52003DC0729

COMMISSION COMMUNICATION - The EU Economy: 2003 Review - Summary and main conclusions -

TABLE OF CONTENTS

Introduction

1. Macroeconomic developments in the euro area

2. Drivers of productivity growth: an economy-wide and industry-level perspective

3. Education, training and growth

4. Wage flexibility and wage interdependencies in EMU

5. Determinants of international capital flows

Introduction

Lesson from 2003: More vigour needed to pursue the economic reforms agreed in the Lisbon strategy and the Broad Economic Policy Guidelines. // The economic performance of the EU economy in 2003 has underlined the need to pursue the Lisbon strategy with more vigour. The recovery that started in 2002 proved short-lived and did not initiate the dynamics necessary to bring economic activity back to potential. Although economic growth failed to rebound, employment withstood the slowdown better than in the early 1990's, suggesting a stronger resilience in the labour market after reforms implemented in the second half of the 1990s. However, employment growth has stalled in 2003 and the rate of unemployment has slightly risen. Moreover, public finances deteriorated. Investment has been a major drag on economic activity and was held back by the required adjustment in corporate balance-sheets and depressed profit margins. At the same time, the euro appreciation weighed on exports, while the sluggish decline in inflation did not stimulate private consumption. These developments have urged policy makers to intensify efforts to design and implement structural reforms in line with the Lisbon targets, the Broad Economic Policy Guidelines and the Employment Guidelines.

Reason for EU's disappointing economic record can be mainly found in domestic conditions. // The reasons for the subdued economic record are mainly to be found in domestic conditions. True, a series of global economic shocks have initiated the slowdown from 2000 onwards and 2003 was not free from further disturbances at the global scale. Oil prices were high and volatile, the global conflicts added to economic uncertainty and world trade did not rebound to former strength. Growth nevertheless picked up in some economic regions, most prominently in the USA and Japan. Among the possible domestic reasons for the European Union's tepid economic performance, structural rigidities figure prominently. Despite progress in recent years, activity rates and labour force utilisation are still too low. Key macroeconomic price variables such as real unit labour costs and consumer price inflation, adjusted only sluggishly to weak growth and deteriorating labour market conditions.

The EU Economy 2003 Review provides analytical support to key issues of the EU economic policy agenda. // The 2003 edition of the EU Economy Review analyses four specific topics that have been chosen for this year in the context of current economic policy challenges. Two chapters elaborate on key determinants of economic growth. The review starts with a chapter on recent macroeconomic and policy developments in the euro area and provides an in-depth discussion of possible reasons behind slow growth in the euro area. Chapter 2 deals with the drivers of productivity growth and analyses this from both an economy-wide and a sectoral perspective. It tries to identify the reason behind the gap between accelerating labour productivity growth in the USA and decelerating labour productivity growth in the EU. Despite widespread attention in policy circles devoted to education and human capital in the recent past, little is known about the contribution of education to economic growth. Chapter 3 provides a detailed analysis of education and growth. The experience from the early years of EMU as regards wage flexibility and wage interdependencies are analysed in Chapter 4. Finally, Chapter 5 deals with important aspects of the process of international capital flows.

// 1. Macroeconomic developments in the euro area

The euro area is to record economic growth below potential for three years in a row, being off course from 2001 to 2003. // Despite signs of a pick-up in economic activity in the second half of 2003, the euro area is set to record economic growth significantly below potential for the third year in a row. Sluggish economic activity can be associated with two main factors at work in 2003. Firstly, global economic uncertainty persisted throughout the spring of 2003. The Iraq conflict dominated headlines, stock markets nose-dived and the euro continued to appreciate rapidly, especially against the US dollar. These events hit an economy that was already coping with the aftermath of past major shocks. Secondly, there is some evidence that adjustment to these past economic disturbances has been more anaemic than analysts and forecasters had assumed. Market forces that usually initiate recovery seem to have worked less efficiently or strongly, implying that the economy, which recovered in early 2002, was not resilient to further adverse events.

