At its monetary policy meeting on January 22, 2015, the Governing Council of the European Central Bank (ECB) unveiled an expansion of its asset purchase programme to include bonds issued by the public sector. Following a brief review of the motivations that led to this decision, I will discuss two main concerns that have been raised regarding the possible side effects of the programme.1
Motivations behind the expanded asset purchase programme
Inflation developments in the euro area
In 2014, the euro area economy struggled to recover at a satisfactory pace, unemployment remained high and inflation continued to fall: after a long series of downward surprises, in December the 12-month percentage change of consumer prices was negative in 11 countries as well as in the area as a whole; it did not exceed 1% in any single country.
Such developments indicate clearly that inflation was low and falling, not because of a virtuous adjustment of relative prices within the euro area, connected with efforts of the weakest countries to increase their competitiveness. Rather, it reflected a widespread trend. Moreover, the collapse of oil prices was only one of the reasons behind the falling inflation, with the weakness of aggregate demand also playing an important and increasing role.2
The continued decrease in inflation progressively affected expectations to an unprecedented extent, including during the financial crisis when the collapse in oil prices reached a level similar to that observed during the second half of last year. Since the summer of 2014, longer-term inflation expectations – that is, at horizons for which they should not be affected by current inflation – had also begun to decline substantially and had become increasingly sensitive to developments at shorter horizons. The fall in market-based long-term inflation expectations was particularly strong, signalling a concrete risk of de-anchoring from the target.3
Between the beginning of 2014 and the first weeks of 2015, the five-year, five-year-ahead inflation-linked swap rate fell from 2.2% to slightly below 1.5%, a level inconsistent with the ECB’s definition of price stability; inflation was not expected to return in line with the target before 2025. These developments were worrying, as the formation of expectations is not a linear process: big changes can materialise rapidly in a discontinuous manner.
The risks of too-low inflation for too long
Economic literature suggests there are several constraints in the decisions of households, firms and policy-makers that can become tighter when prices fall or even when, albeit remaining positive, inflation declines to an excessive extent. As a result, markets may not clear and the central bank is limited in its ability to keep inflation in line with the target.4
When the zero lower bound (ZLB) on nominal interest rates becomes binding, the real short-term interest rate is determined solely by inflation expectations. Since rational agents anticipate that the ZLB may limit the central bank’s action, medium- and long-term expectations may de-anchor from the target and real rates increase still further. In terms of the monetary stance, this is equivalent to an unwanted tightening of monetary policy conditions. In the euro area, the three-month real rate was -0.6% in January 2014; by the end of the year it had increased to -0.1% – notwithstanding a cumulative reduction by 20 basis points of the interest rate on the ECB’s main refinancing operations – due to the decline of inflation expectations. Negative short-term real rates are consistent with the still wide negative output gap in the euro area. In the last quarter of 2014, one-year real interest rates for the maturities up to four years ahead also increased as a result of this decline.
As private sector borrowing depends on the expected future real income and the value of the assets pledged as collateral, such a decrease in inflation – by increasing the debt service cost and reducing the collateral value – also raises the probability that credit constraints become binding and forces agents to deleverage, increase savings and reduce consumption and investment. Inflation would fall still further, raising the real burden of debt and hindering the deleveraging process. This is known as the “debt-deflation mechsnism”, after Fisher (1933).5
In relation to these considerations, there are two arguments I deem to be relevant when assessing the appropriate policy responses to falling inflation.
First, according to a common classification, deflation is considered ‘good’ or ‘bad’ depending on whether it is determined by a supply or a demand shock, respectively.6 But at the ZLB the nature of the shock is less relevant for its macroeconomic effects, as this classification ignores the fact that when the constraints described above are binding, the response of the economy to any given shock is modified. Consider a positive, temporary supply shock, such as a fall in oil prices. If the ZLB and the borrowing constraints are both binding, the induced fall in the general price level increases the real interest rate, which in turn depresses aggregate demand and raises the real cost of servicing debt. The result could be a contraction in aggregate demand, despite the more favourable supply conditions.7 The positive supply shock thus morphs into a negative demand one.
