It is a great pleasure for me to be here today in Madrid to deliver the closing remarks at this important conference on the interactions between monetary and fiscal policies.
Before starting my intervention, let me recall that we are presently in the purdah period, i.e. within the one-week period before the next decision meeting of the Governing Council. Thus, nothing that I will say today is intended to have implications on our future monetary policy decisions, nor should it be interpreted as such.
Undoubtedly, this conference has addressed a very timely topic. Because of the financial crisis, substantial policy challenges have emerged for which we need to find answers. And I am not sure whether it gives me much comfort that these challenges are not only faced by policymakers, but, in a deeper sense, by the economics profession as a whole. I think that for all of us much remains to be done until we will be able to say that we have ‘understood’ the lessons from the financial crisis.
The range of topics that have been covered over the past two days is impressive. I will not try to comment on all the papers. This would be too much for a single speech. In any case, at the end of an intensive conference this would probably not be in line with your preferences. Instead, I will focus my remarks today on the monetary and fiscal reaction of the euro area to the financial crisis. But before doing so let me start by making a few short, inevitably selective remarks on some of the papers. One paper which I found particularly relevant is the one by Eric Leeper, because I very much agree that the current crisis is a period in which uncertainty over future policy regimes poses particular challenges. As government debt reaches unprecedented levels in peacetime and ageing populations will significantly add to the fiscal burden in many industrialised countries in the coming years, the notion of a ‘fiscal limit’ that might be reached at some point in the future is immensely important. And this limit matters already today because of forward-looking expectations. Another interesting point which came out of the presentations by Wieland, Corsetti, Lane, Faia and others, is that the current environment has been conducive to a renewed and very rich debate about the determinants and the likely size of fiscal multipliers, about which there is still much uncertainty. Next, it seems clear to me that the paper presented by Loisel addresses a topic which will stay with the profession for a long time to come, namely the adequate integration of financial market frictions into state-of-the-art DSGE models. And finally, it may not surprise you that the paper presented by Cooper addresses a topic which does not escape my attention these days. By this I mean that monetary unions – which are characterised by the very special combination of a single, centralised monetary policy and many national fiscal policies – require a particular monetary and fiscal architecture.
This brings me to the theme on which I would like to devote my attention today, namely the reaction of euro area monetary and fiscal authorities to the financial crisis. I would like to focus my remarks on the fact that, while similar in many ways, in certain respects the reaction of the euro area authorities to the crisis has differed to that of other, major industrialised economies, such as the United States or Japan. And in this regard, an interesting question is the extent to which this can be linked to the specific architecture of Economic and Monetary Union. Before addressing this question, let me briefly recall the main elements of this architecture.
Indeed, when we discuss monetary and fiscal policy in the euro area, we should remember that we are talking about an economic area sui generis; an economic area within which the member countries have given up their own national currencies in favour of a single, area-wide monetary policy. Governments do not sign up to such ventures lightly. Monetary union only became possible in Europe because a sufficient consensus emerged regarding the basic principles of sound macroeconomic policy. And before going further, it is worth recalling these basic principles, which were written into the Maastricht Treaty.
Central bank independence.
A stability-oriented monetary policy with the primary objective of maintaining price stability.
And an obligation for Member States to treat their economic policies, in particular their fiscal policies, as a matter of common concern.
In order to protect monetary policy from profligate fiscal policies, the EU Treaty prohibits monetary financing and direct bail outs of governments in financial difficulty; it calls on governments to avoid excessive deficits, and it provides for an excessive deficit procedure ultimately leading to sanctions in case of flagrant breaches of European fiscal rules. Before the introduction of the euro, these rules were further elaborated and strengthened by the adoption of the Stability and Growth Pact.
In short, the premise of Economic and Monetary Union is a clear allocation of responsibilities among policy makers; a commitment of both monetary and fiscal authorities to keep their own houses in order; and a set of safeguards to ensure that all parties abide by the rules. Any analysis of monetary and fiscal policy in the euro area – not just in normal times but also in crisis times – should take into account the fact that euro area macroeconomic policy should be guided by these basic principles; because they are the principles on which the EMU contract was signed.
