Event with investors on the ECB. January 20th 2022 (notes)

On January 20th, I presented some thoughts about the ECB. My notes for this event are below. The numbers in the text refer to the slide presentation, also published in the blog.

 SLIDE 2

I will cover three issues, moving from the longer term to the medium and then the short term:

  1. Do we have to revise our view about the ability of central banks, including the ECB, to control inflation?
  2. What are the medium-term macroeconomic scenarios that could prevail after the impact of COVID would have dissipated?
  • What are the next likely policy moves of the ECB?

 SLIDE 3

  1. Do we have to revise our view about the ability of central banks, including the ECB, to control inflation?

Let me start with the longer-term issue.

The prevailing, and indeed mine all through the years of my long central banking career, view about the ability of central banks, including the ECB, to control inflation is that the central bank determines the medium-term inflation path and that exogenous shocks cause short term deviations around that path.

I have been harbouring doubts about this view for some time, which I find difficult to reconcile with three pieces of evidence.

 SLIDE 4

The first piece of evidence is in figure 1.

Figure 1. HICP price level and two trends (based on Goldman Sachs predictions)

Source:  ECB Statistical Data Warehouse; Data for December 2021 is preliminary.

Looking at the figure, the immediate conclusion would be that the ECB has been successful in the first, non-shaded period in the figure, ending around the beginning of 2014, in keeping inflation on its 2% objective, while it has failed it in the subsequent, 8 year long, shaded period, in which average inflation was about a half of the objective, too long to be consistent with the medium-term orientation of the ECB. At the end of the line, you find today’s sharp price level increase, which I will cover in a minute.

 SLIDE 5

Partitioning the life of the ECB so far along with the tenure of its 4 presidents, as in figure 2 below, one would have to conclude that the first 2 presidents have been much more successful than the third and the fourth in keeping inflation, on average, at 2%. Also, in terms of deviations from the 2.0% norm, the record of Duisenberg and Trichet would appear to be better than that of Draghi and Lagarde. However, this is not a conclusion one could really reach without a thorough assessment of the very different conditions in which the 4 presidents exercised their mandate, which is not something for today.

Figure 2. Rate of consumer inflation during the tenure of the first four ECB Presidents

Source: ECB Statistical Data Warehouse; Data for December 2021 is preliminary.

SLIDE 6

The second piece of evidence is in figure 3, developed by my colleague at Bruegel, Zsolt Darvas.

FIGURE 3. Actual core inflation and ECB inflation projections.

Source: updated by Zsolt Darvas, Bruegel from Darvas, Z. (2018). Forecast errors and Monetary Policy Normalisation in the Euro Area, Bruegel Policy Contribution, Issue nº 24, December. https://www.bruegel.org/2018/12/forecast-errors-and-monetary-policy-normalisation-in-the-euro-area/.

The figure shows actual inflation in the thick line, and ECB inflation projections, in the lines that depart from it. The inevitable conclusion from the figure is that the ECB systematically overestimated inflation, projecting that it would move towards its objective.

In sum, the ECB in the last few years not only missed its objective but also systematically overestimated the “potency” of its policy to reach its inflation objective.

SLIDE 7

The third piece of evidence is from the recent book, published by a number of economists led by Massimo Rostagno, especially their figure 6.22, reproduced here as figure 4.

Figure 4. Estimated effects of extraordinary monetary policy on real GDP growth and inflation, 2014-2018.

Source: ECB calculations. Notes: The distance between the actual data (solid line) and the counterfactual path in the absence of ECB’s policy measures is the estimated impact of those measures (dashed line). The impact is disaggregated into the contribution of the individual measures (coloured bars). The results are based on a BVAR.

The reading of this figure by the ECB authors is as follows: We find that, in the absence of the package, GDP would have been roughly 2.7 lower by end-2018, and annual inflation one-third of a percentage weaker on average over 2015-2018. 30 basis points of average annual additional inflation have been precious, given the behaviour of the economy. Still, after 5 years of forceful monetary easing, they look quite modest and, obviously, not enough to reach the 2% holy grail: a meagre result for the gigantic effort deployed by the ECB. Contrary to the interpretation given by the authors, I find this evidence as a sign of the limited effectiveness of monetary policy in steering inflation rather than as a sign of its potency.

