II Trade and Financial Effects of EMU on Selected Transition and Mediterranean Countries
  • 1 https://isni.org/isni/0000000404811396, International Monetary Fund
  • | 2 https://isni.org/isni/0000000404811396, International Monetary Fund


Robert A. Feldman and Heliodoro Temprano-Arroyo

Robert A. Feldman and Heliodoro Temprano-Arroyo

Focusing on trade and financial linkages, this chapter analyzes several aspects of how the European Economic and Monetary Union (EMU) might affect the transition countries of Central and Eastern Europe (CEE) and a group of selected Mediterranean countries.1 The first part of the chapter attempts to gauge the extent to which EMU—through its impact on the economies of participating countries—might affect trade flows and thereby economic activity in the countries under analysis. The second part looks into financial linkages. The linkages considered largely operate through changes in capital flows that might accompany EMU and reflect several important channels of influence such as changes in interest rates in the euro area, portfolio diversification effects, and the possible diversion of foreign direct investment (FDI) flows. We also examine how possible changes in euro-area interest and exchange rates might affect the external debt services of the countries under study, and the potential effect of EMU-related developments on the structure and functioning of their financial systems. The final section provides a summary and the main conclusions of the study.

Linkages Through External Trade

External trade linkages are potentially an important channel through which EMU will affect the CEE countries and selected Mediterranean countries. This section considers first the CEE countries, beginning with a concise description of recent trade developments between this group of countries and the EU. The fact that the exports of the CEE countries have become highly dependent on the EU market suggests that EMU could have a significant effect on them: this section examines with greater specificity some of the relevant factors that could come into play, A similar, but briefer, analysis is undertaken for the selected Mediterranean countries.

Importance of the EU Countries for CEE Country Trade: Recent Developments

Since the beginning of the transition process, CEE exports to and imports from the EU have grown very rapidly, leading to a major reorientation of CEE countries’ foreign trade toward the EU (Tables2.1 and 2.2). By the mid-1990s. CEE countries were directing on average 56 percent of their merchandise exports to the EU market and acquiring a similar proportion of their merchandise imports from the EU (Figure 2.1). As discussed below, while this reorientation of trade toward the EU has been accompanied by some restructuring in the commodity composition of exports, the manufacturing sector continues to be a mainstay of CEE exports and it has been an important contributor to recent export growth.

Figure 2.1.
Figure 2.1.

EU’s Share of CEE Countries’ Trade1

Source: IMF. Direction of Trade Statistics database1 Based on EU-15.
Table 2.1.

Growth of CEE Countries’ Trade with the European Union 1

(Annual percentage change)

article image
Source; IMF, Direction of Trade Statistics database.

Refers to EU 15. Calculated from data valued In U.S. dollar

Table 2.2.

Reorientation of CEE Countries’Trade Toward the European Union1

(In percent of each CEE country’ total exports/imports)

article image
Source: IMF, Direction of Trade Statistics database.

Based on EU 15.

Several factors contributed to the dynamics of expanding CEE-EU trade. First was trade liberalization, through the extension by the EU of the Generalized System of Preferences, and the signing of new trade agreements.2 A second element was the tendency for trade to gravitate toward western Europe once markets became freer. Based on gravity models, several studies predicted a substantial increase in trade with the EU. implying a rapid and substantial reorientation of exports toward western Europe from Council of Mutual Economic Assistance (CMEA) structures based on bureaucratic commands.3 Third, there is some econometric evidence to suggest that increased exports to the EU reflected in part a redirection of goods from the CMEA to other markets through significant cuts in price, or so-called distress exports.4 Finally, the successes achieved in the area of macro-economic stabilization and the supply effects of structural reforms in the various transition economies also contributed to growing trade.

Drawing on various studies on the commodity composition of CEE exports reveals that manufacturing exports dominated CEE exports to the EU in the early 1990s. Sheets and Boata (1996), using data up to 1994, also found that over 80 percent of the net increase in exports to the EU of Czechoslovakia, Hungary, and Poland were in the three one-digit industries principally engaged in manufacturing (manufactured goods, miscellaneous manufactures, and machinery and transport equipment).5 More recent empirical work by the European Bank for Reconstruction and Development (EBRD) (1997) supported the conclusion that manufacturing has been a particularly dynamic segment of CEE exports to the EU. An earlier study by Halpern (1995) reported that the structure of CEE exports to the EU was similar to that of Greece and Portugal.6 Another earlier study, by Graziani (1994), also used data through 1991 and indicated that the relative specialization of several CEE countries (Bulgaria, Czechoslovakia, Hungary, Poland, and Romania) had expressed itself in those sectors where competition with the newly industrialized countries was comparatively strong and competitive phenomena appeared among CEE countries themselves. In all, the concentration of trade in manufactures and the presence of other competitors would suggest that the exports of individual CEE countries could be rather sensitive to relative prices and, therefore, to movements in their real effective exchange rates.7

Trade Effects of EMU on the CEE Countries

To undertake an analysis of the potential effects of different euro-area growth and exchange rate scenarios on the exports of the CEE countries to the euro area, it would have been helpful if econometric estimates, such as those produced by standard trade equations, were generally available. Such estimates would have given, for each of the CEE countries, an idea of the elasticity of exports to the euro area with respect to both euro-area real GDP growth and the real effective exchange rate of the individual CEE country. Unfortunately, it is difficult to isolate such effects, especially given the relatively short time series available since the transition began and the structural changes that have occurred and will continue to occur, which tend to render estimates unstable and uncertain. Nevertheless, the importance of the euro-area market to exports from the CEE countries and the composition of those exports suggest changes in euro-area real GDP growth and EMU-induced changes in the real effective exchange rates would influence CEE trade significantly.

Activity Effects

By relying on data concerning euro-area imports, we attempted to gauge the order of magnitude of the effects of potential variations in real growth in the euro area on CEE exports to that market (Table 2.3).8 First, the impact of higher euro-area growth on euro-area imports was estimated by applying the income elasticities of imports, obtained from the IMF’s multi-region macroeconometric model (MULTIMOD). Second, the effect of higher euro-area import demand on CEE countries’ exports was estimated assuming that the market shares of each of the CEE countries in each of the EU countries remain at their 1996 levels.

