The Contribution of Monetary Policy to Economic Growth
Lucas Papademos, Vice-President of the European Central Bank, 31st Economics Conference, Vienna, 12 June 2003
1 Introduction 
I am delighted at the opportunity to speak at this year's Economics Conference organised by the Oesterreichische Nationalbank. This series of conferences has established a long tradition -more than 30 years old - of focusing on issues which are both important and topical. The theme of this year's conference is no exception. "Fostering Economic Growth in Europe" is the key economic policy priority considering the moderate average growth of the European economy over the past 20 years, its recent weak economic performance, and the expected modest economic recovery.
To foster economic growth in Europe requires both an accurate diagnosis of the factors determining or constraining its growth performance and an appropriate policy prescription regarding the macroeconomic policies and structural reforms needed to achieve higher and sustainable growth. Today, I will discuss the contribution of monetary policy to economic growth, an issue which has long been the subject of theoretical and policy debates among economists. At the present juncture, with weak growth and even talk of deflation risks, and given actual or perceived constraints on economic policies, this topic is attracting increasing interest, the debate has occasionally become rather heated, and there have been numerous calls from politicians and academics for monetary policy to pay more attention to growth. Against this background, I welcome this chance to add my own views to the ongoing discussion. I will do so by examining a number of fundamental issues concerning the role of monetary policy in fostering economic growth.
There are five questions that I will endeavour to answer: Firstly, can monetary policy contribute directly to the attainment of a high but sustainable rate of growth? Secondly, can monetary policy promote economic growth indirectly by maintaining an environment of price stability? Thirdly, can monetary policy effectively influence the pace of growth over the short and medium term, and thus help stabilise output fluctuations consistently with its overriding objective of price stability? Fourthly, how and to what extent has the growth performance of the euro area been influenced by the single monetary policy which has been implemented since the launch of the euro? Finally, what is the role of the other policies in fostering sustainable economic growth in Europe over the coming years?
2 Monetary Policy and Long-Term Economic Growth
In examining the effects of monetary policy on economic activity and growth, it is useful, both for conceptual and for policy reasons, to distinguish between long-term and short-term effects or, alternatively, between permanent and transitory effects. I will begin by considering whether and how monetary policy may influence economic growth in the long run, reviewing first the theoretical arguments on the links between monetary expansion, inflation and economic growth, and then assessing the available empirical evidence.
2.1 Theoretical propositions
A key issue in monetary theory is whether changes in the stock of money or in the rate of growth of money can have lasting effects on real economic variables. In particular, the question concerning the so-called superneutrality of money - whether a permanent change in money growth has no long-term effects on the real interest rate, capital accumulation and output growth - has been the subject of extensive theoretical analysis since the early 1960s.
In a seminal contribution, James Tobin (1965) showed that in a simple model with agents saving for future consumption only out of current income, by either holding money balances or investing in real capital assets, an increase in monetary expansion can lead to higher growth. Thus, Tobin's analysis refuted the superneutrality of money by relying on a fairly straightforward mechanism related to the role of money as an asset and a store of wealth. An increase in money growth leads to a higher rate of inflation that reduces the own rate of return on money and induces a portfolio shift in favour of real capital. This generates an increase in the capital stock and a higher level of output per person in the long run.  In an earlier contribution, Robert Mundell (1963) had also emphasised a link between anticipated inflation and the real interest rate. His analysis, however, examined the short-term positive effect of a permanent increase in inflation on real saving and the demand for capital and not the long-term effects of inflation on the real rate of interest and economic growth.
Over the last forty years, the theories advanced regarding the relationship between money, inflation and growth have refined and extended Tobin's analysis in several ways. They have also challenged his finding that monetary expansion has a positive and lasting effect on growth. Theories on money and growth have become more sophisticated by deriving results from the optimisation of utility by economic agents who are treated either as "infinitely lived" or as belonging to "overlapping generations". Theories have become more complete by incorporating the other functions of money in the real economy. These generalisations, however, have not led to unambiguous and robust conclusions. For example, in models where economic agents are "infinitely lived" and under certain additional assumptions, monetary expansion cannot affect the real rate of interest and economic growth (the superneutrality of money is valid).  On the other hand, the alternative approach using "overlapping generations" models can provide a formal justification for the Tobin effect in an explicitly optimising framework.  The effects, however, of monetary expansion on economic growth under either of these two types of theoretical models, also depend on other underlying assumptions.
A key factor influencing the conclusions of the theoretical studies is the role of money in the real economy and how that role is incorporated in the models. If real money balances and capital perform complementary functions, and are not seen as substitutes as in the Tobin model, higher monetary growth and inflation reduce capital accumulation and the long-term rate of growth. Thus, in models in which agents employ their own money balances to finance consumption and investment, and therefore there is a "cash-in-advance" constraint on spending,  or in models in which money is treated as a factor of production in its own right,  or when the services provided by money holdings affect the resource constraint facing economic agents  (rather than affecting directly the utility or production functions), higher inflation usually leads to lower output per person and output growth in the long run. Hence, different hypotheses about the functions of money imply conflicting conclusions about the size and sign of the permanent effect of monetary expansion on growth. Moreover, the results derived from alternative theories in some cases are not robust with respect to small variations in other underlying hypotheses concerning the preferences of economic agents.
