A route for Europe
Address by Mario Draghi, President of the ECB
at the day in memory of Federico Caffè
organised by the Faculty of Economics and the Department of Economics and Law at the Sapienza University
Rome, 24 May 2012
A teacher, says Eco, “ teaches that everyone must become individual and different”.  Professor Federico Caffè, albeit with a coherent vision and deeply held convictions, was a teacher. He taught his students to think for themselves and did not pass on a binding creed. He helped his students – economists, thinkers, servants of the state and of the institutions, alert citizens – to discover themselves.
I’ll start with the subject which, without a doubt, was the most precious to Caffè, namely welfare. Probably nothing in his intellectual heritage is more topical than this painful protest of his: one cannot, he would say, “ accept the idea that an entire generation of young people should consider themselves as having being born at the wrong time and having to suffer job insecurity as an inevitable fact”. 
Work: a European matter
“ Full employment is not only a means of increasing production..., it is an end in itself, since it leads to overcoming the servile attitude of those who find it hard to obtain a job opportunity or live in constant fear of being deprived of one. In other words, the benefits of full employment are considered as well, and above all, in terms of human dignity.” These words of Caffè do not surprise those who knew him and those who have read his works. They express the fundamental inspiration of his professional and public life: it is a duty of economic policy to act so that the economy can get as close as possible to full employment.
In 1975, he formulated it more precisely: “ The goal of dignified work for all, however, is not compatible either with situations of privilege, which have now become destabilising, nor with excessive labour and social security rights, which results in job opportunities evaporating away” The issue that Caffè raises here is one of fairness. We find it again today: the international crisis has affected everyone, and young people especially.
In the European Union, between 2007 and 2011 the unemployment rate rose by 5.8 percentage points among the 15-24 year olds, by 3.5 points among the 25-34 year olds and by 1.8 points in the 35-64 age range. Qualitatively, the profile is similar almost everywhere; the clear exception is Germany, where the unemployment rate among 15 to 24 year olds in the first quarter of 2012 was 8%; in Italy it was 34.2%, in Spain 50.7% and the euro area average was 21.9%. These trends reflect a fundamental question: they confirm the particular vulnerability of this essential part of our workforce. The unequal sharing of the “cost of flexibility”, only affecting young people, an eternal flexibility with no hope of stabilisation, leads among other things to companies not investing in young people, whose skills and talents often decline in jobs with low added value. The underuse of their resources reduces growth in various ways: it makes the creation of start-ups less likely – and they are on average more innovative than others – it causes a decline in skills in the long run, slowing down the assimilation of new technology and acting as a brake on efficient production processes. In addition to undermining society’s sense of fairness, it is a waste that we cannot afford.
I think it’s essential to ask how economic policy conducted in various Member States has done its duty in the way desired by Caffè.
Social progress is one of the key objectives of the European integration process: “ The Union shall work for the sustainable development of Europe based on balanced economic growth and price stability, a highly competitive social market economy, aiming at full employment and social progress … It shall combat social exclusion and discrimination, and shall promote social justice and protection, equality between women and men, solidarity between generations and protection of the rights of the child”.
(Article I-3 of the draft European Constitution). Welfare is not only a remedy for the failure of insurance markets, but also a tool to promote inclusion, solidarity and a sense of fairness.
In the three post-war decades (the so-called “Golden Age”), which especially in Europe were marked by high growth rates, use of advanced technologies, high growth employment, stable lifetime employment, welfare started to emerge, at different times and on different scales depending on the country, as an integrated system that protects its citizens from significant risks.
The European model redistributes many more resources for social purposes than the US and Japanese systems: on the eve of the crisis, the total expenditure on pensions, unemployment benefit, for children and families was, in relation to GDP, more than twice that of the US and Japan. This can take different forms as regards the composition of social policies and the degree of labour protection. In Italy, with an overall welfare spending ratio of GDP in line with that of the rest of Europe, spending on support for the unemployed, for households, particularly those at risk of falling into poverty, is at a level less than half that of elsewhere in Europe, while spending on pensions is significantly higher. The weakness of the “social shock absorbers” is that relatively high job protection for those in employment is accompanied by weaker protection for those out of work, contrary to what happens in the Nordic models, where the so-called “flexisecurity” combines an extensive social safety net with less job protection.