Table 2.3.

Estimated Effects of Changes in EU-11 Real GDP on Exports from CEE Countries

(in million of U.S. dollars, unless otherwise indicated)

article image
Sources: IMF; Direction of Trade Statistics, database: IMF, World Economic Outlook database (for GDP only);and IMF staff calculations.

These calculations are, however, very rough. One important reason is that EMU, by eliminating currency risk and currency-related transaction costs, will reduce the cost of intra-euro-area trade. This process, which is equivalent to a reduction in tariffs between euro-area countries, will tend to increase the competitiveness of euro-area producers relative to non-EU suppliers and will, therefore, tend to divert trade away from the latter. Thus, any increase in the euro area’s demand for imports stemming from higher real growth may fall more heavily on intra-euro-area trade and thereby reduce the share of imports from nonmembers.9 In this sense, the calculations contained in Table 2.3 should be considered upper bounds.10

As shown in the table, the effects of a 1 percent increase in real GDP in each of the EU-11 countries results in an increase in exports from the CEE countries that ranges from 0.7 percent to 1.6 percent of their total exports in 1996. Such increases in exports represent 0.2 percent to 0.5 percent of the 1996 GDP of the individual CEE countries, effects of a meaningful magnitude. Thus, should it turn out that EMU has a positive (or negative) effect on real growth in the EU countries, CEE countries could benefit (be harmed) in significant ways. The results of a survey of IMF country desk staff generally corroborated these conclusions: a 1 percent increase in real GDP in each of the EU-11 countries was estimated to increase exports by 0,5 percent to 1.1 percent, except in Albania, Poland, and Slovenia, where exports were estimated to rise by more than 2 percent. The range of estimated effects on real GDP growth tended to be somewhat higher than the corresponding figures in Table 2.3, reflecting the impact of multiplier effects

Effects Through Changes in Real Effective Exchange Rates

It is more difficult to gauge the potential effects of EMU-induced exchange rate changes than to estimate the effects of changes in economic growth in the euro area. However, even though empirical estimates of the effects of (exogenously treated) changes in real effective exchange rates on trade are generally unavailable, on a priori grounds there is reason to believe that CEE exports should be sensitive to these changes. As mentioned earlier, the structure of CEE exports is one important factor. Some econometric evidence is also consistent with this conclusion.11

The implications for the effective exchange rates of CEE currencies of a given change in the exchange rate of the euro against other major currencies will depend on the interplay between the exchange rate regimes of the CEE countries and the geographical composition of their trade. Table 2.4 summarizes the exchange rate regimes of the CEE countries. Several of them (e.g., Bulgaria, Croatia, the former Yugoslav Republic of Macedonia, and Slovenia) currently peg or maintain their currencies within narrow bounds against the deutsche mark and can be assumed to replace it as a reference currency with the euro, upon the launching of EMU. In the event of, for example, an appreciation of the euro, the currencies of these countries would appreciate in effective terms and these countries would thereby lose competitiveness, the more so the lower the share of the euro area in their trade. A survey of IMF economist staff suggests that, for these countries, a 10 percent appreciation of the euro could result in a current account deterioration of 1–3 percent of GDP and could lower cumulative real GDP growth by roughly the same amount over a four-year period. Countries that peg to a basket would experience comparatively minor effects as a result of an appreciation of the euro to the extent that the basket peg closely approximates the geographical pattern of their trade (e.g., Hungary and the Slovak Republic). For a third group (notably Poland), the imbalance between the basket peg and the direction of trade could result in more significant effects: an effective depreciation of the currency if the euro appreciates, which would tend to have a positive effect on real GDP growth and contribute to narrowing the current account deficit.

Table 2.4.

Exchange Rate Regimes, July 31,1998

article image
Sources: IMF, Annual Report on Exchange Arrangements and Restrictions 1997; and IMF staff reports.Note: DM = deutsche mark; ECU = European currency unit; F = French franc; C = pound sterling; SWF = Swiss franc: USJ = US, dollar; ¥ = Japanese yen.

The National Bank of Slovakia decided on October 1, 1998, to float the koruna.

In June 1997, the authorities approved a widening of the band from 14 (±7) percent to 28 percent but only by adjusting downward the depreciation limit At the same time the slope (the rate of crawl) of the appreciation limit was decreased to 4 percent a year from 6 percent while the slope of the depreciation limit was maintained at 6 percent, implying a widening of the band by an additional 2 percentage points over the following year

Mediterranean Countries

Our estimates of the potential effects of EMU on selected Mediterranean countries through changes in economic activity in the euro area are set out in Table 2.5. It shows that the effects on Malta are potentially very large: a 1 percent increase in real GDP in the EU-11 is estimated to raise total exports from Malta by almost 3 percent; this increase in exports represents about 1 1/2 percent of GDP. Higher growth in the EU would raise exports in Turkey by 0.8 percent according to the estimates, but this increase represents only 0.1 percent of GDP since exports are a relatively small proportion of economic activity. For both Cyprus and Israel, the effect on exports is much smaller (a 0.4 percent increase for Cyprus and a 0.5 percent increase for Israel), representing a minor portion of economic activity in each country (0.1 percent).12

Table 2.5.

Estimated Effects of Changes in EU-11 Real GDP on Exports from Selected Mediterranean Countries

(in millions of U.S. dollars, unless otherwise indicated)

article image
Sources: IMF, Direction of Trade Statistic: database: IMF. World Economic Outlook database (for GDP only): and IMF staff calculations.

In terms of exchange rate changes, a survey of IMF economist staff indicated that an appreciation of the euro might be expected to have relatively little impact on Malta since it pegs its currency to a basket closely approximating its direction of trade. Such an appreciation could result in some effective depreciation of the Israeli shekel and the Turkish lira, which would be expected to increase exports and real growth in Israel and Turkey. By comparison, Cyprus, which pegs to the ECU, would be likely to experience some effective exchange rate appreciation and a consequent reduction in exports and real GDP growth.

