THEMATIC

8 July 2010

Materials
Mineral Processing 101: Base And Precious Metals
Materials | Metals & Mining | Australia

Event:
In this note we outline the common processing methods relevant to the stocks in our coverage. This primarily includes gold, copper, zinc, nickel and platinum group elements (PGE).

Analyst: Ian Preston +61 3 9679 1453 ian.preston@gsjbw.com

Key take Outs:

In order to understand the economics of mineral processing, we have also included some estimates of capital and operating costs associated with these activities. This article is intended as a companion piece to our previous introduction to underground mining methods (Daily Cable, 8 June 2010) and as a guide to the cost implications of one method over another. Process costs are typically a smaller percentage of the total on-site costs than the mining cost. Within the metallurgical process, the largest cost item is generally crushing and grinding driven by the cost of power and the cost of grinding media. For those metals where a concentrate is produced for sale to a thirdparty smelter/refiner, the additional costs of smelting and refining (treatment and refining charges or TC/RC's) can typically add a further 10% to processing costs, although this may be offset (fully or) to some degree by revenue from by-products (e.g. gold in a copper concentrate, or cobalt in nickel concentrate). In the case of PGEs, the companies that we cover typically receive a percentage (typically ~80%) of the revenue stream, rather than incurring a cost for the processing of concentrate. Unit processing costs are mainly a function of the throughput rate; the higher the rate, the lower the unit cost. However, normally the lower processing rate can be a function of higher grades and thus the costs in terms of /lb or $/oz will be low.

Analyst: Paul Hissey +61 3 9679 1779 paul.hissey@gsjbw.com

Associate Analyst: Brett Cottrell +61 3 9679 1565 brett.cottrell@gsjbw.com

Investment View:
The ability to recognise value in a project ahead of the market presents a significant value opportunity for investors. A better understanding of the likely processing method and costs can offer the potential to form a measured opinion of a project while other participants are waiting for feasibility study results. We do not suggest that the data presented herein is definitive, however it may be approximated with a level of accuracy commensurate with a scoping study (+/- 30%). An understanding of the potential throughput rate and process required for extraction of the economic minerals at an early stage can allow early estimates of capital and operating costs. These assumptions can help assign some initial estimates to the likely value of a project (at least to orders of magnitude) and therefore its significance to the market. Different ore types will have different in-situ values and hence the economics of the scale of mining and processing is the deciding factor in assessing feasibility.

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised Important disclosures and the Reg AC Certification appear at the back of this report

8 July 2010

Materials

Mineral Processing 101

Introduction
A key component of any mining operation is the mineral processing which takes place to extract the metal-bearing mineral from the other deleterious materials surrounding it. In this article we provide a primer on some of the more common techniques used by the companies in our coverage universe. In order to understand the economics of mineral processing, we have also included some estimates of capital and operating costs associated with these activities. This article is intended as a companion piece to our previous introduction to underground mining methods (Daily Cable, 8 June 2010) and as a guide to the cost implications of one method over another. We have not included the downstream processes of smelting and refining, because most of our companies sell concentrate to third-party smelters for this stage of the metal recovery process. First, a few key terms. Comminution: the crushing and grinding of rock in order to liberate individual particles of ore minerals containing only the component minerals of interest. Gangue: waste rock which has no economic value. This typically includes quartz, clays, etc. We note that the valuable minerals - the ore minerals - are typically chemical compounds of a metal or metals in combination with sulphur or oxygen. Chalcopyrite, for example, is a copper-iron sulphide (CuFeS2). Only the copper is of ultimate economic value, but we do not consider the sulphur and iron as "gangue", because they cannot be separated from the copper at the mine site, but only at the smelting and refining stage. Hydrocyclone, or "cyclone": Item of plant designed to classify differently sized particles, generally from the slurry exiting ball or SAG mills. The slurry is introduced into a circulating fluid mass (conceptually a bit like a whirlpool), whereby larger (heavier) particles drop out the bottom (the "cyclone underflow") and smaller (lighter) ones overflow out of the top ("cyclone overflow"). Flotation: the process of separating ore minerals, from gangue minerals by utilising differences in chemical properties (such as attraction/repulsion from water molecules). P80: A P denoted with a subscript number usually precedes a measurement. This notation refers to the size of the openings ("mesh size" in a sieve or "screen"), which allows 80% of the feed material to pass through. For example: P80 = 100mm would refer to a crusher product of which 80% will pass through a screen with 100mm mesh. For milling, P95 is more commonly used (thats 95% passing) and the mesh size is commonly measured in microns.