In some countries, even if, 40 years on, the assumptions of that model are still valid, reflections on it began some time ago.
Structural factors have changed the context within which the European social model operates: the growing competition from emerging countries, the reorganisation of production processes on a global basis, the speed of innovation, the increasing fragmentation of career paths with ever looser ties to a “permanent position”, the greater instability of families, declining fertility, the prospective decrease in the workforce, an ageing population. The set of risks faced by individuals throughout their life has changed significantly.
The social protection systems are therefore constantly evolving; substantial corrections have taken place in recent years in many countries, including France, the United Kingdom and Germany, the country where the reform process began a decade ago. In Italy, the recent pension reform which approves the full transition to a contribution system completes the necessary correction of the pension spending dynamics which was started years ago. As Germany shows very well, large and effective welfare systems can be made more efficient without compromising social goals.
We are living at a critical juncture in the history of the Union. The sovereign debt crisis has exposed serious weaknesses in the institutional framework; in this context, the difficulties in finding common solutions are having a negative impact on market valuations. The extraordinary measures taken by the ECB have gained us time; they have preserved the functioning of monetary policy.
But we have now reached a point where European integration, in order to survive, needs a bold leap of political imagination. It is in this sense that I have referred to the need for a “growth compact” alongside the well-known “fiscal compact”.
A growth compact rests on three pillars and the most important one, from a structural viewpoint, is political: the economic and financial crisis has challenged the myopic belief that monetary union could remain just that, and not evolve into something closer, more binding, into an arrangement whereby national sovereignty on economic policy is replaced by the Community ruling. If the governments of the Member States of the euro define jointly and irrevocably their vision of what the political and economic construct that supports the single currency will be and what the conditions to reach that goal together should be. This is the most effective answer to the question everyone is asking: “Where will the euro be in ten years’ time?”.
The second pillar is that of structural reforms, especially, but not only, in the product and labour markets. The completion of the single market and the strengthening of competition are crucial for growth and employment. Labour market reforms that combine flexibility and mobility with a sense of fairness and social inclusion are essential.
Growth and fairness are closely connected: without growth, and the events of recent months also reflect this, the temptation to “circle our wagons” gains strength, and solidarity weakens. Without fairness, the economy breaks up into multiple interest groups, no common good emerges as a result of social and economic interaction, and there are negative effects on the capacity to grow. Recent Italian history has no shortage of examples.
These reforms have long been indispensable in a global economy very different to the one which witnessed the creation of the institutions still operating today. In the political structure that will emerge from the crisis it is likely and desirable that for these reforms a system of European rules will be introduced similar to that for the fiscal compact, a discipline leading over time to the European harmonisation of objectives and tools.
The third pillar is the revival of public investment: the use of public resources to push forward investment in infrastructure and human capital, research and innovation at national and European levels. (The proposed strengthening of the EIB and the reprogramming of Union structural funds in favour of less-developed areas go in this direction).
Thus, a growth compact complements the fiscal compact, because there can be no sustainable growth without orderly public finances. In this regard I have noted on other occasions the extraordinary progress made by all governments of the euro area in terms of fiscal consolidation, but, once the emergency is overcome, they need to make improvements by cutting current spending and taxation.
Let me now consider some issues more directly related to the ECB’s monetary policy and action.
