RELATIONSHIP BETWEEN CLIMATE CHANGE STRATEGY, RISK MANAGEMENT, CARBON ACCOUNTING AND CORPORATE FINANCIAL PERFORMANCE

................................................................................................................ VII ACKNOWLEDGEMENTS ......................................................................................... IX CONTENTS .................................................................................................................. XI LIST OF TABLES .................................................................................................... XIII LIST OF FIGURES .................................................................................................... XV LIST OF ABBREVIATIONS AND ACRONYMS ................................................. XVI LIST OF APPENDICES ......................................................................................... XVII


INTRODUCTION
Climate Change has been widely acknowledged as one of the major sources of risk by the global community. As a natural phenomenon, is has been associated with global temperature rise which directly affects global meteorological patterns. However this global temperature rise is not considered a natural phenomenon but it is rather being attributed to human activities, which during the last one hundred years have resulted in the rise of Greenhouse Gases (GHGs) emissions, and particularly Carbon Dioxide (CO 2 ), at concentration levels that are beyond normal. Taking as a reference the current flow of CO 2 emissions, it has been estimated that there is a chance ranging between 77% and 99% that the global temperature will rise more than 2°C in the next twenty years (Stern 2006). Such an increase in global temperature is most likely to cause rapid changes to current climatic models affecting directly the natural environment. However, the risks climate change poses are not confined strictly to the environmental and direct physical impacts of global temperature rise but they also involve social, economic and financial impacts.
The first agreement on climate change, the creation of the United Nations Framework Convention on Climate Change (UNFCCC), took place in the 1992 Rio Earth Summit. The aim was to achieve an agreement on the stabilization of the GHG concentrations in the atmosphere at a level that would not pose a threat to the climatic models. However, the main driver in the development and adoption climate change mitigation policies was the creation of the Kyoto Protocol Haque and Deegan, 2010), which went into effect in 2005 and forms a legally binding agreement that sets emission reduction targets for industrialized countries. The need to reach Kyoto targets gave rise to the adoption of national and international climate change mitigation policies. These include putting a price on carbon emissions in the form of carbon taxes, mandatory process and product standards or by establishing carbon trading programs (Bebbington and Larrinaga-Gonzalez, 2008). As a result, GHG-intensive activity sectors will face direct regulation risks in the form of increased costs, related to their obligation to have GHG emission allowances matching the amount of their emissions, which will eventually affect their profitability.
Specific economic sectors, such as agriculture, fishing, forestry, insurance, real estate and tourism, that are particularly exposed to the direct physical impacts of climate change due to their dependency on the natural environment, will face increased physical risk (Wellington and Sauer, 2005;Lash and Wellington, 2007). On the other hand, GHG intensive business sectors such as oil and gas, aluminum, cement and pulp industries are more exposed to risks associated with current or impending regulation regarding the reduction of GHG emissions (ABI, 2005;Busch and Hoffmann, 2007). Finally, non-intensive business sectors will most likely face the indirect effects of climate regulation through increased energy prices, which could affect their production costs.
These increased risks that companies face have caught the attention of various stakeholders, such as institutional investors, banks, accounting firms, governmental agencies, NGOs and consumers, who have been demanding information disclosure regarding the corporate climate change practices of firms. On the financial front there is evidence that institutional investors view climate change as a major source of risk (Solomon et al., 2011). Consequently, they have begun to pay more attention to the corporate climate change practices of firms and to demand information regarding corporate climate change mitigation policies. The reason why institutional investors insist on companies disclosing information, especially those belonging to GHG-intensive sectors, is that inadequate disclosure of climate change related risks could reduce an investor's ability to estimate a firm's performance and future cash flows (Venugopal et al., 2009). Finally, the need to accurately measure a company's GHG emission profile has given the accountancy profession a new role regarding the physical and financial accounting of carbon emissions and emission allowances. Therefore, accounting firms have also stressed the importance of climate change related corporate disclosures and place special attention on the accurate and precise accounting and reporting of GHG emissions.
The aim of this study is to examine the various climate change management practices adopted by firms and how these practices affect their financial performance. We are going to examine three dimensions of climate change management: Strategy, Risk Management and Carbon Accounting and Reporting. We claim that devoting corporate resources in the implementation of climate change strategies, in the identification, assessment and management of climate change risks and in the accurate and precise accounting and reporting of carbon emissions, enhances the financial performance of firms. We also demonstrate that this relationship between corporate climate change practices and financial performance has the tendency to become stronger throughout the years. Finally, we claim that reducing carbon emissions enhances a company's financial performance. In order to prove the validity of our claims we have developed specific climate change corporate indexes that measure the level of corporate commitment regarding climate change strategy, risk management and carbon accounting and Reporting and Advanced Carbon Accounting Practices. We will use the above indexes to assess the level of corporate commitment of the largest companies in the oil and gas sector and in the banking sector. In order to measure the level of corporate commitment regarding climate change practices we weighed the corporate indexes according to the level of difficulty in implementing the climate change practice included in each respective index. In order to assign proper weighs to our indexes, we developed and validated critical factors of climate change strategy, climate change risk management and carbon accounting by using respective survey instruments. Three questionnaires were developed according to extensive literature review, which were sent to three different survey target groups. Regarding carbon emission reductions we used CO 2 emissions data as provided by the Carbon Disclosure Program (CDP). Finally, we examined the relationship between our indexes and the financial performance of the largest companies in the oil and gas sector and in the banking sector.