Prolonged period of slow growth rather than a sharp fall in growth. // In a broader perspective, 2001-03 can best be described as a period of sustained growth slowdown rather than mild recession. A comparison of the last three major downturns in the region that now forms the euro area shows that they all started from a similar level of a positive output gap between 2 and 2 ½ per cent. The current change of the output gap is broadly comparable to that observed in the early 1980s and early 1990s. In international comparison, the deterioration of the euro-area output gap has not been particularly large. Moreover, the cross-country perspective points to a consistent relationship between the size of the output gap in 2000 and its subsequent deterioration. Those countries that witnessed the strongest deterioration in the output gap 2002-2003 were also those where actual GDP was higher than potential in 2000, and vice versa. This suggests that the recent slowdown should not be analysed in isolation but with reference to the events during the previous boom period.

Market adjustment was sluggish, suggesting an economy not resilient to shocks. // The fact that the slowdown has persisted for three years suggests that supply-side dynamics has been important and the growth weakness cannot solely be attributed to demand shocks. Against the background of both receding inflation and a considerable weakening of labour productivity growth at the early stage of the slowdown, steady nominal wage growth contributed to a marked increase in nominal unit labour costs. Both employment and private consumption growth decelerated broadly in line with the weakening of overall economic activity. Compared to historical experience, the fact that employment growth remained slightly positive despite a considerable weakening of economic activity is indicative of improved labour market resilience, and reflects a different path in job creation and destruction than in previous slowdowns as a result of labour market deregulation measures implemented in several Member States. Finally, exchange rate movements had a pro- rather than a counter-cyclical effect. During the previous period of strong growth in 1999-2000, the weakening euro increased price/cost competitiveness while the strengthening euro did not support export demand when economic growth slowed in 2002-2003. Interest rates declined in accordance with the slowdown in economic activity. Nominal interest rates have not been so low for some 50 years and the real long-term interest rates have not been as low as they are now since the late 1970s. Nevertheless, investment activity remained particularly weak, reflecting the importance of macroeconomic factors such as weak demand prospects, a worsening of profit margins and a low degree of capacity utilisation but also the increase of risk aversion and high debt in the corporate sector despite the ongoing correction of corporate balance sheets.

Corporate adjustment to slow growth yielded depressed profit margins and a pronounced weakness of investment. // The perception of risk seems to have fundamentally changed due to economic (slowdown in growth), financial (bursting of the stock market bubble) and political factors (terrorism). All these factors raised corporate capital costs. In a nutshell, the typical euro-area company adjusted to the erosion of revenues by trimming down capital costs whereas the US company reduced both capital and labour costs. The effect was a profound weakening in the growth of labour productivity in the euro area, which translated into depressed profit margins. Investment was cut considerably on both sides of the Atlantic. The main difference was that, whilst almost all the adjustment in investment in the USA took place in the years 2001 and 2002, in the euro area weak investment performance lasted until 2003.

Forces of recovery are well intact. // Optimism as regards the outlook for the euro-area economy was and still is based on significant structural improvements in the euro area that imply a clear break with past patterns. Four positive factors stand out: (1) a stability-oriented macroeconomic policy framework; (2) growing resolve to tackle structural reforms; (3) continuously moderate wage growth; and (4) technological advances providing scope for improvements in labour productivity growth.

Monetary policy has been accommodative. // Monetary policy had to act against the background of only slowly receding rates of headline and core inflation. While most of the increase in headline inflation in 2001 was related to one-off effects (oil price hikes and food price hikes linked to bad weather and BSE), there was a substantial risk of second-round effects. Despite this, the ECB cut interest rates from May 2001 onwards by a cumulative 275 basis points. A positive lesson from the recent experience is that the monetary policy stance has been accompanied with continuously low and stable inflation expectations. Forward interest rates suggest that financial market participants seem to consider that neither the strong growth in monetary aggregates nor the currently low level of money market rates represents a threat to price stability in the short to medium-term.

