Sibanye Gold - BNP Cadiz Initiating Report

INITIATION
COMPANY REPORT
EQUITIES RESEARCH SGL SJ
HOW WE DIFFER FROM THE STREET

BNPP Target Price* (ZAR) EPS 2013 (ZAR) TARGET* CLOSE* UP/DOWNSIDE ZAR700.00 ZAR745.00 -6.0% Market Recs * in cents EPS 2014 (ZAR) 700.00 1.15 (0.44) Positive 2 Consensus 1072.75 2.14 2.57 Neutral 3 % Diff (37.7) (46.3) N/A Negative 3
SIBANYE GOLD LTD

SOUTH AFRICA / METALS & MINING
HOLD
INDUSTRY OUTLOOK
KEY STOCK DATA

YE Dec (ZAR m) Revenue Rec. net profit Recurring EPS (ZAR) EPS growth (%) Recurring P/E (x) Dividend yield (%) EV/EBITDA (x) Price/book (x) Net debt/Equity (%) ROE (%)
Feb-13 1,693 1,493 1,293 1,093 893 693
Sunset on Sibanye
INITIATION
A new South African gold mining company with old assets Sibanye was unbundled from Gold Fields to become the tenth largest gold producer in the world. We expect it to produce 1.2moz this year from marginal deep underground mines and surface rock dumps in South Africa. Production has been falling sharply and the company is the only one amongst its peer group with a declining production outlook.
2013E 16,623 842 1.15 (75.0) 6.5 3.9 2.4 0.4 22.6 6.3
2014E 14,813 (320) (0.44) (138.0) neg 0.0 4.2 0.3 35.3 (2.0)
May-13
2015E 15,074 (795) (1.09) 148.5 neg 0.0 5.8 0.3 39.4 (4.6)
PLANS SIBANYE CANT AFFORD

Sibanyes new plans a hard buy Management intends to hold production steady but we expect it to fall further before stabilising. New projects could sustain production in the longer term. However, plans to mine pillars in the short term are unlikely to add value, while increasing safety risk. These and other new plans will require capital that the company cannot afford at current prices.
4 (6) (16) (26) (36) (46) (56) (66) (76) Sibanye Gold Ltd Rel to FTSE JSE All Share (%)
VALUATION
Initiating coverage with a HOLD and TP of ZAR7/share at spot Our DCF-based target price of ZAR7/share (based on the current spot gold price) factors in lower-than-guided production but includes upside from potential reserves. The company is highly geared and earnings are hedged by the USD/ZAR exchange rate. Buyers of this stock should expect the gold price to rise or the rand to continue to weaken.
493 (ZAr)
Share price performance Absolute (%) Relative to country (%) Next results Mkt cap (USD m)
1 Month 3 Month 12 Month (21.3) (24.2) (42.7) (46.8) -
August 2013 568 7.4 89 Investec (15%) 1630.00/673.00 62.9 SBGL US 3.01 733
3m avg daily turnover (USD m)
RUNNING OUT OF OPTIONS

Long-term fundamentals look weak Although short-term metrics look attractive (forward PE of 6x and FY13E dividend yield of at least 4%), the long-term fundamentals are likely to get worse. Surface rock dumps will contribute about 23% to free cash flow this year but their declining reserves and falling grades means Sibanye expects this number to fall to 7% real in two years time. This will make funding for new projects even more difficult. In addition, declining production and fixed costs of 80% will likely move Sibanye even more towards the top of the global cash cost curve.
Free float (%) Major shareholder 12m high/low (ZAR) 3m historic vol. (%) ADR ticker ADR closing price (USD; 24 May 13) Issued shares (m)
Sources: Bloomberg consensus; BNP Paribas Cadiz Securities estimates
Adrian Hammond
adrian.hammond@bnpparibascadiz.com +27 11 088 2181
BNP Paribas Securities (Asia) Ltd. research is available on Thomson One, Bloomberg, TheMarkets.com, Factset and on http://eqresearch.bnpparibas.com/index. Please contact your salesperson for authorisation. Please see the important notice on the back page.
PREPARED BY BNP PARIBAS SECURITIES ASIA THIS MATERIAL HAS BEEN APPROVED FOR U.S DISTRIBUTION. IMPORTANT DISCLOSURES CAN BE FOUND IN THE DISCLOSURES APPENDIX
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
CONTENTS
Executive summary __________________________________________________________________________ 4 Introduction _________________________________________________________________________________ 5 Sibanyes plan _______________________________________________________________________________ 5 Potential reserves ____________________________________________________________________________ 5 Is pillar mining a game changer?_______________________________________________________________ 6 Can declining production be stopped? __________________________________________________________ 6 Surface rock dumps (SRD)_____________________________________________________________________ 7 Costs aligned with industry ___________________________________________________________________ 7 Risks _______________________________________________________________________________________ 8
Cash costs _____________________________________________________________________________________________________ 8 Once empowered, always empowered? ______________________________________________________________________________ 8 Loss of skills ___________________________________________________________________________________________________ 8 More safety-stoppages ___________________________________________________________________________________________ 8
Earnings ____________________________________________________________________________________ 8 Valuation ___________________________________________________________________________________ 9 Sensitivity __________________________________________________________________________________ 9 Investment recommendation _________________________________________________________________ 10 Appendix 1: Company background _____________________________________________________________ 11 Appendix 2: Geology and mining operations ____________________________________________________ 12
KDC _________________________________________________________________________________________________________ 12 Beatrix _______________________________________________________________________________________________________ 12 Pay-limits and rationalising the operations _________________________________________________________________________ 13
Appendix 3: Reserves ________________________________________________________________________ 14

Underground reserves __________________________________________________________________________________________ 14 Where did all the reserves go? ____________________________________________________________________________________ 14
Appendix 4: Life-of-mine and amortisation _____________________________________________________ 16

Historical life-of-mine (LoM) _____________________________________________________________________________________ 16 Impact on amortisation _________________________________________________________________________________________ 17
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 5: What is left in the lease? __________________________________________________________ 18

Managements strategic plans for KDC _____________________________________________________________________________ 18 Kloof ________________________________________________________________________________________________________ 20 Driefontein ___________________________________________________________________________________________________ 21 Beatrix _______________________________________________________________________________________________________ 22 Opportunities beyond the lease area _______________________________________________________________________________ 23
Appendix 6: Potential upside _________________________________________________________________ 24 Appendix 7: Surface rock dumps ______________________________________________________________ 25 Appendix 8: Tailings project __________________________________________________________________ 26
Assumptions and results ________________________________________________________________________________________ 26 Environmental challenges _______________________________________________________________________________________ 27
Appendix 9: Current life-of-mine plans _________________________________________________________ 27 Appendix 10: Historical production ____________________________________________________________ 29 Appendix 11: Production outlook ______________________________________________________________ 29 Appendix 12: Costs and capex ________________________________________________________________ 30
Cost inflation __________________________________________________________________________________________________ 31 Cost savings ___________________________________________________________________________________________________ 31 Global cash costs ______________________________________________________________________________________________ 32 Capex ________________________________________________________________________________________________________ 33
Appendix 13: Valuation ______________________________________________________________________ 34

Model assumptions and inputs ___________________________________________________________________________________ 34 Other cost assumptions: _________________________________________________________________________________________ 34 Results _______________________________________________________________________________________________________ 34
P&L, Balance Sheet and Cash Flow ____________________________________________________________ 36
To find out more about BNP Paribas Equities Research:

Visit : http://eqresearch.bnpparibas.com/ For ipad users : http://appstore.apple.com/BNPP-equities/
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Executive summary
Management intends to hold production steady at between 1.2moz to 1.3moz over the next three years. However, production has been falling sharply over the last decade as a result of declining grades, volumes, development metres and productivity. Sibanyes new plans will attempt to arrest this trend by maintaining development rates and mining pillars and secondary reefs. Details of their new plans will be revealed later this year. After taking a comprehensive look at their assets, we estimate there are no less than 3.7m oz of potential reserves from pillars and projects below infrastructure. Plans to mine pillars could take two years to implement fully and new projects below infrastructure could start in 2017. Secondary reefs are unlikely to add any value, in our view. There are no other attractive growth options or opportunities with neighbouring mines, in our view. Pillar mining is a low volume and slow process and will likely increase safety risks. To realise any benefit, managements first step will be to arrest the current decline in production first in order to keep unit costs from escalating. Additional capital will be required for development but, in our view, the company cannot afford any new projects at current prices. We expect production is unlikely to recover from FY12 levels and management could take about two years to stabilise production. Q1 FY13 development rates indicate a good recovery from labour unrest last year, but further improvement in these rates is necessary. Surface rock dumps (SRD) are an important source of low-cost ounces for the business, contributing as much as 23% to group free cash this year. However reserves will soon be depleted and will make the challenge for management even harder. We estimate that plans to replace these cash flows from a tailings project are unlikely to add value. Although Sibanyes C1 cash costs are one of the highest in the industry, C3 costs are in line with peers. We estimate that real C3 cash costs could be maintained at about ZAR380,000/kg for three years if production is sustained above 1.15moz and cash savings of at least ZAR1b (10%) are achieved, as expected. Beyond this period, real costs are likely to rise as production falls. Sibanye is currently cash generative and the short-term metrics look attractive. The share is highly geared to the gold price and the USDZAR exchange rate has provided a good hedge to earnings from falling gold prices. The share is currently trading at a forward PE of 6x and we expect a dividend yield of at least 4% this year. However the long-term fundamentals will get worse. Buyers of this stock need to hold the view that the gold price will rise or the Rand will continue to weaken. The company cannot generate any earnings or free cash in the next year at current prices . Based on the current life-of-mine plan, we value Sibanye at ZAR11/share. This is a result of our more bearish outlook on production in the early years as well as more capital to develop new projects. Based on our realistic view of production as well as potential reserves, we value Sibanye at ZAR7/share, at current spot prices. We initiate coverage with a HOLD recommendation.
3 1 2
1 2
C1 = mining costs C3 costs = mining costs + corporate overheads + ongoing capital 3 Current prices taken at USD1,380/oz gold and ZAR9.4/USD
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Introduction
Gold Fields [GFI SJ, Not Rated] unbundled its mature South African assets into a separate vehicle, Sibanye Gold, which was listed on 11 February 2013. Sibanye holds the Kloof, Driefontein and Beatrix mines located in South Africas Witwatersrand Basin. It is the tenth largest gold producer in the world. About 92% of production comes from underground and 8% from surface rocks dumps. Management is reviewing two options for a new tailings project. These mines are mature, deep-level and marginal. Costs are as much as 80% fixed and are highly geared to the ZAR gold price. Kloof and Driefontein mines are currently cash generative while Beatrix isnt. Reserves have declined over 70% in the last five years as a result of higher pay-limits, lower mine call factors and a more strict SAMREC code. The average life-of-mine has fallen from 20 years to 14 years over this period. This has caused amortisation charges to increase almost 70% which we expect will continue to 4 rise. We have low confidence in the current life-of-mine plans . Beatrix West section is expected to close this year and further shaft closures remain an option for management. KDC and Beatrix together produced over 3moz per annum 10 years ago. Production has steadily fallen to around 1.3moz. Whilst under the Gold Fields umbrella, capital programmes at KDC and Beatrix had been held back. The new management team plans to re-vitalise the business by reviewing previously withheld projects. Sibanyes new plan is set to be released later this year. The company is being led by Neal Froneman who previously managed Gold One and Uranium One.
Sibanyes plan
The company is reviewing current LoM plans and will reveal its new strategy by the end of the year. Some of its new plans have already begun to be implemented and include the following: 1 Hold production steady at between 1.2moz to 1.3moz over the next three years by doing the following: a b 2 3 4 Maintain development rates and improving the blast frequency Assess potential reserves from pillars, secondary reefs and new projects below infrastructure
Reduce the workforce by 4,000 to 5,000 people, of which 3,000 have already been retrenched Reduce costs by 20% within 18 months, with about 5% achieved so far Develop a new tailings project to address the decline in SRDs
Potential reserves
Based on our assessment of opportunities within and beyond the lease for KDC and Beatrix, we estimate 5 there is no less than 3.7m oz of potential reserves (EXHIBIT 35) . Of this about 1moz lies below infrastructure and the balance comes from safety-related and shaft pillars at KDC. Mining has progressed towards the lease boundaries. There are no other attractive growth options, in our view. EXHIBIT 1: Potential reserves
Source Potential reserve (koz) Pillars Shaft pillar Projects below infrastructure Sources: Sibanye Gold; BNP Paribas Cadiz estimates 2,1006 560 1,030 11.0 4.3 3-6 Life (years) Annual production (koz) 190 130 220 2015 2017 2017 Start year
There are also potential reserves from secondary reefs, although we do not expect mining them will be 7 value accretive to the business, but only necessary to de-stress the mining of pillars below.
4 5 6
See Appendix 9: Current life-of-mine plans for details See Appendix 4 and 5 for our detailed assessment Assumes all the pillars that were previously removed can be brought back 7 The secondary reefs are low grade and management are prepared to mine them at cost
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Is pillar mining a game changer?