This study contributes to the discussion on the need of further enhancing corporate commitment of firms towards the implementation of climate change practices by identifying specific climate change strategies, risk management and carbon accounting practices that are most likely to have a positive effect on corporate financial performance. International academic research has also studied the relationship between environmental management and corporate performance. Several indicators have been used to measure corporate commitment regarding environmental management, sustainability or corporate disclosures. However, majority of this research focuses on general environmental management or corporate sustainability. Very few studies have attempted to address the issue of climate change management separately from the general concepts of environmental or sustainability management. Furthermore, majority of research on climate change and its relationship to business management is theoretical. There is little empirical research that examines the effect of climate change of corporate performance. Moreover attempts to measure climate change corporate performance have focused on developing aggregate indexes.
Thus the novelty of this study lies first to the fact that it focuses only on climate change and its relationship to corporate performance. Furthermore, we do not use aggregate indexes to measure climate change corporate commitment. Instead, we examine three dimensions of climate change management, strategy, risk management and carbon accounting and reporting.
Moreover, we develop different indexes to measure each climate change management dimension, which allows us an in-depth analysis of the various climate change practices of firms. Indexes were developed by using data collected with the help of three survey questionnaires which were sent to climate change experts in three different fields: strategy, risk management and accounting. For the questionnaires used to develop critical factors regarding climate change practices, the target sample included climate change experts who are either currently employed in companies included in the Financial Times Global 500 (FT 500) list or work as external partners with these companies. Consequently, this study contributes to international research by exploring and analyzing patterns of corporate responses related to strategy, risk management and carbon accounting, and how they affect the financial performance of firms.
The structure of this study is the following: in the next chapter we are briefly going to discuss some basic scientific evidence regarding climate change as a natural phenomenon, as well as the mitigation policies that have been adopted and regulations that are current in place. The reason we go through this stage is because we believe that increased corporate commitment to climate change management during the last ten years has been the result of the creation of an international climate change regime involving climate change mitigation policies and regulation at national and international level. Therefore we consider it important to briefly discuss climate change policy and regulation. Next, we will examine the effects of climate change on business activities and analyze why there is a need for companies to develop climate change management practices. Then we will proceed with a review of previous academic research related to climate change strategy, risk management and carbon accounting and reporting. Along with the relevant literature review, we will develop our initial climate change indexes, which we will empirically test for validity and reliability. To do so, we will conduct an extensive survey, during which, climate change experts will assess the difficulty level of implementing the specific climate change practices included in each respective climate change index. After the validation of the indexes is completed, we will use them to assess the level of corporate commitment of selected firms regarding their climate change strategy, risk management and carbon accounting and reporting practices, and examine their relationship to financial performance. The above procedure and the results of our empirical analysis will be presented in the methodology section. Finally we will provide a discussion of our findings, outline the limitations of our research and present propositions for future research. 23

CLIMATE CHANGE AND MITIGATION POLICIES
The climatic conditions on Earth are determined by the continuous flow of energy from the sun. Solar thermal energy penetrates the Earth's atmosphere and warms its surface. As the temperature increases on the surface, the Earth itself sends, in the form of infrared radiation, thermal energy back from its surface into the atmosphere. A portion of this energy is absorbed by gases in the atmosphere -such as carbon dioxide, methane, nitrous oxide etc. (commonly known as greenhouse gases GHGs) -thereby causing this energy to be trapped into the atmosphere and maintain the average temperature of the earth at about 15 ° C. These temperature levels are necessary for maintaining life on earth for humans, plants and animals.
Without these gases, the Earth's temperature would be fixed on -18 ° C, making our planet uninhabitable for most life forms. Carbon dioxide (CO 2 ) is the most important of GHGs for the maintenance of the desired temperature levels on Earth. The natural CO 2 emission processes and the CO 2 absorption processes are responsible for maintaining a balanced concentration of CO 2 in the atmosphere. Through plant and animal decomposition and respiration, volcanic eruptions and ocean release, CO 2 is released into the atmosphere and it is then reabsorbed by photosynthesis and dissolved in water (Figure 2.1).
Source: Cooperative Research Centre for Greenhouse Gas Technologies (CO2CRC), Australia

Figure 1: The greenhouse effect
The cause of climate change is the excessive use of fossil fuel resources, such as coal and lignite, oil and natural gas, the combustion of which releases vast amounts of CO 2 into the atmosphere. In this way the 'blanket' of GHGs which covers the Earth holds more and more energy which, in turn, increases the average temperature of the planet. Human intervention in the natural CO 2 cycle over the last 150 years has been decisive. The extensive burning of fossil fuels, the intense livestock farming which contributes to increased methane emissions, coupled with deforestation which decreases Earth's absorbing capacity of GHGs, they have irreversibly disturbed the balance in the CO2 cycle. Negative impacts on ecosystems and human populations are already being observed-such as the melting of sea ice in the Arcticeven with the current increase in temperature at 0,8 ° C compared to pre-industrial levels.
