Table 1.1. A comparison between the present forecast and the previous forecast in selected areas.

Transcription

1 English summary 1. Short term forecast Since the beginning of 1 the international economy has experienced relatively low growth rates. This downturn in economic growth has been followed by a substantial expansion in both fiscal and monetary policy especially in USA, but also in Europe. The expansion of economic policy has together with automatic fiscal stabilisers averted a stronger decline in the international growth rates, although also postponed a restoration of private and public balance sheet problems. The growth rates of GDP in USA and EU are expected to be moderate in around.3 per cent in USA and 1. per cent in EU. The stock market decline has reduced consumer and business confidence. Hence, the recovery of the global economy has been postponed and is now expected to gather momentum later in 3. As a result the growth rate of GDP in 3 in both USA and EU has been revised downward by around ¾ percentage points to.3 per cent and. per cent respectively, cf. table 1.1. Table 1.1. A comparison between the present forecast and the previous forecast in selected areas. 1 3 Aug. Dec. Aug. Dec. Dec Percentage change GDP growth USA GDP growth EU Denmark: GDP growth Private consumption Percentage change Consumer prices, Hourly earnings, private sector persons Unemployment Source: Statistics Denmark, EU-Commission, and own calculations. Economic Survey December 1

2 In the growth rate is estimated to reach. per cent in USA and, per cent in EU. In Denmark, real GDP is expected to increase by 1. per cent in 3 following a moderate growth of 1.5 per cent in, cf. figure 1.1a. The estimate for 3 has been revised downward by. percentage points compared to the Economic Survey, August. This is mainly due to a downward revision of growth in net exports and fixed business investments. Growth in is forecast at.1 per cent which is a little above the estimated future trend growth. Figure 1.1a. Real GDP growth in Denmark and EU. Figure 1.1b. Growth in private consumption expenditure in Denmark and EU Denmark EU Denmark EU-15 - Source: Statistics Denmark, EU-Commission, and own calculations. GDP components Following three years of stagnation, private consumption is expected to increase by around ¼ per cent annually in, 3, and, cf. figure 1.1b. Private consumption is therefore to a large extent expected to be the driving force of domestic demand in the forecast period. The relatively strong growth in private consumption growth in is among other things due to relatively high growth in retail sales and passenger car registrations, cf. box 1.1. The outlook for private consumption is also related to a comfortable financial situation of Danish households. Total private savings and financial wealth are still at relatively high levels, indicating that the short-term growth prospect rests on solid ground. This is emphasized by the continuous surpluses on the external current account. Economic Survey December

3 Box 1.1. Composition of private consumption, -. The overall growth of private consumption remains unchanged in compared to the survey in august. However, the composition of the consumption growth is different is due to a higher increase in the sales of retail products and goods, but a lower sale in services etc., cf. table a. Based on an analysis of structural effects on car sales for the next ten years the present low level of new car registrations is due to three factors: the age distribution of the car park, the expected growth in population in the near future and the historical increase in number of passenger cars per head. On this basis and due to the expected development in user cost, interest rates etc. car sales are expected to increase by 5- per cent in 3 and, which supports the increase in the overall private consumption over the period. The border trade-package will lower prices on especially alcohol and cigarettes from October next year and will imply an increased domestic sale of these products. Hence, the overall real growth in the sales of beverages is expected to increase by per cent in, cf. table a. Tabel a. Real growth in private consumption and some components ) Aug. Dec. Aug. Aug. Dec. Bill. Dkr Percentage change Private consumption ,,3,,3,, Of which Retail sales... 3,3,7 3,,,5 3,3 Of which - Beverages... 3,3 1,7 1, 1,,5, Car sales ,,, 7,, 5, Services ,,1,3,5,,3 Others ,3,, 1, 1, 1,1 1) Average for the period Source: ADAM s data bank and own calculations. Furthermore, employment prospects are good and real disposable income of households is expected to increase by 1.7 per cent in and.1 per cent in both 3 and 1. Public consumption is expected to show real growth of 1.3 per cent in, cf. table 1.. The estimate is based on the Fiscal Budget for, which was passed by Parliament in March. Based on the Budget draft for 3 and an agreement with local governments real public consumption is assumed to grow by.7 per cent in 3. Public investment is estimated to increase by. per cent in and remain unchanged in 3. 1 Figures are adjusted for special factors concerning pension funds. Economic Survey December 3

