Danmarks Nationalbank

Transcription

1 Danmarks Nationalbank Monetary Review 4th Quarter 2006 D A N M A R K S 4 N A T I O N A L B A N K

2 Danmarks Nationalbank Monetary Review 4th Quarter 2006

3 MONETARY REVIEW 4th QUARTER 2006 The small picture on the front cover shows the "Bankers" clock, which was designed by Arne Jacobsen for the Danmarks Nationalbank building. Text may be copied from this publication provided that Danmarks Nationalbank is specifically stated as the source. Changes to or misrepresentation of the content are not permitted. The Monetary Review is available on Danmarks Nationalbank's website: under publications. Managing Editor: Jens Thomsen Editor: Anders Møller Christensen This edition closed for contributions on 24 November, The Monetary Review can be ordered from: Danmarks Nationalbank, Information Desk, Havnegade 5, DK-1093 Copenhagen K. Telephone (direct) or Inquiries: Monday-Friday 9.00 a.m.-4 p.m. Datagraf Auning A/S ISSN (Online) ISSN

4 Contents Recent Economic and Monetary Trends... 1 The Cyclicality of Domestic Prices Bo William Hansen and Dan Knudsen, Economics This article views the IMI index in relation to cyclical fluctuations and to other price and wage indices, while also discussing whether IMI can be seen as an indicator of future inflation. In addition, delineation of the IMI index is discussed in the Appendix. The USA's External Imbalance in a Financial Perspective Per Flink Iversen, Economics This article looks into the financing of the USA's current-account deficit and international investment position. Particularly Asian and OPEC economies have played an important role. In addition, the article seeks to elucidate why the USA's net capital income is still positive despite its large external debt. The "dark matter" hypothesis has been offered as an explanation. Slovenia: An Economic Portrait of the New Euro Area Member State Niels Peter Hahnemann, Economics On 1 January Slovenia is the first of the new EU member states to join the euro after only just over 2½ years' membership. Slovenia is the most prosperous of the former Communist states. Its economy has been stable for a prolonged period with moderate inflation and relatively modest deficits on both its government budget balance and current account of the balance of payments. The article reviews Slovenia's compliance with the convergence criteria and the structure of its economy. The Relevance of GDP Growth Rates Christian Ølgaard, Economics An evaluation of a country's economic performance often focuses on GDP growth. However, this is associated with several pitfalls. This article argues that adjustment of GDP growth for demographic factors and terms of trade gives a truer measure of the country's economic performance. Settlement of Foreign-Exchange Transactions Lone Natorp, Market Operations and Tina Skotte Sørensen, Payment Systems The article describes various foreign-exchange settlement methods, including settlement via CLS, and the related risks. BIS has launched an international survey of settlement risk in relation to foreign-exchange transactions. Some of the results of the Danish part of the survey are outlined.

5 IMF Review of the Financial Sector in Denmark Gitte Wallin Pedersen, International Relations In Denmark was subject to a detailed IMF review of its financial sector, FSAP. Overall, the IMF found the systems to be resilient and well-supervised. The IMF's review should be viewed in the light of the systems' importance to economic stability. Articles in the Monetary Review 4th Quarter rd Quarter Press Release Tables and Graphs Section Vol. XLV, No.4

6 1 Recent Economic and Monetary Trends This review covers the period from the beginning of September to late November SUMMARY Growth in the global economy is high and broadly based. The upswing in the USA has moderated in the last few quarters, while the euro area has gained ground, and there are upswings in the UK and the Scandinavian countries. Oil prices declined from the beginning of August to the middle of November, and long-term interest rates have generally decreased since July. The drop in oil prices has contributed to curbing price increases, and despite high global growth inflation is moderate. The Danish economy continues to boom, with strong growth in both domestic and external demand. The capacity pressures have intensified considerably, and imports have increased, but Denmark continues to show a sound current-account surplus due to rising oil exports, among other factors. Employment has risen rapidly in 2006, and unemployment has fallen to the lowest level for more than 30 years. The pressure on the labour market has intensified. All regions, and more and more sectors, including large parts of the public sector, report shortages of labour, and wage increases have accelerated during the year. In 2007, the pressure on the Danish labour market is expected to intensify further, which will increase the competition for labour. Labour-market flexibility will be put to the test, and there is a considerable risk of detrimental overheating of the Danish economy. INTERNATIONAL ECONOMIC BACKGROUND USA In the USA, economic growth has abated after several years of strong cyclical development. According to preliminary national accounts data, economic growth was 0.4 per cent from the 2nd to the 3rd quarter of This is the lowest quarterly growth rate since the 1st quarter of The moderation in economic growth is especially attributable to a cooling of the housing market after strong price increases for a number of years. House prices were lower in August and September than one year before, cf. Chart 1, and the last two quarters have seen declining

7 2 Monetary Review 4th Quarter 2006 ANNUAL GROWTH IN US HOUSING PRICES Chart 1 Per cent, year-on-year Existing home sales, median price Note: Median observation of registered sales prices for existing homes. Source: EcoWin. residential investments. The housing market's cooling was not reflected in private consumption, however, which continued its unabated growth in the 3rd quarter of In this period, private consumption and business investments were stimulated by decreasing oil prices and long-term yields, and consumers and business enterprises retain an optimistic nearterm outlook, albeit less optimistic than in the spring. Considerable imbalances still characterise the US economy. Despite the significant improvement of the federal budget, the US deficit was 248 billion dollars or just over 2 per cent of GDP in the fiscal year 2006 which ended on 30 September. The principal factor behind the improvement of government finances is higher tax revenue than expected, in view of the favourable economic development. The current-account deficit has remained almost unchanged in 2006, amounting to 6.6 per cent of GDP in the 2nd quarter. In contrast, the OPEC countries and Asia posted strong surpluses, cf. the article on the US balance of payments on p. 33ff. The considerable, and growing, goods and capital imbalances in the regions of the world, comprising very large deficits in some regions and very large surpluses in others, are hardly likely to be sustained in the long term, cf. Box 1. US employment rose in the first 10 months of the year, and unemployment has dropped to a low level, intensifying capacity pressures on the economy. Wage increases have been moderate during the upswing, although some indicators point to higher wage increases, and the

8 3 GLOBAL IMBALANCES Box 1 The US current-account deficit has increased considerably since 2001, amounting to 792 billion dollars in 2005, equivalent to 6.4 per cent of GDP. The increase is primarily attributable to the course of the trade balance. The US current-account deficit is the most obvious sign of global imbalances. In particular it reflects the strong domestic demand and very low savings in the USA. Higher energy import costs in recent years have also contributed to the growing deficit. The fact that a number of other countries have been willing to finance consumption in the USA, due to their growing savings ratios, has allowed the USA to build up the deficit. One of the underlying factors is the demographically determined preference for a high level of savings in Asia and increasingly also in Europe. 1 In addition, the high oil prices have boosted the current-account surpluses of the OPEC countries. So far, the global imbalances have not given rise to major problems in the global economy. Persistent large current-account deficits in the USA will entail a level of external debt that can be financed in the short term, but may not be sustainable in the long term. Sooner or later, the imbalances must be reduced, and there are several courses to take. However, it is hardly realistic to expect that expansion of US exports alone can redress the deficit, in view of the size of US imports compared to exports. It is necessary to achieve a higher savings ratio in the USA, for both households and the public sector. Tightening fiscal policy would help to achieve this. A lower propensity to save in Asia and the OPEC countries would contribute to restoring global balances. The economic slowdown in the USA is likely to reduce US imports and gradually improve the balance of payments. However, this is a cyclical downturn that cannot in itself redress the imbalances in the longer term. Whether the imbalances are redressed by sudden and extensive adjustment or prolonged moderate adjustment is of great significance to the global economy and especially to the US economy. Gradual adjustment is most likely since outside the USA the propensity to save will hardly begin to decrease for some time yet, due to demographic factors. A rapid redressing of the global imbalances may on the other hand have extensive negative consequences for global growth as it would require significantly higher savings in the USA and thus lead to lower economic growth. This would affect growth in the rest of the world. Rapid adjustment of the imbalances is likely to weaken the dollar, reinforcing the adverse impact on global economic growth. In addition, significant depreciation of the dollar may lead to uncertainty and instability in the global financial markets. 1 Cf. e.g. (2005), World Economic Outlook, September and Ben Bernanke (2005), "The Global Saving Glut and the U.S. Current Account Deficit". Remarks at the Homer Jones Lecture, St. Louis, Missouri, April 14. domestic price pressure, measured by core inflation, has mounted since the beginning of the year. However, the moderation in economic growth may ease the price pressure. In October, prices were 1.3 per cent higher than in October last year in CPI terms, but 2.7 per cent higher excluding energy and food. Referring to the moderation in economic growth and receding inflationary pressures, the Federal Reserve held the fed funds target rate at

9 4 Monetary Review 4th Quarter per cent at its meetings in September and October. Most market participants expect the US fed funds target rate to remain at this level for the rest of the year. Japan and China In Japan, the economic upswing continued with high export growth, and business and consumer confidence was still high in the 3rd quarter of Notwithstanding the upswing, CPI inflation has been modest this year. In September 2006, consumer prices were 0.6 per cent above the level in September 2005, but consumer prices excluding fresh food and energy were 0.5 per cent lower than one year before. The Bank of Japan has kept its official interest rate unchanged since July when it was raised from zero to 0.25 per cent. In connection with Japan's zerointerest-rate policy, purchases of e.g. US securities have been extensively financed by yen-denominated loans. This "carry trade" has contributed to the yen's depreciation by around 5 per cent since the spring. The very strong growth in the Chinese economy continued in the first three quarters of 2006, with annual GDP growth rates of some 10 per cent. The upswing is driven by e.g. the strong increase in investments, leading to concerns about overinvestment. As a result, several political measures were implemented to dampen the propensity to invest. Furthermore, the People's Bank of China has raised its lending rate by a total of 54 basis points in 2006 so far. The trade surplus has grown further this year, and already after the first nine months the total trade surplus exceeded the surplus in 2005 overall. The substantial trade surpluses have brought a large inflow of foreign exchange to China. With the aim of preventing the renminbi from appreciating, the People's Bank of China has continued recent years' large-scale purchases of foreign exchange, and in September China's foreign-exchange reserves reached 988 billion dollars, equivalent to approximately 44 per cent of China's GDP. However, the People's Bank of China has allowed the renminbi to appreciate a little. In mid-november the renminbi was almost 5 per cent stronger against the dollar than in July 2005 when the People's Bank of China abandoned its strict fixed-exchange-rate policy vis-à-vis the dollar. Europe Economic activity in the euro area is increasing. In the 2nd quarter of 2006 GDP was 2.7 per cent higher than in the 2nd quarter of 2005, and the preliminary national accounts data point to continued growth in the 3rd quarter. Growth is primarily driven by domestic demand, in particular fixed investments. Growth in private consumption has been more

10 5 moderate, but consumer confidence in the euro area rose from June to October, and retail trading data indicates growth in private consumption in the 3rd quarter. From the turn of the year, VAT in Germany will be raised from 16 to 19 per cent for large areas of private consumption. The latter is therefore expected to rise up to the turn of the year. According to the preliminary national accounts data for the 3rd quarter, consumers are beginning to respond to the VAT increase. Industrial production rose during the summer in the euro area, and business confidence was still high in October, indicating that the favourable cyclical development will continue. Based on the two German confidence indicators, Ifo and ZEW, in the 3rd quarter of 2006, business enterprises and analysts in Germany expect slightly weaker growth six months ahead. This probably reflects Germany's VAT increase and expectations of higher financing costs due to tighter monetary policy. Fiscalpolicy tightening measures are expected in Italy too. The fiscal tightenings, the effect of interest-rate increases over the past year, and the contagion effects of the slowdown in the USA may dampen the economic growth in the euro area in The upswing in the euro area has increased employment and brought unemployment down to 7.8 per cent in September. In 2006 so far, wage increases have shown a slightly rising trend from a low level. Energy prices have decreased since the summer, and in September HICP inflation in the euro area fell below 2 per cent for the first time since the beginning of Core inflation, i.e. HICP excluding food, energy, alcohol and tobacco, has been relatively low and stable since the beginning of 2005, cf. Chart 2. The European Central Bank, ECB, raised the minimum bid rate by 25 basis points to 3.25 per cent in October on the grounds of upside risks to inflation in the euro area due to higher economic growth and strong growth in money and credit. This was the fourth increase of the ECB's official interest rate in The minimum bid rate is still low, and stimulates activity, while the money market has taken a further increase this year into account. In the UK, GDP grew by 0.7 per cent from the 2nd to the 3rd quarter of 2006, as in the preceding four quarters. The upswing has led to an expansion of the labour force, which has increased rapidly in The background includes immigration from especially the new EU member states, as the UK takes a very liberal stance on this immigration. The increased supply of labour has helped to curb wage increases. Since the spring, inflation has been just over the target of 2 per cent in terms of annual CPI inflation, and the Bank of England raised the official bank rate to 5.0 per cent at its meeting in November.

11 6 Monetary Review 4th Quarter 2006 INFLATION IN THE EURO AREA Chart 2 Per cent, year-on-year HICP HICP excluding energy, food, alcohol and tobacco Source: EcoWin. The Scandinavian economies are also gaining momentum. Both Sweden and Norway are in an upswing, and the confidence indicators point to a growing shortage of labour. Inflation has fallen below the central banks' targets, and wage increases have remained moderate in both countries so far. Sveriges Riksbank raised its monetary-policy interest rate by 25 basis points to 2.75 per cent in late October, and Norges Bank raised its monetary-policy interest rate to 3.25 per cent at the beginning of November. INTERNATIONAL COMMODITY AND FINANCIAL MARKETS Most commodity prices have increased in 2006 so far. In mid-november the oil price was above the average level in 2005, even though the oil price fell by 25 per cent from the beginning of August to October. Nonenergy commodity prices rose strongly in the first 10 months of the year, and in October the prices reached the highest level for many years in terms of the IMF index. In view of the moderation in economic growth in the USA and lower inflation expectations, 10-year US government bond yields began to decline in July 2006, and the decline continued in the autumn, cf. Chart 3. Short-term US interest rates have been relatively stable since July, reflecting that the Federal Reserve has kept the fed funds target rate unchanged in this period. Long-term yields in Germany have decreased a

12 7 SHORT-TERM AND LONG-TERM INTEREST RATES IN THE USA AND THE EURO AREA Chart 3 Per cent p.a Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Source: EcoWin. 10-year government bond yield, USA 3-month money-market interest rate, USA 10-year government bond yield, Germany 3-month money-market interest rate, euro area little since July, while short-term interest rates in the euro area have increased, thereby narrowing the spread between long-term and shortterm interest rates, cf. Chart 3. In both the USA and Europe, stock prices have risen almost continuously since June, and more than recovered the ground lost in the spring. The increase in stock prices can be attributed to such factors as high growth in earnings, receding long-term yields and market expectations that the US fed funds target rate has now peaked. When the fed funds target rate peaked in February 1989 and again in February 1995, stock prices rose in the following months. The dollar/euro exchange rate has been relatively stable since May within an interval from 1.25 to 1.30 dollars per euro. The dollar moved beyond this interval in the last part of November, to approximately 1.31 dollars per euro towards the end of the month. DANISH MONETARY AND FOREIGN-EXCHANGE CONDITIONS Since September the Danish krone has been stable against the euro at a level close to the central rate in ERM II of kr per euro. Danmarks Nationalbank did not intervene in the foreign-exchange market in September and October, and the foreign-exchange reserve was approximately kr. 180 billion at the end of October. Danmarks Nationalbank has only intervened on a few occasions and for modest

13 8 Monetary Review 4th Quarter 2006 amounts since mid-february when the lending rate was raised by 10 basis points in order to curb a considerable outflow of foreign exchange. Over the entire period since February 2006, the portfolio inflows to and outflows from Denmark have virtually balanced in overall terms. The maturity-adjusted differential between 10-year government bond yields in Denmark and Germany was stable at around 10 basis points from the spring to the beginning of September, after which it narrowed a little. On 6 October, mirroring the ECB's interest-rate increase, Danmarks Nationalbank raised its interest rates by 25 basis points. The increases were expected by the market, and no market reactions were observed. The Danish increase in October brought the lending rate to 3.50 per cent and the discount and current-account rates to 3.25 per cent. On 1 January 2007, Slovenia joins the Economic and Monetary Union, cf. the article on the Slovenian economy on p. 49ff. Slovenia thus leaves ERM II, which then consists of 7 member states and the ECB. The euro is at the core of ERM II, and the other participating currencies have central rates against the euro, but not against each other. If a participating currency reaches one of the fluctuation limits, only the central bank of the relevant member state and the ECB are obliged to intervene. The other participating member states have no intervention obligation in this situation. Slovenia's departure from ERM II therefore has no effect on the conditions for the Danish krone. THE DANISH ECONOMY The upswing in Denmark has continued in 2006 with strong and broadbased growth in demand. Private consumption and business investments have grown just as strongly as in Export growth has been robust, and Danish business enterprises gained market shares in the first half of the year. The shortage of labour has probably contributed to the rising corporate investments in machinery, software, etc. since the start of the upswing. The investment growth has enhanced production capacity, giving more room for the upswing. According to the confidence indicators for building and construction, industry and service, the high growth in the Danish economy will continue. Consumer confidence is also high, indicating continued expansion of private consumption. The households' optimism and expanding consumption mainly reflect higher disposable income. Other contributing factors are high job security, a continued low level of interest rates and the significant increase in home equity due to recent years' strong increase in housing prices. Since, during the current upswing, private consumption has grown at a

14 9 CONSUMPTION AND NET WORTH RATIOS Chart 4 Consumption ratio 0.80 Net worth ratio Consumption ratio Net worth ratio (right-hand axis) Note: The ratios are calculated as private consumption and household net worth in relation to total disposable income in the private sector. Net worth includes the value of homes, but excludes pension savings, and is calculated as at the beginning of the year is partly estimates. Source: Statistics Denmark and Danmarks Nationalbank. considerably slower pace than net worth, the households have only to a minor degree converted increasing capital gains to consumption. The proportion of income channelled to consumption is thus smaller than during the upswing in the mid-1980s, cf. Chart 4. Lending by banks and mortgage-credit institutes to households and business enterprises has increased by around 14 per cent year-on-year in 2006 so far. The households have embraced the new loan products, and deferred-amortisation loans are gaining more and more ground, in October accounting for almost 38 per cent of lending by mortgage-credit institutes for owner-occupied and holiday homes. The ratio of households' variable-rate loans from banks and mortgage-credit institutes has been stable at around 60 per cent since the beginning of 2005, but is greater than e.g. five years ago. Consequently, the households are more exposed than before to increases in short-term interest rates. Taking into account that the households also have interest income, the interestrate increases over the last year will reduce the households' disposable income by 0.3 per cent in 2007, cf. Box 2. The interest-rate increases particularly affect domestic demand via housing prices, net worth and investments, but the dampening effect in 2007 is expected to be limited. According to the statistics of the Association of Danish Mortgage Banks, the rising trend for housing prices continued in the 3rd quarter of 2006, albeit at a slower pace. Furthermore, the number of homes for

15 10 Monetary Review 4th Quarter 2006 EFFECT OF INTEREST-RATE INCREASES ON HOUSEHOLDS' INTEREST EXPENSES Box 2 Adjustable-rate loans have gained more and more ground in recent years, cf. the Chart. This has made the households' disposable income more vulnerable to fluctuations in short-term interest rates. Assessed on the basis of the interest-rate terms in mid-november 2006, refinancing of 1-year mortgage-credit loans can be expected to cost the households just over 1 per cent p.a. more in December 2006 than in December At the end of the 3rd quarter of 2006, households' loans from mortgage-credit institutes with interest-rate adjustment within one year totalled around kr. 500 billion. Furthermore, the households' loans from banks amounted to almost HOUSEHOLDS' LOANS FROM BANKS AND kr. 450 billion. The interest-rate terms MORTGAGE-CREDIT INSTITUTES for these loans can also be expected to Per cent generally follow the short-term interest rate. Against this background, the households' interest expenses before tax are estimated to be just over kr billion higher in 2007 than in 2006 as a result of the interest-rate increases Q during This is equivalent to just Bank loans Other mortgage-credit loans with interest-rate adjustment over kr. 7 billion or around 0.9 per Mortgage-credit loans with interest-rate adjustment (=<1 year) Fixed-rate mortgage-credit loans cent of private consumption after Note: End of period. adjustment for the taxation value of Source: Danmarks Nationalbank. interest expenses. At the same time, the households hold financial assets at variable interest rates. Households' deposits with banks totalled almost kr. 600 billion at the end of the 3rd quarter, and their holdings of very short-term mortgage-credit bonds issued to finance adjustable-rate loans amounted to a further kr. 30 billion on estimate. Taking this into account, the households' net interest expenses after tax are estimated to be only just over kr. 2 billion higher in 2007 than at the beginning of 2006, due to higher short-term interest rates. This corresponds to around 0.3 per cent of both disposable income and private consumption. Since some households have loans and deposits for which the interest rate is linked to the very short-term interest rate, the change from 2006 to 2007 in full-year terms is likely to be minor. The effect on consumption of changes in disposable income probably varies among the individual households, depending especially on whether the household has net interest income or net interest expenses. The liquidity situation is normally good for households with positive net interest income, and they also have good access to borrow, making them less vulnerable to changes in disposable income than households with net interest expenses. A reduction of the borrowers' disposable income is therefore likely to have a relatively stronger effect on consumption than an increase in the disposable income of households with positive net interest income. sale on estate agents' books has risen considerably since February, cf. Chart 5. The year-on-year rate of price increase is still high, but prices for owner-occupied flats in the Greater Copenhagen area stagnated from the 2nd to the 3rd quarter, and a number of factors indicate normal-

16 11 HOMES FOR SALE AT BOLIGSIDEN.DK AND INCREASE IN HOUSING PRICES Chart 5 1,000 homes Per cent, year-on-year Homes for sale Price of one-family and terraced houses (right) Price of owner-occupied flats (right) 0 Note: Most recent observation for homes for sale is 20 November Source: The Danish Association of Chartered Estate Agents and the Association of Danish Mortgage Banks. isation of this segment of the housing market after several years of strong price increases. The number of owner-occupied flats for sale has risen strongly in 2006, and it takes longer to sell them. In addition, more and more flats are sold at a reduced price compared to the initial sales price. NEW CONSTRUCTION OF MULTI-STOREY HOUSING Chart 6 1,000 m Greater Copenhagen region Rest of the country Note: Buildings under construction. Greater Copenhagen comprises the City of Copenhagen and the City of Frederiksberg, as well as the counties of Copenhagen, Frederiksborg and Roskilde. Source: Statistics Denmark, national accounts.

