This Selected Issues paper analyzes the current account performance of Denmark in 1993–98. The paper presents a brief review of structural features of the external current account. It looks at the decline in export market share and concludes that it reflects primarily cyclical factors and the unwinding of an unsustainable export market gain immediately after the German unification. The paper also examines implications for fiscal policy of Denmark’s decision to remain for the time being outside the European Monetary Union.

Abstract

This Selected Issues paper analyzes the current account performance of Denmark in 1993–98. The paper presents a brief review of structural features of the external current account. It looks at the decline in export market share and concludes that it reflects primarily cyclical factors and the unwinding of an unsustainable export market gain immediately after the German unification. The paper also examines implications for fiscal policy of Denmark’s decision to remain for the time being outside the European Monetary Union.

Denmark: Basic Data

article image

Staff projections.

Change as percent of previous year’s GDP.

In percent of labor force.

Denmark: Basic Data (concluded)

article image

Staff projections, unless otherwise stated.

The balance on the central government’s current, investment, and lending account.

Data for 1999 refer to April 1999 compared with April 1998.

Data for 1999 refer to June 23,1999.

Data for 1999 refer to June 1999.

Based on consumer price index; data for 1999 refer to April 1999.

Introduction

1. While some imbalances have emerged in the Danish economy, there are few pressing short-term problems requiring immediate attention. The three background studies included in this paper thus attempt to shed some light on issues of interest from a more medium-term perspective: the causes of recent developments in the current external balance; the implications for fiscal policy of Denmark’s decision to remain for the time being outside the EMU; and an assessment of the potential role that foreign-currency denominated debt could play in countering the costs of fluctuations in domestic output.

2. For the quarter century prior to 1990, Denmark ran current account deficits. This legacy has influenced the sensitivity of policy makers to the most recent dip of the current account into deficit following eight straight years of current account surpluses. Chapter I analyzes the factors associated with the recent deterioration in the current account to provide an assessment of the underlying strength of the external position and decomposes the current account balance into a cyclical component and a structural component. While several factors are implicated in the worsening external performance—for instance, the unwinding of an unsustainable increase in the export market share following German unification—the underlying current account is found to remain in surplus, amounting to about ½ percent of GDP. However, the analysis suggests that the underlying surplus has also diminished in recent years. To counter this tendency, policies which strengthen private savings would be called for. Recent measures introduced in Denmark are expected to raise private savings and improve the external current account.

3. Denmark has demonstrated a long-standing commitment to stable exchange rates by fixing its exchange rate (most recently with a peg to the newly-introduced euro) and maintaining economic policies consistent with its level. Because of Denmark’s open capital markets, the fixed exchange rate policy has limited its ability to pursue an independent monetary policy and thus fiscal policy has been the major instrument for stabilization policy. Since Denmark has chosen to remain an EMU “outsider,” but at the same time has anchored its exchange rate to the euro, Chapter II asks whether this special arrangement places an increased burden on fiscal policy in mitigating output fluctuations. A tentative answer is provided within the context of a two-country stochastic model calibrated to reflect some characteristics of the Danish economy and its links with the euro area. Simulations are performed to quantify the expected variability of output and inflation under an “outsider” case, and an “insider” counterfactual scenario, in which Denmark is assumed to join EMU. These results suggest that there are elements of the outsider arrangement (in this case, related to variations in the risk premium) which lead to an increased need for fiscal policy flexibility and under certain circumstances the difference between the two regimes can become substantial.

4. Long-run challenges for Danish fiscal policy also stem from a more domestic source—the aging of the population in combination with the comparatively high level of social benefits provided to the populace in retirement. To prepare for the future pressure on the fiscal accounts, the official strategy attaches significant weight to reducing the level of sovereign debt. In this context, but also with a view to strengthening the external position, the authorities also aim to reduce and eventually eliminate net external debt. Even though the fiscal constraints likely to arise in the future represent sound reasons for reducing net external debt, recent theoretical work suggests that there may be other reasons, namely macro-economic risk diversification considerations, for maintaining a positive level of foreign currency-denominated debt. Uncertainty in consumption due to output variations can potentially be “shared” with foreigners by issuing foreign currency-denominated debt. Chapter III examines whether these potential diversification benefits are significant enough to make it desirable for Denmark to maintain a positive level of gross foreign currency-denominated debt. For this risk-sharing method to be effective, fluctuations in output and inflation between Denmark and its trading partners need to take on a specific configuration. Using Denmark’s largest trading partners, Sweden, Germany, and the 11 EU countries, the empirical evidence shows that the issuance of foreign currency-denominated debt would not aid Denmark in hedging output fluctuations primarily because output fluctuations are only loosely tied to those of its trading partners.