To what extent has the euro area response to the financial crisis been guided by these basic principles? Let me first very briefly recall the monetary and fiscal policy actions that have been undertaken in the euro area since the financial crisis intensified in autumn 2008. I will then discuss in more detail the rationale behind these actions.
Starting with monetary policy, about which I can obviously say most, during the crisis the Governing Council of the ECB has taken unprecedented action via both standard and non-standard measures. Given rapid changes in the assessment of risks to price stability, the ECB's main refinancing rate was cut by a cumulative 325 basis points between October 2008 and May 2009 bringing it to its current level of just 1%. This is a level not seen in the countries of the euro area since at least the Second World War. In addition, to support financing conditions and credit flows above and beyond what could be achieved through interest rate cuts alone, non-standard measures were adopted, the so-called “enhanced credit support". These measures – which for very good reasons are of a temporary nature, with some of them already being discontinued, as I will discuss at the end of my talk – can be grouped into five categories:
First, fixed rate tenders with full allotment in all liquidity-providing operations.
Second, additional refinancing operations with one-month and three-month maturities, as well as the provision of funding at longer maturities of six months and one year.
Third, a broadened collateral framework, notably the lowering of the rating threshold to BBB- and the acceptance of selected foreign currency assets and of securities issued in some non-regulated markets, all these with commensurate additional risk control measures.
Fourth, the covered bond purchase programme.
And fifth, the introduction of foreign currency-providing operations.
These extraordinary policy measures represented the ECB’s response to extraordinary shocks. They have ensured adequate provision of liquidity to the euro area banking system at favourable conditions, so as to support the provision of credit to households and firms and thereby contribute to the revival of the euro area economy. As a consequence of these ample liquidity conditions, the EONIA rate is set close to the deposit facility rate, i.e. 0.25%.
Euro area governments also responded to the crisis with exceptional measures.
First, euro area governments committed substantial public funds to support individual financial institutions and help stabilise the financial sector. Support came in various forms, in particular recapitalisations, the setting up of “bad banks” to remove toxic assets from balance sheets, and the guaranteeing of interbank lending, deposits and bond issuance.
Secondly, euro area governments have used (or at least allowed) fiscal policy to support economic activity through the crisis. In 2007, the euro area government deficit ratio was just 0.6% of GDP. In all likelihood, the euro area deficit reached around 6% of GDP last year and is likely to remain around this level or even rise closer to 7% this year. This is a significant fiscal impulse.
Let me now move from facts to their interpretation. In doing so I will have to be selective, focusing on aspects related to the conference theme of monetary and fiscal interactions as well as on aspects reflecting the specific architecture of EMU.
Focusing first on monetary policy, let me take as a starting point for my reflections an observation made in a recent overview BIS paper by Claudio Borio and Piti Disyatat (2009). Borio and Disyatat argue that the unprecedented responses of the world’s major central banks to the crisis have two distinct types of action in common, namely, interest rate policy and balance sheet policy. To stress this broad commonality is important: from the onset of the crisis all major central banks shared a broad consensus that an extraordinary monetary policy response was necessary. And this response went beyond interest rate policy.
Let me cite a few numbers for the euro area, which serve to illustrate what is meant by the term balance sheet policy. In summer 2007, before the crisis, the total assets of the consolidated balance sheet of the Eurosystem stood at EUR 0.9 trillion. This was around 10% of euro area GDP. At the height of the crisis, in January 2009, these total assets exceeded EUR 1.8 trillion. Currently, they stand at around EUR 1.5 trillion. Similar balance sheet movements can be observed also for other central banks.