Thus far, I have focused on the relationships between money, inflation and growth derived from traditional growth models in which the rate of technological progress is the fundamental determinant of long-term growth. The more recent "endogenous growth" theories allow for the determination of the long-term growth rate endogenously, for instance by human capital or investment in R&D. A few attempts have been made to include and analyse the effects of money within such a more realistic framework. It has been found that because higher inflation lowers the return on work, it leads to a temporary decline in the supply of labour. Since human capital is thought to benefit from a "learning by doing" effect, this decline in labour supply reduces human capital and thereby lowers the growth rate of the economy (Gomme (1993)). In contrast, it has been shown by others (e.g. Ho (1996)) that higher inflation can increase the capital stock (via a Tobin effect), thereby raising the long-term growth rate.
One strand of this endogenous growth literature has examined the effects of inflation on investment in the presence of nominal rigidities in the tax system (e.g. Jones and Manuelli (1995)). In particular, with nominally specified depreciation allowances, a permanent change in money growth can alter the effective real marginal tax rate on investment income, thereby changing the after tax real rate of return. In such a set-up, a permanently faster monetary expansion causing higher inflation leads to lower capital accumulation and output growth. If, however, the government's budget constraint is taken into account, the additional revenues resulting from higher inflation (including extra seignorage revenue) imply that "ordinary" taxes can be reduced, thereby increasing the net return on human capital and speeding up investment and growth. Thus, the overall impact of monetary expansion on growth depends on the relative magnitude of each of these effects. 
What conclusions can we draw from the theoretical literature on money, inflation and growth? In 1970, Jerome Stein surveyed the literature available at that time and noted that "my main conclusion is that equally plausible models yield fundamentally different results". Two decades later, Orphanides and Solow (1990) noted that "all we have is more reasons for reaching the same conclusion". A more recent survey (Haslag (1997)) also tends to come to the same conclusion. Although these views about the inconclusive nature of money and growth theories may be warranted when one reviews the whole spectrum of models in a neutral way (that is, without assessing the realism of underlying assumptions), my own reading and interpretation of the theoretical findings is less agnostic. The theoretical analyses which employ more general and realistic assumptions regarding (i) the role of money in the economy, (ii) the endogenous determination of factors shaping long-term growth, and (iii) the existence of nominal institutional rigidities in the economy, imply on the whole the existence of a negative association between monetary expansion and inflation, on the one hand, and economic growth, on the other. It should also be pointed out that the existence of a positive association between inflation and long-term growth derived from the models of Tobin and others must be confined to relatively low rates of inflation, otherwise we would reach the absurd conclusion that hyperinflation would drastically improve the real economy's performance.
The view that higher monetary expansion and inflation should adversely affect long-term growth is further supported by other theoretical analyses regarding the welfare costs of inflation (e.g. Fischer and Modigliani (1978), Issing (2001)) and the negative effects on output growth of the increased economic uncertainty induced by inflation (Lucas (1973, 2003)). The costs of inflation, including costs resulting from features of the economy's institutional structure, clearly imply a negative impact of inflation on growth. Moreover, the increased uncertainty due to high and variable inflation impairs the efficiency of the price mechanism and can be expected to reduce both the level of and the rate of increase in productivity and thus economic growth. Therefore, on the whole, theory implies that an expansionary monetary policy leading to permanently higher inflation will have, or is very likely to have, a negative effect on long- term growth, even for moderate rates of inflation. Moreover, this effect can be expected to increase nonlinearly as inflation rises. Nevertheless, the magnitude of the expected negative relation between inflation and growth cannot be determined a priori and has to be assessed on the basis of the available empirical evidence.
2.2 Empirical Evidence
What can the available evidence tell us about the link between monetary expansion and economic growth? A clear majority of studies find that inflation and long-term growth are systematically and negatively related. In other words, higher inflation tends to reduce growth in the long run.  The result is not unanimous, as some papers find no correlation between long- term growth and inflation.  There are very few empirical analyses that have identified a positive and stable long-term relationship between inflation and growth, but this relationship holds only for low rates of inflation.
Nevertheless, it should be recognised that research in this area has been hampered by data problems and difficulties in establishing reliable causal links between inflation and growth. The results of studies using data from just one country may be distorted by a few exceptional periods - such as the marked movements in energy prices during the 1970s. To overcome such possible distortions, researchers have often sought to utilise cross-country data, so that a variety of inflation and growth paths can be compared. These cross-sectional analyses, however, face other difficulties, for example how to adequately account for individual characteristics in different countries.
