ENVIRONMENTAL PROFIT & LOSS (E P&L ) 2014 GROUP RESULTS

Transcription

1 ENVIRONMENTAL PROFIT & LOSS (E P&L ) 2014 GROUP RESULTS

2 03 FOREWORD Contents Foreword page 03 part A: 2014 E P&L Results page 04 part b: A focus on GHGs and leather page 13 part c: Innovation and new research page 32 It is essential that individuals, businesses, and society more broadly, create and share solutions that can help reduce our collective impact on the environment, and replenish our natural resources. It is only in this way that we will avoid the acceleration of the impacts that climate change is already having on people and nature. At Kering, we designed our pioneering tool, the Environmental Profit and Loss (E P&L) account, to measure and monitor our impacts on nature. We are inputting our learnings and methodology from the E P&L to the Natural Capital Protocol, which will help companies to better address environmental disruptions and ultimately help change the climate trajectory we are currently on. This new 2014 Group E P&L report is an update of our 2013 Group E P&L released earlier this year. It showcases progress on our results and our learnings, as well as the improvements we have made to the methodology itself. The release of this report is evidence of our commitment to transparency, and the comparison between 2013 and 2014 results is the first of the annual benchmarks that Kering will be providing. This demonstrates how fundamental the E P&L is to the Group s business reporting, and how regular deployment of the E P&L enables the incorporation of sustainability analytics into day-to-day business decisions. In turn, this adaptive approach helps us to build resilient business growth in the face of climate change. As illustrated in this report, we continue to be dedicated to ongoing testing and improvement of our E P&L methodology to maximise its effectiveness within our own business. As such, over the next months we are examining how to integrate data that helps us assess climate risk alongside other E P&L metrics. We will also be working with a number of global experts to find ways to better integrate the business impacts on biodiversity into an E P&L context. Furthermore, showing that the E P&L can be easily repeated year-to- year will advance efforts to build broader business engagement and the mainstreaming of corporate natural capital reporting. We are continuing to contribute to help drive natural capital accounting adoption by making our open-sourced E P&L methodology even more accessible to the business community. As such, Kering will also open-source the key multipliers of our E P&L to help companies calculate and value their environmental footprint. We look forward to presenting an even more in-depth 2015 Group E P&L which will include a 3-year trend analysis on our E P&L findings and links to our Sustainability Targets achievement. Marie-Claire Daveu Chief Sustainability Officer and Head of international institutional affairs

3 05 part A: 2014 E P&L Results What is an E P&L? An Environmental Profit & Loss account (E P&L) measures and values the costs and benefits generated by a company s environmental impact, both within its own operations and across all of its supply chains. Developed by Kering, the E P&L is an innovative tool that allows us to make better decisions as it provides a clear understanding and accurate measurement of our impacts; whether it be from the production of raw materials, to our own operations, transport and right up to the boutique floor. As a result we can: translate our environmental impacts into a language of business; identify the most significant drivers of impact in our business; compare different environmental impacts with each other, as well as comparing the impact of sourcing in, different locations, which was not directly possible previously; implement targeted projects PART A: 2014 E P&L Results SUMMARY OF GROUP RESULTS This is the second year we have produced and published the full results of our Group E P&L. As we continue to integrate the findings of our 2013 E P&L into our sustainability strategy, we are seeing some positive results in terms of our use of lower impact raw materials and manufacturing techniques. This journey is just beginning and we recognise these projects will take time to scale up, but our efforts are already starting to yield results; over the past year our revenue growth has outstripped the growth of our impacts. In 2014 our Group environmental impacts are estimated to be 793m, representing a 2.2% 2 increase relative compared to 2013, and revenue growth over the same period was 4.5% 3. For more details on what is an E P&L and how and why we developed it see our 2013 E P&L report 1. The results are not related to Kering s financial results past, present or future and do not represent a financial liability or implied or actual cost to Kering. Rather they are a new way of estimating the costs to society of the changes in the environmental that result from our business activities and the activities of the whole of our supply-chain. In contrast to financial accounting, there are not established and agreed on standards for estimating these values contained herein. These figures are not intended to represent a forward looking statement or any other financial obligation by Kering of any kind. Figure 1: OVERALL 2014 GROUP REVENUE AND 2014 GROUP E P&L RESULTS 2013 PF 2014 Revenue 10,038m Revenue +4.5% Revenue 9,656m E P&L 776m 4 E P&L 793m E P&L +2.2% Excluding licenced products 3 Including licenced products 4 Pro forma 2013 E P&L results are 776 million.

4 06 07 part A: 2014 E P&L Results Figure 2: E P&L impacts across the tiers, split by impact area AIR POLLUTION GREENHOUSE GAS EMISSIONS TIER 0: STORES WAREHOUSE OFFICES TIER 1: ASSEMBLY TIER 2: MANUFACTURING TIER 3: RAW MATERIAL PROCESSING TIER 4: RAW MATERIAL PRODUCTION TOTAL IN MILLIONS: 9% % Figure 2 shows that our most significant impacts are generated in the supply chain (93%), and in particular by the production and processing of raw materials (49% and 25% respectively). Our own operations represent only 7% of the impacts. This represents a minimal change from overall distribution of impacts across the tiers of our supply chain and operations. In part this is due to the fact that leveraging changes across the supply-chain is a longer term process and thus will not yield immediate results in some instances. Amongst the raw materials, leather continues to be the major driver of impacts, followed by textiles from plant fibres, synthetics and animal fibres (Figure 3). Metals have moved down from fourth to fifth in this list, in part due to reduced metal use in some products. Our key focus areas for reducing impacts will therefore continue to be: Identifying sustainable production practices and sourcing locations for raw materials; Reducing impacts from our manufacturing suppliers; Increasing efficiency of our retail, offices & warehouses. Figure 3: E P&L contribution of major groups of raw materials, and quantity of consumption LAND USE TOTAL MATERIAL IMPACTS 28% M AIR POLLUTION GREENHOUSE GAS EMISSIONS WASTE WATER CONSUMPTION WATER POLLUTION TOTAL IN MILLIONS: 7% % % % % % % % % RAW MATERIAL QUANTITY (KG) E P&L VALUED IMPACT ( ) 150M 100M 50M 0M 0M 10M 20M 30M 40M LEATHER PLANT FIBRE SYNTHETIC FIBRES ANIMAL FIBERS METAL NATURAL STONES PLASTIC RUBBER LAND USE WASTE WATER CONSUMPTION WATER POLLUTION SYNTHETIC STONES PAPER 50M TOTAL MATERIAL QUANTITY