Financial Linkages

EMU may also affect the CEE countries and the selected Mediterranean countries through a number of financial channels. The most obvious of these financial linkages operate through capital flows between the euro area and the non-EU countries under analysis; they can be divided into two main categories. The first is changes in capital flows other than FDI, resulting from EMU-induced changes in interest rates in the euro area, portfolio diversification effects, and the increased attractiveness of lending and borrowing in the financial markets of the euro area as EMU increases their depth and liquidity. The second category is changes in FDI flows, resulting from EMU-induced changes in economic growth, in the marginal productivity of capital, and in transaction costs within the euro area. Other financial repercussions of EMU include its possible impact on the external debt service of the countries under analysis and the potential impact on the structure of their financial systems.

This section introduces the analysis by focusing on the various ways in which EMU may encourage movements in non-FDI capital flows. Since the extent to which these flows will be affected by EMU will depend on the degree of international financial integration of the countries concerned, this section examines a number of relevant indicators and, when available, econometric tests. It then discusses the possible impact of EMU, in combination with the full implementation of the EU directives in the area of financial services, on the structure of the financial sectors of the countries under study. Finally, the effects through FDI flows and debt service are discussed.

Effect on Capital Flows Other than Foreign Direct Investment

As highlighted above, there are several channels through which EMU may have an effect on capital flows to the CEE and selected Mediterranean countries. Some of these may result in increased capital inflows to them, while others would tend to attract capital into the euro area. All told, it is difficult to know how these various effects may balance out in the end or how investors’ appetite for risk vis-à-vis emerging market assets might change over time. Thus, the focus of the discussion that follows is on understanding the various considerations that may come into play rather than on their relative quantitative importance.

Changes in the Level of Interest Rate in the Euro Area

The level of interest rates in the euro area following the realization of EMU is perhaps the most evident financial variable through which EMU will affect outside countries. In conjunction with possible changes in expectations about the exchange rate of the euro against other world currencies, any changes in interest rates in the euro area will affect interest-sensitive capital flows between the EMU participants and the CEE countries and selected Mediterranean countries.

There is, however, considerable uncertainty among analysts about how EMU may actually affect interest rates in the participating countries. Moreover, the discussion on the interest rate implications of EMU is often couched in a wider policy context that takes into account not only the new regime for monetary and exchange rate policies, but also changes in fiscal and structural policies that are only partly related to EMU itself. Thus, while it is often argued that a successful EMU could bring about a decline in interest rates in the euro area, this favorable scenario normally assumes that EMU also serves as a catalyst for accelerating fiscal consolidation and structural reforms (particularly in the labor market). Downward pressure on interest rates results from lower budget deficits and public debts, as well as from the favorable supply-side effects of structural reforms. Furthermore, these factors could allow the European Central Bank (ECB) to pursue a less restrictive monetary policy without compromising its credibility, possibly reducing the anti-inflation bias that some expect the ECB to have in the early years of EMU.

Of course, EMU need not be accompanied by these virtuous fiscal and structural policies, even though fiscal policies will be constrained by the EU’s Excessive Deficit Procedure and the Stability and Growth Pact. In this regard, the different interest rate implications of two scenarios for EMU and their accompanying fiscal and structural policies were presented in the October 1997 World Economic Outlook. These scenarios are reproduced in Table 2.6.13 Under a first, “successful EMU scenario,” long-term interest rates gradually fall to about half a percentage point below the baseline level by 2003. Under a second, “reform fatigue scenario,” by contrast, interest rates end up about half a percentage point above the baseline level.14

Table 2.6.

Effects in Euro Area of EMU Scenarios Under Different Assumptions Regarding Fiscal and Structural Policies

(Deviations from baseline, in percent, unless otherwise indicated)1

article image
Source: IMF World Economic Outlook (October 1997).

Baseline in the World Economic Outlook of October 1997. Adherence Co the Stability and Growth Pact is assumed in both scenarios. It is also assumed that all current EU member countries participate in EMU from the start.

While the actual behavior of euro-area interest rates will therefore depend significantly on the accompanying fiscal and structural policies, other factors—not necessarily EMU-related—could also come into play. Some of these could tend to make interest rates higher in the first years of EMU. First, interest rates in the euro area are currently at historically low levels. Should the ongoing cyclical recovery of the economies of prospective EMU countries gather momentum, the ECB may have to raise interest rates in the early years of EMU to keep inflationary pressures at bay.15 Second, the ECB could initially show a deflationary bias in order to establish credibility. Third, the introduction of a single currency is expected to reduce transaction costs, enhance total factor productivity, and boost investment demand within the EU.16 This is particularly likely under the favorable scenario, where structural reforms will contribute to productivity improvements, and investors’ “animal spirits” could be positively affected by a climate of growth and optimism. To the extent that EMU promotes investment demand, upward pressure on real interest rates should ensue.

Whatever the effect of EMU on interest rates in the euro area, the impact on outside countries through the capital flow channel will depend on the extent of their international financial integration (discussed below) and on their monetary and exchange rate regimes. In countries that peg their currencies to the euro, or attempt to maintain relatively stable exchange rates against it, domestic interest rates and domestic demand are likely to respond more to changes in euro-area interest rates than they will in countries that maintain relatively flexible exchange rates, where most of the adjustment is likely to take place via changes in the exchange rate.

Table 2.7 illustrates this point by highlighting the results of a survey of IMF country desk economists concerning the possible impact of a change in interest rates in the euro area on growth, the current account, and debt service in the countries under study. The analysis, expressed in terms of estimated deviations from baseline, is based on examining the effect of a 2 percentage point increase (relative to baseline) in short-term interest rates in the euro area in 1999–2000 and a 1 percentage point increase (relative to baseline) in 2001; long-term interest rates partially reflect this increase. In order to concentrate on the impact of interest rate changes, it is assumed that the euro/dollar exchange rate and economic growth within the euro area remain constant, but the analysis takes into account the particular exchange rate regimes that the different CEE countries and selected Mediterranean countries are likely to have in place when EMU is launched.

