Overview
Relevant assessment and reporting criteria

The following key assumptions and parameters, unless otherwise stated, were used in the compilation of the estimates in this declaration:

• Implats developed a Group-wide protocol for the estimation, classification and reporting of mineral resources and mineral reserves in 2010 to enhance standardisation and to facilitate consistency in auditing. This protocol is updated annually with the aim to improve and specifically guide the classification of mineral resources and to ensure compliance with the SAMREC Code
• Implats introduced a depth cut-off in 2010 whereby mineralisation below a certain depth is excluded from the mineral resource estimate. This depth cut-off is applicable to the Bushveld Complex setting and is reviewed annually considering a range of assumptions, specifically the virgin rock temperature (VRT), cooling requirements, available technology, support design and other cost, prices and mining depth limits presently in the platinum industry. It is recognised that while the actual depth cut-off could vary from area to area and over time as conditions vary. The depth cut-off of 2 350m was applied from the 2013 Implats mineral resource estimates and equated approximately to a VRT of 73° C. A depth cut-off of 2 000m below surface was introduced in 2014, additional to the depth cut-off areas various mineral resource blocks are considered on a case-by-case basis and this has resulted in areas where the eventual economic extraction is in doubt. These mineral resources will be reported as exploration results and are excluded from the summation of total mineral resources per area and the attributable mineral resources
• Mineral resource tonnage and grades are estimated in situ. The mineral resources for the Merensky Reef are estimated at a minimum mining width, and may therefore include mineralisation below the selected cut-off grade. Mineral resource estimates for the UG2 Reef reflect the main UG2 chromitite layer widths only and do not include any dilution. Implats prefers to estimate the UG2 chromitite layer separately from the low-grade or barren hangingwall and footwall units as this approach supports improved grade control and ore accounting practices. This practice to report the UG2 chromitite layer as the mineral resource estimate and disclosing the actual estimated layer width is most transparent and compliant with the SAMREC Code
• Note that the main UG2 chromitite layer widths in the case of Impala and Marula are narrower than a practical minimum mining width. For further clarity a comparative summary is listed in these sections where the standard estimates are compared with estimates that include dilution up to a minimum mining width
• Mineral resource estimates for the Main Sulphide Zone are based on optimal mining widths. Such mining widths are reviewed from time to time given varying economic and operational considerations
• Mineral resource estimates are reported inclusive of mineral reserves, unless otherwise stated
• Mineral resource estimates allow for estimated geological losses but not for anticipated pillar losses during eventual mining, except where these pillars will never be extracted, such as legal, boundary and shaft pillars
• Mineral reserve estimates include allowances for mining dilution and are reported as tonnage and grade delivered to the mill
• Rounding-off of figures in the accompanying summary estimates may result in minor computational discrepancies. Where this occurs it is not deemed significant
• It is important to note that the mineral resource statements in principle remain imprecise estimates and cannot be referred to as calculations. All inferred mineral resources should be read as “approximations”
• Exploration samples are mainly assayed for all PGEs and Au, using the nickel sulphide fire assay collection method and determining the elements with an inductively coupled plasma mass spectrometer (ICP-MS). Base metal content is determined by an atomic absorption (AA) spectrometer using partial digestion in order to state metal in sulphide that is amenable to recovery by flotation processes. All these analyses are undertaken by Intertek Genalysis in Perth
• Underground samples are mainly assayed for Pt, Pd, Rh and Au using the lead collection method by the in-house laboratories at the respective mines. A partial digestion at the in-house laboratories is used to determine the base metal content of samples using AA
• All references to tonnage are to the metric unit
• All references to ounces (oz) are troy with the factor used being 31.10348 metric grams per ounce
• The mineral resources and mineral reserves reported for the individual operations and projects are reflected as the total estimate (100%). The corresponding estimates relating to attributable mineral resources and mineral reserves are only given as combined summary tabulations
• Mineral reserves are that portion of the mineral resource which technical and economic studies have demonstrated can justify extraction at the time of disclosure. Historically, Implats has only converted mineral resources to mineral reserves on completion of a full feasibility study for a project and LoM I for an operating mine (as per SAMREC). The exception to this in the past has been at Zimplats where the basis of a pre-feasibility study was applied, as permitted by the JORC Code 2012. This practice is in line with the SAMREC 2009 clarification that only a pre-feasibility study is required for such conversions. The conversion of mineral resources to ore reserves for Zimplats has now been aligned to the Implats standard
• No inferred mineral resources have been converted into mineral reserves
• There are only limited changes in the estimation principles and reporting style as at 30 June 2015 relative to the previous report. The key change is:
  • The mineral resources previously reported as “mineral resources under review” and “areas excluded from mineral resources” will now be reported separately from the individual operations where applicable
  • additional details on nickel and copper are disclosed in the tables
  • a summary section on chromium is included
• The term ore reserve is interchangeable with the term mineral reserve
• Implats uses a discounted cash flow model that embodies economic, financial and production estimates in the valuation of mineral assets. Forecasts of key inputs are:

  Relative rates of inflation in South Africa and the United States Rand/dollar exchange rate Metal prices

  Capital expenditure Operating expenditure Production profile Metal recoveries

• The outputs are net present value, the internal rate of return, annual free cash flow, project payback period and funding requirements. Metal price and exchange rate forecasts are regularly updated by the marketing department of Implats. As at 30 June 2015, a real long-term forecast for revenue per platinum ounce sold of R31 246 was used. Specific real long-term forecasts include:

  Platinum US$1 568/oz Palladium US$1 066/oz Rhodium US$1 540/oz Ruthenium US$59/oz Iridium US$571/oz

  Gold US$1 237/oz Nickel US$19 851/t Copper US$6 940/t Exchange rate R11.30/US$

• The spot basket price calculated for Implats as at 30 June 2015 was R19 981 and the equivalent real long-term consensus basket price is R23 889 per ounce
• Rigorous profitability tests are conducted to test the viability of the mineral reserves, references to this are listed in the sections per operation and highlight the spot price scenarios. A summary graph showing the price sensitivity of the total Group mineral reserves is depicted below.

A mineral resources by definition is “a concentration or occurrence of material of economic interest in or on the earth’s crust in such form, quality and quantity that there are reasonable and realistic prospects for eventual economic extraction”. The interpretation of such “eventual economics” varies significantly, however, it implies some form of high-level view in terms of either “yard-stick comparisons” or high-level scenario models. On this basis Implats has excluded significant mineralisation (a) initially below 2 350m below surface, (b) then 2 000m below surface, (c) selected areas based on geology and potential infrastructure (see section “Areas excluded from mineral resource estimates” in this document). In total some 58Moz Pt has been excluded from current statements on this basis. However, under a present price regime and outlook the bulk of the Implats’ South African mineral resources are marginal at best and require long-term metal prices higher than current estimates. Work is underway to identify opportunities on a scenario scale to optimise such areas in terms of potential output, production costs and future capital expenses. Notably the Zimbabwean mineral resources are reasonably robust in terms of “eventual economic extraction” and require long-term prices in the order of US$1 500 per Pt oz and the deeper Rustenburg mineral resources require prices around US$2 000 per Pt oz.

Environment

Our activities associated with the exploration, extraction and processing of mineral resources result in the unavoidable disturbance of land, the consumption of resources, and the generation of waste and atmospheric and water pollutants. Growing regulatory and social pressure, increasing demands for limited natural resources, and the changing costs of energy and water all highlight the business imperative of responsible environmental management, particularly as our underground operations become deeper and consume greater amounts of energy and water. This involves taking measures not only to address security of resource supply (for example through efficiency, recycling and fuel-switching), but also to actively minimise our impacts on natural resources and on the communities around our operations. Taking such measures has direct benefits in terms of reduced costs and liabilities, enhanced resource security and improved security of our licence to operate.

Implats has an environmental policy that commits the company to running our exploration, mining, processing and refining operations in an environmentally responsible manner and to ensure the well-being of our stakeholders. The policy also commits to integrating environmental management into all aspects of the business with the aim of achieving world-class environmental performance in a sustainable manner.

Our management of the environmental impacts of our operations and processes involves the following focus areas:

• Promoting responsible water stewardship by minimising water use and water pollution
• Minimising our negative impacts on air quality
• Responding to climate change risks and opportunities and promoting responsible energy management
• Managing our waste streams
• Promoting responsible land management and biodiversity practices

All our operations are ISO 14001 certified. In line with our environmental management system expectations, all operations are required to identify and report on environmental incidents. Systems are in place to investigate and determine the direct and root causes of high-severity incidents, and to address and close out these incidents.

To ensure continued assurance of legal compliance to all authorisation requirements and conditions, legal, geology and environmental departments are implementing the best practice land management software FlexiCadastre®, for the management of mineral rights and contractual commitments.

Further details regarding the materiality of environmental aspects, management processes, performance and commitments are reported in the 2015 Implats sustainability report and also summarised in the 2015 Implats integrated report (refer in particular to the notes in the annual financial statements). These reports will be published at www.implats.co.za at the end of September 2015. The financial provisions for the rehabilitation can be summarised as follows:

	Name	Current cost     estimates     R million*	Financial       provision       R million**
	Impala	(766.5)	(458.6)
	Springs	(70.7)	(35.7)
	Marula	(100.7)	(49.1)
	Afplats	(16.4)	(8.2)
	Zimplats	(412.8)	(236.4)
	Totals	(1 367.1)	(788.0)

*	The current expected cost to restore the environment disturbances as estimated by third party experts excluding VAT, P’s & G’s and contingencies
**	Future value of the current cost estimate discounted to current balance sheet date as provided in the annual financial statements of the Group