Portugal: Selected Issues

A range of indicators point to a competitiveness gap of 10–20 percent with respect to euro area competitors. Closing the competitiveness gap will require an extended adjustment period, even with a jump in total factor productivity (TFP) growth and strong wage moderation. This paper reviews several factors that could help explain the boom and bust behavior of corporate investment. Investor sentiment will recover with the deepening of structural reforms, but high corporate debt level is likely to slow the pace of investment growth in the near future.

Abstract

A range of indicators point to a competitiveness gap of 10–20 percent with respect to euro area competitors. Closing the competitiveness gap will require an extended adjustment period, even with a jump in total factor productivity (TFP) growth and strong wage moderation. This paper reviews several factors that could help explain the boom and bust behavior of corporate investment. Investor sentiment will recover with the deepening of structural reforms, but high corporate debt level is likely to slow the pace of investment growth in the near future.

I. Portugal: How Large is the External Competitiveness Gap?1

A. Overview

1. Portugal’s persistent loss of market share and weak export growth in recent years have raised concerns about its external competitiveness. To some extent, the weakening of Portugal’s external position can be seen as a natural consequence of the dynamics that characterized the economy in recent years: the falling risk premium associated with Portugal’s entry into the euro area, and the resulting increase in investment and wealth, the jump in consumption, and the demand boom that characterized the Portuguese economy until the end of the 1990s, was associated with an appreciation of the real exchange rate, some loss of competitiveness, and a widening of the current account deficit. But other mechanisms were also at play: the pro-cyclical fiscal policy of recent years and rapidly rising relative unit labor cost growth, much above the euro area average, contributed to the emergence of a competitiveness gap. To look into competitiveness more directly, this note reviews aggregate measures of competitiveness, and examines disaggregated trade data to assess recent changes in the competitiveness of the Portuguese economy.

2. Assessing Portugal’s relative competitiveness position is not straightforward, however, as different indicators suggest different interpretations. Aggregate broad-based measures of competitiveness—such as the World Economic Forum’s (WEF) and the International Institute for Management (IMD)’s international competitiveness rankings—which attempt to look beyond economic performance to consider economies’ official sectors, business efficiency, and infrastructure quality—provide somewhat divergent assessments of Portugal’s degree of competitiveness. In the WEF’s global competitiveness ranking for 2006, Portugal was placed thirty-fourth out of 125 countries, while under the IMD’s methodology, the Portuguese economy ranked forty-fifth out of 60 countries in the 2005 listing. These measures encompass a number of subjective elements, however, and more importantly, take a broader view of competitiveness than that commonly used to measure external performance.

3. While the various common methods of assessing competitiveness applied in this chapter are subject to known shortcomings, on balance they suggest a substantial competitiveness gap existed at the end of 2005. More traditional aggregate measures of competitiveness show a consistent and substantial decline in competitiveness (Section B): ULC-based real appreciation since the mid 1990s—benchmark years for Portugal—reached 10 percent in 2005, while Portuguese unit labor costs in the manufacturing sector rose some 20 percentage points faster than those of competitors in the euro area in the same period. The ratio of wage cost per employee to value added suggests that Portugal’s cost advantage of the mid-1990s has disappeared in relation to Spain and Italy. In the meantime, the gap between export profit margins in Portugal and the average for the euro area widened some 15 percentage points in the last 10 years, in favor of the euro area. As a complement to these measures, this chapter reviews macro model-based and/or econometric estimates of the equilibrium exchange rate, all of which point to a significant loss of competitiveness in recent years, with a competitiveness gap in the range of 10-20 percent by the end of 2005.

4. Export performance has reflected the loss in competitiveness. Applying constant market share (CMS) analysis to Portuguese exports over the period 1992-2004 suggests the bulk of export market loss in the 1990s was associated with deteriorating competitiveness (Section C). The CMS analysis, based on the value of exports, suggests market losses moderated since 1998, which may reflect a substantial compression of export margin, with a resulting negative impact on investment and employment in the tradables sector. Data on market share in volume terms, computed as the real growth of exports versus the weighted growth of import volumes in the main destination markets, however, suggests steady losses since the mid-1990s, particularly in the last two years.