5 08 09 part A: 2014 E P&L Results In terms of the geographic distribution of our sourcing, the overall pattern remains the same across the group from the prior year s results, as demonstrated in Figure 4 below. Figure 4: Map of impacts and key drivers North America The US is a key source of leather and cotton. It is also one of the key operational locations with some impacts driven by stores, offices and processing. Europe Italy is the main manufacturing location for luxury brands. Italy, France, Germany, Spain, Poland and Switzerland are also key locations for sourcing bovine leather. Whereas Turkey is a key source of lamb leather and cotton. Asia China is the main manufacturing location for sports and lifestyle brands. It is also a key sourcing location for cotton and cashmere, for example. A number of countries in Asia including Vietnam, Thailand and Pakistan are also key sourcing locations for cotton and other textiles. South America South America are key sources of wool and leather driving land use and GHG impacts. Africa Gemstone mining in South and Central Africa drives land use impacts. While sheep and goat rearing is responsible for impacts in Nigeria. Water pollution impacts in Chile, Bolivia and Peru are driven by metal mining. AIR POLLUTION GREENHOUSE GAS EMISSIONS LAND USE WASTE WATER CONSUMPTION WATER POLLUTION Australia and New Zealand Australia and New Zealand are key sourcing location for wool and sheep leather. VALUED RESULT 50M 100M 150M 200M

6 10 11 part A: 2014 E P&L Results UNDERSTANDING 2014 VerSus 2013 results As with our 2013 E P&L, to calculate our results we worked with our suppliers to collect environmental and operational data, validated it, and completed any gaps using best available techniques. The calculations themselves were run using our newly developed online E P&L tool (see Figure 5). Figure 6, opposite, shows the main drivers of change between 2013 and There have been some updates to the scope, methodology and input data since the 2013 analysis; see Appendix (page 35) for details of 2013 pro forma calculation. figure 6: Description of main drivers of change between our 2013 Pro Forma and 2014 results Type of Change Change Reason for Change TieR 0-2 million Improvements in energy efficiency in some of our brand s core operations Manufacturing efficiency improvement - 9 million Efficiencies in our Tier 1 and 2 suppliers manufacturing and product assembly processes. figure 5:, SHOWING RELATIVE INFLUENCES OF CHANGES SINCE 2013 PRO FORMA RESULTS Increased manufacturing volume + 15 million Increased production volumes compared to Reduction in impact Increase in impact 2014 E P&L 2013 pro forma E P&L Raw material production and processing + 12 million Changes in sourcing locations and quantities of raw materials purchased. A more detailed breakdown of this change is provided below. EP&Limpact ( millions) Other + 1 million Includes several small effects including changes in locations and operational procedures PRO FORMA TIER 0 MANUFACTURING EFFICIENCY IMPROVEMENT INCREASED MANUFACTURING VOLUME RAW MATERIAL PRODUCTION & PROCESSING OTHER 2014

7 12 13 EXECUTIVE SUMMARY Much of our sustainability efforts continue to focus on reducing the impacts of the raw materials we use in our products. A closer look at the impact associated with the use of raw materials (Figure 7) shows that whilst there have been some improvements associated with certain materials, overall our raw material impact has increased. This is largely to do with an increase in production volume, resulting in larger quantities of raw materials required. figure 7: A closer look at changes in raw material impacts since 2013 pro forma result E P&L impact (millions of ) Reduction in impact Increase in impact 2014 E P&L 2013 pro forma E P&L PART B: A FOCUS ON GHGS AND LEATHER PRO FORMA CASHMERE CALF LEATHER METALS DIAMONDS PAPER COTTON RUBBER SYNTHETIC STONES OTHER 2014