I. Current account performance in 1993-981

A. Introduction and Summary

5. The external current account in Denmark swung from a surplus of 3½ percent of GDP in 1993 to a deficit of 1.4 percent of GDP in 1998. Over the same period, the manufacturing sector experienced significant losses in market shares both at home and abroad. This chapter analyzes the factors accounting for these developments in an attempt to facilitate an assessment of the underlying strength of the external position. It begins with a brief review of recent developments and a discussion of structural features of the external current account. The next section looks at the decline in export market share and concludes that it reflects primarily cyclical factors and the unwinding of an unsustainable export market gain immediately after the German unification. In the fourth section an attempt is made to decompose the current account balance into a cyclical/temporary and structural component. It is found that while both factors have contributed to the weakening of the current account, the underlying current account remained in surplus, equivalent to about ½ percent of GDP; in 1998.

B. Background

6. The external current account in Denmark recovered from a deficit of 5½ percent of GDP in 1986 to a surplus of 3½ percent of GDP in 1993; but this remarkable improvement was subsequently followed by a gradual weakening culminating in a deficit of 1.4 percent of GDP in 1998 (the first deficit in a decade, Figure I-1)2. The initial improvement was supported by fiscal consolidation, policies that stimulated private savings, and strong foreign demand; increasing off-shore oil and gas extraction and low growth (average annual rate or 0.8 percent in 1987-92) have also been contributing factors 3. The subsequent weakening of the external current account reflects a combination of strong domestic demand and weak external demand. A package to inter alia eliminate excess demand and strengthen the external position was adopted in mid-1998. It is to be phased in over 1999-2002 and is expected to raise the savings ratio and the current account by 1 percentage point of GDP by 2002.

Figure I-1.
Figure I-1.

Current Account Balance

(In percent of GDP)

Citation: IMF Staff Country Reports 1999, 107; 10.5089/9781451811070.002.A001

7. Trade in goods and nonfactor services accounts for the bulk of current account flows and dominates the evolution of the current account balance; both merchandise trade4 and nonfactor services exhibited very simitar patterns (Figure I-2), Net factor income remained largely flat in the 1980s but improved in the 1990s as a result of external debt reduction and the decline in interest rates. Net transfers, mainly to the EU, rose in the 1990s but their overall impact on the current account was small. The rest of this chapter focuses on the trade balance and its components.

Figure I-2.
Figure I-2.

Current Account and its Components

(In percent of GDP)

Citation: IMF Staff Country Reports 1999, 107; 10.5089/9781451811070.002.A001

8. It is useful to decompose the trade balance into terms of trade and volume effects5 and to examine how the latter can be further decomposed into volume changes of exports and imports (Figure I-3). The terms of trade for goods and nonfactor services improved by 6.3 percent in 1988-98. This contributed to the improvement of the trade balance in 1988-93 and mitigated its deterioration in 1994-98, but the effect was quite small compared to the overall change in the trade balance. At constant prices, the trade surplus peaked at 6¼ percent of GDP in 1993 and declined by 6 percentage points of GDP in the subsequent five-year period.

Figure I-3.
Figure I-3.

Decomposition of the Trade Balance, 1988-98

(In percent of GDP)

Citation: IMF Staff Country Reports 1999, 107; 10.5089/9781451811070.002.A001

9. There is a clear negative correlation between the trade balance and the output gap (Figure I-4). This correlation suggests that strong demand may have been the driving force behind the weakening of the current account since 1994. The composition of demand has also shifted: whereas domestic and foreign demand contributed roughly equally to GDP growth before 1994, in the subsequent period GDP growth was driven by strong domestic demand and weakening net external demand.

Figure I-4.
Figure I-4.