17 12 Monetary Review 4th Quarter 2006 In view of the high housing prices, new construction is an attractive alternative, and construction of new flats has risen considerably in 2006, particularly in the Greater Copenhagen area, cf. Chart 6. This expansion of supply has contributed to curbing price increases. Balance of payments, foreign trade and government finances Indicators point to a smaller current-account surplus than in In the first nine months of the year, the surplus totalled kr. 33 billion, which is almost kr. 11 billion less than in the same period last year. The strong domestic demand and limited capacity in the Danish economy have pushed up imports and reduced the trade and current-account surpluses. Oil prices and energy export values were higher in the autumn of 2006 than at the start of the upswing in Consequently, the upswing's adverse impact on the trade surplus is illustrated more clearly by the trade balance in constant prices, cf. Chart 7. The Chart shows that the trade surplus in constant prices has been almost halved since the upswing began. However, the surplus is still sound, and the fact that Denmark is a net exporter of oil and gas is not the only contributing factor, since exports of machinery and medical and pharmaceutical products have also risen considerably since In addition, the balance of payments is positively influenced by increased earnings from sea freight and rising investment income from abroad. DENMARK'S TRADE BALANCE Chart 7 Kr. billion Current prices Constant 2004 prices Note: 12-month moving sum. In the statistics at constant prices, seasonally-adjusted exports and imports are deflated by the respective unit value indices. Source: Statistics Denmark and own calculations.

18 13 GOVERNMENT BUDGET BALANCE AND SAVINGS BALANCES Box 3 The government budget balance has improved significantly since the start of the upswing in In 2005, the government surplus totalled approximately kr. 70 billion, or around 4.5 per cent of GDP, cf. the Chart. The surplus to a high degree reflects the strong economic development, which has led to increased tax revenue. In addition, the extraordinarily high revenue from pension yield tax and the taxation of oil and gas production in the North Sea should also be taken into account, cf. the main article. The government surplus can be attributed to the central-government surplus since the cyclical gains and the extraordinary tax revenue have first and foremost fallen to the central government. The regional and local authorities taken as one have seen increasing deficits in recent years. In 2005, the regional and local authorities posted deficits of respectively kr. 3 billion and kr. 2 billion, according to preliminary data. SAVINGS SURPLUSES Per cent of GDP Private Public Abroad Source: Statistics Denmark and own calculations. The private savings surplus has deteriorated in recent years and in 2005 was reversed to a deficit, cf. the Chart. 1 This means that the private sector's demand for goods and services was partly financed by reducing financial assets. It is normal for the private savings surplus to diminish during an upswing that is driven by consumption and investments. The private sector comprises both business enterprises and households. The deterioration in the savings balance in 2005 can be attributed primarily to the households, which invested in new homes, and whose consumption increased faster than disposable incomes. The decline in the households' savings ratio in 2005, cf. the text in Chart 4, can be regarded as a relatively modest reaction to the strong increase in home equity. The households' rapidly increasing borrowing has to some extent been offset by higher savings in financial assets, including pension savings. The business enterprises have financed the increasing investments via earnings, and maintained their financial savings in 2005.

19 14 Monetary Review 4th Quarter 2006 CONTINUED Box 3 For a number of years, the sum of private and public savings surpluses has been stable at around 3 per cent of GDP. This overall national savings surplus can, by definition, also be regarded as a savings deficit vis-à-vis abroad, cf. the Chart, and is by and large equivalent to the current-account surplus. 2 The balance of payments did not deteriorate to the same extent as was observed during the upswing in the 1990s. The explanation is that the higher revenue from sea freight, and especially from the North Sea activities, has either gone to the public sector as tax revenue or to a great extent remained in the corporate sector without being channelled to the households. Furthermore, the current upswing is characterised by lower growth in the consumption ratio than the development in the households' net worth would otherwise warrant. This has contributed to curbing the deterioration in the private savings balance to a less significant level than in previous upswings. 1 2 The savings surplus is calculated as savings less real investments and capital transfers, etc., and is a measure of (net) savings in financial claims. The savings surplus is also called financial savings or net borrowing/net lending. Modest net capital transfers, etc. to abroad account for the difference between the external savings deficit and the balance of payments. The large government surplus in 2005 gave Denmark a national savings surplus in overall terms, cf. Box 3. The Ministry of Finance's August forecast anticipates a surplus of kr. 50 billion in 2006, equivalent to 3.1 per cent of GDP, but the surplus could very well exceed this amount, as indicated by the larger central-government surplus than expected in September. The revenue from taxation of pension yields varies strongly from year to year. In 2002, it was close to zero, while in 2005 it accounted for 2.4 per cent of GDP, and is around 1 per cent of GDP on average if pension assets remain unchanged. The revenue from taxation of oil extraction in the North Sea may also vary considerably. The Ministry of Finance expects a revenue increase to more than 2 per cent of GDP in 2006 due to the high oil prices in 2006 so far. In the longer term, oil production in the North Sea will decline as the reserves are exhausted. At the same time, government expenditure on pensions, and especially on care of the elderly, will increase as a ratio of GDP in the coming years. In order to balance government expenditure and revenue in the long term, the most appropriate course is to use the present surplus to reduce government debt. Viewed in isolation, this will contribute to improving government finances in the longer term when the pressure on expenditure is expected to increase as a result of an ageing population. Labour market, wages and prices Employment has risen strongly over the last year, especially in the building and construction sector which attracts labour from other sectors as

20 15 BOTTLENECKS ACCORDING TO REGIONAL LABOUR MARKET COUNCILS Chart 8 Number Manufacturing Service Building and construction Public sector Note: The regional labour market councils define a bottleneck as more than temporary demand for labour with certain professional skills that cannot be met by resources among the unemployed. The number of bottlenecks is compiled on the basis of jobs available at AF job centres and reports of unfulfilled needs for labour not directly reported to the AF job centre system. Source: The Danish Labour Market Agency. well as other EU member states. In September, unemployment had declined to 117,100, i.e. 4.2 per cent of the labour force, or 3.5 per cent measured by the EU's harmonised definition. This was the lowest unemployment rate in the EU in September. The lower unemployment rate has increased the pressure on the labour market, and the regional labour market councils report a growing shortage of labour in all sectors in 2006, particularly in the construction and service sectors, cf. Chart 8. Danish business enterprises and institutions increasingly use foreign labour to counter the shortage. This is e.g. reflected in the number of employees from the new EU member states, which has increased rapidly since the enlargement of the EU in 2004, cf. Chart 9, and accounts for around 10 per cent of the growth in employment in this period. Furthermore, Danish business enterprises have hired a considerable number of new employees from Sweden and Germany. During the first 10 months of 2006, more than 2,500 Swedes took jobs in Copenhagen, i.e. twice as many as in the same period last year. The increased supply of foreign labour is also attributable to the higher number of foreign business enterprises operating in Denmark. The number of VAT-registered foreign building and construction enterprises has risen considerably in the course of the upswing. The number of employees is unknown.

21 16 Monetary Review 4th Quarter 2006 ACTIVE WORK PERMITS FOR CITIZENS OF NEW EU MEMBER STATES Chart 9 Number 9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1, Active work permits, total Agriculture, horticulture and forestry Other sectors Note: Seasonal adjustment of the series is not possible. Source: The Danish Labour Market Agency. The monthly statistics for work permits issued to citizens of the new EU member states, cf. Chart 9, disclose the number of eastern Europeans employed by Danish business enterprises. There are no equivalent timely statistics for the number of people from the old EU member states employed in Denmark as they do not require work permits. Despite the higher influx of foreign labour, the contribution to the total labour force, especially from the new EU member states, is still modest. The Danish government has launched a number of initiatives to accommodate the demand for labour, including initiatives to attract more foreign workers to Denmark. The aim of these initiatives is to help Danish employers recruit people abroad, and to provide better information about job opportunities in Denmark. However, it takes time for these measures to have an impact. The need for more foreign labour in Denmark is emphasised by the fact that the population groups with the highest participation rates are diminishing by approximately 20,000 per annum. This reduces the demographic basis for continued growth in the labour force and employment, cf. Chart 10. Nevertheless, the labour force has actually increased a little in recent years, which can be attributed to several factors. Firstly, approximately 40 per cent of the rapidly increasing group of year-olds are part of the labour force. The demographically determined labour force has thus

22 17 POPULATION OF YEAR-OLDS, EMPLOYMENT AND LABOUR FORCE Chart 10 Million people 3.1 Million people year-olds Total employment (right) Total labour force (right) 2.5 Note: Employment is calculated in accordance with the national accounts, thus comprising people over the age of 14. The labour force is the sum of employed and unemployed people. Estimates are applied as from Source: Statistics Denmark and own calculations. decreased by only approximately 6,000 per annum since Secondly, in recent years the participation rate of the over-55 age group has increased. Part of the explanation is the gradual phasing out of the access to transitional allowance. Thirdly, as stated, there has been a considerable influx of foreign labour to Denmark. Overall, the rising participation rates and the influx of foreign labour have contributed 10,000 people annually, whereby the labour force has increased by approximately 4,000 per annum since 2003, notwithstanding the unfavourable demographic trends. The participation rates of the older age groups may possibly continue to rise, and initiatives have been launched to attract more foreign labour to Denmark. The labour force will automatically diminish by 6,500 per annum from 2006 to 2010, if the age-determined participation rates remain constant. Maintaining the labour force at its current level therefore requires an equivalent influx of foreign labour or an increase in Danish participation rates, including for the older age groups. The current level of employment is expected to remain by and large unchanged, as it will be difficult to further reduce unemployment. The shortage of labour is apparent from the wage statistics of most sectors in 2006, especially building and construction. There is a risk of the pressure on the labour market generating higher wage increases.

23 18 Monetary Review 4th Quarter 2006 REAL WAGES AND NOMINAL WAGE INCREASES Chart = 100 Per cent, year-on-year Real wages Nominal wage increase (right) Note: Nominal wage increases are the annual growth in hourly wages in industry. Real wages are calculated as hourly wages in industry deflated by the consumer price index is an estimate. Source: Statistics Denmark and the Ministry of Finance. This applies both to the usual local negotiations and to this winter's collective bargaining in several major areas, including building and construction, as well as industry. Strong increases in unit labour costs will diminish opportunities for continued growth and rising real incomes. The strong, uninterrupted growth in real wages in the last decades has been associated with moderate wage increases, cf. Chart 11. The tighter labour market and intensified capacity pressure in 2006 have reinforced domestic price pressures, cf. the article on the cyclicality of domestic prices, p. 21ff. Although underlying inflation can thus be considered to be rising, consumer price inflation has been modest in the period. The explanation is that the external price pressure has abated as a consequence of lower energy prices, and in October the consumer price index, HICP, was only 1.4 per cent above the level in October All in all, the continued upswing and the expected trends in the labour force and in the employed groups will put the Danish labour market's flexibility to the test. The shortage of labour also curtails output, which can hardly grow more than productivity. The latter has risen by 1.9 per cent per annum over the last 15 years, measured in terms of GDP per employee. The pressure on the labour market will grow in the light of the expected increase in demand for Danish goods by per cent in 2007.

24 19 In the long term, it is in the interest of business enterprises and wage earners to prevent wage increases from jeopardising business earnings. Already at this point, the pressure on the labour market is so strong that there is a considerable risk of detrimental overheating, which will entail excessive wage increases. This tends to reduce competitiveness, so that the Danish economy may risk a hard landing in a few years' time, which could push down employment and cause unemployment to rise. The Finance Bill 2007 is aimed at 1.3 per cent growth in public consumption, i.e. 0.3 per cent more than in the original proposal. According to the Bill, the fiscal effect is neutral, so government finances will not contribute to curbing growth in domestic demand. This only serves to reinforce the need to increase the supply of labour.

25

26 21 The Cyclicality of Domestic Prices Bo William Hansen and Dan Knudsen, Economics INTRODUCTION Since 1984 Danmarks Nationalbank has on a regular basis published an index of domestic market-determined inflation (IMI), i.e. the overall consumer-price index stripped of energy, food, administered prices and imports. Compilation of the IMI price index is described by Hansen and Knudsen (2005). Inflation is the persistent increase in the general level of prices in an economy. The word "persistent" indicates that inflation cannot merely be measured as the year-on-year increase in the consumer price index. Various efforts are therefore made to rinse the consumer price index of random fluctuations so that the underlying trend in price increases becomes more visible. Energy and food are two categories where prices fluctuate substantially, and one approach is to eliminate these two categories from the consumer price index. The increase in this reduced index is referred to as core inflation. A more general approach is to eliminate or attach a low weight to any good or service for which the price has risen or fallen significantly. Like the index of core inflation, the IMI index eliminates energy and food, but it is important to stress that the IMI index is not aimed at capturing the underlying inflation trend. The primary purpose is to capture the part of the price development that is determined in a Danish market 1. Decisive to the usefulness of the IMI index is therefore not whether it shows a stable trend, i.e. price increases with low volatility, but that it can be related to economic development in Denmark. This article views the IMI index in relation to cyclical fluctuations and to other price and wage indices, while also discussing whether IMI can be seen as an indicator of future inflation. In addition, delineation of the IMI index is discussed in the Appendix. 1 Similarly, the Bank of England has for some years compiled an expression of "domestically generated inflation", showing price developments in the British market, cf. Bank of England (1998, 2006).

27 22 Monetary Review 4th Quarter 2006 THE PRICE OF VALUE ADDED The IMI index relates to the price of value added. The method applied when compiling the IMI index is to strip the overall consumer price index of components seen as "exogenous" to price formation in the domestic market. These are energy, food (including beverages and tobacco), administered prices and indirect taxes. In addition, the price impact from the import content in the remaining goods and services is eliminated in the IMI index, as is a small contribution from the indirect energy content, cf. Hansen and Knudsen (2005) on the calculation method. The calculation from consumer price index to IMI index is shown in Table 1. The aggregate eliminated items have considerable weight in private consumption, so that only just under 35 per cent of total price formation is attributable to the Danish market. It is a point of discussion what to eliminate from the overall price index in order to obtain the best index of the domestic market-determined price. Most goods categories contain both exogenous and domestic market-determined price elements. For example, food is subject to a high degree of domestic processing. The Appendix shows that if processed food and beverages are included, IMI increases to 45 per cent of private consumption, but the movement in the IMI index changes only slightly. If the impact from fluctuations in import prices is eliminated, the result is the price of value added on supplies for private consumption. The value added is equivalent to production less input of raw materials, and expresses the proceeds available for remuneration of the factors of production. The development in the IMI index thus corresponds to the development in wages and gross profits on supply of the goods and services included in the IMI index. FROM CONSUMER PRICE INDEX TO IMI INDEX Table 1 Consumer price index (overall inflation) - Energy and food Consumer price index excluding energy and food (core inflation) - Products with administered prices Consumer price index excluding energy, food and administered prices - Indirect taxes Index of net retail prices excluding energy, food and administered prices - Input-output calculated effect of energy price (indirect energy content) - Input-output calculated effect of import price excluding energy (import content) IMI index (domestic market-determined inflation)

28 23 IMI INDEX AND GROSS VALUE ADDED DEFLATOR Chart 1 Per cent, year-on-year IMI index GVA deflator for private non-agricultural sector Note: Annual averages. Source: Mona databank and own calculations. The IMI index relates to value added from the manufacture of goods for private consumption, less the value added in energy production, food production and some parts of the public sector. This is a broad definition, which includes large parts of the manufacturing, transport, post, telecommunications and health sectors, as well as hotels and restaurants and a number of enterprises within financing and business service. The building and construction, iron and metal and chemical industries, which primarily supply products for investment and export, are only included to a limited extent. We cannot really identify an IMI sector in the national accounts since the value-added deflators of the sectors, e.g. the electronics sector or textile industry, in the national accounts reflect not only supplies for private consumption, but also the price of supplies for investment, export and public consumption. The IMI index approximately resembles the deflator for the private non-agricultural sector 1, cf. Chart 1. The covariation with an official price index from Statistics Denmark is reassuring since the IMI calculation is based on the monthly consumer price indices and price indices for the domestic supply of goods and is therefore not necessarily close to the prices in the national accounts, which in principle reflect all price statistics. 1 Compiled as the overall private sector excluding the sectors agriculture, rent (housing), energy extraction and net exports of other services than tourism.

29 24 Monetary Review 4th Quarter 2006 THE IMI INDEX AND OTHER PRICE AND WAGE INDICES In many ways the IMI index resembles the overall consumer price index. For example, the average rate of increase in the IMI index has been 4.8 per cent year-on-year since the beginning of 1975, against 4.7 per cent in the consumer price index. In addition, both indices show that inflation has been lower after 1990 than before, cf. Chart 2. However, there are differences; for instance IMI has been fluctuating more than consumer prices in recent years. In the late 1970s, domestic inflation was substantially below overall inflation, which may have reflected the weak economy with high unemployment. In the mid-1980s the economy was booming, and domestic inflation was relatively high, while wages were accelerating, cf. Chart 3. The most recent gaps between the IMI index and the consumer price index were seen in 2000 and notably in 2005 when the IMI index actually fell year-on-year. This may reflect that the capacity pressure in 2005 was too moderate to allow price rises for imported goods and energy to be passed on to consumers. In 2006, the IMI index has begun to accelerate, cf. Recent Economic and Monetary Trends. In connection with the above episodes there tends to be a negative correlation between increases in the consumer price index and the IMI index. Particularly import prices separate the two indices. Thus, the CONSUMER PRICE INDEX AND IMI INDEX Chart 2 Per cent, year-on-year Consumer price index IMI index IMI index including imports Note: Quarterly observations. The most recent observations are from 3rd quarter Source: Mona databank and own calculations.

30 25 WAGES AND IMPORT AND ENERGY PRICES Chart 3 Per cent, year-on-year 25 Per cent, year-on-year Hourly wages in manufacturing Import prices Energy prices (right-hand axis) -30 Note: Quarterly observations. The most recent observations are from 3rd quarter Source: Mona databank and own calculations. development in the consumer price index is closer to the development in the IMI index before import prices are eliminated. As described above, the IMI index resembles a value-added deflator, and therefore payroll costs are a significant explanatory factor behind the development in the IMI index. In addition, the sluggishness in price formation means that the development in import and energy prices is reflected in the IMI index. More specifically, the IMI index drops in connection with import price hikes 1. The same applies to a value-added deflator if the rise in import prices is not immediately passed on fully to product prices. COVARIATION BETWEEN IMI AND CYCLICAL FLUCTUATIONS Since the IMI index comprises goods and services for which Danish market conditions determine the price formation, we would expect the IMI index to be more sensitive to domestic cyclical trends than e.g. the consumer price index. Moreover, we expect the reaction pattern of a price index for value added to differ from that of a price index for consumer goods and services. In order to investigate this further we will look into the covariation between GDP 2 at constant prices on the one hand and the level of vari- 1 2 For a description of a behavioural relation for the IMI index, see Danmarks Nationalbank (2003). Similar results are achieved by replacing GDP with private consumption or domestic demand.