Crushing and Grinding

A first step for almost all ore types (regardless of commodity, host rock or grade) is the reduction in size of the rock delivered to the plant from the mine. Crushing Primary crushing is usually undertaken by a gyratory or jaw crusher and reduces blasted rock down to a size which can be transported using conveyor belts - commonly <200mm, and fed into the grinding circuit. This stage of processing is essentially a material-handling necessity. Secondary and tertiary crushing can also be used where a much finer feed is required prior to milling. This is generally carried out using cone, small gyratory and roll crushers and can crush down to around 1cm. Screens are commonly used at the discharge point of secondary and tertiary crushers to separate material that is sufficiently small to proceed to the milling stage, from the oversize material which is rerouted back through the crushers. Crushers do not generally provide a bottleneck for processing and rates of 2,000-5,000t/hr are common amongst our companies. We also note, however, that given the violent and abrasive nature of primary crushing, there is generally an expectation of lower availability than other parts of the plant. For this reason, crusher capacity is typically over-engineered.

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised

8 July 2010

Materials

Grinding Crushed material is fed into one or more mills, which further reduce the size of the particles, in order to liberate individual particles or ore minerals. There are several types of mill, with the primary design being a rotating cylinder which reduces particle size through the impact of the rock against either other rocks, or against grinding media (steel balls or rods). The action of the rock being lifted up and then falling back (much like a cement mixer), imparts energy onto the rock which reduces its size through abrasion and impact. Usually a grinding medium is added, such as steel balls, which assist the process. Most mills would be used to reduce the size of particles to <<1mm. Mills are generally lined with steel or rubber liners, which periodically need replacing. For example, the SAG mill at Prominent Hill (OZL) requires a reline approximately quarterly, and this can take up to 144 hours (depending on size). In the same plant, the ball mill only requires a reline annually, and this can take up to 70 hours. These planned maintenance events will affect production and need to be allowed for. Often a regrind is introduced after milling those particles that have not been fully reduced in size to that required for post-processing. It is also important to note that a significant amount of energy is required for both crushing and grinding and, as a result, material should only be ground small enough to liberate the particles of interest. Over-grinding can result in "sliming" which adversely affects metal recovery. There is clearly a trade-off between finer grinding (which generally increases recovery) and increased energy costs. Most companies would undergo this optimisation process once plant commissioning is complete. Common mills amongst our companies include: SAG Mill: Semi-Autogenous Grinding (SAG) mills primarily rely on the rock itself as the primary agent of abrasion (but can have supplemental grinding media added for assistance). As a result, SAG mills are generally much larger in diameter (to allow a larger fall from the top of the rotation) and shorter in length than a ball mill. SAG mills can replace secondary and tertiary crushers, and can typically handle feed material up to 15cm in diameter. The addition of grinding media assists in increasing the throughput. Ball Mill: Generally cylindrical or conical in shape and charged with steel balls as the grinding media. As a general rule, the diameter and length of both ball and SAG mills control the throughput and the length influences the grind size. As a comparison, ball and SAG mills generally achieve the same outcome (although ball mills have the ability to grind finer), however SAG mills are generally less efficient, but require less ongoing maintenance, and consume less grinding media. As a reference point, PNA uses its ball mills to achieve a P80 of approximately 100-110m. The finer the material entering the mill, the smaller the diameter of the grinding media which is added. ISA Mill: These mills are of a similar shape to a ball mill, except they are approximately 1/2 the size and are used for ultra-fine grinding (down to ~10m). The Isa mills also use a grinding medium, which can be ceramic balls (similar to bean bag beans) and are assisted by a series of rotating discs within the mill. This technology was developed by Mount Isa Mines in the 1990s in order to treat the fine zinc ore from Macarthur River and is now used around the world. It is now commonly used to optimise the recovery of ores, (for example at Prominent Hill and Phu Kham), by grinding the material that has passed through the initial flotation stages prior to cleaning and scavenging (see below).

It is also important to note that crushing and grinding often represent the largest cost component of processing plants (particularly for base metals). This is primarily due to the consumables, maintenance (this is a high-wear activity) and power. To highlight this, we note the difference between the cost component of EQN and PNA, which have access to hydroelectric power, compared to AQP and PLA, which have coal-fired power.

