Delphi report 160217 final
- 1. Project Delphi The Material ESG Factors and Metrics that Drive Value February 2016
- 2. INTRODUCING PROJECT DELPHI 1 INTRODUCING PROJECT DELPHI | February 2016 CONTENTS P.2 What is Project Delphi? P.2 What is the real problem that Delphi was set up to solve? P.3 Definitions ESG Materiality Parsimony P.5 Who is behind the Delphi Framework? P.6 The Delphi Process P.8 What makes Delphi different from existing ESG guidelines? P.8 Who can use the Delphi Framework? P.9 The Future of Delphi? P.10 Conclusion P.11 Appendix 1 – Taxonomy P.12 Delphi Timeline The whole or any part of this work may not be reproduced, copied or transmitted or any of its contents disclosed to third parties without express written consent from the Project Delphi Technical Group. The information provided does not constitute investment advice and it should not be relied on as such. It should not be considered a solicitation to buy or an offer to sell a security. It does not take into account any investor’s particular investment objectives, strategies, tax status or investment horizon. You should consult your tax and financial advisor. All material has been obtained from sources believed to be reliable. There is no representation or warranty as to the accuracy of the information and Project Delphi shall have no liability for decisions based on such information. The views expressed in this material are the views of the Project Delphi Technical Group through the period ended 15th February 2016 and are subject to change based on market and other conditions. This document contains certain statements that may be deemed forward-looking statements. Please note that any such statements are not guarantees of any future performance and actual results or developments may differ materially from those projected.
- 3. INTRODUCING PROJECT DELPHI 2 What is Project Delphi? The Delphi Project is an investor-led, collaborative initiative with aims to enhance ESG (Environmental, Social, Governance) integration in asset management by clarifying the ESG investment process. In order to achieve the aim, the project developed the Delphi Framework that is open for all users to implement. Version 1.0 of the Framework will be launched in the first quarter of 2016 and will be available for public use by interested asset managers and asset owners. It contains a list of the most material Factors per sector, to be considered for ESG integration. The ESG Factors are organized in three value drivers: Growth, Return on Capital and Governance & Risk Management. For each Factor, the Framework provides detailed metrics that can be collected from data providers. What is the real problem that Delphi was set up to solve? We believe that there are inefficiencies in the allocation of resources, due to excessive focus on the short-term and a failure to take wider ESG considerations into account. We believe that this failure derives from a lack of consensus as to which are the key ESG factors to be taken into account, inadequate reporting of them and, consequently, incomplete data for analysts and investors to work with. There is an incomplete dialogue taking place between investors and reporting companies on the subject of ESG. Companies tend to think that they have gone to the trouble of completing their sustainability disclosure reports and that investors do not seem to take them into account. Investment analysts tend to think that the disclosure is on issues that they do not deem to be sufficiently material or that it is insufficiently quantifiable. There is a real need to improve this dialogue by achieving a consensus on the part of the investment community as to which issues or factors are significant for them and then communicating this to reporting companies. ESG is high on the agenda of most institutional investors today. The GSIA 1 reports that global “sustainable assets” now stand at $21.6 trn, up by 61% in the last 2 years. These ESG assets now represent 59% of all professionally managed assets in Europe. To facilitate the selection of ESG assets requires, there is a need to focus on the most material Factors and Metrics by industry, which Delphi did. The Delphi framework was designed to give clearer focus on the most meaningful ESG factors by industry 2 classification, as defined by a committee of experts on ESG investing and specialists on particular sectors. The number of factors and metrics highlighted for each sector is relatively small (an average of 14.5 factors per sector). Thus, reporting companies are able