Monetary and Exchange Rate Policies of the Euro Area: Selected Issues

This paper examines monetary and exchange rate policies of the euro area. The paper reviews the European Central Bank’s definition of price stability, and examines the factors determining “the optimal rate of inflation” in the euro area. It reviews the benefits of price stability, including the reduction in the distortions of savings and investment behavior that stem from the interaction between nominal tax systems and inflation. It then goes on to evaluate arguments for maintaining a small positive inflation rate in the context of the euro area.

Abstract

This paper examines monetary and exchange rate policies of the euro area. The paper reviews the European Central Bank’s definition of price stability, and examines the factors determining “the optimal rate of inflation” in the euro area. It reviews the benefits of price stability, including the reduction in the distortions of savings and investment behavior that stem from the interaction between nominal tax systems and inflation. It then goes on to evaluate arguments for maintaining a small positive inflation rate in the context of the euro area.

IV. Euro Area Trade Flows and the Exchange Rate: How Much Disconnect?1

A. Introduction

1. Three stylized facts form the backdrop to this chapter:

  • The euro area’s real effective exchange rate—as measured by the Fund’s exchange rate index based on normalized unit labor costs in manufacturing—fell sharply during 1996-2001, the inverse mirror image of the U.S. dollar’s appreciation during the same period (Figure 1, left panel).

  • The area’s current account balance varied little during the same period, particularly when contrasted with the unprecedented increase in the U.S. external deficit (Figure 1, right panel).

  • As an accounting implication of the second stylized fact, at the global level the ballooning U.S. external deficit had its counterpart in improved current account positions outside the euro area, with offsetting improvements in external positions mainly occurring in the smaller advanced economies and the developing countries (Figure 1, right panel).2

Figure 1.
Figure 1.

Exchange Rates and Current Account Balances, 1995-2001

Citation: IMF Staff Country Reports 2002, 236; 10.5089/9781451812985.002.A004

Source: WEO, IMF.1/ Based on normalized unit labor costs in manufacturing, 1995=100.2/ In billions of U.S. dollars; euro area balance is sum of individual countries.3/ Not including global current account discrepancy.

2. Prima facie, the three stylized facts appear to fly in the face of conventional empirical wisdom on the link between exchange rates and trade flows. In this view, a depreciation should, all other things being held equal, be accompanied by an improvement in the current account position, albeit perhaps with considerable lags. And a widening external U.S. deficit accompanied by an appreciation of the U.S. dollar against other major currencies should at least in part be mirrored by rising external surpluses in the euro area and Japan. Confidence in this vintage of conventional wisdom on the link between current accounts and exchange rates received a major boost from the global adjustment experience of the 1980s when the U.S. dollar rose and fell sharply against the (at that time implicit) euro exchange rate while the changes in the area’s current account position conformed quite closely to the predictions of conventional thinking Figure 2.3

Figure 2.
Figure 2.

Exchange Rates and Current Account Balances, 1980-1989

Citation: IMF Staff Country Reports 2002, 236; 10.5089/9781451812985.002.A004

3. Why then did the euro area’s current account balance fail to improve in the face of a plunging euro during the second half of the 1990s? A priori, and abstracting from several measurement problems related to the euro area’s trade data that are briefly touched upon below, two alternative views seem plausible:

  • Conventional wisdom on the link between exchange rates and the current account is based on partial equilibrium thinking, encapsulated in the “all other things being held equal” clause. And there are many “other things” that could have changed in tandem with the exchange rate, and these other factors could have offset the exchange rate’s partial equilibrium impact on the current account.

