Reflation, not stagflation
Speech by Isabel Schnabel, Member of the Executive Board of the ECB, at a virtual event organised by Goldman Sachs
Frankfurt am Main, 17 November 2021
The recovery from the pandemic has been exceptional on many accounts.
In the euro area, it has been faster and stronger than any recovery since the Second World War. It has also been bumpier than most previous recoveries, with recurring phases of accelerating and decelerating growth. And it has been characterised by an unusually distinct mismatch between supply and demand, causing the prices of many products, including energy, to rise measurably.
With the recovery now in a more advanced stage, and with growth momentum moderating, some observers see current developments as raising the spectre of stagflation, reminiscent of the 1970s when a surge in oil prices stifled growth and pushed inflation levels up into double digits across most oil-importing economies.
In my remarks today, I would like to explain why fears of a return of stagflation appear unfounded. My reasoning will rest on two broad arguments on why comparisons with the 1970s may be misleading.
The first argument discards the risks of stagnation, pointing to the factors that are driving the current slowdown in economic activity. The second dissects the risks of inflation, highlighting the lessons central banks around the world have drawn from the inflationary episode of the 1970s.
The euro area growth outlook: no signs of stagnation
Let me start with the risks to economic growth – that is, with the spectre of stagnation.
In the second and third quarter of 2021, euro area real GDP expanded at an annualised rate of nearly 9% (Slide 2, left-hand chart). The unemployment rate declined from 8.1% in April to 7.4% in September – the same level as in February 2020, before the pandemic hit the global economy (Slide 2, right-hand chart). Over the same period, the number of workers in furlough schemes fell from 6.6% to 1.5% of the euro area labour force.
All in all, the pace of recovery over the summer was extraordinary and indeed much faster than many of us expected, thanks in large part to the marked rise in the number of people who are fully vaccinated.
Incoming data over the past few weeks, however, suggest that economic activity globally and in the euro area has started to moderate. This did not come as a surprise. Given the exceptional pace of expansion over the summer, a moderation was to be expected. Even a visible slowdown would still leave growth at historically high levels.
But there are concerns that the slowdown may be more abrupt and more enduring as the headwinds of higher energy prices and more persistent supply chain disruptions continue to feed through the global and euro area economies.
Our analysis suggests that these concerns are largely unwarranted.
Consider the manufacturing sector.
Supply delivery times remained close to record highs in October, with few signs of imminent easing in sight (Slide 3, left-hand chart). Our most recent telephone survey of non-financial firms showed that over 30% of companies now expect supply constraints to last for another year or longer (Slide 3, right-hand chart).
These supply chain disruptions, in turn, have contributed to a marked fall in inventories across all industries, leading to production slowdowns and longer wait times for consumers.
But supply bottlenecks, by their very nature, do not diminish growth potential. They merely shift activity over time. Port congestions cannot last indefinitely. In a competitive economy, supply will eventually catch up with demand, even if it takes longer than previously expected.
More importantly, supply-side constraints only explain a fraction of the current delay in delivery times. According to ECB staff estimates, about two-thirds of longer global lead times currently reflect exceptionally strong demand outpacing pre-pandemic supply (Slide 4, left-hand chart).
Contrary to previous recessions, euro area household balance sheets have been largely unscathed by the pandemic, thanks to the unprecedented fiscal and monetary policy support (Slide 4, right-hand chart). By the end of the second quarter of 2021, households had accumulated an estimated amount of more than €800 billion of savings in excess of their pre-pandemic level, or 7% of pre-pandemic euro area GDP.
So, contrary to the 1970s and much of the past decade, we are now facing a macroeconomic environment characterised mainly by deficient supply rather than deficient demand. Supply bottlenecks are tempering what has become one of the strongest demand shocks in recent history.
Companies operating at the earlier stages of the value-added chain are producing at full capacity, while firms at the final stages are facing an unprecedented wedge between the amount of orders on the one side and production activity on the other.
In September, industrial orders in the euro area were more than 10% above their pre-pandemic level, whereas production was slightly lower than before the pandemic (Slide 5, left-hand chart). Expected capacity utilisation, taking into account the level of order books, is at its tightest since 2007 (Slide 5, right-hand chart).
The implication is that there is a large potential for demand to sustain growth at elevated levels in the foreseeable future as supply capacity catches up with demand and bottlenecks are gradually resolved.