Base Metals
Sulphide Ore With some exceptions, the majority of ore sourced for any particular project is primary - that is, the rock is fresh and as a result the common ore minerals are sulphides. Taking this into account, most projects will construct a plant that is designed to treat sulphide material. However, the initial stages of many mines (particularly open pits) will produce oxidised material, (usually termed weathered, or "oxide zone") grading to transitional ores, and then

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised

8 July 2010

Materials

primary (sulphide) ore at greater depth, and, as such, the plant must be able to treat this material also, albeit to less than optimal parameters. The processing fundamentals for most base metal sulphides are very similar. Once the ore is ground to a sufficient size to liberate the ore minerals (e.g. individual chalcopyrite or sphalerite grains) it is subject to flotation, thickened and filtered to produce a concentrate which is shipped to a smelter. For the purpose of this document we will discuss the production of a concentrate, given that this is the most common end-process for the companies in our coverage. A generic flow sheet for a flotation circuit is shown below:

Source: OZ Minerals

Flotation is the process whereby fine ore and water are agitated in large tanks, while air is blown through the agitators (from the bottom up) and reagents are added. The most important of these reagents is a "collector". One end of the collector molecule has an affinity for the ore mineral, and therefore sticks to it; the other end repels water, and therefore adheres to the air bubbles rising through the tank. These air bubbles form a froth at the top of the tank which overflows into a launder and which can be thickened and dewatered to form a concentrate. Most plants will contain a combination of rougher, cleaner and scavenger flotation cells, which operate using the same chemical processes. These different methods use different scale throughput and reagent consumption rates. One set of operating parameters will be insufficient to optimise recovery of ore, so using a combination of these three cells allows a first-pass to float the bulk of the concentrate (rougher), and then sequentially more selective circuits to extract the remainder, or clean the material claimed from the rougher. Each plant will have an optimum operating condition, which trades off throughput with recovery. There are generally relationships between the size of the particles in the slurry which is fed into the flotation cells and the speed at which the minerals will float and, therefore, the required residence time in the cell required to achieve separation. Some minerals will also float preferentially to others, which can complicate the circuit when the ore is polymetallic. In some copper process circuits, the gold is contained within the copper minerals, and rather than including an additional circuit to extract gold on-site, many companies sell a copper concentrate and are then paid a credit by the smelter/refiner for the gold contained. Once the slurry has been recovered from the flotation cells it is dewatered, usually through the use of a thickener tank, and then a filter press which, simplistically, removes the water from the fines to produce a concentrate. The moisture level of the concentrate is generally in the order of 8-12% and is optimised as transport to the smelters is charged by weight with the moisture content deducted from the total. We would also note that treatment and refining charges (TC/RCs) for those companies shipping their concentrate to a smelter would generally be of similar order to those costs incurred on-site.

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised

8 July 2010

Materials

A table outlining common base metal ore minerals, and their actual metal endowment, is included below:
Sulphide Mineral Native Copper Chalcopyrite Bornite Chalcocite Covellite Galena Sphalerite Pentlandite
Source: Webmineral.com

