	8.0	4.7	3.2
Slovenia	1994-2003	5.0		−20.4	−2.2	−6.9	1.7	3.0	−18.2	−1.8	−16.4

Sources: IFS; national sources; and staff calculations.

^1/2002

^2/1996

^3/1997

where B is the net international investment position, CA the current account balance, KG the capital gains/losses on net foreign assets (defined as the difference between the value of outstanding stocks between two contiguous years and the corresponding balance of payments flow, all valued in national currency), and E includes capital account transfers and net errors and omissions.⁹ The current account component can be separated into its goods and services and unrequited transfers portions, BGST, and the balance on factor payments, given by i^A_tA_t-1–i^L_tL_t-1 where A and L are gross external assets and liabilities, respectively, and the i’s are the corresponding nominal factor payments implied by balance of payments flows. The above expression can be restated as shares of GDP, using lower-cased letters, as follows:

b_t—b_t-1 = bgst_t + (i^A_tA_t-1—i^L_tL_t-1 + KG_t)/GDP_t—b_t-1*(g_t + π_t)/[(1 + g_t)(1 + π_t)] + ε_t

where g is the growth rate of real GDP, n is the inflation rate, and s includes the share of errors and omissions and capital transfers in GDP. This is similar to the standard dynamic equation for changes in the domestic public debt ratio, but with a few differences. First, given that “b” is a net external investment ratio, its evolution depends on the differential between returns earned on external assets and paid on external liabilities, both measured in domestic currency terms. Second, this includes capital gains and losses, which are potentially significant, given exchange rate changes and potentially stock market valuation effects for foreign direct investment and portfolio equity positions. Third, the final term capturing errors and omissions could also be sizable given the survey-based nature of both debt stock and flow data.

15. Current account flows comprise significant shares of the changes in net international investment positions across new member states, while economic growth has moderated these movements. Cumulative trade deficits have exceeded 60 percentage points of GDP over the last decade in Latvia, Lithuania, and Malta, and by almost 40 percentage points in Estonia and Slovakia. Cumulative factor payments have also averaged some 30-40 percentage points of GDP in the Czech Republic (somewhat surprisingly, given its moderate NIIP), and in Estonia and Hungary. In fact, cumulative current account flows more than accounted for the entire change in NIIP positions in all NMS except Cyprus, Estonia and Slovenia. However, significant increases in nominal GDP has worked to moderate changes in NIIP shares in many economies, especially in Hungary, Estonia, and Poland. All told, these traditional drivers of change in NIIPs account for most of the observed developments in Cyprus, the Czech Republic, Latvia, Poland and Slovakia.

16. However, the influence of other factors have also been sizable for a number of economies. Cumulative capital transfers and errors and omissions have had significant positive influences on Lithuania’s and Malta’s NIIPs (Table 3). These have been almost entirely accounted for by errors and omissions, and uncategorized capital inflows have exceeded 5 percent of GDP in Malta in recent years. Capital gains have also been exceptionally beneficial to Malta’s external position, but have significantly worsened Estonia’s NIIP, and to a lesser degree positions in Hungary and Slovenia. Portfolio investments have yielded large capital gains in Malta, while increased market values of negative net foreign direct investment and net portfolio equity positions have sizably worsened Estonian and Hungarian NIIPs.¹⁰ This is a somewhat paradoxical result in that improved domestic economic conditions, evidenced by higher market valuations of largely foreign-owned enterprises, resulted in a worsening of these economies’ NIIPs.¹¹

Table 3.

EU New Member States: Capital Transfers, Errors and Omissions, and Capital Gains and Losses, 1995-2004 (Percent of GDP)

Sources: IFS; and staff calculations.

Table 3.

EU New Member States: Capital Transfers, Errors and Omissions, and Capital Gains and Losses, 1995-2004 (Percent of GDP)

