The ferrous complex stabilised on July 24, 2021 with spot up slightly and paper too. Steel has not updated:
Since January, I have held a “sell the rallies” position for iron ore. I am now moving to an outright “sell” position. There are five reasons.
First, there is early evidence that this time it is different in Chinese attempts to curb steel production and drop commodity prices. In the past, this strategy has always failed because a fragmented steel mill sector boosted output into improving margins as steel prices rose. This drew in more iron ore, passing price rises to upstream sectors.
This time China looks more determined. If officials can herd the steel mill cats and get them to stick to lower output then the nexus with iron can be broken. The push has already played some role in big output falls to date. The lastest mid-July output number from CISA was a meek bounce out of the EOFY slam:
H1 steel output grew at 10.4%. It is now near zero. An impressive pullback. Chinese authorities want to see total 2021 output below 2020. To achieve that, H2 output is going to have to fall 11.8%, which would look like this from here (see the red line):
This outcome is highly unlikely given it requires a cut of 60mt of steel output in the second half to produce the equivalent 2021 overall output level. Steel prices would go mad. But there are very good reasons to think that the output caps can make considerable progress.
To wit, point two. Demand is going to slow. Throughout this year, I have argued that there is a “pig in the python” in apparent Chinese steel demand because 18 months’ worth of construction activity has been compressed into the last year. This is catch-up growth as COVID-delayed projects during H1 2021 restarted alongside normal project starts.
Now they are all rolling off at once, completions are soaring, and the stock of activity is resetting lower, along with steel demand.
We can add point three. The credit tightening of the last nine months has already dropped the flow of property starts to 2019 levels through H1. Infrastructure is also falling back sharply. This is 70% of Chinese steel demand heading back to 2019 levels. And it is still under pressure from the Three Red Lines policy and could go lower still. Especially so if circumstances around Evergrande deteriorate further.
If we aggregate these three points we can conservatively estimate that H2 steel output will fall by an aggregated 20mt versus H1. That’s 30mt less iron ore needed or a 60mt annualised lower iron ore demand run rate.
Point four is about supply. It is largely steady except for Vale and RIO. Last week’s production reports show that between them they will need to ramp up H2 output by 70mt annualised to hit 2021 iron ore output targets. Again, reduce the number and expect them to miss. The likelihood is that they’ll manage at least half of it.
Putting the observations about demand and supply together, we get an annualised loosening from H1 to H2 of about 100mt in the iron ore market balance.
Point five is about seasonality. Whenever the iron ore market weakens, the biggest impact is felt in the Sep/Oct period. It is the softest period of the year for steel mills as they clear lingering inventories and destock raw materials. The above swing towards market loosening for iron ore is the perfect set-up for a large fall in this seasonally weak period. That we are already weak in the seasonally strong July period is a bad sign.
In short, the five points are more consistent with an iron ore price that is <$150 than it is >$200.