Japanese investment in Australian coal assets through the demise of concessional financing Jason West n Griffith University, Department of Accounting, Finance and Economics, Brisbane, QLD 4111, Australia HIGHLIGHTS c Past foreign investment practices sacrificed profit to create oversupply of Australian coal. c Only a small amount of equity capital was required to exert influence over coal output. c Foreign investors can no longer exploit information advantages to obtain favourable prices. c Information advantages from partial ownership do not influence supply costs. c Foreign investors in Australian mining now achieve similar profits to domestic firms. article info Article history: Received 20 January 2011 Accepted 28 September 2012 Available online 22 October 2012 Keywords: Mining Foreign direct investment Energy markets abstract The Australian coal industry has been described as being a perpetual case of ‘profitless prosperity’. This implies that foreign companies invest in low-margin mining activities with motives other than profit. It is argued that foreign investors and Japanese trading companies in particular used government investment concessions and subsidies to help create oversupply in the seaborne coal market. The aim of this strategy is to depress contract prices at the cost of achieving reasonable profitability levels, which have historically been well below that of other investors in the Australian mining sector. This study shows that the quasi-integration via concessional funding arrangements is not a credible strategy employed by Japanese trading companies or the Japanese Government. The analysis rejects the hypothesis that via foreign direct investment, Japanese companies are securing coal at below average prices. Furthermore we find no clear evidence of Japanese trading companies using their information advantage as equity investors to secure coal at favourable prices during contract negotiations. Finally we examine the investment behaviour of new entrants in the Australian coal production sector to highlight the differences in investment strategy between Japanese companies and other foreign investors regarding the security of supply. & 2012 Elsevier Ltd. All rights reserved. 1. Introduction Black coal production in Australia has risen dramatically over 1997–2011. Production for both domestic consumption and export has doubled in fifteen years to around 420 million tonnes per annum in 2011. This is largely due to increased demand for raw materials needed for energy and steel production in Asia and the reversal of China as a coal exporting nation to a major coal importer coupled with static consumption levels in other major import centres such as Japan, South Korea, Taiwan and Western Europe. The growth has been partly driven by an increase in liquidly traded contracts for seaborne coal helped by the availability of standardised contracts and common coal trading platforms. There has also been a shift from annual negotiations to quarterly or monthly index price fixing for coking coal supply contracts. This growth has also seen foreign ownership of Australian coal mining rise from 43 per cent in 1997 to around 75 per cent in 2010 with profitability levels steadily increasing over the same period. This has occurred despite a physically constrained supply chain. Japanese investors comprise around 20 per cent of the Aus- tralian coal mining sector and Japanese interests have vastly dominated the interests of all other foreign investors in Australia since the 1970s. Given Japan’s dominant position in the Asia– Pacific coal market it was shown by Colley (1997) that the production growth driven by Japanese trading companies, using concessional funding from Japanese Government agencies, cre- ated an oversupply in the market to depress coal supply contract prices. The cost of pursuing this strategy was a period of very low Contents lists available at SciVerse ScienceDirect journal homepage: www.elsevier.com/locate/enpol Energy Policy 0301-4215/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.enpol.2012.09.075 n Tel.: þ61 737 354 272. E-mail address: j.west@griffith.edu.au Energy Policy 52 (2013) 513–521
Transcript
Energy Policy 52 (2013) 513–521
Contents lists available at SciVerse ScienceDirect
Energy Policy
0301-42
http://d
n Tel.:
E-m
journal homepage: www.elsevier.com/locate/enpol
Japanese investment in Australian coal assets through the demiseof concessional financing
Jason West n
Griffith University, Department of Accounting, Finance and Economics, Brisbane, QLD 4111, Australia
H I G H L I G H T S
c Past foreign investment practices sacrificed profit to create oversupply of Australian coal.c Only a small amount of equity capital was required to exert influence over coal output.c Foreign investors can no longer exploit information advantages to obtain favourable prices.c Information advantages from partial ownership do not influence supply costs.c Foreign investors in Australian mining now achieve similar profits to domestic firms.
a r t i c l e i n f o
Article history:
Received 20 January 2011
Accepted 28 September 2012Available online 22 October 2012
The Australian coal industry has been described as being a perpetual case of ‘profitless prosperity’. This
implies that foreign companies invest in low-margin mining activities with motives other than profit.
It is argued that foreign investors and Japanese trading companies in particular used government
investment concessions and subsidies to help create oversupply in the seaborne coal market. The aim of
this strategy is to depress contract prices at the cost of achieving reasonable profitability levels, which
have historically been well below that of other investors in the Australian mining sector. This study
shows that the quasi-integration via concessional funding arrangements is not a credible strategy
employed by Japanese trading companies or the Japanese Government. The analysis rejects the
hypothesis that via foreign direct investment, Japanese companies are securing coal at below average
prices. Furthermore we find no clear evidence of Japanese trading companies using their information
advantage as equity investors to secure coal at favourable prices during contract negotiations. Finally
we examine the investment behaviour of new entrants in the Australian coal production sector to
highlight the differences in investment strategy between Japanese companies and other foreign
investors regarding the security of supply.
& 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Black coal production in Australia has risen dramatically over1997–2011. Production for both domestic consumption andexport has doubled in fifteen years to around 420 million tonnesper annum in 2011. This is largely due to increased demand forraw materials needed for energy and steel production in Asia andthe reversal of China as a coal exporting nation to a major coalimporter coupled with static consumption levels in other majorimport centres such as Japan, South Korea, Taiwan and WesternEurope. The growth has been partly driven by an increase in liquidlytraded contracts for seaborne coal helped by the availability of
ll rights reserved.
standardised contracts and common coal trading platforms. Therehas also been a shift from annual negotiations to quarterly ormonthly index price fixing for coking coal supply contracts. Thisgrowth has also seen foreign ownership of Australian coal miningrise from 43 per cent in 1997 to around 75 per cent in 2010 withprofitability levels steadily increasing over the same period. This hasoccurred despite a physically constrained supply chain.
