Public sector security purchases and monetary dominance in a monetary union without a fiscal union
Speech by Peter Praet, Member of the Executive Board of the ECB,
at the Conference The ECB and Its Watchers XVI,
Contribution to the Panel on Low-interest-rate Policy and Non-standard Monetary Policy Measures: Effectiveness and Challenges,
Frankfurt am Main, 11 March 2015
The Governing Council decision to start purchases of public sector securities has demonstrated its ability to meet the mandate of price stability and bring inflation rates back to levels below but close to 2 percent. With this policy, it makes full use of all legal and effective monetary policy instruments. The design of such purchases has taken into account the specificities of the institutional set-up, notably the fact that the euro area is a monetary union that is not a fiscal union. This is an assertion of monetary dominance, in compliance with the principles enshrined in the Maastricht Treaty.
Nevertheless the overall success of the euro area depends on all stakeholders doing their job.
Thank you very much to the organisers for inviting me to participate this morning.  In my contribution I would like to address the topic of this panel from a particular perspective, namely monetary dominance and the purchase of public sector securities in a monetary union without a fiscal union.
To preview my reasoning, monetary dominance is not about how the central bank implements its policy – i.e. the assets that it buys – it is about how much control the central bank has over its policy – i.e. why it buys those assets.
Indeed, the independence that has been given to the ECB (and, in particular, the purpose of the monetary financing prohibition enshrined in Article 123 of the Treaty) is precisely to ensure that the central bank has full control over its balance sheet – that it cannot be forced by governments into monetising deficits or inflating away debts – and hence that monetary dominance is preserved.
The decision taken by the Governing Council in January to expand the purchase programme to include public sector securities as well as the specific design elements of the programme, demonstrate that we can make full use of all instruments of monetary policy to deliver on our mandate of maintaining price stability. This is an assertion of monetary dominance, in compliance with the principles enshrined in the Maastricht Treaty.
Central bank purchases of public sector securities are a sensitive topic because, by definition, they constitute the most direct link between central bank actions and the budget constraint of the fiscal authority. The existence of this link naturally leads to the question: which constraints or institutional safeguards are necessary to have in place so that a central bank can fulfil its mandate of maintaining price stability?
A “single economy” perspective
Let me start with the case of a single economy perspective – by which I mean a country with one central bank and one fiscal authority – and I will then turn to a monetary union with one central bank and many fiscal authorities.
The relationship between monetary and fiscal policy from a single economy perspective goes back to the classical paper by Thomas Sargent and Neil Wallace on the ‘unpleasant monetarist arithmetic’.  The premise of their argument is that the central bank contributes to the budget of the government via seigniorage income. In its canonical version, this income stream captures the ability of the central bank to issue non-interest bearing central bank liabilities in exchange for interest-bearing assets (either outright or via revolving refinancing operations). The issuance of central bank money comes with a positive return differential, leading to an income stream that is eventually returned to the fiscal authority.
For the central bank to be unconstrained in its ability to ensure stable prices, it is crucial that in all states of the world the amount of that seigniorage income will be solely determined by monetary policy concerns. In other words, drawing on the taxonomy established by Sargent and Wallace, the central bank must be able to exercise monetary dominance. As we all know, monetary dominance, when translated into the realm of institutional design, corresponds closely to the concept of a strictly independent central bank bound by a statute that assigns it to the pursuit of an overriding objective of price stability. This, in turn, is the single most important requirement for the ability of the central bank to deliver price stability.
The second, and more controversial, concept I would like to mention goes back to the so-called ‘Fiscal theory of the price level’, as originally presented by Chris Sims, Michael Woodford and Eric Leeper.  It is conceptually different from Sargent and Wallace. It establishes a link to the budget constraint of the government, which operates through revaluations of outstanding amounts of nominal government debt, via adjustments in the price level. The idea is that such revaluations have the ability to ensure that government debt, when perceived to be on an unsustainable path, can be stabilised without outright default. In this concept, the requirement of the independent central bank that actively engages in setting its policy in the exclusive pursuit of its objective of price stability is no longer sufficient to ensure that the central bank has control over the price level. What is also needed is a solid fiscal framework that ensures that the fiscal authority follows sound fiscal policies which are sustainable at the going price level.
What is striking about this theory is that there is no room for the possibility of outright government default. This can be defended under the conditions of a fiat currency in a closed economy. But these narrow conditions have led to a controversial reception of the theory in the profession – after all, we can all cite examples where countries have defaulted outright rather than implicitly through price level adjustments. At the same time, this theory offered an early conceptualisation of a relevant insight that was well understood by the founding fathers of the Economic and Monetary Union in Europe: an independent central bank with a clear mandate of maintaining price stability should be supported by a fiscal framework which ensures sound fiscal policies.