However, Borio and Disyatat point out that there are interesting differences between the particular balance sheet policies chosen. According to their categorisation, these range from ‘credit policies’ to ‘quasi-debt management policy’, ‘bank reserves policy’ and ‘exchange rate policy’. The enhanced credit support provided by the Eurosystem is seen as a prime example of what the study refers to as credit policies. At the same time, the Eurosystem does not belong to the large group of central banks which have resorted to ‘quasi-debt management policy’. This reflects the fact that the Eurosystem has not considered buying euro area government bonds for monetary policy purposes. Similarly, the Eurosystem has not considered requesting public guarantees to support the expansion of its balance sheet. In sum, what sets the Eurosystem somewhat apart from other major central banks is a stylised feature which one may call the ‘minimisation of the fiscal implications of our monetary policies’.
Let me comment on this feature from a number of perspectives.
Clearly the avoidance of outright purchases of euro area governments’ debt (which in any case would have been clearly restricted to purchases in secondary markets) respects the above discussed clear separation of monetary and fiscal responsibilities enshrined in the EU Treaty. As a corollary of this separation, let me add: in monetary unions, with many sovereign issuers of government debt, there does not exist a single sovereign yield curve. Hence, outright purchases of government debt would inevitably be fraught with distributional concerns between countries to be avoided at the outset.
But the decision of the Eurosystem not to go the route of purchasing government bonds also reflected a number of additional considerations, of which I would like to stress four in particular.
Differences between the measures adopted by the Eurosystem and, for example, the Fed reflect different histories and different operational frameworks. In the United States, in normal times, repo operations are only used to correct temporary fluctuations in the liquidity needs of the banking system, while outright purchases and sales of treasury bonds with short maturities are used on a daily basis to implement monetary policy. This tradition is succinctly summarised, for example, by Marvin Goodfriend’s statement that "monetary policy refers to open market operations that expand or contract high-powered money (bank reserves and currency) by buying or selling Treasury securities". Given that tradition, it may be advantageous under non-standard circumstances to refine this procedure by significantly expanding the volume of purchases and focusing on bonds with longer maturities.
The Eurosystem faced at the outset the challenge to combine various traditions from its member countries. This has led to a broad-based framework in which “reverse transactions” – on the basis of repurchase agreements or collateralised loans – are the single most important instrument in open market operations. Given this starting point it has been a natural step to adapt the tender procedures of our refinancing operations and to adjust collateral requirements. Moreover, these adjustments were successful because access to these refinancing operations has traditionally been granted to a large pool of counterparties and the range of assets accepted as collateral, including private paper, has always been very broad since the start of Monetary Union in 1999.
Of course, under the Eurosystem’s refinancing operations balance sheet connections with fiscal authorities are not entirely avoided. But their nature is different from those in the United States. In particular, it is the counterparties of the Eurosystem’s monetary policy operations who decide which type of eligible asset they pledge as collateral. Interestingly, in the last two years (i.e. 2008 and 2009) the share of government securities in the total eligible collateral that has been posted has been smaller than in preceding years (while the share of asset backed securities, uncovered bank bonds and non marketable assets has increased).
The enhanced credit support has been tailored to the financial structure of the euro area in which banks play a particularly important role. Indeed, the bulk of the external financing of non-financial corporations comes from the banking sector. This feature put a premium on a broad-based reduction of funding constraints for banks rather than on direct interventions in multiple market segments via outright purchases of assets. Let me add: this feature also puts a premium on the recapitalisation and, where needed, restructuring of European banks, as repeatedly stressed by the Governing Council. Where appropriate, however, direct interventions were undertaken, as exemplified by the covered bonds purchase programme.