to focus on the small number of factors identified as being meaningful to the investment community. Today, ESG goes beyond ethical and socially responsible investment and involves a more dynamic process. It consists of incorporating ESG factors into investment decision-making, thereby recognizing the materiality of ESG issues, whether positive or negative. The next generation of ESG investing involves harnessing the ESG information to improve the investment outcome, in other words alpha generation and risk management. The alpha signal is integrated into the investment model or the portfolio construction. It is this next generation ESG analysis that is primarily at the heart of Project Delphi. 1 GSIA – Global Sustainable Investment Alliance (2015), Global sustainable investment review 2014, January available at http://bit.ly/1FSoj9C (accessed: 13.01.2016) 2 Using ICB classification, available at http://www.icbenchmark.com/
- 4. INTRODUCING PROJECT DELPHI 3 We seek to explain the gap between market value and book value which now, on average, represents 80%+ of company value 3,4,5,6. . This ‘unexplained value’ gap emerges from a misalignment of accounting convention which fails to accommodate a range of relatively new and value generating intangible assets, and investment market convention which increasingly capitalizes expected future earnings growth. This is a gap where ESG factors are either a significant contributor to or defined risk for both intangible asset performance and earnings growth trajectory. In either case, corporate managers and investors /analysts are having to take decisions based on incomplete information. Companies do report on ESG Factors, either in the Annual Report or in the CR Report, but there seems to be a serious disconnect between what is reported and what analysts would consider important and usable information. The most common corporate reporting frameworks are not derived from an investor perspective and do not necessarily require companies to report data that investors find material or useful. In many cases, the data is not standardised and may be of questionable quality. In addition, the data sets are still relatively short. And current accounting and corporate reporting is almost exclusively based on backward looking information, making it difficult to provide usable information on intangible assets and their impact. Are management and investor decisions being taken on the basis of subjective estimates of what the future value of companies will be and with almost no guidance as to what the impact that such decisions will be? Definitions The Delphi definition of ESG: Within the scope of Project Delphi, ESG (Environmental, Social and Governance issues) is defined as relating to: Those factors relevant to the understanding of a company which help to assess the company's capacity to remain productive over time AND its potential for long-term maintenance of profitability through: the responsible use of natural resources both in their own operations and the operations of their respective value chains; respecting social norms in their markets of operation and those markets where their products and services are in use; being accountable to providers of equity and debt capital. Delphi on Materiality Analysts tend to use established patterns when making valuation decisions. Many ESG factors do not make it into these patterns, because materiality may not have been adequately established or, quite simply, because these factors have never played a role in traditional analysis. “As the complexity of the fund manager's decision task expands exponentially, knowingly or not, most fund managers jettison the classical decision making model in favour of pattern recognition. If, however, extra-financial factors have not played a part in prescribing the patterns that fund managers consider worthy of attention during their professional development, they are unlikely to appear on a fund manager's radar screen in the future - unless, that is, they are alerted to their materiality to business and stock price performance by a trusted source of 3 Daum, J. H. (2003). Intangible assets and value creation. John Wiley & Sons. 4 Lev, B., & Daum, J. H. (2004). The dominance of intangible assets: consequences for enterprise management and corporate reporting. Measuring business excellence, 8(1), 6-17. 5 Ballow, J., Thomas, R. J., & Roos, G. Future Value: the $7 trillion challenge, Accenture 2004. 6 Tomo, O. (2015). Annual Study of Intangible Asset Market Value from Ocean Tomo, LLC.