  • Alternatively, conventional wisdom could be defective at its very core in that the presumed “connect” between exchange rates and trade volumes for a large open economy is weaker than postulated. More recently, this alternative view has gained traction, partly as a consequence of empirical research that has found little pass-through of exchange rates to consumer prices, particularly for the largest industrial countries;4 partly as a consequence of theoretical research that has begun to develop coherent models of full “disconnect” between exchange rate swings and the real economy, models that suggest that exchange rates among the large common currency areas may be highly volatile precisely because they have little effect on the real economy.5

4. This chapter takes a crude first pass at interpreting the link between the euro exchange rate and (non-oil) trade volumes and prices during 1997-2001. The chapter draws four conclusions:

  • Subject to several caveats regarding the quality of euro-area trade statistics and the liberal use of the “all other things are equal” clause, there appears to be strong evidence for disconnect between euro-area exchange rate changes and trade developments during 1997-2001. The disconnect is found, however, to be more pronounced for import volumes and prices, and most strikingly so for capital goods imports. At the same time, however, versions of the disconnect hypothesis based on local currency pricing (i.e. the assumption that import prices are preset in the importer’s rather than the exporter’s currency) are clearly incompatible with the recent behavior of euro-area trade data, both on the import and export side.

  • The chapter’s empirical findings suggest caution in using standard econometric model simulations for gauging the macroeconomic effects of a reversal of the euro’s depreciation. These models are usually calibrated to produce a close link between exchange rates and trade volumes for the euro area as a whole, reflecting the assumption of close-to-one pass-through of exchange rate changes to import prices at the individual member country level. Following the formation of EMU, however, the exchange rate-trade link for the area as a whole may no longer match up well the econometric relationships implicit in the simple aggregation of individual member countries’ trade equations (which are based on the pre-EMU environment).

  • The chapter’s empirical findings also suggest that assessments of whether the area’s prevailing real exchange rate is consistent with medium-term fundamentals based on the macroeconomic balance approach should be viewed with caution. The macroeconomic balance approach compares the “underlying current account balance” (i.e. the current account adjusted for output gaps and lagged exchange rate effects) at the present level of the real exchange rate with an estimated saving-investment balance norm. The deviation of the underlying current account from the saving-investment norm is then “translated” into a deviation of the prevailing exchange rate from its medium-run equilibrium level—this step assumes that there is conventional connect between exchange rates and trade flows. Thus, in a situation where the connect between exchange rates and trade flows is in fact weaker, the macroeconomic balance approach might incorrectly signal that because the underlying current account is close to the investment-saving norm the prevailing exchange rate must also be close to its medium-run equilibrium level.

  • The implications of the findings for interpreting the apparent global asymmetry in current account adjustments highlighted in Figure 1 are less clear.6 If exchange rate disconnect is more pervasive for exports to the major common currency areas than for exports to smaller open economies, it is plausible that a deterioration of the U.S. current account could be mostly mirrored by current account improvements outside the major common currency areas. More work on this issue using a multi-country model of trade flows, however, would be needed to draw firmer conclusions.

B. Why Has the Euro Been So Weak?

5. The euro’s depreciation since 1996 has defied a coherent and widely accepted explanation.7 While the euro’s nominal bilateral exchange rate fell during 1996-2001 against the currencies of its three main trading partners (United States, United Kingdom, and Japan), the decline was particularly pronounced with respect to the U.S. dollar (Figure 3, left panel). As regards the fundamental sources of the euro’s weakness, there is little evidence that the history of the 1980s repeated itself, with shifts in the stances of monetary or fiscal policies providing the key impulses for the U.S. dollar’s rise and the euro’s decline. Moreover, neither interest differentials nor current account developments appear to be of much help in understanding the observed exchange rate patterns.8 Most “stories” of the driving forces behind the euro’s decline have as a common theme, however, that the unprecedented runup in real asset prices (including equities, and residential and commercial real estate) in the United States relative to the euro area during the second half of the 1990s (Figure 3, right panel) should have played some role, although the fundamental reasons for these relative asset price developments remain controversial.

Figure 3.
Figure 3.