31 26 Monetary Review 4th Quarter 2006 COVARIATION WITH GDP Chart 4 Correlation (GDP(t), Price variable (t+j)) IMI index Hourly wages in manufacturing IMI index including imports Consumer price index Note: The X axis shows quarters. The correlation coefficients relate to Source: Mona databank and own calculations. ous price and wage indices on the other hand. The period under review is since Denmark's economic policy was restructured in the early 1980s. All series are seasonally adjusted quarterly data, i.e. a total of 64 observations. To start with, each of the quarterly series is broken down into a cyclical component and a trend 1. Both GDP and price and wage indices have a pronounced increasing trend. Then the cyclical development is studied by viewing the correlation pattern for the cyclical components. The simple correlations between cyclical components describe the pattern of the covariation between variables for real activity and nominal variables. Chart 4 shows the correlation between the cyclical GDP component and the various indices with a lead of up to 12 quarters. The correlation between GDP and the consumer price index is negative from the outset and remains so for some quarters. Six quarters later, for example, the correlation between GDP and the consumer price index is still marginally negative. Only after eight quarters does this correlation increase notably. In other words, it takes some time before an inflationary effect of high GDP filters through to the overall consumer price. Cyclical developments in consumer prices are one of the controversial 1 Time series are broken down into cyclical components and trends using the "HP filter", cf. Hodrick and Prescott (1997). The data period for the calculation of the filtered series is Since it is difficult for the HP filter to estimate the trend and cyclical component close to the end points, 12 observations are eliminated at either end.

32 27 topics in the study of business cycles, cf. Kydland and Prescott (1990), who believed that the negative correlation showed that the business cycles were mainly created by supply shocks. In contrast, a positive correlation between GDP and the IMI index is seen at an earlier stage. This could indicate that the IMI index reflects activity pressure more rapidly. In the first quarters, however, the correlation is negative, and sudden hikes or dives in GDP cannot be expected to be reflected in the IMI index immediately. In the calculation of the transition from the consumer price index to the IMI index, particularly the subtraction of the import content sustains the correlation between GDP and the IMI index. This is reflected in the fact that an interim result where the import content has not yet been subtracted correlates negatively with GDP in a way that is closer to the GDP correlation with the consumer price index. The clear difference in the GDP correlations for, respectively, the IMI index and the IMI index including imports reflects how high import prices, i.e. poor terms of trade, have typically been linked to both low GDP and a low IMI index. Hourly wages can be seen as a domestic market-determined price formed in the Danish labour market. High cyclical GDP means that activity is high. This increases pressure on the labour market, so that a lagged positive correlation between GDP and wages should be seen. In the period reviewed, the correlation between GDP and hourly wages in manufacturing is indeed positive with a delay of four quarters. The COVARIATION WITH UNEMPLOYMENT Chart 5 Correlation (Unemployment(t), Price variable (t+j)) IMI index Hourly wages in manufacturing IMI index including imports Consumer price index Note: The X axis shows quarters. The correlation coefficients relate to Source: Mona databank and own calculations.

33 28 Monetary Review 4th Quarter 2006 effect from GDP on wages greatly resembles the effect from GDP on the IMI index. Instead of GDP, unemployment can be used to describe the state of the economy. Chart 5 shows the correlation between the cyclical component of unemployment and the nominal variables. If the impact on wages goes via the labour market, there must be a clearer and more rapid cyclical effect in the correlation pattern between the cyclical components of unemployment and wages than between GDP and wages. It is in fact seen that wages immediately have a negative correlation with unemployment. The IMI index's correlation with unemployment is weaker than the wage correlation, but the development over time resembles the wage correlation with unemployment. On the other hand, the correlation of the consumer price index with unemployment is visibly slower and weaker, and again we see that the pattern of the IMI index including imports resembles that of the consumer price index. The key information element in the compilation of the IMI index therefore seems to be elimination of import prices. THE IMI INDEX AS AN INDICATOR OF FUTURE INFLATION? If the IMI index reacts faster to cyclical developments than the consumer price index, this indicates that the IMI index, like wages, can predict the direction of overall consumer prices. A simple way to examine the predictive properties of IMI is to relate the current year-on-year increase in the consumer price index and the IMI index to the realised future consumer price inflation: π = α + β π + β π + ε t+ h 1 t 2 IMI t t Applying monthly data over a horizon of e.g. 12 months (h=12), the left side of the equation shows the year-on-year rate of increase in the consumer price index one year ahead. On the right side of the equation, many other explanatory variables can potentially be useful for predicting inflation. As a benchmark for our results, IMI is replaced in the equation by core inflation defined as the consumer price index excluding energy and food. Core inflation is typically less volatile than both the overall increase in consumer prices and IMI, cf. Chart 6, and internationally it is often used to predict fluctuations in the consumer price index. Table 2 presents a number of regression results for h=12 and h=24. The data period is January 1985 to September 2006.

34 29 REGRESSION RESULTS Table 2 Degree of explanation (R 2 ) Constant Consumer price inflation Core inflation IMI Left-hand variable: Inflation, 12-month lead (3.9) 0.10 (0.6) 0.45 (3.5) - Inflation, 12-month lead (4.8) 0.23 (1.9) (4.3) Inflation, 24-month lead (4.8) 0.24 (1.4) 0.20 (1.2) - Inflation, 24-month lead (6.0) 0.09 (0.9) (3.8) Note: Left-hand variable is consumer price inflation. The data period is January 1985 to September The Table presents the estimated coefficients and the degrees of explanation. Heteroscedastically adjusted and autocorrelation-adjusted t statistics in brackets. Source: Statistics Denmark and own calculations. With a horizon of 12 months, both core inflation and IMI contribute to predicting consumer price inflation. In both cases, the estimated coefficient is significant, and the degree of explanation is slightly higher when IMI is included in the equation. At h=24, the degree of explanation declines, particularly for core inflation. At the same time we see that only the coefficient of IMI is significant, and thus the regression analysis indicates that a rise in IMI is often followed by a higher overall rate of increase in consumer prices. In other words, the development in IMI may contribute to predicting the development in consumer price inflation two years ahead, while core inflation does not have a similar explanatory power. CONSUMER PRICES, CORE INFLATION AND IMI Chart 6 Per cent, year-on-year Consumer prices Core inflation IMI Note: Monthly observations. The most recent observations are from September Source: Statistics Denmark, Eurostat and own calculations.

35 30 Monetary Review 4th Quarter 2006 CONCLUSION Our analysis shows that the IMI index is more exposed to cyclical fluctuations in Denmark than the overall consumer price index. The IMI index responds faster than the consumer price index to GDP and unemployment and in this respect resembles the index of hourly wages. In spite of its higher volatility, the year-on-year increase in the IMI index is a better indicator of overall consumer price inflation than core inflation is, particularly with a horizon of more than one year. The acceleration in the IMI index in 2006 reflects, inter alia, the tight labour market, and the rising domestic inflation pressure indicated by the acceleration in the IMI index will have an impact on overall inflation in LITERATURE Bank of England (1998), Inflation Report, August. Bank of England (2006), Minutes of The Monetary Policy Committee Meeting 5 & 6 July Danmarks Nationalbank (2003), MONA a quarterly model of the Danish economy. Hansen, B. W. and D. Knudsen (2005), Domestic Market-Determined Inflation, Danmarks Nationalbank, Monetary Review, 4th Quarter. Hodrick, R. J. and E. C. Prescott (1997), Postwar U.S. Business Cycles: An Empirical Investigation, Journal of Money, Credit and Banking, Vol. 29, No. 1. Kydland, F. E. and E. C. Prescott (1990), Business Cycles, Real Facts and a Monetary Myth, Federal Reserve Bank of Minneapolis Quarterly Review, Vol. 14, No. 2.

36 31 Appendix: A Broader Delineation of IMI All food and beverages are subtracted in the calculation of IMI, cf. Hansen and Knudsen (2005). IMI is thus in line with the Eurostat expression of core inflation in which energy, food, and beverages and tobacco are eliminated from the HICP index. However, Eurostat also publishes another expression of core inflation, defined as HICP excluding energy and unprocessed food. The European Central Bank, ECB, often uses this expression in its analyses. If the calculation of IMI is linked to this expression of core inflation, the IMI index must be extended to include beverages and tobacco, and processed food. In the most recent weight basis for the index of net retail prices 1, food and beverages, including alcohol and tobacco, constitute 14.4 per cent of consumer spending, while unprocessed food (meat, fish, fruit and vegetables) only constitutes 5.1 per cent. The weight basis for IMI is thus increased by 9.3 ( ) percentage points to 60 per cent if only unprocessed food is eliminated. Chart 7 shows the increase in two stripped indices of net retail prices in the period January 2001 to July In spite of their different content, the two indices show a considerable degree of covariation. The consumption components (IMI products) represented in the stripped index of net retail prices are identified in Statistics Denmark's input-output table for At the most detailed level, private consumption is broken down into 73 components. If processed food and beverages and tobacco are included, the IMI sector increases from 41 to 55 of these consumption components, but in spite of the different composition, the input-output weights do not change materially. Import prices are measured using price indices from the statistics for the domestic supply of goods. "Processed food" and "beverages and tobacco" are represented by the "raw materials for agriculture" and "beverages and tobacco" import price indices. Processed consumer goods (e.g. dairy products) are also included in the "foods of animal origin, excl. fish" import price index, but since this index is dominated by a number of unprocessed meat products, it is of no avail. The "new" import price indices are co-weighed with the import price indices used so far, "raw materials for other industries" and "other consumer 1 The most recent update of the weight basis for Statistics Denmark's index of net retail prices took place in January 2006, based on the consumption pattern in 2003.

37 32 Monetary Review 4th Quarter 2006 DIFFERENT DEFINITIONS OF INDEX OF NET RETAIL PRICES AND IMI INDEX Chart 7 Per cent, year-on-year Index of net retail prices excluding energy, unprocessed food and administtered prices Index of net retail prices excluding energy, food and administered prices Per cent, year-on-year IMI Alternative definition of IMI Note: Monthly observations. Source: Statistics Denmark and own calculations. goods". The indices are co-weighed in accordance with the weight distribution (from 2000) in the price index for the domestic supply of goods: Import price index capital goods = (26.1*"raw materials for other industries" + 1.6*"raw materials for agriculture")/( ) Import price index consumer goods = (12.3*"other consumer goods" + 0.7*"beverages and tobacco")/( ) Since the weights of the two new indices are relatively modest (1.6 and 0.7 per cent, respectively), the import price indices do not change significantly, cf. Chart 8. Neither the stripped index of net retail prices, the input-output weights nor the import price indices change materially if processed food and beverages and tobacco are included in the calculation of IMI. Even though the IMI products' total share of consumer spending increases from just under 35 per cent to around 45 per cent, the impact from the exercise on the IMI index is limited, cf. Chart 7. DIFFERENT DEFINITIONS OF IMPORT PRICE INDEX Chart 8 Per cent, year-on-year Raw materials for other industries Raw materials for other industries + Raw materials for agriculture Per cent, year-on-year Other consumer goods Other consumer goods + beverages and tobacco Note: Monthly observations. Source: Statistics Denmark and own calculations.

38 33 The USA's External Imbalance in a Financial Perspective Per Flink Iversen, Economics INTRODUCTION AND SUMMARY There has been considerable focus on global imbalances, above all the USA's external imbalance, in recent years. This article looks into two remarkable circumstances in relation to the financial aspects of the USA's current-account deficit and international investment position. Firstly, it is interesting that the world's largest economy, the USA, has been able to finance a very large current-account deficit without this leading to significant instability in the international financial markets. The article therefore examines the financing of the USA's external deficit. Particularly Asian and OPEC economies have had substantial financial assets to invest, while a large proportion of the USA's borrowing has been via sale of Treasury securities. So the USA's external debt primarily comprises liabilities that are sensitive to fluctuations in interest rates, and higher global interest rates will therefore reduce the USA's net capital income significantly. Secondly, it is remarkable that the USA's net capital income is still positive, despite its large external debt. One reason is that returns on foreign direct investments (FDIs) from the USA are significantly higher than returns on FDIs in the USA. It is difficult to state one specific reason for this difference, however, since it seems unrealistic that US enterprises operating abroad can systematically generate higher returns than foreign enterprises operating in the USA. The obvious conclusion is therefore that one or more of the statistics used are problematic. This article reviews some of the explanations offered by various commentators, including the discussion of "dark matter". In any case, the USA's current-account deficit is substantial, and its external debt considerable. In a future perspective, a sustained currentaccount deficit will increase the external debt further, and net capital income will decline. As a consequence of the internationally integrated financial markets and large gross capital flows it is not given, however, that the USA will have difficulties financing its external deficit in the future.

39 34 Monetary Review 4th Quarter 2006 USA's BALANCE OF PAYMENTS Chart 1 Billion dollars Current account Trade balance Net income Other transfers Note: Net income comprises net capital income and net wage income. Source: U.S. Department of Commerce, Bureau of Economic Analysis. FINANCING THE US CURRENT-ACCOUNT DEFICIT Current-account deficits and global imbalances With 1991 as the exception, the USA has had a current-account deficit every single year in the period , cf. Chart 1. Moreover, the deficit has grown year by year, reaching 792 billion dollars, or 6.4 per cent of GDP, in The growing current-account deficit is primarily attributable to the development in the trade balance. In view of the current-account deficits, the USA has accumulated considerable external debt, whereas before it was a net creditor, cf. Chart 2. Particularly since the mid-1990s there has been a substantial increase in foreign net investments in or net lending to the USA. In 2005, the USA owed more than 2,500 billion dollars, equivalent to just over 20 per cent of GDP. The US current-account deficit is the most visible characteristic of the global imbalances and is mainly related to high domestic demand in the USA. A low interest rate and higher asset prices have increased demand and been key factors behind the deterioration of the balance of payments in recent years. At the same time, large parts of the global economy outside the USA have seen growing savings ratios and have thus been willing to finance US consumption by acquiring ever larger portfolios of US bonds. This has been referred to as a global "saving glut". 1 1 See e.g. IMF, 2005 and Bernanke, 2005.

40 35 USA's EXTERNAL DEBT Chart 2 Billion dollars 3,000 Per cent of GDP 30 2, , , , , External debt External debt as a percentage of GDP (right-hand axis) Source: U.S. Department of Commerce, Bureau of Economic Analysis. Moreover, high oil prices have led to growing current-account surpluses in the oil-exporting countries. Who finances the USA's external deficit? The USA's current-account deficit is set off every year by an equivalent capital inflow from abroad. However, capital imports from a given country, e.g. Japan, need not correspond to the country's current-account surplus vis-à-vis the USA. The financing of the USA's external deficit may take a completely different course from the bilateral current-account surpluses and deficits, and the nature of the capital flows may thus have a separate impact on the USA's external imbalance. Owing to data availability, the following description of the development in the capital inflow to the USA is based on the development in net investments in longterm securities 1. Long-term securities make up a large share of the total foreign portfolio of US financial assets and of the total net capital inflow, which otherwise primarily comprises direct investments and financial investments with shorter maturities than one year. The development in the USA's financial net investments by the largest regions shows that most of the US capital imports in the late 1990s came 1 Long-term securities are securities with maturities of one year or more, cf. the Treasury International Capital System breakdown. Long-term securities include US Treasury securities, bonds issued by US government-owned and -subsidised institutions, US corporate bonds, US stocks, and foreign bonds and equities. Direct investments are not included. The breakdown of long-term securities is not in accordance with the IMF's Balance of Payments Manual from 1993 (BPM5).

41 36 Monetary Review 4th Quarter 2006 USA's FINANCIAL NET INVESTMENTS IN LONG-TERM SECURITIES Chart 3 Billion dollars Total Europe Asia Other Source: Treasury International Capital reporting system. from Europe, cf. Chart 3. Asia's share grew significantly around the millennium rollover, and in recent years the capital inflow from Asia has been on a par with the inflow from Europe, except in 2005, cf. below. 1 Net investments from other economies, not least Latin America, have also increased in recent years and now account for approximately 20 per cent of total net investments in the USA. Capital imports from Europe relate to a large extent to the UK, whose net investments in 2005 rose substantially to 316 billion dollars, cf. Chart 4. The increase was evenly distributed on all types of securities. The high investment volume does not reflect the US current-account deficit vis-àvis the UK, which was only 21 billion dollars in 2005, but rather the considerable capital flows from other countries into the USA via the UK. Particularly the OPEC countries in the Middle East have had large current-account surpluses in recent years as a result of the high oil prices. However, these countries have by and large not increased their net investments in the USA. Presumably, political considerations prevent the governments of these countries from purchasing US government securities directly, so instead they operate via e.g. UK investment banks. In addition, retained profits in European enterprises have increased, cf. IMF (2006). 1 With the exception of Egypt, all countries usually defined as the Middle East are included in Asia in Chart 3.

42 37 USA's NET INVESTMENTS IN LONG-TERM SECURITIES Chart 4 Billion dollars UK Japan China Caribbean Source: Treasury International Capital reporting system. Especially Japan and China have contributed to Asia's positive net investments in recent years. Japan's current-account surplus has been a major factor behind the acquisition of US assets. The Bank of Japan has intervened in the foreign-exchange markets to stabilise the exchange rate, cf. Chart 5 (left), which has led to ultra-high net investments in the USA. However, Japan's net investments in the USA declined considerably in 2005, to 46 billion dollars, despite a current-account surplus of 99 billion dollars, compared to 94 billion dollars in In fact Japan's portfolio of US Treasury securities decreased because the Bank of Japan stopped intervening in the foreign-exchange market in FOREIGN-EXCHANGE RESERVES (LEFT) AND PERCENTAGE OF TOTAL US LIABILITIES HELD BY FOREIGN OFFICIAL INSTITUTIONS (RIGHT) Chart 5 Billion dollars 1,200 1, Japan China Per cent of liabilities Total foreign official assets Foreign official holdings of US Treasury securities Note: Foreign official investors in the USA are primarily ministries of finance, central banks and regional and international organisations. Source: EcoWin and U.S. Department of Commerce, Bureau of Economic Analysis.

43 38 Monetary Review 4th Quarter 2006 FOREIGN HOLDINGS OF US TREASURY SECURITIES Chart 6 Billion dollars Japan China UK Caribbean Taiwan Beginning of 2005 Beginning of 2006 OPEC Korea Germany Hongkong Mexico Other Note: The UK, the Caribbean and to some extent Hong Kong are financial centres. Consequently, many capital flows are registered in these economies, even though the capital originates from other economies. The large UK, Caribbean and Hong Kong investments in US Treasury securities thus only to a limited extent indicate bilateral current-account surpluses vis-à-vis the USA. Source: Treasury International Capital reporting system. Even though Japan's net investments in US Treasury securities were negative in 2005, Japan still holds a large portfolio of Treasury securities, cf. Chart 6. In terms of the overall securities portfolio, Japan is also the largest creditor, followed successively by the UK, Luxembourg, the Cayman Islands and China. 1 China's current-account surplus has been growing in recent years. Moreover, direct investments in China have been considerable. Since China has imposed restrictions on private capital exports, the People's Bank of China has bought up the capital flowing in. This has boosted the foreign-exchange reserve considerably, and the renminbi has been kept almost unchanged against the dollar. The central bank has thus to a large extent determined how the capital has subsequently been invested. Consequently, a large proportion of China's net investments in the USA have been purchases of Treasury securities by official institutions. The mid-2005 foreign-exchange reserve was estimated at 711 billion dollars, while China's portfolio of US assets was 527 billion dollars, of which US Treasury securities accounted for almost 300 billion dollars. The percentage of the USA's total external liabilities held by foreign official institutions has been growing since 2000, cf. Chart 5 (right). 2 This 1 2 See Department of Treasury, Foreign official investors in the USA are mainly ministries of finance, central banks and regional and international organisations.

44 39 USA's INTERNATIONAL INVESTMENT POSITION Chart 7 Per cent of GDP Total Interest-sensitive assets Direct investments Equities Other Note: International Investment Position, total, equals (minus) external debt. Source: U.S. Department of Commerce, Bureau of Economic Analysis, and own calculations. is, inter alia, attributable to investments from Japan and China. In 2005, this share was 16 per cent, which is lower than in the early 1980s, however. The large capital inflow from the Caribbean is partly attributable to the tax legislation of the Cayman Islands, which has attracted a large number of banks and management companies. The impact of interest rates on capital income Net capital income is included in the current account and is an expression of income from a country's assets less payments on its liabilities. The USA's net capital income is vulnerable to changes in interest rates. A breakdown of the US international investment position, which equals the external debt (with sign negative), into net portfolios of interestbearing assets, equities, FDIs and other items shows that the USA is a net debtor in terms of interest-sensitive assets, cf. Chart 7. Interest-sensitive assets include e.g. bonds and deposits in banks. 1 On the other hand, the USA is a net creditor in terms of equities and FDIs. 1 This breakdown is in accordance with Higgins et al. (2005). Interest-sensitive US assets include e.g. deposits in banks, bonds and foreign-exchange reserves, while interest-sensitive liabilities primarily comprise Treasury securities, other bonds and deposits in banks. Direct investments are defined as equity portfolios exceeding 10 per cent of the share capital. Unlike investment data, the external debt is not broken down as short- and long-term assets.