These impacts could potentially lead to even more global warming and further environmental, social and economic implications.  Figure 2). Furthermore, it is clear that global warming does not necessarily mean a warmer climate for everyone, in all regions of the world. As the planet's temperature increases, the climate system changes and this results to the increase of extreme and unpredictable weather. Other areas will be become warmer, others colder, while global humidity levels with be affected accordingly resulting either to increased drought levels or excessive amounts of rainfall.

Figure 2: Deviations of global annual mean temperatures
The 4th Assessment Report of the IPCC further stresses that: 1. In the coming decades, water supplies stored in glaciers and snow-covered areas will be reduced, causing water shortage for more than 1 billion people 2. Between 20% and 30% of all living organisms on the planet will be at increased risk of extinction if the rise in global average temperature exceeds 1,5-2,5 ° C.
3. At lower geographical latitudes, especially in dry and tropical regions, even a small temperature increase of 1 ° C -2 ° C, is expected to endanger food provision security.
4. After 2080 millions of people will be affected by floods because of the rise in sea level every year. At particular risk will be densely populated areas and areas which are at a low altitude and with limited adaptation capacity. The current regional climate change risks related to physical, biological and human systems are presented in Figure 3. Regarding the physical risks of climate change and their implications to natural ecosystems, these are summarized in Table 1.  Emission Reductions (CERs), that they may later use to achieve their emission targets or trade in ETSs. JI is the same as CDM but instead of investing in developing countries, industrialized countries must invest in those countries whose economies are in a transition phase such as the ex Soviet bloc (Chasek et al. 2013). Finally, the ETS allows the buying and selling of emissions credits among industrialized countries in order to meet their emission reduction commitments.
The need to reach Kyoto targets gave rise to the adoption of national and international climate change mitigation policies. These include putting a price on carbon emissions, through carbon taxes, the establishment of carbon trading programs, setting mandatory process and product standards for industry or provide incentives to invest in lowcarbon technologies (Bebbington and Larrinaga -Gonzalez 2008 involved initial free allocation of emission credits for most industry sectors and a penalty of €100 for each tonne of emissions for installations that did not have the required allowances, Phase 3, which is currently at place, runs from 2013 to 2020, will involve auctioning of the initial emission allowances with only a small portion being allocated for free (Lovell et al., 2010). Furthermore, the cap of emission allowances will gradually decrease each year, in order for the EU to achieve its emission reduction targets.
However, although near-term, moderate emission reduction goals can be achieved with the use of such economic price instruments, ambitious emissions reduction targets in the future, will be very hard if not impossible to achieve without the pervasive diffusion of low-carbon technologies (Sandén and Azar, 2005). Diffusion of renewable energy sources (RES) in the electricity sector could provide the basis for achieving mass reductions in CO 2 emissions in the long-term future. In order to reduce its overall CO 2 emissions, the EU has set a target of 20% final energy consumption based on RES. Furthermore EU leaders have committed to a 80% -95% reduction of CO 2 emissions at 1990 level by 2050 which, as it has been estimated, will not be possible unless a 95-100% decarbonisation of the electricity sector is achieved by 2050 (EU, 2011). Building on the above, the need for technological, institutional and social transition of the electricity system towards the use of low carbon technologies in electricity production is of utmost importance.
As we have discussed above, the negative impacts climate change on ecosystems and human populations are already being observed. Along those lines, the urgent need to reduce GHG emissions has resulted in adoption of the climate change regulations at national and international level. These include putting a price on carbon emissions in the form of carbon taxes, mandatory process and product standards or by establishing carbon trading programs (Bebbington and Larrinaga-Gonzalez, 2008). Consequently, companies, who operate in this climate change regime, are also expected to take into account the negative implications of climate change on their business activities, and implement corporate climate change practices that address climate change risks, and adopt corporate climate change mitigation strategies. As we will discuss in the next section, specific economic sectors are directly influenced by climate change, either through its physical consequences on corporate activities or through climate change regulation. Energy production companies, for example, are particularly exposed to climate change regulation, as they are among the highest CO 2 emitting sectors. On the other hand, companies who do not belong to GHG intensive sectors are also expected to be influenced by climate change regulation. For example, as there is a tendency for energy suppliers to pass costs of regulation to consumers through increased energy prices (Lass and Wellington, 2007;Wellington and Sauer, 2005), companies, that utilize low energy-efficiency technologies in their production process, will also face increased operational costs.
In the next section we will discuss the effects of climate change on business. We will then examine the various strategies that business adopt in order to reduce their exposure to the adverse effect of climate change. We will also analyze the various climate change risks and how they affect corporate activities. Finally, we will analyze the corporate practices related to the accounting and reporting of carbon emissions, which is considered of outmost importance in assessing the climate change exposure of a company.

