Journal Issue

Portugal—Selected Issues

International Monetary Fund
Published Date:
January 1998
  • ShareShare
Show Summary Details

III. The External Sector: Some Key Forces at Work38

A. Introduction and Summary

95. The real effective exchange rate of the escudo underwent a significant appreciation in the decade since Portugal’s accession to EU in 1986. This appreciation, however, has not been accompanied by a fundamental deterioration in the external accounts. After hovering around balance through 1993, the current account has more recently shifted to a deficit in the order of 2-2½ percent of GDP, owing in part to a decline in private transfers and a growing deficit in the income account. These recent developments are partly due to financial innovation, and do not appear to reflect competitiveness factors; they may also be overstated by data difficulties.39 Looking to the future, however, major changes in the external environment facing Portugal are expected to take place. In particular, the advent of EMU and EU enlargement to Central and Eastern European countries are likely to have significant effects on the Portuguese economy.

96. The sustained real appreciation of the escudo coincided with Portugal’s accession to the EU, that is, with a period of profound structural changes in the economy, accompanied by high levels of investment and rapid growth. The hypothesis that this shift in economic fundamentals justified a higher value for the escudo, i.e., that the observed appreciation was an equilibrium phenomenon, has been extensively investigated. The bulk of the research undertaken in recent years has indeed come to this conclusion, linking the appreciation of the exchange rate to economic fundamentals. The staff also came to a similar finding when it last analyzed the issue in detail.40

97. Taking the conclusion of these studies as a starting point, it would appear worthwhile to revisit the issue of export profitability and to examine what developments in exports have underpinned the maintenance of external competitiveness in the face of the sustained escudo appreciation. In the process, an attempt could also be made to infer from these developments Portugal’s ability to adjust to the prospective changes in the external environment.

98. The next two sections will explore these topics. Section B reviews the literature on the appreciation of the equilibrium exchange rate and attempts to measure the direct impact of the real appreciation of the exchange rate on the profitability of the exporting sector. Section C studies the main structural changes in the composition of Portugal’s exports. It also attempts to evaluate the implications of the concentration of structure and geographical destination of exports (drawing inter alia on so-called “gravity” models), and gauge the relative competitive advantages of Portugal vis-à-vis Central and Eastern European candidates for admission to the EU.

99. The investigation underscores that standard real exchange rate indices have tended to overstate the loss of competitiveness suffered by Portugal’s exporting firms in 1986-92. It also finds that, while Portugal’s export base has widened and moved upward on the value-added ladder, scope for further geographical and composition diversification persists. Finally, it highlights that taking up such scope is all the more important in view of the finding that the candidates for the next round of EU enlargement appear to have comparative advantages vis-à-vis Europe similar to those of Portugal.

B. The Real Appreciation of the Escudo: an Equilibrium Phenomenon?

100. Conventional indices of the real effective exchange rate (REER) show a significant appreciation of the escudo in the decade since EU accession in 1986. Both the ULC and the CPI-based REER indices follow a similar trend, albeit with differences in magnitude (Figure 11, top panel). They show a sharp appreciation between 1988 and the first quarter of 1992, followed by a period of instability ensuing ERM entry in April 1992, and a greater stability of the real effective exchange rate since early 1995 (at a level virtually equivalent to that prevailing from mid-1992 to early 1993). These periods broadly reflect (i) the softening and eventual abandonment of the crawling peg regime in a context of relatively high wage and price inflation vis-à-vis main trading partners through early 1992; (ii) turbulence in the exchange rate mechanism and realignments of the escudo’s central parity in the context of the ERM crises of 1992-93; and (iii) more subdued conditions within the ERM since mid-1995, which have allowed a consistent pursuit of the Bank of Portugal’s objective of exchange rate stability and a consolidation of the disinflation process.

Figure 11.Portugal: Real Effective Exchange Rate and Terms of Trade


Sources: IMF, International Financial Statistics; information Notice System; World Economic Outlook; and staff calculations.

1/ Shift to nominal exchange rate peg.

2/ ERM entry.

3/ Central rate realignments of November 1992, May 1993, and March 1995.