B. Aggregate Measures of Competitiveness

Real effective exchange rate measures

5. Portugal has experienced significant CPI-based and ULC-based real appreciation since the mid 1990s. This was the case against both euro-area and non-euro-area competitors. This reflected a significant rise in unit labor costs and the impact of the euro’s sharp appreciation in 2002 and 2003. Most of the appreciation took place between 2000 and 2005, when Portuguese unit labor costs in the manufacturing sector rose 6.7 percent faster than for euro-area competitors and almost 10 percent faster than for competitors in the rest of the world (Table 1 and Figure 1). In 2005, unit labor costs in Portugal were basically flat, caused by wage moderation and a cyclical recovery of labor productivity growth. This ended previous years of worsening cost competitiveness against its competitors (Table 2).

Table 1.

Unit Labor Costs, Manufacturing, 1998-2005

(Annual changes in percent)

article image
Sources: Ameco database, OECD, and Fund staff calculations.
Figure 1.
Figure 1.

Manufacturing Unit Labor Cost (1995-2005)

Citation: IMF Staff Country Reports 2006, 386; 10.5089/9781451832228.002.A001

Table 2.

Share of Portuguese Exports in the World Market at SITC 2-digit

(In percent)

article image
Source: UN COMTRADE, and IMF staff calculation.
Figure 2.
Figure 2.

Real Effective Exchange Rate – CPI based (1995-2005)

Citation: IMF Staff Country Reports 2006, 386; 10.5089/9781451832228.002.A001

Source: OECD Analytical Database for ULC and International Financial Statistics (IMF) for REER.

6. The RER analysis is useful to assess changes in competitiveness, but it provides limited insight on the level of competitiveness and only if an equilibrium base period can be identified. For Portugal, the mid-1990s can be associated as “benchmark” years—based on current account and export developments. However, with changes in the quality and composition of production, entry into the euro area, and a large margin of uncertainty about the extent of any disequilibrium, even in the mid-1990s, these developments do not allow us to make clear statements about Portugal’s current competitiveness gap. Another caveat is that comparator countries are weighted by actual trade shares, with possibly too little weight on actual and potential third country competitors.2

Profit share indicators

7. The ratio of wage costs per employee to value added (in current prices) per person in manufacturing provides a measure of relative profit shares in the tradables-intensive sector of the economy.3 Alternatively, we also calculate the export margin by dividing the deflator of exports of goods by the unit labor cost in manufacturing. This measures improves on ULC-based REERs by taking into account variations across countries in the price of tradable output/exports (Lipschitz and McDonald, 1991). Nevertheless several caveats must be borne in mind. First, relative profit shares in manufacturing are not a good guide to differences in the rate of return on capital if there are significant differences in production technology. Comparisons of profit shares between countries at roughly similar stages of development should be more meaningful, although even here different product mixes can distort level comparisons. Analysis of changes in relative profit shares are meaningful because changes in production technology typically occur slowly. Second, the aggregate indicators could hide large differences in profit shares within the manufacturing industry.

8. The export margin shows a steady and gradual decline in Portugal (Figure 3) while the ratio of wage cost per employee to value added suggests that Portugal’s cost advantage of the mid-1990s has disappeared in relation to Spain and Italy (Figure 4). To the extent that comparisons of the levels of these measures are meaningful (that is, production technologies are similar), 2005 data suggest that since 1995 the gap between profit shares in Portugal and, for example, Germany, widened some 10 percentage points, as the share of wage costs in

Figure 3.
Figure 3.

Export Margin 1/

Citation: IMF Staff Country Reports 2006, 386; 10.5089/9781451832228.002.A001

1/ The Exports Margin is calculated dividing the deflator of exports of goods by the unit labor cost in manufacturing.Source: OECD Analytical Database, INE and European Commission.
Figure 4.
Figure 4.

Ratios of Wage Costs to Value Added, 1995-2005 1/

Citation: IMF Staff Country Reports 2006, 386; 10.5089/9781451832228.002.A001

1/ Wage bill per employee in manufacturing, as ratio of value added per person employed.Sources: OECD, STAN Database; OECD, Analytical Database; and IMF staff estimates.