Metal Cu Cu Cu Cu Cu Pb Zn Ni

Metal Abundance in Mineral 100% 35% 63% 80% 66% 87% 43% 34%

Chemistry Cu CuFeS2 Cu5FeS4 Cu2S CuS PbS ZnFeS (Fe,Ni)9S8

Finally, a critical element of any processing is the concept of recovery. For not only sulphide flotation, but all processing methods, the recovery is a variable which has a significant impact on the economics of a project. There will be a component of any ore which cannot physically be recovered, and there will also be a further component which can be won, but will come at the expense of incremental increases in power consumption and reagent usage which render it uneconomic. As a result, all processing will be optimised to manage this trade-off between recovery and throughput. Lateritic Ore (Nickel) Lateritic ore is different to primary rock in that the minerals have been oxidised or partially oxidised as a result of their proximity to the surface and therefore atmospheric elements. A common feature of lateritic ores is that they generally contain cobalt as a by-product. The weathered nature of lateritic ore translates into shallow, sparse ore bodies that facilitate open pit mining (and, as a result, mining costs makes up a smaller portion of the total unit cost). The treatment of lateritic ore is not as straightforward as flotation and, as a result, there are varying examples of success. The lateritic nickel deposits in Western Australia (Murrin Murrin, Cawse and Bulong) were initially plagued by processing difficulties, however similar processes in Cuba (Moa Bay) and at Coral Bay in the Philippines have worked well. Unlike common sulphide flotation, there does not appear to be a standardised approach to the treatment of lateritic nickel ores, which adds some level of technical risk in the transition from bench/lab to full scale. High pressure acid leach (HPAL) involves acid leaching at high temperatures using sulphuric acid in a high temperature and pressure autoclave to dissolve the nickel and cobalt. The autoclaves use a mixture of ore slurry and sulphuric acid at 280C under ~5,400 kilopascals of pressure. Following this, for example, Murrin Murrin (MRE) precipitates the metals as sulphides, redissolves them with oxygen under pressure and finally separates with solvent extraction and hydrogen reduction. The final product is produced after an electro-winning process to give very high grade nickel briquettes. HPAL processing is employed for two types of nickel laterite ores. a. b. Ores with a limonitic character such as Goro, where nickel is bound in goethite and asbolite. Ores where nickel is bound within clay or secondary silicate substrates in the ores (such as Murrin Murrin).

The advantages of HPAL plants are that they are not as selective toward the type of ore minerals, grades and nature of mineralisation. The disadvantage is the energy required to heat the ore material and acid, and the wear and tear hot acid causes upon plant and equipment. Higher energy costs demand higher ore grades. The process, by nature, leads to higher capital costs (due to plant complexity and the requirement for bespoke plant designs) and higher operating costs due to the high pressures and temperatures leading to greater wear rates on components. An additional processing method which deserves mention is on-site smelting. Some nickel laterites are amenable to on-site smelting, as an alternative to HPAL. In these instances, ore is fed directly to a smelter at the mine, along with a reductant and flux, in order to produce a ferro-nickel alloy. This material is generally shipped directly to third-parties for use in the production of stainless steel. An example of this processing stream is Cerro Matoso (BHPBilliton).

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised

8 July 2010

Materials

Source:Fluor: Construction of Murrin Murrin plant

Precious Metals - Gold

There are several processes which can be used to win gold from its host constituents. The most appropriate method is generally determined by the nature of the gold mineralisation within the gangue minerals. We would also like to note that it is unusual for gold ore bodies to be solely composed of one of the following ore types. There are often combinations and, as a result, processing plants are generally required to accommodate various feed types. Therefore, an understanding of the relative percentages of different feed material and their likely timing through the plant is essential at the design stage. A common gold plant would include a crushing/grinding circuit, followed by a gravity circuit which separates the coarse gold as individual particles, followed by a Carbon in Leach/ Carbon in Pulp (CIL/CIP) plant to recover the finer component. It is common for approximately 5-10% of the gold put into the front of the plant to be lost to the tailings (i.e. it is not recovered). The three main components of a circuit used to produce gold dore from gold-rich ore are; gravity, CIL/CIP, roasting and high-pressure leaching. It is also important to note that gold circuits commonly contain a combination of these three methods to ensure optimal recovery. Gravity Gravity circuits generally require that the ore is crushed and ground small enough so that individual gold grains are separated from their host mineral. Following this, the heavy material is separated using cyclone and/or knelson/falcon concentrators to recover the gold. This method is made possible by the density contrast of the different minerals. A key criterion for this method is the coarseness of the individual gold particles and the overall size distribution, which determines the optimum grind size (given that there is a trade-off between recovery and cost - finer grinding to liberate the particles / power required for additional grinding). Bendigo is a good example of the large percentage of the total gold that can be recovered from gravity. Cyanidation The leaching process is supported by two facts: (i) gold is soluble in dilute solutions of cyanide, and (ii) the pregnant solution can be processed successfully by using zinc powder or activated carbon and electrowinning to remove the gold from the liquid. There are numerous variations and combinations of leaching processes for gold (CIL, CIP, Acacia Reactor), however these are beyond the scope of this introductory article. Cyanidation is very common in Australia and, despite the connotations, the use of cyanide is generally regarded as safe. A typical gold plant would utilise a gravity circuit to recovery the coarsest gold, followed by some form of cyanidation to recover the finer fraction. In all gold process plants the residual cyanide in the liquor after gold extraction is destroyed in a separate process before being discharged onto the tailings storage facility (TSF) such
Goldman Sachs JBWere Investment Research All figures in A$ unless otherwise advised 6