Management believes mining pillars holds significant upside. Some pillars are high grade and mining them 8 would appear to be highly profitable . However, pillar mining is a low volume and slow process. Additional capital will be required for development. In addition, current fixed costs are as high as 80%. To therefore realise any benefit from pillar mining, management will need to arrest the current decline in production first to keep unit costs under control. The potential reserves are not likely to become reserves until updated pre-feasibility studies are complete and/or further drilling is done. We therefore expect these plans to take at least two years to implement.
Can declining production be stopped?

Management faces a difficult task to buck a 10-year trend of falling grades, volumes, development metres 9 and productivity . Below we show the current Life-of-Mine Plan and the BNP Case. EXHIBIT 2: Sibanye Gold production outlook
(moz pa) 2.5 2.0 1.5 1.0 0.5 0.0
2013E 2014E 2015E 2016E 2017E 2018E 2019E 2020E 2021E 2022E 2023E 2024E 2025E 2026E 2027E 2028E 2029E
2016E (moz) 1.2 1.3 1.17 1.16
Current LoM plan
BNP case
2007
2008
2009
2010
2011
Source: BNP Paribas Cadiz estimates
EXHIBIT 3: Potential reserves

2012 (moz) Management guidance Life-of-mine plan BNP Case Source: BNP Paribas Cadiz estimates 1.22 1.22 1.22 2013E (moz) 1.29 1.29 1.22 2014E (moz) 1.2 1.3 1.25 1.15 2015E (moz) 1.2- 1.3 1.26 1.16
The BNP Case assumes production is unlikely to recover from FY12 levels and it will take at least two years 10 to stabilise production . New projects should come online by 2017.
8 9
EXHIBIT 25 Since 2008, tonnages have fallen 8% pa, grades 4% pa, and development metres 6%pa. Blast frequencies have halved. 10 See Appendix 11: Production outlook for further details on our assumptions
2012
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Surface rock dumps (SRD)

Surface rocks dumps are an important source of low-cost ounces for the business. Sibanye expects SRDs to contribute 23% to group free cash in 2013. However reserves are almost depleted and production will shrink to one third by 2015. EXHIBIT 4: LoM production from SRDs
(koz) 120 100 80 60 40 20 0 2012 2013E 2014E 2015E 2016E 2017E Kloof Driefontein Beatrix
EXHIBIT 5: Real free cash from SRDs

(ZAR m) 600 500 400 15 300 10 200 100 0 2012 2013E 2014E 2015E 2016E 2017E 5 0 Real free cash before tax Contribution to group (%) 25 20
Real free cash from SRDs is expected to diminish by 2015 and the contribution to the group is expected to fall from 23% this year to 7% by then. Management is considering a new tailings project to replace the SRDs. From our analysis, the project could yield positive cash flows in the first two to three years but the 11 project as a whole is NPV negative .
Costs aligned with industry

Sibanye has one of the highest C1 cash costs in the industry. However, its C3 cash costs, which are more reflective of sustaining cash flows, are similar to the rest of its South African peer groups (EXHIBIT 6). Sibanye is currently cash generative at current gold prices with C3 cash margins of about 10%. EXHIBIT 6: Spread between peer group C1 and C3 cash costs
(USD/oz) 1,400 1,300 1,200 1,100 1,000 900 800 700 C1 C3 Harmony Gold Sibanye Gold AngloGold Gold Fields
12 13
Normalised cash costs for Q1 2013, we include capex for South Deep as part of Gold Fields C3 costs Source: BNP Paribas Cadiz estimates
We estimate that Sibanyes real C3 cash costs can be maintained at about ZAR380,000/kg for three years if 14 production is sustained above 1.15moz and cash savings of at least ZAR1b (10%) are achieved .
11 12 13
This is due to declining grades. Our NPV-neutral price is USD1,750/oz gold and USD65/lb U3O8.See Appendix 8: Tailings project BNP Paribas Cadiz mining cost definition:C1 = mining cash costs BNP Paribas Cadiz mining cost definition:C3 costs = mining costs + corporate overheads + on-going capital 14 We estimate about ZAR0.5b in savings have already been achieved largely as a result of retrenchments.
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Risks
Cash costs Sibanye already has one of the highest cash costs in the industry at the C1 level. It is the only company amongst its peer group with a declining production outlook. In addition, fixed costs of almost 80% are higher on average compared to other deep-level South African mines. If management cannot arrest the falling production trend, Sibanyes real C3 cash costs will likely rise above USD1,400/oz in FY14, and become the highest cost South African producer. Once empowered, always empowered? Sibanyes direct Black Economic Empowerment (BEE) ownership is 11.75%. However the company has a legal opinion that it complies with the minimum 26% ownership requirement based on Mvelas sale of its 15% stake in Gold Fields in 2009. Although the Mining Charter is not part of the MPRDA Mining Act, companies appear to be interpreting it in different ways. The risk is that not maintaining a direct 26% ownership could come back to haunt shareholders. Loss of skills Management has already retrenched 3,000 employees this year so far, with more expected. Plans to cut out middle management could cause senior managers to be spread more thinly between operations. This could have a negative impact on productivity. More safety-stoppages Pillar mining is a high-risk activity. Safety and regional pillars are left over from previous mining activities to provide support. These pillars are located between 2km and 3.5km below surface. Management will have to thoroughly understand the rock mechanics in order to minimise seismicity if they decide to mine these pillars. We think this could result in more safety stoppages and negatively impact production.
Earnings
We forecast earnings of ZAR1.150/share for FY13. We expect the company should also manage to fund a dividend this year of about 30cpsfrom free cash. However, at constant prices, the company is unlikely to generate any future earnings. A gold price of USD1,450/oz or ZAR440,000/kg should sustain positive earnings for two more years. EXHIBIT 7: Sibanyes earnings outlook
2012 (ZAR m) Gold price USD/oz ZAR/USD Revenue Cash costs Gen. & Admin. Rehab Restructuring EBITDA Net interest Net profit before tax Tax Deferred tax Tax rate Net income HEPS Source: BNP Paribas Cadiz estimates 1,669 8.21 16,554 10,874 229 48 124 5,935 21 2,615 475 (840) (14%) 2,978 4.08 2013E (ZAR m) 1,460 9.3 16,623 11,837 229 48 500 3,821 359 1,171 90 240 28% 842 1.15 0% (320) (0.44) 0% (795) (1.09) 2,363 393 (83) 2,054 478 (613) 2014E (ZAR m) 1,380 9.4 14,813 12,086 200 48 2015E (ZAR m) 1,380 9.4 15,074 12,681 216 48
We forecast earnings at constant gold prices but using nominal costs. We include real cost savings of ZAR1b over the next 18 months. We expect restructuring costs of about ZAR500m this year, largely for retrenchments. Interest charges are likely to rise as the need for debt increases. The company maintains a dividend policy of 25% to 35% of earnings.
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Valuation
We value Sibanye at ZAR7/share based on a real DCF methodology at current prices of USD1,380/oz and 15 ZAR9.4/USD. Below we show the NPV of the mines including net debt to arrive at our target price . If we 16 based our valuation on the current LoM plan, our valuation would be ZAR11/share . This is a result of our more bearish outlook on production in the early years as well as more capital to develop new projects that will sustain production for longer in the later years. EXHIBIT 8: Valuation of Sibanye Gold based on BNP Case and current LoM plan
(ZAR/share) 12 10 8 6 4 6.1 2 0 Driefontein Kloof Beatrix Net debt BNP Case Current LoM Case 4.5 0.3 3.8 7.0 11.0
Source: BNP Paribas Cadiz Securities
Sensitivity
Below we show the sensitivity of the gold price and ZARUSD exchange rate to our Best Case DCF valuation. EXHIBIT 9: Sensitivity of target price to ZAR gold
-------------------------------------------------------- ZAR/USD --------------------------------------------------------9.0 1,350 1,380 USD/oz 1,400 1,450 1,500 1,550 1,600 Source: BNP Paribas Cadiz Securities 1 3 4 8 11 14 16 9.4 5 7 8 12 14 17 20 10.0 11 12 14 26 20 23 26 10.5 15 17 18 22 25 28 32 11.0 19 21 23 26 30 33 37
High C1 and C3 costs as well the high fixed/variable cost ratio makes the share highly geared to ZAR gold.
15 16
We do not include any environmental liabilities since these funds are currently fully funded. This is due to higher production in the early years and declining capex from next year on.
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Investment recommendation
Sibanye is currently cash generative, has an attractive forward PE of 6x and dividend yield of at least 4%. Although Sibanye has one of the highest C1 cash costs in the industry, its C3 cash costs are in-line with peers and the weaker rand has helped boost cash margins above its competitors. Based on historical trends and the current LoM plan, the fundamental outlook looks poor. However, management has a bullish view that production of between 1.2moz to 1.3moz pa can be sustained over the next few years. We think this is optimistic. Production has been falling sharply over the last decade in line with falling development. To arrest this trend development rates will need to be maintained.Q1 FY13 development rates indicate a good recovery from last years labour strikes, although further improvement is necessary. In addition, blast frequencies need to be improved. This will require a change to labour working schedules, something management will likely struggle to negotiate in the upcoming wage review. Plans to mine pillars and secondary reefs will take at least two years to implement although mining them will be unlikely to provide any upside but more safety risk, in our view. The company cannot afford any new projects at current prices. It currently has net debt of ZAR2.8b and will not be able to generate enough free cash as production from the low-cost SRDs begins to fall. We value Sibanye at ZAR7/share at current prices. Upside and downside risks include the gold price, USD/ZAR exchange rate, production and potential safety stoppages. Buyers of this stock need to hold the view that the gold price will rise or the rand will weaken further. We initiate coverage with a HOLD recommendation.
10
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 1: Company background

