An examination of iron ore market forces and their relationship with producers
Since 2014, a challenging operating environment has developed for iron ore producers. Prices have declined and growth in demand has fallen short of forecasts. Within Australia, there has been some controversy over the reasons behind these new market conditions. In particular, there has been significant media and political comment that major local iron ore producers may have manipulated the market to drive out competitors or engaged in suboptimal market behaviour requiring rectification through policy intervention. Is there any truth to this commentary?
The response of Australia’s iron ore producers
Cycles, or irregular fluctuations between periods of expansion (booms) and contraction (recessions), are endemic to many industries, and indeed, to whole economies. It is generally true that more capital-intensive industries, like seaborne iron ore supply, experience larger cyclical fluctuations than consumer-oriented or service industries.
As a consequence, iron ore producers tend to discount many increases in demand as being temporary and not capable of underwriting profitable capacity expansions. In addition, the capital intensity, and especially the large economies of scale in the seaborne iron ore industry, implies that it may be optimal to delay expansions until larger ones become economic.
During the recent boom, it unavoidably took time for the Australian iron ore market to respond to rapid growth in Chinese demand while it determined how much of the growth was permanent. It was only when the growth in demand was large enough and sustained for a long enough period of time that Australian producers found it attractive to make the large investments required to grow supply (Figure 1).
While China’s growth in demand for iron ore took off in the early 2000s, the iron ore price remained fairly stable (Figure 2) and in fact declined in real terms in 2002. During this period, Australia’s major producers, BHP Billiton (BHP) and Rio Tinto (Rio), responded to this rising growth in demand for seaborne iron ore by drawing additional capacity from existing systems through increased efficiency and stockpile draw down. In 2003, BHP stated that its port operations were running at levels around ten per cent higher than ‘capacity’ in response to increased demand (Matthys, 2003). Rio indicated that production from its mines was stretched to achieve ‘record levels’, placing ‘cost and other operating stresses on the production system’ (Rio Tinto, 2003).
In 2004, iron ore prices took off in a manner that signalled that a persistent iron ore shortage had developed (Figure 2). The iron ore price rose rapidly from this point as the willingness of steelmakers to pay more to secure supply increased. At the same time, China’s demand growth continued to exceed all forecasts, providing confidence that there was a structural shift occurring in the country’s steel production and consumption.
These factors encouraged Australian producers to believe in a long period of sustained growth from China. The Chinese experience also encouraged people to predict the similar emergence of other high-population developing economies such as India and Brazil as sources of substantial new demand. The result was a decade of unprecedented investment activity in the Australian iron ore sector. Between 2004 and 2014, BHP spent US$29B and Rio spent US$33B on expansions and new mine development (Figure 3).
Market forecasts were also sufficiently optimistic to attract new entrants to the local industry. Fortescue Metals Group (FMG), Atlas Iron, BC Iron, Arrium Mining and Hancock Prospecting all joined the ranks of iron ore producers during the boom. By far the most significant of these new producers was FMG. From 2004 to 2014, new entrants invested a total of US$31B in the Australian iron ore industry, of which US$21B was from FMG. The result of all this investment was a near fourfold increase in Australian iron ore production (Figure 4).
The addition of new supply during this period changed the industry’s short-run marginal cost curve (Figure 5). The curve initially rose as higher-cost producers entered the market and lower-cost producers incurred additional costs as they pushed infrastructure harder. This was reversed with the subsequent addition of lower marginal cost production after large capital investments by industry majors. The position of BHP and Rio on the industry cost curve clearly shows that they have a marginal cost advantage compared to their competitors, both locally and abroad.
In 2010, China’s growth in demand for iron ore began to slow for the first time since 1998, and in 2011, the iron ore supply growth rate exceeded the demand growth rate. This was the tipping point for the iron ore price, which reached its peak in February 2011. Reversion to real prices that were more consistent with historical norms began as supply balanced demand.
In response to this price reversion, BHP did not approve any iron ore growth investments after 2011. In 2012, it announced the deferral of 180 Mt/a of additional production. While still growing production, Rio stepped back from a 420 Mt/a target to 360 Mt/a and shelved a number of capital projects in favour of efficiency and productivity improvements. FMG grew its production to achieve its targeted 165 Mt/a in 2015 and is now focused on ‘running efficient cost effective operations’ (FMG, 2015).
