The global 3-5 year stockbuilding (inventory) cycle probably bottomed in Q2. The negative contribution of inventories to the annual change in G7 GDP is estimated to have been comparable with or larger than at the trough of the 2008-09 recession. Destocking appears to have been mostly involuntary, reflecting covid restrictions having a larger negative impact on goods production than demand.

A Q2 trough would imply a cycle length of 4.25 years – the prior low occurred in Q1 2016. The previous two cycles – 2009-12 and 2012-16 – lasted 3.5 and 3.25 years respectively. The average length of the cycle historically (i.e. since the 1960s) was 3.5 years*.

Industrial commodity prices usually weaken during cycle downswings but rebound in the restocking phase. The first chart shows the annual change in an index of industrial commodity prices and the contribution of stockbuilding to the annual G7 GDP change – a positive correlation is apparent. Commodity prices fell by 9.6% on average in the 12 months leading up to a cycle trough, rising by 14.3% in the subsequent 12 months.

Strong global monetary expansion suggests an increased probability of a commodity price surge. Global annual narrow money growth is at a post-WW2 record. The second and third highest readings occurred in November 1972 and August 2009 respectively. These monetary surges preceded the two highest peaks in annual industrial commodity price inflation, of 53.5% (December 1973) and 74.5% (March 2010).

Commodity price strength could drive an early and sharp pick-up in producer price inflation. Strikingly, the input and output price indices in the global manufacturing PMI survey have already returned to pro-covid levels, following only modest weakness during the economic contraction phase – second chart.

A commodity price upswing could sustain the recent uptrend in bond market inflation expectations, in turn raising questions about the credibility of central bank guidance that super-easy monetary policies will be maintained far into the future.

*Cycle dates are assessed judgementally taking into account average cycle length, the contribution of inventories to the annual G7 GDP change and business survey inventory responses.