Exchange Rates and Asset Prices, 1980-2001

Citation: IMF Staff Country Reports 2002, 236; 10.5089/9781451812985.002.A004

Sources: Bank for International Settlements (BIS); WEO, IMF; own calculations.1/ Index, the 1980-2001 average is equal to 100.2/ Assets include equities, residential real estate, and commercial real estate (inflation-weighted index 1980=100); euro-area data are based on BIS data for Germany, France, Italy, Spain, the Netherlands, Belgium, and Finland.

6. Unfortunately, understanding the underlying sources of exchange rate movements is important for tracing their impact on the current account. From a general equilibrium perspective, the effect of the exchange rate on trade flows can be very sensitive to the ultimate source driving the changes in the exchange rate. For example, econometric model simulations of exchange rate shocks in Bryant et.al. (1988) suggest that an effective depreciation of the U.S. dollar by about 10 percent over the five-year period 1985-89 would have improved the U.S. current account after the five years by 60 billion U.S. dollars if the depreciation was driven by a U.S. fiscal contraction, by only 8 billion dollars if it was caused by a portfolio shift against the U.S. dollar, and by nil if the depreciation reflected a U.S. monetary expansion.9 Thus, the lack of a coherent account of the factors that caused the euro’s depreciation during 1997-2001 is a serious obstacle to understanding its impact on trade flows.

C. Euro-Area Trade Data: Measurement Issues

7. The poor state of the euro area’s external account statistics adds another layer of uncertainty to this chapter’s analysis. First, the area’s official data for the current account balance in Figure 1 were obtained by summing individual member countries’ current account balances. Because of biases in the measurement of intra-area trade flows, this sum-of-countries measure of the current balance overstates the area’s current account surplus by about ½ percent of GDP. In part this overstatement reflects underreporting of intra-area imports by smaller companies following the introduction of higher reporting thresholds in 1992 (intra-area exports are more likely to be reported because they tend to originate from larger companies that are not exempt from the reporting requirement). The official estimate of the area’s current account balance published by the ECB suggests that the area’s current account balance in 2001 was zero while the sum-of-countries measure yields a surplus of 0.4 percent of GDP (Figure 4, left panel).

Figure 4.
Figure 4.

Euro Area Current Account and Goods Trade Balance Data, 1995-2001 1/

Citation: IMF Staff Country Reports 2002, 236; 10.5089/9781451812985.002.A004

8. The data situation is even more diffuse when it comes to measures of the area’s trade balance for goods. Analysts here have the choice among three measures: (i) the sum-of-countries measure; (ii) the official measure published by the ECB as part of its balance of payments statistics, which tries to correct for the already mentioned intra-area biases in trade flow statistics; and (iii) a Eurostat measure based on extra-area flows in goods trade, which in turn is not directly comparable with the ECB measure because of differences in definitions, coverage, and the timing of recording. As can be seen from Figure 4 (right panel), the differences between the current account balances based on sum-of-member measure and the ECB measure (about ½ percent of GDP) reflect similar differences in the underlying trade balance measures.10

9. The empirical analysis in the rest of the chapter will be based on Eurostat’s trade statistics, largely because trade in goods is the major driver of the overall current account balance in the euro area, and only the Eurostat data provide detailed extra-area statistics for export and import volumes and prices including for different goods categories. A major disadvantage of the Eurostat data set is, however, that the price data are only provided as unit value indexes rather than true price indexes, entailing well-known measurement problems. Finally, because pass-through of exchange rate changes to oil prices is likely to be almost immediate and full,11 the focus of the empirical analysis throughout the rest of the chapter will be on non-oil trade flows and the non-oil trade balance as provided by Eurostat’s trade statistics (Figure 4, right panel).

D. Gauging Exchange Rate Disconnect

10. The nominal trade balance (TBt) as a percent of nominal GDP (YtPt) (where Yt stands for real GDP and Pt for the GDP deflator) is equal to the difference between exports (defined as the product of real exports (Xt) and export prices (PXt) and imports (defined as the product of real imports (Mt) and import prices (PMt)):

(1)TBt/(YtPt)=(Xt/Yt)(PXt/Pt)(Mt/Yt)(PM/tPt).