PPP exchange rates

9. In contrast to REERs, the ratio between the actual exchange rate and the PPP exchange rate aims to assess levels of current exchange rates against their long-term equilibria. The PPP exchange rate compares the cost (in national currency) of a similar basket of goods (typically that of GDP) in two countries. For countries at close to the same level of development, ratios of the market rate to the PPP rate above one indicate overvaluation and below one undervaluation. For such countries, this is a particularly powerful tool because it measures over or undervaluation directly, rather than indirectly via the presumption that any change in the real value of a currency is a movement toward or away from a static equilibrium.

10. Using the PPP exchange rate ratio to analyze exchange rates of countries at different stages of development is more complicated. Countries with lower GDP per capita have lower wages in the service (or nontradable) sector and therefore lower prices in this sector expressed in a common currency—the ratio of the market rate to the PPP rate should be below one and rising (De Broeck and Sløck (2001)).4 In Portugal, the PPP exchange rate ratio rose as convergence proceeded. At end-2005, the actual PPP exchange ratio was 76 percent of that of the euro area. Using the PPP exchange rate ratio consistent with Portugal’s GDP per capita as a norm, the actual Portuguese PPP exchange rate ratio was relatively high in light of its GDP per capita. Also, Portugal’s relative position with respect to Germany or France deteriorated as the ratio of the market rate to the PPP rate in these countries declined.

Figure 5.
Figure 5.

PPP Exchange Rate Ratio and GDP per Capita, 1998; 2005 in bold

Citation: IMF Staff Country Reports 2006, 386; 10.5089/9781451832228.002.A001

1/ Both relative to the euro area. A higher ratio indicates a more appreciated market rate.2/ Euro area average = 100.Sources: OECD; IMF, World Economic Outlook; and IMF staff estimates.

Macro model-based and/or econometric estimates of equilibrium exchange rate

11. The macroeconomic balance approach and related calculations of the fundamental equilibrium exchange rate (FEER) are based on a structural model of the economy, focusing on trade equations.5 Using the methodology in IMF (2006),6 the macro approach compares the underlying external current account with a norm or target. The underlying current account is derived by adjusting the actual current account for “transitory” elements, including the cyclical position and the impact of (all) past real exchange rate changes. The norm is derived from medium-term savings and investment balances or from current account positions needed to achieve a certain net foreign asset position. The gap between the underlying current account and the norm is then mapped into a gap between the actual and equilibrium exchange rate.7 In the case of Portugal, the misalignment implied by the C/A norm (a deficit of 3 percent) yields a misalignment of 7 to 10 percent. The key caveat is that this approach assumes that excess current account deficits are due solely to misalignment of relative prices. In Portugal, however, factors including consumption smoothing, volatile and bulky capital flows, and structural changes in savings behavior can produce temporary large current account deficits not due to a misaligned exchange rate.

12. Recent studies have calculated some concept of equilibrium exchange rate for Portugal:

  • Smidkova and Bulir (2004) use a model of fundamental real exchange rates (for a discussion, see Smidkova, Barrell and Holland, 2002) based on empirically estimated trade equations that relate exports and imports to the real exchange rate, the terms of trade, and domestic and foreign economic activity, with a view to measuring how far real exchange rates are from values corresponding to their economic fundamentals. They define the external balance in terms or stocks rather than flows. For Portugal, they find for the period 1992-2003—covering seven years of preparing for euro adoption as well as a brief post-adoption period—that the escudo remained in line with economic fundamentals until 1999 albeit close to overvaluation. However, they also find that by the end of 2003, the euro was some 10-20 percent too strong in real terms for Portugal.

  • Using the methodology in IMF (2006),8 the equilibrium exchange rate can be derived from reduced-form panel cointegration regressions, relating the real effective exchange rate to a set of underlying fundamentals.9 The dataset used includes 48 industrial countries and emerging markets, and covers the period 1980-2004. The real exchange rate misalignment for Portugal as of March 2006 is estimated to range from 8 percent to 11.3 percent, for the 2005 and 2011 fundamentals, respectively. The main shortcomings of such studies are the large estimation errors and the fact that results are contingent on the assumptions of particular models.

