How Metal Inventories and Demand Shape Futures Curve Direction
Summary
The document explains why different metals can have different futures curve shapes, using copper in backwardation and aluminum in contango as examples. It starts from a cost-of-carry relationship involving financing, storage costs, and convenience yield, then emphasizes that current curves also reflect changing expectations for supply and demand across maturities.
The answer points to differences in the metals’ recent price recoveries and considers possible demand drivers, including construction activity and electric vehicle production. Stronger copper demand from electric vehicles, alongside potentially weaker aluminum demand, is presented as one possible contributor. These explanations are hypotheses rather than a demonstrated causal analysis: construction indicators are mixed, and the answer does not establish that vehicle production is the dominant force. It gives no quantitative test or general rule for predicting curve shape, so the examples should be read as context-specific explanations rather than proof that one factor determines contango or backwardation.
Key ideas
- Futures curve shape can reflect expectations about supply and demand over different maturities.
- Financing, storage costs, and convenience yield provide a framework for understanding contango and backwardation.
- Different price recoveries in copper and aluminum may help explain their different curve shapes.
- Construction activity and electric vehicle production are possible demand factors, but the evidence presented is mixed.
- The proposed explanations are plausible contributors, not a quantified attribution of the curve difference.
Tags
Full text
# Why are some metals in contango (inverted) forward curve and some in backwardation (normal) forward curve?
# Why are some metals in contango (inverted) forward curve and some in backwardation (normal) forward curve?
I am scrolling through the various metals on lme.com and some are in contango and some in backwardation. For example:
Copper: backwardation
Aluminium: contango
Further examination of other metals reveals the same. Why is this? Why is one metal in contango and another in backwardation?
I guess if I take
$$ f^T(t) = S(t)e^{(r+g-c)(T-t)} $$
where $r$ is the risk free rate, $c$ is storage cost and $g$ is convenience yield then depending if $r+g-c$ is negative or positive you get the two curves. But why would $r+g-c$ be that different for 2 different metals? Could it have something to do with inventories - e.g. full inventories would raise $c$, the cost of storage?
## Answer by kurtosis (score 4)
https://quant.stackexchange.com/a/57931
There could be any number of explanations for copper to be backwardated and aluminum to be in contango right now.
The simplest (and most correct) explanation is the most vague: that these futures curves represent time-varying expectations of supply and demand over the next half year or so.
More helpful is to know that copper and aluminum took a sharp dip earlier this year due to the precipitate economic slowdown; however, copper has more than recovered while aluminum's recovery has been weaker. Therefore, another reason for copper to be backwardated is that cash and the front contracts have been dragging the rest of the curve upward -- while aluminum has not done that.
We could also look for explanations in the macroeconomy: there is a bit of a recovery in China and Europe and that will lead to more homes being built as well as more vehicles being sold.
Contango in rebar futures (traded in Shanghai) makes the housing story less likely. Figures on construction in China also point to mixed results: growth in residential but shrinkage in commercial properties -- and an across-the-board decline in completions and large declines in sales for commercial and industrial properties.
Vehicle production is another possibility. An increasing number of vehicles are electric, so that would also increase demand for copper while decreasing demand for aluminum (used to make engine blocks). Furthermore, electric vehicle production in July 2020 was up 77% versus July 2019.
Is a shift in vehicle production toward electric vehicles likely to be the explanation behind these two differing curves? If we consider that EVs are changing the marginal vehicle being produced, then that should certainly be part of the reason and should affect prices. However, the real question is if that is a dominant force behind the difference. Without a much more in-depth study, we cannot be certain if shifts to electric vehicle production are the major driving force for the difference.Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.