8 July 2010

Materials

that typically the water that is recovered from tailings dams for re-use is of drinking water quality. Refractory Gold Some gold can be contained within the chemical structure of other minerals (typically arsenopyrite or pyrite), and this material is said to be refractory and is not suitable for recovery using simple cyanide-leaching processes, unless very low levels of recovery are acceptable, typically ~50%. For these ore types, additional processes must be undertaken prior to recovery of the gold using the methods mentioned above. These methods include: Roasting is an older technology in which the sulphide in the host mineral is physically burnt off using the equivalent of a sinter plant. The arsenic and iron sulphides will be decomposed at high temperatures to produce elemental arsenic and sulphur vapour that react with the oxygen present in air and are released as gas. Bio-oxidation relies upon bacteria to break down the sulphide mineral matrix in the ore being treated. The bacteria attach themselves to the metal sulphide surfaces in the ore, resulting in the accelerated oxidation of the sulphides. Goldfields has patented a particular method of bacterial leaching called BIOX. Sino Gold used this process at its operation, Jinfeng, in China. Pressure Leach (such as Lihir Island), is a process that removes sulphur in an autoclave at high pressures and temperatures. Ultrafine grinding - such as the ISA mills mentioned above. This method is not always appropriate as the gold may be too fine to grind sufficiently small to liberate the individual particles.

Precious Metals - PGEs

We include here the process used in separating PGEs. Although it utilises some of the processes mentioned above, there are some differences. Firstly, PGE (platinum group elements) is a term referring to the group of metals comprising, in order of importance, platinum, palladium, rhodium, gold (often referred to as 3Es + Au or 4Es, and measured in basket oz), iridium, osmium and ruthenium. Given the majority of PGE mining occurs in the Bushveld Igneous complex in South Africa, the mineralogy is generally similar, with mining activities producing an ore which is associated with the PGE-bearing sulphides. In the case of the UG2, this ore is chromite-rich as opposed to the ore derived from the Merensky Reef which has no/little chromite. A table outlining the common PGE ore minerals is included below.
Ore Mineral Sperrylite Braggite Stibiopalladinite Laurite
Source: Webmineral.com

Metal Pt Pt, Pd, Ni Pd Ru

Chemistry PtAs2 (Pt, Pd, Ni)S Pd3Sb RuS2

The ore is generally processed using the sulphide flotation method outlined above to derive a concentrate. For the companies under our coverage (AQP and PLA) this concentrate is sold to third-parties who smelt and refine the concentrate into its constituent parts for sale. It is also worth pointing out that the concentrate produced on site by both of our PGE stocks (AQP and PLA) contains economic quantities of both copper and nickel, which are generally removed at the smelting stage, with the companies receiving a credit for these metals. In the case of AQP, given the particular nature of the ore it is mining (primarily sourced from the UG2 reef in the Bushveld), the chromite levels are sufficient to warrant pre-treatment (before sale) in order to ensure the chromite content of the product is sufficiently below certain values to avoid any penalty rates. This is due to differing metallurgical properties of chromite, which can otherwise hinder the smelting process. As an example, we have included a table below which highlights some average feed grades of 4E ore produced by both AQP and PLA (our PGE stocks) and their individual components. From a cost perspective, with the exception of the chromite pre-treatment, treatment costs

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised

8 July 2010

Materials

would be comparable to those outlined above for similar plant types (eg crush, grind, flotation and dewatering).
Company AQP PLA
1 2

Plant Kroondal1 Smokey Hills2

Feed Grade g/t 2.56 3.13

Last Q Throughput kt 1580 101

Avg Recovery 79% 81%

4E Metal
koz

Pt 59% 44%

4E Breakdown Pd Rh 30% 45% 11% 9%

Au 0.5% 1%

103 8.3

Source: Company data; GSJBW research estimates Mar Q 2010, 100% basis Mar Q 2010

Dollars and Cents - What Does this Mean?