Objectives and instruments
The first issue involves the relationship between objectives and instruments of economic policy in a macroeconomic reference model that changes over time. For Caffè, one of the first scholars of Frisch and Tinbergen, the optimal allocation of tools and objectives occurred in the reference model prevalent in the 1970s, in which the goal of economic policy was to make best use of the available resources, in particular full employment. In pursuit of this goal, monetary policy was subordinated to fiscal policy, with an ancillary role for the central bank vis-à-vis the Treasury. The reference macro-model was based on a static mechanism of elaborating expectations, which were formed by extrapolating from past observations to the future. This mechanism amplified the immediate effect of public spending on aggregate demand. Monetary policy – in charge of credit conditions – was entrusted with the task of alleviating, through a careful policy of accommodation, the impact that government borrowing would have exerted on the cost of private debt. The central bank was financing the Treasury by creating money. Under these conditions, an increase in public spending could “add demand” – where this would be lacking for the goal of full employment – without taking away resources from other uses.
Since then, the theory of economic policy has followed two paths that have led it to invert the ranking of relative potential of tools and to enhance the definition and measurement of the objective.
As for the instruments, a different theory of the formation of expectations – no longer extrapolative – has highlighted the strong impact of monetary policy and has weakened the expected effects of fiscal policy. In the models that we use today, agents, when formulating their expectations, are attentive to the sustainability conditions of the choices in the long term. Economic policies deemed unsustainable in the long run are ineffective. For example, an active deficit-financed fiscal policy is limited by fears about the government’s ability to refinance the debt from which that policy originates. These fears may lead to behaviours that weaken the private sector – or, at worst, completely neutralise – the impact of public spending as a means of controlling demand. Monetary policy, by contrast, is strengthened by this. Acting through the expectations channel, it can have a lasting effect on the expected flows of financial revenue. Affecting the real rate of intertemporal discount, it can deeply affect decisions on savings and consumption.
The definition of the objective is now wider. Price stability has become an essential parameter in defining and measuring prosperity. From taking a relaxed approach to inflation and considering it secondary, nowadays low and predictable inflation is a pre-eminent criterion of economic performance. Why? High inflation hits savings – and therefore investment and future consumption – with a tax on real returns that rewards the risk of uncertain inflation. Low inflation, however, frees up resources that individual choices can allocate to increasing the fixed capital. A policy that neglects inflation gradually destabilises the economy. The costs of uncertainty about the value of the money, initially unimportant, then overlooked, subsequently become evident, and then are judged intolerable. At that point, voters express a strong preference for policies that promise rapid disinflation. But such policies impose high costs in terms of job losses, which have to be included in the dynamic calculation of the costs of inflation. Also, it should not be forgotten that inflation affects the poor more than the rich, and is therefore a tax on the weaker members of society.
Under the influence of the neo-classical synthesis of Samuelson and Solow, the long-term correlation between inflation and growth was perceived as positive in the early 1970s: a slight increase in inflation would have led – within limits – to an increase in employment and growth.  But by the end of that decade, studies by Bob Lucas and Tom Sargent were to show that long-term inflation and growth are not correlated.  Monetary policy, while very effective in the short term, only affects inflation in the medium and long term. It is, in other words, neutral.
Today, new models and advanced computational techniques allow us to simulate the effects of inflation on incentives to save and to work, on the formation of physical capital, and therefore on the prospects for growth. The correlation has become negative: higher inflation reduces growth and employment. For example, vector autoregressive models with non-constant parameters allow the identification of a range of values that quantify the cost in terms of growth failure for every two percentage point increase in steady-state inflation. This cost implies lower growth of between 3 and 5 percentage points over a period of ten years.
Finally, monetary policy is a powerful tool. When used improperly, it may cause permanent damage. But it can become an effective, stabilising factor and contribute to collective prosperity in an independent and active way. The sine qua non for this is to build monetary policy decisions into a systematic and predictable strategy, based on price stability, which drives expectations and guides the economy but doesn’t shock it. This is perhaps the most important practical difference to what was studied in the early 1970s.
The ECB’s monetary policy strategy
The ECB’s monetary policy strategy is based on the new theory of the instruments I have tried to outline, and takes into account the enrichment of the theory of objectives, which emphasises the contributions of price stability to the “general prosperity”. It also provides continuity for a monetary tradition in continental Europe that has ensured inflation-free growth for more than 60 years.