Sibanye Gold owns and operates the KDC complex and Beatrix mines. The location of these mines is shown in EXHIBIT 10 below. EXHIBIT 10: Location of mines
Source: Sibanye Gold
The KDC complex is situated in Carletonville outside Johannesburg and consists of KDC East (Kloof) and KDC West (Driefontein). Their lease areas are contiguous although their mining operations are not. Beatrix is located near the town of Virginia in the Free State province, on the south-western corner of the Witwatersrand Basin. The company is also developing a tailings project in partnership with Gold One. Below we show the corporate structure of Sibanye Gold and top shareholders. EXHIBIT 11: Corporate structure of Sibanye Gold EXHIBIT 12: Top shareholders as at 26 April 2013
Shareholder Ownership (%) Investec Allan Gray 15.21 10.13 5.40 4.90 4.34
Sibanye Gold
100%
First Eagle Van Eck
Kloof
Driefontein
Beatrix
Tailings project
PIC
Source: BNP Paribas Cadiz Securities; Sibanye Gold
11
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Below we show the relative size of each operation based on production in FY12. EXHIBIT 13: Annual production (FY12A) of existing operations
Kloof
Driefontein
Beatrix
SRDs (oz) 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 450,000 500,000
Appendix 2: Geology and mining operations

KDC Currently three main reefs are mined; VCR, Middlevlei and Carbon Leader. The VCR is the dominant reef mined. This reef dips at 22 to the south-southeast. There are secondary reefs although these account for less than 10% of current reserves. Mining takes place at depths of between 600 and 3,350 metres below surface (mbs).The main mining method is breast stoping with dip pillars. The KDC lease area is effectively split into two by the Bank Fault. Kloof lies to the east and Driefontein to the west. Operations consist of 12 separate shaft systems, although they are highly integrated with respect to ventilation and ore transport systems. Costs are allocated per shaft, grouped together into five business units (EXHIBIT 14).We expect that these units may be consolidated as part of cost cutting measures. Any such details will be released this year when management reveals its new strategy for the company. EXHIBIT 14: KDC Business Units
Business unit 1 2 3 4 5 Source: Sibanye Gold Mine Driefontein Driefontein Driefontein Kloof Kloof Shafts 1 and 5 2 and 4 6, 7, 8 and 10 3 and 4 Main, 7, 8 and 10
Beatrix The reefs being mined include Beatrix Reef, Aandenk Reef, VS5 Reef and Kalkoenkrans Reef. The dip of these reefs is to the north between 0 and 15 except for the deeper Kalkoenskrans Reef which dips to the east at 8 to 15. There is a deeper lying Beisa Reef although it is too fractured to be mined at current gold prices. Beatrix includes 1 and 2 shaft (South section), 3 shaft (North section) and 4 shaft (West section).Four shaft was initially developed as a uranium mine. Mining takes place at depths of between 600m and 2,155m below surface. South section reaches depths of 1,000mbs;North: 1,500mbs and West: 2,500mbs. Over 80% of operating profits are generated by North section. The main mining method is conventional breast mining. The mine has high levels of methane gas which is a hazard to safety and production although this appears to be well managed.
12
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Pay-limits and rationalising the operations Kloof and Driefontein are mining at levels 37% above the pay-limit, as of 2012. Beatrix however has been operating at the margin (EXHIBIT 15). This is due to the high costs associated with the West section. The reserve pay-limit is 1,690cmg/t, well below the average mined value of 1,045cmg/t for (EXHIBIT 16). The pay-limit for Beatrix would fall to about 900cmg/t without Beatrix West. The West section is plagued by depth, heat and long haulage distances. In addition, the presence of smectite in the footwall has caused excavations to close making mining technically challenging in this area. Gold grades at this mine are4.3g/t on average. EXHIBIT 15: Pay-limits and mined values
(cmg/tonne) 2,500 2,000 1,500 1,000 500 0 Kloof
Sources: Sibanye Gold; 2012 figures
EXHIBIT 16: Pay-limit at Beatrix North, South and West sections

Mined value (cmg/tonne) 1,800 1,600 1,400 1,200 1,000 800 600 400 200 0 780 Average mined value: 1,045cmg/tonne 900 1,690
Reserve pay limit
Driefontein
Beatrix
North
Sources: Sibanye Gold; 2012 figures
South
West
Pre-developed ore reserves have been significantly depleted as a result of the strikes in Q4 2012 as well as from a fire earlier this year. Although the mine only contributes 6% to group production, cash costs are currently about R550,000/kg which places this mine in a cash negative position. Management at both KDC and Beatrix plan for unpay/pay ratio of 30%/70%. Beatrix West has not been profitable since 2011 and is currently in a Section 189 process, the outcome of which will determine whether the mine will remain open or not. Management has two choices: 1 2 Continue operating the mine at a cash loss of at least ZAR300m pa at current prices. Place the mine on care and maintenance and keep it as an option on gold and uranium. This will come with a once-off cost of about ZAR250m for the retrenchment of about 3,000 people. Uranium grades at Beatrix West are 0.4kg/t on average and the breakeven price is about USD55/lb U3O8.
Management intends on pursuing the second choice in order to maintain the option on uranium.
13
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 3: Reserves
As of December 2012, Sibanye Gold held 13moz of underground reserves and 520koz of surface rock dump material. Underground reserves Of the three main operations, Kloofholds the majority of the reserves (42%), has the highest reserve grade of 8.3g/t and the longest life of 15 years. Below we show the change in reserves and reserve grades for each mine since 2007. EXHIBIT 17: Historical reserves per mine
(moz) 25 20 15 10 5 0 Kloof Driefontein Beatrix 21.2 19.5 18.0 13.3 10.7 10.2 7.9 7.0 2007 2008 2009 2010 2011 2012
EXHIBIT 18: Historical reserve grade per mine

(g/tonne) 12 10 8.30 8 7.50 2007 2008 2009 2010 2011 2012
11.9 8.4 5.5 6.4 4.2 6.7 6.4 5.4 4.9 3.3
6 4.30 4 2 0 Kloof Driefontein Beatrix
Sources: Sibanye Gold; reserves for underground operations only

17
Sources: Sibanye Gold; reserve grades represented for underground operations only
Despite the rise in gold price , reserves have fallen 70%in the last five years. Reserve grades have fallen from 8.6g/t to 7g/t over the same period. Where did all the reserves go? Below we show the reconciliation of underground reserves since 2007. EXHIBIT 19: Reconciliation of underground reserves
(moz) 50 45 40 35 30 25 20 15 10 5 0
42.9
8.0 10.8
2.1
?
8.9
13.0
2007 reserves
Production
Removal of reserves below infrastructure
Exclusion of pillars due to safety
Other exclusions
2012 reserves
Sources: Gold Fields; Sibanye Gold
Since 2007, the following major changes were made: 10.8moz (31%) reserves lying below infrastructure18
17
KDC and Beatrix Technical Short Form Report 2011 for a 10% increase in the gold price, it is expected that reserves should increase about 10%.
14
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
2.1moz (6%) from pillars (as a result of safety concerns) The outstanding 8.9moz were removed for economic reasons which includes the following: 2.5moz due to lower MCF and higher pay-limits. 2.6moz from the closure of D8 and K8 shafts. This was done to conserve cash in the short term by focusing on the most profitable mining areas only. Approximately 1moz were removed from D1 shaft after further drilling confirmed those reserves were not actually there. This leaves 2.8moz unaccounted for although we expect this was removed for similar reasons discussed above. Below we list the four main reasons for the decline in reserves and discuss the first three: 1 2 3 Rising pay-limits Lower mine call factor A more strict adherence to the SAMREC Code, which details the rules for declaring reserves in South Africa Closures of certain mining areas to conserve cash in the short term
Rising pay-limits Resources above the pay-limit can be considered reserves. A primary input in calculating the pay-limit is an estimation of costs. Below, we show historical costs as well as the pay-limits in 2009 compared to 2012. EXHIBIT 20: ZAR/tonne costs at KDC and Beatrix
(ZAR/tonne) 2,500 2,000
1,200
EXHIBIT 21: Pay-limits in 2009 and 2012

(cmg/t) 1,600 1,400 2009 2012
KDC
Beatrix
1,500 1,000
1,000 800 600
500 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Source: Gold Fields
19
400 200 0 Kloof

Source: Gold Fields
Driefontein
Beatrix
Historical unit costs at KDC and Beatrix have risen 14% and 16% y/y since 2005, respectively. The rate at 20 which costs have been rising has also been on the rise. This is largely due to falling tonnages but also poor mining practice. Channel widths have widened from about 180cm to 140cm. This has resulted in higher dilution, raising the pay-limit in the estimation of reserves. It is likely that in calculating reserves historically, future costs may have been underestimated. This would have artificially inflated estimates for reserves. As costs have been rising unexpectedly, increasing the paylimits, hence reserves have fallen.
18 These reserves did not form part of a Feasibility Study and since management took the decision to follow SAMREC compliance more strictly, had to be removed from the reserve statement. Not all South African companies apply the same principles in this regard. 19 CAGR calculated based on quarterly data 20 See Appendix 12: Costs and capex
15
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Lower mine call factor The mine call factor has fallen as much as 10% since 2008 (EXHIBIT 22). Management cites the reasons to be poor discipline and the depletion of old gold which does not get accounted for in the MCF calculation. EXHIBIT 22: Historical mine call factor
MCF 2008 (%) Kloof Driefontein Beatrix Source: Sibanye Gold 86 90 88 2009 (%) 85 91 85.5 2010 (%) 83 90 84 2011 (%) 85 85 74 2012 (%) 81 80 79
SAMREC In 2009, the South African Code for the Reporting of Exploration Results, Mineral Resources and Mineral Reserves (SAMREC CODE), made a significant change to its guidelines. This included the requirement of a Pre-Feasibility Study, as opposed to only a Concept Study, in the declaration of reserves. In the case for KDC, the KEA project at Kloof and the tailings project were removed, a total of 3.7moz.
Appendix 4: Life-of-mine and amortisation

Historical life-of-mine (LoM) The downward adjustment to the reserves has significantly impacted Sibanye Golds life of mine. Below we show the estimated life-of-mine since 2007. EXHIBIT 23: Historical LoM estimates
(LOM calendar year) 2045 2040 2035 2030 2025 2020 2015 2007
Sources: Gold Fields; Sibanye Gold
Kloof LoM
Driefontein LoM
Beatrix LoM
2008
2009
2010
2011
2012
Driefonteins LoM has been reduced from 2041 to 2023, in line with reserves. Kloofs LoM has increased from 2023 to 2027 but has been volatile following the inclusion and then the exclusion of the KEA project. Beatrix LoM has increased from 2020 to 2025.
16
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Impact on amortisation Amortisation of capital is largely determined by LoM estimates. The rapid fall in reserves, and consequently the fall in life-of-mine, together with steady stay-in-business capital, has resulted in larger amortisation charges. We illustrate this trend in EXHIBIT 24 below. EXHIBIT 24: Life of mine and amortisation
(Rm) 1,800 1,700 1,600 1,500 1,400 1,300 1,200 1,100 1,000 900 800 2007 2008 2009 2010 2011 2012 2028 2026 2024 2022 Amortisation LOM avg (LoM calendar year) 2036 2034 2032 2030
Sources: Gold Fields; Sibanye Gold; BNP Paribas Cadiz Securities
We expect capex to remain at least steady and LoM is unlikely to increase. Amortisation charges to at least remain steady.
17
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 5: What is left in the lease?