13. Given the theoretical and practical difficulties associated with estimating equilibrium real exchange rates, we calculated, alternatively, the improvement in competitiveness (as measured by the real effective exchange rate) required to achieve a trade deficit that would stabilize net external liabilities close to their current level. The misalignment estimated is based on the gap between the projected current account deficit and the current account deficit that stabilizes NFA at 2005 levels:

cas=gt+πt(1+gt)(1+πt)bs

where cas is the NFA-stabilizing current account/GDP; g is real GDP growth rate; π is the GDP inflation rate; and, bs is the target NFA/GDP. The estimate assumes a current account deficit consistent with a stable NFA position. The midpoint REER misalignment estimated is 14.5 percent (within a range of 12.4 to 20.6 percent) assuming trade elasticities of 0.7 for exports and 0.92 for imports. 10

C. Constant Market Share Analysis

14. In recent years, Portuguese export performance, in volume terms, has reflected the loss in competitiveness. Real export growth (goods) averaged 2.6 percent in 2001-04, at a time when the growth in the Portuguese export market was averaging above 4.4 percent annually according to WEO data. Real export growth (goods) in 2005 was about 1.6 percent, compared to 6.6 percent market growth.

15. Based on the UN’s COMTRADE trade database, the Portuguese share of the world export market was halved since the early 1990s (Table 2). Most of the decline occurred during the 1990s, with only a slight decline since 1998. However, the world market shares reflect the pattern of specialization in terms of either the country of destination or in the particular commodity bundle exported. To shed light on the relative contributions of these factors, in what follows we disaggregate the trade flows by applying constant market share analysis.

Constant market share analysis (CMS)

16. This approach can be expressed by the following equation:

X1X0=rΣiXi0+Σi(rir)Xi0+ΣiΣj(rijri)Xij0+ΣiΣj(Xij1Xij0rijXij0)

where

Xt=ΣiΣjXijt=ΣiXit,t=1,0

and

  • Xijt = the value of Portuguese export of commodity i to market j at time t,

  • r = the rate of growth of world exports,

  • ri = the rate of growth of world exports of commodity i,

  • rij = the rate of growth of world exports of commodity i in market j.

17. This approach entails decomposing the change in Portuguese exports between any two periods into four effects:

  • The global market growth effect (first term). This indicates the part of the export growth that is due to the expansion of the overall world trade. The magnitude of this effect shows the potential growth of the Portuguese exports when its share of world export market is kept constant.

  • The commodity composition effect (second term). This is the weighted sum of values of exports of different commodities. The weights are the deviations of the growth rates of individual commodity exports from the growth rate of the aggregate world exports. For instance, the commodity composition effect would be negative if Portugal had concentrated its exports on commodities with relatively slow global growth.

  • The market distribution effect (third term). This measures the change in exports due to market distribution and depends on trade policy and income growth of the countries where the Portuguese exports are destined. In general, this effect would be positive if Portuguese exports had gone to countries where demand growth was faster than the global average.

  • The competitiveness effect (fourth term). This residual term can be used as a measure of export competitiveness, the gain or loss in export growth that cannot be attributed to global growth, growth or trade partners or growth in demand for the products in which Portugal specializes.

18. The decomposition highlights that a stable share of world export markets does not imply stable competitiveness. Instead, a constant share in exports—after adjusting for the commodity and market effects—is equivalent to unchanged competitiveness. However, changes in trade policy can also be at play. Moreover, it would be preferable to do the analysis in volume terms, but data are not available. These caveats suggest caution in interpreting the results.

19. In applying the CMS approach to the Portuguese exports over the period 1992-2004, exports at two digits were regrouped into nine major commodities and various export markets. The nine commodities are: food (0-24), mineral fuels (25-27), chemicals (28-38), raw material (39-49), textile and appeal (50-71), metals (72-83), manufacturing goods (8485), transportation equipments (86-89), and other manufacturing goods (90-97). The analysis looks at the following key Portuguese export markets: 1) Spain; 2) Germany; 3) France; 4) Italy; 5) United Kingdom; 6) United States; 7) euro area, excluding Germany, France, Spain, and Italy; and 8) the rest of world.

20. The results, suggest the bulk of export market loss was associated with deteriorating competitiveness. The total export loss associated with weak competitiveness was some 14 percent of Portuguese exports in the period 1992-2004. Since adoption of the euro, the loss of market share continued, albeit at a more moderate pace (Table 3). Such moderation has been partly achieved through a substantial compression of export margin, as discussed above.11 While such compression has prevented a faster loss of export market shares, it is likely to have had a negative impact on investment and employment in the tradables sector, a key driving factor behind Portugal’s low growth. It is also clear that reductions in export margins cannot be sustained indefinitely.

