Driving assessment

The war in Ukraine has raised long-term inflation expectations

Inflation continues to rise in many countries. Until recently, central banks such as the ECB or the Swiss National Bank viewed the current rise in inflation as temporary, albeit more persistent than expected. They attributed the high inflation rates to base effects, supply bottlenecks and demand shifts, all related to the pandemic, and expected these transitory factors to decline over the course of the year. the year. As long as long-term inflation expectations remained anchored, core inflation should be stable and close to central bank targets.

However, the Russian invasion of Ukraine and the resulting economic disruption could fundamentally upset this assessment. For the world economy, Russia and Ukraine are important suppliers of raw materials such as energy, metals and agricultural products. The war called into question the supply of these resources and accelerated the evolution of their price. It is feared that a further increase in the price of these factors of production will translate into higher operating and manufacturing costs, which in turn translates into higher prices and ever-higher inflation expectations. (D’Acunto and Weber 2022). Indeed, oil price shocks are historically known to fuel inflation expectations and unanchor the expectations of consumers and price makers (Coibion ​​and Gorodnichenko 2015, Coibion ​​et al. 2018).

To determine whether the Russian invasion of Ukraine has affected corporate inflation expectations, I use data from a special survey of Swiss companies conducted by the KOF Swiss Economic Institute at ETH Zurich. The companies surveyed are companies from all economic sectors in Switzerland, with the exception of agriculture. As part of this survey, companies were asked where they expected the annual inflation rate of the consumer price index in Switzerland to be over the next twelve months (short term) and next five years (long term). There were no response options limiting company assessments to answer these quantitative questions. Companies were free to formulate their expectations as a numerical value (in percentage).

The empirical strategy then exploits the fact that the investigation was launched just before and conducted during the Russian invasion. This configuration makes it possible to identify the effects induced by the war on companies’ inflation expectations by comparing the answers of companies which answered the survey before February 24 (the first day of the war) with those of companies which responded after the outbreak of war. As the survey was conducted electronically, it is possible to determine the exact response time of the companies: 652 companies responded before the invasion, 258 companies since. The analysis takes into account all responses received at the time of writing (9 March).

Figure 1 shows the distributions of 12-month (left panel) and 5-year (right panel) inflation expectations responses separately for groups of firms that responded before and during the war.

Figure 1 Inflation expectations of Swiss companies

To note: This figure plots the distributions of 12-month (left panel) and 5-year (right panel) inflation expectations responses separately for groups of firms that responded before and during the war. The vertical lines show the group-specific mean values. Responses greater than 20% in absolute values ​​are excluded.

In the short term, inflation expectations are almost the same for both groups. Before the start of the war, Swiss companies expected the inflation rate to reach 2.10% (median: 2.00%) on average over the next twelve months. Since the beginning of the war, the expected increase in consumer prices is 2.12% (median: 2.00%). Thus, Swiss companies estimated that the inflation rate over one year would be higher than the inflation rate known to most participants at the time of the survey. Actual inflation was 1.6% in January 2022.1 Moreover, the median and mean values ​​are close to each other, indicating an overall fairly condensed distribution. The cross-sectional standard deviation of inflation expectations is only 1.54% before the war and 1.55% during the war.

In the long term, inflation expectations have increased since the Russian invasion from 2.37% (median: 2.00%) to 2.75% (median: 2.00%). We can use a regression model to test if this difference is statistically significant. The advantage of such a model is that other factors can also be taken into account, such as the size or the sector of activity of a company. The estimate shows that the increase in five-year inflation expectations is statistically significant at the 5% significance level. This suggests that after the outbreak of the war, Swiss companies expect higher consumer prices in the long term than before the war.

Figure 2 documents the systematic heterogeneity across firms in long-term inflation expectations by showing the effects of conditional mean treatment with 90% confidence limits. Differentiated by sectors, the left panel shows that companies in the manufacturing sector are the source of higher expected inflation since the start of the war. Their long-term inflation expectations have increased by 0.7 percentage point on average. This result seems plausible because these firms are closer in the value chain to input factors such as energy and other raw materials, the prices of which rose significantly during the war. On the other hand, there is no significant change in the expectations of companies in the construction or service sectors.