The conclusions of previous studies

101. Several studies have estimated the escudo’s “fundamental exchange rate” drawing on the methodology developed by Williamson (1985), based on the principle that the equilibrium level of the exchange rate is one that promotes internal and external equilibrium simultaneously.41 The main conclusion of these studies has been that the improvement in the terms of trade (Figure 11, bottom panel), and the increase in both foreign investment (Freitas, 1992) and official transfers associated with EU membership (Luis, 1996) could explain a real appreciation of the escudo that would not be detrimental to the external balance. Comparing the estimated equilibrium exchange rate to the actual value of the real exchange rate, Freitas (1992) concluded that the escudo was still undervalued in 1990. A similar conclusion was reached by Manteu and Mello (1992), who estimated that a small gap remained to be closed by 1992. Finally, Luis (1996) estimated that the process of REER appreciation had led actual and equilibrium exchange rates to converge by 1994.

102. Another strand of the literature attempted to explore directly the potential correlation between economic growth and the level of the real exchange rate first suggested in Balassa (1964). Esteves (1993) used data from 1948-92 to test the hypothesis of cointegration between Portugal’s real GDP vis-à-vis developed economies and the real exchange rate. Having verified that the income of Portugal relative to that of the United Kingdom and the United States was cointegrated with the exchange rate of the escudo vis-à-vis pound sterling and the U.S. dollar (that is, moved together in the long run), the study concluded that the real appreciation of the escudo observed between 1987-92 was an equilibrating process. As such, the rise in the relative price of nontradable to tradable goods (evident in Figure 12) was likely to reflect faster gains in productivity in the latter sector, owing to the increase in the capital stock and the acceleration of economic growth following EU accession. In line with Balassa (1964), the study argued that it should thus not be interpreted as a signal of a deterioration in external competitiveness.

Figure 12.Portugal: Export Profitability


Sources: Portugal authorities; and staff calculations.

103. Finally, Rebelo (1992) and Gaspar and Pereira (1995) provided micro foundations for these empirical results. Rebelo (1992) developed an optimization model which differentiated between traded and nontraded goods and was calibrated to reproduce characteristics of the Portuguese economy. Through simulations, the study showed that increases in the capital stock, productivity gains in the tradable sector, and the pattern of government consumption observed in Portugal could—in equilibrium—explain a persistent differential between inflation in the nontradables and the tradables sectors of the order of 1 to 3 percent a year. Gaspar and Pereira (1995) used a similar approach to gauge the effects of official transfers on the equilibrium exchange rate (through the accumulation of capital, and in contrast to an increase in foreign debt). Their simulations indicated that the long-run real exchange rate (defined as the one prevailing at steady state after the year 2005) would be 8½ percent above the level that would prevail in the absence of such transfers.

104. In sum, although a few studies have dissented from the prevailing view that the appreciation of the escudo has been largely an equilibrium phenomenon (for example, Cunha and Machado, 1993),42 the bulk of the research has attributed the (equilibrium) behavior of the exchange rate to economic fundamentals. The staff had come to a similar conclusion when it last analyzed the issue in detail (SM/94/217, Supplement 1, 8/23/94). In particular, the staff noted that, consistently with the implication of the literature reviewed above (which postulated that if the escudo’s appreciation has been prevalently an equilibrium phenomenon, one would expect to find no persistent deterioration in Portugal’s current account), movements in the current account did not provide any signs of a serious erosion of competitiveness (the performance of the current account over the last decade is briefly reviewed in Box 1). In the latter study the staff also pointed to some shortcomings of the standard indices used in these assessments, noting that they tended to overstate the extent of real effective appreciation.

The evidence from direct indicators of export profitability

105. An aspect of the studies discussed above is that, while the “equilibrium” appreciation of the exchange rate rests largely on the assessment that productivity in the tradable sector outpaced the increase in labor costs in the sector, only indirect indications in this sense are usually provided. In fact, standard indices aimed at capturing the effects of the nominal appreciation on the export sector systematically show unit labor costs increasing much more rapidly than export prices in the late 1980s and early 1990s (Figure 12, middle panel). Taken at face value, these indices would suggest that export profit margins halved in the period—a phenomenon that is hard to reconcile with the almost doubling of total exports over the same time span. Two reasons for this apparent dichotomy can be identified. The first is related to the fact that the industrial production index used in the computation of most ULC indicators does not incorporate new firms. The second is that, by taking manufacturing as a whole, these indices do not differentiate sufficiently between tradable and nontradable sectors, while the “Balassa effect” and similar mechanisms postulate a growing productivity differential between the two sectors (as illustrated in Rebelo, 1992) that sustains an equilibrium real appreciation.