45 40 Monetary Review 4th Quarter 2006 This structure means that a higher international level of interest rates will increase the USA's net interest payments to abroad, whereby net capital income is reduced. Ceteris paribus, a global increase in interest rates by 1 percentage point in 2005 would have reduced the US capital income by around 40 billion dollars. Over the last year, the rate of interest has risen in both Europe and the USA, and the USA's external debt has also grown in Against this background, the USA's interest payments to abroad are expected to increase considerably this year. Since the net capital income is included in the current account, a sustained increase in interest rates will augment the necessary improvement of the trade balance for the external balance to be restored. CAPITAL INCOME AND EXTERNAL DEBT Positive net capital income The USA's net external debt has been mounting in recent years, reaching just over 2,500 billion dollars at end Nevertheless, the USA's net capital income vis-à-vis abroad has remained positive and was 18 billion dollars in 2005, cf. Chart 8. It is difficult to point to one specific reason for this remarkable circumstance. It is natural first to compare returns on the USA's assets and liabilities, respectively. The return is calculated as income from the asset in relation POSITIVE NET CAPITAL INCOME Chart 8 Billion dollars Billion dollars 1, , , , , , International investment position Net capital income (right-hand axis) -150 Note: International investment position, total, equals (minus) external debt. Source: U.S. Department of Commerce, Bureau of Economic Analysis, and own calculations.

46 41 RETURN ON INTEREST-SENSITIVE ASSETS AND EQUITIES (LEFT) AND RETURN ON FDIs (RIGHT) Chart 9 Per cent Interest-sensitive asserts, income Interest-sensitive assets, expenses Equities, income Equities, expenses Per cent FDIs, income FDIs, expenses Note: Income and expenses are stated as percentages of the related assets. Source: U.S. Department of Commerce, Bureau of Economic Analysis, and own calculations. to the value of the asset. The income on interest-sensitive assets comprises interest income, while the income on equities includes dividend, but not capital gains. The income from FDIs is calculated as the reported profits from corporate financial statements. The USA is a net debtor for interest-sensitive assets and a net creditor for equities and FDIs, cf. above. Since FDIs involve greater risk for the investor than investments in interest-sensitive assets, it is not surprising that the return on FDIs has exceeded the return on interest-sensitive investments, thereby contributing to higher capital income. 1 The USA's interest-sensitive net liabilities are, however, so much larger than the asset portfolio of equities and FDIs, also in net terms, cf. Chart 7, that a traditional surplus return on the latter can by no means explain the positive capital income. Consequently, the composition of the USA's assets and liabilities is not the primary explanation for the positive net capital income. Instead, it is to a large extent attributable to the return on US foreign investments compared to the return on equivalent foreign investments in the USA. While the return on equities and interest-sensitive assets has been more or less the same in the USA and elsewhere, cf. Chart 9 (left), the return on FDIs from the USA has been significantly greater than the return on FDIs in the USA, cf. Chart 9 (right). On average, income from US FDIs has exceeded the return on FDIs in the USA by approximately 5.3 percentage points since This has brought considerable net income from FDIs, which thereby has exceeded net interest payments on the interestsensitive portfolio, cf. Chart Likewise, the return on investments in equities should, in theory, exceed the return on interestsensitive assets because equity investments entail greater risk. However, the capital income from equities solely comprises dividend, since capital gains on the equity price are not included in the statistics as income, cf. above, but solely registered as appreciation of the balance-sheet item on the balance of payments. This explains the apparently low return on equities.

47 42 Monetary Review 4th Quarter 2006 NET INCOME FROM VARIOUS FINANCIAL ASSETS Chart 10 Billion dollars FDIs Equities Interest-sensitive assets Source: U.S. Department of Commerce, Bureau of Economic Analysis, and own calculations. It should be noted that in the last few years other factors have also had a positive impact on US net capital income. Until recently, falling interest rates have reduced interest payments on US liabilities significantly, and depreciation of the dollar has increased the dollar value of returns on US investments abroad. Explanations for the USA's higher FDI return The economic literature offers various explanations of why returns on US FDIs have consistently exceeded the returns on FDIs in the USA. Higgins et al. (2005) find it strange that foreign investors are still willing to invest so much in the USA since the return is relatively low. They analyse this in more detail by looking into FDI returns by countries and industries. This does not change the pattern, however. For example, the return on US FDIs in Europe and Japan has consistently exceeded the returns on European and Japanese FDIs in the USA. They then seek to explain the difference by suggesting that US enterprises abroad are more efficient than foreign enterprises operating in the USA. This is not supported by data. According to Higgins et al., another possibility is that the Americans have been better at investing in more profitable enterprises and that foreign enterprises in the USA are newer and therefore have higher up-front costs. Furthermore, it is pointed out that competition is keener in the USA than in other economies, leading to a lower profit ratio.

48 43 Finally, Higgins et al. also mention tax aspects as an explanatory factor behind the differences in FDI returns, cf. Laster and McCauley (1994). Buiter (2006) and Gros (2006) also attribute the difference in returns to taxation. Buiter argues that multinational corporations apply internal settlement methods to transfer profits from their US entities to companies in countries with lower corporate income tax, in order to reduce tax payments. This can e.g. be achieved by letting US subsidiaries pay an artificially high price for imported goods from foreign parent enterprises, i.e. "transfer pricing". This reduces the booked profits of the US companies, while increasing the profits of the companies outside the USA. The profits of US enterprises abroad and foreign enterprises in the USA comprise observable payment flows from dividend payments, as well as unobservable payment flows known as reinvested earnings. The latter are calculated residually as the reported profits of the enterprises less dividend payments. Gros believes that foreign enterprises in the USA underreport their reinvested earnings, since they wish to reduce tax payments in the USA. 1 US enterprises abroad, on the other hand, have no incentive to underreport their reinvested earnings for the US statistics since profits from such enterprises are not taxable in the USA until the profit is repatriated. Consequently, the net return from US FDI assets and liabilities, and thereby net capital income, is overrated in the statistics. Gros estimates that a more uniform compilation of returns from US and foreign enterprises in the USA would reduce net capital income by approximately 100 million dollars per year. Hausmann and Sturzenegger (2006) reverse the issue and argue that the positive net capital income should not be viewed as a singular return on external debt. Instead, the capital income reflects the USA's position as a net creditor vis-à-vis abroad. In support of this argument Hausmann and Sturzenegger explain that if income from one asset is higher than income from another asset, this means that the value of the first asset is higher than the value of the second asset, irrespective of the book value of the assets. The primary reason for the USA's positive net capital income is therefore that the value of US assets abroad is underrated in the official statistics since the assets contain a considerable element of immeasurable values. These hidden values are known as "dark matter". In relation to FDIs, "dark matter" comprises e.g. know-how, branding 1 Reinvested earnings are not an observable capital flow, but an estimated entry in the balance of payments, based on enquiries. The reporting of profits for the balance-of-payments statistics by foreign enterprises forms the basis for assessment of their profits, and thus their tax payments, in the USA. See Gros (2006) for further details.

49 44 Monetary Review 4th Quarter 2006 ASSESSMENTS OF OFFICIAL DATA ACCORDING TO THE HYPOTHESES Table 1 Hausmann and Sturzenegger Buiter Gros Deficit on the trade balance... Overrated Overrated -- Net capital income... As data shows Overrated Overrated Current-account deficit... Overrated As data shows Underrated USA's IIP... Positive As data shows Overrated Note: The Table shows the views of the various hypotheses on whether the published data for the deficit on the trade balance, net capital income, the current account deficit and the USA's international investment position (IIP) is systematically over- or underrated. The USA's IIP equals (minus) the external debt, i.e. when the USA's IIP is e.g. overrated, the external debt is in fact greater than the official data shows. and R&D. 1 As an example, Hausmann and Sturzenegger calculate the USA's net assets in 2005 at approximately 352 billion dollars, whereas the official data says -2,500 billion dollars, on the basis of the positive net capital income of 18 billion dollars and an arbitrary 5-per-cent return on all assets. Implications of the hypotheses for the balance of payments and the international investment position How the positive net capital income is actually to be explained is of paramount importance to the further development in the US balance of payments and international investment position. Hausmann and Sturzenegger belong to the most optimistic school. According to their hypothesis, the USA's international investment position is positive, and moreover the trade balance is close to zero, cf. Table 1. The explanation is that the USA has been exporting "dark matter" for many years. Hausmann and Sturzenegger also believe that the high net capital income is sustainable since US enterprises will continue to export "dark matter". Against that background, the issue of the USA's external deficit and thus the global imbalances is negligible. Buiter's hypothesis also implies that the deficit on the trade balance is, in reality, smaller than the official data suggests since the value of the USA's imports has been artificially high. On the other hand, Buiter's explanation entails that the official data for net capital income is too positive, and consequently the current account is presumably more or less as the data shows 2. Buiter's hypothesis has no significant implications for the assessment of the USA's international investment position. 1 2 Hausmann and Sturzenegger (2006) also mention other contributions to "dark matter" that do not relate to FDIs. These are: seignorage generated by the large foreign holdings of dollars, and an unhedged risk premium in connection with financial transactions. It is not certain that the net effect on the balance of payments of Buiter's hypothesis is zero, since taxation could entail that larger corporate profits, in the absence of transfer pricing, do not lead to correspondingly higher dividend payments.

50 45 According to Gros' analysis, the current-account deficit is greater than the statistics show since net capital income has been overrated. External debt is also greater than shown in the statistics as the value of FDIs in the USA does not sufficiently reflect the accumulated profits of the companies. Higgins et al. do not reflect on whether the difference in returns will also be seen in future, nor on whether the international investment position and current-account deficit are under- or overrated in the statistics. Presumably the reason is that they cannot give a definitive explanation for the differing returns on the USA's FDI assets and liabilities. An assessment of the robustness of the analyses Most of the above explanations of why the USA's FDI return has consistently been higher than the return on FDIs in the USA are difficult to verify. Buiter's hypothesis can, however, to some extent be disproved. If transfer pricing has been widely used the US terms of trade will gradually have deteriorated over the last decades since transactions between multinational corporations have expanded more rapidly than foreign trade. 1 This has not been the case. The Gros analysis suffers from a validation problem as it is based solely on assumptions that foreign enterprises do not report the correct data for reinvested profits. Gros does not assess the extent to which it is actually possible for enterprises to hide their profits from the tax authorities. The "dark matter" hypothesis suffers from an equivalent validation problem since it is not supported by an empirical analysis. A more material weakness in Hausmann and Sturzenegger's theory is, however, that the hypothesis cannot convincingly explain away the global imbalances, including Japan's current-account surplus. In addition, Hausmann and Sturzenegger's calculations have been criticised for assuming the same return on different types of assets, which is unrealistic. 2 This criticism is justified, but the calculations merely serve to illustrate a point and should not be taken at face value. Moreover, the assumptions regarding returns on assets are neither closely linked to the reasoning behind the theory nor to the overall conclusions. As regards the difference between FDI returns in the USA and elsewhere, it seems implausible that foreign investments in the USA systematically yield lower returns than US FDIs, particularly since more than half of the USA's FDI assets and liabilities are held in Europe, where the structure of the economy does not differ materially from that of the 1 2 See Gros (2006) for further details. See e.g. Buiter (2006).

51 46 Monetary Review 4th Quarter 2006 USA's FDIs, END-2005: ASSETS AND LIABILITIES Chart 11 Billion dollars 2,500 2,000 1,500 1, Total Europe Asia Latin America Canada Other USA's FDIs FDIs in the USA Source: U.S. Department of Commerce, Bureau of Economic Analysis. USA, cf. Chart 11. It seems unlikely that foreign investors are willing to invest in the USA when they can apparently make higher profits at home, and that US investors should be better at generating profits, as stated by Hausmann and Sturzenegger and by Higgins et al. It is also strange that foreign investors can achieve a return on equities in the USA that is equivalent to US investors' returns on equity investments abroad, until the point where a foreign investment in a US enterprise exceeds 10 per cent of the share capital, which is the threshold between equity investments and FDIs. Consequently, the differences in FDI returns indicated by the statistics are deemed to be misrepresentative and should not be given too much weight. Buiter's, Gros' and Hausmann and Sturzenegger's explanations for the differences between FDI returns in the USA and elsewhere may all apply to some extent. Even though Hausmann and Sturzenegger's theory presumably paints a far too positive picture of the US debt situation, it cannot be ruled out that the hypothesis contains a small element of reality. It is probable that the USA is to some extent able to export know-how and branding, etc. and that the USA's foreign assets are therefore underrated. It also seems plausible that (foreign) enterprises do their utmost to reduce tax payments. On the other hand, neither explanation is satisfactory in itself. In that case, net capital income is lower than the statistics show, and will continue to be so in the future, since Buiter's and Gros' explanations

52 47 entail that net capital income is overrated, while Hausmann and Sturzenegger take net capital income at face value. Gros' theory has no direct implications for the trade balance, while Buiter's and Hausmann and Sturzenegger's explanations improve the balance of goods and services compared to the official data. Overall it cannot be precluded that the deficit on the balance of goods and services is smaller than indicated by the foreign-trade statistics, although it must currently still be considerable. As regards the USA's net holdings of FDIs, and thus its international investment position, the hypotheses point in opposite directions. Consequently, it is impossible to say whether there is systematic distortion in the compilation of the international investment position. CONCLUSION Major uncertainties are attached to the data concerning the external balance of the USA. There are indications that the deficit on the trade balance is overrated, while net capital income is actually somewhat lower than the data shows. In other words, the current-account deficit may be either greater or smaller than indicated by the data. The actual international investment position may also deviate from the official data. However, it can hardly be ruled out that the current-account deficit is large and the external debt considerable. The US current-account deficit is financed to a large extent by Asian and OPEC economies. A sustained current-account deficit will add to the external debt, and combined with the predominance of interest-bearing liabilities and rising interest rates this year, it will make even greater demands on the US trade balance if the global balances are to be restored. It has been easy for the USA to finance its current-account deficit, owing to the internationally integrated financial markets and large gross capital flows. There is a risk that a sudden resolution of the global imbalances sets in, resulting in a major slowdown in the USA and increased instability in the international financial markets. However, the sustained US current-account deficit in recent years indicates that the depth of the financial markets is so great that major global imbalances may also be seen in future without necessarily resulting in dramatic adjustments.

53 48 Monetary Review 4th Quarter 2006 LITERATURE Bernanke, B. (2005), "The Global Saving Glut and the U.S. Current Account Deficit". Remarks by Governor Ben S. Bernanke at the Homer Jones Lecture, St. Louis, Missouri, April 14. Buiter, W. (2006), "Dark Matter or Cold Fusion". Global Economics Paper No. 136, Goldman Sachs Economic Research, January 16. Department of the Treasury (2006), "Report on Foreign Portfolio Holdings of U.S. Securities as of June 30, 2005". Federal Reserve Bank of New York, June Gros, D. (2006), "Why the US Current Account Deficit is Not Sustainable", International Finance, Vol. 9, No. 2, summer Hausmann, R and F. Sturzenegger (2006), "Why the US Current Account Deficit is Sustainable", International Finance, Vol. 9, No. 2, summer Higgins, M., T. Klitgaard and C. Tille (2005), "The Income Implications of Rising U.S. International Liabilities". Current Issues in Economics and Finance, Vol. 11, No. 12, Federal Reserve Bank of New York. International Monetary Fund (2005), World Economic Outlook, September. International Monetary Fund (2006), World Economic Outlook, April. Kitchen, J. (2005), "Sharecroppers or Shrewd Capitalists? Projections of the U.S. Current Account, International Income Flows, and Net International Debt". Unpublished manuscript, Office of Management and Budget. Laster, D. and R. McCauley (1994), "Making Sense of the Profits of Foreign Firms in the United States". Federal Reserve Bank of New York, Quarterly Review 19, No. 2 (summer-fall).

54 49 Slovenia: An Economic Portrait of the New Euro Area Member State Niels Peter Hahnemann, Economics INTRODUCTION Until 1991, Slovenia was part of Yugoslavia. Since its independence the country has undergone a rapid and successful economic restructuring process. On 1 January 2007 Slovenia adopts the euro as the first of the former Communist states to become part of one of the world's largest currency areas. Slovenia has a population of a little over 2 million, of whom just over 300,000 live in the capital, Ljubljana. The demographic structure is relatively homogenous. The population census from 2002 shows that approximately 83 per cent of the population are Slovenes and 5 per cent are Serbs, Croats or Bosnians, with the remainder not stated. In 2005 Slovenia's GDP in current prices amounted to 27.6 billion euro. This is equivalent to 0.3 per cent of GDP in the euro area, which is slightly below the ratio for the euro area's hitherto smallest economy: Luxembourg, whose population is only around 400,000, however. In terms of purchasing-power-adjusted prices Slovenia's GDP in 2005 was 37.8 billion euro, and thus a little above Luxembourg s GDP. In 2005 its GDP per capita was equivalent to 81 per cent of the overall EU average. After Cyprus, Slovenia is the most prosperous of the new EU member states with an income level on a par with Greece and just above Portugal, the two least affluent economies of the euro area, cf. Chart 1 (left). Slovenia appears to be ready to join the euro area. The economy has enjoyed a long period of stability with moderate inflation and relatively modest deficits on both its government budget balance and current account of the balance of payments. The high economic growth since 2004, cf. Chart 1 (right), when the Slovenian tolar was pegged closely to the euro, underpins the assessment that the economy is prepared for the single monetary policy and the single currency. This article presents an economic portrait of the new euro area member state with emphasis on the convergence criteria and the challenges ahead for Slovenia within Europe's monetary union.

55 50 Monetary Review 4th Quarter 2006 GDP PER CAPITA (LEFT) AND GDP GROWTH (RIGHT) Chart 1 EU25 = Slovenia Czech Republic Hungary Portugal Greece Per cent Slovenia Euro area Note: Purchasing-power-adjusted GDP per capita. Source: Eurostat. THE CONVERGENCE CRITERIA To be able to adopt the euro, an EU member state must fulfil a number of convergence criteria. In view of its stable economy, with high growth and limited macroeconomic imbalances, Slovenia has been able to fulfil the convergence criteria relatively quickly. The assessment of Slovenia by the European Commission and the European Central Bank, ECB, was conducted in May 1 and approved by the European Council in June, i.e. after only two years' EU membership for Slovenia. Below, Slovenia's results are compared with each of the criteria. Development in prices The inflation criterion states that the average year-on-year rate of increase in HICP (the Harmonised Index of Consumer Prices), observed over a period of 12 months before the examination, may not exceed by more than 1.5 percentage points that of, at most, the three bestperforming member states in terms of price stability. In March, the month for which the latest data was available when the assessment of Slovenia was conducted, the criterion value was 2.6 per cent, Slovenia's 12-month average inflation was 2.3 per cent, and actual headline inflation was 2.0 per cent year-on-year. Slovenia s inflation was thus at the euro area inflation level of 2.3 per cent. The development in Slovenia's average rate of inflation clearly shows its convergence with the euro area. In January 2006 average inflation dropped below the criterion value and has since then at no time exceeded it, even though during 2006 the inflation tendency reversed to moderately rising prices, cf. Chart 2 (left), due especially to increasing energy prices. All of the new EU member states are more vulnerable to 1 Cf. the European Commission, Report from the Commission. Convergence Report 2006 on Slovenia, May 2006, and the ECB, Convergence Report, May 2006.