Table 2.7.

Estimated Effects of an Increase in Euro-Area Interest Rates on CEE Countries and Selected Mediterranean Countries1

(Deviations from baseline)

article image
Source: IMF staff estimates.

Simple averages of estimated effects on Albania, Bulgaria. Croatia. Cyprus, the Czech Republic. Hungary, Israel, former Yugoslav Republic of Macedonia. Malta. Poland, Romania, the Slovak Republic, Slovenia, and Turkey. The estimates assume no change in the Euro/dollar exchange rate and no change in the rate of economic growth within the euro area.

Debt service as a percentage of exports of goods and nonfactor services.

The survey suggests that an increase of interest rates in the euro area would, on average, result in some decline in GDP growth, some improvement in the current account, and some deterioration in the debt-service ratio. These results in turn reflect various underlying factors. First, on average, the countries under review experience an increase in their interest rates in response to the increase in euro-area interest rates. Second, on average, the negative impact of higher interest rates on growth (via lower domestic demand) outweighs any possible impact on growth through the depreciation of the domestic currency against the euro (and higher net exports) in countries that have floating exchange rates or adjust their pegs downward. And third, on average, the negative impact on debt service resulting from higher interest payments on euro-denominated foreign debt more than offsets the favorable impact on debt service stemming from an improved current account and foreign debt position.

Portfolio Diversification Effects

Whatever its impact on the average level of interest rates in the euro area, EMU should result in a reduction in interest rate differentials between local-currency-denominated debt issued by participating southern countries and comparable debt issued by core EMU countries. By eliminating exchange rate risk, EMU should make the spreads on local-currency government bonds move in line with those observed on foreign-currency public debt, so that they will reflect only differences in perceived credit risk and, to a lesser extent, differences in liquidity, settlement, and legal risks. In fact, this effect has essentially already taken place over the past two years in Italy, Portugal, and Spain, as these countries have made progress in the areas of Fiscal consolidation and inflation reduction, and as the perceived probability of EMU taking place and including these countries has increased (see Figure 2.2). Long-term yield spreads between domestic-currency-denominated debt issued by these three southern countries and comparable German treasury bonds are now similar to those observed between their foreign-currency-denominated debt and the foreign-currency-denominated debt issued by core EMU countries.

Figure 2.2.
Figure 2.2.

Long-Term Interest Rate Differentials vis-á-vis Germany1

(in percentage points)

Source; IMF staff.1 Differentials between government bonds of 10-year or nearest maturity. For Greece, differential between 12-month treasury bills.

Table 2.8 shows the spreads over comparable U.S. Treasury issues paid by different EU sovereigns on their dollar-denominated debt in recent years. Spain and Italy paid, depending on the maturities, only between 5 and 15 basis points more than the EU country with the lowest spread. This provides a reference for the spreads that are likely to prevail on domestic-currency debt under EMU. The spreads between sovereign issues could, in fact, all but disappear if a country’s participation in EMU increased the perception that it will be bailed out by other member countries, the EU budget, or the ECB in the event of a budgetary crisis. However, the “no bail-out” clause contained in Article 104b of the Maastricht Treaty, the prohibition of ECB financing of EMU governments, and the experience of federal states such as Canada or the United States (where the cost of provincial or state debt bears a close relation to the assigned credit ratings) make this rather unlikely. Thus, some interest rate spreads will probably remain.

Table 2.8.

Spreads on Long-Term U.S. Dollar-Denominated International Bonds1

(In basis points)

article image
Sources; Paribas Capital Markets, London; and Bloomberg Financial Markets LR

Spreads between bonds of 10-year or nearest maturity issued by EU sovereigns and comparable bonds issued by the U.S. Treasury.

Difference between the average spread shown by the country in question in the three months of observation and the average spread shown by the EU country with the lowest spread among those for which data were available.

Average of September 1996 and June 1997.

The convergence of long-term interest rates in southern EMU countries toward those prevailing in core EMU members could stimulate portfolio capital flows from the EMU area (as well as from other industrial countries) into the CEE countries and Mediterranean countries under consideration, as investors try to maintain the desired diversification of their portfolios between assets with different yield/risk combinations. In this respect, it should be noted that portfolio diversification by institutional investors in industrial countries seems to have been an important driving force behind the surge in private capital flows to developing countries seen in the 1990s. With spreads for potential southern EMU participants having narrowed to unprecedented levels, this effect may have already begun to operate ahead of the launching of Stage III of EMU.

Some of the countries under analysis are particularly well suited to replace the southern EU countries in the portfolios of international investors. This would appear to be the case for the Czech Republic, Hungary, Poland, and Slovenia, all of which offer assets with a higher-yield/higher-but-mode rate-risk combination that broadly resembles those offered until recently by southern EU countries. These are all European countries, have or are expected to have annual inflation rates in the 5–15 percent range by the time EMU is launched, have made substantial progress in the area of macroeconomic stabilization and structural reform, have been granted investment-grade credit ratings by the major international rating agencies (discussed in more detail below), have relatively developed stock exchanges, are liberalizing their capital accounts, and enjoy a rather stable political situation and close links with the EU (including the prospect of early accession to the EU).17 In addition, the Czech Republic, Hungary, and Poland have recently become members of the Organization for Economic Cooperation and Development (OECD), an important factor since many institutional investors are by law or by their own statutes limited to investing in OECD countries. Thus, these CEE countries seem the most natural substitutes for the southern EMU countries in the portfolios of institutional investors.

OECD membership could also facilitate a portfolio shift toward Turkey, although higher inflation, weaker links with the EU, and higher macroeconomic and political risk make this country a less obvious substitute.18 Israel is neither a European nor an OECD country and may be seen to exhibit higher political risk than the above-mentioned CEE countries. But having an inflation rate in the 5–15 percent range, an open capital account, relatively developed capital markets, and an investment-grade credit rating, it could also benefit significantly from portfolio shifts.