8 14 15 part b: a focus on ghgs and leather The 2015 Paris Climate Conference of Parties (COP), will see delegates from across 190 governments and leaders from business and civil society meet to forge a response to the challenges presented by our changing climate. Without concerted climate mitigation efforts, global mean surface temperature could increase by more than 4 Celsius by the end of the century. This is double the threshold of 2 C degrees, above which it has been globally agreed climate change risks become unacceptably high. However, more action is needed if we are going to avoid average temperature rises above 2 C. PwC analysis shows that the reduction commitments announced by governments in the run up to the Paris COP will go some way towards reducing these risks, but remain insufficient. It s clear that business must play a central role if we re to limit warming to 2 C; the E P&L is our tool to help us identify and prioritise opportunities to do so. figure 8: breakdown of total GHG impacts and key drivers TIER 0 (11%) TIER 1 (13%) TIER 2 (7%) TIER 3 (38%) TIER 4 (31%) TIER 0 (11%) TIER 1 (13%) TIER 2 (7%) TIER 4 (31%) In this section we focus on GHGs and leather, as these are the key driver of emissions in our supply chain. We hope this analysis can help demonstrate how all businesses can understand their contribution to climate change and identify actions to lead us all onto a more sustainable path. We focus on GHGs here because of the relevance to the upcoming COP. However, when it comes to decision making it is really important to consider all environmental impacts, which is why the E P&L is designed to provide a picture of all environmental impacts. Kering s contribution to GHG emissions In 2014 Kering s total GHG emissions totalled 4.6 million tonnes. These are largely driven by the production and processing of raw materials (Figure 8). Principle among these is leather, which accounts for 29% of our total GHG emissions. TIER 3 (38%) figure 9: breakdown of total GHG impacts and key drivers key impact drivers WHAT WE Are doing tier 0 tier 1 tier 2 TIer 3 tier 4 Electricity consumption of retail stores and offices. Smart Sustainable Stores and Smart Sustainable Offices - We have developed detailed guidelines and good practice examples that reduce energy consumption at our stores and offices. We have also been investing in onsite production of renewable energy at our facilities through solar panels. Smart Metering - We have piloted the use of smart metering technology in selected stores to monitor consumption and identify areas of improvement in energy efficiency. Electricity consumption of assembly process in our supplier sites and their ancillary suppliers. Electricity consumption of cutting process in our supplier sites and their ancillary suppliers. Smart Supplier Financing - We have continued to help our suppliers access national or regional level financing mechanisms for environmental efficiency projects. Processing of raw materials, particularly tanning of leather, cutting of stones, and spinning weaving and dyeing of textiles. Clean by Design - We have been working to promote best practices in energy management and identify opportunities that are cost, energy, electricity and fuel efficient. 25 of our Group s textile mills have already achieved 30% reductions in energy use and GHG emissions. Innovation in Manufacturing - In 2013, Gucci developed and launched a metalfree tanning process, a first for the luxury sector. This innovation reduces water and energy consumption. Animal rearing of cows, sheep and goats for leather and other fabrics. Smart Sourcing Strategy - Through our Leather programme we have been working to identify lowimpact bovine leather suppliers. We have also conducted an evaluation of our current sourcing practices for all our brands to better understand where and how we can make improvements. 5 IPCC Working Group III. 6 Low Carbon Economy Index 2015, PwC.

9 16 17 part b: a focus on ghgs and leather SAMPLE PRODUCT EXAMPLES To demonstrate the impact of GHGs relative to other environmental impacts, we compare and contrast three product examples on the following pages. As shown, it is quite a significant factor in several different component raw materials. figure 10: Impacts of a leather handbag Example calculation for land use, water and GHG impacts for leather, brass and linen inputs to a handbag Figure 11: IMPACTS of a silk-wool dress Example calculation for land use, water and GHG impacts for silk, wool and wood inputs to a dress ONE HANDBAG MADE FROM LEATHER, BRASS AND SILK REQUIRES 18.7m 2 LAND 0.9m 2 of water... AND PRODUCES 40.7kg C0 2 emissions AN IMPACT OF One silkwool dress made from silk, wool & wood requires 18.2 m 2 land 14.4 M 3 OF WATER and produces 76 kg C0 2 emissions an impact of kg leather from France or 0.2 calf hides......needs 18.6m m 3 of water kg C0 2 e g of silk needs 4 m 2 14m 3 of water 51kg CO 2 e ONE HANDBAG 150g of brass......need 0.1m m 3 of water... 1kg C0 2 e 2.99 ONE DRESS 180g of wool or 6% of a sheep s annual coat needs 14 m 2 0.4m 3 of water 25kg CO 2 e g linen lining......need 1/100 th m /30 th m 3 of water kg C0 2 e g of wooden buttons needs 0.2 m 2...trace amount of water... 3/1000 th kg CO 2 e Fraction of a cent GHGs GHGs Land use Land use Water use Water use Figures 10, 11 and 12 show the relative impacts of different materials we use in our production of some sample types of products. This helps put in simple terms the comparative impact of different types of products and are based on the average of data we collected. The E P&L itself contains over 2 million data points, but illustrations like this can show just how much resources it takes to create the items we wear or use in our daily lives. For handbags, the key driver of different impacts is the quality of the grazing lands; poorer grazing quality means lower stocking rates and higher land use impacts alongside more GHGs as the animals digestion releases more methane. The impacts of producing the silk and wool make up the majority of the impacts of the dress. Silk impacts are driven by water use, essential for irrigating the mulberry bushes used to feed the silk worms, and for processing the cocoons into silk threads. Impacts of silk are typically much lower in areas where irrigation is not required. Land use and GHGs are the main driver of impacts of wool; these are lowest where sheep are raised on higher quality pasture because they take up less land and produce lower volumes of methane from the digestion of more nutritious grass.