Japanese investors comprise around 20 per cent of the Aus-tralian coal mining sector and Japanese interests have vastlydominated the interests of all other foreign investors in Australiasince the 1970s. Given Japan’s dominant position in the Asia–Pacific coal market it was shown by Colley (1997) that theproduction growth driven by Japanese trading companies, usingconcessional funding from Japanese Government agencies, cre-ated an oversupply in the market to depress coal supply contractprices. The cost of pursuing this strategy was a period of very low
profitability, particularly during the 1990s. Koerner (1998) andSwan et al. (1999) showed that benefits of being a minority equitypartner in a producing mine such as access to better informationon marginal production costs and capital returns also assistedJapanese trading companies during contract price negotiations.They used hedonic regression to investigate behaviour in annualcoking coal negotiations and showed that inefficient pricingconsistently occurred in the coking coal market due to informa-tion advantages.
D’Cruz (1983) examined the impact of so-called quasi-integra-tion resulting from Japanese trading companies’ establishment oflong-term purchasing agreements for coking coal supplies on theprice and volumes purchased from producers over the 1970s. Hishypothesis was that quasi-integration would assuage the influ-ence of market power during cyclical phases of supply anddemand imbalance. The results demonstrated that during periodsof coal producer dominance (during periods of steel productiongrowth), coking coal producers directly linked with Japaneseconsumers would receive lower prices and experience greatervolume stability than independent coal producers, thereby ben-efiting from quasi-integration. Conversely during periods ofindustry downturns quasi-integrated coal producer contractprices were shown to be higher than for an independent coalproducer, while also achieving superior off take volumes. D’Cruzconcluded that any positive effects of quasi-integration on pricewere minor compared with the detrimental effects of pricediscrimination practised by the Japanese firms in the Pacific coalmarket. Colley (1997) updated this thesis to show the existence ofoversupply was driven by low profitability targets of Japanesetrading companies across both thermal and coking coal marketsover the period 1986–1996.
Despite the boom in Australian coal production over 1975–1997 Colley (1997) showed that average profitability and returnon capital in the coal mining industry were much lower thanother extractive sectors. Since 1998 and more noticeably over2005–2011, coal production and profitability have both grown ata much higher level than other extractive sectors. The character-istics of the mining industry globally have changed significantlyover this period due to a surge in commodities demand from largeemerging economies, and it is claimed that this has altered theinvestment behaviour of foreign trading companies looking toacquire ownership interest in Australian mining assets.
This study will analyse the use of concessional funding fromgovernment agencies by Japanese trading companies over 1998–2011 to determine if a quasi-integration strategy continues toachieve below average prices and volume stability, as well asinformation advantages for use in contract price negotiation.Furthermore we will also investigate whether such strategiescontinue to be employed at the expense of profitability andcapital return when measured against peers in Australia. Finallywe will examine the investment behaviour of new entrants to themarket from South Korea, China and India to determine if priceand cost information advantages serve as the main motive behindtheir acquisition strategies, or if other reasons such as the securityof supply are driving foreign investment in coal assets.
2. Preliminaries
2.1. Current industry status
The majority of Australian coal production is lower qualitythermal coal. In 2010 thermal coal was 56 per cent of total outputrepresenting 208 million tonnes for both domestic consumptionand export (International Energy Agency (IEA), 2011). Metallurgi-cal coal production in 2010 was 165 million tonnes (International
Energy Agency (IEA), 2011). Growth in coking coal productionover this period has been driven by rising exports of hardcoking coal (HCC) and coal suitable for pulverised coal injection(PCI) technologies, with a more modest growth in semi-softcoking coal.
Queensland accounted for 57 per cent of Australian black coalproduction in 2010 with over half of the state’s output beingmetallurgical coal sold on the export market. New South Walesaccounted for 40 per cent of Australian black coal production in2010, the majority of which is thermal coal sold for export.Relatively small volumes of thermal coal are produced in SouthAustralia and Western Australia driven by local power stationdemand with small exports. The Bowen Basin in Queensland isthe largest coal producing basin in Australia, accounting for 49per cent of total output in 2010. The second largest producingarea is the Sydney Basin in New South Wales which accounted for38 per cent of the total.
Infrastructure has been the main constraint on export volumesover 1998–2011. The utilisation of Australia’s export infrastruc-ture, as indicated by the ratio of actual exports to nominal portcapacity has spiked to around 80–85 per cent of nominal portcapacity. Expected growth in global demand is driving higherinvestment in coal mining operations in both Queensland andNew South Wales with a total net capital expenditure of US$6.6bn per annum increasing to over US$10.4 bn per annum over the2007–2010 period (Australian Bureau of Statistics, 2010).
New grants for coal tenements in Australia have more thandoubled over the period 2007–2011. New lease areas were domi-nated by exploration permits in Queensland, mostly in the well-established Bowen Basin due to the availability of high-quality andtherefore high profit margin hard coking coal. Tenements withdeeper coal seams that require more difficult extraction techniqueshave recently been shown to be profitable to mine, particularlywith the recent advancement in underground mining technologies.Emerging coal basins such as the Galilee and Surat in Queenslandare also attracting considerable exploration interest although theymostly contain lower margin thermal coal.
The rapid growth of investment has continued despiteincreases in state-based royalties and Federal resource taxes.Where the ownership of coal is vested in the State, the relevantState government receives a royalty payment for the right to minethat coal. Royalty charges are applied to the value of productionafter beneficiation (total revenue less allowable deductions). Forresource taxes the allowable deductions are limited to beneficia-tion, port charges, despatch and demurrage and research anddevelopment costs, while freight and marketing costs are non-deductible. Australian mining operations are also subject to anumber of other levies such as the coal research levy, the minesafety levy, the mine subsidence levy and the mines rescue levy.However, the imposition of royalties, levies and resource taxesappears to have had little impact on the investment growth inAustralian coal mining over 1975–2011.