Whilst under the Gold Fields umbrella, capital programmes at KDC and Beatrix had been held back. Sibanye now wish to re-capitalise the business by re-visiting previously withheld projects. These projects include the following areas of potential: 1 2 3 Mining of pillars, including safety-related pillars, regional pillars and shaft pillars at KDC Mining of secondary reefs including the Middlevlei and Kloof Reefs at KDC Mining below infrastructure at both KDC and Beatrix
Managements strategic plans for KDC Pillars The extraction of remnant pillars between 2km and 3.5km below surface at KDC is being reviewed. Management intends mining these pillars by using new technologies, de-stress mining and over-stoping techniques. De-stress mining is intended to be achieved by mining the secondary reefs that lie above the pillars. Pillars were removed from reserves in 2010. This was largely a management decision due to safety concerns. We expect that it will be difficult for the company to bring all these ounces back into reserves for the following reasons: 1 KDCs normalised ZAR/tonne costs have increased 2.5x since 2005. Research on the extraction of pillars at KDC Main Shaft in 2005 indicated an average cost of ZAR750/tonne or about ZAR1,900/tonne in todays money. High cost inflation over the last few years will likely have an impact on the economics of extracting these pillars. The pillars are scattered over a large area and their economic viability will largely depend on a shaftby-shaft basis. KDC has relatively high fixed costs (80%) and in many cases, we think it will be difficult to maintain a single shaft by extracting pillars alone.
21
In addition, due to the inherent safety risk in mining pillars, management will need to get consent from the DMR. The high safety risk relating to pillar mining is largely due to challenging rock mechanics.
21
Adjusted for production based on a fixed: variable split of 80:20.
18
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Below we show the location of the pillars at KDC. We also show the cost to extract certain pillars at Main 22 Shaft . EXHIBIT 25: Location of pillars
Pillars at main Shaft: between 2km and 3.5km deep, 63kt, 22g/t average grade, 45k oz, mined over 21 months, mining cost of ZAR1,900/t or R86,000/kg.
dip
We make the following observations: The pillars are spread over a wide area which would require many shafts to remain open. The extraction of shaft pillars is a slow incremental process and volumes would be significantly lower compared to current volumes22. Some pillars are high grade and mining them would appear to be highly profitable. However, development will be required to access the secondary reefs in order to de-stress the pillars. This is not factored into the mining cost. Based on the above, we do not expect all the pillars to re-enter reserves. We also do not expect pillar mining to add a significant benefit to the business due to low volumes and the existing high fixed cost base 23 of the business . Secondary Reefs There are three secondary reefs located at Kloof mine: Middlevlei, Kloof and Libanon Reefs. They are low grade but shallow. No resource data has been made available. Management is expected to provide plans by July 2013 for potential ways in which to mine these reefs. Of the three reefs, management appears most interested in the Middlevlei Reef. Due to existing infrastructure, accessing this reef would require less-thannormal development and costs should be relatively lower. As mentioned previously, management intends mining pillars in conjunction with secondary reefs. By mining the secondary reefs first, it will allow for de-stressing of the pillars. However, this practice alone poses a high safety risk. The Middelvlei Reef lies 50m to 75m above the Carbon Leader (at Driefontein) and mining this ground above a void makes seismicity, hence falls of ground, more likely. This risk is similar to the limitation miners have in the number of lifts one can mine at once using the block-cave mining 24 method .
22 23
Adapted from estimates from Project report pillar mining at No.1 and 2 sub-shaft on Kloof Gold Mine by S. Van der Merwe (SAIMM) As much as 80% of cash costs are fixed. 24 A good example is the mining approach for Wafi Golpu.
19
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Shaft pillars Mining has commenced at K1 shaft pillar at Kloofand D4 shaft pillar at Driefontein. Management indicates that the only other shaft pillar they could potentially mine is D1. Other resources and inventory In addition to the above three areas of potential, management has indicated there are significant resources 25 and inventory they are busy reviewing to bring back into reserve (EXHIBIT 26). EXHIBIT 26: Resources and Inventory at Kloof and Driefontein
Kloof (moz) Resources Inventory Total Source: BNP Paribas Cadiz Securities 12.2 10 22.2 Driefontein (moz) 8.2 11 19.2
The company intends to provide an update on its strategy that will incorporate the above ideas by July 2013. Below we assess some of these ideas that management has discussed in conjunction with the latest mine plans. Kloof Below we assess the Kloof lease area. EXHIBIT 27: Kloof lease area EXHIBIT 28: Kloof reserves by mine shaft (December 2012)
(koz) 3,000
2,500
Grade too low Pillars: potential to re-open 10m oz resource
2,000
1,500
Middlevlei opportunity
High grade but below infrastructure Gold Fields South deep area
1,000
500
Low grade, below infrastructure, too far
0 8 shaft
Source: Sibanye Gold Source: Sibanye Gold
1 shaft
7 shaft
3 shaft
2 shaft
4 shaft
25 Inventory is a relatively uncommon category of mineral classification. It refers to resources that exist although not declared based on managements decision not to mine.
20
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Recent changes to Kloof mine include a reduced level of mining at7 shaft. This has effectively brought mining in the southern part of Kloofs lease area to an end, primarily as a result of low grades. We therefore expect that reserves at 7 shaft (456koz) will likely be removed. Mining of the shaft pillar at Main Shaft has begun although this is already in reserves. The Middlevlei Reef offers opportunity in the 8 shaft area. This reef is close to existing infrastructure. There is significant shallow un-mined ground at Venterspost (6 shaft) to the north although grades here are too low (about 2g/t) for economic extraction. What is effectively left for Kloof mine is 4, 3 and 2 shaft, and a new project called the Kloof Extension Area, which we discuss below. Kloof Extension Area (KEA) A 2.26moz resource potential lies below infrastructure on the eastern boundary at Kloofs 4 shaft. Management intends on accessing these potential reserves via a spiral decline (46 decline) which would extend 44 level to eventually 48 level. Extension beyond the boundary into South Deeps lease area is not possible due to displacement by the West Rand fault. Drilling results indicate grades of between 11g/t and 12g/t. Management estimates that production could begin in no less than three years at a cost of ZAR0.5b to gain access to 46 level. An article by Mining Weekly in October 2006, when the project was first planned for, cited a cost of ZAR1.4b. This higher number is based on extending down three levels. It also indicated peak production of about 200koz pa. This represents significant potential to the lease area although depths would need to be increased beyond 4km below surface. This would not come without an increase to real 26 costs (largely due to needed refrigeration ). Based on a resource/reserve ratio of 36% , we estimate about 800k oz reserve potential for the KEA zone in the next three to four years. However this project is likely to only offer some replacement to the mines falling production profile. Driefontein Below we assess the Driefontein lease area. EXHIBIT 29: Driefontein lease area
Pillars and secondary reef potential 1moz
27
EXHIBIT 30: Driefontein reserves per mine shaft (December 2012)

(koz) 2,000 1,800 1,600
D1 shaft pillar
1,400 1,200 1,000
9 Shaft difficult to access, requires new shaft
800
D5 shaft deeps
600 400 200 0 10 shaft 6 shaft 8 shaft 2 shaft 1 shaft 4 shaft 5 shaft
26 27
Virgin rock temperatures are of the order of 55C Derived from the resource/reserve ratio of 4 shaft zone, of which the KEA zone is an extension of.
21
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Mining at 2 and 8 shaft has been very successful for Driefontein to date. The majority of reserves now remain at 5 and 4 shaft. Management is considering entering ground below infrastructure at 5 shaft. This area has potential for three new levels. Each level has a potential of 300koz reserve. We expect that at least two years would be needed to gain access and produce gold from this area. However, depths would reach below 3.5km and we would expect this to remain only an option for now at current prices. There is potential to mine the ground at 9 shaft although this would need to be done from its neighbour, AngloGolds Tau Tona mine. Nine shaft had been ideally positioned to access reef from both Driefontein and Tau Tona, as a joint venture project. The project started in 2008 but was then stopped due to power supply concerns and capital commitments for South Deep. Management indicates that there is a potential of 8.5m oz resource. Mining at 9 shaft remains as a long-term option. Management intends on mining 1 Shaft pillar at Driefonteinvia 2 Shaft. The company has indicated that there is 560k oz of mineable reserves which they expect to mine from 2017 onwards. This amount has not yet been included in the reserve statement. We expect these ounces to provide about 2 years life extension at the indicated LoM production rates. There is about 1m oz potential from pillars at 10 and 6 shaft. These pillars would be mined in conjunction with the Middlevlei and Kloof reefs. Beatrix Below we assess the Beatrix lease area. EXHIBIT 31: Beatrix lease area EXHIBIT 32: Beatrix reserves per mine section
(koz) 2,500
2,000
Too deep and far
1,500
1,000
Vlakpan
Down-dip extension
500
0 South
Sources: Sibanye Gold; BNP Paribas Cadiz Securities
West
North
West section is no longer economic, as previously discussed. We therefore exclude production from this shaft in our forecasts. This mine however does still hold potential in addition to holding an option on uranium. The Beisa Reef which lies below the Kalkoenkrans Reef, holds 7.5moz in resource. However, it is highly fractured and too costly to mine at current prices. South section has about three years life remaining based on current reserves of about 250koz. Growth or replacement of reserves at Beatrix can only come from below infrastructure within the current lease area. This includes the Vlakpan ground west of North Section and down-dip opportunities at North shaft: Vlakpan Early development work to access the Kalkoenkrans Reef between 16 level and 22 level west of Beatrix North shaft has begun. The area is relatively shallow (1,500mbs). Management is considering access via a down-dip decline or a sub-vertical shaft. The project could add 230koz potential reserves.
22
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
North Section The down-dip extension at North section would unlikely result in growth but replacement of lower cost with higher cost ounces. In summary: There are only two capital opportunities the company intends to pursue within the current lease area: the KEA project at Kloof mine and Vlakpan project at Beatrix mine. The resources at these areas lie below infrastructure are not significant enough to either grow or maintain production, or reduce unit costs, in our view. Opportunities beyond the lease area There are a number of opportunities adjacent to the current lease areas at Beatrix and Driefontein mine, as shown in EXHIBIT 33 and EXHIBIT 34 below. These include: Driefontein: As mentioned previously, 9 shaft can only be mined from the neighbouring Tau Tona mine. This ground could be sold to AngloGold although AngloGold management does not intend pursuing this. Beatrix:Two options exist. The first one concerns Wits Golds Bloemhoek lease area. This area lies north of the down-dip extension from Beatrix North.Bloemhoek has an estimated probable reserve of 5.4Moz at a head grade of 5.33g/t. Wits Gold is an exploration company at present although it has been pressured by shareholders to execute on existing projects. The second option concerns Joel mine belonging to Harmony Gold. This ground lies east of South shaft. At Beatrix however, there are effectively three companies with contiguous ground that offer potential opportunities for corporate action. Management however does not expect any corporate action concerning KDC or Beatrix in the short to medium term. EXHIBIT 33: Beatrix lease area and neighbours EXHIBIT 34: Driefontein lease area and neighbours
AngloGold West Wits Wits Gold
Harmonys Joel mine
23
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 6: Potential upside