A “monetary union without a fiscal union” perspective
To what extent can this theoretical apparatus be exactly applied to the special environment of a monetary union with a single monetary policy and many fiscal policies? The economic literature suggests that, in general terms, it remains valid, but with the requirements for ensuring monetary dominance becoming stricter.
The reason is that in a monetary union the single monetary policy establishes a link between otherwise separated budget constraints of the many fiscal authorities. This can be easily seen if one assumes that central bank incomes or losses from monetary policy operations are shared between governments before they are returned to them. Under this assumption, central bank purchases of government debt from any single member country, will in fact affect the budget constraints of all the other member countries.
The fact that in a monetary union without a fiscal union the single monetary policy establishes a crucial link between otherwise separated budget constraints of governments has been addressed in a number of thoughtful studies (like the ones from Bergin or Chari and Kehoe ). At the risk of over-simplifying their nuanced messages, these studies suggest that compared to a single economy, more binding constraints and institutional safeguards are required in a monetary union in order to protect a stability oriented monetary policy. In particular, monetary dominance needs to be credibly exercised against a large number of fiscal authorities. Consequently, this calls for a governance structure which not only endows the single central bank with a large degree of independence, but it should also entail a fiscal framework which recognises that sound fiscal policies in every single member country are a matter of common concern.
Central banks’ balance sheet structure
These issues can be further understood if one reflects on the structure of the central bank’s balance sheet in a monetary union compared with a single economy. To illustrate what I mean it is instructive to draw a comparison with the single economy case and carry out the following experiment: let us try to import to the euro area Marvin Goodfriend’s taxonomy where he contrasts monetary policy versus credit policy, which he applied to the United States. 
Goodfriend argues that genuine monetary policy exists only when the central bank purchases and sells Treasury securities. When the central bank engages in operations that carry credit risk, such as lending to the private sector, this is not monetary policy, but credit policy and, ultimately, fiscal policy. Goodfriend draws the conclusion that the Federal Reserve should avoid credit policies – in which it became heavily engaged when the crisis started – and stick to proper monetary policy based on confining itself to the Treasury market.
But the notion that Treasury securities do not carry credit risk – so that Treasury securities are treated as nominally risk-free from the central bank’s perspective – can only arise in an institutional framework in which there is full consolidation between the balance sheet of the central bank and the balance sheet of the fiscal authority. It is only in such a case that Treasury securities can become risk-free in nominal terms. The reason is that a central bank can consider as nominally risk-free a security if and only if the central bank itself is ready to guarantee its full and timely payment in cash in all states of the world.
In a monetary union without a fiscal union like the euro area, however (i.e. in an environment where governments do not pool sovereign risks), there is no outside risk-free asset from the point of view of the single monetary policy. The only risk-free assets in the euro area are the ECB’s own liabilities. All other assets including Treasury securities carry credit risk and can in theory fail to be converted at par into cash. Therefore, the ECB’s monetary policy always involves credit risk: private or public. There is no distinction on that front.
Implications for euro area governance
Applying these various insights to the euro area governance framework, it is clear that they provide a strong analytical framework and full backing to rationalise the institutional architecture in the Economic and Monetary Union. That is, the strong independence of the ECB with a deliberately focused mandate on maintaining price stability as a primary objective, which also receives protection from the prohibition of monetary financing; and a well-defined fiscal framework organised around the Stability and Growth Pact and supported by the no-bail out clause, which – absent a common risk-free asset - is intended to underpin medium-term sustainability.
However, it is also clear that the experience of the crisis has given us new information on how this framework operates. We have learned two key lessons in particular.
The first is that focusing only on monetary-fiscal interactions may be insufficient to ensure monetary dominance. A central bank may also find its choices constrained by the situation in the financial sector – so-called “financial dominance”. For instance, without a credible supervisory framework, a central bank may unwittingly end up in a position of financing counterparties that it thinks are illiquid but solvent – as it would be appropriate by its own statute - but transpire to be outright insolvent. Or, even worse, without an effective resolution framework, a central bank may find itself in a position of deliberately funding banks that are not viable because failing to do so may mean that the banks fail in a disorderly manner with systemic risks for the economy and ultimately, for price stability. The bank-sovereign nexus aggravates this captivity: a central bank may be prevented from tightening monetary conditions as would be otherwise appropriate, if it fears that, by doing so, banks may suffer losses and see their fragile health conditions undermined.
In all of these instances, the fiscal authority may not be the most proximate cause of the problems. We should think of this as a new dimension of risks to central bank independence and refer to this as financial dominance.