10 18 19 part b: a focus on ghgs and leather Figure 12: IMPACTS of a gold watch Example calculation for water use, water pollution and GHG impacts for gold, metal and synthetic stone inputs to a watch One watch made from gold, metals, and synthetic stones requires 8m 3 of water and produces 1.8 kg of water pollution and produces 2,045kg C0 2 emissions An impact of 382 Details of the E P&L methodology to estimate GHG emissions from luxury leather Step 1 Decide what to measure Leather is the largest material group, by weight, used in our production of luxury products. We use more than 40 types of leather, although 98% is cow, calf, sheep or goat leather. Figure 13 presents a high level summary of the types and sources of leather used in our products in g of 18-karat gold 7m 3 of water 1.7kg 1,127kg CO 2 e 262 In our 2013 E P&L report, published earlier this year 7, we introduce the 7 step process we developed to guide the E P&L calculation process. Here we present this process in more detail for luxury leather production, focusing in particular on its contribution to climate change. Each type of leather has a different impact as it involves unique agricultural and tanning processes. To ensure the results are of sufficient quality to support decision making, it is critical these differences are taken into account in our analysis. At this stage it is only necessary to identify the variety of production systems, Step 3 will use this to set the data collection, estimation and modelling priorities. One watch 40g of iron, nickel and zinc...1/100 th m 3 of water 1/1000 th kg 0.1kg CO 2 e 0.02 figure 13: Kering 2014 leather use, by source region 10g of synthetic stones and sapphire glass...1m 3 of water...0.1kg 918kg CO 2 e ,000 GHGs Water use Water pollution The impacts of the gold watch are dominated by water pollution associated with gold mining. The impacts per kg of other metals, such as iron and zinc, are much lower than gold because the ores contain a considerably larger quantity of metal, so less mining is needed to extract a greater quantity. Furthermore, their ores tend to contain lower quantities of toxic heavy metals. The sapphire glass, used for the watch face, and synthetic stones, such as rubies used for the mechanism, contribute almost a third of the total impact, as a result of the energy used in their production and cutting. Where these processes are supported by renewable energy, such as in parts of Switzerland, the impacts are lower. LUXURY LEATHER QUANTITY BY SOURCE REGION, TONNES 30,000 25,000 20,000 15,000 10,000 5,000 SHEEP AND GOAT OTHER PRECIOUS SKINS COW AND CALF EUROPE AFRICA ASIA AUSTRALIA NORTH AMERICA south AMERICA 7

11 20 21 part b: a focus on ghgs and leather Step 2 Map the supply chain The objective of this Step is to identify all the processes, together with their inputs and outputs, which are required for the production of the different leathers. It is only through understanding these processes in detail that we can define the required data in Step 3 to produce robust results. There are three principle processes to the leather supply chain; animal raising, abattoir and tanning. It is the inputs and outputs of these process which ultimately drive the creation of impacts. Figure 14 summaries the processes and estimates of these impact drivers are required for each type of leather in each location identified in Step 1. Step 3 Identify priority data This Step is split into two parts: defining the data needs, and identifying the best available methods for collecting or estimating the data. Define data needs Given the diversity in types and sources of leather it is important to prioritise the data needs. We initially focus on the main types of leather and principle sourcing locations to estimate base models. We then consider how the most significant drivers of impact differ in production systems in other locations in our supply chain so we can adjust these base models to reflect the key differences (Figure 15). For example, in relation to calf and bovine leather we estimate three different base models to represent the impacts of producing leather from young calves, calves and adult cows in France. In order to adjust these base models for different locations we consider how the different production systems affect the amount of methane released by the cows, the type and quantity of feedstock and fertiliser used, and the amount of land required for grazing. The impacts of GHGs and land use from these variables represent more than 95% of the total impacts of raising cattle. By taking these differences into account we are therefore able to produce sufficiently accurate local estimates around the world to support identification and development of lower impact sources of leather. figure 14: Summary of the processes, inputs and outputs in the leather supply chain figure 15: base models and adjustments for quantifying impact drivers for different leathers Land, water, electricity and fuels Leather group Specific base models to be estimated for animal raising Adjustments to base models to reflect different locations Feedstock farming Fertiliser and herbicide production Tanning agent production Calf and BOVINE - French young calf - French calf - French cow - Grazing quality affecting methane production - Stocking rates and land use intensity - Use and source of offsite feedstock (e.g. soybeans) Sheep and goat - Nigerian goat - European sheep and lamb - Stocking rates and land use intensity ANimal Raising Breeding > Grazing > Fattening Livestock Abattoir Slaughterhouse & Waste-water treatment Salted hide Tanning Beamhouse > Tanning > Finishing Finished leather Coproducts e.g. milk Coproducts e.g. meat, offal Coproducts e.g. fertiliser Emissions to air, soil and water, solid waste