2.2. Industry structure and Japanese investment
The Australian coal industry has a high degree of foreignownership with over 75 per cent of production generated fromforeign-owned companies in 2009. The concentration of coalproduction is high and has been dominated by four majorcompanies who account for 65 per cent of exports by value. Butactual ownership within the industry continues to be diverse asmine-sites are typically incorporated under a variety of jointventure agreements among domestic operators and foreigninvestors. A useful measure of market concentration and there-fore competition is the Herfindahl–Hirschman Index (HHI),defined as the sum of the squares of the market shares of all
J. West / Energy Policy 52 (2013) 513–521 515
firms in a given industry. The result is proportional to the averagemarket share weighted by market share which results in anoutput that ranges from 0 to 1. For the Australian coal industrythe HHI measure was approximately 8.882 per cent in 2010which has fallen from 10.317 per cent in 2005. This indicatesthat there is very little concentration among producers.
There is also a significant level of ownership volatility in theindustry with a high rate of merger and acquisition activity,especially among coal-exploration companies who look to sellassets that have been extensively surveyed and are ready for minedevelopment and production.
Of the foreign investors the Japanese trading companiesconstituted 19.3 per cent of total production in 2010. Thistranslates into a total share of 14.5 per cent across the entireindustry. This share has remained relatively static since 1997meaning that Japanese investment in the industry, along withother investors and developers, has grown in absolute terms ataround 5.7 per cent per annum. While some companies hold amajority interest, most Japanese investors maintain an influentialbut essentially minor interest in coal assets. The Australia–Japanproducer–consumer coal trade relationship has developed overseveral decades and stems from Japan’s strategic need to secure along-term supply of raw materials. An extensive history isprovided in Barnett (1994). Prior to 1997 coal industry growthwas relatively weak, with seaborne coal prices, hence profit-ability, at very low levels compared with other industries(Barnett, 1994).
Table 1Excess capacity in seaborne hard coal to Japan, Mtpa (source: IEA 2009, BHP
Billiton).
Thermal coal Coking coal
2.3. Commodity price increases and industry effects
The contract prices for both coking and thermal coal haveincreased markedly from 2006 to 2011, primarily driven by netimports of each type of coal to China and India, both of whomprior to 2008 were net exporters of coal. The commodities boomin 2008 triggered increases in average CIF import thermal coalprices by 77 per cent and CIF import coking coal prices by 108 percent in Japan. Some contributing factors to the price spike includetight supply induced by record consumption growth, coupledwith infrastructure and production inefficiencies in major export-ing countries and a halt to Chinese exports to meet domesticneeds. The increase in thermal coal contract prices traded throughthe GlobalCOAL platform in 2005 as well as the switching ofcoking coal contract negotiations from annual to quarterly in2010 has enhanced the liquidity of the global trade in theseproducts. This has established a more efficient market-basedpricing mechanism for exports of Australian coal. Fig. 1 illustratesthe free-on-board (FOB) price of both Queensland hard cokingcoal contracted to Japan and Newcastle thermal coal in 2010 USdollars.
-
50
100
150
200
250
300
1980 1984 1988 1992 1996 2000 2004 2008
US$
/t FO
B C
ontr
act P
rices
(in 2
010
USD
)
QLD Hard Coking CoalThermal Coal
Fig. 1. Queensland hard coking coal and Newcastle thermal coal price history
1980–2010 in 2010 US dollars.
3. Japanese capital
Colley (1997) noted that the distinction in capital formation ofthe parties in the Australia and Japan coal trade is vital forunderstanding the reasons behind the oversupply and depressedprices during the 1990s. Colley claimed that quasi-integration,which is investment that only occurs in one direction such asJapanese-based coal consumers taking minority equity or jointventure stakes in Australian mines, is the primary cause of supplyimbalance. Japanese shareholders display strong patterns ofcross-shareholdings (Koerner, 1998) which have persisted over1997–2011 as opposed to the more diffuse ownership structure ofmost Australian mining firms. Furthermore the fragmented andvolatile ownership structure of the Australian industry lacks‘coherence’ relative to the tightly coordinated and stable owner-ship structure of Japanese trading companies.
These general conditions are similar to the conditions in the1990s when Japanese investment in the Australian mining sectorsupported via loan concessions from the main Japanese exportagencies was at a peak. Koerner (1993) and Colley (1997) showedthat excess capacity was a feature of world coal supply anddemand conditions from 1985 to 1995. This analysis howeverignores the spread of coal qualities across world production,especially as Japan is a buyer of high quality coal with strictspecification limits on levels of ash, sulphur and nitrogen andvolatile matter. Demand and supply conditions for Japaneseconsumers have in fact largely been in balance from 1995 to2009, with some excess capacity leading to slightly lower pricesevident during the financial crisis in 2008. This is illustrated inTable 1.
Coal sold into the Japanese market usually attracts a premiumto the benchmark price due to the greater limitations on qualityspecifications which requires greater processing. It remains fea-sible that global excess capacity depressed prices significantly inthe 1990s but this excess has not persisted in the 15-year periodbeyond 1995.
4. The practice of quasi-integration
Colley (1997) and Koerner (1998) concluded that that thepractice of quasi-integration of Japanese-based companies andthe co-ordinated strategy behind it were a significant contribut-ing factor to the poor profit performance of the Australian coal
Fig. 2. JBIC loan commitment to Australia, global natural resource development
and global investment loans for natural resources in f billion 1999–2010.
Table 2Direct correlations between growth in JBIC commitments to global funding for
natural resource projects, global natural resource development and funding
Australian mining investment, 1998–2011.