Just as it takes time during the ‘take-off’ phase to determine if demand growth rates are permanent, it unavoidably takes time to determine when that growth rate has plateaued. Continued supply expansions, or lack of contraction, in the face of declining metal prices can be explained by several market factors.
The first is the completion of capital commitments made during a higher price environment. The development time required for capital intensive iron ore projects can be several years. Once capital is committed and spending is in progress, it is generally completed. This behaviour is not exclusive to the major producers.
A second reason is that higher marginal cost producers continue to produce, even when they cannot cover their marginal costs. If high-cost producers believe that the downturn is temporary, continuing to produce at a loss may be better than paying the large fixed costs associated with closing and reopening a mine. Only after producers accept that prices are likely to stay below their marginal costs for a substantial period of time do they find it worthwhile to exit. The higher supply in the meantime exacerbates the price decline.
A third reason for continued supply despite falling prices is that some high-cost producers may be kept in business for longer than normal with non-competitive intervention. Prominent firms in this category would be the many Chinese domestic companies that have government backing. Another, more subtle, example would be Indian firms insulated from external prices by their export ban.
Investigating the market manipulation hypothesis
BHP and Rio dominate production in the Australian iron ore market. Combined, they own 62 per cent of Australian production capacity and provide 38 per cent of annual global seaborne exports. There is little doubt that these large producers are aware that their production and expansion decisions influence the market price of iron ore. A situation of dominance like this is not uncommon in an industry that requires large and costly infrastructure to supply the market.
But just because BHP and Rio could influence iron ore prices, was it in their best interests to do so? In particular, can the recent price decline be cited as evidence that BHP and Rio strategically expanded supply in a deliberate attempt to force lower prices in order to pressure smaller producers to exit the market? Typically, one would expect firms exercising monopoly power to reduce output and drive prices higher in order to maximise profits.
Given the size of their reserves and their access to capital, both BHP and Rio could increase supply to a level that would force competitors out of the market. But this action would only represent half of a profit-maximising strategy. The second half would require them to realise higher prices once competitors were forced from the market by ensuring that supply reductions were maintained or further limited.
This presents a significant challenge, and there are a number of reasons why it is unlikely to be a profit-maximising strategy.
1. Major Australian iron ore companies cannot effectively constrain supply
The major Australian producers compete directly against each other for customers and market share. Neither BHP nor Rio have sufficient market share to independently control the market, meaning that collusion would be required. However, collusion between private companies is an illegal practice in their operating country (Australia), their listing countries (Australia, the UK and the USA) and a number of their consumer countries (Japan, Korea and the EU nations).
In 2009, BHP and Rio attempted to enter into a 50-50 iron ore production joint venture, combining the operations and management of their Western Australian iron ore assets. The merger was formally scrapped in 2010 after being met with enormous opposition from global regulators and steelmakers, including the European Commission, the Australian Competition and Consumer Commission and the Japan Fair Trade Commission amongst others.
Even if collusion were legal, it is unlikely to be successful without including additional parties. BHP and Rio control 38 per cent of the annual seaborne iron ore export market. That control would increase to 49 per cent in combination with FMG or to an impressive 68 per cent in collusion with Vale. But each company is only likely to participate in such a cartel if it believes that it can extract more benefit from it than from a competitive free market.
The fundamental challenge with this is that cartels require production to be reduced while creating a strong incentive for each member to expand output to benefit from the higher prices. For this reason, most cartels of private firms do not succeed. Cartels such as the Organization of the Petroleum Exporting Countries (OPEC) have persisted because they involve sovereign countries controlling national resources. However, even OPEC has to contend with continual cheating on production quotas by its members.
2. Competitors may not exit the market as predicted
The purported manipulation strategy relies on higher-cost competitors exiting the market when prices fall. However, this may not eventuate for several reasons or take so long to occur that the strategy would not be profitable in net present value terms.
First, when prices drop, so do the costs of many variable inputs, which means that operating costs will decline. Firms with higher variable costs will benefit more from drops in input costs and may be able to reposition themselves on the cost curve and maintain profitable operations.
Second, if competitive producers believe that the low prices are only temporary and that prices will subsequently be raised to extract monopoly rents, they may accept short-run operating losses and not exit the market. This is of particular relevance in today’s market, where there is still the possibility of demand take-off from growth in other emerging markets. Finally, government backing, through policy intervention or financial assistance, may allow otherwise unprofitable operations to continue producing.