Box 1.The Performance of the Current Account in the Last Decade

The performance of the Portuguese current account in the period 1986-97, which can be divided in three major phases (a deterioration in 1986-88, virtually balanced positions in 1989-93, and the reoccurrence of deficits after 1994), supports the view that there has been no persistent deterioration following the escudo’s real appreciation through 1992.

The current account balance weakened appreciably in 1986–88, mostly as a consequence of the trade balance, whose deficit more than doubled in relation to GDP. The widening of the trade deficit was due mainly to the rapid growth of real imports, which averaged 22 percent a year in the period, more than offsetting the contemporary improvement in the terms of trade and relatively robust export growth. The rise in imports stemmed from the conjunction of trade liberalization and a strong economic expansion. Domestic absorption and import growth decelerated as from 1989. With the contribution of continued improvements in investment income, due to the cumulation of reserves, the current account strengthened and fluctuated around a balanced position between 1989–93.

Since 1994, the current account has shifted into deficit in a range of 1 to 2½ percent of GDP. This deterioration has been determined mostly by a sharper decrease in net private transfers—with also a negative contribution of services. Private remittances have been declining since the early 1980s, reflecting lower migration from Portugal. The more acute decrease of such inflows after 1993 may be linked to the weakness of economic activity in emigrant host countries, and to financial innovations that have diverted part of repatriated savings to financial instruments recorded in the nonmonetary financial account. The weakness of economic activity in Europe also has dampened tourism receipts, while the increase in the income of Portuguese households has been associated with greater expenditures on tourism and other services (preliminary data for 1997, however, suggest a strong pick up in receipts). The trade balance has remained broadly stable.

Movements in the current account can be decomposed into changes attributed to fluctuations in the exchange rate (a “price” effect) and changes reflecting the economic cycle (an “income” effect). The adjustment to “income” effects would reflect trade flows in the absence of output gaps in Portugal or its trade partners, based on Portugal’s estimated import elasticity to the domestic output gap (2.2) and the elasticity of exports with respect to income of main trading partners (3.5), Such cyclical adjustment, aimed at isolating the effect of the exchange rate on the current account (Figure 13), shows that the performance of 1989-93, when the current account hovered around balance, can be viewed as distinctly more positive in light of the relative output gaps of Portugal and its partners at the time (this performance was not fortuitous, rather reflecting a substantial increase in taxes and the adoption of credit ceilings, including for the purchase of durable goods, aimed at cooling down the economy). By 1993, the importance of the cyclical adjustment factor declined, but as the recovery in Portugal has out paced that in most partner countries, the adjusted current account deficit has again become smaller than the actual deficit.

In summary, most of the deterioration of Portugal’s current account in 1986-88 and 1994-97 can be explained by the removal of obstacles to imports associated with EU accession at a time of strong aggregate demand growth in the first instance, and the decline in net transfers and net service flows in the second episode. The overall stability of the trade balance also suggests that the significant appreciation of the escudo over the past decade has not had an adverse impact on Portugal’s competitiveness.

Figure 13.Portugal: Current Account 1/

(In percent of GDP)

Sources: Bank of Portugal; IMF, World Economic Outlook; and staff estimates.

1/ Shaded area indicates staff projections.

106. Indices based on a constant sample of firms underestimate the economy’s competitiveness to the extent that new firms are in general more productive, being more capital intensive and technologically advanced. While a halving of export profit margins (as shown in Figure 12) is consistent with the findings of Farinha and Mata (1996), who accompany existing firms for the period 1986-94,43 the magnitude of the phenomenon is considerably smaller when indicators that better capture the actual evolution of the economy are used. For instance, taking value added in the manufacturing sector (from the Bank of Portugal’s recently released “long series” of the national accounts) as a proxy for activity in the export sector, would point to a much lower increase in unit labor costs over the period, suggesting inter alia that the actual real appreciation of the escudo in 1986-92 was closer to 20 percent, instead of the 40 percent yielded by standard industrial production based indices.44

107. A further indication that the export sector was able to remain competitive in the face of the appreciation of the currency in the late 1980s can be found in Cabral (1996). This study explores the information in the balance sheets of some 2,000 firms with an exporting activity in 1986-92 (the number of firms increases from 1,471 to 2,139 in the period). It finds that, while exports from domestic firms were sensitive to changes in the exchange rate, those from foreign firms—whose participation in total exports increased by 25 percent in 1986-91—were not. The data indicates that unit labor costs in export firms grew in line with export prices (Figure 12, bottom panel). In fact, such data indicate that the profitability of export firms increased in the late 1980s, falling only in 1991-92 (by some 10 percent from their 1986 level).45 This body of micro data also points to a sharp increase in the capital intensity of exporting firms (particularly foreign-owned firms, whose number increased by 20 percent in the period), buttressing the hypothesis that capital accumulation was a major factor sustaining the equilibrium appreciation of the exchange rate.