Global NRloans
Global NRdevelopment
Loans toAustralia
Global NR loans 1.000
Global NRdevelopment
0.601 1.000
Loans to Australia �0.317 �0.585 1.000
J. West / Energy Policy 52 (2013) 513–521516
industry from 1986 to 1996. Colley specifically focussed on theco-ordinated set of activities between private Japanese companiesand the Japanese Government to explain the effect of quasi-integration on profits from Australian coal mining and demon-strated that the profit performance of these investments waspoor. By corollary so was the performance of equity partners inthe same projects.
These studies drew upon and emphasised the argumentclaimed in D’Cruz (1983) and Anderson (1987) that the coal tradebetween Japanese industry and some of its major suppliersexhibits the characteristics of so-called quasi-integration. Quasi-integration is referred to here as the existence of elements ofvertical integration limited by fact that the degree of commonownership is less than a controlling interest. It was shown inAnderson (1987) and that the primary purpose behind this formof vertical integration is to achieve an oversupply of a particularresource in order to depress prices over an extended periodof time.
Dowling (1987) and Koerner (1998) suggest that there areother motives and advantages for Japanese trading companiesresulting from minority equity investment in coal mining opera-tions. These include access to greater knowledge of marginalproduction costs and capital returns gained from partial owner-ship in producing mines. The trading arms of the Japanese firmsmay make this knowledge available to their respective negotiat-ing team who bargain with individual Australian producers onbehalf of the consumption arm of their firm. Koerner (1998)demonstrated that this arrangement significantly contributed tothe depressed prices for coking coal during the period 1988–1998.
Japanese general trading companies, also known as sogo
shosha, have traditionally maintained strong links with govern-ment. Strong profit performance has not been the primaryconsideration for typical Japanese trading houses as they tradi-tionally operate on very thin margins, in the order of a gross profitof rarely more than 3 per cent. They are renowned for their lowprofitability despite significant diversity in balance sheet assets.A study by the Australia–Japan Economic Institute in 1996 rankednet profit against the global trading volumes of each company toshow that none of the nine largest Japanese trading housesachieved a net profit greater than 0.35 per cent of sales between1989 and 1995, while the average was 0.1 per cent (Australia–Japan Economic Institute, 1996). Up to the late 1990s Japan’sgeneral trading houses traditionally made their money from high-volume, low-margin transactions in raw materials and commod-ities. But these activities no longer offer much opportunity forgrowth, and the trading companies, especially the larger ones,appear to be investing in more profitable undertakings in man-ufacturing and infrastructure projects, and where possible,mining projects.
The Japanese Government assists Japanese trading companiesto privately invest in assets abroad using a range of concessionalfinancing instruments from a variety of agencies including theJapanese Ministry of International Trade and Industry (MITI), theJapan Development Bank, the Export-Import Bank of Japan, theOverseas Economic Cooperation Fund (OECF) and the New Energyand Industrial Technology Development Organisation (NEDO)which is a subsidiary of MITI. After the promulgation of the JapanBank for International Cooperation (JBIC) Act in April 1999 JBICwas established and assumed the operations of the Export-ImportBank of Japan and also those of the Overseas Economic Coopera-tion Fund (OECF). JBIC has assumed the formal role of facilitatingJapanese trading company investment abroad.
The concessional financing strategy or kaihatsu yunso policypromulgated in the MITI White Paper in 1970 aimed to enableJapanese companies to place a larger reliance on direct investmentin foreign ventures to augment long term supply agreements, take
advantage of economies of scale and invest in large scale develop-ment projects and diversify investments across regions.
Fig. 2 illustrates JBIC loan and investment commitment toAustralia compared with global commitments for naturalresource development and investment loans in natural resourcesfrom 1999 to 2009. Committed funds increased steadily to 2006and then stabilised. Japan’s public sector commitment wasaround f2300 bn from 2005 to 1908 with around f2800 bncommitted in 2009. This translates into an average commitmentof around US$25–33 bn in foreign natural resource investment.Much of this was committed to Asia and Africa and only arelatively minor amount was committed to Australian-basedprojects over 1999–2009, even accounting for the large invest-ment made in 2008. The overall diminishing level of investmentsuggests that the strategy of quasi-integration investment prac-tices in Australia has been discontinued, despite record coal priceincreases over the latter half of the period. Coal mining inparticular has not received substantial JBIC investment loans forover a decade. The correlation between investment growth inglobal natural resource projects and other development activities,and growth in Australian investments is significantly negative.The correlations between each investment category are providedin Table 2.
The relationship between the year-on-year change in resourceprices and JBIC loan commitments abroad over 1998–2011 can betested. We conducted a regression to test the predictive capacityof loan commitments from JBIC for global resource developmentand project loans and found them to be inversely related to thechange in hard coking coal (HCC) prices, thermal coal prices andthe Goldman Sachs Commodity Index (GSCI) over 1998–2011. TheGSCI is the global benchmark for investment in the commoditymarkets and is a tradable index on the Chicago MercantileExchange. The GSCI does not include HCC or thermal coal in itsconstruction but represents most other energy commodities aswell as metals and agricultural commodities. The change in JBIC
J. West / Energy Policy 52 (2013) 513–521 517
investments in Australian projects however was found to bestatistically significant and positively related to the change inHCC and thermal coal prices over 1998–2011 but there was nosignificant link with the GSCI.
Table 3 provides the results. While each price series wasmodelled separately, the regression results are reported together.Significance at the 1 and 5 per cent levels are indicated by n and y,respectively with t-statistics in parentheses. The correlationbetween HCC and thermal coal prices is significant at over 0.65however the correlation between both types of coal prices and theGSCI is insignificant at around 0.08.