In the absence of a quick exit of competitors, further pressure is created on the iron ore price, increasing the future earnings that the majors would need to realise in the event that they were able to implement the required output restrictions and raise the iron ore price.
3. Alternative supply sources and substitutes exist
If supply restrictions established by major Australian producers subsequent to the exit of competitors resulted in an iron ore price rise, this could stimulate supply expansion by global competitors or encourage the use of alternative products.
High prices could encourage the expansion of existing global competitors (eg Vale, Kumba) or the development of entirely new iron ore provinces (eg Guinea, Sierra Leone) where the high fixed start-up cost currently presents a barrier to entry.
Any attempt to substantially raise prices could trigger Asian steel firms that previously invested in the Australian iron ore industry to invest in other locations to avoid the consequences of monopolistic behaviour. This already occurs in the market to some degree as China tries to diversify its supply options. China recently committed US$4B to help finance Vale’s S11D expansion project. It is also committed to funding a share of the US$20B high-grade Simandou project in Guinea.
High iron ore prices could also stimulate the use of substitutes such as scrap steel in place of iron ore as an input or the invention of new supply or demand technologies to get around the resulting higher prices for steel – for example, the use of plastics, carbon fibres and aluminium in the developed world.
A recent McKinsey study predicted that by 2020, China will be in a position to make a serious shift towards the use of scrap metal as infrastructure from the first building boom reaches the end of its life (Zeumer and Bekaert, 2013). Processing technology could shift from raw steelmaking to steel recycling facilities if the former became more costly.
While such alternatives may take years to develop, BHP and Rio have reserves equal to many years’ supply of iron ore that would decrease in value if suitable alternatives became available.
4. The complex relationship between iron ore producers and consumers
Many steelmakers jointly participate in iron ore mining ventures and are privy to information about the profitability and strategy of the operations. This greatly constrains the ability of iron ore producers to manipulate production to their advantage and to the disadvantage of their customers.
Both BHP and Rio have shared ownership arrangements with the consumers of their products. In 2014, 15 per cent of BHP’s Australian production and 13 per cent of Rio’s Australian production was owned by Asian partners. These alliances would mean that both companies would face significant long-term damage to their reputations with partners were they seen to be manipulating supply and/or price to the detriment of their partners.
There is no denying that BHP and Rio have significant influence in the Australian and global iron ore markets. However, it is unlikely that there has been a ‘strategic’ oversupply of iron ore by either company aimed at removing local competition. The primary reason for doubting this explanation for the recent decline in prices is that this strategy is very unlikely to be profitable. To succeed, it would almost surely require illegal collusion between otherwise competing companies and implausible future reactions by other producers and consumers.
The response of the major Australian producers to the recent market boom was similar to the response of new local market entrants and major overseas competitors. All added similar production tonnage, spent similar capital and grew reserves by similar amounts. It is hard to single out the Australian majors as having behaved in a suboptimal fashion given that the response was the same across the industry.
This article is based on a UWA Business School Discussion paper titled ‘Riding the Iron Ore Cycle: Actions of Australia’s Major Producers’ DP16-15.
Fortescue Metals Group (FMG), 2015. ‘2015 Annual Report’. http://fmgl.com.au/media/2590/fortescue-annual-report-fy15.pdf
Matthys J, 2003. ‘BHP Billiton Carbon Steel Materials Group – A diversified resource base for China’s Steel Industry’. Third China International Steel and Raw Materials Conference. http://www.bhpbilliton.com/~/media/bhp/documents/investors/reports/2003/juliusmatthypresentation.pdf
Reserve Bank of Australia (RBA), 2014. ‘Statement on Monetary Policy – August 2014’. http://www.rba.gov.au/publications/smp/2014/aug/box-b.html
Rio Tinto, 2003. ‘Annual report and financial statements. SEC Washington DC 20549 Form 20-F’. https://www.sec.gov/Archives/edgar/data/863064/000102123104000209/b743424-20f.htm#p37
‘Steel Statistical Yearbook’, World Steel Association. http://www.worldsteel.org/statistics/statistics-archive/yearbook-archive.html
Zeumer B and Bekaert F, 2013. ‘Steel and the challenges of volatile raw materials, flat margins, and overcapacities’. McKinsey on Metals and Mining. Scarcity and Saturation, Number 08, Spring 2013.