Table 3.

Portugal: CMS Analysis of Exports Changes

(Value in billion US dollars; unless otherwise indicated)

article image

Based on the commodity composition of exports as of 1998.

Source: IMF Staff estimates.

D. Concluding Remarks

21. A range of indicators point to a competitiveness gap on the order of 10-20 percent relative to euro-area competitors (text table):

Summary Competitiveness Assessment, 2005 (in percent, unless otherwise noted)

article image
Sources: National authorities; Eurostat; OECD; and IMF staff estimates.

The benchmark year is when the exchange rate was considered to be appropriately valued considering factors including the size of the current account deficit, export growth, and GDP growth.

ULC-RER, unit-labor-cost-based real exchange rate.

Ratio between the market exchange rate and the PPP exchange rate (both relative to euro area).

Norm is the PPP exchange rate ratio consistent with a country’s GDP per capita.

ERER Approach, using 2011 fundamentals.

The implicit trade elasticities vary from 0.13 to 0.30, according to the degree of openess.

As of end-2005.

Export loss due to competitiveness gap, in percent of total exports (1992-2004).

While part of this gap can be seen as a natural consequence of the dynamics that characterized the Portuguese economy in recent years, other mechanisms were at play. It is not surprising that the falling risk premium associated with Portugal’s entry into the euro area, and the resulting increase in investment and wealth, the jump in consumption and the demand boom that characterized the Portuguese economy until the end of the 1990s, was associated with an appreciation of the real exchange rate, some loss of competitiveness, and a widening of the current account deficit. But, the pro-cyclical fiscal policy of the late 1990s and rapidly rising relative unit labor cost growth since the mid-1990s, much above the euro area average, are key factors that help explain the competitiveness gap.

22. Closing the competitiveness gap will likely require an extended adjustment period, even with a jump in TFP growth and strong wage moderation. Assuming future ULC growth of 1 percent annually for the euro area (as in the April 2006 WEO projections), even if Portuguese ULCs were to fall at a rate of 2 percent annually it would still take three to seven years to close the competitiveness gap with the euro area. Moreover, if capital deepening were to contribute 1 percent annually to labor productivity growth over that period (slightly above the rate experienced in 2001-04), even an increase in annual TFP growth to 1 percent (the euro-area average for those years) would leave no room for nominal wage growth in Portugal under this scenario. While these calculations are simply illustrative and are subject to considerable uncertainty, they demonstrate the scale of the problem Portugal faces, and the critical role that reforms to boost productivity and promote wage moderation must play in restoring competitiveness.

References

  • Blanchard, Olivier, 2006,Adjustment with the Euro. The Difficult Case of Portugal,Working Paper No. 06-04 (Cambridge: Massachusetts Institute of Technology).

    • Search Google Scholar
    • Export Citation
  • Bulír, Aleš, and Kateřina Šmídková, 2005,Exchange rates in the new EU accession countries: What have we learned from the forerunners?,Economic Systems, pp. 16386.

    • Search Google Scholar
    • Export Citation
  • Doroodian, K., Chulho Jung, and Ahmet Yucel, 2002,Estimating the equilibrium real exchange rate: the case of Turkey,Applied Economics (Athens: Ohio University).

    • Search Google Scholar
    • Export Citation
  • Edwards, Sebastian, 1988,The Determination of Equilibrium Real Exchange Rate,UCLA Working Paper No. 508 (University of California at Los Angeles and National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • European Central Bank, 2005,The New EU Member States Convergence and Stability” (Frankfurt).

  • Feyzioğlu, Tarhan, 1997,Estimating the Equilibrium Real Exchange Rate: An Application to Finland,IMF Working Paper No. WP/97/109.

    • Search Google Scholar
    • Export Citation
  • Halpern, Lğszló, and Charles Wyplosz, 1995,Equilibrium Real Exchange Rates in Transition,Discussion Paper Series No. 1145 (London: Center for Economic Policy Research).