Budgetary policy: easing did not stimulate economic activity. // In terms of both actual budgetary developments and as regards the implementation of the EU framework for fiscal surveillance the past few years have been a difficult period. The play of automatic stabilisers in the context of the slowdown implied a considerable worsening of government finances. But the increase in the nominal deficit for the euro area as a whole reflects also discretionary loosening by some Member States. Available evidence suggests that the impact of the tax cuts, which were enacted in several EU Member States (D, F, I, NL, and A) since 2001, did not yield the hoped-for increase in private consumption and investment. The less energetic pursuit of budgetary consolidation may, also in view of the growing awareness of the need to reform pension systems, have dented private consumption through negative confidence effects. Moreover, worsening public finances may have prevented any further lowering of interest rates.

Budgetary consolidation needs to resume to tackle on time the looming budgetary implications of ageing ... // The deterioration of public finances witnessed since 2000, particularly in some Member States, has cast doubts on the commitment of several euro-area countries to achieve sound public finances over the coming years. This unfortunate development has been clearly marked by a breach by some Member States of the EU's fiscal rules. In responding to this it is important that fiscal authorities do not settle for short-term solutions that undermine the EU fiscal framework and the need to pay adequate attention to sustainability issues. Indeed, the increased focus on the quality of public finances has highlighted that about half of the Member States face a serious problem of sustainability. Achieving sound public finances is an important prong in the strategy to tackle on time the looming budgetary implications of ageing. Member States should demonstrate a clear willingness to pursue the medium-term strategy that in some cases has already delivered periods of high and sustained growth.

... supplemented by further progress with encouraging labour market participation and economic growth. // Moreover, encouraging labour market participation and economic growth will be key to alleviating the problem of ageing populations. For example, enhanced efforts to help parents combine work and family life, which Member States are committed to undertake, may contribute to raise employment rates. In order to bolster and speed up implementation of the Lisbon strategy, the European Initiative for Growth seeks to mobilise investment in areas that will reinforce structural reforms, stimulate growth and create jobs. It targets public and private investment in networks and knowledge.

// 2. Drivers of productivity growth: an economy-wide and industry-level perspective

A new growth pattern has emerged in the USA and a small number of the EU's Member States since the mid-1990s. // A new growth pattern has emerged in the US and a small number of the EU's Member States since the mid-1990s. For the first time since World War II, the EU is now on a lower trend productivity growth path than the USA. Over the 1996-2002 period, the EU has proved incapable of reversing the long-run decline in its productivity growth performance whereas the USA has enjoyed a notable recovery in its secular trend.

Deterioration in EU productivity growth is due to inadequate investment and innovation. // The 1 percentage point decline in EU labour productivity growth experienced over the 1990s emanates from two factors. Half of the decline can be attributed to a reduction in the contribution from capital deepening. Within this category, whilst investment in information and communication technologies (ICT) contributed positively (but not as much as in the USA), the rest of investment performed poorly. The remaining half emanates from deterioration in total factor productivity (TFP). This should probably be seen as the greatest source of concern for policy makers. Improvements in TFP are generally attributed to a more efficient resource utilisation emanating from enhanced market efficiency; from technological progress resulting from investments in human capital, R&D and information technology; or from the natural catching-up process of the less developed EU countries through increased business investment in general.

Economic growth in the EU in the 1990s is characterised by more labour input and less productivity. // In terms of GDP growth, the EU and the USA experienced significant breaks in the 1990s not only in terms of labour productivity but also with regard to labour input. The EU in fact achieved a sharp increase in its contribution from labour which, as mentioned above, was accompanied by equally sharp reductions in the contribution from productivity. The opposite pattern emerged in the USA. These divergent labour input and labour productivity trends are clearly linked. Up to one quarter of the 1 percentage point slowdown in EU productivity growth can be attributed to the higher employment content of growth. No policy trade-off should, however, be implied since boosting employment rates through bringing low-skilled workers into employment only leads to a temporary reduction in measured productivity growth, with no effect on the long-run productivity growth of the existing workforce.