This “search for higher yields” effect is likely to be reinforced by other EMU-induced portfolio shifts. For example, EMU could lead to a higher degree of synchronization of macroeconomic performance and financial asset price movements between the different EMU countries.19 This would encourage investors to find new opportunities for portfolio diversification outside the euro area, including in the countries under analysis. In addition, EMU (in combination with other factors such as the development within EU countries of parallel, privately managed pension pillars) is expected to lead to an acceleration in the process of securitization that is already being observed in the financial systems of EU countries.20 This will expand the size of the portfolios of institutional investors located in the euro area, and these investors are likely to place part of this increase in their available funds in assets of neighboring non-EU countries. Also, as some CEE countries and Mediterranean countries join the EU and start to approach participation in EMU, the markets may further increase their holdings of bonds denominated in those countries’ currencies under so-called convergence plays, where traders bet that the gap in bond yields between peripheral countries and EMU countries will shrink rapidly. Finally, EMU will ease the constraints imposed by currency-matching requirements on foreign-currency-denominated investments by insurance companies, pension funds, and mutual funds located in the euro area. These rules aim at limiting the foreign exchange risk associated with having obligations to policyholders or shareholders denominated in local currency while some investments are denominated in foreign currency. With EMU, foreign currency investments made by affected institutions in other EMU currencies will automatically be reclassified as domestic-currency investments, thus increasing the margin available to those institutions for investing in non-euro currencies. Some national (or statutory) regulations define currency-matching requirements in terms of the nationality of the issuer, in which case the requirement would continue to be binding despite EMU.21 Whatever their exact definition, however, matching requirements within the EU are expected to be eased in the context of either the future Pension Funds Directive or a stricter interpretation of the directive of 1998 liberalizing capital movements.22 The OECD is also bringing into the discipline of the Capital Movements Code certain restrictions on portfolio investment abroad by pension funds and insurance companies, which would have the effect of easing constraints on diversification.23

In sum, EMU is likely to trigger or intensify a number of portfolio effects that, in combination, should contribute significantly to increasing portfolio flows from the EU and other industrial countries into the countries under analysis. Among these countries, those that stand to benefit most from these portfolio shifts are those with relatively open capital accounts and developed domestic securities markets, with a higher degree of macroeconomic and political stability, with relatively low inflation rates (although not as low as in the euro area), with close links with the EU, and that are members of the OECD.

Increased Investment and Borrowing in Euros by Non-EU Countries

It is often argued that, by creating more highly integrated, larger, and more liquid securities markets in the euro area, EMU will attract international investment flows to that area.24 Those non-EU countries that have made more progress in liberalizing capital outflows (particularly portfolio flows) and developing an institutional investor base could therefore experience an increase in outflows to the euro area as a result of EMU, although this effect is likely to be moderate. The enhanced depth and liquidity of EMU bond and equity markets will also tend to increase their relative attractiveness for investors located in the countries participating in EMU and other industrial countries. This could divert certain investment flows away from non-EU emerging economies and into the euro area, thus partially offsetting the portfolio effects referred to in the previous section.

EMU is also likely to increase borrowing in euros by non-EU countries (for a given interest rate differential vis-à-vis the euro area), since non-EU borrowers would have access to the larger, more liquid, and more competitive financial markets that EMU will generate. First, EMU is expected to increase substantially competition in the markets for the underwriting of bond issues and syndicated bank loans.25 This will tend to reduce the cost of issuing in the European bond markets and of obtaining syndicated loans. A second reason, noted by McCauley and White (1997), is that, to the extent that the government bond market in the euro area becomes more liquid, the costs to underwriters of hedging a new issue through taking a short position in government bonds in the cash or futures market would decline, and this saving would also be expected to be passed through to the borrower. Third, EMU is expected to increase the efficiency of the euro-area payment system and improve its linkages with those of surrounding non-EU countries (discussed below). Combined with ongoing deregulation and technological progress in the financial systems of both EMU and non-EU countries, and capital account liberalization in the latter, these factors should also stimulate borrowing by non-EU countries in the euro area. Finally, EMU is likely to make EU equity markets more efficient by accelerating the process of competition, integration, and technological innovation that has characterized them in recent years. Equity issues by non-EU countries in the EU stock exchanges may therefore increase to take advantage of the latter’s higher efficiency and wider investor base.

This expansion of euro-denominated borrowing would tend to increase the still relatively low share of the currencies of prospective EMU participants in the total foreign debt of non-EU countries, which would have implications for the sensitivity of their debt-service payments to changes in euro interest and exchange rates (discussed below). Moreover, when combined with the stimulating effect of KMU on portfolio flows from non-EU countries into the euro area and with the various other portfolio effects discussed earlier, higher euro borrowing by non-EU countries should result in an increased degree of financial integration between the two areas.

Measuring the International Financial Integration of the Countries Under Study

The actual response of capital flows to EMU-induced changes in interest rates and to the other factors discussed above will depend not just on the extent of financial interaction between EMU participants and non-EU countries, but also, more generally, on the overall degree of international financial integration of the latter group. For example, a decline in yield spreads in the southern EU countries as a result of their participation in EMU could induce a portfolio reallocation toward certain non-EU countries by institutional investors located in, say, New York or Hong Kong SAR, in addition to those located in the EMU countries. The extent to which changes in interest rates in the euro area influence capital flows to the non-EU countries will also depend on the ability of investors located in third countries to shift funds easily between the euro area and the non-EU countries.