108. In sum, direct indicators of competitiveness (e.g., export profitability), computed from data that attempt to reflect the substantial changes that occurred in the export sector in the period and more narrowly focused on exporting firms, would strengthen the conclusion reached by most of the literature regarding an equilibrium appreciation of the real exchange rate since EU accession.

C. Factors Underpinning the Equilibrium Appreciation and Future Prospects

109. The maintenance of external profitability in the face of the real appreciation of the escudo was underpinned by a major transformation of the external sector. Associated with the increase in capital intensity noted above, there was an important shift in the structure of exports toward higher value-added goods. This process is not yet complete, and may even accelerate as the economy meets new demands and becomes increasingly integrated in the global economy.46 In the following paragraphs this structural change, as well as some directions in which further changes may take place, are highlighted.

The changing structure of exports

110. The positive export performance since EU accession can be traced to a movement of exports toward higher value-added products. In the immediate aftermath of accession, exports of light manufacturing—notably clothing and footwear—increased by more than 50 percent (Figure 14, top panel), sustained by the effects of greater access to European markets and supplanting more traditional exports, such as agricultural products. These effects were countered, however, by the continued appreciation of the escudo in the later part of the 1980s. The steady rise in relative unit labor costs weighed increasingly on the profitability of such labor-intensive industries as clothing and footwear.47 Overall exports thus leveled off in the early 1990s, and fell marginally in 1993, in the face of the weakening of economic activity in the rest of Europe. This pause was, however, short-lived as new productive capacity came on stream following a peak in direct foreign investment in the early 1990s. In this more recent phase, export growth has been driven particularly by sales of machinery and transportation equipment, that doubled between 1994 and 1996 (Figure 14).

Figure 14.Portugal: Export Growth

(Volume, 1986=100)

Sources: Portugal authorities; IMF, World Economic Outlook; and staff calculations.

111. The structural change in exports since the mid-1980s is well-illustrated by the near doubling of the share of machinery and transportation equipment in total export values in the decade to 1996 (Figure 15).48 At the same time, important categories of traditional exports—wood, cork, and paper pulp (all classified among “crude materials”), agricultural products, and textiles49—saw their relative importance decline appreciably.

Figure 15.Portugal: Export Composition by Sector

(In percent of total exports)

Sources: IMF, Trade Analysis and Reporting System.

112. This structural shift in exports toward higher valued-added products, which has undoubtedly strengthened the country’s export base, has not however been fully accompanied by a diversification of geographical destination, which remains highly concentrated. In 1991, exports to EU12 countries accounted for three-quarters of the total, with exports to Germany alone accounting for about one-fifth. Moreover, the reliance on traditional markets has increased since then, with EU12 countries absorbing 80 percent of Portuguese exports in 1996. Such a concentration on relatively slow-growing European markets has led to a paradoxical situation: while market shares measured as inroads in traditional markets have increased by close to 15 percent since 1990, the share of Portuguese exports in world trade has actually declined slightly over the same period (Figure 14, bottom panel).

Geographic diversification: indications from a gravity model

113. The possible over-reliance on traditional markets can also be illustrated by the comparison of actual trade patterns with those predicted by a “gravity” model. Gravity models postulate that bilateral trade between two countries can be explained by each country’s size and income per capita, the distance between them, and other factors such as membership to trade blocs (see Appendix for a brief discussion of the model and its application to Portugal). Although in some cases the failure of actual trade flows to attain the levels predicted by the model is attributed to the existence of trade restrictions, the model can also shed light on potential markets that, for one reason or another, have not been fully tapped. In this context, gravity models, while originally used for analyzing aggregate trade flows (Helpman and Krugman, 1985; Baldwin, 1994), have proved useful in estimating potential levels of sectoral trade, for which differences in transportation costs and income elasticities may be significant (e.g.,Vittas and Mauro, 1997).