Of course, we can find explanations for these disappointing results. The book by Rostagno and other identifies one in a series of deflationary shocks. Another is that unconventional measure may be less efficient than interest rate changes, which were no longer possible when the effective lower bound was reached. The two explanations would lead to thinking that the effectiveness of monetary policy in controlling inflation is asymmetric: more potent in avoiding too high inflation, but much weaker in raising inflation towards the objective in deflationary conditions.

The inflation situation has radically changed since the summer of last year. Instead of being too low, inflation has been much higher than the objective at 5%, the highest level recorded since the introduction of the €. But recent developments don’t help me get rid of the disturbing thought that central banks control and forecast of inflation is less precise than we want to think. The ECB was clearly surprised by recent inflation developments, which have been more dynamic and sustained than it thought: the 2022 projection has been more than doubled, from 1.5 to 3.2 % between September and December of last year and projections for 2023 and 2024 have also been increased, albeit less dramatically. This raises the question of the right approach in looking at monetary policy: do exogenous shocks disturb the medium-term inflation path determined by the central bank, or does the main inflation story occur outside the central bank?

Let me be very clear here: I am not criticising the ECB. I agree with the predominant view that it is an institution with an effective and smooth governance system that serves the €-area well and is endowed with excellent leaders and staff. I am raising a more fundamental issue regarding the relative role of monetary policy and other factors, e.g. demographics (Goodhart), in determining the medium-term inflation path.

I leave you with this doubt. It does not immediately lead to operational conclusions, but taking it into account can help take the right decisions, especially in currently unsettled conditions. The practical advice is: don’t be fixed on monetary policy in thinking about inflation; broaden your perspective, admittedly at the cost of a more complex assessment.

SLIDE 8

  1. What are the medium-term macroeconomic scenarios that could prevail after the impact of COVID would have dissipated? 1)This section draws on a post jointly written with Pauline Weil and Monika Grzegorczyk, Is the risk of stagflation real?, https://www.bruegel.org/2021/11/is-the-risk-of-stagflation-real/

In thinking about the future of monetary policy, it is useful to figure out which are the plausible macroeconomic scenarios going forward.

Stagflation seems to be the new worry in Europe and several other advanced economies. The experience of the 1970s supply-side shocks, raising inflation and pushing down growth, is actively revisited. At the October press conference, ECB President Lagarde had to reassure listeners that no such thing was on the horizon. Isabel Schnabel did the same with more details in a more recent speech (November 17th), which is worthwhile reading.

So far, we had the “stag” component in 2020, with a fall of real income by more than 6 % in the euro area, while we are having now the “flation” component, with consumer price inflation at 5 per cent, but not the two at the same time, as stagflation would require. Similar experiences are taking place in other advanced economies. The interesting question is what could happen in the medium term, say over the next 2-3 years: could the “stag” and the “flation” components happen contemporaneously and persist for some time?

More generally, on which path will the euro-area economy settle after the Covid shock?

In principle, three main scenarios are plausible:

  1. Move to higher growth and inflation path, say 2.0-3.0%
  2. A return to the pre-Covid situation, with both EU growth and inflation at 1.0-2.0%
  3. A move to stagflation, where growth and inflation no longer move in the same direction, say with inflation higher than 3.0% and growth lower than 1.0%.

The first scenario is, under some conditions, the most attractive. If central bank control of inflationary forces is more effective than control of deflationary trends, as per the point above about the asymmetry in inflation control, the ECB could control inflation by tightening monetary policy while avoiding a recession. The second scenario is lacklustre, but not worse than what we had in the last decade or so. If my doubts about the low potency of monetary policy to fight too low inflation come true, this situation could last for long, with the ECB continuing in its attempt to reach 2 per cent before eventually realising that this is not possible. The third scenario is the most worrying, with two “bads”, high inflation and low or negative growth, and the central bank in a dilemma on which “bad” to fight.