Based on our assessment of opportunities within and beyond the lease, we summarise the potential reserves that we think are probable, apart from the current LoM plan. We also show the estimated annual production and start year. EXHIBIT 35: Potential reserves
Source Potential reserve (koz) Pillars Secondary reefs KEA project D1 pillar Vlakpan Sources: Sibanye Gold; BNP Paribas Cadiz Securities 2,100 800 560 230 11.0 6.2 4.3 2.6 Life (years) Annual production (koz) 19,028 13,029 13,029 9,030 2015 2017 2017 2021 Start year
We estimate there is no less than 3.7m oz of potential reserves, assuming all the pillars that were previously removed can be brought back. These reserves are not expected to increase production, but to sustain it and potentially extend the current life of mine by no more than three years. We do not account for potential reserves from secondary reefs which could further sustain production or extend life-of-mine. We do know that secondary reefs form part of the resources and inventory shown in EXHIBIT 26. Management is prepared to mine secondary reefs at cost. We do not expect mining them will be value accretive to the business, but necessary to de-stress the mining of pillars below.
28 29
Calculated using reserve over remaining Life of Mine for KDC Calculated based on 400kt pa at an average grade of 10g/t 30 60kt pm(Beatrix CPR Report 2012), grade based on reserve grade of 3.9g/t at North Section
24
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 7: Surface rock dumps

All three mines produce gold from surface rock dumps (SRD). This material should not be confused with lower grade tailings from processed ore. SRDs are waste material that has been stockpiled over the years. The SRDs are expected to contribute 23% to group free cash in 2013 and therefore form an important part of the business, although we think this is often overlooked by the investor community. Below we assess the expected life and profitability of the SRDs. Kloofs SRD reserves are the largest in the group (66%), followed by Driefontein (27%) and Beatrix (7%).Total reserves are 520koz. The resources and reserves for SRDs are typically well defined since they lie on the surface and have recently undergone an intensive drilling campaign. The profitability of these three operations all vary as a result of different head grades, production volumes and operating costs. Below we forecast key operating information for FY13. EXHIBIT 36: SRDs key operating data for FY13F
Kloof Tonnes Head grade Ounces Cash cost Cash cost ktpa g/t oz/a R/t R/kg 3,300 0.66 47,622 79 175,000 Driefontein 3,200 0.60 40,129 173 443,000 Beatrix 1,250 0.26 5,305 83 630,000
Sources: BNP Paribas Cadiz Securities; estimates based on current reserves; LoM plan and historical quarterly information
Additional capacity of 100kt pm is expected to be installed at Kloof by June 2013 and a further 40kt pm in FY14.This will boost total capacity by 28%. Operations at Kloof are the most profitable, followed by Driefontein, while Beatrix is likely to be cash negative from FY13. The differences here are largely a function of grade, and to a lesser extent, rock hardness. Below, we show expected production and the outlook for real free cash. Kloof has the largest reserves, largely as a result of a major sampling campaign. SRDs are expected to become fully depleted by 2017. EXHIBIT 37: LoM production from SRDs
(koz) 120 100 80 60 40 20 0 2012 2013E 2014E 2015E 2016E 2017E Kloof Driefontein Beatrix
EXHIBIT 38: Real free cash from SRDs

(ZAR m) 600 500 400 15 300 10 200 100 0 2012 2013E 2014E 2015E 2016E 2017E 5 0 Real free cash before tax Contribution to group (%) 25 20
Sibanye expects real free cash from SRDs to more than halve by 2015 while the contribution to group free cash is expected to fall from 23% this year to 7% in 2015. This is a result of two main reasons. 1 2 Grades from Kloof fall from 0.66g/t this year to 0.4g/t in 2015. Driefontein reserves become fully depleted by 2014.
The contribution to the group rises y/y in 2013 as a result of higher capital budgeted for the underground operations, as well as a lower spot gold price.
25
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
The company intends replacing the cash flows from these operations with a new project to process surface 31 tailings. However, we dont think this project will be value accretive .
Appendix 8: Tailings project

Management is considering two options for processing tailings. Option 1: Last year Gold Fields and Gold One engaged in a joint venture study for developing a large scale surface reclamation plant to produce gold and uranium. Gold would effectively be processed from Sibanyes dumps and uranium from Gold Ones dumps. Results from a Scoping Study placed the capital cost between ZAR8b and ZAR14b with a production rate of 4mt pm feed to produce an average of 300koz Au eq. per annum over a 20-year life. The Scoping Study has been progressed to a Pre-feasibility Study with a focus to reduce capital costs. Management expects that the capital cost should not exceed ZAR10b. This project, if executed, would make it the largest tailings facility in the world. Currently, Ergo Mining, belonging to DRDGold, is the largest with a production rate just over 2mt pm. Option 2: A smaller Pre-feasibility Study is being done to assess the viability of producing gold only from KDCs 13 tailings dumps, and eventually from run-of-mine material as well. Management expects the project will cost no more than ZAR6b for a 2.4mt pm processing plant. Operating costs are expected to 32 be less than ZAR100/t. Current resources are estimated at 372mt at an average grade of 0.32g/t to give about 3.7m oz gold. We estimate recoveries would be about 35% on average based on recoveries achieved at Ergo. Hence potential reserves are estimated at about 1.3m oz. Assumptions and results Below we summarise the assumptions for each option. Our DCF valuation (at current prices and 10% real discount rate) results in a negative NPV for both options. Our NPV-neutral price is USD1,750/oz gold and USD65/lb U3O8. Although both options are cash positive in the early years, gold grades begin to decline to below 0.3g/t. Recoveries could be enhanced by the use of ultra-fine grinding. This could reduce our breakeven price from USD1,750/oz to between USD1,500/oz and USD1,600/oz. EXHIBIT 39: Summary of Option 1 and Option 2 tailings project
Option 1 Capacity Recoverable reserves Gold Recovery % g/t moz Uranium Recovery % kg/t mkg Capex Opex NPV Breakeven price Gold Uranium Sources: Sibanye Gold; BNP Paribas Cadiz Securities USD/oz USD/lb 1750 65 1750 ZARbn ZAR/t ZARm 80 0.19 12.6 9 100 <0 5 60 <0 35 0.32 1.65 35 0.32 1.3 mtpm 4 Option 2 2.4
31 32
Appendix 8: Tailings project Grades vary between 0.5g/t and 0.23g/t
26
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Environmental challenges As with DRDGold, mining surface dumps brings with it many environmental responsibilities and potential resistance from lobby groups. The primary concerns with any mining dump are dust, uranium and contamination of ground water. Preventative measures can be taken although often not enough is done. In the case for the tailings project, material would be re-deposited from 13 dumps scattered around KDC to one mega deposition site at South Deep. This could be a net positive impact on the environment as long as the new mega dump is built to a world-class standard. Management will need a revised Record of Decision from government, which would allow them permission to deposit sulphur rich, and uranium rich (in the case for Option 2) material on the new mega dump. This will likely receive resistance from lobby groups.
Appendix 9: Current life-of-mine plans

To assist in our outlook for production, we consider the 2012 Competent Persons Reports prepared by management at each mine. In EXHIBIT 40 we show the historical and planned head grades relative to the reserve grades. In EXHIBIT 41 we show historical and planned production tonnages. In EXHIBIT 42 we show historical and planned development. These plans do not include any potential upside from remnant pillars or shaft pillars. Firstly, we make the following observations with respect to grades: 1 2 3 Head grades at all three mines have been trending down in line with reserve grades. Head grades on average have been significantly lower than reserve grades. Planned grades however have been forecast to rise at all three mines. In addition, grades at Kloof and Beatrix have been forecast to rise above reserve grades.
EXHIBIT 40: Head grades and reserve grades

(g/tonne) 12.0 10.0 8.0 6.0 4.0 2.0 0.0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Kloof Beatrix Driefontein reserve grade Driefontein Kloof reserve grade Beatrix reserve grade
Sibanye LoM Plan
Secondly, we make the following observations with respect to production tonnes: 1 Volumes have been trending downwards in line with declining development. This has resulted in less face availability. Planned volumes are expected to remain relatively steady.
27
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
EXHIBIT 41: Production tonnes (quarterly)

(tonnes) 1,200 1,000 800 600 400 200 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Kloof
Driefontein
Beatrix Sibanye LoM Plan
Thirdly, we make the following observations with respect to development metres: Ore reserve development is a significant driver for production volumes and maintaining flexibility . Although development continues to fall, production volumes have been forecast to remain steady at Kloof, Driefontein and Beatrix. This could be possible a few years prior to closure of a mine where only pre-developed ore reserves are mined, although the LoM plans extend beyond 2020. EXHIBIT 42: Kloof historical development metres and tonnes
(dev metres) 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
33
Development metres
Underground tonnes (000) Sibanye Forecast
(tonnes) 1,200 1,000 800 600 400 200 0
Based on the above, we have low confidence in the current LoM plans. We do not expect grades to rise above reserve grades but we do expect them to remain at least steady. We expect volumes to continue to fall if development continues to fall, as previously discussed. In addition to falling development, tonnages have fallen as a result of a fall in the blast frequency from 1 in 2 days to 1 in 4 days. This is largely a result of two factors: 1) less time at the stope face due to longer travelling times. 2) More time being spent on support activities (such as support) instead of stoping activities. Management hopes to improve the blast frequency by extending shift hours and optimising holiday periods. This will likely be difficult to negotiate with the labour unions.
33
By flexibility we imply face availability as measured by the number of stopes/panels available to be mined at any time.
28
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 10: Historical production

Gold production from KDC and Beatrix has been reduced from just over 3.1m oz in 2002 to 1.22m oz in 2012(EXHIBIT 43). During this period there have been two rapid falls in production. The first event occurred in 2008 as a result of power cuts and seismicity issues. Production fell from 2.4m oz to 1.9m oz but did not recover thereafter. The second event occurred last year as a result of prolonged labour unrest. This was a significant stress event and we expect that some areas of the mines suffered a permanent loss of face availability. This has already been observed at Beatrix West which is now under consideration for closure. We do not expect production in 2013 to recover to previous levels. EXHIBIT 43: Historical gold production
(moz) 3.00 2.50 2.00 1.50 1.00 0.50 0.00 2005
Source: Gold Fields
2.74
2.59 2.41 power cuts seismicity 1.90
1.86 1.59 1.44 strikes 1.22
2006
2007
2008
2009
2010
2011
2012
Appendix 11: Production outlook

Management is currently reviewing the life-of-mine plans but intend on maintaining production at current levels of about 1.24m oz. We expect production to remain flat y/y in FY13 at about 1.22m oz, falling to 1.0m oz by FY15 (EXHIBIT 44). We take note of the following in our forecasts: New plans have yet to be fully developed and it is likely this will take two to three years to implement. The mining of pillars will likely result in small volumes over a long period, making no significant impact to overall production. We expect it will take management at least two years to stabilise production in line with current development rates. Production from SRDs will reduce to one third within two years. New projects within the lease are unlikely to provide any growth but rather replacement. We dont expect any projects beyond the lease in the short to medium term. By 2017, new projects (namely Vlakpan, D1 shaft and KEA) can be expected to come online and production should be sustained beyond what the current life-of-mine plan indicates. We illustrate this in the chart below.
29
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
EXHIBIT 44: Sibanye Gold Production Outlook

(moz pa) 2.5 2.0 1.5 1.0 0.5 0.0 Current LoM plan BNP case
2013E
2014E
2015E
2016E
2017E
2018E
2019E
2020E
2021E
2022E
2023E
2024E
2025E
2026E
2027E
2028E
Appendix 12: Costs and capex