Several Member States have outperformed the USA in terms of labour productivity growth. // A much more nuanced picture emerges at the individual EU Member State level. As regards labour productivity growth, seven EU Member States (Belgium, Greece, Ireland, Austria, Portugal, Finland and Sweden) performed well above the EU productivity average and even above that of the USA. Three of the seven, namely Ireland, Finland and Sweden were also capable of combining both strong productivity and high labour utilisation rates. The aggregate EU productivity gap therefore reflects the particularly poor performances of a number of the larger Member States, most notably Italy.

The industry-level analysis shows that superior US performance is concentrated in four ICT-producing and ICT-using industries. // The industry-level analysis shows that the superior performance of the USA in ICT-producing manufacturing and intensive ICT-using service industries is the principal source of the diverging productivity trends in favour of the USA. Whilst productivity in ICT-producing manufacturing industries has been growing at a significantly faster pace than in the associated ICT-using service industries, the latter account for by far the greatest proportion of the USA's upsurge in productivity. Labour productivity growth seems to be dominated by just five, out of a total of fifty-six, industries. All of these are among the ICT-producing and intensive ICT-using areas of the respective economies. The USA outperforms the EU in four of these five, namely in one ICT-producing manufacturing industry (i.e. semiconductors and other electronic equipment) and in three intensive ICT-using service industries (i.e. wholesale trade; retail trade; and financial services). On a more encouraging note, the EU is dominant in one ICT-producing service industry, namely telecommunications.

But with ICT contributing also positively to EU productivity growth, the slowdown has occurred in the non-ICT part of the economy. // The industry analysis also re-affirms that ICT is only part of the story behind the rising US and declining EU labour productivity trends. Just like in the USA, ICT also contributes to both capital deepening and TFP in the EU (although the extent of the gains in the USA is larger). The origin of the deterioration in EU productivity over the 1990s stems therefore from developments in the non-ICT, more traditional, group of industries, including services. Indeed, data reveal that both capital intensity and overall efficiency patterns in these sectors appear to be deteriorating. Accounting still for nearly seventy percent of total EU output, these developments are particularly worrisome. In addition, these are the parts of an enlarged EU economy which are facing the greatest competitive challenges from globalisation.

Productivity growth differentials appear to be related to some fundamental structural differences at the individual Member State level. // The key policy question addressed is whether the EU countries that experienced high productivity growth and the USA shared certain common characteristics that could explain their superior performance? More specifically, what were the channels via which the more fundamental factors driving growth (i.e. institutions, trade, market size, education and labour supply/demographics) affected investment and TFP in these countries, and how did these latter two factors interact to generate labour productivity growth? A model-based analysis shows that EU-US productivity differentials are indeed related to some fundamental structural differences at the individual country level, with five areas being identified as being quantitatively important and relevant in an EU context, namely the level of regulation, the structure of financial markets, the degree of product market integration, the size of knowledge investment and the ageing of the labour force.

"Lisbon strategy" simulation highlights the difficulties for the EU in becoming the most competitive, knowledge-based, economy in the world. // A "Lisbon Strategy" simulation, whilst explicitly concentrating on regulatory reform and the knowledge economy, implicitly highlights the importance of these five factors in determining the EU's long-run productivity growth rate and therewith for its ambitions to outperform the US in terms of potential growth. [1] In terms of boosting investment via regulatory reform, the "Lisbon Strategy" simulation showed that even a relatively rapid policy of deregulation towards equivalent US levels would not lead to sufficiently large productivity gains over the next seven years to close the present 10% efficiency gap with the US. Such a policy approach would appear to yield static efficiency gains rather than the dynamic efficiency benefits needed to achieve an outward shift of the "technology frontier". This suggests that deregulation alone, whilst crucial for investment, would be insufficient to meet the strategic Lisbon goal. It must therefore be accompanied by concerted efforts aimed at boosting the production of knowledge.

[1] The Lisbon simulation captures two supply-side initiatives linked to the Lisbon strategy, namely (1) a reduction of the level of regulation in the EU to the US level and (2) higher spending on third level education, software and R&D. It suggests that the effect of implementing such a large package of reforms would be to significantly boost EU potential growth rates, on average by ½ to ¾ percentage point annually over a 5-10 year horizon.