At this descriptive level, the change in the nominal trade balance (as a percent of GDP) can (approximately) be decomposed into trade balance effects due to changes in export and import volumes (“volume effects”) and trade balance effects due to changes in export and import prices (“terms of trade effects”):

(2)(TBt/(YtPt))(μxxtμmmt)+(μxpxtμmpmt),

where μx and μm are the nominal export-and import-GDP ratios, respectively, and small letters for trade variables denote their growth rates; the first term in parentheses on the right hand stands for the “volume effect” while the second term reflects the “terms of trade effect.”

11. The relative contributions of trade volumes and prices to the changes in the euro area’s non-oil trade balance (as a percent of GDP) are set out in Table 1. The non-oil trade balance during 1996-2001 declined by ¼ percent of GDP, with a small positive volume effect ½ percent of GDP) overcompensated by a negative terms of trade effect (¾ percent of GDP). As a consequence, the area’s non-oil trade balance during 1996-2001 was essentially flat.

Table 1.

Euro Area: Decomposition of Changes in Non-Oil Trade Balance, 1996-2001

article image
Sources: Eurostat; own calculations based on equation (2) in text; volume and terms of price may not add up because equation (2) holds approximately.

12. As the next step, consider an analytical decomposition of the change in the trade balance (as a percent of GDP) in terms of responses to the exchange rate (Et) and in terms of responses to “all other things”, where it is convenient to think about “all other things” as being broken down into those factors affecting volumes (AVt) and those factors impacting on the terms of trade (ATt):

(3)(TBt/(YtPt))=[AVt(μxβx+μmβm)et]+[ATt(μxγx+μmγm)et].

In equation (3), ²x and ²m denote price elasticities of export and import volumes with respect to the exchange rate (reflecting “trade volume sensitivity”), and ³x and ³m are the elasticities of export and import prices with respect to the exchange rate (capturing “pass-through to trade prices”). The assumption that the trade variables respond only to current exchange rate changes is made for expositional convenience (the empirical work will allow for up to 4 year lags). The terms in the first bracket can be interpreted as a decomposition of the overall volume effect in equation (2) in the portion explained by the exchange rate and “all other things,” while the second bracket provides the analogous decomposition for the overall terms of trade effect in equation (2).

13. As a benchmark for characterizing conventional wisdom on the exchange rate-trade link, the empirical analysis in this chapter will use representative short- and long-run estimates of price elasticities for trade volumes and prices Table 2. Similar estimates are embedded in the trade volume and price equations of many standard econometric models, and they correspond closely to what Krugman’s summing up paper in the Bergsten (1991) volume termed “empirical conventional wisdom.” According to this benchmark, exchange rate depreciations are assumed to raise import prices one-to-one, albeit with one-year lag, while export prices are assumed to increase only modestly. Thus, one clear implication of the conventional wisdom view is that the terms of trade should deteriorate about one-to-one in response to an exchange rate depreciation. While export and import volumes are assumed to react much more slowly to exchange rate changes than prices, the representative long-run price elasticities in Table 2 clearly fulfill the Marshall-Lerner condition (that the sum of the two elasticities exceeds one).

Table 2.

Representative Trade Elasticities Based on Conventional View of Linkage Between Trade Rows and Exchange Rate

article image
Sources: Own calculations; trade elasticities based on Bayoumi and Faruqee (1998) and the results of the six econometric models reported in Bryant, Holtham, and Hooper (1988).

14. An illustrative path for the trade balance in response to a 10 percent depreciation of the exchange rate based on the elasticities in Table 2 is shown in Table 3. Following a first-year deterioration in the current account owing to a small J-curve effect (import price pass-through is assumed to be relatively fast compared with trade volume responses in the first year), the current account improves by about 1¼ percentage points of GDP, with most of the trade balance adjustment completed after three years.

Table 3.