		1995	1996	1997	1998	1999	2000	2001	2002	2003	2004
Capital transfers/errors & omissions
Cyprus		−0.8	−0.4	−1.9	0.5	−0.4	−0.6	−0.3	−0.9	0.5	0.9
Czech Republic		1.1	−1.5	1.3	0.5	0.0	−0.5	0.8	0.4	0.7	−1.2
Estonia		0.2	−0.8	−0.5	0.1	−0.1	0.5	0.5	0.5	0.1	0.0
Hungary		3.0	2.1	−0.4	−0.2	−0.7	0.2	0.7	0.5	0.3	0.3
Latvia		…	…	1.6	1.7	0.7	−0.7	0.5	−0.6	0.9	0.8
Lithuania		3.9	0.9	2.0	2.5	−0.4	1.1	1.3	1.0	1.4	2.1
Malta		0.9	3.4	3.1	3.4	−1.8	1.0	7.4	2.0	5.3	5.7
Poland		−0.2	0.3	0.9	−0.3	1.3	0.2	1.0	−0.8	1.1	1.6
Slovakia		1.0	0.9	1.3	−1.2	0.7	0.7	0.9	1.7	0.4	0.2
Slovenia		−1.0	0.0	0.4	0.3	0.2	0.2	0.3	−1.6	−0.5	−1.3
Capital gains/losses
Cyprus		…	…	…	…	…	…	…	…	−4.6	…
Czech Republic		0.9	0.5	3.0	1.1	2.8	1.6	2.3	−0.9	0.6	−4.4
Estonia		…	…	12.5	3.0	−12.3	1.2	−2.6	0.2	−7.2	−15.8
	of which: net FDI and portfolio equity securities	…	…	12.1	4.8	−11.4	−0.3	−1.0	−3.2	−8.6	−16.4
Hungary		…	…	…	−3.0	−7.5	1.4	2.0	−0.1	−10.7	−3.0
	of which net FDI and portfolio equity securities	…	…	…	−0.7	−7.4	1.9	0.2	1.0	−9.6	−3.6
Latvia		…	0.3	1.8	−5.6	−0.3	0.6	−5.7	0.0	0.8	−2.7
Lithuania		−0.6	0.6	0.9	3.4	0.3	1.7	1.9	3.7	2.2	0.7
Malta		2.3	1.2	0.6	3.8	4.1	0.8	4.5	21.2	7.9	…
	of which: net portfolio securities	−6.2	0.5	−1.9	1.1	2.7	1.0	0.3	12.0	4.1	…
Poland		1.0	−2.5	−1.5	−2.3	−2.5	2.3	1.8	−3.2	−6.6	…
Slovakia		−0.8	−1.3	−0.4	−4.4	1.1	1.6	2.5	9.0	−4.0	−1.4
Slovenia		−0.1	−5.1	−0.5	−2.4	−2.8	−0.9	4.1	0.1	−8.9	…

Sources: IFS; and staff calculations.

View Table

Table 3.

EU New Member States: Capital Transfers, Errors and Omissions, and Capital Gains and Losses, 1995-2004 (Percent of GDP)

		1995	1996	1997	1998	1999	2000	2001	2002	2003	2004
Capital transfers/errors & omissions
Cyprus		−0.8	−0.4	−1.9	0.5	−0.4	−0.6	−0.3	−0.9	0.5	0.9
Czech Republic		1.1	−1.5	1.3	0.5	0.0	−0.5	0.8	0.4	0.7	−1.2
Estonia		0.2	−0.8	−0.5	0.1	−0.1	0.5	0.5	0.5	0.1	0.0
Hungary		3.0	2.1	−0.4	−0.2	−0.7	0.2	0.7	0.5	0.3	0.3
Latvia		…	…	1.6	1.7	0.7	−0.7	0.5	−0.6	0.9	0.8
Lithuania		3.9	0.9	2.0	2.5	−0.4	1.1	1.3	1.0	1.4	2.1
Malta		0.9	3.4	3.1	3.4	−1.8	1.0	7.4	2.0	5.3	5.7
Poland		−0.2	0.3	0.9	−0.3	1.3	0.2	1.0	−0.8	1.1	1.6
Slovakia		1.0	0.9	1.3	−1.2	0.7	0.7	0.9	1.7	0.4	0.2
Slovenia		−1.0	0.0	0.4	0.3	0.2	0.2	0.3	−1.6	−0.5	−1.3
Capital gains/losses
Cyprus		…	…	…	…	…	…	…	…	−4.6	…
Czech Republic		0.9	0.5	3.0	1.1	2.8	1.6	2.3	−0.9	0.6	−4.4
Estonia		…	…	12.5	3.0	−12.3	1.2	−2.6	0.2	−7.2	−15.8
	of which: net FDI and portfolio equity securities	…	…	12.1	4.8	−11.4	−0.3	−1.0	−3.2	−8.6	−16.4
Hungary		…	…	…	−3.0	−7.5	1.4	2.0	−0.1	−10.7	−3.0
	of which net FDI and portfolio equity securities	…	…	…	−0.7	−7.4	1.9	0.2	1.0	−9.6	−3.6
Latvia		…	0.3	1.8	−5.6	−0.3	0.6	−5.7	0.0	0.8	−2.7
Lithuania		−0.6	0.6	0.9	3.4	0.3	1.7	1.9	3.7	2.2	0.7
Malta		2.3	1.2	0.6	3.8	4.1	0.8	4.5	21.2	7.9	…
	of which: net portfolio securities	−6.2	0.5	−1.9	1.1	2.7	1.0	0.3	12.0	4.1	…
Poland		1.0	−2.5	−1.5	−2.3	−2.5	2.3	1.8	−3.2	−6.6	…
Slovakia		−0.8	−1.3	−0.4	−4.4	1.1	1.6	2.5	9.0	−4.0	−1.4
Slovenia		−0.1	−5.1	−0.5	−2.4	−2.8	−0.9	4.1	0.1	−8.9	…