The results suggest that growth in coal prices has motivatedJBIC to grow its investment in Australian resource projects.However when we eliminate investments in non-coal assets andmeasure JBIC investment growth in coal assets only against theyear-on-year change in HCC and thermal coal prices, we observe astatistically significant negative relationship as indicated in thefinal line of Table 3. Similarly global JBIC investment commitmentgrowth appears to be negatively related to global resource pricesas measured by the GSCI.
Table 4 shows that while natural resource funding commit-ments from JBIC have increased to over 25 per cent of totalcommitments, funding and financial guarantees made available tocoal mining activities has declined from over 7 per cent to around2 per cent as a proportion of natural resource funding in 2009.Over the period 2007–2009 JBIC invested in three coal minesglobally for a total amount of less than f15 bn (US$200 m). Thislevel of funding would be insufficient to fund the development of
Table 3Regression of the year-on-year change JBIC commitments to global funding for
natural resource projects, global natural resource development and funding
Australian mining investment, 1998–2011.
Variable HCC (US$/t) Thermal (US$/t) GSCI
Global NR loans �0.088 �0.151n�0.270n
(�0.957) (�2.261) (�2.473)
Global NR development �0.427n�0.379n
�0.659n
(�2.665) (�2.683) (�2.829)
Loans to Australia 4.802y 4.542n 3.541
(2.084) (2.326) (0.865)
Loans to Australia (coal) �3.667n�2.956n n/a
(�3.012) (�3.461)
Table 4Natural resource loans by item, Japanese Bank for International Cooperation
Financial Report FY2010 (units: billions of yen, %).
a small coal mine in Australia, let alone significant enough tocreate excess supply to depress global prices.
JBIC overseas investment assistance in the form of loans,investments and guarantees have grown from f523.7 bn (43 percent of total commitment) in 1999 to f2193.7 bn in 2009 (65 percent of total commitment). For natural resources development inAustralia the outstanding total of funds increased from f1135.2bn in 1999 to f1645.4 bn in 2009. It is clear that while theabsolute level of funding for Japanese projects has increased inabsolute terms, the amount made available for foreign coalmining activities has dramatically decreased since the 1990s.Resource investments have increased but no significant financialassistance from JBIC has been invested specifically in Australiancoal assets over the period 2004–2011.
Table 5 illustrates that funding available for natural resourcedevelopment and overseas investment loans represents around25 per cent of the total commitment made by JBIC in 2009, whichhas increased from around 14 per cent of the total in the ten yearsprior. The growth from 1999 to 2009 has been gradual and fundsavailable for natural resources have increased in both absoluteand relative terms for investment in oil, natural gas and iron oreproduction. Over the same period Japan has not significantlydecreased its consumption of coking or thermal coal. The conclu-sion reached in Colley (1997) and Koerner (1998) that thedeployment of concessional Government-backed funding for thequasi-integration of Japanese trading companies to depress coalprices does not appear to have been a major strategic element ofJBIC since at least 1999.
The decline in funding coal investments in Australia com-menced prior to the commodity price boom in 2007–2008 as perFig. 1. Seaborne coal prices remained relatively low during theperiod 2000–2004 followed by increasing profitability from 2004to 2011. The recent price growth has stimulated global invest-ment interest in Australian coal assets. Increased investment byJapanese trading companies has occurred despite the lack of JBICparticipation over 1998–2011. This suggests that Japanese firmsare seeking greater investment returns in contrast to the 1990sstrategy of creating excess supply to depress prices. In fact over2005–2011 a significant proportion of Japan’s Australian subsidi-aries have secured loans in the Australian domestic capital marketfor the investment in, and ownership of, coal mining assets witheither limited or no recourse funding from their Japanese parent.This represents a clear shift in the funding practice of Japanese
Table 5Outstanding liabilities by purpose of financing Japanese Bank for International
Cooperation Financial Reports FY1999 and FY2010 (units: billions of yen, %).
trading companies away from obtaining concessional loans orguarantees through the JBIC entity.
4.00
6.00
US$
/t
5. Profitability measures
The hypothesis that Japanese trading companies have shiftedtowards investment returns and away from coal supply strategiescan be tested. The Australian subsidiaries of Japanese tradingcompanies have experienced profitability levels that greatlyexceed those of their parent over 2005–2011. We examine thelevel of profitability as measured using both return on assets(ROA) and return on equity (ROE). Australian subsidiaries arefunded with both parent-guaranteed debt and equity and so thecapital structure of the incorporated entities is not always clear.However uncertainty around the actual degree of financial lever-age in the balance sheet can be overcome by comparing both ROAand ROE of the subsidiary and the parent. The uncertainty aroundthe leverage implicit in the ROE measure does not affect the ROAmeasure.
ROA is an indicator of how profitable a company is beforeleverage is considered and is comprised of profit margin and assetturnover. Since the figure for total assets of the company dependson the carrying value of the assets, some caution is required forcompanies whose carrying value may not correspond to theactual market value. For resource companies operating in Aus-tralia, ROA is a reasonable initial estimate for performancebenchmarking because the majority of their assets will have acarrying value that is close to their actual market value. Resourcescompanies are very asset-intensive businesses which typicallyexhibit relatively low ROA. In contrast ROE measures the rate ofreturn on the level of ownership interest (shareholder equity) inthe firm. ROE is simply the product of ROA and the degree offinancial leverage employed by the firm as measured by the ratioof assets against shareholder equity. ROE generally measures theability to generate cash internally. The degree of leverage acrossJapanese-based companies in Australia varies significantly. Whilethis may clearly distort the ROE measure for comparisons acrossthe industry, ROA is also included in this analysis.