56 51 INFLATION (LEFT) AND EXCHANGE-RATE DEVELOPMENT (RIGHT) Chart 2 Per cent, year-on-year Headline 12-month moving averaget Criterion value SIT per euro ERM II entry Central rate Note: In the chart for the exchange rate an increase is equivalent to depreciation. The central rate is tolars per euro. The agreed intervention limits of +/- 15 per cent are respectively and tolars per euro. Source: Eurostat and EcoWin. energy price hikes than the euro area. Energy accounts for more than 13 per cent of Slovenia's CPI (consumer-price index), compared to 9 per cent in the euro area. Exchange-rate stability A condition for adopting the euro is participation in ERM II for at least 2 years without "severe tensions", including unilateral devaluation. The Slovenian tolar has been part of ERM II since 28 June 2004, i.e. for more than 2 years. Prior to its ERM II entry, Slovenia's exchange-rate policy was to write down the exchange rate vis-à-vis the euro by a fixed rate. This policy was discontinued when Slovenia entered ERM II, after which the exchange rate vis-à-vis the euro has been very close to the central rate, cf. Chart 2 (right). The market rate has not deviated from the central rate by more than +/- 0.2 per cent, even though with the exception of August 2006 Slovenia's central bank, Banka Slovenije, has not followed the ECB's raising of interest rates. In addition, interest rates were lowered on three occasions: in February, March and June. The spread between the official interest rate, 60 day tolar bills, and the ECB's interest rate has therefore narrowed considerably, cf. Chart 3 (left). Tolar-denominated money-market instruments are thus less attractive to commercial banks than euro-denominated instruments. In July the Ecofin Council resolved that Slovenia would adopt the euro on 1 January 2007 at the present central rate, i.e tolars per euro. Development in long-term interest rates A requirement for adoption of the euro is that, observed over a period of one year before the examination, a member state has had an average nominal long-term interest rate that does not exceed by more than 2 percentage points that of, at most, the three best performing member

57 52 Monetary Review 4th Quarter 2006 OFFICIAL INTEREST RATE (LEFT) AND LONG-TERM YIELD (RIGHT) Chart 3 Per cent ERM II entry Per cent ECBs minimum bid rate Banka Slovenije's 60-day tolar bills Slovenia Euro area Germany Italy Greece Note: 60-day tolar bills are the yield on certificates of deposit issued by Banka Slovenije. It is the bank's key monetarypolicy interest rate. Note that the maturity differs from the ECB's 1-week minimum bid rate. The long-term yields are the monthly average yield on a government bond with around 10 years left to maturity, for the euro area weighed with the nominal outstanding amount for each country. This is the interest-rate series applied to the convergence assessment. Source: Banka Slovenije, Eurostat and EcoWin. states in terms of price stability. In March the criterion value was an upper limit of 5.9 per cent, but the average yield on long-term Slovenian government bonds was only 3.8 per cent, i.e. considerably below the criterion value and also lower than the unweighted average for the three member states on which the assessment was based. However, the criterion value included two non-euro area member states, i.e. Poland and Sweden. Compared to the euro area Slovenia had a small, but positive average yield spread of 0.4 percentage point over the 12-month period up to March. The long-term yield has fallen in recent years in step with the global decline in bond yields, the stable macroeconomic development and the resulting perception that Slovenia could soon join the euro area. At the same time the yield spread to the euro area has narrowed as a result of the dampening of inflation. The yield spread has narrowed further since the end of 2005, cf. Chart 3 (right). Fiscal policy According to the convergence criteria the government budget deficit may not exceed 3 per cent of GDP, and the government debt may not exceed 60 per cent of GDP is the reference year for the decision on adoption of the euro in In 2005, Slovenia's budget deficit was equivalent to 1.8 per cent of GDP, and the debt was 29.1 per cent of GDP. In recent years the budget deficit has been relatively stable in the range of 2 to 2.5 per cent of GDP, except for , when the deficit rose, cf. Chart 4, among other reasons because commitments to a public war injury fund had to be carried as expenditure.

58 53 GOVERNMENT BUDGET BALANCE AND DEBT Chart 4 Per cent of GDP 30 Per cent of GDP Budget balance (right) Gross debt (left) Source: Eurostat, ECB Convergence Report, May The convergence reports from both the ECB and the European Commission pointed out that government expenditure related to the population's ageing is expected to increase particularly strongly in Slovenia in the coming decades, and in fact more strongly than in both the new EU member states overall and the old EU. The reason is that in 20 years' time Slovenia will have a larger retired population segment than other member states. SLOVENIA IN THE EURO AREA Slovenia's rapid and successful achievement of the demanding nominal convergence process is a vital indication that the economy is ready to enter the euro area. This is confirmed by high economic growth, continuously declining unemployment, and strongly rising employment in the last couple of years during which the tolar has been pegged closely to the euro. However, the strong economic progress has entailed a modest current-account deficit. In the long term, participation in a monetary union requires an adaptable economic structure. In recent years Slovenia has undergone a comprehensive restructuring process and as in the other new EU member states the economy has been integrated into the European production structure. Especially the manufacturing sector has undergone structural adjustment.

59 54 Monetary Review 4th Quarter 2006 CORRELATION OF INDUSTRIAL PRODUCTION WITH GERMANY Chart 5 Correlation coefficient SIT CZK HUF PRT GRK AUT Note: The correlation coefficients are calculated vis-à-vis Germany for year-on-year increases in the monthly seasonallyadjusted industrial production index. SIT = Slovenia, CZK = Czech Republic, HUF = Hungary, PRT = Portugal, GRK = Greece, AUT = Austria. Source: Own calculations, Eurostat, EcoWin. The structure of the new member states' manufacturing sector has thus been adjusted to that of the highly-developed EU member states through rationalisation, down-sizing and modernisation, and the private service sector has been developed. This entails among other things that intra-industrial trade in intermediaries and semi-manufactures will account for an increasing share of the new member states' total industrial trade. This development has also been observed in Slovenia. The growth in industrial production is therefore relatively closely correlated with industrial production in the euro area's "locomotive", Germany, although not as closely as in the core member state Austria, but closer than in member states towards the periphery of the euro area such as Greece and Portugal, cf. Chart 5. Slovenia's economic structure nonetheless still differs substantially from the rest of the euro area, including Greece and Portugal. As in the other new EU member states, the manufacturing sector accounts for a considerably larger share of the overall economy than in the euro area. However, the special aspect in Slovenia's case is that its manufacturing sector is also larger than those of many of the other new EU member states, with the exception of the Czech Republic. The private service sector is equivalently less developed. For many years Slovenia has received less foreign direct investment, FDI, than any of the other new EU member states. In the period

60 55 FDI corresponded to only 2.3 per cent of GDP, compared to an average of 5 per cent in the new member states. The speed of the privatisation process, including within the financial sector, has been characterised as slower than in the other new EU member states. 1 There are no signs of any problems with Slovenia's competitiveness. Exports are rising steadily, the real exchange rate is stable, 2 and wage and price increases are moderate. However, the average rate of increase in exports since the mid-1990s has been lower than in e.g. the Czech Republic and Hungary, and imports have also shown slower growth. This indicates that the "catch-up" effect has not been as strong a driver of Slovenia's economy as in the other new member states, due to Slovenia's relatively favourable starting point. At the outset of the transition process in the early 1990s Slovenia was already considerably more affluent than the other transition economies that have since joined the EU. Slovenia's export performance within the EU is in line with that of Greece and Portugal, but unlike these two member states and most of the other new EU member states, Slovenia's current-account deficit is relatively modest. All in all, Slovenia thus already appears to be relatively well-integrated into intra-eu trade. Slovenia's overall labour participation rate is slightly above the euro area level, whereas for most of the other new EU member states it is lower as a consequence of the reform process in recent years, which for a period has reduced the demand for labour. Moreover, the labour participation rate of the older age groups is significantly lower than in the euro area, and also lower than in most of the other new member states, cf. Chart 6. Slovenia's low participation rate for its older age groups reflects that it has the lowest average retirement age in the EU. In 2003 the effective retirement age was slightly above 56, compared to around 61 in the euro area, and in Poland, Hungary and the Czech Republic. The existing pension system, for instance, is an incentive to retire early. The tax and social security system also affects the supply of labour. Interview surveys thus show far lower unemployment than is actually registered, probably because a large proportion of the registered unemployed are not actually looking for work. The latter is necessary for them to be counted as unemployed in the interview surveys. This may be related to the interaction between a progressive tax 1 Cf. IMF, Republic of Slovenia Selected Issues, June 2006, Table 3 p. 104 and the European Commission, Convergence Report 2006 on Slovenia, May 2006, p It is particularly difficult to calculate equilibrium exchange rates for economies that, like Slovenia, are in the throes of a structural reform process. In 2006 the real effective exchange rate for the Slovenian tolar was at the same level as in According to Schadler et al., Adopting the Euro in Central Europe, IMF Occasional Paper No. 234, 2005, p. 77, the year 2001 can be taken as the benchmark equilibrium exchange rate in Slovenia.

61 56 Monetary Review 4th Quarter 2006 LABOUR PARTICIPATION RATE Chart 6 Per cent of the population Slovenia Czech Republic Hungary Euro area Note: The labour participation rate is the number employed in the age group as a percentage of the population in the same age group. Source: EcoWin. system with high tax rates (the highest income tax rate is 50 per cent) and social benefits at the OECD level or higher. 1 The structure of Slovenia's labour market is characterised by some of the rigidity seen in many of the euro area member states. For example, labour market legislation is relatively stringent, 2 wage formation is centralised at sector or national level, and a large proportion of the overall employment is covered by collective agreements. 3 Unlike the other euro area member states Slovenia conducts an incomes policy. Under the latest agreement between the government and the social partners, wage increases must be at least 1 percentage point below productivity increases. Another special characteristic is that Slovenia still conducts a type of cost-of-living adjustment since wage development is closely linked to inflation via indexation to expected inflation. The low supply of labour has consequences for the flexibility of Slovenia's fiscal policy. In 2005 government expenditure accounted for 48 per cent of GDP, which is higher than the averages of 42 per cent for the new EU member states and 44 per cent for the old member states. Especially social security expenditure is high. For most budget areas the Cf. IMF (2006), pp. 71 and 73. In relation to the OECD's index for the legal protection of employees Slovenia is exceeded only by the euro area member states Portugal and Greece, cf. Schadler et al. (2005), Table 3.7 p. 25. Thomas Gruber, Employment and Labour Market Flexibility in the New EU Member States, ONB Focus I/2004, Table 3 p. 111 and Chart 9 p. 114.

62 57 cyclicality of government expenditure is a good deal lower in Slovenia than in both the new and the old EU member states. 1 CONCLUSION Slovenia has a well-functioning economy and is already well-integrated into the European economy. In some aspects, however, the economy is less flexible. For instance, there is a relatively high degree of budgetary rigidity in government finances. Together with the lack of expenditure flexibility, the ageing of the population will exert pressure on government finances. A need for considerable fiscal-policy adjustments in the years to come can already be foreseen in order to keep both the deficit and the debt in line with the Maastricht Treaty's requirements. 1 Cf. IMF (2006), pp. 6-8 and Table 1 p. 10.

63 58 Monetary Review 4th Quarter 2006 THE NATIONAL SIDES OF THE SLOVENIAN EURO COINS

64 59 The Relevance of GDP Growth Rates Christian Ølgaard, Economics INTRODUCTION An evaluation of a country's economic performance often focuses on GDP in constant prices. Countries with high economic growth over a period of time are often pinpointed, and the explicit or implied conclusion is that these countries operate with competitive economic models that can be a lesson to low-growth countries. On the surface, this may seem to be the case, and, depending on the agenda of the sender, it may be the basis for policy recommendations to politicians and the general public. However, a comparison of simple GDP growth rates across countries is associated with several pitfalls even though all countries follow the same manual for compilation of national accounts. This article focuses on pitfalls that are (too) often overlooked: The significance of demographic factors. The significance of terms of trade. The conclusion of the analysis is that demographic factors and terms of trade can be of great significance to the evaluation of the relative performance of the individual countries. DELINEATION OF THE ANALYSIS For the sake of clarity, the analysis is limited to 13 OECD countries covering the full range from high to low GDP growth. The period under review is The choice of period is not immaterial since, among other factors, the cyclical position of the individual country in the first and last year of the period impacts on growth in the interim period. The choice of a 12-year period diminishes this problem, as does the selection of countries that were either at or close to a trough in 1993, measured in terms of the OECD's output gap. At the same time, the period is so long that the differences among the countries are not likely to be merely a coincidence, and not too long for the results not to be relevant today.

65 60 Monetary Review 4th Quarter 2006 GDP IN CONSTANT PRICES, AVERAGE ANNUAL CHANGE FROM 1993 TO 2005 Chart 1 Per cent p.a Japan Italy Germany France Netherland Denmark Sweden UK Norway USA Spain Finland Ireland Note: The standard deviation for the growth rates in the Chart (a measure of the spread) is Source: OECD, Economic Outlook 79 (database), June 2006 and own calculations. In terms of simple GDP growth rates, Ireland clearly tops the list with an average annual growth rate of 7.4 per cent in the period , cf. Chart 1. It is followed by Finland, Spain and the USA with growth rates of around 3.5 per cent. Denmark's growth rate of 2.5 per cent puts it in the middle, while Germany, Italy and Japan are at the bottom of the list. These are well-known facts. Disregarding Ireland, the differences in growth rates may seem modest, but even minor differences may accumulate to substantial levels over a 12-year period. Sweden's average growth rate has been 0.5 percentage point higher than the Danish rate, which means that over the period Sweden's GDP growth has exceeded Denmark's by 6 per cent. THE SIGNIFICANCE OF DEMOGRAPHIC FACTORS Demographic trends are significant to a country's growth potential. All other things being equal, growth will be higher in a country with an increasing population of working age than in a country with no such increase. If higher growth in country A than in country B solely reflects that there are more hands to do the work, the gap cannot be taken to mean that country A has outperformed country B in economic terms. This elementary aspect is often overlooked perhaps in the belief that "the gap is negligible anyway".

66 61 POPULATION OF WORKING AGE, CHANGE FROM 1993 TO 2005 Chart = Per cent Japan Italy Germany France Netherland Denmark Sweden UK Norway USA Spain Finland Ireland Denmark Ireland USA Germany Note: The delineation of the age groups of working age follows OECD, Economic Outlook. The delineation varies across countries. Most countries define the age groups of working age as year-olds or similar, but the US definition includes everyone over the age of 15. The exact delineation is hardly of material importance here as the focus is on changes. Source: OECD, Economic Outlook 79 (database), June The fact is, however, that demographic factors can vary considerably across countries over prolonged periods. In the period , countries with high GDP growth rates also showed high growth in the population of working age, and vice versa, cf. Chart 2. The number of people of working age rose by 26 per cent in Ireland in the period under review, and by 16 per cent in the USA, but fell in Japan, Germany and Italy. This difference in the supply of labour impacts on the growth potential of the economies, and thus on actual growth. Adjustment of GDP growth rates for the change in the number of people of working age 1 reduces the growth gap, cf. Chart 3. Ireland still has the highest growth rate by far, but the adjustment brings the growth rates of the other countries much closer together. A change in the order of the countries is also observed. A particularly noteworthy aspect is that the USA's growth rate is now lower than Denmark's and only slightly higher than e.g. Germany's rate. Contrary to the argument often put forward, the USA's relatively high economic growth in the last years is first and foremost attributable to a strong influx of labour. After adjustment for this effect, the development in the USA has been close to the average. 2 THE SIGNIFICANCE OF TERMS OF TRADE Recent years have seen increasing focus on the fact that GDP growth rates adjusted or unadjusted for demographic factors do not capture 1 2 On the adjustment of GDP growth for demographic factors, a measure of GDP growth per capita is normally applied. Here a measure of the number of people of working age is applied because this group is more closely associated with the output potential than the total population. If a country's population growth can be attributed to a higher birth rate or lower mortality, this will have no immediate significant effect on the country's supply of labour. It should be noted that adjustment for demographic factors is not relevant to identifying in which markets the sales potential is greatest. All other things being equal, higher population growth also entails greater sales potential.

67 62 Monetary Review 4th Quarter 2006 GDP IN CONSTANT PRICES PER PERSON OF WORKING AGE, AVERAGE ANNUAL CHANGE FROM 1993 TO 2005 Chart 3 Per cent p.a Japan Italy Germany France Netherland Denmark Sweden UK Norway USA Spain Finland Ireland Note: The dots show GDP growth from Chart 1. The standard deviation for the growth rates in the Chart (a measure of the spread) is 1.04, i.e. considerably lower than in Chart 1. Source: OECD, Economic Outlook 79 (database), June 2006 and own calculations. the effect of ongoing differences in terms of trade, cf. e.g. Beier and Pedersen, 2005, Ministry of Finance, 2005, and OECD, Finland in particular is often cited as an example. In simplified terms, Finland has seen rapid growth in the ICT (information and communications technology) sector since the mid-1990s, with Nokia as a case in point. Today ICT accounts for a much larger share of total output than in other countries. ICT output has been characterised by high productivity gains, not only in Finland, but worldwide. The greater weight of the ICT sector in the Finnish economy compared to most other economies has given Finland high growth in labour productivity. This is the background to Finland's rapid growth in GDP in constant prices. However, the high productivity growth in the ICT sector has not been accompanied by equivalent growth in earnings. Thanks to the strong competition in the ICT sector, the productivity gains have first and foremost been to the benefit of consumers and ICT-consuming business enterprises, as lower prices. On the other hand, the opposite factor a strong drop in sales prices for ICT products, causing Finland's terms of trade to deteriorate by more than 10 per cent since 1993, has no impact on real GDP growth. The example illustrates that statistics for real GDP or productivity growth can be misleading. A strong increase in productivity does not enhance a country's performance if the country at the same time has to sell goods and services at falling prices.

68 63 TERMS OF TRADE, CHANGE FROM 1993 TO 2005 Chart 4 Per cent Japan Italy Germany France Netherland Denmark Sweden 57% UK Norway USA Spain Finland Ireland 1993 = Denmark Finland Norway Japan Note: The terms of trade are calculated as the ratio between the deflator in the national accounts for exports of goods and services and the deflator for imports of goods and services. Source: OECD, Economic Outlook 79 (database), June Finland is not the only country whose terms of trade changed significantly in the period Sweden's terms of trade deteriorated almost as much, and Japan's even more, cf. Chart 4. In contrast, Norway's terms of trade have improved strongly since the late 1990s, first and foremost driven by rising oil prices. Denmark's terms of trade have improved too, but more gradually. Only a minor share of Denmark's improvement can be attributed to North Sea oil and gas production. All other things being equal, an increase in GDP in constant prices increases a country's real incomes, while deterioration in the terms of trade has the opposite effect. The question is: to what extent is a given level of additional GDP growth offset by deterioration in the terms of trade? There is no clear answer. The terms of trade reflect certain price effects which GDP in constant prices as such is designed to eliminate. This makes it difficult to equivalate GDP and terms of trade. An estimate of the significance of the terms of trade can be obtained by calculating what GDP would have been if export prices had developed in parallel with import prices. This corresponds to calculating export volume as the volume of imported goods that export revenue can buy. More specifically, the value of exports is deflated by the import price index rather than the export price index: p x yk = y x + x p m where y is GDP in constant prices, y k is adjusted GDP, x is exports in constant prices and p x and p m are the national accounts' deflators for respectively exports and imports of goods and services. The same method is applied in Ministry of Finance (2005) and OECD (2006). 1 This static 1 The former applies a measure of GDP per capita adjusted for terms of trade, while the latter focuses on productivity adjusted for terms of trade. The OECD otherwise uses the term "command GDP", i.e. available GDP, for GDP adjusted for terms of trade.