Sources: IFS; and staff calculations.

D. Institutional Sector and Financial Instrument Influences on Gross and Net International Investment Positions

17. In examining the economic and institutional composition of gross international investment positions, the differences between those in Cyprus and Malta, and in the other new member states is stark. In the former two economies “other” assets and liabilities dominate. These largely comprise loans and currency and deposits by the banking system (Figures 4 and 5). While this pattern is taking on increasing importance in Latvia as well, other components, including inward foreign direct investment and “other” sector liabilities (largely borrowing by the nonbank private sector) remain relatively prominent.

View Full Size

View Full Size

EU New Member States: Foreign Assets and Liabilities by Economic Component, 1993-2004

(Percent of GDP)

Citation: IMF Staff Country Reports 2005, 395; 10.5089/9781451812503.002.A001

Source: IFS.

• Download Figure
• Download Figure
• Download figure as PowerPoint slide

View Full Size

View Full Size

EU New Member States: Foreign Assets and Liabilities by Institutional Sector, 1993-2004

(Percent of GDP)

Citation: IMF Staff Country Reports 2005, 395; 10.5089/9781451812503.002.A001

Source: IFS.

• Download Figure
• Download Figure
• Download figure as PowerPoint slide

18. The composition among other new member states reflects, inter alia, competing influences of financial structure, government indebtedness and privatization experiences. Foreign direct investment assets and liabilities vary widely among NMS. FDI assets are all but negligible except in Estonia, where claims on banking institutions in other Baltic states predominate. FDI liabilities are also most prominent in Estonia, exceeding 80 percent of GDP in 2004, and are also sizable—at about 50 percent of GDP—in the Czech Republic and Hungary. Hungary’s persistently large share reflects its initial decision to privatize state-owned assets to strategic foreign investors as a means to import technical and managerial know how as well as foreign financing. Estonia’s large increase, in addition to privatization, reflects sizable greenfield investments. As noted before, capital gains in the market value of inward FDI has been significant in Estonia and Hungary. The influence of fiscal policy is also reflected in general government assets and liabilities, with Hungary’s persistently large deficits resulting in large negative gross external liabilities (predominantly bonds), while Estonia’s persistent surpluses have been generally held as foreign debt securities. Monetary authority assets are notably sizable in Slovakia and Slovenia (at about 30 percent of GDP), reflecting large foreign exchange reserves; they are also a very large share of GDP in Malta, but a much smaller share of total external assets. Bank’s external assets are relatively large in Latvia and the Czech Republic, both as a share of GDP and in relation to total assets. In fact, the banks’ net external assets are positive only in the Czech Republic and Malta, both of which have large holdings of foreign portfolio debt. Aside from Latvia, banking sector liabilities have grown rapidly in Estonia, where highly ranked domestically-owned offspring of Nordic parent institutions have issued securities internationally to fund domestic credit demand. The “other” sectors’ external assets are relatively inconsequential except in Slovenia and Estonia, where in the latter case they have been rapidly growing in recent years. The former reflects large overseas loans, while Estonia’s growing holdings reflect asset claims by nonbank financial intermediaries following the widespread adoption of a funded second pension pillar by the overwhelming majority of the working population. The other sectors’ external liabilities have been generally larger than external assets across NMS, comprising largely direct bank borrowing and loans from abroad, foreign holdings of domestic portfolio equities, and trade credits.