12 22 23 part b: a focus on ghgs and leather For each of the base models it is necessary to collect data on the drivers of impact; the inputs and outputs for each process identified in Step 2. Figure 16 presents the key data requirements for the tanning process as an example. Many of these impact drivers lead to GHG emissions. figure 16: Summary of data needs for tanning of leather Core tanning processes: Data on impact drivers required (per m 2 finished leather): Data for inputs and outputs is required for each of the sub-processes: Beamhouse (raw hide to wet blue) Tanning (wet blue to crust) Finishing (crust to finished leather) Chemical production for inputs to tanning processes: Data for inputs and outputs is required for each of the types of chemical inputs, taking into consideration the whole chemical supply chain: Biocides, fungicides, surfactants, degreasers, acids, alkalines, tanning agents, dyes, resins, lacquers and other finishing agents Inputs: Electricity (kwh) Water (m 3 ) Land use (ha) Fuel (m 3, split by type of fuel) Chemicals (kg, split by type of chemical) Outputs: Waste (kg, split by type of waste) Water pollutants (kg, split by specific pollutant, e.g. arsenic, chromium III, phosphates) Co-products (kg, split by type of co-product) Wet blue/ Crust/ Finished leather (m 2 ) Data on impact drivers required (per m 2 finished leather): Inputs: Electricity (kwh) Water (m 3 ) Land use (ha) Fuel (m 3, split by type of fuel) Chemicals (kg, split by type of chemical) Outputs: Waste (kg, split by type of waste) Water pollutants (kg, split by specific pollutant) Co-products (kg, split by type of co-products) Chemical product (kg) Identify best available methods for obtaining data There are broadly five different types of data which can be sourced to quantify the impact drivers: Primary data from Kering and its brands; Primary data from supplier surveys; Secondary data from life cycle assessment (LCA) studies, national and industry statistics; Secondary data from material flow analysis or productivity modelling; and, Secondary data from economic models. In order to get results which are representative of the context and practices in our supply chain it is important to select the best available data for each process. Our 2013 E P&L methodology report discusses the general considerations for selecting data in more detail so it is not repeated here. Given the importance of leather to the overall E P&L result, we use as much primary data from our operations and those of our suppliers as possible, and compliment this with LCA data: Tanning: Over the last three years we have assessed 70 tanneries operated by direct suppliers of Kering Group. This provides the majority of the data required, particularly for core inputs and outputs such as energy, water and waste. We supplemented this with LCA data on the production of chemical inputs and the associated emissions to water. We are continuing to refine this data and have recently commissioned another study focusing on the benefits of the alternative chrome-free and metal-free tanning technologies being developed and used by our brands. Abattoir: LCA data is used to estimate the average impacts of abattoir operations. The relative value of the hide and other co-products, including meat, is used to allocate the impacts to our leather demand. Animal raising: The data framework for animal raising is provided by a comprehensive LCA of the French cattle industry. The level of detail provided by this framework allows us adapt specific details to reflect the range of production practices in our supply chain. For example, we used data from the FAO on grazing quality and stocking rates to adjust for the quantity of methane produced by the cattle, the production feedstock, and the land area required. We are using this, together with primary data on production practices gathered directly from farms by Origem (a specialist consultancy) to identify sustainable sources of bovine leather. Steps 4 and 5 Collect primary and secondary data Collecting accurate data that is representative of our operations is crucial to the success of the E P&L. Our 2013 methodology report presents the process of primary and secondary data collection and validation. Here we present the results of these efforts for GHGs released through the production of calf leather (Table 17 and 18 overleaf). For each process we show two different scenarios to illustrate the range in impacts across different production systems and the importance of using data which is representative of our operations and supply chain locations. The difference in tanning emissions is primarily driven by the efficiency of the tanning operations, as well as the emission intensity of the energy being consumed. In Asia, compared with Italy, more than twice as many GHGs are release for every kwh of electricity consumed from the grid because it uses a higher proportion of coal in its energy mix. This confirms another reason, beyond higher quality, that our sourcing leather from Europe with a focus on Italy is beneficial for our luxury brands. The impacts of chemical and fuel production in this table represent industry average which we have not sought to differentiate between countries because these inputs are both sourced by the tanneries from a common global market.

13 24 25 part b: a focus on ghgs and leather Figure 17: Example data for GHG emissions from tanneries in Italy and Asia Figure 18: Example data for GHG emissions from animal raising and abattoirs in France and Argentina Italy 2.63kg CO 2 e per kg of leather Asia 3.34kg CO 2 e per kg of leather France 17.1kg CO 2 e per kg of leather Argentina 48.6kg CO 2 e per kg of leather tanning Raw hide to wet blue: Tanning operations Raw hide to wet blue: Chemical production Raw hide to wet blue: Fuel production Wet blue to finished leather: Tanning operations Wet blue to finished leather: Chemical production Wet blue to finished leather: Fuel production Animal raising - animal emmisions Animal raising - farm operations & application of fertilisers Animal raising - chemical production Animal raising - offsite feedstock production Abattoir The emissions released through animal raising are far more significant than those from tanning, and the difference between locations is also much more important. The key difference is due to the quality of the grazing lands. Lower quality Argentinean grazing leads to more methane emissions released by the animals through enteric fermentation, and increased use of organic fertilisers with associated N 2 O emissions (this is slightly offset by reduced use of chemical fertilisers). We have not differentiated emissions of abattoirs by country because their contribution to the overall GHG impact is minimal.

14 26 27 part b: a focus on ghgs and leather Step 6 Determine valuation Figure 19: Impact pathway for greenhouse gas emissions The objective of the valuation exercise in Step 6 is to estimate the extent of societal impacts, now and in the future, resulting from emissions of GHGs today. As with the rest of the E P&L, we use a societal valuation approach to achieve this. Here, we first discuss why we believe this is the right approach for our purposes, before presenting further detail behind the valuation methodology. The societal cost of carbon The societal cost of carbon (SCC) represents the total economic costs, now and in the future, associated with the emission of one tonne of CO 2 e today. This includes damage to the environment, human health, property and agriculture, and disruption of our economies. Alternative approaches present different types of value: 1) Marginal abatement cost approach: An abatement cost is the financial cost of achieving a certain level of GHG emissions reductions given prevailing conditions; typically by investing in technological solutions or substitutes. However, this is not appropriate for the E P&L because it is not a measure of impact. Abatement cost is a measure of the cost to business of a given reduction, rather than the consequences for society of a change in emissions. 2) Market price approach: In some places, for some industries, there are market prices for carbon enforced by governments through cap-and-trade systems or taxes. Market prices do not reflect societal impacts either. Prices in cap and trade markets are a function of the supply and demand for credits within relatively inflexible regulated markets. As such, they give the current private cost of GHG emissions for regulated installations, but tend to be a poor proxy for the societal impact of those emissions unless the price happens to coincide with the societal cost of carbon. Carbon taxes are set by governments, and in theory, can be set at a level equal to the societal cost of carbon. However, carbon taxes vary significantly between countries and sectors and rarely reflect the true extent of externalities imposed on society by GHG emissions. Market prices are therefore inappropriate for use in our E P&L. We chose to use the SCC as a measure of the impact associated with GHG emissions as it is the only approach consistent with the theory of welfare economics, seeking to measure changes in human wellbeing associated with corporate environmental impacts. It is therefore consistent with the wider conceptual framework for the E P&L. The other valuation approaches are useful, but it is only through integrating the SCC into business decision making that we can properly understand our impacts and prioritise actions to avoid or reduce these impacts. Calculating the societal cost of carbon On completion of Step 5 we have a complete picture of the quantity of GHG emissions associated with leather production the impact drivers. There are two further parts to understanding the societal cost of these GHG emissions: - First we must evaluate how emissions of GHGs contribute to a changing climate the environmental outcomes. - Secondly, we need to consider what the consequences of a given change in climate change will be for populations around the world the societal impacts. PwC s impact pathways describe the relationship between impact drivers, environmental outcomes and societal impacts (Figure 19). Corporate Activities IMPACT DRIVER (Source: PwC) GHG emissions: CO 2 CH 2 N 2 O HFCs PFCs SF 6 Other GHGs Increasing concentrations of atmospheric GHGs ENVIRONMENTAL OUTCOMES Shifting climate patterns Sea level rises Increasing extreme weather events Rising mean temperatures SOCIETAL IMPACTS HUMAN HEALTH: Malnutrition due to increasing frequency of droughts and floods, reduced agricultural output, spread of disease and heat related deaths BUILT ENVIRONMENT: Damage from extreme weather events and interesting adaptation costs due to shifting patterns of climate ECONOMIC DISRUPTION: Economic loses caused byproduction and supply disruptions, particularly for manufacturing and agricultural supply chains AGRICULTURE AND TIMBER: Crop losses; changes in growth and yields DESERTIFICATION: Loses of productive and habitable land OTHER ECOSYSTEM SERVICES: Widespread changes affecting biodiversity and many associated ecosystem services 8 In fact, we do use these values to compliment the SCC in decision making because we also need to understand the financial costs of different interventions (e.g. technological abatement costs) which can be compared to the savings in E P&L impact (i.e. societal impact).