3 CLIMATE CHANGE AND BUSINESS
As climate change affects both directly and indirectly corporate activities, some companies will face more risks than others depending both on the nature of their business, and on the strategies they adopt regarding climate change (Porter and Reinhardt, 2007;Schwartz, 2007). Climate change and GHG reduction policies create systemic risks across the global economy, by affecting energy and food prices, national income and health expenditures. On the other hand they create direct and indirect risks at sector and company specific level (Hoffman, 2006;Wellington and Sauer, 2005;ACCA, 2009;Ceres, 2010a). Most of these risks have been identified even partly estimated by insurance companies, which have become increasingly concerned about climate risks and how they affect both firms' physical assets, potential regulatory costs and even litigation costs regarding corporate environmental liabilities (ACCA, 2009).
Specific economic sectors, such as the agricultural sector, fishing, forestry, real estate and tourism are particularly exposed the direct physical impacts of climate change due to their dependency to the natural environment (Lash andWellington, 2007, Wellington andSauer, 2005 ; ACCA, 2009) therefore facing increased physical risk.
On the other hand, GHGintensive or energy intensive business sectors such as energy -production, aluminum, cement and pulp industries are more exposed to risks associated with current or impending regulation regarding CO 2 emissions (Busch and Hoffman, 2007;ABI, 2005). As a result, some companies will most likely face increased operational costs as we move towards a carbon constrained economy (Busch and Hoffman, 2007). These costs will most certainly affect their profitability and create significant competitive risks not only for GHG intensive industries such energy production but also for the majority of industrial manufacturing sectors, as there is the tendency for energy suppliers to pass the increased costs of regulation to their consumers (Lash and Wellington, 2007;Wellington and Sauer, 2005). With the dramatic increase in climate change impacts, there is a pressing need for business to develop appropriate climate change mitigation strategies for the risks posed by the projected climate change policies.
On the financial front there is evidence that institutional investors view climate change as a major source of risk (Solomon et al., 2011) and have also began to pay more attention to corporate climate change practices of firms, demanding more information disclosure from companies regarding their corporate climate change mitigation policies. If a company is not seen as adequately addressing and managing physical or regulatory risks posed by climate change, investors might be reluctant to invest in the company until it has developed a specific corporate climate change strategy (ACCA, 2009). What investors are particularly interested in, when assessing the climate risk profile of a company, is information regarding GHG emissions, exposure to carbon regulations, such as taxes or mandatory product and process standards, or corporate strategies regarding renewable energy investments (Solomon et al., 2011;Delloite, 2007). Mainstream institutional investors, for example, like Goldman Sachs, Bank of America, JP Morgan and Citigroup have already adopted sustainability criteria such valuating GHG -intensity of projects and promoting low -carbon technologies such as renewable energy (Hoffman, 2006;ACCA, 2009 The reason why various stakeholders and institutional investors in particular, insist on companies disclosing information, especially those belonging to GHGintensive sectors, is the fact that inadequate disclosure of climate change related risks will reduce the investor's ability to accurately measure and manage the exposure of his portfolio to the reverse effects of climate change (Venugopal et al., 2009) therefore causing him to falsely estimate firms' performance and future cash flows. Moreover, academic research on credit risk management has shown that by incorporating sustainability criteria in financial assessment of projects, for example, banks can not only lower the chances of having to deal with credit default cases but have also more chances of gaining an advantage against their competitors by improving their credit risk management assessment procedures (Weber et al., 2010;Weber et al., 2008;Nitsche and Hope, 1996;Thompson, 1998).
Additionally, accounting firms have also stressed the importance of climate change related corporate disclosures and place special attention on both the physical and financial accounting of GHGs. Accountants were initially involved with climate change reporting issues due to the creation on the EU ETS (Lovell and MacKenzie, 2011) and the financial implications of the treatment of emission allowances in financial reports. Indeed, during the last years carbon markets have began to have material impacts on the balance sheets with carbon allowances being treated as assets or potential liabilities and affecting financial cash flows (Ascui and Lovell, 2011;Lovell et al., 2010;KPMG, 2008). These impacts are also expected to rise during the third phase of the EU ETS in which the majority of allowances are expected to be auctioned rather than being given away for free, thereby further affecting the financial results of firms.
Since now we have only referred to climate change regulation only as a source of business risk. Nevertheless, climate change regulation could under certain circumstances help companies gain a competitive advantage and increase their profitability. This claim is based on the fact that competiveness is not related only to the cost of raw materials or to the ability for mass productions of goods, but also to the ability of firms to be innovative and improve their production processes (Porter & Van der Linde, 1995;Williams et al., 2002). Therefore, according to Porter and Van der Linde (1995), well designed environmental regulation can stimulate the development of innovative technologies and production processes, which on their turn could make up for the regulation costs imposed on firms and even offer a competitive advantage.
For example, investing in carbon reduction projects could generate a positive return on investment by offering tax credits for energy reduction or the development of carbon capture and storage (CCS) technology could leave firms with a surplus of emissions credits to be traded in carbon markets thus offering a new source of revenue (Ernst & Young, 2010). However empirical research has shown that firms are not always able to counterbalance the costs of environmental regulations, the probability to do so increases when the production process is relatively flexible, when the company belongs to a high competitive sector and when environmental regulation is adaptable, i.e. there are pollution permits trading systems in place (Ambec and Lanoie, 2007).