- 5. INTRODUCING PROJECT DELPHI 4 information and advice.” 7 In the absence of clearly demonstrated materiality of non-financial indicators, investors tend to focus on a relatively narrow set of indicators and to use simple earnings based heuristics when making their recommendations. 8 Nevertheless, there clearly is a role played by factors other than the traditional indicators of financial performance in stock evaluations. There is a large percentage of stock values which cannot be explained by fundamental or tangible criteria. Correlations are notoriously difficult to identify in this respect. The process is imperfect and therefore leaves opportunities open for investors with more encompassing patterns. Investors, consultants and analysts have a fiduciary duty to take account of “material” factors, whether they be tangibles or intangibles, fundamental criteria or sustainability ones. More efficient capital allocation and better alignment with the interests of end beneficiaries requires this. There is mounting pressure, both on the part of regulators and long-term investor institutions to take these considerations into account as a growing body of evidence is tending to establish the materiality of certain ESG factors. Materiality, then, is the central issue here and it is the very cornerstone of our Project Delphi. Within the scope of Project Delphi, an ESG factor is considered material if it: has a high probability of having a financially material impact on the defined value drivers of investment value of Delphi within a given time horizon, and allows prioritization of issues for investment decision-making and/or corporate engagement, and is based on research or sufficient evidence on its impact, and is rigorous i.e. replicable, defensible, and credible, and practical i.e. simple enough to be widely used and communicated, and sophisticated enough for the needs of investment decision-making including modelling and calculation e.g. of DCF, NPV etc. Project Delphi started off with 250 ESG factors. Many of these may be considered to be values based and of importance from an ethical standpoint, but whose financial impact may be less clear than that of other ESG factors. We refer to these latter as “material ESG factors”. By focusing on the materiality of these factors, we underline the necessity for companies to report reliably on this reduced number of factors. Delphi on Granularity and Parsimony The Delphi process followed a principle of granularity - the level of detail required to depict a given industry and its specifics - matched with a principle of parsimony - i.e. striving to find the lowest possible number of metrics within an industry. These principles of parsimony and granularity are characteristics that distinguish Delphi from other existing projects. We consider a good framework as giving a limited and fairly precise picture, even for a complex issue such as ESG integration. A framework fails to meet the parsimony criterion if it does not focus on only a few factors, or is accompanied by lengthy publications, to which we prefer webinars. As a result, there were metrics or factors that were initially included in the framework but were removed by consensus for reasons of parsimony. 7 O'Loughlin, James (JOL Consulting) and Raj Thamotheram (USS). “Enhanced Analytics for a New Generation of Investor. How the Investment Industry Can Use Extra-Financial Factors in Investing”, May 2006. 8 Bradshaw, Mark T "How Do Analysts Use Their Forecasts in Generating Stock Recommendations?" Harvard University, December 2000)
- 6. INTRODUCING PROJECT DELPHI 5 Who is behind the Delphi Framework? The Delphi Project was initiated by institutional investors who recognize the need to develop a tool for ESG integration. Initial discussions took place between Michael Polya and John Swannick and the other leaders of a collaborative project called “Valuing Non-Financial Performance”, which had been set up under the Enterprise 2020 initiative. Ralf Frank, MD of the DVFA, the German federation of Financial Analysts and Director of EFFAS (European Financial Analysts Society), agreed to participate and to head the “Metrics” workstream. The DVFA, under Ralf Frank had already drawn up an impressive compendium of ESG KPIs, which has served as inspiration for much of the Delphi work. 9 Chris McKnett, Head of ESG at State Street Global Advisors, agreed to head the “Asset Managers” workstream and Frank Curtiss, Head of Governance at Railpen pension fund, took on the role of head of the Asset Owners’ workstream. Richard Lacaille, Global CIO of State Street Global Advisors, kindly agreed to be the sponsor of the project. Michael Polya agreed to coordinate the project as Project Manager. A Steering Committee was established, comprising representatives from the various groups of stakeholders in the project (Asset Managers, Asset Owners, Consultants, Corporates and academics). All participants were involved pro bono. A Technical Group was set up to reach a consensus on the KPIs based on the proposal of asset managers and asset owners. The process was observed and documented by an academic. The members of the Technical Group and the asset manager and asset owners groups are as follows: Technical Group: Michael Polya, SSGA, Christopher McKnett, SSGA Neil Brown, Alliance Trust, UK Daniela Carosio, Sustainable Equity Value, IT Cécile Churet, RobecoSAM, Switzerland Ralph Frank, DVFA / EFFAS, Germany Hendrik Garz, Sustainalytics, Germany Andrew Howard, Didas Research, UK John Swannick, Consultant, UK Stéphane Voisin, Kepler Cheuvreux, France Niamh Whooley, SGCIB, UK Christel Dumas, ICHEC Brussels Management School and University of Gent – Belgium (ex officio) Other organisations who participated are: State Street Global Advisors DVFA EFFAS Railpen Investments Axa IM Didas Research Alliance Trust Aon Hewitt Schroders Aegon Pension Protection Fund Kepler Cheuvreux KKR Mercer Consultants 9 Garz, H., F. Schnella, and R. Frank. "KPIs for ESG. A Guideline for the Integration of ESG into Financial Analysis and Corporate Validation. Version 3.0, Frankfurt, DVFA/EFFAS, 2010." http://www.effas- esg.com/wpcontent/uploads/2011/07/KPIs_for_ESG_3_0_Final.pdf