Trade Balance and Domestic Output Gap, 1988-98

Citation: IMF Staff Country Reports 1999, 107; 10.5089/9781451811070.002.A001

10. The following structural characteristics of the Danish external sector are useful for the subsequent discussion:

  • Exports and imports of goods and non-factor services account each for about 35 percent of GDP (Figure I-5). The ratios of trade flows to GDP exhibit a pronounced upward trend when calculated at constant prices, apparently the result of increasing globalization. They appear trendless when they are at current prices, reflecting a faster growth of the GDP deflator relative to the trade deflators.

  • Since the mid-1980s Denmark has been a small net exporter of crude oil and gas. Thus oil price gyrations have only a small direct effect on the trade balance. However, large investment related to oil exploration exacerbated the current account deficit in the first half of the 1980s and transfers to foreign investors partially offset the improvement of the trade balance in the subsequent period.

  • The EU is the destination for 65 percent of Danish merchandise exports. The main trading partner is Germany (21 percent of Danish exports) followed by the United Kingdom, Sweden, Norway, and the United States. The main changes in the geographic pattern of trade over the past decade have been: the rise in the share of Germany since unification; the drop in the share of the United Kingdom; the doubling in the share of CEECs to 7 percent; and the rise in the share of Asia from 6 percent in the mid 1980s to 9 percent currently.

Figure I-5.
Figure I-5.

Trade in Goods and Services

(In percent of GDP)

Citation: IMF Staff Country Reports 1999, 107; 10.5089/9781451811070.002.A001

C. Export Performance of Danish Manufacturing

11. Export performance, as measured by the share of Danish exports in foreign imports, has deteriorated sharply since 1993. The loss of market share was particularly pronounced (14 percent) in the case of manufacturing (Figure I-6). The deterioration was larger than the one experienced in the upswing during the first half of 1980s and was accompanied by increased import penetration, as measured by the share of Danish imports to final total demand.

Figure I-6.
Figure I-6.

Export Performance, 1981-98

Citation: IMF Staff Country Reports 1999, 107; 10.5089/9781451811070.002.A001

12. The following equation for manufactured exports was estimated to help quantify the causes of the recent deterioration in export performance:

ln(xmv/xmv-1)=0.93ln(xmvmkt/xmvmkt-1)-0.11ln(reer/reer-1)-0.006DGAP(0.23)(0.19)(0.003)-0.49[ln(xmv-1)-4.34-1.19ln(xmvmkt-1)+0.70ln(REER-1)-0.02REL-1+0.06D9498)](0.13)(0.88)(0.06)(0.20)(0.01)(0.05)

Where xmv denotes the volume of manufacturing exports, xmvmkt measures the size of foreign market in manufactured goods (weighted average of manufacturing import volumes of Danish trading partners); REER denotes the real effective exchange rate (unit labor cost-based), DGAP denotes the deviation of the final domestic demand from its trend; BEL stands for the relative price of manufactured exports to the GDP deflator; finally, D9498 is a dummy for the period 1994-98 intended to capture structural breaks in export performance. The equation was estimated with annual observations for 1975-98 from the OECD Economic Outlook databank.

13. The reduced form export equation contains explanatory variables pertaining to domestic and foreign demand for and the supply of manufactures. The coefficient estimates are plausible and reasonably robust to changes in the econometric specification.

  • The coefficient of the error correction term in the above equation indicates a fairly rapid response of exports to deviations from their longer-term path: half of the deviation is eliminated within one year. Moreover, export growth moves in tandem with contemporaneous export market growth (its coefficient, 0.93, is insignificantly lower than unity).

  • The greater than unity long-term elasticity with respect to export market growth is influenced by sizeable market share gains in the early part of the sample. Re-estimating the equation in the post 1980 period yields an elasticity slightly below unity which is consistent with a creeping erosion of market share during that period.

  • The real effective exchange rate accounts for only a small part of the variability in exports. The size of the elasticity is generally low, although it varies somewhat depending on the measure used: it is larger when the price of manufactured exports relative to those of foreign competitors is used as an explanatory variable instead of the unit labor cost based real effective exchange rate.

  • DGAP is a proxy for domestic demand pressure and capacity utilization. Its negative coefficient captures the tendency of producers, at times of buoyant domestic demand, to shift from foreign to the domestic market (but due to its correlation with REER, it captures other effects as well). Such shifts tend to be accompanied by rises in export prices above those of competitors’ and an appreciation of the real effective exchange rate.