This section and the accompanying Table 2.9, through Table 2.6, and Figure 2.3 and 2.4) examine a number of indicators of the degree of financial openness of the countries under review.26 The bottom line of the analysis is that, while international financial integration seems to be increasing in most of the countries examined, and at a relatively fast pace in some of them, the degree of integration already achieved differs considerably across countries. Financial openness seems highest in the Czech Republic, Hungary, Israel, and Poland. Thus, these four countries could be significantly more affected than the others by EMU-related financial shocks, even if it is difficult to estimate the magnitude of the corresponding effects. Croatia, Cyprus, Malta, the Slovak Republic. Slovenia, and Turkey show an intermediate, though significant and growing, degree of integration. At the lower end of the financial integration spectrum—though with differentiation within the group—are Albania, Bosnia-Herzegovina, Bulgaria, the former Yugoslav Republic of Macedonia. Romania, and the Federal Republic of Yugoslavia. These countries continue to impose restrictions on most of their capital transactions and have weak credit ratings and limited or no access to the international capital markets; they are the least likely to be affected by financial developments in the euro area.

Figure 2.3.
Figure 2.3.

Stock Market Indices in Selected CEE Countries1

(January 1995 = 1OO)

Source: International Finance Corporation (IFC), Emerging Markets Database.1 IFC weekly global indices.
Figure 2.4.
Figure 2.4.

Stock Market Indices in Selected Mediterranean Countries1

(January 1995 = 100)

Sources: International Finance Corporation (IFC), Emerging Markets Database: and Bloomberg Financial Markets LP.1 IFC weekly global indices. For Israel, local market index (monthly data).
Table 2.9.

Selected Capital Controls in CEE Countries and Mediterranean Countries

(Position as of March 31,1997)

article image
Table 2.9
article image
Table 2.9
article image
Table 2.9
article image
Sources: IMF, Annual Report on Exchange Arrangements and Exchange Restrictions 1997 and various Recent Economic Developments reports: and Organization for Economic Cooperation and Development.

With the exception of investments in sectors normally considered sensitive or of strategic national interest.

Borrowed or extended by residents Other than banks.

Securities with more than one year of original maturity, such as shares and other securities of a participatory nature, and bonds.

Securities with one year or less of original maturity, such as treasury bills, certificates of deposit, and bills of exchange.

The index can take values between 0 and 100, with 100 representing the maximum degree of liberalization of the capital flows under consideration, The index for a given country is constructed by adding up the values obtained in each category of capital flows and dividing this total by the maximum possible score. Flows not subject to controls (“no” in the table) are assigned a value of 2 flows classified as being subject to partial controls (“partial” in the table) are assigned a value of I; flows subject to serious controls (“yes” in the table) are given a value of zero. When information on a given capital transaction is not available, a value of zero is assigned in both the numerator and the denominator.

Capital Account Regulations

The degree of financial integration should be related to capital account regulations to the extent that these regulations are enforced. In examining such regulations, this section emphasizes when relevant any specific rules applied only to flows with prospective EMU participants.27

Most CEE countries lifted the restrictions on FDI inflows (except in a few sensitive or defense-related sectors) at the beginning of their transition to a market economy. They also guaranteed in most cases the free repatriation of both profits (current account convertibility) and FDI capital. FDI inflows were welcomed as a not-easily-reversible source of foreign exchange and because of the usual externalities associated with FDI (introduction of new technologies and managerial skills, access to export networks, and so forth). In addition, FDI was to play a key role both in these countries’ privatization and enterprise restructuring processes and as a source of government revenues. Most CEE countries initially required authorization for FDI outflows, but the authorization was normally granted liberally when the investment was thought to help exports. Since early in the reform process, most CEE countries have also applied a quite liberal treatment to trade credits (in the case of outward trade credits because of their positive effect on exports), and allowed individuals to hold and operate foreign exchange accounts at local banks, a privilege that most industrial countries accorded only at the last stage of liberalization of their capital accounts.28

With the exception of FDI-related transactions, however, most other capital movements remained very restricted in CEE countries, as these countries feared the potentially destabilizing effects of an open capital account at a time when they were still trying to put their macroeconomic houses in order. In particular, financial credits, portfolio flows, and real estate operations remained severely restricted for years in practically all countries. A general pattern, and one that has endured until today, was the existence (legally and de facto) of much tighter controls on outflows than on inflows and more serious restrictions on short-term than on long-term transactions.29

Since 1995, a number of CEE countries have been gradually easing restrictions on capital movements. This process has been led by the Czech Republic, Hungary, and Poland, all of which have undertaken substantial liberalization commitments as part of their accession to the OECD in December 1995, May 1996, and November 1996, respectively.-30 But other CEE countries have also taken measures in recent years to ease their capital controls.

In the context of their accession to the OECD, the Czech Republic, Hungary, and Poland have liberalized FDI outflows (vis-à-vis OECD countries and countries having signed bilateral investment protection agreements in the case of Hungary and Poland), outward trade credits, and personal capital movements. They have also liberalized other outflows (particularly long-term outflows) to varying degrees. Thus, the Czech Republic has fully liberalized the purchase abroad by residents of any foreign security, as well as Czech investment in real estate abroad, although it continues to require authorization for the admission of foreign securities to its capital and money markets. Hungary has allowed residents to invest in securities of more than one year of original maturity, issued by OECD governments or OECD-based companies with an investment grade rating. Poland has liberalized, for maturities of more than one year, outward financial credits and purchases of securities issued by OECD countries. On the side of inflows, the Foreign Exchange Law passed by the Czech Republic on October 1, 1995, removed the requirement of prior authorization for most inflows, while Poland and Hungary have removed all restrictions on medium- and long-term inflows (maturities of more than one year).31

Overall, these measures have contributed to a substantial liberalization of flows (including portfolio flows) in these three countries, particularly with respect to medium- and long-term flows and on the side of inflows. The acceptance of the OECD Capital Movements Code also implies for these countries the need to comply with its standstill, nondiscrimination, and transparency clauses.32 Finally, the Czech Republic and Hungary have also committed themselves to abolishing all remaining capital controls within three to four years of their dates of OECD membership, if the macroeconomic situation permits, while Poland has committed itself within the OECD to removing all remaining restrictions by end-1999.