Another problem is the robustness of the empirical results. It is often found that slight variations in the specification of these regressions lead to substantially different results. Some studies find that, once other determinants of growth are included, a previously observed negative relationship between inflation and long-term growth disappears.  A key issue is how to estimate the trend growth of output. It is important to be able to identify changes in trend growth, yet it is equally important for the estimate of the trend not to be polluted by cyclical fluctuations.
As previously noted, the relationship between inflation and growth can be expected to depend on whether inflation is initially high or low. It is sometimes argued that the estimated negative correlation between inflation and growth is due to the inclusion of high inflation countries and that it is much harder to find such a negative relationship among countries with relatively low inflation.  Recent research, however, by Andrés and Hernando (1999), focusing on OECD countries, finds that even in low or moderate inflation countries, there is evidence of a robust negative relationship between inflation and output in the long run.
A number of studies have considered whether there are any non-linearities in the relationship between inflation and growth by examining the possibility that there are "threshold levels" in the relationship.  It has been found that the effect of an increase in inflation on growth may depend on whether inflation is abo ve or below some threshold level: while higher inflation above this level is associated with lower growth, this does not appear to be the case for inflation rates below the threshold.  Indeed, some studies (e. g. Ghosh and Phillips (1998)) suggest that for very low inflation rates, and within a very narrow range, inflation and growth may be positively correlated.
This last finding lends support to the view, which can be traced back to Vickrey (1955) and Tobin (1972), that "small doses" of inflation may be helpful for growth and employment, or that a little inflation is necessary to "grease the wheels of the economy". It has been argued that, because of downward rigidity in nominal wages, a certain amount of inflation is required in order to enable real wages to adjust to changing economic conditions. This argument does not rest on money illusion, but on the idea that workers will resist relative wage cuts and that, as a consequence, inflation provides a means of synchronising real wage reductions across the economy.
These propositions have been further developed by Akerlof et al. (1996, 2000) who have calculated that, in the face of downward nominal wage rigidity, an attempt to reduce inflation from 3% to zero would raise US equilibrium unemployment by 2.6 percentage points. Therefore, a permanently higher rate of unemployment can emerge at a very low rate of inflation. The evidence, however, for the existence of such rigidities is mixed and it may well be the case that they are removed or mitigated under low inflation.  Moreover, inflation in the presence of nominal rigidities can also "put sand in the gears" of the labour market.  The existence of "menu costs" and fixed nominal contracts implies that changes in the general price level may not be evenly transmitted throughout the economy and may therefore lead to unintended and disruptive changes in relative prices.
This debate is unresolved, but such arguments and evidence have been used to provide justification for allowing central banks' definitions of price stability to encompass low positive rates of inflation rather than literally aiming for a stable price level.  Nevertheless, I do not believe that the evidence about "greasing the wheels" of the economy is sufficiently convincing compared with the favourable effects of price stability. Even if some trade-offs have been found statistically to exist between inflation and output at very low rates of inflation, it is not at all clear that they are either stable - and would therefore persist during a prolonged period of price stability - or that they could successfully be exploited by policy makers.
The general conclusion that I would draw from this review of the money, inflation and growth literature is that the weight of evidence does not support the notion that monetary policy makers could sustainably raise growth by tolerating higher inflation. On the contrary, theoretical analyses (regarding the real effects and welfare costs of inflation) as well as the bulk of empirical evidence strongly suggest that price stability is conducive to long-term growth.
3 The Stabilising Role of Monetary Policy
In sum, monetary policy cannot be expected to directly contribute to raising long-term economic growth, though it can foster sustainable growth by maintaining an environment of price stability. It is often argued, however, that monetary policy can and should seek to stabilise output around its potential growth path in the short and medium run. As we all know, this has been one of the most widely debated issues of economics since Keynes (1936) made the case for stabilisation policies. And still today the debate regarding the stabilisation of output fluctuations remains very much active, as indicated by the Presidential Address of Robert Lucas (2003) at the recent annual meeting of the American Economic Association.
When considering this issue, it is essential to realise that the potential for stabilisation policies depends on both the size and the nature of cyclical fluctuations. We are all aware that economic cycles are caused by various factors and processes. They can be triggered and driven by shocks of various types as well as by changes in policies affecting demand and supply in product and financial markets. The magnitude and duration of economic cycles are also determined by technological processes, agents' behaviour and expectations, and institutional features of the economy. Policies influence the cycle not only by directly affecting aggregate demand and supply but also by shaping expectations and institutions. The stabilising effects of monetary policy depend crucially on the nature of the public's expectations.
Overall, theory and evidence support the view that it is possible for monetary policy to influence aggregate economic activity in the short and the medium term. This conclusion, however, does not mean that it is necessary or desirable for monetary policy to play a stabilising role. There are several reasons for being cautious in assigning such a role to monetary policy. Indeed, because there are indications that economic cycles have diminished in industrialised countries, there may be less of a need for an active stabilisation policy. A recent paper by Blanchard and Simon (2001) has pointed to a long-term decline in US output volatility, a phenomenon which can be traced back at least to the 1950s. They report that, with the notable exception of Japan, other G7 economies, including Germany, France and Italy, have also experienced a downward trend in output volatility. A number of explanations have been put forward for the apparent moderation of the economic cycle. These include the increasing relative importance of services in aggregate output, improvements in inventory management, and the stabilising effects of monetary policy.