  • REL intends to capture the supply response to the declining price of manufactures relative to the GDP deflator. This decline, which has been ongoing since the early 1980s, must have triggered some reallocation of resources away from manufacturing and toward sectors with rising relative prices (e.g., services). The coefficient of REL has the appropriate sign but is insignificant numerically and statistically.

14. The coefficient of the dummy variable D9498 (although not statistically significant) indicates that in the last cyclical upswing export performance deteriorated by 6 percent relative to earlier periods. This deterioration may reflect factors not captured by the equation such as the lower savings rate, the larger share of high import content items in total demand during that period, and some permanent deterioration in export performance related to:

  • The opening up of trade of several newly industrialized countries and Central and East European countries. Liberalization led to rapid export growth in these countries and, indirectly, contributed to Denmark’s loss of export market share. Meanwhile, Danish exporters maintained their market share in imports of OECD countries from other OECD countries.

  • The change in the pattern of specialization towards high value-added high quality products and the relocation of production lines (especially the labor intensive ones) to low wage countries.6 This is a desirable change that improves terms of trade and at the same time reduces export volumes; export market shares, adjusted for quality, may have been maintained. In this connection it is noteworthy that market shares in nominal terms have shown smaller declines (higher prices can be thought of as proxies for better quality).

  • The return of export market shares in Germany to more sustainable levels (see below).

15. The table below summarizes the contribution of the various explanatory variables to the increase of manufactured exports. Whereas in 1988-93 export market growth and domestic demand pressures contributed roughly equally to export growth, the increase in exports in 1994-98 was driven by export market growth; meanwhile, domestic demand pressures, the appreciation of the real effective exchange rate and unaccounted factors lowered export growth and the export market share. At the same time, the overall profitability in manufacturing remained comfortable and shortages in skilled labor seemed to have constrained supply.

Contributions to the Increase in Manufacturing Export Volumes 1/

article image

Percentage changes are approximated by the change in the logarithm.

It includes also the effect of the dummy variable.

D. The Underlying Current Account

16. This section shifts focus to the overall external performance of the Danish economy by looking at the underlying current account, namely the level of the external current account that would prevail if temporary and cyclical factors were absent and if trade volumes and prices had adjusted fully to the current exchange rate.7

17. The calculation of the underlying current account requires estimates of trade elasticities and a quantification of the effect of special factors.

18. Elasticities of exports and imports in goods and nonfactor services with respect to economic activity and the exchange rate were estimated using annual data for the period 1975 to 1998.8 Several specifications were tried to test for the robustness of the estimates. Besides export market volume and the real effective exchange rate, the export equation included also the domestic output gap as explanatory variable, to test for the hypothesis that in periods of strong domestic demand producers switch from foreign to the domestic market. Moreover, a dummy for the post-1990 period was introduced to capture the effects of the German unification. In the import equation the dependent variable (the volume of imports) was redefined by excluding the estimated import content of exports.9

  • The elasticity of exports with respect to export market volume hovered around 1.2 and differed marginally from unity (which would had been consistent with a constant market share). The elasticity of imports (adjusted for the import content of exports) with respect to domestic activity was about 1.4 and significantly higher than unity. This greater than unity elasticity raises the amplitude of cyclical movements in imports and, together with a trend increase in total demand, leads to increasing import penetration.

  • The elasticity of exports and imports with respect to the real effective exchange rate depended on the definition used. Relative prices of exports relative to those of competitors tended to give the highest coefficient values; CPI and GDP deflator based measures gave slightly lower estimates (around 0.8 in the case of exports and 0.7 in the case of imports), and those based on unit labor costs (ULC) gave the lowest. The low elasticity with respect to ULC-based indicators suggests that profits act as a shock absorber: lower profit margins offset wage increases to maintain competitiveness.

  • The coefficient of the domestic output gap in the export equation had the right (negative) sign but was insignificant (both numerically and statistically). This is probably due to the correlation between the gap and the real effective exchange rate.