114. The comparison between “potential” (that is, predicted by a gravity model) and actual exports to a number of countries (Figure 16) highlights two prominent features of Portugal’s external trade performance. First, considering the country’s income level (with its attendant labor costs) and geographical location, the potential for substantial export growth in sectors such as textiles,50 clothing, and footwear appears very limited or even negative. Second, in sectors with greater value added and significant scope for growth, some markets appear close to saturation.

Figure 16.Portugal: Potential Exports 1/

(In million of US dollars)

Sources: TARS and Staff calculations.

1/ Gray areas indicate the amount of exports of Portugal to respective countries in 1995. White areas indicate the slack vis-à-vis potential (i.e., predicted by the gravity model) export flow. A negative white area indicates that actual trade exceeds that predicted as “potential”.

115. Although these results should be interpreted with caution, as the model does not fully capture some specificities of trade flows,51 they do indicate a need for greater geographical diversification, which is indeed an objective of the government’s trade policy. Interestingly, the model suggests that while large markets such as the United States generally offer ample opportunities for significant increases in trade, small economies also hold potentially valuable trade opportunities in certain sectors.52

Looking ahead and eastward: challenges and opportunities

116. The potential drawbacks of the geographical concentration of exports to a few EU countries acquires greater and more immediate relevance in view of the prospective enlargement of the EU to Central and Eastern European countries. Among the candidates for early accession to the EU are the Czech Republic, Hungary, Poland, and the Slovak Republic, which (with the exception of the Slovak Republic) are more populous than Portugal and closer to Portugal’s most important export market. In addition, from a sectoral standpoint, these countries as a group (hereafter called CEEC-4) appear to have similar comparative strengths to those of Portugal.53

117. Indicators of revealed comparative advantages (RCA)54 between Portugal and seven EU countries55 (hereafter called EU-7) show that Portugal has enjoyed a comparative advantage vis-à-vis these countries in the sectors of crude materials and miscellaneous manufactured products, such as clothing and footwear (Figure 17). The main trends captured by the RCAs have been the deterioration in Portugal’s position in chemicals and miscellaneous manufactured products, and the improvement in machinery and transport equipment.56

Figure 17.Portugal: Revealed Comparative Advantages

Sources: IMF, Trade Analysis and Reporting System

118. RCAs for the CEEC-4 vis-à-vis those EU-7 countries (Figure 17, bottom panel) show that the CEEC-4 have enjoyed comparative advantages in the areas of food products, crude materials, mineral fuels (e.g., coal), miscellaneous manufactured goods, and, to a lesser degree, basic manufactures (which include mainly textiles, paper, cement, and steel). The main characteristics of the evolution of the CEEC-4 RCAs have been the significant erosion of the advantage held in food products, mineral fuels, and manufactured goods, and the improvement in the area of machinery and transport equipment (see Vittas and Mauro, 1997 for an account of possible causes for this erosion).

119. The similarities between the revealed comparative advantages of Portugal and the CEEC-4 vis-à-vis EU-7 countries (with the notable exception of food and fuels) thus constitute a strong case for pursuing a greater geographical diversification of Portugal’s exports. This is especially the case given that proximity to Germany would appear to favor CEEC-4 exports once they have full access to the European single market. As noted by Luz (1997), this diversification effort should, however, embrace the CEEC-4 themselves, in relation to whom Portugal’s revealed comparative advantages in numerous sectors have increased substantially, following market liberalization in those countries. Figure 18 indicates that in the case of basic manufactures Portugal’s disadvantage was halved, while it was reversed in the case of machinery and transportation goods; Portugal has also retained a comparative advantage in clothing. Also based on a gravity model, Luz concludes that, if taken up by Portugal, the potential for exporting to the more than 60 million consumers in these fast growing CEEC-4 economies could in good measure offset the competitive loss of Portugal vis-à-vis German markets entailed by CEEC-4 accession to the EU.

Figure 18.Portugal: Revealed Comparative Advantage Vis-à-Vis CEEC4

Sources: IMF, Trade Analysis and Reporting System.

D. Conclusions

120. Portugal’s external competitiveness has been preserved despite the real appreciation of the exchange rate in the late 1980s and early 1990s. This adjustment has been underpinned by a change in the structure of exports toward higher value-added products, accompanied by an increase in the capital intensity of the export sector and an attendant increase in labor productivity in the sector. Reflecting the impact of EU accession, the share of exports to other EU countries has become dominant. Looking forward, it would appear that the ongoing diversification in the structure of exports should be accompanied by greater diversification of geographical destination. In particular, “potential” export flows to the United States predicted by a gravity model are well above their actual levels. That model also suggests that numerous small economies offer the potential for absorbing a greater share of Portugal’s exports. Finally, the study highlights that taking up such scope is all the more important in view of the finding that the candidates for the next round of EU enlargement appear to have comparative advantages vis-à-vis Europe similar to those of Portugal.