The strong rise in oil prices is unlikely to change this picture fundamentally. To see this, it is worth looking back at the 1970s.
When the first oil price shock hit the global economy in 1973, roughly one barrel of oil was required to generate USD 1,000 of GDP in 2010 prices (Slide 6, left-hand chart). Today, less than half that amount of oil is needed to generate the same level of output.
In other words, changes in consumer preferences, improvements in energy efficiency as well as the decline in the growth contribution of energy-intensive industries all imply that rising oil prices hurt growth and the economy much less now than they did in the past.
ECB analysis corroborates this view. It finds that the impact of an oil supply shock on global industrial production may be about half of what it was in the 1980s and 1990s (Slide 6, right-hand chart).
The ongoing green transition can be expected to further strengthen the economy’s resilience to oil price shocks. New research finds that green investments are not dampened by higher oil prices.
In addition, many governments are currently imposing tax cuts, price caps or rebates to shield the most vulnerable households, which typically spend a substantial share of income on energy, from the sharp rise in gas, fuel and electricity prices.
Economists here at Goldman Sachs estimate that such mitigating measures, in conjunction with a generally low pass-through from wholesale to retail energy prices, will substantially limit the economic impact of the current energy price shock.
Survey data point in a similar direction, suggesting that consumers remain upbeat about economic prospects.
Neither their assessment of the general economic situation over the next 12 months, nor their intentions to make major purchases have so far taken a visible hit from the surge in energy prices (Slide 7, left-hand chart). Survey data also point to further improvements in the labour market, supporting the prospect of higher incomes and increased spending.
There are two further reasons to be optimistic about future growth expectations.
The first relates to fiscal policy. Unlike over much of the past decade, fiscal policy will remain supportive of growth as the recovery matures. In the euro area, broader discretionary fiscal stimulus is expected to amount to more than 1.5% of GDP in 2022 and around 1.25% in 2023. The EU recovery fund will boost public investment over the coming few years.
The second reason for optimism concerns the services sector.
Contact-intensive services have rebounded sharply since the spring and have remained an important source of growth since then. But activity in these sectors has not yet fully returned to its pre-crisis level. By the end of the third quarter, value added in contact-intensive services was still approximately 8% below its pre-pandemic level (Slide 7, right-hand chart).
New or existing containment measures and labour shortages mean that activity will only resume gradually. But as the pandemic abates and economies reopen further, activity in more contact-intensive services is likely to keep expanding.
All in all, stagnation looks like a distant threat – a view that is shared by most of the forecasting community.
According to Consensus Economics, revisions to expected growth have been moderate over the past few weeks, in particular for the euro area (Slide 8). Real GDP in the euro area is expected to grow by 5% in 2021 and by 4.3% in 2022, well above estimated potential growth.
The euro area inflation outlook: rising uncertainty favours a risk-management approach
Let me now turn to the inflation part of the stagflation concern.
The first oil price shock hit the euro area economy in a strong inflationary environment. In 1972, before the shock, euro area inflation was already at 6.3%. In 1974, it rose to 13.2%. Wage growth stood at 12% in 1972 and increased to around 18% in 1974.
The monetary policy response to this spiral of rising prices and wages was insufficient to secure people’s purchasing power. The real short-term interest rate fell into deep negative territory in 1974 and remained negative for the next four years despite inflation hovering at double-digit rates.
As a result, expectations of high inflation became embedded in the minds of employers and employees, making it significantly costlier in terms of lost output and higher unemployment when central banks finally picked up the fight against inflation in the early 1980s.
The experience of this episode dramatically changed the conduct of monetary policy in the following decades. It forged a consensus among policymakers and academia on two fundamental pillars of modern central banking.
First, there is no stable, long-run trade-off that monetary policy can exploit to permanently lower unemployment at the expense of modestly higher inflation. Hence, the best contribution that central banks can make to growth and welfare is to maintain price stability.
Second, central bank independence is important for safeguarding stable prices. Today, central banks in almost all advanced economies and many emerging market economies are politically independent.
The “Great Moderation” that took off in the 1980s built on these two pillars – a narrow, well-defined mandate and central bank independence. Over time, a clear commitment to price stability has anchored inflation expectations at very low levels.
Maintaining this level of trust is what will guide the response of central banks to the challenges we are facing today.