The strategy is based on the objective of “ maintaining price stability” that the Treaty has entrusted to the central bank and the quantitative definition that the Governing Council subsequently gave the objective. The studies I have mentioned helped to define a range within which inflation is no longer a factor that distorts economic choices. The ECB pursues, as an objective of monetary stability in the medium term, an inflation rate below, but close to, 2%, which is the upper limit of this range.
The macroeconomic model on which the ECB’s monetary policy strategy rests is imbued with contemporary macroeconomics, based on dynamic optimisation and on the centrality of forward-looking expectations.  At the same time, the model is broader and well structured and corrects the simplifications of the pre-crisis neo-Keynesian paradigm with its prescriptive approach to optimal monetary policy.
The weakness of this paradigm was, and is, its inability to recognise the importance of financial frictions and the role of credit and money. This has to do with the fragility of the theoretical foundations that formalise the links between the real economy, financial imbalances and the level of confidence. Ignoring money is tantamount to assuming an absence of risk and uncertainty. Without risk, Keynes would have said, there would be no money. The preference for liquidity is not justified in an economy without uncertainty. But the neo-Keynesian model excludes the possibility of default. In it, risks in the financial sector can be isolated and therefore have no effect on the real economy.
The financial crisis has clearly highlighted the weaknesses of this system. Macroeconomic theory has begun to reflect on the neo-Keynesian system and these studies are now one of the liveliest areas of analysis. These studies – at least their early results – confirm the farsightedness of some of strategic choices of the ECB. Liquidity, money, credit have always been – since 1998, when the Bank received its mandate from the founders of the monetary union – qualifying variables of the ECB’s reference model and its strategy. The monetary analysis requires a constant monitoring of banks’ assets and liabilities as sources of information on the assessment of risk in the markets and the economy as a whole. This analysis commits the Governing Council to adjust the tenor of monetary policy to ensure the long-term growth of monetary aggregates and credit consistent with the potential for economic expansion.
In this sense, the monetary pillar of the strategy can be interpreted as a strategic reinforcement that helps to prepare correction mechanisms in situations where macroeconomic imbalances are having difficulty in manifesting themselves in inflationary pressures.
The monetary analysis gave important warning signals in the years preceding the crisis regarding the existence of deep macroeconomic and financial imbalances. In the autumn of 2005, in conditions of inflation observed and projected to be “normal”, the ECB’s monetary analysis began to record a change in the composition of M3 growth: from a model of growth explained by money demand factors – that we would define as irrelevant to the evolution of spending and prices – to a dynamic associated with increased credit creation – that is, to banks’ money supply factors. These changes were accepted as indications that the tenor of monetary policy, despite the moderation of inflationary pressures observed and projected, had become too lax. A new cycle of monetary policy tightening was initiated in December of that year, based on considerations inspired by the monetary pillar, where monetary and financial stability is an intermediate objective for attaining a more balanced development of macroeconomic variables and hence price stability over the long term.
In a globalised world, the international financial crisis has not spared the euro area. But today we know that preventive action by the ECB, implemented mainly by considering monetary indicators, has mitigated the impact on incomes and inflation. 
The two operations of three-year refinancing (LTROs)
The monetary analysis has also been an essential strategic tool in diagnosing and managing the crisis. In this regard, the analysis that led to the more recent non-standard monetary policy measures can illustrate how the systematic monitoring of banks’ liabilities – money in its broadest sense – can provide guidance on the risks faced by the economy as a whole.
Towards the end of 2011, with a decision unprecedented in the history of the euro, the ECB’s Governing Council decided to conduct two three-year refinancing operations. At the end of February, or when the second three-year operation was completed, the net increase in loans granted to counterparties was around €520 billion.