As the effects of climate change become more intense and as carbon regulation evolves and becomes more specialized in addressing carbon reduction issues, corporate stakeholders will demand that companies develop and implement climate change mitigation strategies, address climate change risks and provide information regarding their carbon emissions profile. The aim of this study is to examine corporate practices related to climate change strategy, risk management and carbon accounting and to develop corporate indexes that measure the level of corporate commitment to the above practices. Moreover, we aim to prove that companies who devote resources in implementing climate change practices achieve enhanced financial performance. In order to develop our climate change indexes we are going to conduct an extensive survey, during which, climate change experts will assess the difficulty level of implementing the specific climate change practices included in each respective climate change index. In the next three subsections of this study, we are going to discuss climate change practices related to strategy, risk management and carbon accounting and present the questionnaires that we will use in order to develop our climate change indexes.

Climate Change Strategies
Strategy, as a term, comes from the Greek word "strategia" which can be narrowly defined as "the art of the general". It is a concept that has been adopted from the military and been adapted into a business term describing the means and tactics a firm uses to realize its purposes and missions. Several definitions have been given for the concept of strategy. Henry Mintzberg describes strategy as a "plan" on how to get from one point to another, as a "pattern of actions", as a "position" on which particular products or services should a firm offer to a particular market and as "perspective", a shared mentality of how things should be done. Kenneth Andrews defines corporate strategy as ''the pattern of decisions in a company that determines and reveals its objectives, purposes, or goals, produces the principal policies and plans for achieving those goals, and defines the range of business the company is to pursue, the kind of economic and human organization it is or intends to be, and the nature of the economic and non-economic contribution it intends to make to its shareholders, employees, customers, and communities". Finally, Michael Porter, who focuses on competitive strategy, explains the concept of strategy as a means of "being different" or "deliberately choosing a different set of activities to deliver a unique mix of value".
Regarding climate change, it is only recently that firms have begun to treat it as more than a Corporate Social Responsibility (CSR) issue. According to Porter and Reinhardt (2007), business leaders need to carefully examine the cost of emissions to their business as well as a firm's vulnerability to climate change physical, economic and social impacts. Studying a firm's value chain and assessing its exposure to climate change can help them develop strategies that will not only reduce current and impeding climate change risks but also reveal business opportunities and enhance corporate performance. There is a number of different strategies that a firm can adopt ant those depend both on the strategic choices it top management makes as well as on its available resources (Lee, 2012;Christmann, 2000). Some firms may choose to make incremental changes regarding their business activities while others may choose to make radical changes on their business model.
In one of the earliest papers on climate change strategy, Dunn (2002) briefly describes the technological, economic and policy implications of climate change on firms.
Regarding the technological dimension, special attention is given on the use of alternative fuel sources, such as natural gas and renewable energy sources (RES), as well as the development of efficient Combined Heat and Power (CHP) technologies (please see Table 4). Also, the use of market-based instruments, such as carbon trading is suggested as a way of lowering the cost of reducing CO 2 emissions. Finally, the need of developing climate change risk management processes is also denoted, as insurance and reinsurance companies have already began to quantify the cost of climate change risks on companies, e.g. cost of climate change related extreme weather events.
Weinhofer and Hoffman (2010) divide climate change strategies into three different groups: CO 2 compensation, CO 2 reduction and Carbon independence. CO 2 compensation strategies involve actions taken by firms to balance their carbon emissions, for example, through ETS, buying CERs or investing in emissions reduction projects. CO 2 reduction strategies focus on the improvement of CO 2 emitting production processes or on the design of new products, whose production emits fewer emissions. Finally, CO 2 reduction strategies involve the design of processes and products that are carbon free or which radically reduce carbon emissions.
In their study of emergent corporate climate change strategies, Kolk and Pinkse (2005) also distinguish between innovation strategies and compensation strategies. They describe innovation strategies as process improvements that reduce energy consumption, for example installation of energy-efficient technologies for carbon intensive industries such as oil and gas and chemicals, or processes focused on the reduction of CO 2 emissions through supply chain processes for automotives or electronics companies. In the same line with Weinhofer and Hoffman (2010), compensation strategies include internal transfer of carbon emissions through emissions trading or participation in carbon offset projects. Jeswani et al. (2008) separate corporate climate change strategies into two categories: operational activities for energy efficiency and GHG reduction and management activities. They lump all strategies described by Weinhofer and Hoffman (2010) and Kolk and Pinkse (2005)   The specific corporate practices related to its type of climate change strategy, and the related research, are presented in Table 3. Building on these climate change practices, we are going to create a climate change strategy questionnaire, which we are going to use in order to conduct an extensive survey, during which, climate change experts will assess the difficulty level of implementing the specific climate change strategies included in Table 3. The questionnaire is presented in Appendix A.  Regarding Climate Change Governance, decision-makers in firms are generally considered those who are in position to make changes in an organization, to choose the business environment in which it operates and engrave its course in the long-term. In Linnenluecke et al.'s (2015) study on executives' perception on the need of developing climate change adaptation strategies, the authors find that engagement with climate change science and the perceived degree of the firm's vulnerability are positively related to the choice of developing climate change strategies. Therefore, examining the level of commitment of corporate management in climate change, as well as the degree to which they have access in information regarding climate change impacts on their firm, are of utmost importance, in the development of effective climate change strategies.