From an institutional perspective, this financial dominance problem has now been acknowledged by elevating governance of the banking sector to the European level through Banking Union. The Single Supervisory Mechanism reduces the risk of supervisors showing excessive forbearance towards insolvent banks. And the Single Resolution Mechanism, implementing the Bank Recovery and Resolution Directive, provides a framework for banks to safely fail. Moreover, to the extent that bank failures do still spill over to sovereigns and threaten their market access, the position of the ECB vis-à-vis governments is protected by the design features of Outright Monetary Transactions (OMT), where the requirement for an ESM programme preserves monetary dominance.
The second lesson from the crisis is that the governance framework, in particular on the fiscal side, did not perform as expected. In certain countries, government debt dynamics could only be stabilised with external assistance. The fiscal framework has of course now been strengthened through revisions to the Stability and Growth Pact and through the Fiscal Compact. But these new additions remain largely unproven.
Extended asset purchase programme in the context of monetary dominance
It is in this overall context that one needs to understand both the rationale for the ECB’s recent decision to expand its asset purchase programme (APP) to include purchases of public sector securities, and the specific framework within which they are implemented to take into account the institutional set-up in the euro area.
In terms of the rationale, the key point about this decision is that it reaffirms the central pillar of monetary dominance which is our commitment to our mandate.
We acted because we saw a real risk of not achieving our price stability objective over the medium-term. One may argue that we could have tolerated a deviation of inflation from our aim (namely to stabilise inflation around levels not far from 2% in the price stability range) for longer. Yet in my view this would have extended the notion of the medium term to temporal dimensions that would have been hardly justifiable on the basis of accountability to our objective. By making the medium term too elastic one inevitably increases the degree of discretion and even arbitrariness in delivering on the monetary policy mandate. As I have explained in more detail elsewhere, the mandate is meaningful only if exercised and assessed within reasonable time limits. 
Through acting, we also clearly demonstrated our independence and that we retain full control over our actions. Expanding our asset purchases at a critical juncture demonstrated that we were not dominated in any form. Indeed, the expanded purchase programme represents a continuation of the credit easing measures undertaken over the June-October 2014 period. The Governing Council considered that the outright purchases of public sector securities were the best instrument that could be activated on a sufficient scale to strengthen the measures already in place and deliver on our mandate.
Yet it is important to understand that our programme is also designed to take into account the specific institutional setup of the euro area, notably in terms of the fiscal set-up, and the implications that has on monetary dominance. Factors determining loss-sharing arrangements, the country portfolio composition as well as modalities upholding the prohibition of monetary financing, to name but a few, are crucial parameters that have been carefully chosen. The programme has deliberate safeguards to keep fiscal incentives in place and provide adequate protection for the central bank.
Let me emphasise four crucial elements.
First, purchases of public sector securities are subject to limited loss-sharing. Loss-sharing is reserved for those purchases that are done by the ECB or fall on securities issued by supranational European institutions, amounting in sum to 20%. The remaining 80% of all purchases – falling on purchases done by NCBs – will not be subject to loss-sharing. This mix corresponds approximately to the current allocation of fiscal responsibilities in the euro area, preserving thereby needed incentives for fiscal discipline of euro area governments.
Second, purchases of public sector securities are allocated across issuers from the various euro area countries on the basis of the ECB’s capital key. This fixed key respects that our public sector security purchases are guided by the ECB’s mandate to ensure price stability for the euro area as a whole. But it also effectively limits our degree of discretion towards individual governments, which in turn preserves incentives.
Third, purchase limits on the issue and issuer of public sector securities have been introduced to avoid a blocking minority for collective action clauses, as well as the Eurosystem becoming overly dominant in the sovereign bond market. In addition, the use of a blackout period averts any direct influence on the price formation in the primary market. These elements are important in the context of upholding the prohibition of monetary financing.
Fourth, the eligibility criteria for purchasing public sector securities are clear: securities are required to achieve the Eurosystem harmonised CQS3 rating scale. In other words, this means that we purchase as a rule only sovereign securities that are assigned an investment grade assessment by at least one external credit assessment institution. For those securities that do not reach the required rating, they can only become eligible if a rating waiver applies, which in turn requires the successful adherence to an EU-IMF adjustment programme. Similar to the logic underpinning OMT, this ensures that the central bank has an additional layer of protection vis-à-vis governments with potentially unsound policies.
Forward looking aspects
The strongest defence for monetary dominance, however, will be the success of our programme in strengthening economic activity in the euro area and bringing inflation back close to 2%. And looking at the impact of our monetary policy measures taken since the summer, according to the recent flow of data these measures now appear to be working their way through markets and the economy.