While it appears that output volatility has declined in recent decades in many industrialised economies, the size and frequency of several types of shocks cannot be controlled. There are also reasons to believe that structural changes may have created new sources of instability that policy makers need to monitor very closely. In particular, the role of asset prices in the economic cycle has received a lot of attention recently. Financial markets have gained markedly in importance during the last decade. One implication of the growing size of stock markets is that changes in equity prices are likely to have a more pronounced impact on the economy than in the past. While the development of financial markets should, in principle, improve the allocation of resources, economists have long been aware that financial markets can be characterised by periods when asset prices tend to deviate significantly from their equilibrium values. Such situations can have implications for economic activity and can generate or accentuate output fluctuations.
An important reason why it may not in general be desirable for monetary policy to play an active stabilisation role is that there is evidence that a large - if not the largest - part of cyclical output variability can be attributed to real rather than nominal or demand shocks (Lucas (2003)). Such real shocks, which are driven by technology, cannot be effectively offset by monetary policy. In addition, under "normal circumstances", i. e. when the central bank is confronted with cyclical fluctuations of average magnitude, the systematic pursuit of an activist counter-cyclical policy can be ineffective and may increase rather than moderate output volatility. The risk of this happening emerges because of uncertainty about the magnitude and timing of the effects of monetary policy on output.  In addition, there are uncertainties associated with identifying the types of shocks and assessing the precise cyclical position of the economy.
These considerations lead me to conclude that the conduct of an activist, fine-tuning countercyclical monetary policy involves more risks than potential benefits and should be avoided under normal circumstances. Nevertheless, it is possible to envisage particular circumstances, triggered by severe shocks, when monetary policy can play a role in stabilising output around its potential growth path. Such a policy would have to be implemented carefully and consistently with the central bank's commitment to its primary objective of maintaining price stability. It should also be communicated effectively so that public expectations and the central bank's credibility would not be adversely affected. The precise nature of the policy reaction will, of course, depend on the nature of the shock as, for instance, the response to an adverse supply shock would be very different from the response to a demand shock.
The ECB's mandate and strategy are fully consistent with the theoretical arguments and the empirical evidence regarding the role and effectiveness of monetary policy in preserving price stability and fostering economic growth. A key element of the ECB's strategy is the commitment to maintain medium-term price stability, which is defined quantitatively. The announcement of a quantitative definition of price stability aims to anchor the public's inflation expectations. Another important feature of the ECB's strategy is that it is forward-looking, with a medium-term orientation, which reflects the long time lags in the effects of monetary policy on the price level. The strategy does not justify short-term activism and policies aimed at "fine- tuning" the economy. At the same time, the medium-term orientation of the strategy allows for a gradualist policy response to shocks to the price level and provides some scope and a degree of flexibility which may be needed to address various types of severe shocks. The combination of commitment and flexibility that characterises the ECB's strategy allows for some "constrained discretion" in dealing with cyclical output fluctuations in a way consistent with the preservation of price stability.
4 Economic Reforms to Increase Sustainable Growth in Europe
Two general conclusions emerge about the contribution of monetary policy to economic growth: first, monetary policy cannot be expected to raise growth sustainably in the long run; second, although monetary policy can play a stabilising role over the medium term, the scope of such a role may be limited by the pursuit of the primary objective of price stability, the nature of the monetary policy transmission mechanism, and by other factors, including the uncertainty facing policy makers and the stance of economic policies. These conclusions raise two relevant questions: How can we increase long-term economic growth in Europe and how can we speed up the recovery of the euro area economy towards its potential growth path?
Before addressing these questions, it is useful first to point to some facts concerning the growth performance of the euro area economy and the monetary policy stance. During the four years since the establishment of the Economic and Monetary Union (EMU), economic growth in the euro area averaged a modest 2.1%. It turns out that this is precisely the average growth recorded in the ten years, from 1989 to 1998, before the introduction of the euro. It is rather remarkable that over a longer period of almost 20 years, from 1981 to 1998, before the launch of the single currency, the growth rate in the countries that are currently members of EMU was also 2.1% on average, although during the same period average inflation was about 4.6%, i.e. substantially higher than its level after the adoption of the euro. It is thus evident that over a fairly long period of time, which was characterised by different monetary regimes and policies, the average growth performance of the 12 euro area countries was moderate and unchanged. Although we cannot reach any definite conclusions from these observations, they suggest that the average growth of the euro area countries mainly reflects the influence of non-monetary factors and policies.