Productivity gains from R&D and human capital investments ... // Regarding knowledge production, long run productivity gains seem to stem above all from investments in both education and R&D. Regarding education, investment that fosters higher educational attainment can be expected to yield productivity gains as explained in the next section. Regarding R&D, the focus should be on creating the framework conditions that would promote an increase in total investment in R&D. These conditions include a higher degree of product market integration (e.g. through completion of the internal market) and an investment environment which ensures the development of a more active risk capital market. The reforms in this direction would undoubtedly improve the EU's economic fortunes, even if taken by themselves they would still not allow the EU to overtake the US in productivity terms over the timescale laid out by the Lisbon agenda.

... will be partially offset by the parallel efforts to boost employment growth and from the effects of the EU's ageing labour force // Apart from the time it takes for the reforms to yield visible effects, two further obstacles need to be overcome to reach the productivity target put forward in Lisbon. Firstly the temporary efficiency trade-off faced in attaining the parallel employment target of 70% and secondly the continuous drag on productivity induced by Europe's ageing labour force.

The productivity analysis supports the conclusions of the 2003 Spring Report. // Realising the difficulties of measuring progress in structural reforms, the Commission and the Council devised a set of structural indicators which have become one of the main tools for assessing progress in achieving the Lisbon objectives. This year, the Spring Report presented a simple, but very informative, exercise counting the frequency with which each Member State was amongst the three best or three worst performing Member States in the EU on each indicator. Certain countries appeared again and again amongst the top three Member States, most notably Denmark, Sweden and Finland. It is important to note that these are precisely the same countries that had already undertaken deep and successful reforms well before the launch of the Lisbon strategy. On the other hand, the largest Member States, such as Germany, France and Italy, came out as clear laggards with respect to structural reforms. The strong productivity growth performances of a small number of Member States vindicate the policy framework established by the Lisbon strategy. To bear fruit, however, the strategy has to be backed up by commitment and the timely and thorough implementation of the different reform measures.

Reversing declining labour productivity trends depends ultimately on the policy choices made by governments in the five areas highlighted in the analysis. // Finally, whether recent EU productivity trends are likely to be permanent or transitory will depend on the policy choices governments make. The analysis confirms the importance to the EU's long-run productivity performance of a forceful implementation of a comprehensive reform strategy. It should aim at reducing the regulatory burden, further integrating markets, promoting human capital investment and enhancing the innovation potential of the economy. Implementation of such a wide-ranging reform agenda would create a more flexible, dynamic and investment-friendly business environment. Together with better functioning markets, and more risk-oriented financing mechanisms, this will set the conditions for a significant increase in the EU's underlying labour productivity growth rate.

// 3. Education, training and growth

Investment in education is a powerful influence on economic growth ... // Rising educational attainment has been a major influence on economic growth. Attainment can be defined as the successful completion of a given level of education, such as lower-secondary school or an undergraduate degree. Given the difficulties in comparing education systems in different countries, the number of years of study required to obtain a given qualification is usually used as a proxy. Several recent studies, based on improved attainment data, suggest that an extra year of average attainment in the 25-64 population could raise productivity by as much as 4 to 6 per cent. In the EU, average attainment has grown by about 0.8 years per decade since 1960. This means that education might have accounted for as much as 0.3 to 0.5 percentage points of annual GDP growth. Further possible benefits might result if education indirectly promotes technical progress in the longer term. Whether this continues to be the case in the future depends on many unknowns, not least the nature of technical change and the consequent demand for skills. Nevertheless, educational attainment in the EU as a whole is set to continue increasing in the medium term at a similar pace to that of recent decades. Thus, a similar contribution to growth might be expected, though this will vary considerably among Member States.