To understand what the ECB could do, it is useful to ascertain which medium-term macroeconomic scenario is built in the forecasts of official institutions, private forecasters and, as much as possible, into market prices.

SLIDE 9

Before coming to this, let me give you my own view of the central scenario, which is a variant of the second one (return to pre-Covid), where growth is indeed between 1 and 2 per cent, while inflation is somewhat higher than 2 per cent. Indeed, I believe that there is some justification for the concern on the inflation front, but not for the contemporaneous stagnation.

 SLIDE 10

In December, consumer prices continued to accelerate, reaching 5.0% (Figure 5). This is the fastest growth observed since the beginning of the euro in 1999.

Figure 5: Consumer prices in the euro area 2000-2021 (Not Seasonally Adjusted)

Source: European Central Bank Statistical Data Warehouse

 SLIDE 11

As gathered from market prices, a summary view of expected inflation can be gauged from looking at ILS inflation, as in figure 4. While the fixed leg of inflation swaps with a five-year maturity prevailing over a five-year horizon dipped below 1.0 per cent at the beginning of the Covid recession, it is now close to around 2.0 per cent, not far from the current level of core inflation at 2.6%. This is consistent with the financial market expecting energy increases and other special factors to be temporary.

Figure 6. Break-even inflation is derived from inflation swaps.

Source: Bruegel based on Bloomberg

Source: Bloomberg, Refinitiv, ECB calculations. Notes: HICPxT refers to HICP excluding Tabacco. Latest observation: September 2021 (HICPxT)

Overall, my central expectation is that, indeed, inflation over the medium term, say around 2023-2024, will not be far from the 2% objective but a bit higher and thus higher than the level expected by the ECB. However, I hold this expectation with limited confidence as I cannot exclude less favourable prospects. These less favourable prospects could be generated by the efforts to green the €-area economy. I noted with interest another speech by Isabel Schnabel on the 8th of January, in which she advanced the doubt that the need to make brown energy more expensive on a sustained basis to reach the desired effects on climate could exert upward pressure on inflation on a persistent basis. Indeed, it could be the case that energy price increases are not temporary but quite persistent, unlike in the past.2)L. Cadamuro and F. Papadia, Does money growth tell us anything about inflation? https://www.bruegel.org/2021/11/does-money-growth-tell-us-anything-about-inflation/

In conclusion, I see no imminent alarm on inflation, considering the muted wage developments so far. However, I still stand by what I wrote a couple of months ago: “Attention should be paid to a possible sequence of negative events: if inflation would start to be volatile and money growth remains high, efforts to control inflation could be undermined.”

SLIDE 12

Regarding the “stag” part, the framework that seems most useful to examine the post-Covid scenario is the one developed to assess the hysteresis phenomenon. In its initial formulation, developed by Summers and Blanchard 3)Olivier J. Blanchard and Lawrence H. Summers, Hysteresis and the European Unemployment Problem, NBER Macroeconomics Annual 1986, Volume 1. Stanley Fischer, editor. MIT Press. 1986., the idea was that recession could have permanent, or at least persistent, effects on potential output and that determined macroeconomic easing is needed after a recession to recover from the lingering adverse effects on employment.

I use that framework here without any prior about the consequences of the COVID recession, more as a scheme for interpreting developments than as a device to test a hypothesis.

Figure 7 reports the (log of) real GDP in the euro area between 1996 and June 2021. The line is then extrapolated to 2024 based on Goldman Sachs forecasts (estimates from this investment bank are close to those of other forecasters, so choosing one or the other source does not make much difference).

Two trends are identified in the behaviour of real GDP: a first one prevailing between 1995 and the beginning of the Great Financial Crisis in 2008, a second flatter one, starting between 2008 and 2009.