The nominal cost of bank borrowing for euro area NFCs has declined sharply with some convergence across countries. Indeed according to ECB staff analysis, the pass-through to lending rates in the second half of 2014 was faster than what would be implied by model estimates. Moreover, this was particularly the case for corporate lending rates in Italy and Spain. The cost of borrowing for euro area households is also showing further signs of a moderate decline as well as an increased convergence across countries. Over the course of last year, euro area households’ cost of borrowing for house purchases declined by approximately 50 basis points. While lending for house purchases was explicitly excluded from the targeted longer-term refinancing operation (TLTRO), we are seeing that those banks in more stressed economies which participated in last year’s two TLTROs have reduced their lending rates to households, albeit less notably than to NFC’s.
But the easing of price conditions for credit is only the first link in the chain that the summer measures have set in motion. The picture emerging from the January 2015 Bank Lending Survey attests that euro area banks are engaged in a sustained net easing concerning loans to NFCs and with a continued decline in the disparity across countries. This was mainly driven by the improved cost of funding - across all main market instruments- and balance sheet conditions. But unlike in the past, when improved funding conditions for banks were hardly passed along to their customers, the measures that we took over the summer have generated enough competitive pressures for banks to make their lending behaviour more responsive to the pace at which they see an improvement in their funding situation. Banks have started curtailing lending margins. As lending rates have been reduced in lockstep with the pace of reduction in market rates, demand for loans – once torpid – has been revived as well. Improvements in the net demand for loans to NFCs continued into the final quarter of last year across almost all euro area countries. Why? Because, by reducing the interest rate charged on the marginal loan, a bank makes the investment project that stand behind that marginal loan a profitable project. By reducing lending rates, more investments will break even, and more investors will ask for loans. Importantly, the increase in loan demand that we have observed was largely driven by the respondents financing needs for fixed investment and marked its first positive contribution since mid-2011. Looking ahead, the indications bode well for future economic activity, as euro area banks expect both a further easing of credit standards as well as a continued increase of net NFC loan demand.
Loan dynamics have also continued to recover, with the growth in credit to the private sector turning positive in December for the first time since mid-2012. Moreover, we have also seen a measurable improvement in indicators of business and consumer confidence. Overall, I think it is fair to say that we are seeing positive effects from the credit easing measures introduced last year.
In terms of our overall monetary policy tool-box, it should be kept in mind that the expanded asset purchase programme is just one element among a suite of others – including, for example, also OMT - which comprise the ECB’s comprehensive policy response to all contingencies which may arise in the future. The APP offers broad-based interventions to address base-risks to price stability for the euro area as a whole. In contrast, OMT has a clear country dimension and offers contingent interventions to address tail risks and extreme events.
The APP will contribute towards a sustained adjustment in the path of inflation towards 2%. But as monetary policy provides its contribution, other policy areas need to contribute decisively. Fiscal policy should support economic growth, while ensuring debt sustainability in compliance with the Stability and Growth Pact.. The Macro Imbalances Procedures should be applied strictly. And structural reforms should be implemented promptly and with determination. The overall success of the euro area depends on all stakeholders doing their job.
Let me conclude.
What I have outlined in my remarks today is that purchases of public sector securities in a monetary union that is not a fiscal union are a valid tool to overcome restrictions arising from the lower bound on monetary policy rates such that monetary policy can meet its price stability objective. The Governing Council decision has demonstrated that we are unconstrained in our ability to meet our mandate, and could make full use of all legal and effective monetary policy instruments. And the design of such purchases has taken into account the specificities of the institutional set-up.
This is an assertion of monetary dominance, in compliance with the principles enshrined in the Maastricht Treaty.
But the overall success of the euro area depends on all stakeholders doing their job. The ECB does not exist in a vacuum.
I would like to thank Leopold von Thadden and John Hutchinson for their contributions to this speech.
Sargent, T. and Wallace, N., Some unpleasant monetarist arithmetic, Federal Reserve Bank of Minneapolis Quarterly Review, Fall, 1-17, 1981.
Leeper, E., Equilibria under `active' and `passive' monetary and fiscal policies, Journal of Monetary Economics, 27, 129-147, 1991; Sims, C., A simple model for the determination of the price level and the interaction of monetary and fiscal policy, Economic Theory, 4, 381-399, 1994; Woodford, M., Monetary policy and price level determinacy in a cash-in-advance economy, Economic Theory, 4, 345-380, 1994.
Bergin, P., Fiscal solvency and price level determination in a monetary union, Journal of Monetary Economics, 45, 37-53, 2000; Chari, V. und Kehoe, P., Time inconsistency and free-riding in a monetary union, Journal of Money, Credit, and Banking, 40/7, 1329-1355, 2008.
Goodfriend, M., Central Banking in the credit turmoil: an assessment of Federal Reserve practice, Journal of Monetary Economics, 58, 1-12, 2011.
See my speech on monetary policy and balance sheet adjustment, Sintra, 27 May 2014.