4 For a technical assumption is applied for public consumption growth of 1. per cent and public investment growth of. per cent in line with the medium targets. See section. Table 1.. Use and supply of goods and services Bill. Dkr Percentage changes from previous year Private consumption Government consumption Government investment Residential construction Fixed business investment Stock building 1) Total domestic demand Exports of goods and services Of which manufactures Imports of goods and services Of which goods Gross domestic product GDP at factor costs Of which in private nonagricultural sector GDP per capita (1. Dkr.) ) The percentages indicate the volume changes in stock building in per cent of volume GDP in the preceding year, both in 1995-prices. Source: ADAM s databank and own calculations. Residential investment is expected to fall by per cent in. The estimated growth rate in has been revised down by percentage points compared to August. The forecasted negative growth rate in is primarily due to negative growth in the first two quarters of and a fall in the business indicators for building and construction. In accordance with a new political agreement a pickup in the number of municipal housing projects is expected, and this will contribute to a modest increase in residential investments of 1 and per cent in 3 and. Movements in property prices have showed a significant slowdown since the fourth quarter of 1. The increase in the real estate price of owner-occupied one family houses are therefore expected to increase by only and 1.5 per cent in 3 and respectively, after an increase of.5 per cent in. Business investments are expected to increase in by 1.5 per cent after a 3 per cent growth in 1. This is an upward revision of 1.3 percentage points in since August. The revision in reflects a sur- Economic Survey December

5 prising high level of investment in equipment in the second quarter of. Influenced by the international environment a slow down in the second half of is foreseen. In 3 and a renewed pick up in production growth is expected to increase growth in business investments by.7 per cent and 3. per cent, respectively. Business investment growth in 3 is.7 percentage points smaller than expected in August as a consequence of the postponement of an expected upturn in the international market trends. However, low interest rates are expected to support the growth of business investment. Following an increase in inventory investment of. per cent of GDP in 1, inventory changes are expected to be negative, contributing around -.1 percentage points to GDP growth in. In 3 and no contribution from inventory changes to GDP growth is expected. Figure 1.a. Total investments in Denmark and EU Denmark EU Figure 1.b. External current account in Denmark and EU. of GDP of GDP Denmark EU Source: Statistics Denmark, EU-Commission, and own calculations. Growth in exports of goods and services has not been as strong as expected in the first three quarters of. Hence, export growth in is revised downward by 1.5 percentage points to. per cent. Due to the postponed recovery of the international economy the pick up in exports in 3 will be more moderate than previously expected and export growth is estimated to be 3. per cent in 3, which is a downward revision of 1.7 percentage points compared to the survey in august. In the renewed international upturn will contribute to a rise in the growth rate of exports to 5. per cent. The gradual increase in domestic demand is expected to lead to a similar rise in imports. Thus, the growth in imports of goods and services are Economic Survey December 5