Figure 2 Heterogeneity of long-term inflation expectations between companies

To note: This chart shows the effects of the conditional mean treatment (with 90% confidence limits) of the Russian invasion on firms’ long-term inflation expectations by sector (left panel), firm size (middle panel) and the evolution of profit margins over the last five years (right panel).

The middle panel shows the effect of firm size. Small and medium-sized companies with less than 250 employees tend to report higher long-term inflation expectations than large companies since the Russian invasion. Finally, the panel on the right distinguishes according to the evaluation of companies whether the profit margin of their main product or service has increased or decreased over the last five years. Inflation expectations have risen for companies whose margins have shrunk in recent years. Rising energy and commodity prices are already reducing profits for many companies. Therefore, the result might suggest that firms facing margin pressures are more likely to expect to pass on higher prices for their inputs by raising their own prices.

Indeed, companies confirm that higher operating and manufacturing costs impact their prices. As part of the survey, companies were also asked about the reasons for changing prices.2 Specifically, the survey asked companies to rate a range of factors such as labor cost or competitor price changes based on their importance to price adjustments with points ranging from 1 (“completely unimportant”) and 4 (“very important”). These ratings were collected separately for price increases and price decreases.

The average odds for most factors are higher for price increases than for price decreases. Exceptions to this observation are market conditions such as the intention to gain market share, changes in competitor prices or changes in demand. Taken together, this observation highlights asymmetries in price drivers: cost changes are the most important factor for price increases. On the other hand, market conditions are driving price declines. Figure 3 illustrates these asymmetries by showing the difference in average score for each factor and by sector. The results show a surprisingly regular pattern of positive asymmetries in costs and negative asymmetries in market conditions.

picture 3 Asymmetries of factors influencing prices

To note: Companies were asked to rate the factors according to their importance for price adjustments with points ranging from 1 (“completely unimportant”) to 4 (“very important”). These ratings were collected separately for price increases and price decreases. This figure shows for each factor and sector the difference between its average score for a price increase and its average score for a price decrease.

Beyond that, the figure shows that these asymmetries are not the same in all sectors. In particular, higher supplier prices, higher raw material costs, and higher energy and fuel prices – all of which are rising sharply following the Russian invasion – are more important in explaining the increases. prices in the manufacturing sector than in construction or services.

Conclusion

The results of a special survey on pricing behavior show that following the Russian invasion of Ukraine, the long-term inflation expectations of Swiss companies increased significantly, especially in the manufacturing sector. In manufacturing, rising input costs, such as rising energy and commodity prices, are particularly important in driving price increases. These results reinforce concerns that, as price pressures widen, inflation expectations could become less anchored, making headline inflationary pressures much more persistent.

The references

Blinder, A, E Canetti, D Lebow and J Rudd (1998), Asking about prices: a new approach to understanding price stickinessRussell Sage Foundation.

Coibion, O and Y Gorodnichenko (2015), “Is the Phillips curve alive after all? Inflation expectations and missing disinflation”, American Economic Journal: Macroeconomics 7(1): 197-232.

Coibion, O, Y Gorodnichenko and S Kumar (2018), “How Do Firms Shape Their Expectations? New Survey Data”, American Economic Review 108(9): 2671-2713.

D’Acunto, F and M Weber (2022), “Rising inflation is worrisome. But not for the reasons you think”, VoxEU.org, 4 January.

Fabiani, S, M Druant, I Hernando, C Kwapil, B Landau, C Loupias, F Martins, T Mathae, R Sabbatini, H Stahl, A Stokman (2005), “The behavior of firm prices in the euro zone: new evidence investigation”, ECB Working Paper No. 535.

Endnotes

1 The inflation rate for February was only published on March 3 and amounts to 2.2%.

2 The wording of these questions is adapted from Blinder et al. (1998) and Fabiani et al. (2005).