- 7. INTRODUCING PROJECT DELPHI 6 SG Corp & Investment Bank Aberdeen Asset Management F&C APG KLP, Norway Nestlé Deutschland DNB (Dutch Central bank) Towers Watson Hymans Robertson BofA Merrill Lynch RobecoSAM Sustainalytics First State Investments SAM Research AG Storebrand PGGM (consultative role) Cambridge Associates SEB Goldman Sachs (GS Sustain) Royal London Asset Managers ERAFP Varma, Finland Sustainable Value Investors Sustainable Insight Capital Management WHEB Asset Management AWJ Capital Partners The Delphi Process A key aspect of the Delphi Project is that its framework derives from a consensus. This consensus, resulting from members’ representations of ESG, is what gives the framework its strength and legitimacy. To reach this consensus between the Delphi parties on a short-list of the most significant ESG factors per industry, and the appropriate metrics for these factors, the Delphi Project adopted an iterative process. This process was a constant back and forth between opposite perspectives on each issue. We know that such an enterprise is not like conducting a controlled experiment in the natural sciences. As discussed, the data is not robust enough to allow such certainty. Nor, given its nature, is it ever likely to be. Even traditional financial analysis is not immune from inevitable approximation. So, we are left with the option of leveraging the opinions of experts and looking for indications in the data available. Such subjectivity can lead to tensions. Do we include only factors and metrics on which companies report today? Do we include only factors that are quantifiable? What is the time horizon that we should apply when assessing the material impact of a factor? Questions such as these, and many others, were vigorously debated and in most cases, a consensus view was arrived at. In other cases, it was agreed that the issue should be parked temporarily and that we would have to live with certain paradoxes. The paradoxes were revisited frequently and usually a pragmatic solution was found. For example, it was decided to classify the metrics with a color code: Green – reported by a reasonable number of companies; Blue – requires 3 rd party sources; Red – Reported on by a small number of companies. We were conscious of the fact that some metrics are “measurable” but not reported on because of concerns about confidentiality or other such reasons. It was felt that if a metric was considered important, we should include it, even if reporting on it was thin, in order to send the message that investors were interested in it. The variety of tasks Delphi undertook included structuring the information, with a structure inspired by existing frameworks, and adapted to the needs of the group. A “metrics team” provided a first framework of “value drivers”, “factors” and “metrics” by industry, based on participants’ individual expertise, with a strong EFFAS 10 influence. The asset managers work stream then provided a second framework of value drivers and factors, inspired by a McKinsey framework 11 . 10 http://www.effas-esg.com/?page_id=206 11 Valuation : Measuring and Managing the Value of Companies, McKinsey & Co Inc. – Tim Koller, Marc Goedhart, David Wessels, June 10, 2010
- 8. INTRODUCING PROJECT DELPHI 7 The group soon agreed on adopting a “Value Driver” or “Value Lever” approach which anchors the concept of financial impact into the framework. The three Value Levers from which the analysis derives are: Growth, Return on Capital and Governance and Risk Management. It was essential that the logic should flow from left to right: from Value Lever to Value Driver to ESG Factor and into the metrics and be specific to each of the 10 industry sectors. This being said, the Delphi framework has a starting point which is not a number of ESG factors and their application to the 10 industry sectors. Rather it has a Business Model focus which is grounded in the reality of how businesses function and seek to create value. This is where its strength lies. Put another way, instead of starting from the myriad of factors typically considered “ESG factors”, our analysis started from an assessment of the drivers of growth and value, through the universal value driver model we have relied on. The project’s focus was primarily on understanding how companies’ management of environmental, social and governance issues impact elements of that model in each sector, and tools available to analysts to better identify companies that are managing those issues most effectively or poorly. As such, the framework leads with the topics most relevant to investors – value creation – and ends with ESG metrics rather than the more conventional reverse approach. Determining parameters that fit in the structure, and giving boundaries by excluding items were additional steps that led to the creation of a framework. In parallel, the group produced a taxonomy which facilitated common language, and defined concepts that were used to build the framework, such as the concept of ESG, or materiality. These definitions were written down and approved by the group, building up documentation for the project. Once the structure was in place, the group had to determine the materiality of each factor, per sector, drilling down into the framework with a top down approach, from the more general factors to the specific metrics. This also required reducing the number of factors and clustering factors together, parsimony being a key word for the group throughout the process. When determining whether or not a factor or metric is material, we applied a rule of thumb for a first screening. This consisted of asking three questions: 1. What is the size of the financial impact? 