15 28 29 part b: a focus on ghgs and leather Academic precedents to estimating the SCC Calculating the SCC for the E P&L figure 20: Summary of methodological choices in estimating SCC Unsurprisingly, there has been considerable work by academics and governments to estimate the SCC. The huge body of work underlying IPCC s (Intergovernmental Panel on Climate Change) five assessment reports and modelling scenarios has helped advance our understanding of how anthropogenic emissions of GHGs are contributing to climate change. Economists have used the IPCC scenarios to consider what the associated costs will be, considering for example costs of coastal flooding, changes in agricultural productivity, water availability, extreme weather events, economic productivity and the collapse of ecosystems. The basic steps academic studies take to produce new estimates of the SCC are: Modelling environmental outcomes: 1) Selecting an emissions scenario (typically one of the IPCC scenarios) 2) Constructing a global climate model to project the likely future atmospheric GHG concentrations and resulting changes in climatic conditions Estimating societal impacts: 3) Developing impact assessment models to quantify associated impacts on society 4) Estimating the total economic costs associated with these impacts 5) Discounting back the total cost estimate to the present day using a social discount rate 6) Apportioning the net present value of climate damages according to the volume of anthropogenic GHGs emitted Alternative methods to identify an SCC estimate include producing a new primary estimate following the method described above, selecting a single estimate from a single study, or undertaking a meta-analysis of multiple studies. However, given the wealth of existing research there would be limited value in producing a new study in the absence of new information. PwC s approach applies a meta-analysis rather than selecting a single study because it allows them to take account of the different underlying assumptions and scenarios, reflecting the uncertainty in the SCC. The approach is not a purely statistical meta-analysis (since it incorporates a number of non-statistical factors), but it shares some of the key benefits of a conventional statistical meta-analysis, particularly the ability to incorporate the results of multiple studies applying a range of different methods and scenarios. It also has the significant advantage that, by setting rules for selecting a sub-set of studies, an automatic and un-biased mechanism to update the SCC estimate over time (as new research becomes available) is also established. Criteria considered in PwC s meta-analysis of SCC studies Identifying an appropriate central estimate for the societal cost of carbon is challenging because estimating the cost of greenhouse gas emissions is subject to a range of assumptions. To account for these various assumptions, PwC select a sub-set of academic studies by applying a set of criteria, and calculate the mean SCC estimate from this sub-set. Selection criteria are limited to those considered most fundamental; details are summarised in Figure 20. Factor Quality of study Age of study Discount rate Treatment of outliers Methodological choice in estimating SCC Estimates from peer reviewed studies will be used. Estimates from the ten most recently published peer-reviewed studies in our dataset are included. Estimates that apply Pure Rate of Time Preference (PRTP) = 0% are included. Estimates are not selected according to the values they use for future economic growth rates and income elasticity of marginal utility. Estimates more than three standard deviations from the mean are excluded. Assumptions and justification Peer review is the only widely accepted measure of quality applicable to studies of the societal cost of carbon. The significant and apparently systematic difference in values (peer reviewed values are typically lower) suggests that this is an important criterion. Studies and estimates are generally perceived to improve in quality over time, as both climate modelling and economic damage assessment methods advance. We therefore deem it appropriate to focus on more recent peerreviewed estimates of the SCC that meet our criteria, while maintaining a reasonable number of estimates to reflect the diversity of views about underlying assumptions and avoid the result being skewed by a single study. A discount rate is used to convert future damage costs to their present value. In established economic theory, the discount rate includes the PRTP, a forecast of economic growth, and the marginal elasticity of utility with respect to income. We consider it ethically defensible and aligned with notions of inter-generational equity commonly found in the climate change literature to value the wellbeing of future generations equally to our own and therefore only select studies with a PRTP = 0%. It is not possible to select a subset of estimates that use specific values for income elasticity of marginal utility and economic growth rate because not all studies disclose this information. However, those that do disclose their assumptions show a sample average of 2.5%. Eliminating outliers helps to prevent extreme values from unduly distorting sample statistics. However, the possibility of catastrophic climate outcomes (however remote) is generally accepted, and estimates of the SCC have been observed to follow a fat-tailed distribution. We acknowledge the likelihood of this type of distribution by including estimates up to three standard deviations from the mean, but consider estimates outside this range to be true outliers and exclude them from our sample statistics on this basis. 9 Ramsey, F., 1928, A Mathematical Theory of Saving, Economic Journal, vol. 38,