Second, deflation is not a necessary requirement for the constraints to become binding. Any substantial and sufficiently persistent disinflation may also increase their tightness. The negative effects just mentioned can materialise when inflation runs below expectations. Contractionary effects on consumption and investment plans are observed as soon as the real interest rate starts increasing, which at the ZLB does not necessarily require a decrease in the price level.
The need for more monetary accommodation
At the end of last year, our assessment within the Governing Council was that the measures we had launched in the second half of 2014 had not resulted in a sufficient increase of the Eurosystem’s balance sheet, due to the lower-than-expected out-turns in the first two targeted longer-term refinancing operations (TLTROs) and to a relatively modest contribution from the covered bond and, especially, the asset-backed securities (ABS) purchases. The size of the balance sheet was around €2.2 trillion, about €1 trillion (or 30%) below the peak reached in June 2012. The decline was driven by the gradual reimbursement by intermediaries of the funds borrowed under the two three-year, longer-term refinancing operations carried out in December 2011 and February 2012, and to the lower-than-expected contribution of the take-up in the first two TLTROs. During the same period, the Federal Reserve’s balance sheet had expanded by around 60%.
As a result, the transmission of the measures to the financing conditions of the economy proved weaker than expected. The credibility of the ECB, its main asset and the necessary condition for preserving price stability, was at serious risk; the Governing Council could not afford to adopt an attitude of ‘benign neglect’, also taking into account that market-based measures of inflation expectations were below the definition of price stability for the euro area over the next 10 years. A stronger response became warranted. After the discussion within the Governing Council in the November and December meetings, in January there was a broadly-shared view that the conditions were fully in place for taking additional monetary policy action.
The expanded asset purchase programme (APP), announced by the Governing Council on January 22, encompasses the existing programmes for asset-backed securities and covered bonds, as well as purchases in the secondary market of euro-denominated investment-grade securities issued by euro-area governments and agencies. The Governing Council unanimously viewed the programme as a legitimate tool of monetary policy and voted by a large majority to deploy it immediately.
The purchasing of sovereign debt was the only instrument through which the required increase of the Eurosystem’s balance sheet could be reached. Purchases of corporate bonds would not have done the job. The markets for these securities are much thinner and are concentrated in only a few countries where mainly large corporations, usually without funding problems, can easily issue bonds. The inclusion of these securities could have had distortionary effects across the euro area without determining a satisfactory degree of additional monetary accommodation.
The risks of public sector purchases
The APP is expected to affect economic activity and inflation through a variety of channels. It will lower the yields on public-sector bonds, which are benchmarks for the interest rates on a large set of financial instruments. The liquidity obtained in exchange for the securities sold to the Eurosystem should drive investors towards financial instruments that pay higher returns, thus transmitting the monetary policy impulse to a broader range of assets. The revaluation of financial assets would contribute to the adjustment of the private sector’s balance sheets, thus positively affecting its spending capacity. The banks’ cost of funding should fall, as a result of the reduced yields on the securities purchased, thus easing credit supply conditions. While the exchange rate is not a monetary policy objective, lower yields on euro-denominated assets will be reflected, other things being equal, in a lower euro, which would sustain exports and increase imported inflation and inflation expectations.
Worries about possible side effects of the public bonds purchase programme have, however, been raised within the Governing Council as well as by external observers. They concern, in particular, the risk of moral hazard by governments and the risk of generating financial instability. As for any other important policy action, these worries have been carefully considered in evaluating the balance of the benefits and the possible costs. All in all, my view is that the latter are less significant compared to the benefit of reducing the severe euro area macroeconomic risk and preserving the credibility of the ECB.
The risk of moral hazard
It has been argued that the purchase programme goes beyond the ECB’s mandate of price stability, blurring the distinction between monetary and fiscal policies. I do not agree with such a view. It is evident to all members of the Governing Council that monetary policy cannot solve every economic problem that affects the euro area. The programme has the very clear objective of bringing inflation back on target and therefore is fully within the ECB’s mandate. It is not intended to address fiscal policy issues, or to prevent a country from defaulting on its obligations.