APPENDIX: A Gravity Model for Portugal

121. This appendix briefly describes the gravity model estimated for Portugal. It provides a summary of the model, and the main estimates.

122. Gravity models attempt to explain the importance of geographic proximity and industrial structure in international trade flows. Theoretical foundations for this approach were provided by Helpman and Krugman (1985), while the use of this class of model was popularized in Baldwin (1994). Essentially, gravity models assume that bilateral trade between two countries can be explained by each country’s size and income per capita, the distance between them, and other specific variables, notably the membership in trade blocs.

123. While originally computed for aggregate trade, the use of these models has been extended to analyze sectoral trade (Vittas and Mauro, 1997). As noted there, there is no a priori reason to believe that the extent to which income or distance affect trade flows is the same for all commodity groups. The basic specification (Baldwin, 1994) is the following:

where X is the log of export flows, Y/P is the log of per capita output, P is the log of the population, D is the log of the distance between the two countries, and the dummy variables Border and Trbloc indicate whether the two countries are adjacent and belong to a common trade bloc.

124. The source of data on trade flows is the United Nations trade statistics system (TARS), while GDP and population statistics were drawn from the International Financial Statistics. Distances were compiled from a variety of sources. The sample used for the estimation covered 1988-95. For the purpose of the estimation, averages over the period were used (so as to neutralize cyclical effects, although using subsamples did not change results dramatically). The number of observations in each sectoral panel (i.e., without considering the longitudinal dimension) varied between 300 and 600. Although the sample comprised trade among 34 countries, estimates were based on trade among the 18 largest economies, under the assumption that these flows would reflect the “potential” trade among diversified free economies.

125. The following table gives the parameter estimates obtained through ordinary least squares:

Portugal: Parameter Estimates for the Gravity Model
ChemicalsR squared = 0.69
Estimated Coefficient−2.780.540.430.590.56−0.600.180.11
Standard Error1.
Machinery and Transportation EquipmentR squared = 0.72
Estimated Coefficient−7.290.750.640.780.66−0.520.270.31
Standard Error1.
Transportation EquipmentR squared = 0.67
Estimated Coefficient−11.070.880.771.000.62−0.620.360.53
Standard Error1.600.
PaperR squared = 0.08
Estimated Coefficient4.−0.230.19−0.22
Standard Error2.
Iron and SteelR squared = 0.65
Estimated Coefficient−−0.500.420.18
Standard Error1.
TextilesR squared = 0.75
Estimated Coefficient−4.840.860.591.180.66−1.47−0.16−0.12
Standard Error2.320.
ClothingR squared = 0.34
Estimated Coefficient1.50−0.240.520.300.34−0.290.650.34
Standard Error1.800.
FootwearR squared = 0.57
Estimated Coefficient−13.140.381.250.660.58−0.740.930.31
Standard Error2.890.
Source: Staff Calculations
Source: Staff Calculations

    BalassaB.1964“The Purchasing Power Parity Doctrine: a Reappraisal,”Journal of Political EconomyVol. 72.

    BaldwinRichard1994“Toward an Integrated Europe”Center for Economic Policy Research (CEPR)London.

    CabralSonia1996“Comparative Behavior of Foreign and Domestic Firms in Portugal,”Working Paper No. 8/96Bank of Portugal.

    CoradoC.1995“Textiles and Clothing Trade with Central and Eastern Europe: Impact on Members of the EC”in European Union Trade with Central and Eastern Europe: Adjustments and Opportunitiesed. byR.Faini and R.PortesCEPR, London.

    • Search Google Scholar
    • Export Citation

    CoradoC.V.Benacek and W.Caban1995“Adjustment and Performance of the Textile and Clothing Industry in the Czech Republic, Poland, and Portugal,”CEPR Discussion Paper No. 1260London.

    • Search Google Scholar
    • Export Citation

    CunhaLuis C. and JoseA. F. Machado1993“Real Convergence and Real Appreciation,”mimeo Universidade Nova de Lisboa Working Paper.

    • Search Google Scholar
    • Export Citation

    EstevesP. S.1993“Crescimento Económico e Taxa de Câmbio Real: a Experiência Portuguesa (1948-1992)”Bank of Portugal Quarterly BulletinJanuary.