Response (in percent of GDP) of Euro Area Non-OU Trade Balance to a 10 percent Real Depreciation under Conventional View of Trade-Exchange Rate Link

article image
Source: Own calculations based on the trade elasticities in Table 1.

15. The setup provided by equation (3) suggests at least two interpretations of “exchange rate disconnect:”

  • First, exchange rate disconnect could mean that exchange rate changes do not affect the nominal trade balance (always holding constant the “all other things” terms, which may themselves depend on the exchange rate) because (positive) volume effects are small and could be offset by (negative) terms of trade effects. This view of exchange rate disconnect appears to underlie classic elasticity pessimism (see Obstfeld (2002, pp. 2-3).

  • Second, (positive) volume effects are small because of “local currency pricing,” i.e. the assumption that import prices are preset in the importer’s rather than in the exporter’s currency. Under an extreme version of local currency pricing, export prices in domestic currencies could be viewed as adjusting proportionally to the change in the exchange rate, with a depreciation increasing export prices in domestic currencies and pass-through to import prices zero. Trade volumes would not react to exchange rate changes, but the terms of trade would improve, leading to an overall improvement in the trade balance. In fact, looking only at movements in the trade balance, the conventional and local currency pricing views of the link between exchange rates and trade flows could be observationally equivalent, although the implications of the views for trade volume responses are radically different.

16. The remainder of the chapter uses equation (3) to shed some light on the degree of connect between exchange rates and trade volumes and prices during the period 1997-2001. At the outset, however, it is important to point out that the presence of the unobserved “all other things” terms in equation (3) suggest that this type of partial equilibrium analysis can only support tentative conclusions.

17. As the most simple exercise, assume that AVt and ATt are kept constant and that trade variables only respond to exchange rate changes. Then, a simulation can be used to calculate a path for the various trade variables during 1997-2001 based on the actual real exchange rate path using the representative trade elasticities in Table 2. Table 4 compares the simulated changes in the non-oil trade balance and in underlying trade volumes and prices with the results in Table 1. The results of this simulation can be compared with the results in Table 1. In response to the cumulative depreciation in the real effective exchange rate of the euro by almost 30 percent, the non-oil trade surplus increases to about 4¾ percent of GDP by 2001, rising some 2½ percent of GDP above its actual value in 2001.12 Thus, this benchmark simulation suggests that the path for the area’s non-oil trade surplus would have been significantly above actual outcomes if conventional wisdom on the link between trade flows and the exchange rate had applied and if other things had been equal. Turning to the underlying reasons for the apparent disconnect in the actual data relative to the benchmark case, the most striking “discrepancy” occurs with regard to volume effects, which are found to be much lower in the actual data. As regards the terms of trade effects on the trade balance, the discrepancy between actual and simulated data is less significant than for volumes, but the overall terms of trade effect in actual data is clearly smaller than implied by conventional pass-through assumptions.

Table 4.

Simulated Changes in Non-Oil Trade Balance Conditional on Representative Trade Elasticities and Actual Real Exchange Rate Path During 1997-2001

article image
Sources: Eurostat; own calculations based on equation (2) in text; volume and terms of price may not add up because equation (2) holds approximately.

18. The previous benchmark simulation can only be illustrative because “other things” have clearly not been equal. Short of applying a full-fledged general equilibrium approach—an approach that looks forbidding in view of the mentioned uncertainties surrounding the driving forces behind the euro’s decline—a less ambitious approach consists in allowing for two other key factors (besides changes in the exchange rate) that drive trade flows: (i) the effects of domestic and foreign absorption growth on trade volumes; and (ii) the effects of domestic and foreign cost developments on domestic trade prices.13 The elasticities of export and import trade volumes with respect to foreign and domestic real absorption growth are assumed to be 1.5 and 2.0, respectively, the difference between the two parameters being motivated by the stylized fact that trading partners’ absorption grows somewhat faster than euro-area absorption.14 The elasticities of export and import prices with respect to domestic and foreign cost developments are set at 1.0, and the domestic and foreign inflation rates of the GDP deflators are used to proxy domestic and trading partners’ cost developments.