The Great Financial Crisis conforms well to a case of “super-hysteresis”: not only the dent in GDP caused by the recession is not recovered, but also the rate of growth after the recession is persistently lower4)Two hypotheses have been put forward to explain the low rate of growth after the Great Financial Crisis: the “debt supercycle”, of K. Rogoff, Debt Supercycle, Not Secular Stagnation, in Progress and Confusion, O. Blanchard, R. Rajan, K. Rogoff and L. Summers editors. MIT Press. And the secular stagnation, that Summers restated, following Hansen: A. Hansen, Economic progress and declining population growth (speech for AEA meetings 1938), American Economic Review, March 1939. Vol XXIX, NO. 1, PART I. L. Summers, Rethinking Secular Stagnation after Seventeen months. In Progress and Confusion, O. Blanchard, R. Rajan, K. Rogoff and L. Summers editors. MIT Press.. This is, instead, not the case for the Covid recession, unless the available forecasts for the next three and a half years are badly wrong: GDP is forecasted to return in a couple of years very close to the pre-Covid path: so, neither super-hysteresis nor hysteresis seems to prevail. The difference between the consequences of the two recessions is not surprising: the first one was caused by endogenous disturbances generated in the financial sector, the second was fully exogenous and thus likely to be surpassed when the external shock dissipates.

Figure 7. GDP, Euro area (constant prices)

Source: Is the risk of stagflation real? Bruegel post. November 2nd 2021.

I find convincing the idea that still in 2022, barring negative news about COVID because of omicron, growth would be quite higher than the potential at 4 to 5%, before settling down to a more normal 2% or lower after that. 4.5% is indeed what I answered to economists’ annual Financial Times poll.

So, to sum up, I regard the risk of stagflation as remote: the joint occurrence of very low or negative growth and sustained inflation looks improbable to me. I am less confident on the “flation” part; I am less sanguine than the ECB about inflation being lower than the ECB objective in 2023 and 2024. In addition, inflation could bring negative surprises going forward, even if this is not my central scenario.

SLIDE 13

Table 1 summarises all the information gathered to conclude which of the three scenarios presented above are official forecasters, private forecasters and the market converging. To complete the harnessing of information about real future developments, we would need an indication of which rate of real growth is implicit in financial market prices. Not finding an indication for this variable, we leave the relative cell void in table 1.

Table 1: Medium-term scenarios – growth and inflation for the euro area (green – upgrade, orange – downgrade)

Source: updated to December 2021 from, Is the risk of stagflation real? Bruegel post. November 2nd 2021.

The overall reading of the table is that all projections converge to the second scenario mentioned above, i.e., a return, in the medium-term, to the pre-Covid situation regarding both growth and inflation.

As far as one can read into them, neither official institutions nor private forecasters, nor financial markets foresee either stagflation or higher growth and inflation.

However, this could be a case of extended group thinking, with different forecasters converging to the same view. A collective error would not be so surprising given that forecasts are built with close to no relevant prior experience of a pandemic caused by the recession. Vigilance is thus needed from the ECB, fiscal authorities as well as observers to detect signs that the situation is developing in different directions than expected.

SLIDE 14

As I said, I agree on the little likelihood of stagflation and the growth part, while I see inflation a little higher.

SLIDE 15

  • What are the next likely policy moves of the ECB?

A recurrent topos in documents from economic institutions is to lament the exceptional uncertainty prevailing at the time of writing. Of course, since the complaint is so frequent, the uncertainty cannot always be so exceptional as it is pretended to be. Still, if there is a time when the observation is true, this is today. Recent economic developments have been practically without precedents, and economic forecasts are fundamentally based on precedents. To put it bluntly, you could say that projections are guesses more than forecasts. Omicron is the last uncertainty factor, but there was large uncertainty even before: economic activity has surprised on the upside, both in the US and in Europe, in 2021, but momentum is somewhat fading, and there are fears that the scares of the pandemic will affect supply for years to come. Inflation is higher than it has been in years, even much higher in the United States, Germany and the €-area, but central bankers, most international institutions and economic observers expect a move back to more subdued levels. We have the big progress of the Next Generation EU, but there is, or at least I have, some anxiety about its implementation.