    • Search Google Scholar
    • Export Citation

    FarinhaLuisa and JoseMata1996“A Rendibilidade das Empresas na Indústria Portuguesa: 1986-1994”Economic BulletinBank of PortugalJanuary.

    • Search Google Scholar
    • Export Citation

    FreitasMiguel L.1992“Sobre a Futura Paridade do Escudo: Há Lugar para uma Apreciação Real?”mimeo Universidade Nova de Lisboa Working Paper No. 183.

    • Search Google Scholar
    • Export Citation

    GasparVitor and AlfredoPereira1995“An Intertemporal Analysis of Development Policies in the EU”Universidade Nova de Lisboa mimeo; forthcoming in the Journal of Policy Modeling.

    • Search Google Scholar
    • Export Citation

    HelpmanE. and PaulKrugman1985“Market Structure and Foreign Trade” (MIT PressCambridge, MA).

    KenenP.1969“The Theory of Optimum Currency Areas: An Eclectic View”in Monetary Problems of the International Economyed. byR.A.Mundell and A.K.Swoboda (Chicago University PressChicago, IL).

    • Search Google Scholar
    • Export Citation

    KrugmanPaul1993“Lessons of Massachusetts for EMU”in Adjustment and Growth in the European Monetary Unioned. byF.Torre and F.Giavazzi (Cambridge University PressCambridge, MA).

    • Search Google Scholar
    • Export Citation

    LuisJ. B.1996“A Taxa de Câmbio de Equilíbrio na Economia Portuguesa”Bank of Portugal, mimeo.

    LuzSilvia1997“Alargamento da União Europeia e Comércio Externo Portuguesa”Economic BulletinBank of PortugalMarch.

    ManteuC. and A.S.Mello1992“Taxa de Câmbio de Equilíbrio Fundamental”Bank of PortugalQuarterly BulletinJanuary.

    RebeloSergio1992“Inflation in Fixed Exchange Rate Regimes: The Recent Portuguese Experience,”Bank of PortugalWorking Paper No. 11-92.

    • Search Google Scholar
    • Export Citation

    VittasHari and PaoloMauro1997“Potential Trade with Core and Periphery: Industry Differences in Trade Patterns”in Europe’s Economy Looks East: Implications for Germany and the European Unioned. byStanleyBlack (Cambridge University PressCambridge, MA).

    • Search Google Scholar
    • Export Citation

    WilliamsonJ.1985“The”Exchange Rate System”Institute for International EconomicsWashington, D.C.


Prepared by Joaquim Levy and Claudio Paiva.


Financial innovation has created new investment opportunities for both residents and emigrants. Part of the decline in private transfers, for instance, is attributed to a more widespread use of financial instruments, with transfers to emigrants’ savings to Portugal increasingly being recorded in the nonmonetary financial account of the balance of payments, instead of in the current account. Financial innovation and free movements of capital have also favored an increase in foreign asset holdings by residents. The income flows associated to these foreign assets are not, however, well captured in the financial statistics used to compile the current account.


“The Real Exchange Rates and Competitiveness in Portugal,” (SM/94/217, Supplement 1, 8/23/94).


In practical terms, these studies generally estimate the exchange rate that would equate the current account balance at positions of full employment (that is, internal balance) to the balance of “structural” or “underlying” capital flows (that is, capital flows excluding transactions driven by short-term interest rate differentials and exchange rate expectations).


Cunha and Machado (1993) also used cointegration analysis to determine whether the behavior of the real exchange rate observed in Portugal (and Spain) between 1960–92 shared a common trend with (and can be attributed to) the path of the respective capital stocks and population growth vis-à-vis the rest of Europe. From the results obtained for Portugal (cointegration is observed for the period 1960–88 but not for 1960–92), the authors concluded that in contrast to the peseta, the escudo appreciation against the ECU between 1989–92 could not be explained by real factors alone. Instead it was interpreted as possibly being “the result of a sluggish domestic adjustment to a very stiff anti-inflationary monetary policy,” stemming from less-than-expected disinflation in face of nominal exchange rate stability.


Farinha and Mata (1996) indicate that the ratio of operational income to sales for exporting firms declined from around 8 percent in 1986–89 to some 4 percent in 1991–93. They noted that using different variables (such as gross or net profits) did not alter the thrust of the results. Curiously, when looking at sectors, instead of at individual firms, Farinha and Mata found that profitability margins in “more exported-geared” sectors have not declined in the period, further suggesting that, when the sample comprises new firms, profitability developments appear decidedly more favorable.