2. How probable is it that this impact will occur? 3. Over what time horizon will this impact occur? If the impact is assessed to be high, the probability of occurrence is strong and the time horizon is relatively short, the item was deemed to be material. If doubt persisted on the answers to any of these three questions, the issue was subjected to further discussion. Project management was a key and challenging task throughout the process. Delphi was first organized in three workstreams: the metrics group, the asset managers, and the asset owners, plus a steering committee overseeing the project. After this initial dialectic approach it was decided to set up a “Technical Group” consisting of members of the three workstreams with technical expertise in either standard-setting or ESG valuation, in order to speed up the process of arriving at consensus. Finally the group activities required seeking validation internally, for example by developing a narrative to justify each metric, or externally by presenting the framework to colleagues and potential partners. A series of webinars was conducted at which the framework was presented and commented on by members of the Technical Group. The audience for these were 1) the larger Delphi family 2) Friends of Delphi with long experience
- 9. INTRODUCING PROJECT DELPHI 8 of the field of ESG investment 3) Specific target groups (asset owners, corporates, asset managers, academics, consultants). Feedback was elicited and comments integrated into the framework. To summarise this section, these different activities are illustrated in the Delphi process time line, in the Appendix 2. They contribute to four characteristics that make the uniqueness and value of the Delphi process: 1) Embracing the paradoxes. From the start, be it in its mission statement or in the constitution of workstreams, Delphi adopted an ‘either/and’ view of contradictions, in which the opposites both continue to exist and can be united in an arrangement. 2) Exploring tensions. The process required the group to dive into each dimension, such as Governance, Climate Change and Supply Chains, again and again. We have looked at the choice of factors and metrics through multiple lenses: long-term vs short-term, E, S & G, Capitals, forward looking vs backward looking, exposure vs performance, disclosed vs non-disclosed, inductive vs deductive are but a few examples of the thematics discussed. 3) Engaging in social interaction. An important part of the process consisted in socializing, at many different levels with different actors in the ESG field. This required building a shared understanding of the issues, for example with taxonomies, and definitions. 4) Taking the time. As suggested by all of the above, this process took time. The group decided it wanted to do a good job rather than a job on time. With an army of volunteers working on the project this meant prolonging the deadlines. We believe taking the time is a necessary condition for a process to respect paradoxes. A very different result could be reached under time pressure and strict deadlines. What makes Delphi different from existing ESG guidelines? We believe that what gives Delphi its strength is that: It is based on Financial Value Drivers, linked to enterprise performance It is investment industry led – by investors, with investors, for investors It is a collaborative venture (45 organisations), representing the collective intellectual property of leading industry figures through iterative dialectic. It is Parsimonious. We have sought to identify the most significant ESG factors and metrics in terms of their financial impact. It is predictive with emphasis on future performance not just historic reporting It is dynamic. It is intended to set up an interactive platform to enable regular updating
- 10. INTRODUCING PROJECT DELPHI 9 Who can use the Delphi Framework? The Delphi framework, as it exists at the time publishing, will be available on an open website. It will enter the public domain. It is intended to: Provide guidelines for Analysts and Researchers to scope the material issues by industry and enable them to perform the analysis and research necessary to inform the valuation approaches and investment models of asset managers. Promote greater recognition of the critical ESG issues by Asset Managers and to provide the information and data they need to review their valuation approaches and integrate ESG factors into their investment models. Increase the understanding of Asset Owners and Consultants of the value implications of ESG factors and the priorities they should be setting for asset managers. Assist Corporates to identify the most material factors and metrics on which they should