The second set of facts relates to the monetary policy stance in the period after the introduction of the single currency. During this period, GDP growth in the euro area was negatively affected by several shocks and factors, some of which also had unfavourable effects on the price level. Although monetary policy had to be tightened for some time in order to respond to an unusual number of sizeable adverse shocks to price stability, the stance of monetary policy cannot be considered as the factor which constrained economic activity. Actually, it was accommodative at times. Overall, the ECB's monetary policy did attain a high degree of price stability in the euro area. The average inflation rate in the euro area between January 1999 and December 2002 was 2.1%, marginally abo ve the upper ceiling of the ECB's quantitative definition of price stability. This was achieved because the ECB responded to shocks to price stability in a determined and systematic way, consistently with its mandate and strategy. At the same time, the pattern of inflation and interest rates during this period reflects the medium-term orientation of monetary policy and it reveals that the ECB, in formulating its policy, did take into account its likely impact on the real economy, since policy did not aim to offset fully and promptly the effects of shocks to price stability.
The foregoing observations support a general proposition, which, I believe, is by now widely recognised. Although the preservation of price stability and the implementation of sound macroeconomic policies are necessary to foster sustained growth, structural reforms are essential - the main policy instrument - for increasing long-term growth in the euro area. More specifically, there are two complementary routes through which it is possible to raise trend economic growth. The first is to remove obstacles to the efficient utilisation of resources by improving the functioning of market mechanisms. For instance, by reforming labour market institutions, it should be possible to reduce structural unemployment. The resulting higher labour input should allow the level of output to increase, giving a temporary boost to economic growth until unemployment fell to a new, lower, sustainable level. The second route involves implementing policies that permanently raise economic growth. This requires knowledge of the factors that drive the growth process in developed economies. There is extensive literature that seeks to identify the variables influencing economic growth over the longer run. Physical and human capital accumulation, innovation, entrepreneurship, competition, the rule of law and the magnitude of government investment are important determinants. As mentioned earlier, some of the empirical studies on the link between inflation and growth can be criticised for their failure to take into account both inflation and other factors that may determine economic growth. A study by Levine and Renelt (1992) which sought to take into account such other factors, found a clear and robust association between average economic growth and the importance of investment and trade.
The policies required to increase long-term growth in the euro area will therefore fall largely within the remit of national governments. Such policies, however, are increasingly being co- ordinated at the European Union level. In March 2000, the Lisbon European Council recognised the importance of modernising the EU's regulatory framework and introduced an ambitious reform agenda aimed at making the European Union the "most competitive and dynamic knowledge-based economy in the world by 2010". The functioning of product markets is monitored and evaluated as part of what is known as the "Cardiff process". Similarly, labour market reform is assessed within the "Luxembourg process". Both processes rely on country examinations of reforms and provide input into the Broad Economic Policy Guidelines, which define the overarching economic policy priorities in various fields over the coming three years. Policy recommendations made to each EU Member State are based on these priorities.
In this year's Broad Economic Policy Guidelines (for the 2003-2005 period), several main priorities for policy action were identified.  In addition to reforming pension and health care systems, it was agreed that priority should be given to improving the functioning of the labour market in Europe and to implementing structural reforms to increase investment in human and physical capital. In particular, it was stressed that there is a need to:
(i) facilitate investment in research and development and in infrastructure;
(ii) lift barriers to the business application of technology and foster links between the public and private sectors in order to exploit research findings; (iii) enhance the role of small and medium- sized enterprises in research and development through their participation in integrated projects, and (iv) complete the internal market so as to help increase the competitiveness of industry, thereby fostering productivity and business dynamism. More specifically, it was emphasised that a fully integrated financial market would help to channel savings more efficiently into productive investment.
In order to improve the labour market situation in the European Union, a number of far-reaching structural reforms are required. The main specified priorities in this field are five:
(i) to take action, via reforms to tax and benefit systems, to make "work pay", so that people would not be discouraged from seeking work by the prospect of losing benefits and paying higher taxes;
(ii) to strike a balance between security and minimum standards for workers, so as to favour job creation and offer firms the flexibility they need to be able to respond to changing economic conditions;
(iii) to increase labour market participation, particularly among women, people who are over 50, as well as the low-skilled and the long-term unemployed;
(iv) to promote "life-long learning" and a constant upgrading of skills in order to generate
higher productivity and better jobs and
(v) to encourage closer cross-border co-operation in the setting of standards, so that qualifications and experience can be widely recognised and mobility can be facilitated. Although the list of priorities for policy action is long and ambitious, they are necessary and we should fully support them. I believe that if these priorities lead to concrete and substantial policy reforms, then we can reasonably expect to see a strengthening of Europe's growth potential and an improvement in the labour market situation.
The key issue is whether these reforms will be implemented in a timely and effective manner. Unfortunately, the implementation of previous years' Broad Economic Policy Guidelines has been somewhat patchy. For instance, following last year's Guidelines, there were a number of useful initiatives in the field of labour market reform. Several countries have started, or in some cases continued, to implement measures to make work financially worthwhile or reduce employers' social security contributions. In addition, there have been efforts to improve the job search process by increasing the efficiency of services provided by employment agencies and the adoption of stricter job search requirements. However, despite these positive developments, many labour market reforms have not yet been adequately introduced, including comprehensive reforms to pension systems and early retirement schemes aimed at increasing the labour force participation of older workers, and reforms to employment protection regulations aimed at improving job mobility. It is disappointing that the pace of labour market reform slowed down in most euro area countries in 2002.