19. A mixed pattern also emerges when examining the influences of openness, development and economic size on international investment components. Trade openness tends to be positively correlated with most investment components, strongly so in many cases. It was also associated positively with a preference for equity over debt instruments, especially when banking-centered Cyprus and Malta were excluded. Both of these findings are in accord with those of Lane and Milesi-Ferretti (2000). Economic development was also positively related to international asset accumulation, again in concert with Lane and Milesi-Ferretti’s findings, although less so when relatively prosperous Cyprus and Malta were excluded. However, the linkages with international liabilities was more mixed, reflecting relatively poor Latvia’s and Estonia’s large obligations, in contrast to wealthier Slovenia’s lower liabilities. Economic size was often negatively associated with various investment components, reflecting Poland’s and the Czech Republic’s small international positions in contrast to large Baltic and Mediterranean gross positions.

E. Summary and Conclusions

20. While international financial integration has occurred in all new member states, the degree to which this has occurred has varied widely. Greater financial integration is associated with greater economic openness, but the influence of levels of economic development is mixed, and economic size appears to have been negatively related. Most NMS have experienced a worsening in their NIIPs, the bulk of which is accounted for by current account flows, although errors and omissions and capital gains were important in a number of economies. Finally, asset and liability composition result from a myriad complex of influences, both economic and political.

21. This topic provides a number of interesting areas for further investigation. Among these, the following issues appear to be most interesting. Once more observations become available, more sophisticated statistical analysis could be brought to bear. Relative rates of returns in various assets and liabilities could be analyzed, including both income payments as well as capital gains and losses. Using recently available data on portfolio investment by countries of both origin and destination, combined with similar data on foreign direct investment, one can examine in more detail geographic correlations between trade and investment patterns.

Table 4.

EU New Member States: International Investment Components and their Influences, 2003-04

Sources: IFS; and staff calculations.

Table 4.

EU New Member States: International Investment Components and their Influences, 2003-04

		FDI Assets	FDI Liabilities	Net FDI	Port. Eq. Assets	Port. Eq. Liabilities	Net Port. Equities	Debt Assets	Debt Liabilities	Net Debt	Eq./Debt Liab.
		Spearman Rank Correlation Coefficients
Trade openness
	All NMS	0.29	0.65	−0.66	0.58	0.26	−0.11	0.29	0.14	0.38	0.40
	Excluding Cyprus and Malta	0.61	0.66	−0.59	0.82	0.33	−0.16	0.35	0.10	0.54	0.71
Per capita
	All NMS	0.72	0.26	−0.19	0.51	−0.05	0.47	0.46	0.29	0.54	−0.21
	Excluding Cyprus and Malta	0.58	0.11	−0.06	0.25	−0.06	0.06	0.08	−0.24	0.27	0.44
Economic size
	All NMS	−0.66	−0.44	0.33	−0.61	0.16	−0.49	−0.80	−0.83	−0.53	0.62
	Excluding Cyprus and Malta	−0.50	−0.13	0.06	−0.24	0.11	−0.01	−0.62	−0.67	−0.08	0.25

Sources: IFS; and staff calculations.

View Table

Table 4.

EU New Member States: International Investment Components and their Influences, 2003-04

		FDI Assets	FDI Liabilities	Net FDI	Port. Eq. Assets	Port. Eq. Liabilities	Net Port. Equities	Debt Assets	Debt Liabilities	Net Debt	Eq./Debt Liab.
		Spearman Rank Correlation Coefficients
Trade openness
	All NMS	0.29	0.65	−0.66	0.58	0.26	−0.11	0.29	0.14	0.38	0.40
	Excluding Cyprus and Malta	0.61	0.66	−0.59	0.82	0.33	−0.16	0.35	0.10	0.54	0.71
Per capita
	All NMS	0.72	0.26	−0.19	0.51	−0.05	0.47	0.46	0.29	0.54	−0.21
	Excluding Cyprus and Malta	0.58	0.11	−0.06	0.25	−0.06	0.06	0.08	−0.24	0.27	0.44
Economic size
	All NMS	−0.66	−0.44	0.33	−0.61	0.16	−0.49	−0.80	−0.83	−0.53	0.62
	Excluding Cyprus and Malta	−0.50	−0.13	0.06	−0.24	0.11	−0.01	−0.62	−0.67	−0.08	0.25

Sources: IFS; and staff calculations.