6 expected to rise from 3. per cent in to.3 per cent and. per cent in 3 and, respectively. As a result the surplus on the external current account is expected to remain stable at around. per cent of GDP from to, cf. figure 1.b. Consequently, net foreign liabilities will be reduced from 1.7 per cent of GDP in 1 to 1. per cent of GDP in and 13. per cent and 11.1 per cent of GDP in 3 and. In the period from 1 to the surplus on the external current account will be of broadly the same magnitude as the surplus on general government finances (cf. below). Consequently, private sector savings are sufficient to finance private investment. Even if the savings ratio of households is small compared to other countries, the total private sector is well in balance. Employment, unemployment and labour force Overall employment is expected to fall in by. persons, increase in 3 by 3. and 13. persons in, cf. table 1.3. In and 3 growth has been revised downwards by 1. and 1. persons respectively compared to the Survey from August. In the revisions are based on labour market statistics from the first half of and the expectation of modest production growth in second half of. In 3 the revision follows the change in production growth forecast. Only private sector employment growth has been adjusted compared to the Survey in August. In private employment is expected to fall by 1. persons. This is the first fall in private sector employment on an annual basis since 199. Next year private sector employment is expected to stay at the same level as in while it is expected to grow by 9. persons from 3 to following the international upturn. From January till October (seasonal adjusted monthly) unemployment has increased by 1. persons, a bit more than expected in the Survey from August. Therefore average unemployment in has been revised upwards by. persons compared to the Survey in August to 1. persons. From to 3 the average level of unemployment is expected to increase by 9. persons. This is a revision of approximately 13. persons, which is related both to the revision of growth in 3 and to the expected higher level of (monthly) unemployment at the end of. In unemployment is expected to fall by 5. persons to an average level around 15.. Economic Survey December

7 Table 1.3. Key figures of the Danish economy Percentage changes from previous year ---- Real GDP Trade-weighted GDP abroad Markets for Danish manufactures International competitiveness Export of manufactures, volume Hourly wages Consumer price index Real estate prices for one-family houses Merchandise export prices Merchandise import prices Merchandise terms of trade Productivity in private non-agricultural sector Real disposable income of households 1) Labour market: Labour force (thousands) Employment (thousands) Of which in private sector in public sector Percentage change in total employment Unemployment (thousands) Early retirements (thousands) Persons on leave (thousands) Unemployment rate (per cent) Unemployment rate, EU-def. (per cent) Long term bond yields, exchange rate: 1-year government bonds year mortgage credit bond The effective krone rate (19=1) Balance of payments: Goods and services (bill. Dkr.) Current account (bill. Dkr.) Current account in per cent of GDP Net foreign debt, ult. (bill. Dkr.) Net foreign debt in per cent of GDP Public finances: Government net lending (bill. Dkr.) ) Government net lending in per cent of GDP ) General government gross debt, ult. (bill. Dkr.) General government gross debt ult. in per cent of GDP Tax burden (per cent of GDP) ) Expenditures (per cent of GDP) ) Adjusted for special factors concerning pension funds. ) For the years government net lending and the tax burden include the net savings respectively gross savings in the Special Pension fund (app. ½ per cent of GDP). Source: Statistics Denmark and own calculations. Economic Survey December 7

8 This implies that the unemployment rate will be.3 per cent (standardised figures) in,. per cent in 3 and. per cent in, cf. figure 1.3a. Though unemployment is expected to increase from to 3 the level is still low in a historical perspective. The actual unemployment rate is still below the estimated structural unemployment rate indicating some pressure on capacity in the labour market. However, compared to the Survey in August this pressure has diminished somewhat in 3. Figure 1.3a. Unemployment rates in Denmark and EU. of labour force Denmark (EU-Def.) EU of labour force Denmark (Nat. Def.) 1 1 Figure 1.3b. Consumer price inflation (HICP) in Denmark and EU Denmark EU Source: Statistics Denmark, EU-Commission, and own calculations. The labour force is expected to fall from 1 to by. persons. This is a downward revision of 1. persons compared to the Economic Survey in August. According to a decomposition the fall is primarily due to a cyclical weakening of the labour market and not due to a higher than estimated outflow to early retirement schemes. Overall the labour force is expected to grow by 3. persons from to, cf. box 1.. Inflation and wages Consumer price inflation is estimated at. per cent in,. per cent in 3 and 1, per cent in, cf. figure 1.3b. The estimate for has been revised upwards by.1 percentage point primarily due to higher than expected prices on services and energy. In 3 the estimate has been revised upwards by. percentage points. This follows from a In national figures the unemployment rate will be 5. and.9 per cent in and 3, respectively. Economic Survey December