Table 6 shows that the ROA and ROE achieved by Australian-based Japanese trading companies are on average slightly lessthan the industry average over 2005–2011. The industry averageis assumed to be the average of the producing coal companiesthat comprise the S&P/ASX 200 Resources Index. Japanese com-panies who have significant capital invested in Australia such asMitsubishi and Mitsui are earning returns on par with theindustry average. Each company in the sample has earned ROAand ROE far in excess of their respective parent over 2005–2011.Average ROA for both the Japanese trading companies and theirAustralian subsidiaries in the ten years prior to 1997 was around3 per cent and their profitability as measured by earnings before
Table 6Comparison of ROA and ROE for major Australian-based Japanese trading compa-
nies with their respective parent company averaged over FY2005–FY2011.
Company Australian subsidiary Parent
ROA (%) ROE (%) ROA (%) ROE (%)
Mitsubishi Corp 14.34 26.26 4.05 13.88
Mitsui & Co. 13.48 20.77 2.07 13.11
Marubeni 5.67 24.33 2.34 16.26
Sumitomo 5.98 20.42 2.70 14.08
Idemitsu Kosan 7.37 8.01 3.01 3.27
Itochu 8.93 13.77 2.34 9.54
Sojitz 12.11 24.91 1.51 9.12
S&P/ASX 200 (coal) 14.44 25.70 N/a N/a
interest and tax as a return on total funds employed (EBIT/TFE)was between 6.5 and 10 per cent per annum, well below theindustry average of 15 per cent (Swan et al., 1999).
There has been a sizeable shift in parent companies invest-ment return expectations of their Australian subsidiaries com-pared to that experienced over 1970–1997 (Colley, 1997; Koerner,1998). In the 1970s Yonezawa (1978) first argued that Japanesetrading companies pursued national interest objectives in procur-ing raw materials at a cost that exceeded the benefits of achievingthe lowest price possible. Colley (1997) showed that profitabilityin the coal mining sector, although low, still exceeded the averagereturns earned by their respective parent companies. The aboveevidence shows that on the basis of ROA and ROE measures, thefocus on low profitability for the purpose of achieving below averageprices and volume stability has changed to seek above averagereturns while retaining coal sector information advantages.
6. Information advantages and price negotiation
The next part of this analysis tests whether Japanese tradingcompanies use their informational advantages associated with own-ership to secure below average coal prices and greater volumestability. Thermal coal sales data from 2007 to 2009 was obtainedfrom three Australian coal mining companies who have varyingdegrees of Japanese equity ownership. The differential in observedFOB prices to FOB Newcastle benchmark prices was computed foreach shipment after adjusting for the quality differences in ash,moisture and sulphur for each contract. Fig. 3 illustrates thedifferentials for individual shipments made to non-Japanese Asian-based trading firms. Total shipment volume over this period was4.48 million tonnes. The average differential was US$2.48 per tonnemeaning that a strong discount was achieved over the period. Fig. 3also suggests that no persistent change in the level of the discount isevident.
Fig. 4 illustrates the differentials for individual shipmentsmade to Japanese trading companies. Total shipment volumeover this period was 4.1 million tonnes. The average differentialwas US$1.53 per tonne meaning that while a strong discount wasachieved over the period, it was significantly less than the non-Japanese Asian-based trading firms and often the benchmarkprice itself was achieved. The variation in sales price data is notsignificantly different between the two series, nor is theresignificant variation in the timing of the shipments.
To test for volume stability we conduct a simple distributionanalysis of each time series. Fig. 5 illustrates the q–q plot for coalprice differentials to the FOB Newcastle benchmark in US$/t for
-12.00
-10.00
-8.00
-6.00
-4.00
-2.00
-
2.00
Dec
-07
Mar
-08
Jun-
08
Aug
-08
Nov
-08
Feb-
09
Diff
eren
tial t
o FO
B N
ewca
stle
Fig. 3. Non-Japanese Asian-based trading company premium/discounts to FOB
Newcastle benchmark price 2007–2009.
R2 = 0.9889
-10.00
-8.00
-6.00
-4.00
-2.00
0.00
2.00
4.00
-3 -2 -1 0 1 2 3
J. West / Energy Policy 52 (2013) 513–521 519
non-Japanese Asian-based trading companies. A q–q plot is agraphical method for comparing the distribution of the differen-tials against a normal distribution. If the volume of coal ship-ments were unstable over the period then we would expect to seeprice differentials skew away from the mean value, as coalconsumers would look to buy coal at a much higher discountthan normal, providing significant outliers in the data. Stable coalvolumes should be associated with normally distributed differ-ences in the price differential with few outliers.
The coal prices of each series have again been adjusted forquality differences of ash, moisture and sulphur. The relativelylinear slope of the q–q plot along with the R2 value of0.9539 indicates that there is little dispersion of values fromnormally distributed prices and therefore relatively steady offtake volumes.
Fig. 6 illustrates the q–q plot for coal price differentials to theFOB Newcastle benchmark in US$/t for Japanese trading compa-nies. The relatively linear slope of the q–q plot along with the R2
value of 0.9889 indicates that there is even less dispersion ofvalues from normally distributed prices than non-Japanese Asian-
R2 = 0.9539
-12.00
-10.00
-8.00
-6.00
-4.00
-2.00
-
2.00
4.00
6.00
8.00
-3 -2 -1 0 1 2 3
Fig. 5. Normal q–q plot of non-Japanese Asian-based trading company premium/
discounts to FOB Newcastle benchmark price 2007–2009.
Fig. 6. Normal q–q plot of Japanese trading company premium/discounts to FOB
Newcastle benchmark price 2008–2009.
Table 7Descriptive statistics of premium/discounts to FOB Newcastle benchmark price for
non-Japanese Asian based trading company and Japanese trading companies
2007–2009.