19. While permanent shocks (e.g. the development of off-shore oil production in the mid-1980s) which have lasting effects on the current account should be included in the underlying current account balance, transitory/extraordinary events that are unrelated to, and not captured by the output gap (or other cyclical indicators) and the other determinants of trade flows need to be excluded from it.10 The following special factors were considered:

  • German unification. This led to an export surge in the early 1990s as Danish exporters were quick to take advantage of new business opportunities: their market share in German imports increased from 1.8 percent in 1988 to 2.6 percent in 1993. However, as other exporters caught up, some of the initial gains in market share were eroded and by 1997 market share had receded to 2.4 percent. This temporary surge in exports reached 1 percent of Danish GDP in 1993 and faded out in the subsequent period.11

  • Deviations of the terms of trade from their trend. The terms of trade improved continuously in the past decade adding to the improvement in the underlying current account, measured at current prices.

  • Gyrations in pork exports stemming primarily from the BSE-crisis in the UK in 1996, pig diseases in the Netherlands and the Taiwan Province of China in 1997, and the collapse of exports to Russia in 1998. Corresponding to these shocks, pork prices soared in 1996-97 and collapsed in 1998. The temporary shocks were calculated as the percentage deviation of pork prices from their normal level times the share of pork exports to GDP.

  • Other special events. These include a major strike in 1998 (estimated to have worsened the current account by 0.1 percentage point of GDP); the temporary closure of shipyards in 1998, mainly due to intense competition following the depreciation of the Korean won (estimated to have contributed 0.2 percent of GDP to the weakening of the current account); the temporary increase of electricity exports to Sweden and Norway in 1996-97 due to inadequate rainfall in those two countries (estimated to have improved the current account by 0.1-0.3 percent of GDP).

20. Based on the above methodology, the underlying current account is estimated to have deteriorated by 2 percentage points of GDP in 1994-98, compared with an almost 5½ percentage points of GDP deterioration in the actual current account (Figure I-7, and tabulation below). Cyclically strong domestic demand and temporary events contributed each 1½ percentage points of GDP to the decline in the current account.12 By comparison, the improvement of the current account in 1988-93 was driven primarily by a 514 percentage points of GDP improvement in the underlying current account. In 1998, the underlying current account was in surplus of ½ percent of GDP compared with an actual current account deficit of 1½ percent of GDP and a savings-investment “norm” equivalent to a surplus of 1½ percent of GDP.13

Figure I-7.
Figure I-7.

External Current Account Balance

(In percent of GDP)

Citation: IMF Staff Country Reports 1999, 107; 10.5089/9781451811070.002.A001

Contributions to the Change in the Current Account Balance

(In percent of nominal GDP)

article image

21. In June 1998, a package of measures (Whitsun Package) was adopted to stimulate private savings, cool off the housing market and strengthen work incentives. The measures, which would be phased over the four year period 1999-2002, included inter alia a steep cut in the tax value of interest deductions. This reduction of tax deductibility from 46.4 percent to 32.2 percent would raise the real lending rate by one percentage point and, in turn, is expected to raise the savings ratio (and improve the external current account) by one percentage point of GDP.14

1

Prepared by Anastassios Gagales.

2

Unless otherwise indicated, all ratios are calculated with respect to nominal GDP.

3

The recovery of the current account in the 1980s has been reviewed in SM/95/62.

4

The quality of trade statistics has deteriorated since the replacement in 1993 of the old custom-based system of collecting information on intra-EU trade flows. There have also been large discrepancies between trade statistics and other information on trade (e.g., balance of payments statistics, VAT revenues and other countries’ exports). For instance, Germany’s industrial imports from Denmark fell by more than 30 percent in 1993 whereas Danish exports to Germany fell by only 3 percent. Investigations on these discrepancies are ongoing.

5

The decomposition can be written as (X-M)=(x-m)+[(Px-P).x-(Pm-P).m]/Y where X and M stand for exports and imports measured in percent of nominal GDP; x and m denote exports and imports at constant prices in percent of real GDP; Px, Pm and P denote respectively the deflators of exports, imports and GDP. The bracketed term captures the terms of trade effect; its size depends on the change in the relative price of exports/imports (cumulatively relative to the base year) and the extent to which the country is a net exporter (net exporters benefit more from a terms-of-trade improvement).

6

Large Danish enterprises moved production lines abroad in the early 1990s and smaller ones started to follow suit in the mid 1990s. This should show in the external current account as reduced exports and higher net factor income.

