Oya Celasun, Niels-Jakob Hansen, Mariano Spector, Aiko Mineshima, Jing Zhou 16 March 2022
Supply bottlenecks hindered the industrial recovery and boosted inflation in many economies last year, as a fast recovery in goods demand collided with disruptions to production and distribution. When mobility restrictions were introduced in the spring of 2020 to control Covid-19, consumers cut spending on services and started buying more manufactured goods instead. The reopening of economies initially boosted manufacturing output, but renewed lockdowns and shortages of intermediate inputs from chemicals to microchips caused the manufacturing recovery to stall. Prices of manufactured goods rose rapidly as delivery times reached record highs, sparking a debate about inflation and the appropriate course of monetary policy (e.g. Summers 2021 and Powell 2021).
Understanding the respective roles of increased demand and weakened supply, and how long these shifts may last, is key to getting macroeconomic policies right. Where bottlenecks result primarily from the interaction of strong demand and inelastic supply, only a softening in demand can alleviate inflation pressures in the short run. Conversely, where the bottlenecks primarily result from disruptions, actions to boost supply in the short run could lessen the need for policies to contain demand.
In our recent research (Celasun et al. 2022), we seek to quantify the impact of increased demand versus weakened supply on aggregated manufacturing output and inflation in a small set of countries using sign-restricted vector auto regressions.1 The assumption used to identify demand and supply shocks is that the former move output and prices in the same direction, whereas the latter move them in opposite directions. In this column, we expand that analysis to cover over 30 economies over the period January 2001 to December 2021. We then shed further light on the drivers of the demand and supply shocks by correlating them with the characteristics of the economies in the sample.
It’s both supply and demand
Figure 1 Supply-demand decomposition of euro area manufacturing industrial production (IP) and producer price index (PPI), 2016-2021
A) Contributions to industrial production (IP) fluctuations (%)
B) Contributions to producer price index fluctuations (%)
Source: IMF staff calculations.
We find that, during 2021, the boost to manufacturing output from higher demand was largely, or in some cases more than, offset by supply constraints (Figure 1). Estimated supply shocks are predominantly negative2 in the years before the pandemic, dampening the inflation of manufactured goods and supporting manufacturing output. During the pandemic, however, months with positive supply shocks have been very common, boosting inflation and holding down output. Figure 1 shows the historical decomposition for the euro area. On the left, the solid line is euro area manufacturing industrial production (IP) in deviation from trend, given by the sum of a component capturing the cumulative effects of demand shocks (blue bars) and a component capturing the cumulative effect of supply shocks (red bars). Manufacturing industrial production was on average about 2.5% below trend in the last six months of 2021, with demand shocks pushing output about 5% above trend and supply shocks pushing it down by about 7.5% below trend. Put differently, supply shocks lowered euro area manufacturing output in the second half of 2021 by about 7.5%. Figure 2 shows that the supply shocks experienced in 2021 exerted a staggering 17%-15% drag on manufacturing industrial production in Romania and Slovakia, and drags of about 12.5% in Germany, Czechia, and Hungary.3
The supply constraints on manufacturing production imply large effects on real GDP. We use the simple historical correlation between manufacturing industrial production and GDP to get a rough estimate of the impact of the manufacturing supply shocks on 2021 GDP. The estimates suggest that supply bottlenecks lowered euro area real GDP by about 2.5% in 2021.
Figure 2 Bivariate structural vector autoregression (SVAR): Impact of supply shock on manufacturing industrial production, 2021H2 (percentage points)
Source: IMF staff calculations.
Notes: The values shown in bars are the average impacts of the 2021 supply shocks on manufacturing industrial production for the last six months of 2021.
Supply shocks have also boosted the inflation of manufactured goods prices, but so did demand. Manufacturing producer price index (PPI) inflation in the euro area averaged about 12.5% in the second half of 2021, substantially higher than an average of 1.5% in 2017–19. We estimate that close to half of that upward swing came from the change in the contribution of supply shocks, which had mostly exerted downward pressure on manufactured-goods prices in the pre-pandemic years (to the tune of -0.5 percentage points on average for the euro area in 2017-19). The share of the increase attributable to supply shocks varies across individual countries; it is estimated at about half for the euro area, 60% for France and Germany, 45%–50% in the US and the UK, and about 33% for Italy (Figure 3). Higher demand contributed the bulk of the remaining increase in manufacturing PPI inflation.
Figure 3 Supply shock contribution to the change in manufacturing producer price index (PPI) inflation, 2021H2 relative to 2017-19
Source: IMF staff calculations.
Notes: The bars show how much of the change in inflation between 2017-19 and 2021H2 can attributed to the shift in supply shocks (given by the ratio of (i) the supply-shock impact on producer price index (PPI) inflation in 2021H2 relative to the supply shock impact in 2017-19 to (ii) the difference between manufacturing PPI inflation in 2021H2 and average manufacturing PPI inflation in 2017-19, where all inflation rates are quarter-on-quarter and annualised).
Supply disruptions also had an impact on core consumer prices (i.e. the consumer price index (CPI) inflation excluding energy and food prices). This measure of inflation in the euro area was about 1.4 percentage points higher in the second half of 2021 than in 2017-19; about half of this increase can be explained by supply constraints for manufactured goods. This smaller impact on core CPI inflation, in absolute terms, is not surprising because goods make up less than half of the consumption basket. The prices of services, which account for more than half, are less sensitive than those of goods to manufacturing supply shocks.