... and yields long-term benefits. // The full productivity benefits of investment in young people's education accrue over the whole professional life. Three quarters or more of the increase in average attainment over the next decade will result from investments already made, in some cases many years ago, as older workers retire and are replaced by younger and better-educated cohorts. In comparison, investments made today will have a relatively small impact on average attainment over the next decade. Nevertheless, for the benefits of education to be reaped throughout the working life of an individual, knowledge and skills must be maintained and updated. Indeed, education should be interpreted in the broadest sense of lifelong learning, from pre-school and basic education to adult education and training in the workplace. The impact of education on growth is expected to be highest in countries where enrolment in secondary and tertiary education has risen most rapidly over the past 30-40 years, and lowest in countries where enrolment was already high and has grown less rapidly. There is some evidence of high returns to education particularly in the case of people who would otherwise enter the labour market with low levels of attainment. Since initial education leads to further training opportunities, inequalities in attainment tend to widen over time. Those with few qualifications are faced with a higher risk of unemployment and the need for later and more costly attempts to improve employability.

The quality of education is as important as the number of years spent in education ... // The economic evidence suggests that the quality of educational outcomes - measured by scores in internationally comparable tests - may be at least as important as the number of years spent in school or college. In fact, when quality is taken into account, the estimated growth impact of the number of years of schooling tends to fall. A key question, then, is how quality can be improved. It is self-evident that adequate resources are necessary for a high-quality education system. On the other hand, the link between expenditure and outcomes across countries is weak at best, which suggests that resources are being used with varying efficiency. Improving teachers' incentives to deliver high-quality outcomes may be more of a priority than increasing spending in some countries. Where increased resources are available, decisions on how these are spent - for example, on books, computer equipment, smaller class sizes, higher salaries for staff and so forth - may have important implications for quality.

... and greater efficiency would encourage investment in education. // Greater efficiency in the use of resources would increase the rate of return to investment in education. At tertiary level, for example, high drop-out rates and studies that often last well beyond the standard duration are equivalent to years spent outside the labour market without tangible benefits in the form of higher attainment. At primary and lower-secondary levels, demographic developments mean that the number of pupils is falling. This should in principle free resources. But, in practice, expenditure per student has tended to grow faster than GDP in recent years. If this continues, the additional cost in a decade could comfortably exceed the cost of an ambitious programme to increase enrolment in pre-school, upper-secondary, tertiary and adult education. Reforms in other areas, such as labour markets, tax and benefit systems and retirement incentives, would also increase the returns to education, thus encouraging investment.

Additional public resources should be focused where social returns are highest compared to private returns. // The available evidence suggests that the social returns to an additional year of schooling (i.e. the benefits to the whole economy) are broadly comparable to the private returns (i.e. the benefits to the individuals concerned). But both private and social returns are likely to vary considerably between, and indeed within, specific areas of education and training. There may be a case for targeted increases in public investment where the social returns appear high enough, and where they exceed the perceived private returns (otherwise government would merely subsidise investments that might anyway be made, leaving other more deserving projects unfunded, or unnecessarily raising the tax burden). A good case might be made for broadening access to pre-school education or for increasing upper-secondary participation, especially since these investments have long-lasting benefits and may help to even out inequalities in access to education that tend to widen over time. Where private returns are high and apparent, policy-makers should question whether increased public funding is needed to meet their objectives. Potential external benefits in terms of longer-term technical progress might justify certain public investments, including aspects of tertiary education.

Adult education and training may offer the greatest scope for raising average attainment in the longer term, but policies to encourage it must be efficient. // Since upper-secondary and tertiary participation cannot grow unboundedly, adult education and training is likely to offer the greatest scope for increasing educational attainment in the long term. Of course, the duration of the benefits is shorter than for children and young adults. Nevertheless, theory suggests, with some empirical support, that there are significant failures in the market for training, leading to under-provision. In addition, lifelong learning could play a crucial role in maintaining and renewing human capital acquired earlier in life, something which is not taken fully into account in the basic 'returns to education' framework. Lifelong learning could also help older workers to remain longer in the labour market, thus extending the benefits of earlier investments in human capital. If policies could be designed to address market failures in an efficient way, the returns could be higher than those for traditional schooling. Experience suggests, however, that tax incentives, subsidies and co-financing schemes to encourage training will need to be designed and evaluated much more carefully than in the past. This would help to maximise incentives to undertake genuinely additional training, and to minimise deadweight losses, substitution effects and other inefficiencies that may otherwise quickly consume the potential benefits of such programmes.