I expect that in a couple of years’ time – in light of the then available research on current events – we will understand much better the effectiveness of the various policy options that are available under ‘non-standard’ circumstances at which the short-term interest rate has reached its effective lower bound. I cannot prejudge the outcome of this research. Nevertheless, let me share with you some questions concerning the effectiveness of the options that have been on the table. For the sake of the argument, assume there is a uniquely defined sovereign yield curve. Then, under non-standard circumstances, short-term government debt and monetary balances are close substitutes in private sector portfolios. We know from studies in the spirit of Eggertsson and Woodford (2003) that, under well functioning capital markets, the effectiveness of outright purchases of government bonds tends to be limited, unless such purchases can be effectively used to steer expectations concerning future actions. Yet, given the observed market failures, the practical relevance of this reasoning is controversial, and most central banks were primarily interested in acting directly on liquidity premia. While this can be done via outright purchases, it seems to me that this can be done with similar effect via the maturity structure of tenders in a repo-based operational framework. Indeed, the Eurosystem’s experience has shown that a shift towards long-term refinancing operations can ensure secure funding of activities at low rates over a predictable horizon. But let me say again: I have touched here on questions for which we need more guidance from good academic work, both conceptually and empirically.
The non-standard measures adopted by the Eurosystem were designed with an exit strategy in mind. The Eurosystem's refinancing operations provide liquidity over a fixed time horizon and unwind in a fully predictable way. By contrast, the unwinding of outright purchases, in particular of long-term bonds, typically requires an additional decision, namely whether to hold the securities to maturity – and if not, when to sell. The route taken by the Eurosystem limits such decisions to the covered bonds purchase programme, which is limited in size such that the effect on total liquidity can be easily neutralized.
Before addressing exit considerations in further detail, let me first touch upon fiscal policies during the crisis.
The extent to which the euro area fiscal response to the crisis has been shaped by the EMU architecture, and in particular the European fiscal rules, is not easy to gauge. One could argue that the answer is “not much”. Most government support granted to the financial sector was either below the line or off balance sheet, so it didn’t affect government deficits according to the “Maastricht definition”. Obviously, there has been no concerted decision to keep Maastricht deficit and debt levels within the respective ceilings of 3% and 60% of gross domestic product, regardless of the economic circumstances. If we turn the clock back to autumn 2008, it was already obvious that most countries were heading towards “excessive” government deficits in the sense of the Treaty, if not in 2008, then certainly in 2009. Yet most euro area governments went ahead with fiscal stimulus anyway. More generally, the high fiscal deficits that have emerged in almost all industrialised countries over the past year seem incongruous with the 3% deficit limit of the Maastricht Treaty and the Stability and Growth Pact.
In light of this, to some extent it would be tempting to say that, as far as fiscal policy is concerned, the Treaty and the Stability and Growth Pact were simply ignored; an inconvenient obstacle that was brushed aside. But for at least four reasons, I think such a critical view would be too simplistic and inappropriate.
First, the Stability and Growth Pact has its “flexibility clauses”. According to the Pact, an economic downturn is deemed to be exceptional - potentially warranting a deficit in excess of the 3% of GDP reference value - in case of negative output growth, or a protracted period of low growth relative to potential. By this, or any measure, the circumstances faced at the height of the economic and financial crisis were exceptional and warranted an exceptional fiscal policy response.
Second, at least some euro area countries – including Spain – entered the crisis with relative strong starting positions in terms of low levels of government debt and/or budget surpluses. So in this sense, at least in some countries, there was fiscal room for manoeuvre at the beginning of the crisis.
Third, the extent to which euro area governments engaged in discretionary fiscal stimulus generally reflected the strength of their fiscal starting positions. Fiscal stimulus packages were comparatively large in low debt or surplus countries such as Spain and Germany, but virtually non-existent in high debt countries such as Italy.
And fourth, the discretionary fiscal stimulus in the euro area has not, actually, been that large. According to the European Commission, measures adopted by euro area governments under the umbrella of the European Economic Recovery Plan amounted to about 1% of euro area GDP in 2009. And even this is probably a slight exaggeration, as it ignores the fact that there were also offsetting tax increases and spending cuts in some countries.
The bulk of the increase in government deficits in the euro area during the crisis has been caused not by discretionary stimulus but by the automatic response of fiscal policy to lower output and inflation. Tax revenues fell as a consequence of lower wages, profits and consumption, while social spending increased due to rising unemployment. Moreover – and this is something that tends to get overlooked - the implementation of spending budgets expressed in cash terms in a context of lower than previously expected inflation has meant more spending in real terms; and, of course, higher deficits. By contrast, a larger proportion of the crisis related deficit increase in the United States can be attributed to discretionary stimulus measures. But this may have been justified given the smaller size of the American public sector.