The motivation for the two operations can be summarised by the following strategic view. A central bank is mandated with the crucial task of ensuring the sufficient supply of liquidity to sound bank counterparties in return for adequate collateral. In normal times, “sufficient liquidity” means a volume of refinancing in line with the need for banks to meet the obligatory reserve requirements and the financing of other independent factors which explain the growth over time in the demand for money. In times of increased financial instability, “sufficient liquidity” indicates a volume of available central bank money which avoids the risk that – under such market conditions – the temporary inability of banks to provide refinancing leads to insolvency and thus to a situation of widespread default.
In neither of the two cases – normal times or crisis periods – can the central bank be considered responsible for the survival of bank counterparties that are close to bankruptcy.
The two long-term refinancing operations achieved the purpose for which they were intended. In an environment of a near-shutdown of private credit markets, banks were not able to refinance their assets and were unable to maintain their level of exposure to households and businesses. The extensive long-term refinancing allowed for the partial and temporary substitution of private credit with central bank money and thus avoided a disorderly process of credit to the economy running dry.
Nevertheless, these operations were not without their criticism. These can be summarised in three points:
The growth in liquidity following the two operations will ultimately lead to inflation;
The Eurosystem’s balance sheet is exposed to unprecedented and uncontrollable risks;
Such operations have reinforced the perverse link between banks and sovereign debtors and may be considered a violation of the prohibition of financing public debt with central bank money, laid down in Article 123 of the Treaty. And, the opposite – but conceptually equivalent – criticism that these operations did not provide the economy with finance.
As regards the first point, it is again pertinent to refer to strategy: in the medium to long term, the inflation dynamic reflects developments in broader monetary aggregates. Conditions that encourage the creation of inflation or speculative bubbles are generated by strong and sustained growth in money and credit, not necessarily as a result of an increase in liquidity granted by the central bank to the banking sector. This liquidity constitutes a precautionary supply for sight liabilities that the banks have towards households, non-financial corporations and other banks. The sight liabilities, i.e. deposits, and not the supply of liquidity, demonstrate a high statistical correlation to the price dynamics of goods and assets.
Today, monetary developments do not allow for the identification of risks of inflation or pressure on asset prices in the medium term.
As regards the second point, the main criticism related principally to the expansion of collateral accepted by the national central banks of the Eurosystem as a guarantee in return for liquidity extended to credit institutions. In particular, doubts were raised regarding credit claims that became eligible with the decisions taken in December 2011.
These doubts are founded on an incorrect understanding of the guarantees that are requested by the national central banks to protect against the risk that central bank liquidity is not repaid. In particular, the discount applied to the nominal value of credit claims provided as security in refinancing operations is very high. This means that, for credit claims deposited as collateral with a nominal value of €100, the national central banks accepting the new collateral provide, on average, the equivalent of around €47 in liquidity. This discounting represents a powerful method for absorbing the credit risk involved in such operations. It is also worth highlighting the fact that the main elements of risk control continue to be shared by the Eurosystem: the criteria for acceptance and measures for controlling risk are approved by the Governing Council, which is also responsible for the continuous monitoring of the effectiveness of all of the measures for mitigating risk.
With regard to the third point, it is undeniable that, in some countries in particular (especially Spain and Italy), banks used some of the liquidity acquired via the three-year long-term refinancing operations for temporary investments in government bonds.
Today, central banks do not have an instrument for the precise and targeted allocation of credit to a sector or in favour of a specific financial use. Those who accuse the ECB of an indirect violation of the prohibition on financing public debt with central bank money are making the same conceptual mistake as those who accuse the ECB of not having forced the banks to the use the funds acquired via the LTROs to supply credit to the private sector. In the first case, banks would have had to have been forced not to purchase government securities, and in the second case, to give credit to the private sector. Both groups of accusers forget that the policy of precise allocation of credit was a norm in various countries until the end of the 1970s. In the 1980s the operational system for monetary policy was radically redefined, principally on account of the heavily distorting effects of such an operational framework on economic activity. Moreover, legal arguments relating to the Treaty and to the current contractual form of repos underlying the LTROs, as well as considerations relating to feasibility, would make it difficult to reinstate such a framework. Finally, the behaviour of credit institutions with respect to households and businesses is not uniform across countries: while some countries saw negative credit developments, others saw growth in credit, in some cases even significant growth.