// 4. Wage flexibility and wage interdependencies in EMU

Wages play a key role in macroeconomic adjustment in EMU. // Over recent years, a near consensus view has emerged on the roots of high and persistent unemployment in many Member States, including all the major economies of the euro area and, more generally, on the low employment rates. Broadly speaking, this view regards the poor labour market performance of the countries concerned as the result of the interaction of a series of adverse macroeconomic shocks with unfavourable labour market institutions, and also product market regulations that have significantly limited the capacity to adjust to changes in economic conditions. Obviously, wages as the price of labour have a key role to play in determining the overall balance of supply and demand in the labour market. Furthermore, the formation of economic and monetary union (EMU) is often taken to put further demands on the flexibility of wages to compensate for lack of (national) instruments to deal with economic disturbances. If wages are too rigid, the necessary adjustment will come slowly and with considerable economic and social costs.

The downturn has exposed both the strength and the limits of wage setting mechanisms in the euro area. // Both common macroeconomic shocks and country-specific developments have put the flexibility of wage setting mechanisms in the euro area to a stress test in recent years. It was expected that nominal wage growth would remain consistent with price stability and productivity gains, thereby allowing companies to increase job-creating investment. Regarding actual developments, on the positive side, overall wage discipline has been preserved and risks that the inflation overshoot would lead to extended second-round wage effects have been averted. On the negative side, with nominal wage growth rather invariant to the cyclical situation, the slowdown in labour productivity growth translated into significant increases of nominal unit labour costs in 2001 and 2002. Hence, wage flexibility appears to have provided little, if any, support to the expected cyclical recovery so far.

Wage moderation should be pursued if EMU is to continue to deliver strong job growth. // After a prolonged period of wage moderation, the fall of the share of wages in GDP came to halt at the turn of the decade and remained broadly stable throughout the downturn. There are indications that the wage share will decrease again when the economy gathers momentum in 2004. Moderate real wage increases, consistent with productivity gains and the need of restoring profitability where necessary, help to increase employment and to lower structural unemployment over the medium term, without necessarily compromising domestic demand in the economy. This assertion is backed up by both standard economic theory and by the factual experience of many euro-area countries, in particular in the second half of the 1990s. Hence, in light of still high structural unemployment, further wage moderation is necessary in the euro area. However, it should also be noted that aggregate real wage moderation is a fairly poor substitute for wage differentiation, when it comes to helping to price the low-skilled back into jobs. It needs therefore to be accompanied with specific measures targeted at raising employment among low-skilled workers.

Conventional wisdom has it that wage formation mechanisms in Europe are characterised by a high degree of rigidity and slow adjustment to shocks but evidence is still inconclusive. // Conventional wisdom has it that wage formation mechanisms in Europe are characterised by a high degree of rigidity and slow adjustment to shocks. A number of institutional features in the euro-area labour market could account for a lack of nominal as well as of real wage flexibility. Factors typically mentioned in this context include union power, coordination/centralisation of bargaining, bargaining coverage, the impact of collective bargaining on contract length, the use of wage rules in collective bargaining, including wage indexation, and, last but not least various insider-outsider mechanisms in the labour market affecting the sensitivity of wages with respect to unemployment. However, in line with findings from other studies, formal econometric analysis of Phillips curve-type wage equations suggests that wage inflation persistence in the euro area is not higher than in the USA. The finding of broadly similar degrees of nominal inertia across euro-area Member States, and in the euro area and the USA, makes it difficult to identify institutional labour market characteristics as the major determinants of nominal rigidities. Thus, while institutional and structural factors are a key to an understanding of what determines the level of equilibrium unemployment over the medium term, institutional labour market characteristics appear to be of less importance for the degree of nominal inertia in the economy.