As far as Carbon Reduction Strategies is concerned, particular attention is placed on the technology and carbon reduction business process. During the last decade, in order to reduce CO 2 emissions, governments have been focusing on designing efficient climate change policies such as ETSs or carbon taxes. However, although near-term, moderate emission reduction goals can be achieved with the use of such economic price instruments, ambitious emissions reduction targets in the future, such as the European Union (EU) 2050 targets for 80% reduction of CO 2 emissions at 1990 levels, will be very hard if not impossible to achieve without the pervasive diffusion of lowcarbon technologies (Sandén and Azar, 2005).
Moreover, empirical evidence has also shown that by adopting a proactive climate change innovation strategy firms do actually gain advantages concerning energy efficiency, reducing financial risks, avoiding social shaming from NGOs or targeting and boycotts and flexibility in adapting to stricter environmental regulation (Boiral, 2006;Lash and Wellington, 2007). Furthermore, investing in innovation can give firms competitive advantages that could offer them a larger market share or open the path for new markets. Finally, companies that have already adopted specific climate change strategies can lobby the government for stricter climate change regulations, for example specific production and product environmental standards, and thus use climate change regulation in order to shield themselves from competition or entrants of new firms in the business sector (Lash and Wellington, 2007).
In this study, we are going to examine some indicative climate change technologies based on the research of Cadez and Czerny (2016) on climate change strategies of carbon-intensive firms. The following carbon reduction technologies are examined: natural gas, Combined Heat and Power Technology (CHP), renewable energy sources (RES), Carbon Capture and Storage (CCS). For more information on these technologies please see Table 4. Suitable geological formations can be offshore or terrestrial, for example in depleted oil or gas fields or in water aquifers, and several kilometers below ground or sea level.
The Combined Heat and Power production systems (CHP -also known as cogeneration) generate both electricity and thermal energy in a single, integrated system. This comes in contrast with the conventional electricity production process where electricity is produced at a central station, while separate technology at local level is used for heating and cooling. In a CHP system, thermal energy produced by electricity is recovered and used for heating or cooling. Because CHP harnesses the heat that would otherwise be lost in the conventional electrical or mechanical energy production, the total efficiency of these integrated systems is much higher than that of individual electrical and thermal/cooling systems.
Renewable energy is from an energy resource that is replaced by a natural process at a rate that is equal to or faster than the rate at which that resource is being consumed. Any energy resource that is naturally regenerated over a short time scale and derived directly from the sun (such as thermal, photochemical, and photoelectric), indirectly from the sun (such as wind, hydropower, and photosynthetic energy  "Uncertainty must be taken in a sense radically distinct from the familiar notion of risk, from which it has never been properly separated…. The essential fact is that 'risk' means in some cases a quantity susceptible of measurement, while at other times it is something distinctly not of this character; and there are far-reaching and crucial differences in the bearings of the phenomena depending on which of the two is really present and operating…. It will appear that a measurable uncertainty, or 'risk' proper, as we shall use the term, is so far different from an immeasurable one that it is not in effect an uncertainty at all'' Since then, many academics have given different definitions of risk: "The combination of the probability of a hazardous event and its negative consequences." (Smith, 2013) "Risk is a combination of the chance of a particular event, with the impact that the event would cause if it occurred. Risk therefore has two componentsthe chance (or probability) of an event occurring and the impact (or consequence) associated with that event. The consequence of an event may be either desirable or undesirable…In some, but not all cases, therefore a convenient single measure of the importance of a risk is given by: Risk = Probability × Consequence." (Sayers et al.,2002) "Risk might be defined simply as the probability of the occurrence of an undesired event [but] be better described as the probability of a hazard contributing to a potential disaster…importantly, it involves consideration of vulnerability to the hazard". (Stenchion, 1997) ''Risk is the chance of something happening that will have an impact upon objectives.
Risk is measured in terms of likelihood and consequences'' (AS/NZS 4360 -Risk

Management Standard)
According to ISO 31000, risk is the "effect of uncertainty on objectives" and an effect is a positive or negative deviation from what is expected.
Uncertainty on the other hand is a state where lack of complete or accurate information leads to inadequate or incomplete knowledge or understanding of a situation. In this context, uncertainty exists whenever the knowledge we have about an event, about the consequences of an event, or likelihood of an event taking place, is inadequate or incomplete.
Therefore risk management can be defined as "the systematic application of management policies, procedures and practices to the tasks of identifying, analysing, evaluating, treating and monitoring risk.'' (AS/NZS 4360 -Risk Management Standard) The science of risk management deals with the identification, analysis and perception of risk by offering practical strategies for the prevention, reduction and transfer of risk (UNDP, 2002). For managers, the management aspects of environmental protection are linked with the principles of risk management. The main objective has always been to avoid the costs associated with industrial accidents, consumer boycotts or lawsuits.