69 64 Monetary Review 4th Quarter 2006 TERMS-OF-TRADE-ADJUSTED GDP PER PERSON OF WORKING AGE, AVERAGE ANNUAL CHANGE FROM 1993 TO 2005 Chart 5 Per cent p.a Japan Italy Germany France Netherland Denmark Sweden UK Norway USA Spain Finland Ireland Note: The dots show GDP growth from Chart 3. The standard deviation for the growth rates in the Chart (a measure of the spread) is 1.09, i.e. slightly higher than in Chart 3. Source: OECD, Economic Outlook 79 (database), June 2006 and own calculations. calculation, based on all-other-things-being-equal, does not provide the full picture. Had the terms of trade followed a different pattern, the development in the rest of the economy would also have been different, but such derived effects are disregarded here. Nevertheless, the calculation does give an indication of magnitude. Compared to adjustment for demographic factors, adjustment for terms of trade has a more limited effect on growth rates. Norway accounts for the greatest change, as its annual growth rate rises by 1.6 percentage points. On the other hand, the growth rates of Finland and Sweden are reduced by 0.5 and 0.3 percentage points, respectively, cf. Chart 5. Japan's growth rate is reduced by only 0.2 percentage point, even though Japan clearly accounts for the strongest deterioration in terms of trade. This reflects the more closed nature of the Japanese economy compared to the Nordic economies, so that the deterioration in terms of trade impacts less on the overall economy. Denmark's GDP growth rises by 0.25 percentage point per annum. After adjustment for terms of trade, Denmark's GDP growth is thus slightly higher than Sweden's, and the gap to the USA widens. After adjustment for terms of trade, Norway shows the second-highest growth rate after Ireland. Since Norway's terms of trade are highly sensitive to oil prices, the adjustment will tend to vary more year-on-year

70 65 than is the case for other countries. For as long as oil prices remain high and oil production is substantial, the growth adjusted for terms of trade will exceed the unadjusted growth. After adjustment for demographic factors and terms of trade, Finland is in third place, but Finland's growth rate is only slightly higher than that of e.g. Denmark. It should also be taken into account that in 1993, i.e. the first year in the period analysed, Finland posted a negative output gap of per cent of GDP, after a number of years of severe economic crisis. None of the other countries were equivalently below their output potential in At that time, Denmark posted an output gap of 3-4 per cent after seven "lean" years. In other words, in 1993 the Finnish economy had a far greater output lag than the other countries in the analysis. Finland's relatively high growth since 1993 thus not only reflects structural differences, but also the country's weak cyclical position from the start. TERMS-OF-TRADE EFFECT IN KRONER As an alternative, the effect of the change in terms of trade can be quantified in kroner per capita. Chart 6 (left) illustrates the change in GDP per capita (current prices) compared to a situation with unchanged terms of trade. In 2005, Norway's GDP per capita was 53,000 Danish kroner higher than would have been the case with unchanged terms of trade. Denmark showed a terms-of-trade gain of kr. 7,000 per capita in This corresponds to 3.5 per cent of (private and public) consumption, cf. Chart 6 (right), while the gain for Norway was 24 per cent of consumption. Conversely, Finland's terms-of-trade loss reduced GDP by kr. 11,000 in 2005, corresponding to 6 per cent of consumption. INCOME EFFECT OF CHANGE IN TERMS OF TRADE FROM 1993 TO 2005 Chart 6 Kr. 1, Per capita in 2005 Kr. 53,000 Per cent 8 4 Share of private and public consumption in per cent Japan Italy Germany France Netherland Denmark Sweden UK Norway USA Spain Finland Ireland Japan Italy Germany France Netherland Denmark Sweden UK Norway USA Spain Finland Ireland Note: The amounts are translated into Danish kroner using the OECD's purchasing power parities for 2005 and not market exchange rates. This eliminates the risk of the results being influenced by random exchange-rate fluctuations. The purchasing power parity between Denmark and the USA was kr per dollar in Source: OECD, Economic Outlook 79 (database), June 2006 and own calculations.

71 66 Monetary Review 4th Quarter 2006 The calculations illustrate that terms-of-trade gains accumulated over a prolonged period can have a considerable impact on incomes and consumption, even though the adjustment of annual GDP growth may appear to be limited. CONCLUSION The above calculations have illustrated that it can be misleading to base assessment of a country's economic performance solely on simple GDP growth rates. The adjustments for demographic factors and terms of trade reduce the growth gaps between the countries, and their ranking changes. Norway and, to a certain extent, Denmark move up considerably, while the growth performance of the USA, Finland and Sweden becomes less impressive. Thus, after the adjustments, Denmark's growth in was clearly higher than in the USA and Sweden, and at the level of Finland, while in terms of unadjusted figures Denmark ranked lower than these countries. This article has focused on demographic factors and terms of trade, but other factors, too, can lead to diverging measures of GDP growth, without actually reflecting differences in economic capacity to generate real income. For example, capital is not taken into account. A high growth rate may be attributable to rapid expansion of the capital apparatus. In principle, the consumption of fixed real capital, i.e. depreciation, should be deducted from GDP. The problem is that the capital apparatus is difficult to quantify and that there are no internationally comparable statistics for depreciation. The analysis can be expanded by more broadly considering measures of the countries' standards of living. However, this is a different discussion that includes such aspects as who benefits from the added value. SPREAD BETWEEN GROSS NATIONAL PRODUCT AND GROSS DOMESTIC PRODUCT Chart 7 Per cent 0 Per cent Ireland Luxembourg Denmark Sweden Finland Norway Note: The gross national product (GNP) is the gross domestic product (GDP) plus (net) fixed capital formation from abroad and output and import taxes, etc. from abroad. The spread is calculated as (GNP-GDP)/GDP. Source: OECD and Eurostat.

72 67 Particularly in Ireland and Luxembourg, an increasing share of output has gone out of the country for remuneration of the considerable foreign investments and labour that these countries have attracted over the last decades. As a result, GNP has increased less than GDP, cf. Chart 7 (left). The opposite trend is observed in the four Nordic countries that for a prolonged period have posted large current-account surpluses, cf. Chart 7 (right). In principle, an almost indefinite number of adjustments can be made, so it is important to keep the current focus in mind. The point of departure of the analysis in this article is the change in the countries' GDP over a period. Adjustment of GDP growth for demographic factors and terms of trade results in a measure of the countries' economic performance, or perhaps more accurately the countries' ability to generate income through output, that is better than simple GDP growth. The measure thus describes a key dimension of the countries' competitiveness. LITERATURE Beier, Niels C. and Erik Haller Pedersen (2005), Wages, Competitiveness and the Balance of Payments, Danmarks Nationalbank, Monetary Review, 1st Quarter. Danish Ministry of Finance (2005), Economic-Policy Cooperation (in Danish), November. OECD (2005), OECD Economics Surveys, Denmark.

73

74 69 Settlement of Foreign-Exchange Transactions Lone Natorp, Market Operations and Tina Skotte Sørensen, Payment Systems SUMMARY For the last 25 years, participants in the financial markets have focused on reducing the settlement risk on foreign-exchange transactions. In 1996, the central banks formulated a strategy in accordance with which the private banks have improved their own management of foreignexchange settlement risk. The establishment of the international clearing and settlement system CLS 1 Bank International in 2002 has enabled settlement of foreign-exchange transactions without any mutual credit risk between the parties. As more and more foreign-exchange trading migrates to CLS, it will therefore be possible to reduce the aggregate settlement risk. Surveys show that Danish market participants generally manage settlement risk prudently, but also that some participants can reduce their settlement risk by optimising settlement procedures or by settling via CLS. INTRODUCTION The daily transaction value in the international foreign-exchange markets exceeds 1,900 billion dollars. 2 When a foreign-exchange transaction has been concluded, it is settled via two opposite payments between the parties. A transaction in e.g. kroner against euro involves a payment in kroner by one party to the other and a payment in euro in the opposite direction. Foreign-exchange transactions have traditionally been settled by remitting the agreed amounts via correspondent banks 3. With this settlement form, the parties incur a mutual credit risk equivalent to the traded amount. In view of the very large trading volumes, this settlement risk can potentially have an impact on financial stability. The case Continuous Linked Settlement. According to BIS (2005), the daily value of foreign-exchange transactions concluded in April 2004 was approximately 1,900 billion dollars. The trading value is expected to increase gradually over time. A correspondent bank holds accounts for other banks and effects payments on their behalf.

75 70 Monetary Review 4th Quarter 2006 FOREIGN-EXCHANGE SETTLEMENT RISK AND FINANCIAL STABILITY Box 1 Over time there have been several examples of how settlement of foreign-exchange transactions via the traditional channels for international payments can cause financial problems to spread among participants in the foreign-exchange market. In June 1974, the German banking supervision authority, Bundesaufsichtsamt für das Kreditwesen, withdrew Bankhaus Herstatt's banking licence with immediate effect and Herstatt suspended its payments. The suspension of payments was announced at 3.30 p.m. Central European Time, i.e a.m. US Eastern Time (New York). At that time, several of Herstatt's foreign-exchange counterparties had effected payments of D-marks to Herstatt via their correspondent banks in Germany in the expectation that they would receive dollars from Herstatt later in the day. However, the suspension of payments led Herstatt's correspondent bank in New York to stop all payments from Herstatt's account, which left Herstatt's counterparties with credit exposure equivalent to the principal of the D-mark payments made. In February 1990, serious liquidity problems led to the collapse of the Drexel Burnham Lambert, DBL, group. As the liquidity problems became known to the counterparties of the subsidiary, Drexel Burnham Lambert Trading, DBLT, they became less inclined to assume credit and liquidity exposures in connection with the settlement of foreign-exchange transactions with DBLT. DBLT, on the other hand, did not wish to "pay first", fearing that the counterparty would not pay its leg of the transaction, but would rather set off the foreign exchange received against other claims. Since DBLT could prove that it was solvent, the Bank of England agreed to establish a temporary settlement facility to ensure that both parties supplied the foreign exchange they had sold. In July 1991, Bank for Credit and Commerce International (BCCI) suspended its payments, thereby inflicting losses on British and Japanese counterparties that had purchased foreign exchange from BCCI. A UK counterparty's payment of pounds sterling to BCCI was effected as planned, while BCCI's opposite payment of dollars was delayed and later revoked by BCCI's correspondent bank in the USA when the suspension of payments became known. A Japanese counterparty incurred losses due to the time difference between Japan and the USA. The counterparty had supplied yen to BCCI during Japanese opening hours, but BCCI's accounts in New York were frozen before payment of the dollar leg of the transaction had been effected. of Bankhaus Herstatt in 1974 highlighted this issue, but there are also more recent examples of how settlement problems in the foreignexchange markets have jeopardised financial stability, cf. Box 1. Since 2002, the international clearing and settlement system CLS Bank International has enabled settlement of foreign-exchange transactions in a manner whereby the parties do not incur a mutual credit risk. Around half of the global foreign-exchange trading is estimated to take place via CLS. The remaining transactions are settled via traditional channels. Financial stability is an important objective for the central banks. Since the failure of Herstatt, central banks have therefore worked with super-

76 71 THE BIS STRATEGY TO REDUCE FOREIGN-EXCHANGE SETTLEMENT RISK Box 2 In March 1996, the BIS Committee on Payment and Settlement Systems, CPSS, published the report Settlement Risk in Foreign Exchange Transactions. The report outlined a method to calculate the exposure on settlement of foreign-exchange transactions. A survey of the practices of 80 international banks showed that in many cases more than three days passed from the bank supplied one currency until it knew with certainty that the other currency had been received. This meant that a bank might incur a credit risk equivalent to the aggregate value of several days' trading. Moreover, the survey showed that most of the banks either did not calculate the exposure to settlement risk correctly or did not calculate it at all. Against this background, the G10 central banks formulated a three-track strategy to reduce the banks' foreignexchange settlement risk: 1) The individual banks should take steps to improve their own risk management by measuring and managing their exposures to foreign-exchange settlement risk on an ongoing basis and by optimising settlement procedures with a view to reducing the duration of the exposure. 2) Industry groups should take action to create one or more risk-reducing multicurrency services ensuring that the two legs of a foreign-exchange transaction are settled simultaneously in accordance with the PvP 1 principle. 3) Central banks should seek to support and facilitate private initiatives by offering clear recommendations for measuring and managing settlement exposures, by cooperating with private banks to develop new settlement facilities, and by making relevant enhancements to the national payment systems. In the next two years, BIS/CPSS conducted further surveys with a view to following up the 1996 strategy. 2 These surveys showed that although the banks' awareness and management of settlement risk had improved substantially, more than 60 per cent of the banks in the survey still underestimated their exposure to foreign-exchange settlement risk. In the spring of 2006, CPSS launched a survey of how more than 100 financial institutions and enterprises managed their foreign-exchange settlement risk, including their use of CLS. The purpose was another follow-up on the 1996 strategy. The overall conclusions of the Danish part of the survey are presented at the end of this article. No Danish market participants took part in the previous BIS surveys. The establishment of CLS settlement in September 2002 implemented item 2 of the BIS strategy. It had been preceded by around five years' preparatory work. Throughout the process, the central banks maintained a dialogue with the banks behind CLS with a view to monitoring the development and helping to identify issues in relation to financial stability. 1 2 Payment versus Payment. The results were presented in BIS (1998). visory authorities and private banks to reduce foreign-exchange settlement risk. The work of the central banks is coordinated in the BIS 1 Committee on Payment and Settlement Systems, CPSS. In 1996, CPSS formulated a strategy to reduce foreign-exchange settlement risk, which 1 The Bank for International Settlements (BIS) serves as a banker for central banks.

77 72 Monetary Review 4th Quarter 2006 has been followed up regularly. Most recently, in April 2006, BIS conducted a survey of the market participants' management of foreignexchange settlement risk, including their use of CLS. BIS' work in this respect is described in Box 2. The following presents a general description of various settlement methods and the related risks, as well as an account of Danish participants' foreign-exchange settlement risk, based on the above BIS survey, among other things. DIFFERENT SETTLEMENT METHODS AND THE ASSOCIATED RISKS Settlement risk on a foreign-exchange transaction arises when one party to the transaction has supplied the sold currency and has not yet received the purchased currency. Settlement risk comprises various types of risk: Credit risk is the risk of financial loss as a consequence of a counterparty's inability to meet its obligations, so that the purchased currency is received neither at the time of settlement nor at a later time. Liquidity risk is the risk of incurring a loss because the purchased currency is not received at the expected time. The loss arises e.g. if alternative liquidity can only be obtained at a cost. Operational risk is the risk of losses due to human errors, system errors and external events such as natural disasters, acts of terrorism, etc. In connection with foreign-exchange settlement, operational risk typically results in unexpected credit and/or liquidity risk on the purchased currency. The most frequently used settlement methods and the related risks are described below. The terminology used is taken from the BIS reports on settlement risk. Settlement via correspondent banks and payment systems A foreign-exchange transaction can be settled by exchanging the traded amounts via the traditional channels for international payments. Settlement of e.g. a euro/dollar transaction involves two opposite payments in, respectively, euro and dollars that are effected independently in payment systems for the two currencies, or via the parties' correspondent banks in euro and dollars. 1 1 For a more detailed description of payments via payments systems and correspondent banks, see Chapter 3 of Payment Systems in Denmark, Danmarks Nationalbank 2005.

78 73 STATUS OF A TRANSACTION DURING THE SETTLEMENT PROCESS Chart 1 (1) Revocable (2) Irrevocable (3) Uncertain (4) Settled /(5) Failed Transaction Deadline for unilateral revocation of payment instruction for sold currency Time of final receipt of purchased currency Time of verification of final and failed receipts of purchased currency (1) Revocable: The payment instruction for the sold currency has not yet been submitted, or it may be revoked without the consent of the correspondent bank and/or the counterparty. In this phase there is no settlement risk. (2) Irrevocable: The payment instruction for the sold currency is irrevocable, i.e. it has been settled in the relevant payment system, or consent is required from e.g. the correspondent bank and/or the counterparty if it is to be revoked. A payment instruction may be irrevocable the moment it is sent. Since the purchased currency has not yet been received, there is a settlement risk equivalent to the full amount of the transaction. (3) Uncertain: It is uncertain whether the purchased currency has been received since final verification will not take place until later. Consequently, there is still a potential settlement risk on the purchased currency. (4) Settled: Receipt of the purchased currency has been verified. There is no longer any settlement risk. (5) Failed: It is known with certainty that the purchsed currency has not been received as agreed. A settlement risk still exists until the currency is received. The BIS method for calculation of settlement risk Settlement of foreign-exchange transactions via correspondent banks and payment systems involves settlement risk since the parties must supply the sold currency without having any guarantee that they will receive the purchased currency 1. The exposure is equivalent to the purchased amount and lasts from the time of remittance of the sold currency until the time when the purchased currency has been received. The various settlement phases are illustrated in Chart 1. As a minimum, the total exposure to settlement risk is equivalent to the sum of the transactions where payment of the sold currency is irrevocable, and where the purchased currency has not yet been received, and the transactions where it is known with certainty that the purchased currency has not been received. In Chart 1, this is equivalent to the sum of status (2) and (5) transactions. To this should be added the potential settlement risk on transactions where receipt of the purchased currency has not yet been verified. In Chart 1, this means that a 1 In addition to the settlement risk, there may be credit risk on the correspondent bank, as well as risks related to participation in payment systems. For a more detailed description of the risks connected with settlement of payments, see Chapter 4 of Payment Systems in Denmark, Danmarks Nationalbank 2005.

79 74 Monetary Review 4th Quarter 2006 proportion of the status (3) transactions should be added to the exposure. The maximum exposure is thus the sum of status (2), (5) and (3) transactions. The duration of a market participant's exposure to settlement risk on the individual transaction depends on the currencies traded and whether the transaction is settled via direct entry in payment systems or via the correspondent bank network. It is also of significance whether the two sides of the transaction are settled in different time zones. Surveys from 1997 showed that the settlement risk on a transaction typically had a duration of hours, depending on the currencies traded. In some cases, the duration of the settlement risk was more than three days. Consequently, a market participant's current total exposure to settlement risk could be the aggregate value of several days' trading. Settlement via CLS CLS is an international clearing and settlement system that settles foreign-exchange transactions in 15 currencies at present 1. CLS was established as a bank owned by 69 of the world's largest private banks. In CLS, the two eligible payment instructions relating to a foreignexchange transaction are settled simultaneously, i.e. Payment versus Payment, PvP. The parties to the transaction only supply the currency they have sold if they simultaneously receive the currency they have bought. Consequently, the parties to a CLS transaction incur no mutual credit risk. CLS settlement is described in Box 3. CLS currently has 55 direct participants (settlement members), whose foreign-exchange transactions are settled via accounts with CLS. Only CLS shareholders can join as direct participants. In addition, there are around 750 indirect participants (third parties), who submit and settle their foreign-exchange instructions via direct participants. Direct participants are typically large international banks, while the relatively smaller banks, brokers, funds and non-financial corporations often participate as indirect participants. 2 Transactions in CLS are settled by transferring the traded amounts between the relevant participants' accounts with CLS. The two sides of the transaction are entered to CLS' books simultaneously, i.e. PvP. Even though transactions are thus settled individually, the participants' pay- 1 2 US, Canadian, Australian, New Zealand, Singapore and Hong Kong dollars, euro, Japanese yen, pounds sterling, Swiss francs, Korean won, South African rands, Danish and Norwegian kroner and Swedish kronor. A number of players in the Danish foreign-exchange market settle their transactions via CLS. Nordea and Danske Bank are direct participants, while Amagerbanken, Forstædernes Bank, HSH Nordbank, ISS Finans, ISS Global, Jyske Bank, Nykredit Bank, Saxo Bank and Sydbank are indirect participants.

80 75 ments to CLS are effected on a net basis 1. Net pay-ins are split into several smaller pay-ins, which must be received by CLS at fixed times during the settlement day. Pay-outs to participants are made on an ongoing basis. Pay-ins to and pay-outs from CLS take place via the national RTGS 2 systems. A participant need not be connected to all 15 RTGS systems, but may choose to let a "nostro agent" handle its pay-ins and pay-outs. A total of 48 large banks, typically settlement members themselves, offer nostro services to CLS participants. 3 For a more detailed description of CLS settlement see the Appendix. 4 Settlement risk in CLS The establishment of CLS has changed the risk scenario for foreignexchange settlement. The application of the PvP settlement principle has eliminated the traditional credit risk between the parties to a foreignexchange transaction. Settlement still involves a certain degree of liquidity risk since pay-ins to CLS must observe tight deadlines. In the structure of CLS settlement, great importance has, however, been attached to limiting the participants' liquidity requirements and thus their liquidity risk. A core element is that pay-ins to CLS take place on a net basis. This reduces the liquidity requirement in CLS considerably, cf. Box 3. The liquidity requirement is reduced further by splitting the individual participants' pay-ins in a given currency into several smaller, time-lagged pay-ins. In addition, liquidity management is supported by a number of online facilities, issue of initial pay-in schedules, etc. For indirect participants in CLS settlement, the risk scenario is slightly different. Indirect participants incur no credit risk on their trading counterparties, but instead they may incur a minor credit risk on the direct participant. This is the case if the indirect participant pays in the sold foreign exchange to the direct participant in the morning and receives the purchased foreign exchange during the day. Unlike settlement outside CLS, the settlement risk is always incurred intraday, and the exposure solely corresponds to the net amount for each of the currencies sold As an example, a CLS member has submitted two foreign-exchange instructions: purchase of 100 million dollars against sale of euro, and purchase of kroner against sale of 100 million dollars. The two dollar positions eliminate each other so that the member in question pays only euro and receives only kroner. Real-Time Gross Settlement. An RTGS system is a payment system in which payments are settled individually and immediately. RTGS systems are typically owned and operated by central banks. Most CLS participants use Danske Bank and Nordea as nostro agents for their pay-ins and pay-outs in Danish kroner. SEB, ABN Amro and Jyske Bank also offer nostro services in Danish kroner. CLS is described in more detail in Chapter 8 of Payment Systems in Denmark, Danmarks Nationalbank, Another option is for the indirect participant to obtain intraday credit from the direct participant so that the amount due is payable by the end of business on the same day. In that case the direct participant incurs a credit risk on the indirect participant.