More specifically, it has been suggested that the purchases of government securities may discourage or delay the adoption of structural reforms by national governments or may slow down fiscal consolidation. This worry has some merit, as experience during the run-up to the third phase of the Economic and Monetary Union has shown. Countries characterised by weaker fiscal conditions did not exploit the convergence of interest rates towards the low levels of the most ‘virtuous’ countries to put their own house in order, and we cannot exclude that the more favourable conditions also contributed to the deferment of the necessary adjustments.8
However, other aspects should also be considered. In the current cyclical conditions, the support of monetary policy to aggregate demand might play a relevant role in compensating for the possible short-run negative impact of structural reforms and for the effects of restrictive budgetary policies.
Empirical evidence suggests that structural reforms might have an adverse impact in the short run if they fail to boost confidence immediately. These short-term costs may also act as a political disincentive to the reform efforts. In this respect, the adoption of an asset purchase programme – which stimulates aggregate demand, reduces uncertainty and sustains confidence – can help the economy absorb these possible costs and maintain the necessary political drive and consensus on the need for reforming, thus making it more feasible.
Moreover, structural reforms and monetary policy can complement each other. Monetary accommodation may enable reforms, but the monetary stimulus is not in itself sufficient to strengthen the recovery and usher in sustained and lasting growth. The new monetary measures are designed to defuse the risk of slipping into a self-sustaining downward spiral of low inflation and economic stagnation. They cannot increase productivity or improve the structural features of the economy. Only the determined implementation of structural reforms can boost job creation and raise the euro area’s long-term growth prospects. This is not dependent upon the Governing Council; the responsibility lies entirely with national governments. The reduction in the overall macroeconomic risk that comes from the large monetary stimulus provides a great window of opportunity.
Similar considerations hold for fiscal sustainability. Several euro-area countries have excessively high debt-to-GDP ratios. Reducing them is indispensable. But this adjustment cannot be based only on additional fiscal consolidation. The speed with which the decline of the debt-to-GDP ratio can be attained depends crucially on the dynamics of the denominator, which also affects the numerator through its impact on the budget deficit. A rapid decline can only occur if economic activity quickly returns to a path of stable and sustained growth and inflation moves back in line with price stability. By contrast, an attempt to reduce the ratio only through a sharp fiscal correction, beyond what is needed to achieve the structural budgetary objective, could easily prove to be counter-productive through its negative effects on growth and inflation, as shown by counterfactual simulations carried out at the Bank of Italy for the Italian economy. Indeed, the fiscal multiplier depends on the cyclical phase. During recessions, budgetary measures can have indirect effects on the confidence and liquidity of households and firms, further exacerbating the contractionary effect of fiscal policy.9 As with the reform effort, a more accommodative monetary policy can also make the necessary fiscal adjustment more feasible.
The Governing Council addressed the moral hazard concern by opting for a partial risk-sharing of the purchase programme. The decision to let the balance sheets of the individual national central banks (NCBs) bear the entire risk of losses on the government bonds they purchase reflects the concern within the Council that full sharing might have led to unintended cross-country transfers of resources, without the approval of the competent bodies, and might have induced governments to relax their reform efforts.
In my view, full risk-sharing would have been more consistent with the concept of a single ECB monetary policy and with the Treaty. In this respect, the choice may be seen as a step backwards in our monetary policy-making. At the same time, given the different positions within the Governing Council and the importance of reaching a consensus on all decisions, it was a reasonable compromise. In any case, the risks in the NCB’s balance sheets will prove to be considerably lower if the programme succeeds in reducing the macroeconomic risks faced by the euro-area countries.
The risks of financial instability
Regarding the risks for financial stability, it is important to bear in mind the distinction between what monetary policy can and should do, and what is, instead, the domain of macro-prudential policy.10 In the euro area, monetary policy has the primary objective of maintaining price stability over the medium term. Macro-prudential measures should be used to limit the accumulation of systemic risk and to smooth the financial cycle in specific sectors or jurisdictions. Should any threat to financial stability materialise, specific macro-prudential measures should be implemented by national authorities to deal with local risks, without the need to alter the expansionary stance of monetary policy. In recent years, such measures have been adopted in connection with developments in the housing sector by the macro-prudential authorities of some euro-area countries (Belgium, Ireland and the Netherlands) and other European countries (Norway, Sweden and the UK).
