Commodity inflation is likely to be temporary, but the upward risks of long-term inflation remain

Commodity inflation is likely to be temporary, but the upward risks of long-term inflation remain

Inflation is back. Annual US CPI inflation hit a 40-year high of 8.5% in March 2022. This increase continues the pattern that began in May 2021, when annual inflation exceeded 5% for the first time in 30 years. It has been steadily rising ever since. While the high level of inflation is certainly a concern, a major policy issue is whether the current increase is a transitional or continuous phenomenon. Describing inflation in terms of its persistence (or lack thereof) is not merely an academic one. Until very recently, the Federal Reserve relied on such a strategy to communicate employee opinions about the (potentially temporary) nature of current inflationary pressures.1 Against this background, conventional wisdom holds that the transmission of monetary policy effects takes 12-18 months. This suggests that central banks should only act if they realize that inflationary pressures are likely to persist.

A notable feature of the current episode of inflation is the dynamics of commodity inflation. Figure 1 shows annual goods and services inflation in the United States from the first quarter of 1960 to the fourth quarter of 2021. While goods and services inflation increased during this period, the rise in goods inflation was significantly more pronounced. This can largely be attributed to two factors. First, the disruption to global supply chains and labor shortages that we have seen since the onset of the COVID-19 pandemic have disproportionately affected the shipping and delivery of consumer goods worldwide (Coibion ​​et al. 2020). Second, since early 2021, these supply-side bottlenecks have been countered by increased demand for goods fueled by excess savings, generous fiscal policy, and some service-based expenditures simply not being an option due to lockdown policies (D’Acunto and Weber 2022).

shape 1 Annual inflation of US goods and services

Noticeable: Shaded areas indicate NBER stagnation dates.

Although the current episode of high inflation has highlighted the behavior of commodity inflation, Figure 1 also presents underestimated facts. Specifically, since the 1990s, goods and services inflation has shown somewhat different patterns with goods inflation being more volatile and less (on average) than service inflation.

If we accept the initial assumption that monetary policy must work against the persistent component of inflation, the natural issue is how one should explain the recent behavior of goods and services inflation in terms of its effects on aggregate inflation over an extended horizon. In our recent work (Eo et al. 2022), we set out to understand trend inflation (that is, the rate of inflation that would be expected to prevail if there were no temporal factors) from the perspective of a two-sector empirical inflation model. We construct trend inflation for both goods and services by adopting statistical methods that strip the temporary component (or noise) of each sector’s inflation rate.


Figure 2 presents the estimated trends of goods and services inflation along with the corresponding inflation rates for these two sectors. The striking result is that prior to the early 1990s, the estimated trend in each sector mimicked actual inflation for a particular sector. However, since the 1990s, while commodity inflation has remained volatile, its trend has become quite constant. Accordingly, our model interprets much of the fluctuation in commodity inflation as transient. In contrast, a quick examination of the estimated trend in the services sector indicates that the opposite has occurred with most of the variance in service inflation attributable to its continuous (rather than temporary) component.

Figure 2 Estimated trend of sectoral inflation

Notes: The flat line is the quarterly annual inflation rates for a specific sector, and the dotted lines represent our post-median estimate of the sector-specific trend with its associated credibility group of 67%. All inflation rates on an annual basis. Shaded areas indicate NBER stagnation dates.

Given that goods and services represent the entire consumption basket, the variance of sum Trend inflation can be formally decomposed into the variance of each of these two sectors as well as the correlation between them. Figure 3 displays such a decomposition. We note two striking features. First, the volatility of the overall trend of inflation was high during the hyperinflation of the 1970s, but has decreased significantly since the early 1990s. This result was also prominently documented by Stock and Watson (2007). Second, while both sectors and their firms used to contribute to the volatility of the headline inflation trend in the 1970s, the service sector has dominated this measure since the early 1990s. In other words, inflation in the commodity sector over the past three decades has contributed very little or nothing to the variance in the persistent component of aggregate inflation.

Figure 3 Variance decomposition for trend inflation

NotesInflation trend in annual inflation units on a quarterly basis. Shaded areas indicate NBER stagnation dates.

How should we look at the recent inflation episode?

We turn to understanding recent high inflation in the context of our model. Figure 4 shows the estimated trend inflation along with headline inflation. Our point estimate for trend inflation at the end of 2021 (last data point) is 2.2%. Our sectoral split provides a clear idea of ​​why trend inflation has not (yet) accelerated as much as overall inflation. Since the current high inflation episode is largely evident in the goods sector and our model interprets commodity inflation as mostly transient, much of the increase in headline inflation is transient. Thus, overall trend inflation remains somewhat muted.