For the first time in more than a decade, we are witnessing measurable upward pressure on inflation. In October, harmonised inflation in the euro area increased by 4.1% compared with a year earlier, markedly above our two per cent target (Slide 9, left-hand chart). In November, we expect inflation to hit its highest level since the euro was introduced in 1999.
There is broad agreement on the main factors that are currently pushing up inflation: rising energy prices, shortages of goods, equipment and labour, as well as statistical base effects.
Energy inflation, for example, accounted for more than half of total HICP inflation in October (Slide 9, right-hand chart). Measurable increases in the prices of contact-intensive services, such as restaurants and travel, explain the lion’s share of the rise in service price inflation, while supply bottlenecks, as well as growing producer prices, have led to the fastest increase in durable goods inflation in October on record (Slide 10). And significant base effects – in particular from the reversal of the VAT cut in Germany – are exerting a strong upward impact on measured inflation.
There is less agreement on the persistence of many of these price pressures, and what they mean for the appropriate response of monetary policy.
At our last Governing Council meeting, we concluded that, in spite of significant uncertainty, there is still good reason to believe that euro area inflation will decline visibly over the course of next year and gradually fall back below our target of two per cent in the medium term, meaning that the conditions for raising interest rates, as set out in our forward guidance, are very unlikely to be met next year.
Financial market participants, too, expect a considerable and lasting decline in inflation in 2022 (Slide 11). There is no evidence that markets, or professional forecasters, expect runaway inflation, or inflation dynamics even remotely similar to what we observed in the 1970s.
At the same time, investors are increasingly challenging the view that the euro area economy will fall back into the vicious circle of low growth and low inflation that characterised the pre-pandemic macroeconomic environment. Current forward inflation swap rates indicate that inflation is expected to stabilise at, or close to, our two per cent target over the medium to long term, well above the level that was expected before the pandemic.
It is therefore not surprising that uncertainty around the future path of short-term interest rates as priced by investors has increased over the past few weeks (Slide 12). Such a rise in uncertainty and risk premia is to be expected when the market considers it increasingly likely that inflation will meet the conditions for lift-off, without putting into question the credibility of our forward guidance.
Rising risk premia rather suggest that we are facing an extraordinary degree of uncertainty around the inflation outlook. Since the start of the year, we – just as other forecasters – have repeatedly underestimated inflation in our central scenario, and we have done so by a significant margin (Slide 13). Elevated inflation is now expected to last for longer than we previously thought.
Forecasting errors are nothing unusual. Neither central banks nor financial markets have been particularly good at forecasting inflation over the past decade. But when uncertainty is particularly large, as it is today, there are benefits in applying what Alan Greenspan once called a “risk-management approach” to monetary policy.
At its core, this approach builds on the notion that central banks should not only consider the most likely future path of the economy, but also the entire distribution of risks around that path with a view to keeping sufficient optionality to address all inflation contingencies.
The current distribution of risks as priced by financial markets shows why such an approach may be particularly appropriate at present.
Option-implied probabilities suggest that − within the data sample that is available to us, starting in 2010 − the market is currently putting the lowest weight ever on inflation being below 1% over the next five years, and the highest weight in more than a decade – above 50% – on inflation being above 2% (Slide 14, left-hand chart).
Our forward guidance provides protection against the risks of too low inflation by setting clear conditions that prevent a premature tightening. A full risk-management approach ensures that policy also remains sufficiently robust when the risk distribution is increasingly skewed in the other direction, so as to ensure that inflation expectations remain firmly anchored at our target.
At present, upside risks to inflation largely stem from two sources.
First, a too narrow focus on the supply side of the economy. By all likelihood, the output gap in the euro area will turn positive next year and will be significantly positive in 2023 and beyond, with sustained demand possibly putting more pressure on underlying prices and wages than currently foreseen.
Over the summer, price pressures already broadened significantly. In September, the prices of more than 40% of the goods and services of the HICP excluding food and energy increased at a rate faster than 2% (Slide 14, right-hand side). Such a share was last observed in 2013.
Such broad-based increases, if persistent, and in conjunction with a more protracted period of supply and demand mismatches, could raise the risks of second-round effects.
Second, the global and euro area economy are currently undergoing profound structural change, also as a result of the pandemic.