Overall, there has been some progress in recent years towards addressing the structural weaknesses of the euro area. The approach adopted until now by many countries, however, seems to have taken the form of partial steps rather than comprehensive reform efforts. As it takes time for structural reforms to yield their full benefits, the slow and partial approach pursued in most Member States will make it increasingly difficult to achieve the strategic objectives set in the Lisbon agenda. Furthermore, a lack of determination to implement comprehensive reforms may also be a reason for the low level of confidence in a rapid and sustainable economic recovery. This point makes greater efforts in the field of structural reform all the more important and pressing.
I would like to conclude by stressing a few points relating to some of the questions that I posed at the start. The first question was whether and how monetary policy can contribute to higher sustainable economic growth. The weight of the theoretical arguments and empirical evidence I reviewed is consistent with the notion that the best contribution that monetary policy can make to sustainable growth is to maintain price stability. Because inflation is fundamentally a monetary phenomenon, monetary policy is the only tool that can effectively maintain price stability in the medium and long run. Therefore, it makes sense that price stability is its primary objective. In addition, most of the available empirical evidence and analysis shows that lower inflation is associated with higher long-term growth. While there may be doubts about the robustness of some of these results, there is little empirical support for the view that monetary policy should abandon the pursuit of price stability in order to increase long-term economic growth.
The second question was whether monetary policy can affect the pace of growth in the short and medium term and, therefore, whether it can be used to stabilise output growth, while remaining true to its primary objective of price stability. I have argued that, while fine-tuning of the economy is generally to be avoided, the ECB strategy can provide sufficient scope and flexibility for dealing with unexpected sizeable economic fluctuations consistently with the maintenance of price stability. This flexibility has also been highlighted in the outcome of the recent re view of the ECB's monetary policy strategy. By clarifying that it aims to maintain inflation rates below but close to 2% over the medium term, the ECB has underlined its commitment to providing a sufficient safety margin to guard against the risks of deflation.
The effectiveness, however, with which monetary policy can perform a stabilising role is limited by several factors. These include the uncertainty regarding the timing and magnitude of its effects, which partly reflects the dynamics of the cycle, the nature of expectations and the type of shocks affecting the real economy. For instance, in an environment of very low interest rates and inflation or when the implementation of the appropriate economic policies is constrained, say by political obstacles to structural reform. In such situations, the effectiveness of monetary policy would depend on the underlying causes of the economic weakness and on the stance of other policies. Monetary policy may be more effective if the problems being faced are perceived as short-lived, for example, caused by a temporary decline in consumer confidence. In such circumstances, a temporary monetary easing that does not endanger price stability may provide a stimulus to help economic agents cope with a cyclical slowdown.
If, however, the economic problems are persistent and/or structural in nature, then monetary policy is likely to be less effective. For instance, if firms are reluctant to invest because of structural impediments implying few profitable business opportunities, a short-term monetary easing is unlikely to make much difference. In such circumstances, structural reforms aimed at raising trend growth can help restore confidence and improve the "animal spirits" of investors. In this way a change in the monetary policy stance can become more effective. Moreover, structural reforms that lead to an increase in aggregate supply and greater price flexibility can be expected to reduce upside risks to price stability and therefore give monetary policy more scope for manoeuvre.
At this juncture, in order to speed up economic recovery and achieve higher and sustainable growth in Europe, it is important for economic policies to strengthen confidence and enhance the competitiveness of the European economy. To this end, the implementation of credible fiscal consolidation strategies, based on growth-enhancing measures, can boost confidence and private spending and thus counteract the direct effects of budgetary measures on aggregate demand. More importantly, the introduction of structural reforms to improve productivity and market flexibility is essential for increasing potential growth as well as international competitiveness in a sustained way. These reforms could also help to increase confidence in the euro area's capacity to grow, with favourable effects on aggregate demand and economic activity in the short run. Monetary policy cannot help to solve the structural problems constraining the growth performance of the euro area. The implementation of fiscal consolidation strategies and structural reforms in Member States will facilitate the conduct of the ECB's monetary policy so that it can foster growth combined with the maintenance of price stability.
Akerlof, G. A., W. T. Dickens and G. L. Perry (1996), "The Macroeconomics of Low Inflation", Brookings Papers on Economic Activity, Vol. 1, pp. 1-59.
Akerlof, G. A., W. T. Dickens and G. L. Perry (2000), "Near Rational Wage and Price Setting and the Long-run Phillips Curve", Brookings Papers on Economic Activity, Vol. 1, pp. 1-44.