19. A simulation of the path for the euro-area non-oil trade balance during 1997-2001 based on these more general assumptions largely confirms the results of the benchmark simulation Table 5. In this more refined simulation, the actual non-oil trade balance in 2001 falls short of the balance’s simulated value in 2001 by about 2 percent of GDP. A detailed breakdown in terms of underlying trade volume and price responses of why the simulated non-oil trade balance deviates from outcome data for 1997-2001 suggests:

  • The largest discrepancies between actual and simulated data occur on the import side. Actual non-oil import volumes grew on average much faster than in the simulation, while pass-through of the depreciation to import prices was far from complete.

  • Actual export volume and price developments are, by contrast, closer to the simulated averages. There is only weak evidence for local currency pricing by euro-area exporters as actual export price inflation exceeds simulated average export inflation only by a small margin.

Table 5.

Simulated Changes in Non-Oil Trade Balance Conditional on Representative Trade Elasticities, Actual Real Exchange Rate Path, and Domestic and Foreign Absorption and Cost Developments During 1997-2001

article image
Sources: Eurostat; own calculations based on equation (2) in text; volume and terms of price may not add up because equation (2) holds approximately.

20. As a final exercise, Table 6 provides additional simulations using a breakdown of goods trade in intermediate, capital, and consumption goods. While the results in Table 6 confirm the previously observed asymmetry for all goods categories, i.e. import volumes and prices appear to be more disconnected from exchange rate changes than export volumes and prices, the results additionally suggest that “disconnect” is clearly strongest for capital goods imports—perhaps in part reflecting a rising share of relatively price-inelastic high tech goods—and probably least pronounced for intermediate goods.15

Table 6.

Euro Area: Simulation of Exctiange Rate Impact on Goods Trade Categories, 1997-2001

article image
Sources: Eurostat; and own calculations.

Intermediate goods imports exclude oil.

References

  • Bayoumi, Tamim, and Hamid Faruqee, 1998 “A Calibrated Model of the Underlying Current Account,” in: Isard, Peter, and Hamid Faruqee, eds., 1998, Exchange Rate Assessment: Extensions of the Macroeconomic Balance Approach, IMF Occasional Paper No. 167, pp. 3234 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Bergsten, Fred C., ed., 1991 International Adjustment and Financing: The Lessons of 1985-91 (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation
  • Bryant, C. Ralph, Gerald Holtham, and Peter Hooper, eds., 1988 External Deficits and the Dollar: The Pit and the Pendulum (Washington: The Brookings Institution).

    • Search Google Scholar
    • Export Citation
  • Campa, Jose Manuel, and Linda S. Goldberg, 2002Exchange Rate Pass-Through Into Import Prices: A Macro or Micro Phenomenon?NBER Working Paper No. 8934.

    • Search Google Scholar
    • Export Citation
  • Devereux, Michael B., and Charles Engel, 2002Exchange Rate Pass-Through, Exchange Rate Volatility, and Exchange Rate Disconnect,NBER Working Paper No. 8858.

    • Search Google Scholar
    • Export Citation
  • Engel, Charles, 1999Accounting for U.S. Real Exchange Rate Changes,Journal of Political Economy 107, pp. 50738.

  • International Monetary Fund, 2001 Annual Report by the IMF Committee on Balance of Payments Statistics (Washington: International Monetary Fund).

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

    • Search Google Scholar
    • Export Citation
  • Isard, Peter, and Hamid Faruqee, et.al., 2001 Methodology for Current Account and Exchange Rate Assessments, IMF Occasional Paper No. 209 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Krugman, Paul, 1987Pricing to Market When the Exchange Rate Changes,” in: Sven W. Arndt and J. David Richardson (eds.), Real-Financial Linkages Among Open Economies, pp. 4970 (Cambridge: MIT Press).