EMU is affecting the wage bargaining system in several ways with potentially important implications for the adjustment to shocks in the euro area. // While it is still too early to draw final conclusions on potential channels through which EMU could impact on the incentives faced by its economic agents and on its wage bargaining systems, the picture is nevertheless becoming progressively clearer. Research has already identified a strong positive impact of the euro on product market integration via increased trade and foreign direct investment. This should lead to enhanced competition on product markets. The impact of EMU is somewhat less clear-cut in the case of wage interdependencies. The convergence of wages and unit labour costs has not waited for the single market, let alone EMU, to be completed. Available sectoral evidence suggests that convergence was in fact stronger in the 1980s than in the 1990s. The emergence of higher goods market integration and of stronger interdependencies in wage setting across countries - be it due to EMU or other factors - can affect the way in which shocks are absorbed and transmitted in EMU. Model simulations show that this partly depends on the nature of the shocks. Increased wage interdependency does not lead to major differences in the absorption of supply shocks but entails a more protracted adjustment to demand shocks. In the case of demand shocks, the wage and price response slows down if wage setting is interdependent, with simulations showing that it takes approximately one more year for the output adjustment process to work out than in the case without wage interdependencies.

// 5. Determinants of international capital flows

The creation of the internal market and the launch of EMU have fostered international capital flows. // The strong increase of international capital flows (portfolio flows and direct investments) over the past ten years is the combined result of legal and economic forces. As regards the EU, the full liberalisation of capital movements within the Community was finally accomplished on 1 July 1990 while capital movements between Member States and third countries were fully liberalised on 1 January 1994. The rapid expansion of domestic financial markets and surging international trade have been two of the main driving economic forces. In addition, the adoption of the euro and the resulting elimination of foreign exchange risk within the euro area have accelerated financial integration within the EU.

Increased international capital flows have strong implications for the global economy. // Enhanced financial integration has strong implications for the functioning of the global economy. International capital flows may serve both as a source of growth and as a transmitter of macroeconomic shocks. By smoothing consumption, capital flows play an important role in the adjustment to disturbances. Sudden shifts in the flow of foreign finance can, however, also create major domestic problems, as demonstrated by financial crises in several emerging economies in the past decade. Many emerging economies liberalised their capital flows in the 1990s, while maintaining weak financial institutions and pursuing macroeconomic and financial policies that turned out to be inconsistent with exchange rate stability. The outcome has been large financial imbalances driven by capital inflows and eventually financial crises and distress.

The need to finance high investment ratios without adequate national savings continues to lead to external deficits financed by FDI inflows in acceding countries. // Current account deficits are a common feature in the acceding countries. In several cases they amount to more than 5 percent of GDP, having increased over recent years in connection with rising foreign direct investment. Thus, the current account deficit in most cases is a reflection of large FDI inflows and not the main reason for the worsening of the external accounts. With the notable exception of Hungary, the external deficits are largely covered by non-debt-creating FDI inflows. In some acceding countries, privatisations are still under way. In others, second-round investment in the form of inter-company loans provide an important source of current account financing. On the whole, the acceding countries are likely to run considerable current account deficits for some time to come in order to compensate for their lack of domestic savings. Thus, foreign investments will continue to be a major motor of growth.

Adoption of EU acquis should contribute to financial stability in the acceding countries. // In the area of financial sector development and supervision, in particular, there are striking differences between acceding countries and many other emerging markets. Here the acceding countries have gradually implemented the EU acquis for regulation and supervision and have opened their markets to large-scale foreign ownership. This experience suggests that the acceding countries - by pursuing adequate policies - can avoid the negative experiences in other regions, thereby setting the pre-conditions for strong real convergence in a setting of financial stability.

Improving corporate governance systems should help the EU to attract capital flows. // Countries with good corporate governance systems are likely to attract international capital flows on better terms than countries with weak systems that invite fraudulent behaviour. With rising competition for capital inflows these issues are likely to become important determinants of capital flows in the coming years. The EU has already taken a number of steps to improve corporate governance in Europe, including the Financial Services Action Plan (FSAP) and the Market Abuse Directive. Work is also underway to strengthen accountancy standards, auditor independence and share-holders' rights. This will make the EU more attractive for growth-enhancing capital flows.