7
The underlying current account is calculated by eliminating from the actual current account the effect of temporary and cyclical factors and unrealized effects of past exchange rate changes. It is given by the formula below which is based on standard trade equations that link trade flows to the real effective exchange rate and economic activity at home and abroad and the assumption that the impact of a change in the real effective exchange rate is completed in three years (in proportion to 60:25:15); export prices (in foreign currency) are unaffected by changes in the real effective exchange rate (small country assumption), while import prices are subject to immediate and full pass-through:
(CA/Y)underlying=CA/Y-[(M/Y)βm+(X/Y)βx][0.40(Rt-Rt-1)+0.15(Rt-1-Rt-2)]+(M/Y)γmYGAP-(X/Y)γxYGAPF
where X, M, CA and 7 denote respectively the nominal domestic currency value of exports, imports, current account, and output; exports and imports are defined exclusive of re-exports and the import content of exports. YGAP and YGAPF are the deviations of actual from potential output in the home country and in trading partners (positive values indicate that real GDP is above potential); R is the logarithm of the real effective exchange rate (increase indicates appreciation); and βm, βx γm, and γx are import and export elasticities with respect to the real effective exchange rate and economic activity at home and abroad. The second term on the RHS of the above expression measures the volume effect of past exchange rate changes that have yet to materialize; the third and fourth terms measure the impact of output gaps at home and abroad; and the last term captures the effects of transitory factors. The above formula is an adaptation of a similar formula in Exchange Rate Assessment-Extensions of the Macroeconomic Balance Approach, (1998) Occasional Paper 167, edited by P. Isard and H. Faruqee, It has been modified to take into account the high import content in Danish exports; it also abstracts from within the year changes in the exchange rate given that the focus is on the underlying current account rather than on assessing a particular level of the exchange rate.
8

Lumping together goods and nonfactor services did not alter significantly the econometric estimates. This probably reflects the fact that non-factor services account for 20 percent of total trade and their share has been fairly stable when calculated at current prices (at constant prices it declined by 2 percentage points in the past decade).

9

The import content was set at 20 percent of total exports based on non-sample information. Attempts at directly estimating this coefficient gave implausible estimates.

10

Special factors are not entirely independent from the output gap or the real effective exchange rate. For example, the export surge related to the German unification affected the level of economic activity and, probably, the real effective exchange rate. Such secondary effects were not taken into account in the calculations.

11

The temporary effect was estimated by smoothing (ω-ϖ) MG/GDP where ω,ϖ denote the actual and “normal/sustainable” share of Danish exports in the German market, MG stands for total German imports, and GDP for nominal Danish GDP.

12

The cyclical components of trade flows are assumed proportional to the output gaps. This abstracts from that the different components of demand differ in their import content and, hence, in their impact on the current account: a positive domestic output gap driven by buoyant exports tends to strengthen the current account whereas one driven by strong domestic demand tends to worsen it. Similarly, the import compression in crisis-stricken emerging market economies in 1997-98 may have biased downward the foreign cyclical effect.

13
The norm, estimated by regressing the current account over its fundamental determinants using a panel of 21 industrial countries, is given by:
S-Inorm-3.145(FE+0.210SUR-0.057DEM+0.024YPCAP)
Where SUR denotes the country’s general government structural balance (in percent of GDP) relative to the average of the countries in the panel; DEM is the country’s dependency ratio (population older than 64 and younger than 20 in percent of working age (20-64) population) relative to the average in the panel; YPCAP stands for the country’s per capita income relative to that of the United States; and FE is country-specific constant that incorporates fixed effects and the impact of tax policies that have raised the propensity to save since 1986. See Isard and Faruqee, op. cit.
14

The calculations are based on an estimated unit semi-elasticity of the savings rate with respect to the real after-tax interest rate. The real after-tax interest rate is given by r=(l+R(l-t))/(l+i)-l where R, t, and i denote respectively the nominal interest rate, the marginal tax rate applying to deductible interest expenses, and inflation. With nominal borrowing rate and inflation at 7 percent and 2 percent, the reduction of the marginal tax rate from 46.4 percent to 32.2 percent raises the real borrowing rate from 1.7 percent to 2.7 percent. Ministry of Finance, Finansredegorelse 97, November 1997 and Okonomisk Oversigt, May 1999.

Denmark: Selected Issues
Author: International Monetary Fund