    • Search Google Scholar
    • Export Citation
  • Krugman, Paul, 1989Differences in Income Elasticities and Trends in Real Exchange Rates,European Economic Review 33, pp. 10311054.

    • Search Google Scholar
    • Export Citation
  • Obstfeld, Maurice, 2002Exchange Rates and Adjustment: Perspectives from the New Open Macroeconomics,NBER Working Paper No. 9118.

    • Search Google Scholar
    • Export Citation
1

Prepared by Albert Jaeger (Ajaeger@imf.org).

2

This global pattern of current account balances does not rule out that the euro area may have operated as a “financial turntable” for financing the U.S. current account deficit, with increases in net capital flows from the euro area to the United States financed by increases in net capital flows from other regions to the euro area.

3

See the papers in Bergsten (1991) for a post mortem analysis of the 1980s experience.

5

See Devereux and Engel (2002). Obstfeld (2002) sketches the intellectual history of the various incarnations of the disconnect view, including “elasticitiy pessimism” at the time of the set up of the Bretton Woods system in the 1940s, “real wage rigidity” during the early 1980s, “pricing to market and sunk cost” during the late-1980s, and “local currency pricing” more recently; his paper also provides an overview of the present debate on the international pricing of goods.

6

See Chapter II in the September 2002 World Economic Outlook, pp. 65-81, for a discussion of the evolution of global current account imbalances and associated policy issues.

7

References to the euro’s exchange rate before 1999 always pertain to “synthetic euro rates.”

8

See Chapter II in the May 2001 World Economic Outlook, pp. 66-75, for an analysis linking the euro’s slide vis-a-vis the U.S. dollar to capital flow components. See also last year’s Selected Issues Paper (SM/01/297), which highlights the roles of the global equity price boom and international diversification pressures on the euro. A cursory attempt to compile a list of explanations of “Why is the Euro so Weak?” suggested that around 25 different explanations have been put forward to account for the euro’s persistent weakness.

9

Own estimates based on the eleven model simulations in table C (fiscal contraction), table F (portfolio shift), and table E (monetary expansion) in Bryant et.al. (1988).

10

As an aside, would using the ECB’s bias-adjusted measures of the external current account balance (instead of using the sum-of-countries measure as presently the case) reduce the global current account discrepancy? The answer appears to be no. For example, in 2000 the global current account discrepancy amounted to $128 billion, reflecting a “global goods trade surplus” of $16 billion and combined deficits of $144 billion in the non-goods balance items (see IMF (2001), Table 1). Thus, if the euro area would have reported the ECB’s bias-adjusted current account deficit of $55 billion for 2000 (instead of the sum-of-countries deficit of $ 15 billion), this would have boosted the global current account discrepancy by some 40 billion U.S. dollars. This does not, however, point to any inadequacies with regard to the ECB measure.

11

See Campa and Goldberg (2002) for detailed cross-country estimates of exchange rate pass-through to import prices for oil and coal.

12

Assuming that the real exchange rate remains constant from 2002 onward, owing to the long lags between movements in the exchange rate and trade volumes the simulated trade surplus would rise to 5¾ of GDP by 2004.

13

This approach is roughly equivalent to the “benchmark comparators” approach outlined in Appendix II of Isard et.al. (2001).

14

The relative size of the trade income elasticities has been calibrated to align them with Krugman’s (1989) “45-degree rule,” namely that in the long run the ratio of domestic to foreign absorption growth should be roughly equal to the ratio of the income elasticity for exports to the income elasticity for imports. Advanced trading partners’ absorption growth during 1997-2001 was about 3 percent, while the euro area’s absorption grew by 2lA percent during the same period.

15

This finding echoes the finding in Krugman’s (1987) classic article on pricing-to-market, which reported an unusual degree of disconnect between exchange rate changes and export prices in the case of German exports of machinery and equipment to the U.S.

Monetary and Exchange Rate Policies of the Euro Area: Selected Issues
Author: International Monetary Fund