Andrés, J. and I. Hernando (1999), "Does Inflation Harm Economic Growth? Evidence from the OECD", in M. Feldstein (ed.), The Costs and Benefits of Price Stability, University of Chicago Press, Chicago.
Angeloni, I., A. Kashyap, and B. Mojon (2003), Monetary Policy Transmission in the Euro Area, Cambridge University Press, Forthcoming.
Barro, R. J. (1997), Determinants of Economic Growth: A Cross-Country Empirical Study, The MIT Press, Cambridge, Massachusetts.
Blanchard, O. and J. Simon (2001), "The Long and Large Decline in US Output Volatility", Brookings Papers on Economic Activity, pp. 135-164.
Bruno, M. and W. Easterly (1996), "Inflation and Growth: In Search of a Stable Relationship", Federal Reserve Bank ofSt. Louis Review, Vol. 78(3), pp. 139-146.
Bullard, J. and J. W. Keating (1995), "The Long-Run Relationship between Inflation and Output in Postwar Economies", Journal of Monetary Economics, Vol. 36(3), pp. 477-96.
Cooley, T. F. and G. D. Hansen (1989), "The Inflation Tax in a Real Business Cycle Model", American Economic Review, Vol. 79, September, pp. 733-748.
Cozier, B. and J. Selody (1992), "Inflation and Macroeconomic Performance: Some Cross- Country Evidence", Bank of Canada Working Paper No. 92-06.
Danthine, J. P. (1985), "Inflation and Growth in Utility-Maximising Models", University of Lausanne Working Paper, No. 8511.
European Central Bank (2003), Overview of the Background Studies for the Reflections on the
ECB 's Monetary Policy Strategy, http://www.ecb.europa.eu/pub/strateg y /monetar y polic y strategyreview-overview.pdf .
Feenstra, R. C. (1986), "Functional Equivalence Between Liquidity Costs and the Utility of Money", Journal of Monetary Economics, Vol. 17, pp. 271-291.
Fischer, S. (1993), "The Role of Macroeconomic Factors in Growth", Journal of Monetary Economics, Vol. 32, pp. 485-512.
Fischer, S. and F. Modigliani (1978), "Towards an Understanding of the Real Effects and Costs of Inflation", Weltwirtschaftliches Archiv, Vol. 114, pp. 810-832.
Ghosh, A. and S. Phillips (1998), "Warning: Inflation May Be Harmful to Your Growth", IMF StaffPapers, Vol. 45, No. 4, pp. 672-710.
Gomme, P. (1993), "Money and Growth Revisited: Measuring the Costs of Inflation in an Endogenous Growth Model", Journal ofMonetary Economics, Vol. 32(1), pp. 51-77.
Greenwood, J. and G. W. Huffman (1987), "A Dynamic Equilibrium Model of Inflation and Unemployment", Journal ofMonetary Economics, Vol. 19, pp. 23-28.
Grier, K. B. and G. Tullock (1989), "An Empirical Analysis of Cross-National Economic Growth, 1951-1980", Journal ofMonetary Economics, Vol. 24(2), pp. 259-276.
Groshen, E. L. and M. E. Schweitzer (1999), "Identifying Inflation's Sand and Grease Effects in the Labour Market", in M. Feldstein (ed.), The Costs and Benefits of Price Stability, University of Chicago Press, Chicago.
Haslag, J. H.(1997), "Output, Growth, Welfare, and Inflation: A Survey", Federal Reserve Bank of Dallas Economic Review, Second Quarter, pp. 11-21.
Ho, W. M. (1996), "Imperfect Information, Money and Economic Growth", Journal of Money, Credit and Banking, Vol. 28(4), pp. 578-603.
Issing, O. (2001), "Why Price Stability" in A. Garcia-Herrero, V. Gaspar, L. Hoogduin, J. Morgan and B. Winkler (eds.), Why Price Stability?, European Central Bank, Frankfurt am Main.
Jones, L. E. and R. E. Manuelli (1995), "Growth and the Effects of Inflation", Journal of Economic Dynamics and Control, Vol. 19, pp. 1405-1428.
Judson, R. and A. Orphanides (1996), "Inflation, Volatility and Growth", International Finance, Vol. 2(1), pp. 117-138.
Keynes, J. M. (1936), The General Theory of Employment, Interest and Money, Macmillan, London.
Khan, M. S. and A. S. Senhadji (2001), "Threshold Effects in the Relationship Between Inflation and Growth", IMF StaffPapers, Vol. 48, No. 1, pp. 1-21.
Kormendi, R. C. and P. G. Meguire (1985), "Macroeconomic Determinants of Growth: Cross- Country Evidence", Journal ofMonetary Economics, Vol. 16(2), pp. 141-163.
Levine, R. and D. Renelt (1992), "A Sensitivity Analysis of Cross-Country Growth Regressions", American Economic Review, Vol. 82(4), pp. 942-963.
Lucas, R. E. (1973), "Some International Evidence on Output-Inflation Tradeoffs", American Economic Review, Vol. 63(3), pp. 326-334.