On the one hand, this indicates a significant need to reallocate resources from one sector to another, implying that labour shortages, in particular skilled labour, and associated wage pressures may start to emerge even before aggregate labour market slack vanishes.
On the other hand, deep-seated changes mean that many of the historical regularities that inform our decisions may prove less reliable.
Take energy and the green transition as an example.
Oil and gas prices are unlikely to continue rising at the extraordinary pace we saw over the summer. In our baseline scenario, lower energy prices will significantly help ease inflation over the course of next year. But it is far from clear whether energy prices will decline as rapidly as current futures prices, which underpin our projections, would suggest.
Specifically, the recovery phase has been characterised by an atypically slow response of US shale oil production to rising oil prices (Slide 15). With the green transition in full motion, such investments may no longer prove profitable over the medium term, at least not to the same extent as they did in the past, keeping supply constrained and prices high.
Eventually, if we are serious about fighting climate change, the green transition will need to bring about a measurable further rise in the price of carbon, which can be expected to lead to higher fuel and electricity prices.
These effects may become entrenched in expectations if people start to anticipate them.
Similarly, some of the bottlenecks we are observing today may not only reflect the reopening of our economy but also structural forces, such as the secular shift to electric vehicles. Adapting supply capacities to such shifts in demand can take several years.
In other words, if some factors that are by nature temporary – such as supply and demand imbalances – are stronger and last longer, then inflation may become more persistent and broad-based.
So, all in all, and with this I would like to conclude, the global and euro area economy are, without doubt, significantly better placed today to prevent a recurrence of the adverse consequences of the oil price shocks of the 1970s and 1980s. Reflation, not stagflation remains the defining theme of our times.
Thanks to solid household balance sheets and supportive fiscal policies, the medium-term growth prospects of the euro area economy remain strong and favourable, even if more persistent supply bottlenecks imply that part of the increase in production and demand that was expected for this year and next may only materialise later on.
As a result, inflation will remain higher for longer than previously anticipated. It will decline over the course of next year, but uncertainty has increased around the pace and extent of the slowdown, both of which crucially depend on the ramifications of the green transition on energy prices and the capacity of supply to catch up with the current fast rate of demand normalisation.
In such situations of elevated uncertainty, monetary policymakers need to focus on the entire range of possible outcomes to ensure that they will be able to deliver on their mandate.
On the one hand, this means avoiding the mistake of a premature tightening of monetary policy in response to a temporary and possibly short-lived inflation spike. On the other hand, it means keeping a watchful eye on the upside risks to inflation that financial markets currently anticipate and retain optionality to be able to act if needed, so as to maintain trust in our determination to defend price stability in a symmetric way and prevent a deanchoring of inflation expectations in both directions.
- De Bondt, G. et al. (2021), “Main findings from the ECB’s recent contacts with non-financial companies”, Economic Bulletin, ECB, Issue 7. The exchanges mainly took place between 4 and 13 October 2021.
- Rühl, C. and Erker, T. (2021), “Oil Intensity: The curious relationship between oil and GDP”, M-RCBG Associate Working Paper Series, No 164, Mossavar-Rahmani Center for Business & Government, Harvard University.
- Dutta, A. et al. (2020), ”Do green investments react to oil price shocks? Implications for sustainable development“, Journal of Cleaner Production, Vol. 266.
- Durré, A. et al. (2021), “A Short-Lived Income Hit from Soaring Energy Prices”, European Daily, Goldman Sachs, 16 October.
- See also Andersson, M. et al. (2021), “Economic developments and outlook for contact-intensive services in the euro area”, Economic Bulletin, ECB, Issue 7.
- Based on data for the 11 countries that adopted the euro in 1999. See ECB (2000), “Lessons to be drawn from the oil price shocks of the 1970s and early 1980s”, Monthly Bulletin, November.
- There were, of course, differences across euro area countries. In Germany, for example, real short-term interest rates were, on average, higher than in most other economies.
- Greenspan, A. (2004), “Risk and Uncertainty in Monetary Policy”, speech at the American Economic Association Annual Meeting, San Diego, 3 January.
- Oil futures prices also partly reflect convenience yields rather than actual price expectations.
- Natural gas production, too, remains below pre-crisis levels in the United States, Norway and other economies, reflecting lacklustre interest from investors in fossil fuel energy production.
- For example, some semiconductor developers expect that factory capacity expansions will not boost production until 2023 or 2024.