The ability to recognise value in a project ahead of the market presents a significant value opportunity for investors. A better understanding of the likely processing method and costs can offer the potential to form a measured opinion of a project while other participants are waiting for feasibility study results. We do not suggest that the data presented herein is definitive, however it may be approximated with a level of accuracy commensurate with a scoping study (+/- 30%). Otherwise, the data is taken directly from company reports. Obviously there are significant differences in the processing methods for the various companies in our coverage. The key with all of them, however, is that processing costs will vary according to the complexity of the material being fed to the plant. The higher the intrinsic value of each tonne of ore, the greater the capacity for extraction while retaining sufficient margin for project approval. This relationship can be driven by either cost or value, with higher-value ores having a greater margin to facilitate more expensive treatment. In some instances, the high intrinsic value of the ore may correspond to a low treatment cost, (eg, SFR's potential direct shipping ore at Doolgunna at ~20% Cu), in which case the economics are unlikely to be an impediment to extraction. We have conducted some research amongst the companies in our coverage universe in order to demonstrate some approximate treatment considerations amongst these stocks. Please note that these tables (unless otherwise indicated) represent the costing associated with processing (and processing facilities) only. We note that it is difficult to directly compare capital costs due to the inclusion of various additional infrastructure.

Gold
Company PNA MDL MDL
1 2 3

Mine Ban Houayxai1 Sabodola2 Sabodola3 CIL CIL

Type

Year 2009 2008 2010*

Throughput tpa 4 2 3.5

Capital Cost US$m 75-80 300 40-45

Operating Cost US$/t 7.0-7.50 13-15 12-13

Expansion

Source: GSJBW Research estimates plant only whole project - includes power station and pre-strip - assume US$100-150 for plant excludes gyratory crusher

* currently only proposed

Base Metals
Copper
Company OZL PNA EQN CGG
` nameplate ^ approximate - may include additional infrastructure such as roads, workshops, TSF etc

Mine Prominent Hill Phu Kham Lumwana Jabal sayid

Type Flotation Flotation Flotation Flotation

Year 2007 2008 2005 2010

Throughput`
mtpa

Capital Cost^
US$m

Operating Cost^
US$/t

8 12 20 3

296 300 350 60

8.1 5.7 4.0 10.6

Source: Company data; GSJBW research estimates

Nickel
Company WSA IGO MRE
` nameplate ^ approximate

Plant Cosmic Boy BHP Nickel West Murrin Murrin

Type Flotation Flotation HPAL

Year Built 2007 1998

Current Throughput^
ktpa

Capital Cost
US$m

Operating Cost^
A$/t

250
(toll treatment)

~2900

32 >1000^

47-53 33-37 140-160

Source: Company data; GSJBW research estimates

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised

8 July 2010

Materials

Each of the analysts named below hereby certifies that, with respect to each subject company and its securities for which the analyst is responsible in this report, (1) all of the views expressed in this report accurately reflect his or her personal views about the subject companies and securities, and (2) no part of his or her compensation was, is, or will be, directly or indirectly, related to specific recommendations or views expressed in this report: Ian Preston, Paul Hissey.