be reporting in order to align with the data disclosure needs of investors. This should better enable their investment relations people to understand the importance of disclosure of the material issues. The Future of Delphi? Having achieved consensus on the content of the Delphi Framework by all members, the technical group has developed a narrative accompanying the Framework and will launch it as an open-source tool for ESG integration. Following the launch, two processes are expected to commence. First, asset managers interested in ESG integration will be able to develop investment strategies focusing on the limited list of metrics, and asset owners will be able to engage with companies and with asset managers based on the metrics. Second, data providers would be able, should they so wish, to align the information populating their datasets with the metrics promoted by the Delphi Framework. The Delphi Framework does not aim to constrain the competitive edge of asset managers who hold expertise in generating investment models. Hence, the Framework is flexible enough that people can use it to build their own models. The Framework purely focuses on metrics and Factors that are material in terms of expected financial risk and return. It does not recommend investment strategies, nor does it provide non-material, values based Factors. Anyone interested in developing specific tools internally can do so based on the Framework, but will have to go beyond the information provided in the Framework. The Delphi Group is conscious of the relative unreliability of the data collected on ESG and the short life-span (5-10 years) of the data sets available. The dynamic nature of the framework is designed to accommodate this. It is intended that the framework will be set up on an interactive website and endorsement and suggestions will be elicited from a very wide audience from the investment and corporate world, as well as other interested groups. In this way, the framework will undergo continual improvement and updating, whilst still retaining its focus on the nexus between the investment and corporate communities and the financial impact of the factors.
- 11. INTRODUCING PROJECT DELPHI 10 We would also like to work with the major data providers in order to help them focus their data collection on the most material factors by sector, as defined by the updated Delphi framework. Through the combination of continual updating by the “crowd” and collaboration with data providers, we hope to improve disclosure and the reliability of the data over time. Conclusion With the rise of ESG requests from the investment community comes the need to develop tools and techniques to value non-financial performance. This is what led the Delphi project to develop a framework of KPIs for ESG: a short-list of the most significant ESG factors and the appropriate metrics for these factors. The 10 to 15 material metrics per sector are currently available in an open source framework. This framework will evolve as the ESG market evolves, affecting the materiality of metrics. We see the process followed by Delphi in generating factors and the debates Delphi has had on each of the paradoxes it encountered in ESG, as a real contribution to a better and more complete dialogue taking place between investors and reporting companies on the subject of sustainability. As one Delphi participant stated: “The spreadsheet is useful, but the discussions it makes us have with analysts are brilliant. This is the big change, the better conversations we are having.”
- 12. INTRODUCING PROJECT DELPHI 11 Appendix 1 Taxonomy – Project Delphi As a preliminary to categorizing ESG factors, elements, aspects etc. a coherent and analytically rigorous framework is necessary. Definitions: i. ESG factor: An ESG factor is defined as any aspect that relates to environmental, social or governance issues or shows impacts in the environmental, social or governance performance of a company and potentially or effectively has a monetary or direct impact on the company value, its fundamental performance (earnings), its reputation, its share price or its longevity. ESG super-factor: An ESG super-factor is defined as an ESG factor which shows significant financial impact on a company. These factors may be sector specific. ESG factors and superfactors are presented in the format of a label or a short description e.g. carbon footprint, energy consumption, supply chain management, innovation. They are not identical with a performance indicator or KPI but merely describe an issue, topic or aspect. In the taxonomy of Delphi factors belong to the highest level of abstraction. Factors always need to be translated into one or more quantified elements which can be ESG line items, metrics or ratios. When a line item, metric a ratio is defined i.e. standardized as the leading measurement for a factor it becomes a KPI. While factors apply to many industries (some to all), line items, metrics and ratios may differ from industry to industry. Examples of ESG factors are carbon emissions, energy intensity, supply chain management, innovation. ii. ESG line item: ESG line items are defined as raw data i.e. items as reported by companies or investors which reflect direct performance in absolute terms. Examples of ESG data points are e.g. electric energy consumption FY 2010: 3,456 terajoules, staff turnover FY 2011: 4.5%, fines for non-compliance with safety regulations in €k: 678.
- 13. INTRODUCING PROJECT DELPHI 12 Appendix 2: Time-line of the Delphi Process