    • Search Google Scholar
    • Export Citation
  • Halpern, Lğszló, Charles Wyplosz, 1996,Real Exchange Rates and Exchange Rate Policy in Hungary,Discussion Paper Series No. 1366 (London: Center for Economic Policy Research).

    • Search Google Scholar
    • Export Citation
  • Hofman, David, Francisco Nadal De Simone, and Mark Walsh, 2005,The External Competitiveness of the Dutch Economy: A Short Note on Evidence from both Aggregate and Disaggregate Data,Selected Issues, Country Report No. 05/225, (Washington, DC: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Isard, Peter, and Hamid Faruqee, 1998,Exchange Rate Assessment Extension of the Macroeconomic Balance Approach,Occasional Paper No. 167, (Washington, DC: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Isard, Peter, Hamid Faruqee, G. Russell Kincaid, and Martin Fetherston, 2001,Methodology for Current Account and Exchange Rate Assessments,Occasional Paper No. 209 (Washington, DC: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Mongardini, Joannes, 1998,Estimating Egypt’s Equilibrium Real Exchange Rate,IMF Working Paper No. WP/98/5.

  • Pain, Nigel, Annabelle Mourougane, Frank Sédillot, and Laurence Le Fouler, 1999,The New OECD International Trade Model,OECD Working Paper No. 440.

    • Search Google Scholar
    • Export Citation
  • Soares Esteves, Paulo, and Carolina Reis, 2005,Competitividad Das Exportações Portuguesas: Uma Avaliação dos Pesos da Taxa de C´mbio Efectiva,Boletin Económico(Lisboa: Banco de Portugal).

    • Search Google Scholar
    • Export Citation
1

Prepared by Paulo Drummond (EUR).

2

For a discussion of this point, see “Competitividade das Exportacões Portuguesas: Uma Avaliacão dos Pesos das Taxas de Cambio Efectiva,” by Paulo Soares Esteves and Carolina Reis.

3

This measure is close to wage shares used by Lipschitz and McDonald. Here, however, productivity is calculated per person employed (including self-employed), but wage costs are calculated per employee (excluding self-employed). This avoids a bias due to—sometimes tax system related—differences in the importance of self-employment across countries.

4

For a sample of developing countries, they estimate the log of the PPP exchange rate ratio as a linear function of the log of PPP GDP per capita in dollars and find that an increase in PPP per capita GDP of one percent increases the exchange rate ratio by 0.41 percent.

5

This approach is discussed by Isard and Faruqee (1998) and Isard and others (2001).

6

“Methodology for CGER Exchange Assessments” (forthcoming).

7

A related approach is the estimation of the natural real exchange rate (NATREX). Based on more rigorous modeling of stock-flow interaction in a macroeconomic growth model, it makes a distinction between medium-term equilibrium (with external and internal balance) and long-run equilibrium (with net foreign debt constant and the capital stock at a steady state level).

8

“Methodology for CGER Exchange Assessments” (forthcoming).

9

Lagged net foreign assets to trade, productivity of tradables versus nontradables relative to trading partners, commodity terms of trade, government consumption to GDP ratio, and an index of trade restriction.

10

Isard and Faruqee (1998).

11

The market share analysis uses data in value terms and may differ from other export market analysis in volume terms.

Appendix I. Alternative Estimates of the Equilibrium Debt

1. In this Appendix, we present alternative estimates of the equilibrium levels of debt, by modeling the relationship between the level of debt and its two main determinants, the output level and the nominal interest rate. Firms incur debts to finance their operation, therefore the output level captures the demand for debt. The nominal interest rate reflects the cost of holding debt.

2. The upward trend in the debt figures is apparently driven by the development of the economy and the steady decline of the interest rates throughout the sample period (Figure A1). As can be seen, a break of the trend occurred possibly in the late 1990s. In the late 1990s, the interest rate continued its declining trend, but there was not a sharp drop. It is conceivable that, as the adoption of the euro approached, a number of financing options became open to Portugal, and firms’ expectations on future growth were revised, which caused a revision of the desired debt level. The jump of corporate debt in the period 1998–2000 is also due to a number of public investment projects that were carried out by private companies in the context of the public-private partnerships (for instance, in this period the building of a number of motorways without tolls, SCUTs, were launched) together with the domestic financing of FDI projects of Portuguese companies abroad and a number of mergers and acquisitions operations. Therefore, we present estimates based on specifications both including and excluding a dummy variable for the period after 1998.