Goldman Sachs JBWere Investment Research

All figures in A$ unless otherwise advised

Disclosures of Interest
Important Notice - Aus: Australia - Research Important Notice Company Specific Regulatory Disclosures See company-specific regulatory disclosures for any of the following disclosures required as to companies referred to in the report: manager or co manager in a pending transaction; financial advisor in a strategic corporate transaction; 1% or other ownership; compensation for certain services; types of client relationships; managed/co-managed public offerings in prior periods; directorships; market making and/or specialist role. The following are additional required disclosures: Ownership and Material Conflicts of Interest: Goldman Sachs JBWere policy prohibits its analysts, assistant analysts and their respective associates owning securities of any company in the analyst's area of coverage. Analyst compensation: Analysts are paid in part based on the profitability of Goldman Sachs JBWere, which includes investment banking revenues. Distribution of recommendations: See the distribution of recommendations disclosure on the following page. Compendium Report Please see disclosures at http://www.gsjbw.com/Disclosures. Disclosures applicable to companies included in this compendium report can be found in the latest relevant published research. Global Product; Distributing Entities This report has been prepared by the Goldman Sachs JBWere Investment Research Division for distribution to clients of affiliates of Goldman Sachs JBWere and pursuant to certain contractual arrangements to clients of affiliates of The Goldman Sachs Group, Inc. (Group) (Collectively, Group and its affiliates, ?GS?). Group owns indirectly 45% of the ordinary shares of Goldman Sachs JBWere Group Holdings Pty Ltd, the ultimate holding company of Goldman Sachs JBWere Pty Ltd. Research views, investment opinions and recommendations published by Goldman Sachs JBWere Pty Ltd are developed independently from those published by the Goldman Sachs Global Investment Research Division. This research is disseminated in Australia by Goldman Sachs JBWere Pty Ltd (ABN 21 006 797 897); in Canada by Goldman Sachs Canada Inc. regarding Canadian equities and by Goldman Sachs & Co. (all other research); in Germany by Goldman Sachs & Co. oHG; in Hong Kong by Goldman Sachs (Asia) L.L.C.; in India by Goldman Sachs (India) Securities Private Ltd.; in Japan by Goldman Sachs Japan Co., Ltd.; in the Republic of Korea by Goldman Sachs (Asia) L.L.C., Seoul Branch; in New Zealand by Goldman Sachs JBWere (NZ) Limited; in Singapore by Goldman Sachs (Singapore) Pte. (Company Number: 198602165W); and in the United States of America by Goldman, Sachs & Co. Goldman Sachs International has approved this research in connection with its distribution in the United Kingdom and European Union. Persons who would be categorized as private customers in the United Kingdom, as such term is defined in the rules of the Financial Services Authority, should read this material in conjunction with the last published reports on the companies mentioned herein and should refer to the risk warnings that have been sent to them by Goldman Sachs International. A copy of these risk warnings is available from the offices of Goldman Sachs International on request. Unless governing law permits otherwise, you must contact a Goldman Sachs entity in your home jurisdiction if you want to use Goldman Sachs JBWere's or GS's services in effecting a transaction in the securities mentioned in this material. European Union: Goldman Sachs International, authorised and regulated by the Financial Services Authority, has approved this research in connection with its distribution in the European Union and United Kingdom; Goldman, Sachs & Co. oHG, regulated by the Bundesanstalt f?r Finanzdienstleistungsaufsicht, may also be distributing research in Germany. General Disclosures This research is for clients only, as stated above. Other than disclosures relating to Goldman Sachs JBWere, this research is based on current public information that we consider reliable, but we do not represent it is accurate or complete, and it should not be relied on as such. We seek to update our research as appropriate, but various regulations may prevent us from doing so. Other than some industry reports published on a periodic basis, the large majority of reports are published at irregular intervals as appropriate in the analyst's judgment. Goldman Sachs JBWere and/or its affiliates conduct a global full-service, integrated investment banking, investment management, and brokerage business. We have corporate advisory, corporate finance and other business relationships with a substantial percentage of the companies covered by our Investment Research Division. Our salespeople, traders, and other professionals may provide oral or written market commentary or trading strategies to our clients and our proprietary trading desks that reflect opinions that are contrary to the opinions expressed in this research. Our asset management area, our proprietary trading desks and investing businesses may make investment decisions that are inconsistent with the recommendations or views expressed in this research. We and our affiliates, officers, directors, and employees, excluding equity analysts, will from time to time have long or short positions in, act as principal in, and buy or sell, the securities or derivatives (including options and warrants) thereof of covered companies referred to in this research. This research is not an offer to sell or the solicitation of an offer to buy any security in any jurisdiction where such an offer or solicitation would be illegal. It does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of individual clients. Clients should consider whether any advice or recommendation in this research is suitable for their particular circumstances and, if appropriate, seek professional advice, including tax advice. The price and value of the investments referred to in this research and the income from them may fluctuate. Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur. Certain transactions, including those involving futures, options, and other derivatives, give rise to substantial risk and are not suitable for all investors. In producing research reports, members of Goldman Sachs JBWere Investment Research may attend site visits and other meetings hosted by the issuers the subject of its research reports. In some instances the costs of such site visits or meetings may be met in part or in whole by the issuers concerned if Goldman Sachs JBWere considers it is appropriate and reasonable in the specific circumstances relating to the site visit or meeting. Our research is disseminated primarily electronically, and, in some cases, in printed form. Electronic research is simultaneously available to all clients. Disclosure information is also available at http://www.gsjbw.com/Disclosures or from Research Compliance, Level 42, 1 Farrer Place Sydney NSW 2000. Investment Adviser & Broker Disclosure Statement The disclosure statement for Goldman Sachs JBWere (NZ) Limited and your adviser is available on request and free of charge. Please contact your adviser to obtain a copy. Research Analyst Certification Each equity and strategy research report excerpted herein was certified under Reg AC by the analyst primarily responsible for such report as follows: I, <Name of Analyst>, hereby certify that all of the views expressed in this report accurately reflect my personal views about the subject company or

companies and its or their securities. I also certify that no part of my compensation was, is, or will be, directly or indirectly, related to the specific recommendations or views expressed in this report. Copyright 2010 Goldman Sachs JBWere Pty Ltd ABN 21 006 797 897 AFSL 243346 No part of this material may be (i) copied, photocopied or duplicated in any form by any means or (ii) redistributed without the prior written consent of Goldman Sachs JBWere.