However climate change in relation to other environmental problems such as e.g. the ozone hole, biodiversity protection, etc., has the characteristic that occurs discontinuously, systemic (affecting all economic sectors) and under great uncertainty regime ( for example to what degree will the temperature of the earth increase) (Winn et al. 2011).
Climate change is characterized by a high degree of uncertainty, by the lack of sufficient prior data related to extreme weather events and can lead to 'massive discontinuous changes' (Winn et al. 2011). For that reason, statistical methods based on analysis of trends based on historical data, and the traditional probabilistic models used by insurance companies and the risk management analysis tools that exist today may prove to be insufficient to be able to integrate quantitative and qualitative risk, associated with the impact of climate change in business operation (Winn et al. 2011).
This highlights the need to develop different risk management procedures that incorporate the principles of risk management, climate change adaptation and resilience strategies into one integrated framework. Busch and Hoffmann (2007)  Although climate change mitigation strategies regarding for example CO 2 emission reduction has received considerable attention during the last twenty year, little attention has been given on how the natural environment affects organizational management, i.e. how are organizations adapting on the physical impacts of climate change and how they enhance their resilience capacity to address and recover from natural catastrophes (Winn et al., 2011;Linnenluecke et al., 2008;Hoffman, 2006 Perceived firm vulnerability to climate change physical risks is closely related to perceived impacts of climate change risks, which includes both the direct physical impacts on a firm's business operations as well as the indirect impacts on a firm's market and business environment (e.g. the supply chain). Moreover, it is affected by a firm's past experience with climate change risks, from lessons learnt from previous extreme weather events and form its ability to quantify and assess the financial implications of extreme weather events. Also, important to perceived firm vulnerability is the degree to which climate change impacts can be controlled for and it depends on the rarity of extreme weather events, on the degree of insurance coverage and on the need for coordinated response with business partners.
Finally, all of the above affect a firm's response to the physical risks of climate change which, according to the authors, can be divided into taking "Routine measures" , i.e.
developing risk monitoring and assessment procedures, taking technical measures to endure impacts, using financial instruments for risk sharing, developing emergency and restorations plans etc. and "Non-Routine measures", which could involve assessing the vulnerability of geographic sites of business activities, conducting product portfolio diversification by investing in alternative products or business procedures and driving cooperation within the industry in order to reduce climate change exposure. Linnenluecke et al. (2011) divide corporate responses to extreme weather by using an organizational theory approach. According to the authors, extreme weather events cause organizations to enhance their adaptation, i.e. their long-term readjustment of their processes and products to the physical impacts of climate change and their resilience i.e. their capacity to absorb or recover from extreme weather events.
Adaptation is characterized by incremental changes to business processes, which can be either proactive or reactive to a extreme weather events and includes the reorganization of corporate strategies in order to "achieve improved performance by reaching an equilibrium with the environment". Resilience involves enhancing the capacity of a company to absorb unexpected weather events and to persist despite external environmental disturbances by using non-routine measures.
Despite the fact that implementing strategies that enhance the adaptation and resilience of firms, greatly improves their ability to protect themselves to the adverse implications of climate change, in order for firms to successfully adjust themselves to extreme weather events, they have to link both adaptation and resilience processes into one single action framework, something which, according to the authors, most companies have yet to accomplish. Finally, the authors propose a five step procedure for integrating both concepts which involves the development of anticipatory adaptation measures, resilience and impact resistance measures for addressing extreme weather events exposure, recovery and restoration measures, assessment of the resilience capacity of the company after the incurrence of an extreme event and the readjustment of anticipatory adaptation according to the assessment of the resilience measures taken in the previous stages.
Regarding climate change regulation risk, even though the Kyoto Protocol sets specific GHG reduction targets for most industrialized countries, the fact that policymakers at national level have left many regulation issues open has increased climate change regulation uncertainty (Engau and Hoffman, 2011). Hoffman (2005) suggests that companies should try to participate in climate change regulation development and implementation by collaborating with the government as a means to reduce regulation uncertainty. However, research in international business activity has shown that it is generally difficult for foreign firms to built relationships policy makers compared to domestic firms and since economic activity nowadays has been internationalized to a high degree, lobbying for climate change regulation could prove inefficient as a climate change risk mitigation strategy.
According to Engau and Hoffman (2011), firms adopt different strategies when it comes to copying with climate change uncertainty. The authors divide climate change regulation strategies into offensive, defensive and passive. Firms belonging in the first category attempt to reduce the causes of uncertainty by systematically searching for information regarding climate change regulation risks on their own or in cooperation with other firms in the same sector. Finally, they try to directly influence policy makers and adopt flexible climate change strategies that allow them to easily adjust themselves to a changing regulatory environment. Defensive strategies also focus on cooperation and flexibility but also involve stabilization strategies such as engaging in long-term contracts e.g. with suppliers or customers in order to decrease uncertainty and imitation strategies, where they stay one step behind their competitors. Finally, passive strategies focus on reorganizing internal business processes to reduce uncertainty, postponing business decisions to minimize regulation exposure or withdrawing from specific business environments that are characterized by a high degree of uncertainty.