9 slower decline in the inflation rate of services than earlier expected, thus postponing the expected downturn in inflation. The fall in consumer price inflation from to originates from a fall in the contribution from the underlying inflation a national measure of market determined domestic inflation. Among other things this measure is influenced by lower growth of unit labour costs due to (the estimated) somewhat higher productivity growth and lower wage inflation. The fall in excise taxes on soft drinks, cigarettes and alcohol reduces the inflation rate in by,1 percentage point. Box 1.. Labour force growth, 1- In 3 and the labour force is expected to grow by 1. and. persons respectively. Along with the increase of 3. persons from to this implies that the labour force is expected to grow by 3. persons from to. This is in line with the estimates of the possible development from the government policy strategy ranging to 1. Though in the years to come there is an underlying downward pressure on the labour force due to the demographic development (that is, for instance, a relative increase of older people having relatively low participation rates) it is expected that participation rates will increase. Increasing labour force participation is expected to follow from previous falls in the intake of people leaving the labour force due to early retirement schemes. Also there is an expected effect from the reforms of these schemes and the recent action plan More people in employment. Table a. Contribution to changes in the labour force, - Demographic development Increased participation Total change , persons Labour force Transfer income recipients and others outside the labour force Transitional schemes Early retirement schemes Disability retirement schemes Other year-old Note: The basic scenario (Demographic development) is calculated by keeping the share of the population (according to age, gender, and origin) in the different labour market categories unchanged at the level. Source: Own calculations and Economic Survey, January. Hourly wages in the private sector are expected to grow by.1 per cent in and 3.9 per cent in 3. Compared to the estimates in Economic Survey December 9

10 august the estimate for 3 has been revised a little downward. This is primary due to the expected rise in unemployment in 3. Hourly wages are expected to grow by 3.7 per cent in. Despite the expected gradual decline in the growth of wages the hourly compensation of employees in Denmark remains higher than the average of the other EU-countries. Table 1.. A comparison between the present and the previous forecast. 3 August December August December December Percentage changes from previous year Private consumption Total government demand of which government consumption of which government investment Residential construction Fixed business investment Domestic demand exc. stock building Changes in stock building 1) Total domestic demand Exports of goods and services Of which manufactures Total demand Imports of goods and services Of which goods ) Gross domestic product GDP at factor costs Of which in non-agricultural sector Changes in thousand persons Labour force Employment of which in the private sector of which in the public sector Unemployment ) The volumes indicate volume changes in stock building in per cent of volume GDP in the preceding year. ) Excluding production of oil and gas and maritime. (Continues) 1 Economic Survey December

11 Table 1.. A comparison between the present and the previous forecast (continued). 3 August December August December December Percentage changes from previous year Merchandise export prices Merchandise import prices Merchandise terms of trade Real estate prices for one-family houses Consumer price index Hourly earnings Real disposable income, private sector Productivity in non-agricultural sector Savings ratio in the private sector year government bond yeat government bond Current account , 35. 9, 3, Government net lending , 3.1, 33,5 Unemployment (thousands) GDP growth in OECD countries ,5,9,5 3, International market growth... 1,, 7,,5, Dollar exchange rate (Dkr. Per US$). 7,9 7,9 7, 7, 7, Price of oil (Us-dollar per barrel) Price of oil (Dkr. per barrel)... 19, 197,5 19,3 197,9 19,3 Source: Statistics Denmark and own calculations.. The fiscal stance and government finances. The general government surplus is estimated at 1.5 per cent of GDP in, 1. per cent of GDP in 3 and. per cent of GDP in, cf. figure.1a. This is a downward revision of. and. percentage points in and 3, respectively, compared with the Survey in August, whereas estimates for is included for the first time in this Survey. Part of the revision for and 3 follows from a technical revision by Statistics Denmark of the national accounts for 1999 to 1 3, cf. box.1. The remaining downward revision of the government surplus is almost entirely due to a downward revision of the employment and higher unemployment. 3 See Statistiske Efterretninger, Offentlige finanser :9, Statistics Denmark. Economic Survey December 11