Another institutional process that is having a bearing on the capital account regulations of the countries under analysis is the signing of Association Agreements with the EU and, for a number of countries, the prospect of early accession to the EU. The Europe Agreements concluded between the EU and some CEE countries oblige both parties to liberalize FDI-related flows between themselves. They also encourage CEE countries to take further liberalization measures with the ultimate objective of fully liberalizing flows with the EU. Furthermore, countries joining the EU will have to free capital movements visa-vis all countries from the date of membership, unless they obtain a transitory period for this part of the acquis communautaire:.33 Those countries that started accession negotiations in early 1998 (Cyprus, the Czech Republic, Hungary, Poland, and Slovenia, among the countries considered here), could join the EU as soon as in 2002, meaning that they would in principle have to have a completely open capital account by then. The commitments before the OECD undertaken by Poland to lift all remaining restrictions by end-1999, and by Hungary and the Czech Republic to do so within three to four years, therefore, are consistent with their prospect of EU membership.

Empirical Tests of International Financial Integration for Mediterranean Countries

A number of empirical tests of financial openness have been conducted in the literature, based, for example, on interest parity conditions, saving and investment correlations, or correlations of consumption levels across countries.1 In the case of the CEE countries, however, empirical tests of financial openness are generally not available, reflecting the lack of appropriate national account statistics or sufficiently long time series, and the econometric difficulties in dealing with rapid structural change. For the four Mediterranean countries under analysis, a number of formal estimates have been produced in the context of more general tests for developing countries.

An important finding of the empirical tests of capital mobility in developing countries performed in the 1980s and early 1990s was that these countries are more financially open than had been assumed (see Montiel, 1993). This conclusion was reached even though most of these studies were done before the surge in private capital inflows experienced by a number of developing countries in the 1990s. Available empirical estimates for Cyprus, Israel, Malta, and Turkey, shown in the table, are in line with the main conclusions of this literature. They generally tend to classify Israel as a country with a relatively high degree of international financial integration, and the other three countries as having an intermediate-to-high degree of financial openness.

Haque and Montiel (1990), using the methodology proposed in Edwards and Khan (1985) to test the uncovered interest parity condition, concluded that during the period 1969–87 both Malta and Turkey showed intermediate degrees of international financial integration. Montiel (1989) and Dowla and Chowdhury (1991) tested for the hypothesis of monetary autonomy in a number of developing countries. The idea behind these two studies was that, with perfect international capital mobility and fixed exchange rates, the domestic credit or money supply should not cause, in the Granger sense, movements in nominal income. Using annual data for the period 1962–86, Montiel rejected the hypothesis that these domestic financial aggregates Granger-caused nominal income in Turkey, implying a relatively high degree of financial openness in this country. Dowla and Chowdhury reached a similar conclusion for Israel, using quarterly data for the period 1957–89.

Montiel (1993) performed a number of tests of financial openness with data for several developing countries, including the computation of gross capital inflows and outflows as a percentage of GDP, Feldstein-Ho-rioka type of saving and investment correlations, tests of uncovered interest parity, and correlations of national and world consumption. The data covered, depending on the test and the country, different subsets of years within the period 1960–90. Montiel concluded that the degree of financial integration was high in Israel and intermediate in Cyprus and Turkey. Among the tests performed by Montiel (1993), those that distinguished different subperiods detected an increase in financial integration over time in Cyprus and Israel, but not in Turkey. For Malta, the results were inconclusive.

Mamingi (1993) estimated saving and investment correlations for 58 developing countries over the period 1970–90. He concluded that Turkey had an intermediate and Malta an intermediate-to-high degree of capital mobility. Finally, Obstfeld (1994) tested for the correlation of national consumption with world consumption in a sample of industrial and developing countries including Cyprus and Turkey for the periods 1951–72 and 1973–88. Cyprus was found to have in the second period the highest correlation among the developing countries included in the sample, with the coefficient of correlation (0.64) close to the highest coefficients obtained by industrial countries, implying a relatively high degree of international financial integration.2 The coefficient also increased between the two periods, consistent with the findings by Montiel (1993). For Turkey, by contrast, the coefficient of correlation was very low and (also consistent with Montiel’s results) slightly declining.

Results of Empirical Tests of International Financial Integration in Mediterranean Countries

article image

Combines degree of access to international capital markers total private capital inflows/GDR and a measure of the level of diversification of the country’s financing.

For the period 1992–94.

Not too much emphasis should be put on the suggestion in these studies that international financial integration in Turkey may have actually declined during the 1970s and 1980s. Apart from the general caution with which all the results of these tests should be interpreted, these tests do not cover, as noted, the surge in private capital flows to developing countries of the 1990s, of which Turkey has been an important beneficiary (see below). In this respect, a study recently published by the World Bank, which classified a sample of 65 developing countries according to their degree of financial integration using data up to 1994, concluded that Turkey had moved from an intermediate level of integration in 1985–87 to a high level in 1992–94.3

1 For a discussion of different approaches to measuring international financial integration, see Goldstein and Mussa f 1993), Montiel (1993). and Obstfeld (1994). For a survey of the growing body of literature applying these techniques to developing countries, see Montiel (1993).2 A coefficient of 1 implies, under certain assumptions, perfect capital mobility.3 See World Bank (1997a). The World Bank constructed an overall index of financial integration by combining a measure of access to the international capital markets, a measure of a country’s ability to attract private capital inflows (based on total gross inflows over GDP but adjusted to weight more heavily flows other than FDI), and a measure of the level of diversification of a country’s financing (with higher diversification reflecting higher integration). Interestingly, the World Bank calculations showed Hungary, the only CEE country included in the sample, moving from the group of developing countries with low integration in 1985–87 to the group of highly integrated countries in 1992–94.