This conclusion does not take into account similar adjustments in partner-country ULCs, under the assumption that the need for such an adjustment was more pronounced in Portugal in view of the sharp increase in foreign investment and the creation of new firms in the period under consideration.


The increase in profitability in the late 1980s was driven by the higher profitability of foreign firms, while the decline in 1991–92 was provoked by a 15 percent fall in the profitability of domestic firms (which can probably be traced to the textile and clothing sectors, as suggested by the extensive restructuring undergone by these sectors since 1993). These figures were computed using data from Cabral (1996) and the assumption that the sample is representative of the export sector (i.e., that the volume of exports generated by firms in the sample was proportional to Portugal’s total exports in the period).


A diversified economy is also better prepared to participate in a monetary union (Kenen, 1969; and Krugman, 1993).


This phenomenon is consistent with the findings in Cabral (1996) pointing to a greater sensitivity of changes in the exchange rate on the part of less capital intensive (domestic) firms than on that of increasingly more capital intensive (foreign) firms.


The recent performance of transportation material exports has of course been driven largely by sales of AutoEuropa, a large car manufacturer partially owned by Volkswagen which started operating in 1995. Nonetheless, this sector had already shown considerable dynamism in earlier years, sustaining the diversification of the export base into higher value-added products.


The share of textiles (which are classified among basic manufactures) in total exports almost halved, to around 7 percent.


Corado (1995) and Corado, Benacek, and Caban (1995) have studied trade patterns between Portugal and the rest of the EU, most notably with respect to textiles and clothing. They highlight quality issues in intra-industry trade, noting that Portugal tends to import high quality intermediate and final goods from the rest of the EU, while exporting lower quality goods. This pattern is not unique to Portugal, however, being shared by most Central and Eastern European countries.


While the model does not fully capture specialization in market segments, such as luxury or design clothing, at an appropriate level of disaggregation, it probably captures intra-industry trade and specificities of individual sectors (e.g., that trade in the transportation sector is dominated by large transnational companies). On the other hand, when interpreting results, the “dynamic” aspect of potential trade should also be kept in mind. More specifically, as more disaggregate data are used, any step increase in production of a specific item may lead to export levels that are temporarily above “potential.” For instance, the “over” exporting of transportation equipment to Germany associated with AutoEuropa, while in part linked to the ownership of the plant, is likely to wane as the Portuguese economy grows.


The large gap between actual and “potential” exports of iron and steel, for its part, is traditionally attributed to trade restrictions stemming from quotas and other protectionist barriers.


In 1996, the combined GDP of the CEEC-4 was approximately US$250 billion, with an average GDP per capita of US$4,000. Exports have grown steadily, and the share of CEEC-4 exports in EU-7 total imports tripled between 1986 and 1995, although remaining relatively low (around 2 percent). Exports to the EU-7, which comprise about half of total exports, are composed mainly by manufactured goods (basic and light manufactured goods accounting for 50 percent of total exports, machinery for 25 percent), with foodstuff and chemicals accounting for 7–8 percent to total exports.


The indicator of revealed comparative advantage is given by RCAipc = (Xipc−Mipc)/(Xipc+Mipc), where Xipc (Mipc) stands for the Portuguese exports (imports) of product i to (from) the CEEC-4 in a given period. Note that RCA indicators show actual trade patterns which can be influenced by restrictive trade policies, temporary factors, and, in the case of the CEEC-4, distortions still remaining from the era of central planning. These indicators therefore do not necessarily reflect true longer-term comparative advantages, but can be used as an indication of which export sectors bear the greatest potential for expansion.


Belgium, France, Germany, Italy, Luxembourg, the Netherlands, Spain, and the United Kingdom.


Disaggregated data show that the deterioration in the RCA for basic manufactures has been driven primarily by the losses in the textiles and wood sectors, which could not be compensated by the improvements in all other major categories. The negative performance in miscellaneous manufactured products has been driven by the RCA decline in clothing and footwear. The improvement in the category machinery and transportation equipment can be attributed solely to the performance in transport equipment, indicating that the observed increase in the export of machinery was offset by imports under the same broad category (not surprisingly, in view of the extensive construction of new plants and retooling of old ones in recent years).

Other Resources Citing This Publication