12 Box.1. Revisions of the government surplus. The technical revisions (according to new national account definitions) of the government surplus in and 3 concerns swap-arrangements where the public sector net expenditures are to be calculated excluding net interest settlements due to swap-arrangements. Further, the revision implies that imposed taxes which in fact is never paid by the private households and firms is to be written off. This is now treated as a capital transfer to the private sector. The technical revisions increase the central and local government expenditure burden and reduce the measured general government surplus by approximately.5 per cent of GDP from 1999 onwards. The deterioration of the public finances is completely (and also technically) offset by a lower future net interest burden of the public sector net debt. Thereby, the technical revisions do not impact the sustainability of the public finances. Gross public debt is expected to decline to 39. per cent of GDP by the end of compared to.7 per cent by the end of 1, cf. figure.1b. Figure.1a. General government surplus in Denmark and EU. of GDP of GDP Figure.1b. Gross public debt ratio in Denmark and EU. of GDP of GDP Denmark EU Denmark EU-15 3 Source: Statistics Denmark, EU-Commission, and own calculations. Fiscal policy in 3 is based on the budgets of the local governments and the central government budget agreement for 3. The fiscal effect, which measures the first year impact of discretionary fiscal policy changes on GDP, is estimated at. per cent in and.1 per cent in 3, cf. table.1. The fiscal policy for is based on technical assumptions in line with the target in the Government 1- plan. The actual fiscal policy for is to be determined in 3. GDP-growth is also influenced by fiscal policy measures in previous years including continued structural effects from the saving incentives due to the adjustment of income taxation adopted in June 199 (Whitsun 1 Economic Survey December

13 package). The cumulated activity effect and structural saving effects are expected to dampen activity by.1 per cent in and 3. Thus, the effect on economic activity of overall economic policy is slightly negative for and neutral for 3, cf. annex A1. Table.1. Activity effect of fiscal policy and structural saving effects Fiscal effect Percentage points Cumulated activity effect of fiscal policy Structural saving effects Source: Statistics Denmark and own calculations. This level of fiscal tightness is considered appropriate in view of the current cyclical position of the Danish economy, also taking into account that fiscal policy must be conducted given the monetary policy implied by the fixed exchange rate in the ERM II. The fixed exchange rate implies that monetary policy is effectively determined by the ECB. Thus, the fiscal stance should be considered in the context of sharp interest rate cuts internationally during 1 and the renewed decline of interest rates during the summer and autumn of, a low level of unemployment, and a relatively high level of capacity utilization. Hence, the estimated level of the output gap is still positive, cf. figure.a, and annual wage increases are expected to be approximately 1 percentage point higher than abroad, cf. figure.b. Figure.a. Output gap. Figure.b. Wage increases. Pct. Pct Danmark Euro Area Denmark EU-15 Trading partners Source: Statistics Denmark, EU-Commission, and own calculations. Economic Survey December 13

14 General government finances in a long term perspective It is the Government's objective to reduce public debt and thereby ensuring the long-term sustainability of government finances in the prospects of future demographic changes. Therefore, the size of the current budget surplus must be compared to the future obligations of the public sector, including expenditures for pensions and care for the elderly. The required surplus on the budget balance adjusted for these future obligations is calculated on the basis of the latest demographic projection. It is estimated that a surplus in the range 1½ to ½ per cent of GDP on average until 1 will ensure that fiscal policy is on a sustainable path. The estimated structural surplus is around per cent in both and 3, which is in line with the medium term target, cf. annex A1. 1 Economic Survey December