Figure A1.
Figure A1.

Fitted Debt Based On Equation (A1)

Citation: IMF Staff Country Reports 2006, 386; 10.5089/9781451832228.002.A001

3. We have included the resulting deviations of the actual debt path from its equilibrium path in estimating the investment function (Equation (1)), and the results are similar to those obtained from using the Hodrick-Prescott detrending approach.

Debt equation assuming no trend break

4. We obtain the following relationship for debt (D), output (Y) and the nominal interest rate (i) by OLS:

log(D)=29.96(0.00)+3.35(0.00)log(Y)0.02i(0.03)error.(A1)

p-values are in parenthesis. Unit root tests suggest that all three variables are I(1). Therefore we conduct the Engle-Granger cointegration test on the residual and find that the three series are cointegrated. Figure A1 plots the debt series and the estimated equilibrium debt series. The results suggest that the actual path of corporate debt was below its equilibrium path in the mid to late 1990s and above the equilibrium path in the early 2000s.

Debt equation assuming trend break in 1998

5. As discussed above, there could be a possible trend break in the debt series around 1998. Next we re-estimate Equation (A1) by including a dummy variable for the period after 1998 (D1998), and the following relationship is obtained:

log(D)=24.10(0.00)+2.75(0.00)log(Y)0.01(0.19)i+0.34D(0.00)1998+error.(A2)

6. Since a trend break in the debt series is assumed, we need to modify the Dicky-Fuller tests to test the presence of a unit root in this series. Here we apply the methodology of testing unit roots with structural breaks as developed in Perron (1990). Again the hypothesis that the debt series is I(1) can not be rejected. The Engle-Granger cointegration test shows that the three series in Equation (A2) are cointegrated. Figure A2 plots the debt series and the estimated equilibrium debt series. The results are similar to those obtained in the case without the trend break, suggesting that the debt path was below its equilibrium path in the late 1990s and above the equilibrium path in the early 2000s.

Figure A2.
Figure A2.

Fitted Debt Based On Equation (A2)

Citation: IMF Staff Country Reports 2006, 386; 10.5089/9781451832228.002.A001

References

  • Constâncio, Vítor, 2005,European Monetary Integration and the Portuguese Case,in Carsten Detken, Vitor Gaspar and Gilles Noblet (eds.), The New EU Member States: Convergence and Stability, Third ECB Central Banking Conference, October 21-22, 2004, Frankfurt, European Central Bank.

    • Search Google Scholar
    • Export Citation
  • Fagan, Gabriel and Vítor Gaspar, 2005,Adjusting to the Euro Area: Some Issues Inspired by the Portuguese Experience,paper presented at a Conference organized by the ECB on “What effects is EMU having on the euro area and its member countries?” in Frankfurt am Main on June 16–17, 2005.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2004,Portugal: 2003 Article IV Consultation—Selected Issues,IMF Country Report No. 04/81.

  • Murgasova, Zuzana, 2005,Post-Transition Investment Behavior in Poland: A Sectoral Panel Analysis,IMF Working Paper WP/05/184.

  • Perron, Pierre, 1990,Testing for a Unit Root in a Time Series with a Changing Mean,Journal of Business & Economic Statistics, American Statistical Association, Vol. 8 (2), pages 15362, April.

    • Search Google Scholar
    • Export Citation
  • Ramaswamy, Ramana, 2000,Explaining the Slump in Japanese Business Investment,in Post-Bubble Blues: How Japan Responded to Asset Price Collapse, IMF, Washington, DC.

    • Search Google Scholar
    • Export Citation
12

Prepared by Yuan Xiao (EUR).

13

The argument should be qualified by the possibility of labor market rigidity. Declining profitability could also be interpreted as evidence for the lack of wage moderation during the period.

14

See International Monetary Fund (2004).

15

We have also attempted to estimate the equilibrium debt by modeling the relationship between debt and its two main determinants, output and the nominal interest rate. The implications for Equation (A1) are similar to the Hodrick-Prescott filter approach. For details, see the Appendix.

16

All data are obtained from the Bank of Portugal.

Portugal: Selected Issues
Author: International Monetary Fund