Australia - Research Recommendation Definitions

Sell (S) Hold (H) Buy (B) Other Definitions NR Stock is expected to underperform the S&P/ASX 200 for 12 months Stock is expected to perform in line with the S&P/ASX 200 for 12 months Stock is expected to outperform the S&P/ASX 200 for 12 months

Not Rated. The investment rating has been suspended temporarily. Such suspension is in compliance with applicable regulations and/or Goldman Sachs JBWere policies in circumstances when Goldman Sachs JBWere is acting in an advisory capacity in a merger or strategic transaction involving the company and in certain other situations Coverage Suspended. GSJBW has suspended coverage of this company. Not Covered. GSJBW does not cover this company.

CS NC

Target Price Analysts set share price targets for individual companies based on a 12 month horizon. These share price targets are subject to a range of company specific and market risks. Target prices are based on a methodology chosen by the analyst as the best predictor of the share price over the 12 month horizon. Research Criteria Definitions The above recommendations are primarily determined with reference to the recommendation criteria outlined below. Analysts can introduce other factors when determining their recommendation, with any material factors stated in the written research where appropriate. Each criterion is clearly defined for the research team to ensure consistent consideration of the relevant criteria in an appropriate manner. Prior to 20 July 2007, GSJBW had a dual-horizon recommendation system: Short Term & Long Term. The Short Term factors were Relative Earnings Outlook, Earnings Momentum, News Flow, Relative Performance, and Valuation Support. The Long Term factors were Industry Structure, EVA Trend, Growth Option and Price/DCF. Industry Structure: Based on Goldman Sachs JBWere industry structure ranking. All industries relevant to the Australian equity market are ranked, based on a combination of Porters Five Forces of industry structure as well as an industry's growth potential, relevant regulatory risk and probable technological risk. A companys specific ranking is based on the proportion of funds employed in particular industry segments, aggregated to determine an overall company rating, adjusted to reflect a view of the quality of a companys management team. EVA trend forecast for coming 2 years. Designed to reflect turnaround stories or to highlight companies Goldman Sachs JBWere analysts believe will allocate capital poorly in the estimated timeframe. The percentage change in the current consensus EPS estimate for the stock (year 1) over the consensus EPS estimate for the stock 3 months ago. Stocks are rated according to their relative rank, effectively making it a market relative measure A qualitative and quantitative assessment of a company's long term catalysts that the analyst believes should be considered and possibly recognised by the market. The premium or discount to base case DCF valuation at which the stock is trading relative to the average premium or discount across the market.

EVA Trend:

Earnings Momentum:

Catalysts: Price:Base Case DCF: For Insurers EVA Trend: 1 Balance Sheet:

ROE is used as a proxy for EVA. Rating takes into account the expected level and trend of ROE over the next 23 years. Analysts assessment of the quality and strength of the insurers balance sheet, including conservatism of provisioning, sufficiency of capital, and quality of capital.

For REITs Strategy: EPU Growth: Yield: For NZ Companies Relevant Index:

Used instead of industry structure as many REIT investors are intra rather than inter sector focussed. Ranking of Earnings Per Unit growth relative to other listed Real Estate Investment Trusts. Used instead of EVA Trend. Yield relative to the REIT sector average. Used instead of Earnings revision.

If a research report is published by the New Zealand affiliate of Goldman Sachs JBWere, the recommendation of a company or trust is based on their performance relative to the NZSX 50 Index (Gross) and not the S&P/ASX 200 index. EVA is a registered trademark of the U.S. consultancy firm Stern Stewart Corporate relationship* in last 12 months 5% 42% 53%

Distribution of Recommendations as at 31 March 2010 Recommendation Overall Sell 7% Hold Buy * 55% 38%

No direct linkage with overall distribution as the latter relates to the full Goldman Sachs JBWere stock coverage (>250 companies). The above table combines the corporate relationships and recommendations of both Goldman Sachs JBWere Pty Ltd and its affiliate in New Zealand, Goldman Sachs JBWere (NZ) Limited.

2010 Goldman Sachs JBWere Pty Ltd ABN 21 006 797 897