Lucas, R. E. (2003), "Macroeconomic Priorities", American Economic Review, Vol. 93(1), pp. 1-14.
McCandless, G. T. and W. E. Weber (1995), "Some Monetary Facts", Federal Reserve Bank of Minneapolis Quarterly Review, Vol. 19(3), pp. 2-11.
Metzler, L. (1951), "Wealth, Saving, and the Rate of Interest", Journal of Political Economy, Vol. 59, pp. 93-116.
Mundell, R. (1963), "Inflation and Real Interest", Journal of Political Economy, Vol. 71, pp. 280-283.
Orphanides, A. and R. M. Solow (1990), "Money, Inflation and Growth" in B. M. Friedman and F. H. Hahn (eds.), Handbook of Monetary Economics, Elsevier Science Publishers B. V., Amsterdam.
Palokangas, T. (1996), "Inflation and Growth in an Open Economy", Economica, Vol. 64, pp. 509-518.
Papademos, L. (2003), "Economic Cycles and Monetary Policy", paper presented at the International Symposium of the Banque de France, Monetary Policy, Economic Cycle and Financial Dynamics, Paris, March.
Sarel, M. (1996), "Nonlinear Effects of Inflation on Economic Growth", IMF StaffPapers, Vol. 43, No. 1,pp. 199-215.
Sidrauski, M. (1967), "Rational Choice and Patterns of Growth in a Monetary Economy",
American Economic Review Papers andProceedings, Vol. 57, pp. 534-544.
Stein, J. L. (1970), "Monetary Growth Theory in Perspective", American Economic Review, Vol. 60, pp. 85-106.
Stockman, A.C. (1981), "Anticipated Inflation and the Capital Stock in a Cash-in-Advance Economy", Journal of Monetary Economics, Vol. 8, pp. 387-393.
Tobin, J. (1965), "Money and Economic Growth", Econometrica, Vol. 33, pp. 671-684.
Tobin, J. (1972), "Inflation and Unemployment", American Economic Review, Vol. 62(1), pp. 1-18.
Vickrey, W. (1955), "Stability Through Inflation", in K. Kurihara (ed.), Post-Keynesian Economics, George Allen and Unwin Ltd., London.
Zhang, J. (2000), "Inflation and Growth: Pecuniary Transactions Costs and Qualitative Equivalence", Journal of Money, Credit and Banking, Vol. 32(1), pp. 1-12.
 I would like to thank Julian Morgan for his valuable research contribution.
 The mechanism through which the Tobin effect arises, with inflation impacting on the real rate of interest and thereby on the accumulation of capital, can be traced back to Metzler (1951), albeit in a more limited form.
 An unexpected but permanent increase in the inflation rate causes a decline in real private wealth leading to an increase in real saving and the capital stock, and a consequent decline in the real interest rate.
 Sidrauski (1967) showed that money is "superneutral" in a utility maximising framework of infinitely lived consumers.
 Even if inflation does not affect the level of output it could affect its composition. According to Feenstra (1986) an increase in inflation which leads people to economise on money balances could result in a change in the composition of output from consumption goods to financial services
 See Stockman (1981), Greenwood and Huffman (1987) and Cooley and Hansen (1989).
 See, for example, Zhang (2000).
 The list of papers which establish such a negative relationship includes Kormendi and McGuire (1985), Grier and Tullock (1989), Cozier and Selody (1992), Fischer (1993) and Barro (1997).
 For instance, McCandless and Weber (1995), Bullard and Keating (1995).
 See Levine and Renelt (1992).
 See, for example, Bruno and Easterly (1996).
 See Sarel (1996), Judson and Orphanides (1996), Ghosh and Phillips (1998) and Khan and Senhadji (2001).
 The estimated value of the threshold varies considerably between studies. Using annual data for 87 countries over the period 1970-1990, Sarel (1996) estimates that there is a structural break in the relationship between inflation and growth at inflation rates of around 8%. With a larger sample covering 145 countries for the period 1960-1996, Ghosh and Phillips (1998) find a much lower threshold value of around 2-3% inflation. More recently, Khan and Senhadji (2001) have split the sample into industrial and developing countries to allow for the possibility that threshold values may differ. They obtained an estimated threshold value of between 1-3% for industrialised countries and 11-12% for developing countries.
 These papers are briefly surveyed in Issing (2001).
 See Groschen and Schweitzer (1999).
 Other reasons for tolerating a small positive rate of inflation include the potential for measurement bias in price indices, the possibility of sustained inflation differentials in a monetary union and the prospect of encountering the zero lower bound on nominal interest rates. For a discussion of these issues, see European Central Bank (2003).
 A review and assessment of the theory and evidence concerning the links between monetary policy and economic cycles is presented in Papademos (2003).
 The papers contained in Angeloni et al. (2003) present and discuss the evidence on monetary policy transmission in the euro area.
 This literature is reviewed in Barro (1997).
 Council of the European Union (2003).