81 76 Monetary Review 4th Quarter 2006 CLS BANK INTERNATIONAL Box 3 CLS was established in 2002 by some of the world's largest private banks in response to the BIS strategy's call for development of risk-reducing settlement facilities. CLS currently settles foreign-exchange transactions in 15 currencies. The Danish krone, along with the Norwegian krone and the Swedish krona, joined CLS in September New currencies are added on an ongoing basis. In April 2006, CLS settled an average of 125,000 foreign-exchange transactions daily, for a total value of 1,350 billion dollars. This is estimated to constitute around half of the global trade in foreign exchange. 1 The distribution of CLS settlement by currencies is shown in the Chart below. As the Chart illustrates, a large share of the foreign-exchange transactions have one leg in dollars. The participants' payments to CLS are netted. Total daily pay-ins to CLS in April 2006 averaged 42 billion dollars, compiled as the sum of all net pay-ins translated into dollars. Net pay-ins constitute between 1 and 8 per cent of the gross settlement, which is the total trading value, depending on the currency. The Chart shows the netting effect (net pay-ins as a percentage of gross settlement) for the various currencies. It is seen that the netting effect increases with the gross settlement value for the currency in question. The largest gross daily settlement value in CLS so far was just over 2,700 billion dollars. Total net pay-ins were 51 billion dollars. This record was achieved on 20 September 2006, the date on which quarterly futures settlement took place. The record for number of transactions, just over 250,000, was set on 17 January 2006, the day after Martin Luther King Day in the USA. Transactions in dollars cannot be concluded for settlement on US public holidays. The general pattern in connection with US public holidays is that settlement is postponed until the following day, so that the settlement volume in dollars almost doubles on these days. TRANSACTION VALUE AND NETTING EFFECT IN CLS BY CURRENCIES Value of transactions settled (gross settlement) Billion dollars 1,200 1,000 Per cent Net pay-ins as a percentage of gross settlement USD EUR JPY GBP CHF AUD CAD SEK NZD NOK DKK HKD SGD ZAR KRW USD EUR JPY GBP CHF AUD CAD SEK NZD NOK DKK HKD SGD ZAR KRW Note: Averages for April The CLS settlement figures are not directly comparable with the BIS statistics for global foreign-exchange trading. The concentration of settlement of foreign-exchange transactions in CLS means that risk is concentrated in CLS. CLS is a complex system that in practice interlinks 15 RTGS systems and 55 participants. Operational problems in the CLS system or settlement errors could therefore have a major global impact. For example, lacking or delayed pay-ins to CLS by one participant could delay pay-outs to other participants, so that RTGS systems in other countries might have to stay open for longer.

82 77 As a provision against system failures, parallel operation of the CLS system has been established in London and New York, and a complete "out of region" back-up solution is underway. When this solution has been fully implemented as planned in 2007, CLS will comply with the US authorities' enhanced requirements 1 after the attacks on the World Trade Center in New York on 11 September In addition, further contingency planning ensures that settlement can be concluded even if the participant with the largest negative net position does not meet its payment obligations. Any losses in connection with settlement are covered by a loss sharing agreement between the direct participants, which are large, well-capitalised private banks. Overall, settlement of foreign-exchange transactions via CLS reduces the settlement risk considerably compared with traditional settlement methods. This is confirmed by a survey of the settlement risk of large US CLS participants two years after the introduction of CLS. 2 The aggregate settlement risk in the foreign-exchange market can thus be reduced by increasing the proportion of transactions settled via CLS. CLS is continuously working to introduce new currencies and products in CLS settlement. In future, this will enable CLS participants to settle a larger share of their foreign-exchange transactions via CLS. In addition, an influx of new participants is essential to the further growth of CLS settlement. Primarily the number of indirect participants is expected to rise. FOREIGN-EXCHANGE SETTLEMENT BY DANISH MARKET PARTICIPANTS In the spring of 2006, BIS launched a survey of settlement risk on foreign-exchange transactions. The purpose is to establish the extent to which the objectives of the BIS strategy have been met, and whether further measures are required in order to reduce settlement risk. Approximately 100 financial institutions and enterprises from around 20 countries participate in the BIS survey. The participants have been selected with a view to covering at least 80 per cent of the foreignexchange trading in the country in question. In addition, importance has been attached to the inclusion of different types of market participant, including both CLS participants and market participants without CLS access. For the purpose of the survey, the participants have supplied data for foreign-exchange transactions settled in April 2006 and answered questions about how they manage settlement risk in practice. The survey is based on the BIS definitions, cf. above. BIS expects to publish a report on the findings of the survey in Cf. Board of Governors of the Federal Reserve System et al. (2003). The Payments Risk Committee (2005).

83 78 Monetary Review 4th Quarter 2006 SETTLEMENT METHODS USED BY DANISH PARTICIPANTS Chart 3 Per cent Participant 1 Participant 2 Participant 3 Participant 4 Participant 5 Participant 6 Participant 7 CLS Correspondent banks Own accounts (with risk) Own accounts (without risk) Bilateral netting Source: Stated by Danish participants in the survey in April Foreign-exchange transactions settled by foreign branches of Danish banks are included. Danmarks Nationalbank is responsible for the Danish part of the survey, comprising seven Danish market participants. Two of these do not currently have access to CLS settlement. Some of the results of the Danish part of the survey are outlined below. First, different settlement methods used by Danish participants are described. This is followed by a review of the participants' foreign-exchange settlement within and outside CLS, respectively. For settlement outside CLS, the exposure to settlement risk is calculated, as well as its duration and the concentration of settlement risk on the Danish participants' counterparties. Finally, the Danish participants' management of settlement risk in practice is described. Settlement methods used in Denmark Chart 3 shows the distribution by settlement methods for the seven Danish participants in the BIS survey. The five participants that have direct or indirect access to CLS settle an average of approximately 80 per cent of their transactions via CLS. The remaining transactions are settled via the traditional channels for international payments, or alternatively via own accounts 1 and bilateral netting 2. Indirect participants in CLS, who do not 1 2 Foreign-exchange transactions between a bank and its customers can be settled via "own accounts" if the customer holds accounts with the bank in both the currencies traded. The settlement risk can be reduced if the bank offers a PvP service, so that accounts are debited and credited simultaneously. Settlement on own accounts can therefore take place with or without settlement risk. Two parties set off their mutual claims so that only the net sums are exchanged.

84 79 CLS SETTLEMENT IN KRONER Chart 4 Kr. billion May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Net pay-ins Total transaction value trade on behalf of customers, have the option to settle via CLS only. This is the case for one of the Danish participants in the survey, cf. Chart 3. The two participants in the survey that do not have access to CLS mainly settle foreign-exchange transactions in the traditional manner via correspondent banks and/or payment systems. Consequently, these participants are exposed to settlement risk equivalent to the full amount of the purchased currency. Danish participants' use of CLS The BIS survey shows that CLS is now the most common method for settlement of foreign-exchange transactions among the large market participants in Denmark. This means that the credit risk on foreignexchange settlement has been reduced significantly, which also benefits financial stability in Denmark. Chart 4 shows CLS settlement in kroner on the basis of CLS data. It is seen that net pay-ins are very small relative to the aggregate value of the transactions settled, which illustrates the liquidity savings from net pay-ins, cf. above. On certain days Danish or US public holidays very few transactions in kroner are settled, if any. The reason is that the other leg in krone transactions is virtually always in dollars. The tight deadlines for pay-ins for CLS settlement require strict liquidity management by the participants. For the Danish CLS participants, the largest single pay-ins on a daily basis are typically kr. 1-2 billion, but payins exceeding kr. 10 billion are seen a few times a year. No Danish CLS

85 80 Monetary Review 4th Quarter 2006 participant has indicated that the tight deadlines and the size of the individual pay-ins are problematic. The BIS survey shows that on average approximately 20 per cent of the Danish CLS participants' foreign-exchange transactions are still settled in the traditional manner or via other settlement methods that involve some risk. The primary reason is that not all transactions can be settled via CLS. According to the participants, this applies to transactions in currencies or with counterparties that are not part of CLS, or transaction types, e.g. intraday transactions, that cannot be settled via CLS. Danish participants' settlement via correspondent banks and payment systems Foreign-exchange transactions settled via traditional channels for international payments involve settlement risk on the full transaction amount, cf. above. It is important that the Danish market participants are aware of the size and duration of their exposures when settling outside CLS, and know how to reduce and manage such settlement risk. Size of the settlement risk Chart 5 shows the total average daily exposures to settlement risk for the Danish participants in the BIS survey, except the one participant that DANISH PARTICIPANTS' TOTAL EXPOSURE ON TRADITIONAL SETTLEMENT DAILY AVERAGES IN APRIL 2006 Chart 5 USD billion USD EUR DKK CHF SEK NOK Other GBP JPY CAD AUD ZAR HKD SGD NZD KRW Source: Stated by Danish participants in the survey in April Foreign-exchange transactions settled by foreign branches of Danish banks are included.

86 81 always settles via CLS. The Chart shows that, after dollars, euro and kroner are the most frequently traded and settled currencies. On average, the Danish participants' aggregate daily exposure to settlement risk is approximately 34 billion dollars for all currencies taken as one. The exposures of the individual participants vary considerably in size. In order to show whether these large exposures to settlement risk constitute a serious risk, the exposure of each participant can be compared to its equity capital. For participants with access to CLS, the exposure is lower than the equity capital in some cases, but slightly higher in other cases. For one of the participants without access to CLS, the exposure is several times the equity capital. This means that delayed payments from or failure of just a few counterparties could have serious implications, and consequently this participant must be extra prudent in its management of settlement risk. Duration of the settlement risk The exposure to settlement risk lasts from the time when payment of the sold currency becomes irrevocable to the time when final receipt of the purchased currency is verified, cf. Chart 1. Table 1 shows the exposures in hours of Danish participants in the BIS survey to settlement risk. The duration is calculated using the BIS definition and is shown for selected currencies. The longest and shortest exposures of the Danish participants are shown and averages calculated. It is seen that the duration of the participants' exposures to settlement risk varies considerably. The BIS survey shows that the variation is attributable to major differences in the settlement practices of the participants. One participant has thus reduced the duration of the exposure to settlement risk, so that only intraday risk is incurred for the most fre- DANISH PARTICIPANTS' EXPOSURE, TRADITIONAL SETTLEMENT Table 1 Hours Purchase/sale of foreign exchange Max. exposure Min. exposure Average exposure Time difference 1 Dollars/euro Euro/dollars ½ -6 Dollars/yen ½ 14 Euro/yen ½ 36 8 Dollars/kroner... 42½ 22½ 30 6 Kroner/dollars ¾ -6 Euro/kroner... 42½ 9 27¾ 0 Kroner/euro Source: Stated in the BIS survey from April 2006 by Danish participants settling in the traditional manner in that period. 1 The difference between the time zones of the purchased and sold currencies.

87 82 Monetary Review 4th Quarter 2006 quently traded currencies 1. In contrast, one of the participants without access to CLS settlement is exposed to settlement risk for a far longer time than the other Danish participants. For the individual participant, the duration of the exposure varies, depending on the currencies traded. This is a result of time differences, since the longest exposures are typically seen when the sold currency is from a more easterly time zone than the purchased currency 2. The longest average exposures of the Danish participants relate to purchases of dollars, euro or kroner against yen, but the dollar/euro and dollar/kroner exposures are also long. Concentration of the settlement risk If the purchased currency is received from a small number of counterparties, the settlement risk is concentrated. This implies greater dependency on the individual counterparty meeting its obligations than if the settlement risk is distributed on a larger number of counterparties. The Danish participants in the BIS survey that have access to CLS spread their settlement risk on many counterparties, while the two participants that do not have access to CLS spread their settlement risk on a smaller number of counterparties. Danish participants' management of settlement risk The size and duration of the exposure to settlement risk may be considerable when transactions are settled outside CLS. However, market participants can actively limit their settlement risk by reducing the duration of the exposures. This can be done by postponing payment of the sold currency and/or by checking for receipt of the purchased currency at an earlier time. The BIS survey shows that a few Danish participants have made considerable progress in their efforts to reduce the duration of the exposure. Among other things, they can enter payments directly to several of the relevant payment systems during the actual settlement day. Moreover, they check for receipt of the purchased currency regularly throughout the settlement day. One of the participants employs staff that check for certain currencies until 6 p.m. or later (Danish time) on the settlement day. As Table 1 shows, however, several participants should be able to reduce the duration of their exposures substantially. For participants that settle solely via correspondent banks, the duration of the exposures can 1 2 Cf. Table 1, the column showing the minimum exposures for euro/dollars, kroner/dollars, euro/kroner and kroner/euro. I.e. the currency pairs for which a positive time difference is stated in Table 1.

88 83 be reduced if agreements are concluded to postpone the deadline for remittance/withdrawal of amounts supplied. If settlement takes place via correspondent banks, receipt of the purchased currency cannot be verified until it has been confirmed by the correspondent bank, e.g. via a statement of account. The duration of the exposure may be reduced via an agreement with the correspondent bank to forward confirmation as soon as possible. At present, it is not possible to eliminate settlement risk completely, and consequently market participants must be able to manage the existing settlement risk prudently. The Danish participants in the BIS survey have stated that they set limits to the exposure vis-à-vis the individual counterparties at any given time, depending on the credit ratings of the counterparties. In addition, the participants calculate their exposures to settlement risk on an ongoing basis. In other words, the participants manage their settlement risk in the same way as other types of credit risk. CONCLUSION Danmarks Nationalbank supports the BIS strategy to reduce the settlement risk on foreign-exchange transactions and participates in the work to implement this strategy. The most recent BIS survey shows that the Danish participants generally manage settlement risk prudently. There is, however, scope for further reduction of settlement risk. This particularly applies to market participants whose exposure to settlement risk is very large in relation to their equity capital. Even though the individual market participant can reduce the duration of the exposure, CLS is the only settlement method that ensures settlement of foreign-exchange transactions without credit risk on the counterparty. Indirect participation in CLS significantly limits the credit risk.

89 84 Monetary Review 4th Quarter 2006 Appendix: Structure of CLS Settlement CLS settlement participants hold multi-currency accounts with CLS, in which their balances in the various CLS currencies are registered. On the basis of the trading instructions submitted, CLS calculates the participants' net positions in each currency. Pay-ins to CLS take place in the national RTGS systems by transfer of the relevant amounts from the participants to the CLS account at the central bank and are automatically credited to the participants' accounts with CLS. Foreign-exchange transactions are settled by transferring the amounts between the relevant participants' accounts with CLS. The two eligible payment instructions are settled simultaneously (PvP). Settlement of a transaction takes place subject to a number of risk measures, including that a participant's overall balance for all currencies must not be negative. This ensures that CLS does not incur a credit risk on the participants. 1 In connection with pay-outs, CLS transfers funds from its own central bank account to the participants' central bank accounts via the national RTGS systems. At the same time, the pay-outs are debited to the participants' accounts with CLS. Pay-out are subject to CLS risk measures. If a participant does not have an account in an RTGS system, or does not have access to sufficient liquidity in a given currency, a nostro agent 2 in the relevant currency may be used. The participants' pay-ins to and pay-outs from CLS are effected in the national RTGS systems for the currencies in question, as described above. To allow this, the RTGS systems must be open at the same time across the relevant time zones. Settlement in CLS has therefore been scheduled to take place between 7.00 a.m. and noon Central European Time (CET), i.e. afternoon/evening in Asia/Pacific and night/early morning in North America. 3 The CLS settlement cycle is outlined in Box For a more detailed description of CLS risk management, see Danmarks Nationalbank (2005). Nostro agents are typically CLS participants who offer to send and receive payments in their "domestic currency" to and from CLS on behalf of other participants. The RTGS systems in Asia/Pacific close before noon CET. Therefore it is sought to settle and effect payments in these currencies before a.m. CET.

90 85 CLS SETTLEMENT CYCLE Box 4 The participants submit trading instructions to CLS on an ongoing basis. At CET, CLS calculates each participant's preliminary net position for each currency and sends out an initial pay-in schedule to participants. Until 6.30 a.m. it is possible to submit trades for settlement on the same day. Immediately after 6.30 a.m. CLS sends out the final pay-in schedule to participants. A participant's total pay-in is broken down into three or five pay-ins per currency. Pay-ins must be received by CLS within fixed time limits, cf. the Chart. Initial pay-in schedule Final pay-in schedule Deadline for first pay-in Deadine for third pay-in. Pay-ins and pay-outs in AUD, HKD, JPY, KRW, NZD and SGD concluded Deadline for fifth pay-in. Pay-ins and pay-outs in CAD, CHF, DKK, EUR, GBP, NOK, SEK, USD and ZAR concluded Midnight 6.30 a.m a.m a.m a.m a.m a.m. Noon Pay-outs CET CLS opens for settlemtnt of trandes Deadline for second payin. Settlement of trades sought to be concluded. Deadline for fourth pay-in At 7.00 a.m. CLS opens for settlement of foreign-exchange transactions, and as soon as CLS has received pay-ins from the participants or their nostro agents settlement commences. Foreign-exchange transactions are settled individually by simultaneously entering the two sides of a transaction (PvP) to the respective participants' accounts with CLS. CLS seeks to settle and enter all foreign-exchange transactions by 9.00 a.m. In accordance with CLS' risk measures, settlement of trades may be concluded before all pay-ins have been received. Currency pay-outs do not take place according to a fixed schedule, but on an ongoing basis subject to CLS' risk measures. It is sought to conclude pay-outs in Asian/Pacific currencies immediately after a.m. and in other currencies immediately after noon.

91 86 Monetary Review 4th Quarter 2006 LITERATURE BIS, Settlement Risk in Foreign Exchange Transactions, Report prepared by the Committee on Payment and Settlement Systems, Bank for International Settlements, March BIS, Reducing Foreign Exchange Settlement Risk: A Progress Report, Report prepared by the Committee on Payment and Settlement Systems, Bank for International Settlements, July BIS, Supervisory Guidance for Managing Settlement Risk in Foreign Exchange Transactions, Basel Committee on Banking Supervision, Bank for International Settlements, September BIS, Triennial Central Bank Survey. Foreign Exchange and Derivatives Market Activity in Board of Governors of the Federal Reserve System et al., Interagency Paper on Sound Practices to Strengthen the Resilience of the U.S. Financial System, April CLS Bank International, Regulator's Summary, International Rules and Member Handbook. CLS home page: Danmarks Nationalbank, Payment Systems in Denmark. Norges Bank, 2000: Settlement risk in foreign exchange transactions. Economic Bulletin, Q The Payments Risk Committee, Foreign Exchange Settlement Risk, Report by the FX Settlement Risk Task Force, February 2005.

92 87 IMF Review of the Financial Sector in Denmark Gitte Wallin Pedersen, International Relations The IMF has in recent years stepped up the financial surveillance of its member countries in view of the financial systems' key role in ensuring economic stability. In the IMF reviewed the financial sector in Denmark, providing a good opportunity for an in-depth analysis of the financial systems. Overall, the IMF found the systems to be resilient and well supervised. Notwithstanding the IMF's surveillance, each individual country is responsible for its financial sector. BACKGROUND TO THE IMF's FINANCIAL SURVEILLANCE Since the financial crisis in Asia in , the IMF has reviewed the financial sector in more than 100 of its 184 member countries. In it was Denmark's turn to be reviewed under the Financial Sector Assessment Program, FSAP, whereby the IMF conducts detailed review and assessment of virtually every aspect of the financial system, cf. Box 1. The rationale behind FSAP is that the free movement of capital has made the financial sector decisive to economic stability in individual countries, as well as globally, since a financial crisis in one country can have a contagion effect on other countries. Problems in the financial systems can e.g. lead to economic crises and may require substantial expenditure to support financial institutions, which may affect economic stability. If vulnerabilities in the financial sectors of individual countries are identified and managed in time, there is less risk of new crises in the financial markets, also globally. The IMF's objective for FSAP in the individual countries is thus: to identify strengths and vulnerabilities in the financial system, to determine how key sources of risk are being managed, to ascertain the financial sector's developmental needs, and to help prioritise policy responses to meet these challenges. In the IMF review, everything is questioned in terms of both the structure and functioning of the financial systems. In some countries, the review has disclosed a need for fundamental changes. In countries with

93 88 Monetary Review 4th Quarter 2006 CONTENT OF FSAP Box 1 FSAP is a review of the financial sector in order to assess the stability of the financial systems. The IMF's point of departure is the macroeconomic conditions, and the performance of stress tests. Stress tests are simulations used to examine how various types of hypothetical macroeconomic shocks affect the key financial institutions and thereby the entire financial system. In addition, it is assessed whether international standards for the financial sector are observed for areas such as banking supervision, insurance supervision or the efficiency of payment systems. There are 12 standards in total, but the IMF and the country in question typically select 4-5 standards for the FSAP review after identifying the aspects of the financial systems that are of special interest to financial stability. Issues that are nearly always selected for review are e.g. the country's banking supervision and its measures to combat money laundering and the financing of terrorism. At the same time, the most important players and the structure of the financial sector are reviewed. This assessment comprises the division of responsibility between the country's national authorities, as well as the market conditions, i.e. whether the financial markets are characterised by a few large or many small financial institutions, or by particular financial instruments. Finally, the IMF undertakes an overall assessment of the stability of the financial systems, recommending concrete reforms and development of the financial sector. The IMF typically conducts FSAP up to an Article IV consultation, which is the core element of the IMF's surveillance of the macroeconomic conditions in the member countries. Article IV consultations take place once a year or every other year. The close link between FSAP and Article IV consultations strengthens the relations between the IMF's assessments of the macroeconomic and financial developments in the member countries. Financial system stability Macroeconomic conditions Financial regulatory and supervisory conditions Market and infrastructure conditions of the financial sektor Analysis of structure soundness, and vulnera-bilities of the financial system Assessment of observance of standards and codes Review of financial sector infrastructure and governance framework 1 Overall stability assessment 2 Structure/development assessment 3 Recommended actions and follow-up FSAP is described in more detail in Review of the Financial System in Denmark, Danmarks Nationalbank, Monetary Review, 3rd Quarter 2005, p. 43ff. well-functioning financial sectors, a review is more likely to lead to finetuning and even better systems.