Moreover, while the current high inflation inevitably gives rise to comparison with the hyperinflation of the 1970s (eg Ha et al. 2022), our empirical exercise offers other perspectives on how to view comparisons relating to the 1970s. Estimates of the inflation trend of the 1970s are still much higher than our current estimates. This indicates that inflation expectations remain constant, as did Cascaldi-Garcia et al. (2022) argue. Furthermore, our model suggests widespread inflation during the Great Inflation of the 1970s. Figure 3 shows that the term comovement between the goods and services sectors is used to contribute to the overall trend inflation variance. This term comovement now plays little role in our analysis of what leads to the variance in overall trend inflation at the end of 2021. Thus, the current high inflation episode, at least at this point, does not appear to reflect the persistent, broad-based cross-sectoral inflation observed during the 1970s (See also Borio et al. 2022).

However, it is important to describe some of the nuances associated with our findings by making two specific points. First, the degree of uncertainty associated with our estimate of directional inflation of 2.2% at the end of 2021 is much greater than usual.2 Figure 4 shows that the 67% reliable interval for estimating trend inflation at the end of 2021 is between 1% and 4%, highlighting the possibility that overall trend inflation will be much higher relative to the Fed’s long-term average. Inflation target is 2%. The reliable interval also indicates that the risk on our estimate of directional inflation at 2.2% is largely skewed to the upside. Therefore, from a risk management perspective, one should not necessarily accept unequivocally that the current high inflation is a passing thing.

Figure 4 Estimated overall trend inflation estimate

Notes: Flat line indicates PCE inflation both on a quarterly basis. The dotted lines indicate our subsequent mean estimate of the overall trend inflation with the 67% confidence interval. All inflation rates on an annual basis. Shaded areas indicate NBER stagnation dates.

Second, according to our model, even if the point estimate of trend inflation remained at a level that the monetary authority might be comfortable with, trend inflation was gradually picking up and correlating with moderately higher levels of trend inflation not only in goods but also (and more importantly) in the services sector. The latter may be consistent with the idea of ​​a wage increase, and if so, we know from our work that movement in service inflation is fueling trend inflation. Given the years between the Great Recession and the COVID-19 pandemic, the United States, and most of the world, has consistently experienced below-target inflation, and some may welcome recent increases in trend services inflation as it could turn trend inflation into a range where monetary policymakers would like it to. He is. However, our work also indicates that if services inflation continues to rise, trend inflation is likely to increase beyond its current modest estimates to potentially more worrying levels.


We estimate a model of directional inflation for the goods and services sectors to understand how both sectors contribute to the persistent inflation component. Our main finding is that since the early 1990s, commodity inflation has played little role in driving trend inflation. This contrasts with the hyperinflation of the 1970s, when commodity inflation was a larger contributor to the variance in aggregate inflation. Our results were maintained throughout the COVID-19 pandemic. Therefore, the discretionary trend inflation rate was a relatively modest 2.2% at the end of 2021, and we believe that the high commodity inflation we are currently seeing is likely to be temporary. The latest assessment is largely based on much of the recent high inflation (still) concentrated in the traditionally high-frequency commodities sector. However, we should note that the model finds greater uncertainty typical associated with the estimate of trend inflation during the recovery from the COVID-19 pandemic. It seems important to reiterate the last point since the risk of trend amplification tends to the upside, which remains a relevant consideration from a risk management perspective.

Authors’ note: The opinions expressed are our own and do not necessarily reflect the views of the Bank of Canada.


Borio, C, P; Disyatat, D Xia and E Zakrajšek (2022), “Looking Under the Hood: The Two Faces of Inflation”,, 24 January.

Coibion, O, Y Gorodnichenko and M Weber (2020), “Labour markets during the COVID-19 crisis: an initial look,” NBER Working Paper No. 27017.

D’Acunto, F and M Weber (2022), “Rising inflation is worrisome. But not for the reasons you think,”, Jan. 04.

Cascaldi-Garcia D, F Loria, D López-Salido (2022), “Is the inflation trend at risk of becoming unrecognized? The role of inflation expectations”, FEDS notes, March 31.

Eo, Y, L Uzeda, B Wong (2022), “Understanding trend inflation through the lens of the goods and services sectors”, CAMA Working Papers 2022-28.

Ha J, M Ayhan Kose and F Ohnsorge (2022), “Today’s Inflation and the Great Inflation of the 1970s: Similarities and Differences”,, 30 March.

Stock, J. H. and MW Watson (2007), ‘Why has inflation become so difficult to predict in the United States?’ , Journal of Money, Credit and Banking 39 (Q1): 3-33.


1 See, for example, the FOMC statement of September 22, 2021 at

2 Our estimate of the current trend of inflation and uncertainty is largely consistent with that reported in Cascaldi-Garcia et al. (2022).

Leave a Comment