15 Annex A1. Fiscal policy Fiscal stance Since 1997 the fiscal stance measured by the cumulated fiscal policy contribution to GDP-growth has been negative, cf. table A1.1 and Box A1.1. In and 3 the effects of tight fiscal policy in previous year is expected to ease and the fiscal stance will become more neutral. Table A1.1 Contributions to GDP-growth GDP-growth Fiscal effect Cumulated effects, hereby: Percentage points Economic policy Cumulated fiscal policy Structural saving effects Changing interest rates Other factors Note: The cumulated effects of fiscal policy embrace the impact on growth from policy changes in present and previous years. Source: Statistics Denmark and own calculations. Box A1.1. Contributions to GDP-growth from cumulated effects of fiscal policy and changes of the interest rate, In GDP-growth was decreased by the cumulated effects of fiscal policy tightening including the structural effects of lower tax value of households interest expenditures. In 1 both higher interest rates and fiscal policy tightening has contributed to dampen the growth of GDP For year a modest tightening of the fiscal policy is expected to dampen growth while lower interest rates are supposed to increase the growth of GDP in both and 3. Figure a. Contributions to GDP-growth from cumulated effects of changes in the interest rates and fiscal policy, percentage points. Fiscal loosening Negative effect from interest rate Fiscal policy 1 3 Fiscal loosening Positive effect from interest rate Interest rate effects Fiscal tightening Fiscal tightening Negative effect from - Positive effect from interest rate interest rate Source: Own calculations. Economic Survey December 15

16 Fiscal sustainability The structural surplus is estimated at 1.7 per cent of GDP in and 1.9 per cent of GDP in 33 and, cf. table A1.. This is, considering the uncertainty, at the estimated requirement for long-term sustainability, and in the middle of the medium term target range of 1½ and ½ per cent. Thus fiscal policy is assessed to be on a sustainable path. Table A1.. Sustainability of the public finances per cent of GDP. 3 Stance of public finances: General Government surplus hereby social funds hereby central and local government Central and local Governments net interest expenditures Central and local government primary surplus Deviations between actual and structural primary surplus Central and local government structural primary surplus Requirement for central and local government primary surplus: Public consumption, tax freeze 1 and lower taxes on labour Income 1) Increase in net taxes from private pension funds Taxes from North Sea oil and gas production..... Burden of net expenditures on public net debt Requirement of central and local government structural primary surplus Sustainability indicator ) After 1 a technically assumption of increases in underlying tax revenues in line with GDP-growth is applied. Note: The increase in future net liabilities is calculated as the net present value of: Source: Own calculations. 1 Economic Survey December

17 A. The accuracy of short term forecast The following section examines the (ex post) accuracy of the short-term economic projections made by the Danish government between 19 and 1. Forecasts of GDP-growth track the actual growth rate fairly well, cf. figure A.1a. However, in some years there are large deviations between the forecast and the actual value especially for the year ahead forecasts. Figure A.1a. GDP-forecasts, annual growth rate Actual value Current year Year ahead Figure A.1b. Inflation forecasts Actual value Current year Year ahead Source: Economic Survey (1979-1) and Statistics Denmark. For GDP-growth the forecast error is on average close to zero indicating no systematic bias in the forecast, cf. table A.1. The average absolute forecast error and the root mean square error (RMSE), which are measures of forecasts precision, are.9 and 1.1 percentage point respectively for the year ahead forecast of GDP. The magnitude of forecast precision rises to about,5 percentage points, as the forecast horizon gets shorter. The mean absolute errors of the forecast for the following year are 5-7 per cent of the normal fluctuation in the GDP growth rate. Forecasts of inflation have an average absolute error of about, and,3 percentage points for the year ahead and the current year forecasts, respectively. This low degree of uncertainty is partly due to the relatively low volatility in the inflation rate. The average absolute error of unemployment forecast is about,5 percentage points for the year ahead forecast, which is about ½ of the actual volatility in the unemployment rate. Economic Survey December 17