Clearly monetary policy cannot remain expansionary forever. As the financial and the sovereign debt crises have shown, financial instability can have a strong impact on macroeconomic stability and hamper the transmission mechanism of monetary policy. Indicators of overheating should be considered and particular attention should be paid to credit developments in order to identify any possible build-up of broad-based imbalances in a timely manner. Currently no signs of such imbalances are visible in the euro area: credit is still languishing, assets do not generally appear to be overpriced and the overall propensity to take risks is still low. In any case, we should look at the potential effects of the programme on financial stability in the right perspective. The euro-area economy is still slowly recovering; economic stimulus can hardly be expected from fiscal policy, which is constrained; the risk that the economy will struggle to recover at a sustained pace is still high.
In the same vein, one should add that a risk can originate from regulatory policies that do not look beyond the single institutions and do not take a macro-prudential perspective, with potentially severe pro-cyclical (contractionary) effects. With an economic recovery that is still fragile and subject to downside risks and with inflation not yet on safe ground, we should be careful not to adopt excessively restrictive micro-prudential measures. Under the current circumstances, a macro-prudential perspective is of crucial importance.
A path to sustained inflation
Both price and financial stability risks in the euro area are mainly related to the persistence of an uncertain growth outlook and of too low inflation. Economic recovery, despite recent positive signals, still requires significant monetary accommodation to support aggregate demand and bring inflation back on target. The positive effects of the expanded APP we have seen so far in financial markets depend crucially on the expectation that it will be deployed to its maximum power until late 2016. The ECB’s Governing Council is firmly determined in its commitment to carry out purchases until a sustained adjustment in the path of inflation is observed.
At the same time, it is clear that the current situation, with nominal interest rates at their lower bound, cannot represent an equilibrium in the medium term and must be regarded as part of an ongoing adjustment process that will eventually bring us back to more customary figures for output growth, inflation and interest rates. This is why the programme must be firm, clear and well understood. Effective communication will play a crucial role, not only during its implementation, but also once the goal has been attained and a path of gradual normalisation of the monetary stance has become possible. This will be a new challenge for the Eurosystem, and a possible topic for another article, hopefully in a not-too-distant future.
Ignazio Visco is governor of the Bank of Italy.
1.This article draws on speeches I gave at the Second Annual Meeting of the Zell/Lurie Real Estate Center at Wharton-University of Pennsylvania (London, May 6, 2015) and at the symposium on “Economic growth, monetary policy and structural reform: challenges ahead for the United States and Europe” organised by the Bundesbank and the National Association for Business Economics-NABE (Munich, May 19, 2015). For their useful comments, I wish to thank Paolo Del Giovane, Giuseppe Ferrero, Stefano Neri, Alessandro Notarpietro and Francesco Spadafora.
2. See A Conti, S Neri and A Nobili, “Why is inflation so low in the euro area?”, Banca d’Italia, Working papers, forthcoming.
3. S. Cecchetti, F Natoli and L Sigalotti, “Tail comovement in option-implied inflation expectations as an indicator of anchoring”, Banca d’Italia, Working papers, forthcoming.
4. See M Casiraghi and G Ferrero, “Is deflation good or bad? Just mind the inflation gap”, Banca d’Italia, Occasional papers, 268, 2015.
5. See I Fisher, “The debt-deflation theory of great depressions”, Econometrica, 1(4), 1933.
6. See C Borio, M Erdem, A Filardo and B Hofmann, “The costs of deflations: a historical perspective”, BIS Quarterly Review, March 2015.
7. See S Neri and A Notarpietro, “The macroeconomic effects of low and falling inflation at the zero lower bound”, Banca d’Italia, Working papers, forthcoming.
8. See J. Fernández-Villaverde, L. Garicano and T. Santos, “Political credit cycles: the case of the Eurozone”, Journal of Economic Perspectives, 27(3), 2013.
9. See “The sensitivity of the macroeconomic outlook to variations in the profile of fiscal consolidation”, in Banca d’Italia, Economic Bulletin, 1, 2015, January.
10. See I Visco, “The challenges for central banks”, Central Banking, 25 (1), 2014.