94 89 THE FSAP PROCESS IN DENMARK Around 20 IMF experts participated in the review of the Danish financial sector in order to assess the scope for improvements. A large body of information is available on Danish websites, but the IMF also asked the parties involved to complete a number of questionnaires. In addition, the IMF visited Denmark to meet relevant authorities, organisations and business enterprises. The IMF visited Denmark seven times over a 12-month period. Four visits were for the sole purpose of FSAP, while the other three had a broader focus on the Danish economy and the relations between macroeconomic and financial issues. In August 2005, the IMF gave an initial presentation for the relevant authorities and organisations in the financial sector. This was followed up in May 2006 with a presentation of the IMF's main conclusions on the financial sector. In the interim period, the authorities, organisations and specific financial institutions had readily attended meetings with and answered questions from the IMF. The FSAP process was resource-consuming for both the IMF and the parties involved in Denmark. For example, almost 90 meetings with the IMF were held during a two-week period in November The final FSAP meetings with the IMF were held in June 2006 in connection with the Article IV consultation regarding the Danish economy. FSAP is linked to Article IV consultations in order to strengthen the interrelations between the IMF's assessments of respectively macroeconomic and financial developments. FSAP is an extraordinary review, whereas Article IV consultations are conducted regularly and in Denmark every other year. In the following three months, the IMF completed a number of general reports with specific assessments and recommendations for Denmark. These reports were discussed by the IMF's Executive Board on 29 September 2006 and then published 1. Chart 1 illustrates the FSAP process in Denmark. THE IMF's ASSESSMENT OF DENMARK In Denmark, the IMF's starting points were the country's favourable cyclical position, which generally has a positive impact on the performance of the Danish financial sector, and the substantial increase in lending by credit institutions. Keeping in mind the overall objective of financial stability in Denmark's case maintaining financial stability the IMF 1 Cf. Danmarks Nationalbank's website

95 90 Monetary Review 4th Quarter 2006 TIME FRAME FOR FSAP IN DENMARK Chart 1 IMF preliminary visit to Denmark Meetings with relevant national authorities and organisations Introductory discussions between national authorities and the IMF regarding stress tests and international standards Special IMF mission to Denmark Review of standards regarding banking supervision, insurance and payment and securities settlement systems Determination of framework for stress tests Meetings with national authorities, organisations and business enterprises Special IMF mission to Denmark Review of standard for anti-money laundering and combating the financing of terrorism Second IMF mission to Denmark Review of results of stress tests Provisional conclusions on financial stability Article IV consultation with Denmark Discussion of conclusions regarding FSAP Review of the Danish economy IMF press conference at Danmarks Nationalbank Discussion by the Executive Board of the IMF Article IV consultation FSAP Publication Article IV report and analyses Overall FSAP report Aug Nov Feb- 15 Mar May Jun Sep Oct 2006 was to assess whether e.g. the authorities are taking adequate measures for the favourable development to continue. The IMF's main conclusions after FSAP in Denmark illustrate how the IMF links information on the financial sector with the supervisory authorities' opportunities to influence the financial sector and the impact of macroeconomic policy. The IMF's overall assessment is that Denmark's financial systems are resilient and well supervised. However, it is pointed out that sustaining the strong financial performance and stability requires support from the macroeconomic policy and continued supervisory vigilance, including continuous monitoring and assessment of the economic developments and related risks. It is also stated that any easing of the banks' capital requirements allowing for further expansion of credit should be done gradually, and that the supervisors should make greater use of their regulatory authority to require additional capital as appropriate in individual cases. As regards the housing market, the IMF finds that the current situation warrants close monitoring and strict application of supervisory regulations. The IMF finds this to be reflected in the letter from the Danish Financial Supervisory Authority (DFSA) to the credit institutions in February 2006, in which the DFSA requests the institutions to include in

96 91 IMF MAIN RECOMMENDATIONS TO DENMARK Box 2 Near-term financial stability Any easing of capital requirements under Basel II should be done gradually, and supervisors should make use of their regulatory authority to require additional capital as appropriate in individual cases. The mission supports the regulation limiting the decreases in capital requirements in for institutions that would use advanced methods for determining their capital adequacy. Efforts to better monitor, compile information, and model the impact of various shocks should be intensified through strengthened stress testing capacity at the DFSA. The DFSA should consider more frequent inspections as warranted. Close monitoring of the housing market, strict adherence to supervisory rules, and effective consumer information about risks are recommended to avoid unbalanced housing market developments which could adversely affect financial stability. Structural issues The autonomy and accountability of the DFSA should be entrenched by providing the agency a statutory basis and granting it greater budgetary autonomy. With respect to the latter, separation of its legislative and regulatory activities and supervisory budgets should be considered. The resources available to the DFSA should be reassessed in light of the need to expand the agency s capacity to verify banks internal models to calculate liable capital, undertake stress testing, and ensure an effective framework for anti-money laundering and combating financing of terrorism. While the regulatory framework has worked well, it would benefit from upgrading the guideline on internal controls to a more binding Executive Order and broadening the coverage of fit-and-proper requirements. Source: IMF report "Financial System Stability Assessment" (2006), box 1, p. 7. the calculation of the solvency requirement any risk for the institution concerning loans based on real estate as collateral in the light of e.g. the rising housing prices. 1 In addition, the IMF finds it important to ensure continued public information to consumers on the risks involved in overborrowing. The IMF's main recommendations are presented in Box 2. Two analyses were at the core of the IMF's assessment of Denmark. 1. Performance of a number of stress tests to calculate the impact of hypothetical macroeconomic shocks on the financial sector Review of Denmark's compliance with international standards in the financial area. The IMF and Denmark selected four areas of special relevance to the stability of financial systems in Denmark, i.e. banking supervision, insurance, payment systems and securities settlement systems, and anti-money laundering and combating the financing of terrorism. 1 2 Cf. the DFSA's website Cf. Danmarks Nationalbank's annual publication "Financial stability", which includes stress tests.

97 92 Monetary Review 4th Quarter 2006 Stress testing From the outset, the financial institutions were found to have recordhigh profits and to appear robust. The banks' average solvency ratio, i.e. regulatory capital as a ratio of risk-weighted assets, was thus 13 per cent at the end of 2005, which is considerably above the statutory capital adequacy requirement of 8 per cent. The IMF therefore examined the consequences of possible macroeconomic shocks to the balance sheets of the financial institutions. From the outset, the IMF and Danmarks Nationalbank agreed on three extreme macroeconomic scenarios calculated using Danmarks Nationalbank's economic model: 1. Boom-bust in real estate prices and credit. Asset prices fall and credit growth decreases by 14 per cent. The result is an increase in unemployment to 8.8 per cent after three years with zero growth in GDP over three years. 2. External shock to the effective exchange rate stemming from correction of global imbalances, with the Danish krone (and the euro) appreciating by 29 per cent against other currencies. The result is an increase in unemployment to 8.2 per cent after three years and growth in GDP by 2.6 per cent over three years. 3. Boom-bust in real estate prices and credit combined with an interestrate increase by 250 basis points. The result is an increase in unemployment to 10.1 per cent after three years and a decrease in GDP by 2.9 per cent over three years. Danmarks Nationalbank, the DFSA and selected Danish banks hereafter contributed to the macroeconomic stress tests since the consequences for the banks' solvency were to be calculated by both the IMF and the banks. The DFSA agreed on participation by the individual banks since THE IMF's CALCULATIONS OF SOLVENCY RATIOS (AVERAGES) Table 1 Scenarios Boom-bust in real estate prices and credit. Unemployment rises to 8.8 per cent, with zero growth in GDP over three years External shock to effective exchange rate. Unemployment rises to 8.2 per cent, and GDP growth is only 2.6 per cent over three years Boom-bust in real estate prices and credit and interest-rate increase by 250 basis points. Unemployment rises to 10.1 per cent and GDP decreases by 2.9 per cent over three years Source: IMF (2006) Financial System Stability Assessment.

98 93 THE BANKS' CALCULATIONS OF SOLVENCY RATIO (AVERAGES) Table 2 Scenarios Boom-bust in real estate prices and credit. Unemployment rises to 8.8 per cent, with zero growth in GDP over three years External shock to effective exchange rate. Unemployment rises to 8.2 per cent, and GDP growth is only 2.6 per cent over three years Boom-bust in real estate prices and credit and interest-rate increase by 250 basis points. Unemployment rises to 10.1 per cent and GDP decreases by 2.9 per cent over three years Source: IMF (2006) Financial System Stability Assessment. the participating banks were to represent a very large proportion of the banking sector and have the capacity to handle the macroeconomic scenarios in their internal systems. The calculations of the IMF and the banks showed approximately the same results, although the banks' calculations revealed a stronger decrease in the solvency ratio, cf. Tables 1 and 2. On the basis of the calculations, the IMF concluded that the banks can withstand sizeable shocks, but would experience considerable capital shortfalls if subjected to severe shocks. The mortgage-credit institutes were not included in the analysis, while the insurance companies were assessed using the DFSA's "traffic-light" test. As of the end of 2005, the test flagged only one insurance company as being under the "red light", which entails close monitoring by the DFSA. International standards After a detailed review of Danish legislation, procedures and the structure of the financial systems, the IMF's overall conclusion was that the financial sector in Denmark shows a high degree of compliance with the international standards. Nonetheless, the IMF identified scope for improvement of the systems and supervision in the individual areas banking supervision, insurance, payment systems and securities settlement systems, and anti-money laundering and combating the financing of terrorism. 1 The DFSA is responsible for banking supervision and insurance, while Danmarks 1 The relevant standards are: For banking supervision: Basel Committee's Core Principles for Effective Banking Supervision. For insurance: IAIS Insurance Core Principles. For payment systems and securities settlement systems: CPSS Core Principles for Systemically Important Payments Systems and CPSS-IOSCO Recommendations for Securities Settlement Systems. For anti-money laundering and combating the financing of terrorism: FATF 40+9 Recommendations (Anti Money Laundering and Combating Financing of Terrorism).

99 94 Monetary Review 4th Quarter 2006 Nationalbank holds the primary responsibility for payment systems and securities settlement systems. Furthermore, the DFSA holds the overall responsibility for anti-money laundering and combating the financing of terrorism together with the Public Prosecutor for Serious Economic Crime. As regards banking supervision and insurance, the IMF found that all principles were fully or broadly observed. Nonetheless, the IMF found that considering the challenges ahead, some further strengthening would be desirable, inter alia with respect to the autonomy and resources of the DFSA and its cross-border supervision. Furthermore, the IMF found that the DFSA should broaden the coverage of "fit-andproper" requirements from board members to all key staff in order to enable the DFSA to take action against unfit key staff members. The IMF also emphasised the need to strengthen the authorities' capacity for stress testing of the financial institutions. As regards payment systems and securities settlement systems, the IMF considered the systems to be secure and efficient, but with room for minor improvements. The IMF's review showed high compliance with the standards. The review covered Danmarks Nationalbank's large-value payment system (Kronos), the Danish Bankers Association's system for settlement of retail payments (Sumclearing) and VP Securities Services' securities settlement system. The IMF's recommendations for Kronos comprised further reduction of certain operational risks and further documentation of Kronos' efficiency. As regards securities settlement, the IMF among other things called for consideration of the benefits and drawbacks of introducing a central counterparty in the settlement of transactions. The IMF's recommendations for the Sumclearing included restructuring of certain elements of the settlement and further strengthening of oversight. Danmarks Nationalbank is following up on the IMF's recommendations. In the area of anti-money laundering and combating the financing of terrorism, the IMF found that Denmark has a solid framework for antimoney laundering and combating the financing of terrorism. However, in certain areas the assessment is that Denmark does not fully comply with the international standards. This should be viewed against the background of the entry into force of a new, very extensive Act on 1 March 2006, i.e. during the IMF visit. The IMF was thus not able to assess whether this Act would be implemented effectively. Furthermore, the Act does not cover the Faroe Islands and Greenland, which influenced some of the IMF's assessments. With respect to all the standards reviewed, it is up to the relevant Danish authorities to decide on the response to the IMF's specific recom-

100 95 STATEMENT BY THE DANISH AUTHORITIES TO THE IMF'S EXECUTIVE BOARD Box 3 The FSAP exercise confirms that the Danish financial system appears resilient and is well supervised, underpinned by an effective legal and financial infrastructure. There is good coordination between the various domestic institutions responsible for financial sector stability. The Danish government will thoroughly consider the recommendation concerning the institutional set up of the Danish Financial Supervisory Authority. However, decisions concerning the budget of an agency such as the DFSA are a matter for the Parliament as part of the annual revision of the Danish National Budget. The DFSA is allocated increased resources when given new assignments. On the basis of analyses and stress tests, Danmarks Nationalbank s most recent Financial Stability Report concluded in line with the Fund that the financial institutions are still robust and that there are currently no black clouds on the horizon that could threaten financial stability in Denmark. Stress tests will in the future be given higher priority in the DNB s assessment of financial stability. It should also be noted that the DFSA is developing the use of stress tests for monitoring the soundness of financial institutions. The new Basel II capital requirements could, for institutions which apply for the use of the internal rating based approaches for credit risk and advanced measurement approach for operational risk, imply potentially large decreases in the minimum capital requirements. To ensure that any large potential decrease in capital requirement will be gradual, the Danish regulation has included floors that set the maximum decrease in capital requirements for institutions using the advanced methods in 2007, 2008 and 2009, in accordance with the EU Capital Requirement Directive. Furthermore the DFSA will have the possibility of setting additional capital requirements in individual cases. The DFSA will consider issuing executive orders instead of guidelines. The benefits of such executive orders will have to be viewed against the increased costs that these may imply for the supervised entities. However, there are definitely supervisory benefits in upgrading the guidelines to legislation. Specifically, regarding the recommendation to extend the field of the fit-and-proper regulation to all key staff, and not just board members, our authorities generally concur that it could be useful if the DFSA had the possibility to take action towards unfit staff members. However, it also seems an unnecessary burden on the financial institutions if such key staff members should have a fit-and-proper approval in advance. Such measures would also claim a considerable part of resources from the DFSA, and it is in the opinion of the DFSA doubtful whether the benefits will exceed the costs. It must be stressed that Denmark is at an advanced stage in the implementation of the 3rd Money Laundering Directive. The implementing act has been adopted by the Parliament. Parts of the act have already entered into force. The remaining parts enter into force on January 1, The necessary implementing measures will enter into force on this date as well. The examination has thrown light on some shortcomings and the examination has, therefore, been very helpful for the implementation and the work related to formulating the administrative provisions. The FSAP confirms that the Danish payment and securities settlement systems are safe and efficient. The DNB and the DFSA will thoroughly analyze the recommendations for improvements.

101 96 Monetary Review 4th Quarter 2006 mendations. The Danish authorities have generally been very responsive to the IMF's recommendations, and in many areas the recommendations are being implemented in order to improve the already well-functioning financial systems. The authorities' responses to the IMF's recommendations are summarised in Box 3. 1 Just as FSAP was concluded with an Article IV consultation, the IMF is expected to follow up on Denmark's response to the recommendations in future Article IV consultations. More detailed updates of FSAP and the international standards typically take place after five years or more. However, the IMF and the individual member country can agree on a review of specific standards outside FSAP. In many countries the IMF has reviewed the standards concerning e.g. statistics 2 or fiscal transparency 3 outside the framework of FSAP. IMPACT OF FSAP ON FINANCIAL STABILITY The Independent Evaluation Office (IEO), established by the IMF, has examined whether the IMF's extended financial surveillance has had any impact. In addition to a thorough review of the IMF's reports and procedures, the IEO has, among other things, conducted a questionnaire survey in countries that have been subject to FSAP. 45 countries replied, and the IEO furthermore conducted in-depth reviews of 25 countries. 4 5 The IEO's surveys show that FSAP promotes internal dialogue in the individual countries' administrations, and that many countries regard FSAP as instrumental in raising some "taboo" issues. Virtually all the countries consider it vital to the value of FSAP that the assessment of the financial sector is independent and external. The IEO finds that the impact of FSAP has been greatest in countries whose governments were already committed to reform of the financial sector, either for internal reasons or reflecting strong external considerations, such as EU membership. Approximately 50 per cent of the countries find that FSAP has strengthened their strategic view on development of the financial sector The authorities' responses to the recommendations are reproduced in the IMF report Financial System Stability Assessment (2006). The responses are summarised in the authorities' statement to the IMF's Executive Board attached to the IMF report Denmark: Article IV Consultation-Staff Report (2006). The reports are available at Danmarks Nationalbank's website Compilation and publication of data (the Fund's Special Data Dissemination Standard/General Data Dissemination System). Preparation of the public sector's budgets and accounting practices and ensuring good administration practice in the public sector (the Fund's Code of Good Practices on Fiscal Transparency). Report on the Evaluation of the Financial Sector Assessment Program (FSAP) (2006) is available at the IEO's website Brazil, Bulgaria, Cameroon, Chile, Costa Rica, Dominican Republic, Egypt, Germany, Ghana, India, Ireland, Japan, Jordan, Kazakhstan, Korea, Mexico, New Zealand, the Philippines, Romania, Russia, Singapore, Slovenia, South Africa, Sri Lanka and Tunisia.

102 97 QUESTIONNAIRE SURVEY ON THE IMPACT OF FSAP Chart 2 FSAP has contributed to prioritisation of financial reforms Per cent Per cent "Strongly disagree" or "Undecided" "Strongly agree" "Don't know" or "Agree" "Disagree" Per cent FSAP has contributed to changes in the financial sector "Strongly disagree" or "Disagree" "Undecided" "Strongly agree" or "Agree" "Don't know" Per cent Source: Independent Evaluation Office, IEO. Both the IEO and the respondent countries find that the quality of FSAP is very high. This is also Denmark's experience. Just under half of the 45 countries even find that FSAP has provided new analytical insight into the financial systems, and areas such as stress testing and review of international standards are considered particularly important. Changes in key policies and institutions cannot necessarily be attributed directly to FSAP, but many countries emphasised two particular contributions from FSAP. Firstly, FSAP was found to entail a new approach to risk assessment, and many governments began to publish financial stability reports or develop stress tests after FSAP. Secondly, quite a few countries stated that the credibility of planned reforms was enhanced as a result of the independent expert advice. Approximately 70 per cent of the countries found that FSAP had influenced the prioritisation of reforms, and just as many believed that FSAP contributed to changes in the financial sector, cf. Chart 2. FSAP is widely supported since approximately 90 per cent of the countries reply that they implemented some or all of the IMF's recommendations. This high degree of implementation is by and large confirmed by the IMF's FSAP team leaders. Furthermore, FSAP appears to answer the purpose since almost 40 per cent of the countries believe that vulnerabilities in the financial sector have actually been reduced since the completion of FSAP. The countries' positive responses are probably part of the reason why approximately three quarters of the IMF's 184 member countries have volunteered for an FSAP review. Nearly all the European countries have participated in FSAP, cf. Chart 3. The IMF also seeks to apply some pressure to the member countries to encourage them to sign up for an FSAP, even though FSAP is resourceconsuming and the IMF can conduct FSAP reviews per year at the most. In this connection, the IMF gives priority to countries of great sys-

103 98 Monetary Review 4th Quarter 2006 GLOBAL OVERVIEW Chart 3 Americas Europe and Africa Asia/Pacific FSAP has been completed FSAP is underway No decision to conduct FSAP Note: FSAP for Argentina and the Ivory Coast have been discontinued. Source: IMF.