Non-Japanese tradingcompanies
Japanese tradingcompanies
Mean �2.48 �1.53
Standard error 0.32 0.30
Median �1.65 0.00
Mode 0.00 0.00
Standard deviation 3.34 3.17
Sample variance 11.15 10.11
Kurtosis �0.13 2.21
Skewness 0.68 0.76
Range 14.51 16.56
Minimum �9.36 �8.50
Maximum 5.15 8.06
Confidence level (5%) 0.82 0.60
-10.00
-8.00
-6.00
-4.00
-2.00
-
2.00
4.00
6.00
8.00
10.00
Dec
-07
Mar
-08
Jun-
08
Aug
-08
Nov
-08
Feb-
09
May
-09
Aug
-09
Diff
eren
tial t
o FO
B N
ewca
stle
US$
/t
Fig. 4. Japan-based trading company premium/discounts to FOB Newcastle
benchmark price 2007–2009.
based trading companies and therefore very steady off takevolumes.
Table 7 provides the descriptive statistics for both distribu-tions. Both sets of data have similar distributional characteristicsand therefore there appears to be little information advantageobtained by Japanese trading companies in thermal coal pricenegotiation. We also conduct a two-sample heteroscedastic t-testto determine whether the two samples come from distributionswith equal population means but unequal variances. Thet-statistic of �1.494 is insignificant suggesting the two samplescome from the same population. We therefore reject the hypoth-esis that Japanese trading companies create volume instabilityusing informational advantages relative to non-Japanese tradingcompanies.
To further illustrate the inability to take advantage of coalproduction cost information in negotiating discounts to thebenchmark price, the actual contracted volumes and price differ-entials for both non-Japanese Asian based trading companies andJapanese trading companies were plotted. Fig. 7 shows thedifferentials and volumes of the non-Japanese traders who largelyaccept the market price of the day. There appears to be no
-10
-8
-6
-4
-2
0
2
4
6
8
10
0
10
20
30
40
50
60
70
80
90
100
Dec
-07
Jan-
08
Feb-
08
Mar
-08
Apr
-08
May
-08
Jun-
08
Jul-0
8
Aug
-08
Sep
-08
Oct
-08
Nov
-08
Dec
-08
US
$/t d
iffer
entia
l
Voilu
me
Com
tract
ed (t
hous
and
tonn
es) Differential to FOB Newc
Fig. 8. Volume and price differential plot of Japanese trading company premium/
discounts to FOB Newcastle benchmark price 2008.
Table 8Regression between price differential to FOB Newcastle benchmark price and
volume contracted by non-Japanese Asian based trading companies and Japanese
trading companies 2007–2009.
Constant Volume R2
Non-Japanese �2.51495n�0.00003 0.01
(�2.93530) �1.22379
Japanese �4.46555n 0.00004n 0.13
(�5.21325) (3.08006)
Table 9Comparison of ROA and ROE for major Australian-based foreign tradin
averaged over 2004–2011 (where data is unavailable for the full period
Austral
ROA (%)
Korea 12.11
China 12.01
India �1.00
USA 19.87
Brazil 13.34
Japan 9.69
S&P/ASX 200 Resources average (coal companies) 14.44
-10
-8
-6
-4
-2
0
2
4
6
0
10
20
30
40
50
60
70
80
90D
ec-0
7
Jan-
08
Feb-
08
Mar
-08
Apr
-08
May
-08
Jun-
08
Jul-0
8
Aug
-08
Sep
-08
Oct
-08
Nov
-08
Dec
-08
US
$/t d
iffer
entia
l
Voilu
me
Com
tract
ed (t
hous
and
tonn
es) Differential to FOB Newc
Fig. 7. Volume and price differential plot of non-Japanese Asian based trading
company premium/discounts to FOB Newcastle benchmark price 2008.
J. West / Energy Policy 52 (2013) 513–521520
relationship between contracted volume and price indicating thattheir ability to contract at a larger discount does not necessarilycorrelate with larger volumes.
Similarly Fig. 8 also shows that there may be some relationshipbetween contracted volume and price for Japanese trading com-panies. To test for their ability to contract at a larger discount asvolumes increase we conduct a regression of price differentialsagainst volumes for both non-Japanese and Japanese tradingcompanies. The results are at Table 8. Significance at the 1 percent level is indicated by n with t-statistics in parentheses.
While a positive relationship is detected between price differ-entials and volumes of observed contracts traded by Japanesetrading companies, it is only very slight and the R2 result suggestsit is of little explanatory value. This result highlights that evenif informational advantages exist for Japanese investors no per-sistent advantage is evident in coal price negotiation outcomesnor is there any advantage taken by the investors in the timing ofshipments themselves. We conclude that Japanese equity owner-ship in coal assets does not translate into persistently lower pricesor volume stability.
7. Non-Japanese foreign investors
It was claimed in Colley (1997) that the number of Japaneseinvestors in the Australian coal industry was not necessarily awave of substantial investment from a Japanese investors’ pointof view but rather a small strategic investment undertaken by anumber of Japanese companies as part of a consensus view onmanaging coal production and capacity. From the above analysisas commodity prices have increased over 1998–2011 theassumed aims of equity investment to achieve lower prices andvolume stability through information asymmetry have beenexchanged for above average capital returns for the parentcompany with no real loss of control.
It is prudent to examine the investment behaviour andmotives of other new entrants in the Australian coal market fromSouth Korea, China and India, as well as existing participants fromthe US and Brazil. In particular we wish to determine if theadvantages associated with equity ownership serve as the mainmotive behind their acquisition strategies, or if other reasons suchas the security of supply are driving foreign investment in coalmining. Table 9 provides a comparison of ROA and ROE for majorAustralian-based foreign trading companies by country with theirrespective parent companies averaged over 2004–2011. Wheredata is unavailable for the full period all available data is used.