SLIDE 16

The next policy actions of the ECB depend on how these uncertainties will be solved, in particular regarding inflation developments.

I have given you my, weakly held, central scenario of a return, starting next year, to the pre-covid situation as regards growth, between 1 and 2 %, while inflation could be somewhat higher than 2 %.

If this expectation consolidates in the course of 2022, I think net purchases will be stopped towards the end of this year, probably followed by a small increase in the deposit rate facility still in 2022. I notice here, as an aside, that the 1.8 % inflation projected by the ECB for 2023 and 2024 would indeed be consistent with the 2% objective once Owner-Occupied Costs of 0.2-0.3 % are added, as the ECB mentioned in the last strategy review. This point was made at the latest press conference, but Lagarde ignored it, and Lane did not mention it in his recent interview with an Italian financial newspaper. I also believe that the ECB would somehow move in its forward guidance towards YCC a la Bank of Japan. Not in the sense that it would target a specific level for a long-term yield with its purchases, but rather that it would give an indication that purchases could be re-started if either aggregate rates (German yields and OIS swaps) or spreads between government securities would move away from levels consistent with macroeconomic developments or threaten the homogenous transmission of monetary policy across the €-area. Rate increases would continue in 2023, finally moving away from negative rates.

SLIDE 17

You may wonder whether the debt conditions of some €-area countries would not be an obstacle to this gradual normalisation of monetary policy. You may have in mind the case of Italy. I don’t have a definitive view on this. I can, however, make a few relevant observations, which go in the reassuring rather than the worrying direction:

  • First, it is unlikely that the ECB would refrain from doing what it needs to do to control inflation in the aggregate because of the fiscal situation of a member country,
  • Second, there are two strong incentives for Italy to do what is needed to maintain debt sustainability and adequate market access, namely bond vigilantes and the control exercised by the EU Commission on the reform engagements taken in the framework of the Next Generation EU program,
  • Third, there is a protection of the fiscal situation of Italy, given by the combination of fairly long maturity of sovereign debt and the exceptionally low level of yields from which any increase could start.

Overall, my central scenario is somewhat tighter than what the ECB, in particular Lane, is putting forward.

SLIDE 18

Of course, if the fear about brisker inflation would materialize, the tightening of monetary policy would be accelerated, the stop to APP purchases could come earlier than the end of the year, and the prospect of an interest rate increase 2022 would become more likely. The reverse would be true if inflation would be seen as decelerating below 2.0% in 2023, as foreseen by Lane. Then the expectation of Lane and Lagarde of no interest rate increase in 2022 would be confirmed while net purchases could continue until the end of this year and possibly for a few months more.

 

References

1 This section draws on a post jointly written with Pauline Weil and Monika Grzegorczyk, Is the risk of stagflation real?, https://www.bruegel.org/2021/11/is-the-risk-of-stagflation-real/
2 L. Cadamuro and F. Papadia, Does money growth tell us anything about inflation? https://www.bruegel.org/2021/11/does-money-growth-tell-us-anything-about-inflation/
3 Olivier J. Blanchard and Lawrence H. Summers, Hysteresis and the European Unemployment Problem, NBER Macroeconomics Annual 1986, Volume 1. Stanley Fischer, editor. MIT Press. 1986.
4 Two hypotheses have been put forward to explain the low rate of growth after the Great Financial Crisis: the “debt supercycle”, of K. Rogoff, Debt Supercycle, Not Secular Stagnation, in Progress and Confusion, O. Blanchard, R. Rajan, K. Rogoff and L. Summers editors. MIT Press. And the secular stagnation, that Summers restated, following Hansen: A. Hansen, Economic progress and declining population growth (speech for AEA meetings 1938), American Economic Review, March 1939. Vol XXIX, NO. 1, PART I. L. Summers, Rethinking Secular Stagnation after Seventeen months. In Progress and Confusion, O. Blanchard, R. Rajan, K. Rogoff and L. Summers editors. MIT Press.
Share