In our analysis of unit cash costs, BNP Paribas Cadiz use the following terms, defined as follows: C1 = mining cash costs C2 = C1 + corporate overheads + royalties C3 = C2 + on-going capex34 C4 = C3 + net interest + social expenditure C5 = C4 + exploration + project capital EXHIBIT 45 shows C1 ZAR/tonne costs at KDCs underground operations since 2005. We compare this to a steady-state operation (Barberton Gold Mine) as a benchmark measure for inflation. We also show KDCs adjusted costs. The adjusted costs assume the same level of inflation as Barberton, normalised for tonnes 35 on a 70% fixed, 30% variable basis . We show the compounded annual growth rates on EXHIBIT 45. EXHIBIT 45: Historical C1 cash costs
(ZAR/tonne) 3,000 2,500 2,000 1,500 1,000 500 0 Sep-05 Dec-05 Mar-06 Jun-06 Sep-06 Dec-06 Mar-07 Jun-07 Sep-07 Dec-07 Mar-08 Jun-08 Sep-08 Dec-08 Mar-09 Jun-09 Sep-09 Dec-09 Mar-10 Jun-10 Sep-10 Dec-10 Mar-11 Jun-11 Sep-11 Dec-11 Mar-12 Jun-12 Sep-12 Dec-12 Mar-13
Sources: Gold Fields; Pan African Resources; BNP Paribas Cadiz Securities
Benchmark inflation CAGR = 12%
KDC (unadjusted) CAGR = 14%
KDC (normalised) CAGR = 10%
The normalised costs increased an average of only 10% pa. This compares well to the benchmark of 12% and suggests that the mines at KDC have done well to contain inflation.
34 35
On-going capital= stay-in-business capital + ore reserve development capital Although the current fixed:variable split is closer to 80:20
30
BNP PARIBAS
2029E
2007
2008
2009
2010
2011
2012
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Actual C1 costs at KDC increased 14% pa. This is largely due to falling tonnages which have halved from about 1,800kt pa to 900kt pa over the same period. Other reasons include: A high fixed-cost base which management has indicated has risen to as much as 80%. The high fixed costs are largely due to pumping of water and maintaining the complex ore transport infrastructure at KDC. Employee numbers have remained relatively static. This suggests the mines have not managed their production plans well. Redistribution of profits to Gold Fields other operations instead of re-capitalising KDC and Beatrix would have been part of the reason. Cost inflation Below we show cost inflation for the main cost components as well as its relative contribution to total costs. Our assumptions for inflation rates are based on current actual and expected rates. Overall, we expect inflation of no less than 9% pa for FY13, FY14 and FY15. EXHIBIT 46: Cost breakdown and inflation rates
inflation rate pa (%) labour + contractors consumables electricity other overall Source: BNP Paribas Cadiz Securities 10 8 8 6 9 % of costs (%) 58 22 18 2 100
Cost savings Management intends cutting fixed costs by 20% in the next 18 months. This includes corporate overheads and services (as a result of the unbundling) as well as at least 4,000 retrenchments, of which 3,000 have been achieved. Based on current cash costs of ZAR12b, this would equate to about ZAR1.8b. We estimate that the following is achievable, broken down as follows: ZAR750m for the retrenchment of 4,000 workers. ZAR50m in corporate overheads Total ZAR0.8b In addition, Sibanye management intends removing some middle management employees that they believe are no longer needed. We do not expect this will fill the shortfall of ZAR1b. At most, we expect no more than ZAR1b in cost reductions over the next 18 months. EXHIBIT 47: C1 and C3 real cash costs BNP Best Case
Combined underground and surface Kloof C1 C3 Driefontein C1 C3 Beatrix C1 C3 Combined C1 C3 Source: BNP Paribas Cadiz Securities R/kg R/kg 285,461 382,045 285,612 388,364 270,283 371,973 R/kg R/kg 314,567 413,702 327,727 413,715 328,050 413,431 R/kg R/kg 294,862 369,059 285,409 363,849 260,266 358,242 R/kg R/kg 258,761 376,464 260,841 397,558 248,663 362,891 2013E 2014E 2015E
Real cash costs are expected to remain steady as a result of the cost savings as well as the mining of pillars in 2015.
31
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Global cash costs Below we show where Sibanye is positioned on the global C1 and C3 cash cost curves, relative to its peer group. Sibanyes current C1 cash costs of USD1,073/oz positions it in the upper quartile of the cost curve alongside Harmony, and less favourably compared to Gold Fields and AngloGold Ashanti. However, C3 cash costs are a lot more similar between the peer group, at between USD1,300/oz and USD1,400/oz. EXHIBIT 48: Global C1 cash cost curve (current costs)
(USD/oz) 2,200 2,000 1,800 1,600 1,400 1,200 1,000 800 600 400 200 0 0 10 20 30 40 oz Au per annum 50 60 70 (m) Harmony Sibanye AngloGold Ashanti Gold Fields
Sources: Brook Hunt; BNP Paribas Cadiz Securities
EXHIBIT 49: Global C3 cash cost curve (current costs)

(USD/oz) 2,600 2,400 2,200 2,000 1,800 1,600 1,400 1,200 1,000 800 600 400 200 0 0 10 20 30 40 oz Au per annum 50 60 70 (m) Harmony Sibanye AngloGold Ashanti Gold Fields
Sources: Brook Hunt; BNP Paribas Cadiz Securities
32
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Capex Below we show historical and forecast capex. Capex this year has been budgeted at ZAR3.2b. Based on the current life-of-mine plan, capex is expected to fall to about ZAR1b by 2018 in line with falling development. We expect current SIB capex numbers to remain steady in-line with maintaining development rates at current levels. We factor in ZAR1b project capital for new projects from 2015 to 2017. We also include the expected capital outlay for a potential new tailings project, equivalent to ZAR5b over three years. We do not factor this into our models. EXHIBIT 50: Real Capex Outlook
(ZAR b) 6.0 5.0 Tailings project? 4.0 3.0 2.0 1.0 0.0 2010 2011 2012 2013E 2014E 2015E 2016E 2017E 2018E Current LoM Plan BNP Case Tailings project
If management intends to keep production steady, capex will at least need to remain at about ZAR3b pa. Capex will likely grow even more to bring new projects into production.
33
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Appendix 13: Valuation

We adopt a real DCF method to estimate a 12-month target price. This valuation technique lends itself well to gold companies in South Africa due to it being well established amongst the investment community as well as the transparency from company management. Model assumptions and inputs We calculate the NAV for Sibanye Gold using the following key assumptions: Spot gold of USD1,380/oz. Exchange rate of ZAR9.4/USD. Actual net debt of ZAR2.8b, as at April 2013. Real discount rate of 10%. Life of mine based on stated reserves, as well as the potential additional reserves of 3.7m oz. Additional project capital of ZAR1b real (based on Pre-feas estimates for the KEA project). A starting period of FY13. We do not include the valuation for the tailings project since the results are NPV negative based on current assumptions and prices. Management has indicated that all the Environmental Rehabilitation Trusts are fully funded. In our assessment of life of mine, future expected grades, tonnages, expected ramp-up time and capital cost, we have taken cognisance of management guidance and conversations with the CFO and MRM manager. Other cost assumptions: Below we show the royalty rates and taxes which we incorporate into our DCF valuation. Royalties and taxes for South Africa are formula based: Royalty rate= 0.5 + EBIT/(Revenue x 12.5)% Tax rate = 34 -170/x, where x = taxable income/revenue Results We value Sibanye Gold at ZAR7.00/share broken down as follows: Driefontein: ZAR6.10/share Kloof: ZAR4.50/share Beatrix: ZAR0.3/share Net debt: ZAR3.8/share Our valuation based on the current LoM plan is ZAR11/share.
34
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
EXHIBIT 51: Valuation of Sibanye Gold for BNP Case and based on current life of mine plan
(ZAR/share) 12 10 8 6 4 6.1 2 0 Driefontein Kloof Beatrix Net debt BNP Case Current LoM Case 4.5 0.3 3.8 7.0 11.0
The majority of our valuation lies with Driefontein (56.3%), followed by Kloof (43.4%) while Beatrix holds no value at all (0.4%). EXHIBIT 52: NPV of Kloof, Driefontein and Beatrix
Beatrix 2.9%
Kloof 41.2%
Driefontein 55.9%
Below we show our progressive NPV for each case. EXHIBIT 53: A forward view of Sibanye's NPV
(ZAR m) 12,000 10,000 8,000 6,000 4,000 2,000 0 (2,000) 2013F 2014F 2015F 2016F 2017F 2018F 2019F 2020F 2021F
Source: BNP Paribas Cadiz Securities estimates
BNP CASE
Current LoM
35
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Financial statements
Sibanye Gold Ltd
Profit and Loss (ZAR m) Year Ending Dec Revenue Cost of sales ex depreciation Gross profit ex depreciation Other operating income Operating costs Operating EBITDA Depreciation Goodwill amortisation Operating EBIT Net financing costs Associates Recurring non operating income Non recurring items Profit before tax Tax Profit after tax Minority interests Preferred dividends Other items Reported net profit Non recurring items & goodwill (net) Recurring net profit Per share (ZAR) Recurring EPS * Reported EPS DPS Growth Revenue (%) Operating EBITDA (%) Operating EBIT (%) Recurring EPS (%) Reported EPS (%) Operating performance Gross margin inc depreciation (%) Operating EBITDA margin (%) Operating EBIT margin (%) Net margin (%) Effective tax rate (%) Dividend payout on recurring profit (%) Interest cover (x) Inventory days Debtor days Creditor days Operating ROIC (%) Operating ROIC - WACC (%) ROIC (%) ROIC - WACC (%) ROE (%) ROA (%)
*Pre exceptional, pre-goodwill and fully diluted
2011A -
2012A 16,554 (10,874) 5,680 0 (403) 5,277 (2,363) 0 2,914 (21) 93 0 (373) 2,613 366 2,979 0 0 0 2,979 372 3,350
2013E 16,623 (11,837) 4,787 0 (966) 3,821 (2,291) 0 1,530 (359) 0 0 0 1,171 (330) 842 0 0 0 842 0 842
2014E 14,813 (12,086) 2,727 0 (322) 2,405 (2,333) 0 73 (393) 0 0 0 (320) 0 (320) 0 0 0 (320) 0 (320)
2015E 15,074 (12,681) 2,393 0 (339) 2,054 (2,371) 0 (318) (478) 0 0 0 (795) 0 (795) 0 0 0 (795) 0 (795)
4.59 4.08 0.09
1.15 1.15 0.29
(0.44) (0.44) 0.00
(1.09) (1.09) 0.00
0.4 (27.6) (47.5) (75.0) (71.8)
(10.9) (37.1) (95.3) (138.0) (138.0)
1.8 (14.6) (538.1) 148.5 148.5
20.0 31.9 17.6 20.2 (14.0) 2.0 140.5 -
15.0 23.0 9.2 5.1 28.1 25.0 4.3 0.0 12.6 0.0 4.7 4.4 6.3 5.0
2.7 16.2 0.5 (2.2) 0.2 0.0 14.9 0.0 0.2 0.2 (2.0) (0.3)
0.1 13.6 (2.1) (5.3) (0.7) 0.0 15.1 0.0 (0.7) (0.7) (4.6) (1.8)
Sources: Sibanye Gold; BNP Paribas estimates
36
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd Sibanye Gold Ltd