Korean investment patterns in Australia have changed since2005. Korean investment in Australia has been based on astrategy of obtaining a stable supply of coal for the long term.Existing investments undertaken by Korean companies havetypically been small strategic equity positions (10–15 per cent
g companies by country with their respective parent companies
all available data is used).
ian subsidiary Parent
ROE (%) ROA (%) ROE (%)
24.91 10.15 15.29
22.81 6.40 13.37
�2.25 4.65 12.89
43.07 7.26 17.79
22.75 12.52 30.27
19.78 2.57 11.32
25.70 N/a N/a
J. West / Energy Policy 52 (2013) 513–521 521
at most) linked to marketing rights, but some recent investmentshave been larger with majority ownership positions includingoperating rights being established (e.g., Wallarah 2 Coal Project inNSW is 100 per cent owned by the government-backed KoreaResources Corporation (KORES)). Table 9 illustrates that averageinvestor returns of the Australian subsidiaries are likely to exceedthe average returns to the parent.
The new found confidence in investing in direct ownership andoperating rights is an approach emulated by recent Chineseinvestments, particularly the acquisition of Watermark, FelixResources and the Austar mine in the Hunter Valley by Chinesecompanies. Thus far the Chinese companies have funded pur-chases of Australian coal assets using cash reserves and the needfor concessional funding from government agencies in the futureis expected to remain low. It is expected that in contrast to Koreanand Indian investors, security of supply is not the primaryobjective behind Chinese investment in Australian coal assets asthese represent a very small portion of total Chinese coalconsumption (less than 0.04 per cent). Some Chinese coal com-panies also maintain majority ownership of Australian domesticelectricity assets which indicates a normal strategy aimedtowards achieving superior investment returns for Chinese inves-tors, rather than securing supply.
Indian investors first acquired sizeable interests in operatingassets in 2007 with more investments in reserves and infrastruc-ture opportunities observed in 2010. Historical investmentreturns in Australian assets have been comparatively low andthe trend towards investing in developing assets indicates thatsecurity of supply is the primary objective. For major Indianpower producers Government approvals for plant constructionare often conditional on securing a certain volume of importedbituminous coal so the motive behind the acquisition of coalassets in Australia is expected to be for the security of supplyrather than investment returns.
8. Concluding remarks
The financing and investment practices by the Japanesegovernment and coal consumers in the Australian coal industry,which were found to have stimulated supply to the point whereoversupply was significant enough to affect prices during the1990s, have clearly not continued over 1998–2011. While evi-dence to support investment practices influencing a persistentoversupply of coal is apparent in previous studies such asAnderson (1987) and Colley (1997), the capacity of Japanesetrading companies to influence prices through supply and useinformation advantages of marginal production costs in contractnegotiations has diminished. At the same time the Japanesetrading companies have experienced significant returns as mea-sured by the metrics of return on assets and return on equityobserved in a number of Australian subsidiaries. Moreover theestablished Japanese trading companies have clearly shifted from
minority to majority ownership and in some cases have soughtoutright ownership in both existing and newly developing coalmines in Australia. In 2009–2010 several Japanese companieshave obtained either limited or no recourse funding for a range ofcoal asset investments. Japanese companies have instead turnedto the local capital market for financing rather than rely onconcessional funding from JBIC.
This study has examined the major types of investmentpractices by Japanese trading companies in the form of conces-sional finance and debt guarantees supplied by JBIC for investingin the Australian coal industry. Where the profitability of Japa-nese investments in Australia was poor in the 1990s more recentexperience has shown that returns over 1998–2011 haveexceeded their parent company returns and are reasonable whenbenchmarked against their Australian peers. The experience oflow profitability driven through a strategy to simultaneouslysecure supply and reduce prices has not carried over during thecommodity price boom and the quasi-integration strategyemployed in the 1990s is unlikely to emerge as an investmentstrategy for some time. Japan continues to import significantvolumes of both coking and thermal coal and is now seeking longterm coal supply agreements in a more competitive global coalmarket. Japanese trading companies investing in Australian coalassets are also now competing with Chinese, Korean and Indiantrading companies looking to secure supply. Profitability is seenas a key element of the new investment paradigm and may spillover to other raw material acquisitions by the major Japanesetrading companies in the future.
References
Anderson, D., 1987. An Analysis of Japanese Coking Coal Procurement Policies: TheCanadian and Australian Experience. Centre for Resource Studies, QueensUniversity, Kingston, Ontario.
Australia–Japan Economic Institute, 1996. Economic Bulletin, vol. 4, no. 9.Australian Bureau of Statistics, 2010. 5625.0 Private New Capital Expenditure and
Expected Expenditure, Australia, September.Barnett, D.W., 1994. Australia’s coal industry: fuel for growth. Energy Policy 22 (4),
353–367.Colley, 1997. Investment practices in Australian coal: the practice and profit of
quasi-integration in the Australia–Japan coal trade. Energy Policy 25 (12),1013–1025.
D’Cruz, J.R., 1983. Quasi Integration Strategies in Markets for Industrial RawMaterials University of Toronto Working Paper, Canada.
Dowling, G., 1987. Buying is marketing too—Japan’s influence on the Australiancoal trade. Long Range Planning 20, 35–43.
International Energy Agency (IEA), 2011, World Energy Outlook. Paris.Koerner, R., 1993. The behaviour of Pacific metallurgical coal markets: the impact
of Japan’s acquisition strategy on market price. Resources Policy 19 (1), 66–79.Koerner, R., 1998. The influence of sogo shosha companies on contract bargaining
in the Pacific metallurgical coal trade. Resources Policy 24 (3), 167–177.Swan, A., Thorpe, S., Hogan, L., 1999. Australia–Japan coking coal trade: a hedonic
analysis under benchmark and fair treatment pricing. Resources Policy 25,15–25.
Yonezawa, Y., 1978. ‘Resource trade and economic security: the case of coal’. In:Drysdale, P., Kojima, K. (Eds.), Australia–Japan Economic Relations in the Interna-tional Context: Recent Experience and the Projects ahead. Australia–JapanEconomic Research Project, Canberra.