The bottom line is that government deficits in the euro area would have reached levels well above the Treaty reference value of 3% of GDP even in the absence of any discretionary stimulus. Seeking to prevent this in the midst of the severest recession in many decades and at a time of intense uncertainty was not really an option for most countries. But even in these extreme circumstances, the European fiscal rules were not brushed aside. As actual or planned deficits have exceeded the 3% reference value, excessive deficit procedures have been launched, as foreseen in the Treaty and the Stability and Growth Pact. The impact of these excessive deficit procedures is not negligible. They have obliged the countries concerned, already at a relatively early stage, to put on paper their post crisis fiscal adjustment strategies.
Indeed, as the major economies around the world appear to be stabilising, the attention of policy-markers at the current juncture is increasingly on developing appropriate exit strategies. And in this regard I would argue that, on balance, the euro area is quite well placed, if anything having something of a head start.
As I already mentioned earlier, the non-standard monetary policy measures adopted by the Eurosystem have been designed with their phasing-out in mind. Several of them unwind naturally in the absence of an explicit decision to prolong them. This applies, for example, to the temporary expansion of the list of eligible assets. Others have already been discontinued, such as the provision of non-euro liquidity via swap lines with a number of other central banks, including the Federal Reserve. These operations have been discontinued against the background of improved functioning of financial markets over the past year and limited demand. Moreover, going back to a Governing Council decision already taken last December, the number and frequency of longer-term refinancing operations is gradually being scaled back. In particular, the last one-year operation was conducted in December 2009, and one final six-month operation with full allotment will be conducted next month. Further decisions on the gradual phasing-out of those non-standard measures that are not needed to the same extent as in the past will be announced in the coming months.
The decision to initiate the gradual phasing-out of these measures reflects improvements observed in financial conditions. Money markets are performing better, and the past reductions in the ECB’s key policy rates are increasingly reflected in the interest rates on bank loans to households and corporations. This indicates that the monetary policy transmission process is broadly functioning.
Looking forward, it goes without saying that the overriding concern for the ECB’s monetary exit considerations is the primary objective of price stability to be maintained over the medium term. From this perspective, I would like to stress that the operational framework with its interest rate corridor is flexible, allowing control over interest rates even without the phasing-out of all non standard measures. In other words, the Governing Council is free to choose the way in which interest rate action is combined with the unwinding of the remaining non-standard measures.
What about fiscal policy? I think it is fair to say that the “fiscal exit” - or I think it is better to say “fiscal adjustment” that will be necessary in the coming years is likely to prove even more challenging for governments than the phasing out of non-standard measures for the Eurosystem. During the crisis, governments were right to allow deficits to expand so that fiscal policy provided support to the economy in the short run. But now, government deficits are high – and indeed very high in many countries; government debt-to-GDP ratios are mostly high and rising; and the attention of financial markets is increasingly turning to issues of government solvency. In this context, it is of some comfort that – in part thanks to the European fiscal framework - the aggregate euro area government deficit is currently significantly lower than that of most other major economies. But this is only limited comfort. Prompt action to bring the fiscal house in order is necessary so that high and increasing levels of government debt do not jeopardise longer-term growth prospects. The next crisis must not be a sovereign debt crisis!
The key issue for fiscal policy at this juncture is not “whether” but “how much” and “how soon” to start fiscal consolidation.
Earlier, I argued that the increase in euro area government deficits during the crisis was largely automatic rather than being due to discretionary fiscal stimulus measures. Unfortunately, what automatically goes up does not always automatically come down. The crisis has put our economies on a lower path of potential output, although we still don’t know exactly what this path is. Yesterday’s cyclical deficits are, from today’s perspective, more structural in nature. The government sector needs to adjust to the new reality of a smaller economy. In many countries, this means that challenging decisions will have to be made in the coming years.