18 Table A.1. Government economic projections, Forecast horizon years ahead Year ahead Current year Volatility Autumn Spring Autumn Spring Autumn Number of observations Percentage points GDP-growth: Mean forecast error ,3 -,3 -,1 -,1, RMSE ,1,5 1,11,7,59 Average absolute error... 1,5 1,1,,,,5 - compared to volatility..,1,5,7,9,1 Inflation: Mean forecast error , -,7,13, -, RMSE... -,3,,9,5,55 Average absolute error...,1,51,53,7,5, - compared to volatility.. -,3,5,9,5,9 Unemployment rate: Mean forecast error ,17 -,1 -, -, -, RMSE ,,95,7,1,15 Average absolute error... 1,1 1,,1,5,31,1 - compared to volatility.. -,93,71,,7,1 Note: The forecast error is the difference between the first available realization of the variable (first settled estimate) and the forecast. Volatility is calculated as the average absolute deviation from the mean. Source: Economic Survey, The current year forecast of GDP, inflation and unemployment are all superior to the naïve alternatives, cf. box A.1. This is also the case for the year ahead, although the differences are smaller. Box A.1. Comparison to naïve forecasts. Forecast evaluation traditionally looks to some alternative forecasting procedures to provide a benchmark against which to compare the forecasts under examination. In this case two naïve alternative forecasts are considered: The forecast is equal to the historical mean of the preceding 1 years. The forecast is equal to the last available realization of the variable. Forecasts are compared using the Theil-statistic, which is computed as the ratio of the RMSE of the forecast in question to the RMSE of the naïve alternative. Theil-statistics less than 1 are said to beat the naïve forecast. Forecasts made within the year are for all the variables superior to the alternative forecasts. The relative accuracy of the forecast deteriorates for the year ahead forecast, but continues to outperform the naïve forecasts. The Theil-statistic for the year ahead GDP-forecast is /3 percentage points compare to alternative 1. This also indicates that 1 Economic Survey December

19 the GDP-forecast captures about 1/3 of the overall variation in the GDP-growth rate. Table a. Comparison to naïve forecasts. Forecast horizon Current year Year ahead Alternative 1 Alternative Alternative 1 Alternative Percentage points GDP-growth...,3,,7, Inflation...,19,,33,9 Unemployment rate...,5,1,5,39 Source: Economic Survey, Comparison forecast accuracy with EU-commission and OECD Compared with the forecasts made for Denmark by the EU-Commission and OECD there are only small differences in the forecast accuracy. Apart from the OECD-forecast the average absolute error of the GDP-forecast is below 1 percentage point, cf. table A.. The growth forecasts of the EU-Commission overshoots the actual GDP-growth rate by around, percentage points. Table A.. Forecast comparison (year ahead autumn forecast). Economic Survey EU OECD Percentage points GDP-growth: Mean forecast error... -,1 -, -,1 RMSE... 1,11 1,1 1,3 Average absolute error...,, 1, Inflation: Mean forecast error...,13,37,13 RMSE...,9,7 1, Average absolute error...,7,3, Unemployment rate: Mean forecast error... -, -,15 -, RMSE...,7 1,5,1 Average absolute error...,5,75,7 Source: Economic Survey, EU-commission forecast and OECD Economic Outlook. EU-forecasts of the inflation rate have the lowest absolute error, but the forecasts have a tendency to undershoot the actual inflation rate of about, percentage points. Forecast from both OECD and Economic Survey have a mean forecast error close to zero and an absolute average error around, percentage points. Economic Survey December 19

20 The accuracy of the unemployment forecasts in the Economic Survey is high compared to the forecast of EU and OECD. Furthermore, the forecasts of EU and OECD both have a slight tendency to overshoot the actual unemployment rate. Economic Survey December