Lessons for iron ore from the coal downturn: part 1

There has been a huge divergence between the performance of iron ore and coal (both met and thermal) in the last 12 months.  While iron ore isn't any higher than where it was this time last year, it is currently defying expectations of much weaker prices.

Coal, on the other hand, is weaker than many would have anticipated and has moved into territory that most didn't think possible not too long ago.

 Given the diverse price performance relative to expectation, it worth looking at the similarities and differences of these markets to see if there are lessons that can be drawn.

This first part will provide the background on why thermal coal prices have surprised to the downside and the importance (or non-importance) of cost curves.

The initial weakness thermal coal prices was in part due to stronger supply, particularly as US domestic tonnes were displaced into seaborne markets as natural gas prices collapsed.  2012 was also a year of few supply disruptions and expected decent growth from Australia and Indonesia in particular.

But I the bigger source of weakness is the massive miss in coal demand in China in the last two years. At the end of 2011, most were expecting power demand to travel at ~6-7% in 2012 and coal burn to be somewhat weaker.  The reality was that power demand at 4.5% while coal burn barely rose at all YoY thanks to a surge in hydro generation. In coastal regions, coal burn actually shrank over the year.

This meant that coal consumption in Chinese coastal areas, which draws the marginal domestic supplied tonne and consumes imports, was around 100mt lower than was originally forecast in 2012. This huge miss on demand meant that the balance within China was radically shifted to surplus and very high cost coal was displaced.

Adding pressure to the equation is that an not only has demand shifted substantially lower in China, but new sources of cheaper supply has become available in the last 2 years.

Firstly, cheaper domestic supply has become available as more rail and port capacity has reduced the need for long distance trucking to Bohai sea ports. For example, rail expansions into Huanghua have offset trucked coal in Tianjin and more marginal ports on the North east side of the Bohai Bay.


Second, imports continue to grow. Chinese coal imports should be up ~15mt this year, although perhaps some of this tonnage is coming from seaborne producers who are also loss making like high cost domestic producers.

In all, this adds up to ~150mt of high-cost coal that has been displaced over the last two years, which is a huge number.

Prices have continued to fall in 2013 from already low levels, because 1) demand growth failed to recover and 2) supply from the seaborne market AND within China has remained strong.

Again, the China equation for thermal coal has been much poorer than anticipated.  The rebound in power demand has been tepid, with elasticities to GDP and IP shifting drastically lower in the two years.  Coal burn has also again been impacted by solid growth in hydro generation.

I also think a lot of marginal high cost supply continues to be displaced by rail expansion into key ports, removing the need for trucking over long distances.  This combined with the weakness in demand means that the top end of the cost curve has shifted down quite significantly.

There are some signs that the current level of prices is causing much more pain for Chinese producers. But the bottom line is that prices are much much lower than many would of expected a couple of years ago.

Compounding the problem for global miners is that seaborne supply has remained strong.  This is because marginal supply has proven to be much less flexible than expected.  US and Russian exports have remained at high levels even though producers are making a loss.

Australian export growth has motored on given the long dated nature of capex and the cost of using increased infrastructure, but it would be wrong to single out Australian export growth as the defining reason for seaborne market weakness.  For example, if Australian producers reversed the 23mt of growth in 2012 and the ~20mt of expansion in 2013, its unlikely that prices would be materially higher and producers would be much worse off.

Likewise for Indonesian supply growth, which remains strong and for the most part in the money, although quality is an issue.

For me, there are 3 key lessons from thermal coal that you could apply to commodity analysis in general and that might be useful for the road ahead in iron ore.

1. Cost curves only matter to the degree which supply is flexible:  Supply has proven to be much more stubborn in most parts of the world.  If suppliers are locked into supply schedules thanks to large capex costs, then they are unlikely to cut supply.

2. First reaction to falling price = produce more? Given commodity producers are price takers (in coal at least) the initial reaction to a squeeze in margins is not to make costly supply cuts, but to improve efficiency.

This means either producing the same amount with reduced costs, or producing more at the same cost levels.  The latter is preferable to producers given the negative impact of redundancies, but for the industry as a whole means that supply growth may bring further imbalance to markets and downside risk to prices.

3. Top end of the cost curve can be rendered irrelevant fast:  Just because costs are reported to be high when demand is very strong, doesn't mean they will be when demand slackens.

In part 2, I'll provide some background on the problems that met coal are facing at present