16 30 31 part b: a focus on ghgs and leather table 21: Summary of normalisation steps Factor Weighting of estimates Monetary inflation Methodological choice in estimating SCC A multiple estimates weighting is applied to values from studies which contain more than one estimate. Monetary inflation has been addressed by inflating each SCC estimate using World PPP-adjusted GDP deflators. Assumptions and justification Studies with multiple estimates are weighted such that the sum of weightings for all estimates from a single study is 1. This is as applied by Tol (2011) and prevents individual studies containing large numbers of estimates crowding the sample and distorting the average SCC obtained towards the methods they employed. The technique also attempts to reflect the confidence placed by the author in each estimate. The value of a given monetary unit typically decreases over time as a result of monetary inflation. As the underlying studies relate to different years, the estimates need to be adjusted for monetary inflation to be comparable. Most studies explicitly or implicitly assume constant real exchange rates into the future. In practice real exchange rates have varied materially in the past twenty years; for this reason, World PPP adjusted GDP deflators are calculated for inflating older SCC estimates. Not all studies publish the year the SCC has been calculated for; therefore, the inflation rate from the rounded year of publication of each study has been applied. SCC estimates are not selected according to the equity weighting used or damage valuation approach as no consensus exists on the appropriate method to use. Following the selection of a sub-set of SCC estimates, a set of normalisations are applied so that the estimates are expressed in common units and a common year, to account for changes since the publication of the studies, and to ensure that a single study is not over-represented. Figure 21 opposite summarises the normalisation steps. Finally the estimate of the SCC is calculated from this normalized sub-set of estimates by taking the arithmetic mean and median values. There are valid statistical and ethical reasons for choosing either a mean or median value in this context. We have opted to take the mean, which takes more account of very high estimates derived from potentially catastrophic climate scenarios and therefore reflects a more precautionary approach to potential climate change. The resultant value in 2014 Euros is 62 per tonne of CO 2 e emitted. For readers wanting further detail, PwC s full methodology for estimating the Societal Cost of Carbon is available online here: Step 7 Calculate and analyse your results The final step of the calculation is to multiply the quantity of GHGs emitted by the SCC. For other types of impact the valuations are location specific, but as one tonne of GHGs has the same contribution to climate change wherever in the world it is emitted, the location has no bearing on the valuation of GHG impacts. Figure 22 presents the impacts of sourcing calf leather for the two scenarios presented in Step 5 above; animal raising in France with tanning in Italy, and animal raising in Argentina with tanning in Asia. Figure 22: E P&L impacts of calf leather from sourced and processed in different locations Growth rate of SCC over time Growth rate of SCC assumed to be 3% per year. Unit conversion Conversion of $/tc to $/ tco 2 e has been carried out by multiplying societal costs expressed in tc by the fraction Because the profile of anticipated climate damages is weighted into the future, and GHGs reside in the atmosphere for a limited period, the climate impact of an additional tonne of CO 2 e rises over time. Three percent is the mid-point of the IPCC estimated range (2 4%) for this rate of increase. Estimates of the SCC from the academic literature are typically expressed in: $/tce. We wish to present our results in the industry-standard units, $tco 2 e. PwC therefore adjusts for the difference in weight between a single atom of carbon (atomic mass = 12u) and a molecule of CO 2 (molecular mass = 44u). LUXURY LEATHER QUANTITY BY SOURCE REGION, TONNES AIR POLLUTION GREENHOUSE GAS EMISSIONS LAND USE WASTE WATER CONSUMPTION WATER POLLUTION Distribution of data No fitted distributions are applied for the purpose of producing the SCC. The sub-set of estimates selected (after applying the criteria set out above) does not clearly fit a specific distribution. PwC therefore considers it more transparent to use unfitted data to derive sample statistics. animal raising ABATTOIR ARGENTINA TANNING CHINA ANIMAL RAISING ABATTOIR FRANCE TANNING ITALY

17 33 part C: INNOVATION AND NEW RESEARCH PART C: INNOVATION AND NEW RESEARCH Incorporating climate risks Kering in recently co-published a report with BSR a report entitled, Climate Change: Implications and Strategies for the Luxury Fashion Sector, aims to help luxury brands, and the fashion industry more broadly, understand and respond to their specific climate change risks. The report evaluates the consequences of climate change on primary production and initial processing locations, and highlights the hotspots where action should be focused. We then present a climate resilience agenda as a guide for luxury fashion brands to design holistic solutions to reduce their greenhouse gas (GHG) emissions and build adaptive capacity in their value chains, including for raw material production. As a next step from this work, Kering will incorporate analysis of climate risks as an additional dimension of analysis provided by the E P&L. Improvements outside of the value chain shall be presented alongside the E P&L results, but shall not be netted off from them because they do not relate to the core operations of a company (which is what the E P&L is designed to illustrate). Biodiversity and EcoSystem Services Kering is working with leading global academic experts, conservation practitioners and environmental economists on the way we should measure the impacts of our business on biodiversity. This is part of Kering s ongoing commitment to enhance our E P&L methodology in the broader natural capital context and in conjunction with the Natural Capital Protocol. The results of this work will be shared publicly in accordance with our open source philosophy. Accelerating the E P&L process Alongside scope and methodology developments we have also been focusing on streamlining the process of physically generating E P&L results. In order to use the E P&L as a real-time decision making tool we need to be able to generate results quickly, and easily, provide what-if capabilities and enable a self-service toolset for our brands. To enable this we have developed an online tool that brings the calculation time down from months to minutes. This tool enable us to take over 2 million data points and calculate results in real-time, and then visualize these results in a business analytics tool. It is the investment we have made in understanding the impacts generated by each individual process in our supply chain that makes this possible. Within the tool we have impact data relating to each of these processes and materials we use, so we can construct scenarios to quickly estimate the implications of one decision or another within our business. We will provide more details on this tool and open source these multipliers in Spring of Sharing our multipliers Kering will continue to encourage other organizations to use tools like the E P&L to help lessen their impacts and gain business insight. To facilitate this, we will open source key multipliers (these are the factors we use to calculate and value our footprint) that will help enable companies to identify impacts associated with raw material use and production processes. 10 Accounting for environmental benefits in the E P&L, Kering, 2015