In the long run, fiscal consolidation – in other words moving from unsustainable to more sustainable public finances – is clearly beneficial for economic growth. I think we would all agree on this. However, fiscal consolidation is usually not without costs in the short-run.
Does this mean that under current circumstances fiscal consolidation should be postponed until the recovery is entrenched? Obviously a premature fiscal tightening that plunges the economy back into recession needs to be avoided. But barring this, the sooner fiscal policy adjusts the better. As long as fiscal policy is on an unsustainable course with government debt ratios still rising, the social costs of further delaying adjustment will exceed the costs to society of acting sooner.
We also have to acknowledge that there is a lot of uncertainty regarding the short run effects of fiscal adjustment. There is an extensive literature examining this issue and different studies have yielded quite different results. But broadly speaking, what we can ascertain from this literature is that the short run costs of fiscal adjustment are likely to depend on a variety of conditions. Notably, these costs are likely to be more limited:
If the fiscal starting position is precarious and the adjustment is credible;
If financial markets react by lowering long-term interest rates;
If households have correctly understood the need for fiscal adjustment and have factored this into their spending decisions; in other words, if households are – at least partly - “Ricardian”;
and if monetary conditions are accommodative.
This last condition touches on the very subject of today’s conference: monetary and fiscal policy interactions. And in certain respects, it has to be acknowledged that such interactions are particularly challenging for the euro area.
When we consider the monetary and fiscal exit from a common perspective, it is often argued that it is preferable that fiscal policy adjusts earlier and more forcefully than monetary policy. Ceteris paribus, a tighter fiscal stance means less inflationary pressure and more subdued inflation expectations. For a stability-oriented monetary policy, this should imply a less front-loaded phasing out of monetary stimulus. Yet, this line of reasoning largely ignores that by their nature fiscal policies lack strong built-in mechanisms when it comes to the unwinding of stimulus, and it rests on an assumption that all ‘players’ live up to their ex ante given promises. We know that in practice this assumption does not always hold.
In the euro area there are, indeed, many players. Hence, this reasoning should be further qualified in at least two respects.
First, in the euro area there are multiple fiscal authorities, so predicting the future stance of fiscal policy and correctly factoring this into monetary policy decisions is not a straightforward task. In such environment it would be particularly costly if delayed monetary action compromises on price stability, and, not less important, on the remarkably solid anchoring of medium-term inflation expectations. To put this differently: the stability-oriented monetary policy of the Eurosystem provides the nominal anchor which acts as the single most important implicit coordination device for all other actors with their various responsibilities, including fiscal policymakers. This anchor must not be put at risk.
Second, the task of monetary policy is facilitated if the future stance of fiscal policy is reasonably clear. In the euro area, the Stability and Growth Pact aims to guarantee sound public finances and, thereby, a reasonable degree of fiscal certainty. It is not without reason that the ECB has made repeated calls on governments to abide by their commitments under the Stability and Growth Pact. These commitments are known to everyone. From the current perspective, in December the ECOFIN Council established deadlines for most euro area Member States to correct their excessive deficits. The Council also recommended average annual adjustment paths geared to achieving these goals. These consolidation paths are reflected in the stability programmes that Member States have recently submitted to the Commission and Council in the context of the Stability and Growth Pact.
But a plan or a target is one thing; delivering it is another. In the past, the track record of some euro area governments in actually delivering promised fiscal consolidation has been mixed. This has created a credibility gap as far as euro area fiscal policies are concerned.
To conclude, it is therefore essential that consolidation targets are reflected as soon as possible in specific, detailed tax and spending plans. Credible and detailed fiscal adjustment plans should reassure investors, creating a more favourable, long-term financing environment. Credible and detailed fiscal adjustment plans should provide certainty for households making their saving and investment decisions. Credible and detailed fiscal adjustment plans would also facilitate the task of monetary policy in maintaining price stability, while gradually returning to a more normal, post-crisis monetary environment.
Reproduction is permitted provided that the source is acknowledged.