In Italy, but also in the vast majority of euro area countries, the fall in loans recorded in December has come to a halt, avoiding a much more severe risk of credit restriction which would have had far more serious consequences for growth and monetary stability than the ones we are seeing currently. The Bank Lending Survey registered a gradual normalisation of interest rates set by banks and of the criteria for granting loans to companies. The continued anaemic developments in lending reflect the weakness in demand and the worsening of creditworthiness resulting from an adverse economic cycle. Furthermore, in the countries most adversely affected by the crisis, banks are rationing credit on account of certain prevalent contractual structures.
The large-scale participation in the February operation, in which around 800 banks obtained funding, and the composition of counterparties in terms of their size and type implies that the distribution of liquidity has been widespread and could be even more so in the near future. Furthermore, this widespread distribution of liquidity is also of advantage to small and medium-sized enterprises with which smaller banks have closer relationships. We would wish for the liquidity provided to end up as credit to the private sector. This is the motivation behind the new non-standard monetary policy instrument.
Central bank credit to banks has been a complement to – rather than a substitute for – private credit: in the first quarter of this year, the amount of bonds issued was equal to the total issued in 2011 as a whole, which shows that the operations, at least in the first few months of the year, provided the markets with liquidity, reactivating a number of credit channels. The two LTROs removed one of the obstacles – the only one over which the ECB has any influence – to credit, namely a lack of liquidity. The ECB cannot do anything to make up for a lack of capital, to change intermediaries’ risk perceptions, or to remove other structural obstacles present at national level. More generally, the size and complexity of the two LTROs is such that it will take time for all of their positive effects on the euro area economy to be fully felt. However, it is essential for growth and employment that credit institutions regain their ability to provide refinancing to the economy.
Tiberi, M., “Ricordo di Federico Caffè”, in Rivista italiana degli economisti, April 1997, p. 132.
Caffè, F., La solitudine del riformista, edited by Acocella, N. and Franzini, M., Bollati Boringhieri, Turin, 1990, p. 258.
Paul Samuelson and Robert Solow published their article entitled “Analytical Aspects of Anti-Inflation Policy” in the May 1960 edition of the American Economic Review. The article discussed the causes of inflation and the Phillips curve.
Bob Lucas then assembled the empirical evidence which motivated his study on the theory of neutrality in his Nobel lecture entitled “Monetary Neutrality”, 7 December 1995. The absence of a correlation between inflation and growth is documented in McCandless, G.T. and Weber, W., “Some Monetary Facts”, Federal Reserve of Minneapolis Quarterly Review, No 1931, 1995.
The pre-crisis new-Keynesian paradigm and the monetary policy strategy of inflation targeting derived from it are outlined in Clarida, R., Galì, J. and Gertler, M., “The Science of Monetary Policy: A New Keynesian Perspective”, Journal of Economic Literature, Vol. XXXVII, 1999, pp. 1661-1707. These were superbly expounded in Woodford, M., Interest and Prices, Princeton: Foundations of a Theory of Monetary Policy, Princeton University Press, 2003.
In a recent paper entitled “A monetary policy strategy in good and bad times: lessons from the recent past” (ECB Working Paper, No 1336, May 2011), S. Fahr, R. Motto, M. Rostagno, F. Smets and O. Tristani simulated a “counterfactual” monetary policy strategy which did not include credit and money among its variables of observation and reaction. The authors found that as a consequence of the shock caused by the collapse of Lehman Brothers, the euro area would have entered into a protracted period of deflation and the level of GDP at the end of 2010 would have been 2 percentage points below that which was observed.