Diverse impacts across countries
The extent of supply disruptions – and estimated supply shocks – has been diverse across countries and sectors. For instance, manufacturing production weakened during much of 2021 in Germany and in September and October in Japan, while continuing to recover in the UK and Spain. One reason for this diversity is likely to be the composition of manufacturing output, as some subsectors could be more vulnerable to disruptions than others. For instance, the auto sector suffered severe production declines after 2021Q2 when the shutdowns to contain the Delta variant of Covid-19 in Asia started curtailing semiconductor production – one of the important intermediate inputs for autos. Correspondingly, countries with large auto sectors are estimated to have suffered greater supply shocks (Figure 4, Panel A). Likewise, disruptions were larger in countries with higher shares of foreign value added in gross domestic manufacturing output (Figure 4, Panel B), or where the manufacturing sector operates in the downstream segments of global value chains and is therefore vulnerable to shutdowns that hamper upstream production (Figure 4, Panel C).4
Demand shocks were also diverse, being larger in countries where the consumer spending on goods has grown more relative to pre-Covid times (Figure 4, Panel D). Manufacturing is typically open to trade, and large estimated demand shocks could reflect strong demand at home or spillovers from demand elsewhere.
Figure 4 Impact of supply and demand shocks: Country diversity
Sources: Haver Analytics, OECD, and IMF staff calculations.
Notes: This panel plots the structural vector auto-regression (SVAR) estimated supply and demand shocks (average in 2021H2) against country characteristics. For panel B, supply shocks are residualised from domestic shutdowns; for panel D, pre-Covid goods takes a linear trend, and data for 2021Q4 is available only for France, Japan, The Netherlands, Norway, and the US.
Problems could persist
On average, we find that up to 40% of the supply constraints can be traced to shutdowns, which should have only transient effects on supply and inflation. The same is true of the severe weather and industrial accidents that hindered microchip and auto output in 2021. Other drivers of supply constraints, such as labour shortages due to re-evaluated worker preferences (which we estimate to explain up to 10% percent of supply constraints) and ageing logistics infrastructure, could have more persistent effects than shutdowns.
Late last year, industry experts expected supply shortages for autos to largely dissipate by mid-2022, and broader bottlenecks by the end of this year. The Omicron wave and the war in Ukraine have injected new uncertainty. All in all, supply disruptions could last for longer, possibly into 2023.
The bottom line of the analysis is that the rise in the inflation of manufactured goods prices reflects a mix between supply constraints and demand strength. The supply constraints are likely to last for some time and are likely to be exacerbated by the war in Ukraine. The policy priorities for countries hit by supply constraints fall into three groups.
- First, use regulations to address specific bottlenecks wherever possible – for example to ease the licensing of truck drivers and port workers to resolve logistical logjams, ease immigration rules to bolster labour supply, and set rules that can help keep the virus under control at workplaces.
- Second, fiscal policy should seek to minimise the scarring that can come from prolonged weakness in manufacturing activity, for example by helping firms retain skilled manufacturing workers. Fiscal policy should be actively used to support labour supply, including through the re-skilling and retraining of workers, and ensuring robust child and elder care so people can return to work.
- Inflation pressures from supply bottlenecks might continue for some more time – adding to pressures from surging energy prices. The challenge for central bankers is to foster a full recovery while anchoring medium term inflation expectations at target. For the latter, it is critical that central bankers continue to communicate how they will react to inflation and other pertinent data and movements in inflation expectations, and signal readiness to respond rapidly to any significant change in the medium-term inflation outlook.
Antràs, P and D Chor (2018), “On the Measurement of Upstreamness and Downstreamness in Global Value Chains”, in L Yan Ing and M Yu (eds.), World Trade Evolution: Growth, Productivity and Employment, Chapter 5, London, U.K.: Routledge.
Celasun, O, N-J Hansen, A Mineshima, M Spector and J Zhou (2022), “Supply Bottlenecks: Where, Why, How Much, and What Next”, IMF Working Paper, 22/31.
Kilian, L, N Nomikos and X Zhou (2021), “Container Trade and the US Recovery”, Dallas Fed Working Paper No. 2108.
Powell, J H (2021), “Monetary Policy in the Time of Covid”, Speech delivered at the “Macroeconomic Policy in an Uneven Economy”, economic policy symposium sponsored by the Federal Reserve Bank of Kansas City, Jackson Hole, Wyoming (via webcast), 27 August.
Summers, L (2021), “On Inflation, its Past Time for Team Transitory to Step Down”, Washington Post, 16 November.
1 The methodology follows the one proposed by Kilian et al. (2021), who use sign restricted VARs to estimate supply and demand shocks.
2 We adopt the sign convention that a positive supply shock increases inflation and decreases output.
3 The bars in Figure 2 capture the cumulative effects of the supply shocks of 2021 only (we abstract from the shocks of 2020 and earlier years as the supply disruptions we want to focus on occurred in 2021).
4 On average, manufacturing in Asia tends to be in the upstream part of global value chains, America is typically downstream, and Europe usually locates in between (Antràs and Chor 2018).