Measures in the area of capital account liberalization are also contained in the Euro-Mediterranean Agreement signed in 1995 between Israel and the EU, though not in the Association Agreements concluded by the EU with Cyprus, Malta, and Turkey. Among these Mediterranean countries, the liberalization efforts are currently being led by Israel, which has already achieved a relatively high degree of capital account openness and has announced its intention to do away with virtually all remaining restrictions in 1998, Turkey undertook a substantial liberalization of its capital account during the 1980s and first half of the 1990s, partly reflecting its obligations as a member of the OECD, Cyprus, for its part, has taken steps in recent years to liberalize its capital account and plans to take further measures in the coming years, in preparation for its integration into the EU.34

(Table 2.9) provides a summary description of existing controls on selected capital flows in the countries under analysis. It focuses on FDI, real estate, and portfolio and credit flows not involving the domestic banking system, thus leaving aside the specific provisions affecting banks and other financial institutions, the investment restrictions affecting institutional investors, and the regulations determining the capacity of residents and nonresidents to hold and operate foreign- and domestic-currency-denominated accounts. While some of the regulations excluded from the analysis are admittedly important, they are often too complex to be characterized in a simple way, and the overall degree of liberalization of the capital account might reasonably be expected to show a high degree of correlation with the overall degree of liberalization of the capital flows included in Table 2.9. An index of liberalization of these selected flows was calculated as a proxy for the overall degree of capital account openness; it is also reported as a memorandum item in table 2.9.

This index can take values between 0 and 100, with 100 representing the maximum degree of liberalization. It shows that the Czech Republic, with an index of 74 percent, was, by a substantial margin, the country that had as of early 1997 made the most progress in opening its capital account, followed, in order, by Israel, Hungary, Poland, and Turkey (all with indices in the 50–62 percent range). Bulgaria, Croatia, Cyprus, Malta, and Slovenia stood at an intermediate level of liberalization. All other countries showed indices below 25 percent, although in some of these countries (in particular, Albania and Bosnia-Herzegovina) the enforcement of capital controls is so weak that the index is likely to significantly underestimate the actual openness of their capital accounts.

Evidence from Net Capital Flows

The regulatory framework is only a partial indicator of actual international financial integration. First, controls on capital flows are often circumvented so that it is possible to have a country with a very closed capital account on paper but, because of weak enforcement, a relatively open capital account in practice. Conversely, despite the removal of many capital account restrictions, a country may show little international capital mobility, reflecting underdeveloped domestic capital markets, lack of market access, or other factors. Because of such considerations, the analysis is complemented in the remainder of this section by looking at the trends in total net capital inflows, indicators of market access (such as ratings, spreads, and actual presence in the international capital markets), and, for some countries, correlations between their stock market indices and those of other stock exchanges.

As shown in Table 2.10, most CEE countries have experienced a significant increase in net capital inflows since 1993. After an initial period characterized by weak private inflows (largely limited to FDI inflows and international bond issuance concentrated in a few countries), a predominance of official assistance, and capital flight in some countries, several CEE countries began to experience in 1993 a sharp increase in private capital inflows. Those were largely countries where the stabilization and structural reform efforts were most advanced. In particular, the Czech Republic and Hungary were among the main beneficiaries of the surge in private capital inflows to emerging countries observed since the early 1990s.-35 Net capital inflows into the CEE countries as a whole peaked in 1995, with the Czech Republic and Hungary both receiving record net inflows of nearly 18 percent of GDP. They abated substantially in 1996–97, led by the deceleration and subsequent partial reversal of inflows into the Czech Republic and Hungary, but remained high in most countries (Croatia and the Slovak Republic in particular).36

Table 2.10.

Net Capital Inflows1

(in percent of GDP)

article image
Sources: IMF, Balance of Payments Statistics Yearbook 1997; IMF International Financial Statistics; and IMF staff estimates.

Met capital inflows are defined as the balance on the financial account of the balance of payments, excluding changes in international reserves, plus net errors and omissions. In a number of countries, the net errors and omissions item is large.

Until 1992. refers to the former Czechoslovakia

Bulgaria. Czech Republic, Hungary. Poland. Romania, and the Slovak Republic.

With the exception of Cyprus, the Mediterranean countries have also experienced this trend, with Israel and Malta receiving average net capital inflows in the period 1993–97 of about 8 percent and 10 percent of GDP, respectively, and capital inflows into Turkey resuming at a healthy rate after the 1994 balance of payments crisis.

Ratings and Access to the International Capital Markets

Ratings. The majority of the countries included in this analysis have seen an improvement, very marked in a number of cases, in their credit ratings in recent years (see Table 2.11), Twelve countries have already been assigned long-term foreign-currency sovereign ratings by the major international rating agencies. Among the CEE countries, five (Croatia, the Czech Republic, Hungary, Poland, and Slovenia) enjoy investment grade ratings from both Moody’s and Standard & Poor’s, with the Czech Republic and Slovenia having been granted an A level mark by at least one of those rating agencies.37 The last CEE country to have joined this group is Croatia, which was assigned the lowest investment grade ratings by both Moody’s and Standard & Poor’s in January 1997. The Slovak Republic had also been assigned investment grade ratings by Moody’s and Standard & Poor’s in 1995–96, but Moody’s downgraded the country’s debt from Baa3 to Bal in March 1998, Bulgaria and Romania remain at the subinvestment grade category, and the rest of the CEE countries have not yet been rated. Among the Mediterranean countries considered here, Israel, Malta, and Cyprus all enjoy A level ratings at present. In contrast with this generally positive trend, Turkey has been successively downgraded since 1994 from an investment grade mark to the lowest mark in the noninvestment grade.38

Table 2.11.

Sovereign Credit Ratings1

article image
Sources: Moody’s Investor Service; and Standard & Poor’s

Ratings on long-term foreign currency bonds. The ratings are ranked in descending order according to the degree of creditworthiness as follows:

  • Moody’sStandard & Poor’s

  • Investment gradeAaa, Aa, A, BaaAAA, AA, A, BBB

  • Noninvestment gradeBa, BBB, B

  • Default gradeCaa, Ca, CCCC, CC, C, D

Borrowers within a given grade are differentiated by attaching either numbers from \ (highest creditworthiness) to 3 (Moody’s), or + and - signs (Standard & Poor’s).

Israel was assigned a BBS rating by Standard & Poor’s in October J 986.