Cash Flow (ZAR m) Year Ending Dec Recurring net profit Depreciation Associates & minorities Other non-cash items Recurring cash flow Change in working capital Capex - maintenance Capex - new investment Free cash flow to equity Net acquisitions & disposals Dividends paid Non recurring cash flows Net cash flow Equity finance Debt finance Movement in cash Per share (ZAR) Recurring cash flow per share FCF to equity per share Balance Sheet (ZAR m) Year Ending Dec Working capital assets Working capital liabilities Net working capital Tangible fixed assets Operating invested capital Goodwill Other intangible assets Investments Other assets Invested capital Cash & equivalents Short term debt Long term debt * Net debt Deferred tax Other liabilities Total equity Minority interests Invested capital 2011A 7.83 3.63 2012A 1,455 2,063 3,518 16,376 19,894 0 0 0 1,354 21,248 (512) 2,220 2,000 3,708 4,186 0 11,598 0 21,248 4.28 (0.11) 2013E 1,504 2,063 3,566 19,640 23,207 0 0 0 1,354 24,561 (565) 2,000 2,000 3,435 4,186 0 15,184 0 24,561 2.75 (1.92) 2014E 1,516 2,063 3,579 23,075 26,654 0 0 0 1,354 28,008 761 2,000 3,000 5,761 4,186 0 16,305 0 28,008 2.15 (2.69) 2015E 1,546 2,063 3,609 26,652 30,261 0 0 0 1,354 31,616 1,256 2,000 4,000 7,256 4,186 0 18,418 0 31,616 2011A 2012A 3,350 2,363 0 0 5,713 21 (3,083) 0 2,651 0 (731) 0 1,920 0 (1,991) (71) 2013E 842 2,291 0 0 3,133 49 (3,264) 0 (83) 0 (210) 0 (293) 0 346 53 2014E (320) 2,333 0 0 2,013 13 (3,435) 0 (1,409) 0 0 0 (1,409) 0 82 (1,327) 2015E (795) 2,371 0 0 1,576 30 (3,578) 0 (1,971) 0 0 0 (1,971) 0 1,477 (494)
Adrian Hammond
* includes convertibles and preferred stock which is being treated as debt
Per share (ZAR) Book value per share Tangible book value per share Financial strength Net debt/equity (%) Net debt/total assets (%) Current ratio (x) CF interest cover (x) Valuation Recurring P/E (x) * Recurring P/E @ target price (x) * Reported P/E (x) Dividend yield (%) P/CF (x) P/FCF (x) Price/book (x) Price/tangible book (x) EV/EBITDA (x) ** EV/EBITDA @ target price (x) ** EV/invested capital (x)
* Pre exceptional, pre-goodwill and fully diluted
15.89 15.89
20.72 20.72
22.25 22.25
25.13 25.13
2011A neg -
32.0 18.8 12.5 124.9 2012A 1.6 1.5 1.8 1.2 1.0 2.1 0.5 0.5 0.4
22.6 14.9 (33.1) 0.8 2013E 6.5 6.1 6.5 3.9 1.7 (66.0) 0.4 0.4 2.4 1.2 0.4
35.3 22.9 (12.1) (2.6) 2014E neg neg neg 0.0 2.7 (3.9) 0.3 0.3 4.2 1.8 0.4
39.4 25.6 (4.6) (3.1) 2015E neg neg neg 0.0 3.5 (2.8) 0.3 0.3 5.8 2.7 0.4
** EBITDA includes associate income and recurring non-operating income
Sources: Sibanye Gold; BNP Paribas estimates
37
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
Disclaimers and Disclosures

ANALYST(S) CERTIFICATION
Adrian Hammond, BNP Paribas Cadiz Securities (Pty) Ltd, +27 11 088 2181, adrian.hammond@bnpparibascadiz.com. The analyst(s) or strategist(s) herein each referred to as analyst(s) named in this report certifies that (i) all views expressed in this report accurately reflect the personal view of the analyst(s) with regard to any and all of the subject securities, companies or issuers mentioned in this report; (ii) no part of the compensation of the analyst(s) was, is, or will be, directly or indirectly, relate to the specific recommendation or views expressed herein; and (iii) is not aware of any other actual or material conflicts of interest concerning any of the subject securities, companies or issuers referenced herein as of the time of this certification. Analysts mentioned in this disclaimer are employed by a non-US affiliate of BNP Paribas Securities Corp., and are not registered/ qualified pursuant to NYSE and/ or FINRA regulations.
GENERAL DISCLAIMER
This report was produced by BNP Paribas Cadiz Securities (Pty) Ltd, a member company of the BNP Paribas Group. "BNP Paribas is the marketing name for the global banking and markets business of BNP Paribas Group1. This report is for the use of intended recipients only and may not be reproduced (in whole or in part) or delivered or transmitted to any other person without our prior written consent. By accepting this report, the recipient agrees to be bound by the terms and limitations set forth herein. BNP Paribas analysts prudently perform analysis and create quantitative models and estimates derived from their own review of publicly available data without any assistance from any represented company. BNP Paribas analyst estimates and models reflect the analysts current judgment only; they are neither all-inclusive nor can they be guaranteed. The analysts analysis and models are subject to change based on various other factors. Valuations are based on internal quantitative models and qualitative interpretation. No representation or warranty, express or implied, is made that such information or analysis is accurate, complete or verified and it should not be relied upon as such. Analysts' compensation is not linked to investment banking or capital markets transactions performed by BNP Paribas or the profitability or revenues of particular trading desks. BNP Paribas analysts may participate in company events such as site visits and are prohibited from accepting payment by the company of associated expenses unless pre-approved by authorized members of Research management. This report does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of individual clients. Customers are advised to use the information contained herein as just one of many inputs and considerations prior to engaging in any trading activity. This report does not constitute a prospectus or other offering document or an offer or solicitation to buy or sell any securities or other investments. This report is not intended to provide the sole basis of any evaluation of the subject securities and companies mentioned in this report. Information and opinions contained in this report are published for reference of the recipients and are not to be relied upon as authoritative or without the recipients own independent verification, or taken in substitution for the exercise of judgment by the recipient. Additionally, the products mentioned in this report may not be available for sale in certain jurisdictions. BNP Paribas is not aware of any other actual or material conflicts of interest concerning any of the subject securities, companies or issuers referenced herein as of the time of publication of the research report. As an investment bank with a wide range of activities, BNPP may face conflicts of interest, which are resolved under applicable legal provisions and internal guidelines. You should be aware, however, that BNPP may engage in transactions in a manner inconsistent with the views expressed in this document, either for its own account or for the account of its clients. Australia: This report is being distributed in Australia by BNP Paribas Sydney Branch, registered in Australia as ABN 23 000 000 117 at 60 Castlereagh Street Sydney NSW 2000. BNP Paribas Sydney Branch is licensed under the Banking Act 1959 and the holder of Australian Financial Services Licence no. 238043 and therefore subject to regulation by the Australian Securities & Investments Commission in relation to delivery of financial services. By accepting this document you agree to be bound by the foregoing limitations, and acknowledge that information and opinions in this document relate to financial products or financial services which are delivered solely to wholesale clients (in terms of the Corporations Act 2001, sections 761G and 761GA; Corporations Regulations 2001, division 2, reg. 7.1.18 & 7.1.19) and/or professional investors (as defined in section 9 of the Corporations Act 2001). Canada: The information contained herein is not, and under no circumstances is to be construed as, a prospectus, an advertisement, a public offering, an offer to sell securities described herein, or solicitation of an offer to buy securities described herein, in Canada or any province or territory thereof. Any offer or sale of the securities described herein in Canada will be made only under an exemption from the requirements to file a prospectus with the relevant Canadian securities regulators and only by a dealer properly registered under applicable securities laws or, alternatively, pursuant to an exemption from the dealer registration requirement in the relevant province or territory of Canada in which such offer or sale is made. The information contained herein is under no circumstances to be construed as investment advice in any province or territory of Canada and is not tailored to the needs of the recipient. To the extent that the information contained herein references securities of an issuer incorporated, formed or created under the laws of Canada or a province or territory of Canada, any trades in such securities must be conducted through a dealer registered in Canada. No securities commission or similar regulatory authority in Canada has reviewed or in any way passed judgment upon these materials, the information contained herein or the merits of the securities described herein, and any representation to the contrary is an offence. Hong Kong: This report is prepared for professional investors and is being distributed in Hong Kong by BNP Paribas Securities (Asia) Limited to persons whose business involves the acquisition, disposal or holding of securities, whether as principal or agent. BNP Paribas Securities (Asia) Limited, a subsidiary of BNP Paribas, is regulated by the Securities and Futures Commission for the conduct of dealing in securities, advising on securities and providing automated trading services. For professional investors in Hong Kong, please contact BNP Paribas Securities (Asia) Limited for all matters and queries relating to this report. INDIA: In India, this document is being distributed by BNP Paribas Securities India Pvt. Ltd. ("BNPPSIPL"), having its registered office at 5th floor, BNP Paribas House, 1 North Avenue, Maker Maxity, Bandra Kurla Complex, Bandra (East), Mumbai 400 051 (Tel. no. +91 22 3370 4000 / 6196 4000). BNPPSIPL is registered with the Securities and Exchange Board of India (SEBI) as a stockbroker in the Equities and the Futures & Options segments of National Stock Exchange of India Ltd. and Bombay Stock Exchange Ltd. (SEBI regn. nos. INB/INF231474835, INB/INF011474831). Indonesia: This report is distributed by PT BNP Paribas Securities Indonesia, having its registered office at Menara BCA, 35th Floor, Grand Indonesia, Jl. M.H.Thamrin No.1, Jakarta, 10310, Indonesia. PT BNP Paribas Securities Indonesia is a fully subsidiaries company of BNP Paribas SA and is licensed under Capital Market Law No. 8 of 1995 and the holder of broker-dealer and underwriter licenses issued by the Capital Market Supervisory Agency. PT BNP Paribas Securities Indonesia is also a member of Indonesia Stock Exchange. Japan: This report is being distributed to Japanese based firms by BNP Paribas Securities (Japan) Limited or by a subsidiary or affiliate of BNP Paribas not registered as a financial instruments firm in Japan, to certain financial institutions defined by article 17-3, item 1 of the Financial Instruments and Exchange Law Enforcement Order. BNP Paribas Securities (Japan) Limited is a financial instruments firm registered according to the Financial Instruments and Exchange Law of Japan and a member of the Japan Securities Dealers Association and the Financial Futures Association of Japan. BNP Paribas Securities (Japan) Limited accepts responsibility for the content of a report prepared by another non-Japan affiliate only when distributed to Japanese based firms by BNP Paribas Securities (Japan) Limited. Some of the foreign securities stated on this report are not disclosed according to the Financial Instruments and Exchange Law of Japan. Malaysia: This report is issued and distributed by BNP Paribas Capital (Malaysia) Sdn Bhd. The views and opinions in this research report are our own as of the date hereof and are subject to change. BNP Paribas Capital (Malaysia) Sdn Bhd has no obligation to update its opinion or the information in this research report. This publication is strictly confidential and is for private circulation only to clients of BNP Paribas Capital (Malaysia) Sdn Bhd. This publication is being provided to you strictly on the basis that it will remain confidential. No part of this material may be (i) copied, photocopied, duplicated, stored or reproduced in any form by any means or (ii) redistributed or passed on, directly or indirectly, to any other person in whole or in part, for any purpose
38
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd
Adrian Hammond
without the prior written consent of BNP Paribas Capital (Malaysia) Sdn Bhd. Singapore: This report is distributed in Singapore by BNP Paribas Securities (Singapore) Ptd Ltd ("BNPPSSL") and may be distributed in Singapore only to an Accredited or Institutional Investor, each as defined under the Financial Advisers Regulations ("FAR") and the Securities and Futures Act (Chapter 289) of Singapore, as amended from time to time. In relation to the distribution to such categories of investors, BNPPSSL and its representatives are exempted under Regulation 35 of the FAR from the requirements in Section 36 of the Financial Advisers Act of Singapore, regarding the disclosure of certain interests in, or certain interests in the acquisition or disposal of, securities referred to in this report. For Institutional and Accredited Investors in Singapore, please contact BNP Paribas Securities (Singapore) Ptd Ltd for all matters and queries relating to this report. South Africa: In South Africa, BNP Paribas Cadiz Securities (Pty) Ltd and BNP Paribas Cadiz Stock Broking (Pty) Ltd are licensed members of Johannesburg Stock Exchange and are authorised Financial Services Providers and subject to regulation by the Financial Services Board. Taiwan: Information on securities that trade in Taiwan is distributed by BNP Paribas Securities (Taiwan) Co., Ltd. Such information is for your reference only. The reader should independently evaluate the investment risks and is solely responsible for their investment decision. Information on securities that do not trade in Taiwan is for informational purposes only and is not to be construed as a recommendation or a solicitation to trade in such securities. BNP Paribas Securities (Taiwan) Co., Ltd. may not execute transactions for clients in these securities. This publication may not be distributed to the public media or quoted or used by the public media without the express written consent of BNP Paribas. Thailand: Research relating to Thailand and Thailand based issuers are produced pursuant to an arrangement between BNP Paribas Securities (Asia) Ltd and ACL Securities Co. Ltd. ACL Securities Co. Ltd. is otherwise unaffiliated with BNP Paribas. This report is being distributed outside Thailand by members of BNP Paribas. Turkey: This report is being distributed in Turkey by TEB Investment a member company of the BNP Paribas Group. Notice Published in accordance with Communiqu Regarding the Principles on Investment Consultancy Activities and the Investment Consultancy Institutions Series: V, No: 55 issued by the Capital Markets Board. The investment related information, commentary and recommendations contained herein do not constitute investment consultancy services. Investment consultancy services are provided in accordance with investment consultancy agreements executed between investors and brokerage companies or portfolio management companies or non-deposit accepting banks. The commentary and recommendations contained herein are based on the personal views of the persons who have made such commentary and recommendations. These views may not conform to your financial standing or to your risk and return preferences. Therefore, investment decisions based solely on the information provided herein may fail to produce results in accordance with your expectations. United States: This report may be distributed in the United States only to U.S. Persons who are major U.S. institutional investors (as such term is defined in Rule 15a-6 under the Securities Exchange Act of 1934, as amended) and is not intended for the use of any person or entity that is not a major U.S. institutional investor. U.S persons who wish to effect transactions in securities discussed herein must do so through BNP Paribas Securities Corp., a USregistered broker dealer and member of FINRA, SIPC, NFA, NYSE and other principal exchanges. Certain countries within the European Economic Area: This document may only be distributed in the United Kingdom to eligible counterparties and professional clients and is not intended for, and should not be circulated to, retail clients (as such terms are defined in the Markets in Financial Instruments Directive 2004/39/EC (MiFID)). This document will have been approved for publication and distribution in the United Kingdom by BNP Paribas London Branch, a branch of BNP Paribas SA whose head office is in Paris, France. BNP Paribas London Branch is registered in England and Wales under No. FC13447. Registered Office: 10 Harewood Avenue, London NW1 6AA. BNP Paribas SA is incorporated in France with limited liability with its registered office at 16 boulevard des Italiens, 75009 Paris. BNP Paribas London Branch is authorised and supervised by the Autorit de Contrle Prudentiel and is authorised and subject to limited regulation by the Financial Services Authority ("FSA") for the conduct of its designated investment business in the United Kingdom. Details of the extent of its authorisation and regulation by the Financial Services Authority are available upon request. This report has been approved for publication in France by BNP Paribas, a credit institution licensed as an investment services provider by the Autorit de Contrle Prudentiel whose head office is 16, Boulevard des Italiens 75009 Paris, France. This report is being distributed in Germany either by BNP Paribas London Branch or by BNP Paribas Niederlassung Frankfurt am Main, regulated by the Bundesanstalt fr Finanzdienstleistungsaufsicht (BaFin). Other Jurisdictions: The distribution of this report in other jurisdictions or to residents of other jurisdictions may also be restricted by law, and persons into whose possession this report comes should inform themselves about, and observe, any such restrictions. By accepting this report you agree to be bound by the foregoing instructions. This report is not directed to, or intended for distribution to or use by, any person or entity that is a citizen or resident of or located in any locality, state, country, or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation. All research reports are disseminated and available to all clients simultaneously through our internal client websites. For all research available on a particular stock, please contact the relevant BNP Paribas research team or the author(s) of this report.
1
No portion of this report was prepared by BNP Paribas Securities Corp (US) personnel and it is considered Third-Party Affiliate research under NASD Rule 2711.
IMPORTANT DISCLOSURES
The disclosure column in the following table lists the important disclosures applicable to each company that has been rated and/or recommended in this report:
Company Disclosure (as applicable) -
BNP Paribas represents that: 1. Within the past year, it has managed or co-managed a public offering for this company, for which it received fees. 2. It had an investment banking relationship with this company in the last 12 months. 3. It received compensation for investment banking services from this company in the last 12 months. 4. It beneficially owns 1% or more or the market capitalization of this company. 5. It makes a market in securities in respect of this company. 6. The analyst(s) or an individual who assisted in the preparation of this report (or a member of his/her household) has a financial interest position in securities issued by this company or derivatives thereof. 7. The analyst (or a member of his/her household) is an officer, director, or advisory board member of this company. Additional Disclosures Within the next three months, BNP Paribas may receive or seek compensation in connection with an investment banking relationship with one or more of the companies referenced herein. Target price history, stock price charts, valuation and risk details, and equity rating histories applicable to each company rated in this report is available in our most recently published reports available on our website: http://eqresearch.bnpparibas.com, or you can contact the analyst named on the front of this note or your BNP Paribas representative. All share prices are as at market close on 28 May 2013 unless otherwise stated.
39
BNP PARIBAS
29 MAY 2013
Sibanye Gold Ltd RECOMMENDATION STRUCTURE