18 35 APPENDIX appendix Pro forma 2013 results As we continue to embed the E P&L within our business we are also continuing to improve the E P&L as a decision-making tool. Since 2013 we have expanded the scope of the analysis, developed better protocols to collect information from our supply chain, and refined the measurement and valuation approaches of the impacts themselves. To allow fair comparison we have re-calculated pro forma 2013 Group results to reflect these changes. Figure 23 shows the relative contribution of the changes to the previously published 2013 results, 773 million, resulting in a revised pro forma 2013 E P&L of 776 million. Updated Scope The scope of this report includes all business units and operations that are consolidated into Kering s financial reporting for 2014 excluding all licensees and some minor areas of our business. Together these represent less than 1% of total product revenue, however the impacts associated with these areas are being further researched so as to be included in future E P&L analysis. Figure 23: 2013 pro forma results showing influences of changes since initial 2013 results Reduction in impact Increase in impact PRO FORMA E P&L 2013 ORIGINAL E P&L RESULTS EP&Limpact ( millions) METHODOLOGY CHANGES SCOPE CHANGE CORRECTIONS TO DATA 2013 PRO FORMA

19 36 37 APPENDIX Figure 24: Summary of scope, methodology and data improvements since the 2013 initial results Type of Change Change Reason for Change Type of Change Change Reason for Change Methodology Methodology Methodology More accurate impacts for sheep and goat leather More accurate impacts for calf leather Differentiation of viscose impacts in different locations Key sourcing locations for our sheep and goat leather required further research on ground level impacts. So we wanted to ensure the impacts were reflected as accurately as possible. We looked in-depth at the farming practices, land quality, water consumption and astsociated emissions. For leather sourced from grasslands that do not impact deforestation in the Amazon, we have taken steps to more accurately reflect the impacts for these specific sourcing locations rather than using average data for Peru. This year we have looked in detail at a range of life cycle assessments for viscose grown in different regions. Consequently we can better reflect the impacts associated with sourcing from different sources. There is however much work still to do on improving transparency in the viscose supply chain, we are working to identify sustainable sources and quantify the advantages associated with these sources relative to current industry practice. Scope Scope Scope Added extraction and processing of precious stones and semi-precious stones Added extraction and processing of synthetic stones and sapphire glass Added gold chain making business unit There is limited life cycle data available for gemstone mining which prevented us from including it as part of the E P&L previously. This year we were able to assess impacts from primary data collected from ruby and emerald mines. Although data is likely to vary widely between mine types and locations the data allows us to begin to estimate the likely impacts from the production of precious stones; we will seek to continue to refine this in the coming years. We were able to collect primary data on sapphire glass and synthetic stone production processes from manufacturers in France and Switzerland. We also used life cycle data on ore extraction to get a complete picture of the impacts along with the primary cutting data we collected last year. Continued engagement with our jewellery suppliers meant that we were able to collect data on the impacts from gold chain making this year. This has allowed us to get a more complete picture of the impacts associated with our jewellery manufacturing business. Methodology Use of trade data Transparency for some materials continues to be challenging, particularly those where we are a minority consumer, such as metals. We therefore use trade data based on the country of purchase to understand the potential sources and estimate the impacts accordingly. In 2014 this trade data was updated. Changes to input data Updated 2013 data Continuous improvements to data collection techniques and ongoing communications with suppliers is allowing us to improve the accuracy of the data we collect. Consequently we have reviewed our 2013 data with our brands and suppliers and updated this data in a few cases where more accurate data became available.

20 Disclaimer The Environmental P&L (E P&L) issued by Kering is the product of a methodology developed by to measure the impact of an economic activity on the environment, applying financial metrics. The E P&L is one among other manifestations of Kering s commitment to protect the environment and leadership in sustainability. As such, Kering aims to share the methodology and tool hereby published with the general business community so as to make sure they will be improved and benefit to other actors in their own efforts to minimise the impact of their own industrial and economic activities on the environment. Because of its nature the E P&L cannot achieve the accuracy of financial results nor can it be subjected to financial audits. For any financial information about Kering, readers should refer to Kering s Reference Document (document de reference) and other published information (regulated information disclosed as such). As a result, the E P&L in no way reflects nor has any impact on Kering s past, present or future financial performance. In particular, the E P&L does not create any liabilities, implied costs or any rights to offset any amounts contained therein, nor does it trigger any provisions and neither does it result in any off balance sheet commitments. Finally, Kering makes no express or implied warranty or representation in relation to any information or data contained in the E P&L. Therefore, none of Kering or its representatives will have any liability whatsoever in negligence or otherwise for any loss however arising from any use of the E P&L or its contents or otherwise arising in connection with this presentation or any other information or material comprised in or derived from the E P&L. Kering Société anonyme (a French corporation) with a share capital of 505,065,960 Registered office: 10 avenue Hoche Paris Cedex RCS Paris Tel.: Fax: November 2015

21 E P&L