In the closing paragraph of a recent article on US and European steel prices next year, we introduced the proposition made by Credit Suisse bank in a report to clients that long-term steel costs are likely to rise because new steel mills are not being built at the required rate to meet long-term demand growth.
The bank identifies several factors as contributing to this situation. Returns from the massive investment required for integrated steel plants is too low at current prices to incentivize steel mills to invest in such projects. The cost of new plant additions remains high, the bank says. ArcelorMittal has talked about a US $2000-2500 per ton cost of a new-build integrated mill (slab US$2000/t, and US$2500/t, including rolling). ThyssenKrupp’s cost for CSA (its slab plant in Brazil) was finalized at around US$1350/t. If this was started today, it would probably cost more like US$1500/t. Adding in rolling facilities would make US$2000-2500/t seem reasonable for an up-and-downstream operation. The bank uses a current EBITDA per metric ton of around US $100 as an example, the pretax return would be around 3.5 percent on new-build.
The capital cost of new build in China is about 50 percent of that outside the country, but major new construction is increasingly unlikely in China as the authorities seek to dissuade further investment in what is already a surplus supply market. Nor will Chinese producers be able to dump excess capacity abroad as they once did; a combination of Beijing imposed export taxes and rising protectionism will limit the ability of Chinese producers to access export markets. Nor are integrated new build steel mills easy from an approval point of view. Russia is a closed shop to outsiders and even after five years of being active investors in India, neither Posco nor ArcelorMittal have yet started work on a major integrated steel plant in India.
The capital cost of a mini mill is far lower than for an integrated mill, so surely the future is mini mills, then? EAF mini-mills are far more flexible in times of reduced demand, more environmentally friendly and do not need the same massive size to achieve economies of scale that integrated mills require. But in spite of the success of US mini-mill producers, the percentage of global mini-mill capacity has fallen from about 35 percent in 2004 to 31 percent of crude steel in 2009 for a number of reasons. The scrap market is structurally tighter than iron ore and coal, scrap is all about recycling and the global steel market recycling ratio (and hence the scrap reservoir) has fallen significantly as emerging markets consume lots of steel and produce little scrap. Furthermore, EAFs are electricity-intensive and, in a world short of power, many countries with high power costs appear reluctant to allow permission for mini-millsâ€China, for example has only 10 percent mini-mill capacity. The success of US mini-mills has, and will continue to be, a tough act to follow elsewhere.
What the bank terms de-bottling and brown-field expansion is the most cost-effective source of new capacity, costing only about two-thirds of a new build. Unfortunately the low hanging fruit has already been picked and at best the bank estimates that such investment should achieve about 1.01.5 percent capacity creep each year (1319mt globally) against a long-term requirement of 6090mt per annum.
In the bank’s estimate, the c120mt of excess capacity created by the crisis could arguably be reduced to close to zero within six years. However, the steel industry would likely go structurally tight some time before this. As a result Credit Suisse is predicting that HRC prices could and probably should peak to over US $1,200 per metric ton during that period from approximately US $600 per ton today.