Stock Ratings Stock ratings are based on absolute upside or downside, which we define as (target price* - current price) / current price. BUY (B). The upside is 10% or more. HOLD (H). The upside or downside is less than 10%. REDUCE (R). The downside is 10% or more.
Adrian Hammond
Unless otherwise specified, these recommendations are set with a 12-month horizon. Thus, it is possible that future price volatility may cause a temporary mismatch between upside/downside for a stock based on market price and the formal recommendation.
* In most cases, the target price will equal the analyst's assessment of the current fair value of the stock. However, if the analyst doesn't think the market will reassess the stock over the specified time horizon due to a lack of events or catalysts, then the target price may differ from fair value. In most cases, therefore, our recommendation is an assessment of the mismatch between current market price and our assessment of current fair value.
Industry Recommendations Improving ( ): The analyst expects the fundamental conditions of the sector to be positive over the next 12 months. Neutral ( ): The analyst expects the fundamental conditions of the sector to be maintained over the next 12 months. Deteriorating ( ): The analyst expects the fundamental conditions of the sector to be negative over the next 12 months. Country (Strategy) Recommendations Overweight (O). Over the next 12 months, the analyst expects the market to score positively on two or more of the criteria used to determine market recommendations: index returns relative to the regional benchmark, index sharpe ratio relative to the regional benchmark and index returns relative to the market cost of equity. Neutral (N). Over the next 12 months, the analyst expects the market to score positively on one of the criteria used to determine market recommendations: index returns relative to the regional benchmark, index sharpe ratio relative to the regional benchmark and index returns relative to the market cost of equity. Underweight (U). Over the next 12 months, the analyst does not expect the market to score positively on any of the criteria used to determine market recommendations: index returns relative to the regional benchmark, index sharpe ratio relative to the regional benchmark and index returns relative to the market cost of equity.
RATING DISTRIBUTION (as at 28 May 2013)

Total BNP Paribas coverage universe Buy Hold Reduce 654 346 208 100 Investment Banking Relationship Buy Hold Reduce (%) 8.1 2.4 4.0
Should you require additional information concerning this report please contact the relevant BNP Paribas research team or the author(s) of this report. 2013 BNP Paribas Group
40
BNP PARIBAS
29 MAY 2013

Sibanye Gold - BNP Cadiz Initiating Report

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Sibanye Gold - BNP Cadiz Initiating Report

Încărcat de

Drepturi de autor:

Formate disponibile

INITIATION

EQUITIES RESEARCH SGL SJ

HOW WE DIFFER FROM THE STREET

SIBANYE GOLD LTD

KEY STOCK DATA

PLANS SIBANYE CANT AFFORD

1 Month 3 Month 12 Month (21.3) (24.2) (42.7) (46.8) -

3m avg daily turnover (USD m)

RUNNING OUT OF OPTIONS

Sources: Bloomberg consensus; BNP Paribas Cadiz Securities estimates

Sibanye Gold Ltd

Appendix 3: Reserves ________________________________________________________________________ 14

Appendix 4: Life-of-mine and amortisation _____________________________________________________ 16

Sibanye Gold Ltd

Appendix 5: What is left in the lease? __________________________________________________________ 18

Appendix 13: Valuation ______________________________________________________________________ 34

P&L, Balance Sheet and Cash Flow ____________________________________________________________ 36

To find out more about BNP Paribas Equities Research:

Sibanye Gold Ltd

Sibanye Gold Ltd

Sibanye Gold Ltd

Is pillar mining a game changer?

Can declining production be stopped?

Current LoM plan

Source: BNP Paribas Cadiz estimates

EXHIBIT 3: Potential reserves

Sibanye Gold Ltd

Surface rock dumps (SRD)

EXHIBIT 5: Real free cash from SRDs

Source: BNP Paribas Cadiz estimates

Source: BNP Paribas Cadiz estimates

Costs aligned with industry

Sibanye Gold Ltd

Sibanye Gold Ltd

Source: BNP Paribas Cadiz Securities

Sibanye Gold Ltd

Sibanye Gold Ltd

Appendix 1: Company background

Source: Sibanye Gold

First Eagle Van Eck

Source: BNP Paribas Cadiz Securities; Sibanye Gold

Source: Sibanye Gold

Sibanye Gold Ltd

Source: BNP Paribas Cadiz Securities

Appendix 2: Geology and mining operations

Sibanye Gold Ltd

EXHIBIT 16: Pay-limit at Beatrix North, South and West sections

Reserve pay limit

Sibanye Gold Ltd

EXHIBIT 18: Historical reserve grade per mine

6 4.30 4 2 0 Kloof Driefontein Beatrix

Sources: Sibanye Gold; reserves for underground operations only

Removal of reserves below infrastructure

Exclusion of pillars due to safety

Sources: Gold Fields; Sibanye Gold

Sibanye Gold Ltd

EXHIBIT 21: Pay-limits in 2009 and 2012

1,000 800 600

400 200 0 Kloof

Sibanye Gold Ltd

Appendix 4: Life-of-mine and amortisation

Sibanye Gold Ltd

Sources: Gold Fields; Sibanye Gold; BNP Paribas Cadiz Securities

Sibanye Gold Ltd