In this study we will use the principles of risk management. Typically, risk management involves three stages: risk identification, risk assessment and risk response (Smith, 1995;Merna and Al-Thani, 2008). During the first stage, risk identification, company management determines which type of risk is relevant to its business activities. As argued above, risk identification is highly influenced by risk awareness and the information available to risk managers regarding the effects of the particular risk on their business. In the second stage, risk assessment, the exposure of the firm to the particular risk, the probability of the risk and potential implications to business are estimated. Most times, this stage leads to the calculation of the cost -in monetary value -to the business should the particular risk event take place. Finally, risk response involves all the potential measures a company can take to minimize its exposure to the particular risk. Our purpose is to measure the climate change risk management effort of firms and to examine whether identifying, assessing and managing climate change risks affects a firm's financial performance. More information on the measurement instrument developed for this purpose is given in the Methodology section of this research. Below, we briefly discuss the climate change risks that will be included in this research.
At this point it is appropriate to mention briefly the reasons why businesses adopt strategies to deal with the consequences of climate change. As we noted above, climate change and the impact policy responses that constitute systemic risk sources affecting prices in the energy market, national income, consumption, health and the prices of agricultural products (Hoffmann, 2006;Wellington and Sauer, 2005). However, apart from the systemic effects on the economy, climate change is a source of regulatory, physical, competitive, litigation and reputation risk for companies (Lash and Wellington, 2006;Wellington and Sauer, 2005;Nikolaou et al., 2015). These risks do not uniformly affect all businesses. Some sectors, depending on the nature of business activities, will be more exposed than others. Even within the same industry, the vulnerability of a company will be determined by its ability to develop the individual risk management strategies.

Regulation Risk
This type of risk includes the impacts of regulation regarding GHG emissions both at national or international level on corporate activities. Depending on the nature of their business activities, some companies will face higher costs in adapting to climate change regulation, for example, firms who belong to CO 2 intensive industries, such as oil and gas or chemical companies. However, since the demand of many GHG intensive products (such as energy) is highly inelastic, the increased costs that GHG intensive companies will face, due to climate change regulation, will most likely affect the production costs of the majority of companies, depending on their place in the global business value chain. Therefore, climate change regulation will most likely create risks also for non-GHG intensive firms.
Climate change regulation can be classified into two categories "command and control'' regulation and "market-based" regulation.
Command and control regulation includes policies that prescribe how much CO 2 an individual source is allowed to emit -also known as performance standards -and what type of control equipment it must use in order to meet the standards set by the regulators. Such standards are usually defined in terms of emission intensity rates.
Despite the introduction of potentially more cost-effective methods for regulating emissions, this type of regulation is still commonly used and is sometimes statutorily required. It is almost always available as a "backstop" if other approaches do not achieve desired pollution limits. Command and control regulation also includes technology-based standards which force companies to use a particular pollution control technology, such as installing scrubbers on smokestacks.
Market-based regulation is based on creating incentives for firms to reduce CO 2 emissions. In a market based regulation regime, each firm can choose on its own how to most cost-effectively achieve the required CO 2 emission level. Some firms can reduce emissions more cheaply than others (because of the age of their equipment or the technology they are using), allowing them to reduce their emissions more than other firms and eventually compensate for those firm who face higher costs and can do less regarding emissions reduction. Taken together, the overall environmental objective will be achieved at the lowest possible total costs. " The key criterion in determining if a policy is 'market-based' is that it provides a financial incentive designed to elicit a specific behavior from those responsible for the pollution. Some policy options are applicable as economy-wide solutions where greater efficiencies can be achieved, while others are more generally targeted to a particular market segment or sector." (C2ES, 2015). Classic 'market-based' approaches for emissions reduction involve taxes, subsidies and cap-and-trade schemes. A tax sets a price on each unit of CO 2 emissions. By introducing a carbon tax, the firm or facility that produces emissions incurs an additional cost based on the amount of CO 2 emitted.
Therefore, there is an incentive to reduce the emissions produced by changing production processes or by adopting new energy efficient technology. Moreover, taxes provide a continuous incentive for innovation as the more emissions are reduced, the less tax a company pays. Subsidy programs include government assistance, in the form of tax credits, for specific technologies that reduce CO 2 emissions in a similar way to taxes. They are in fact a form of negative taxes (C2ES, 2015) because they provide a specific financial mechanism to motivate a particular environmentally beneficial outcome.
Instead of putting a price on each unit of CO 2 emissions, a total quantity of emissions (a "cap") is set by the regulator. Then, companies buy and sell emission allowances (tradable certificates that allow a certain amount of emissions) based on their needs.
The limited number of these allowances creates demand and supply market forces. If a firm can reduce emissions below its requirements, so it has excess allowances, it can sell those allowances in emissions trading markets to a firm that finds it more difficult (costly) to reduce emissions.

Physical risk.
Physical risk refers to the impacts of climate change on business activities related to the natural environment: extreme weather events, such as increased intensity of cyclones and storms, floods, hurricanes, typhoons etc, changes in temperature and precipitation extremes, for example impacts such as reduced water supplies, droughts.
Since climate change is characterized by a high degree of uncertainty regarding the time and scale of its impacts in the natural environment as well as the fact climate change has different impacts across regions and economic sectors, extended research has been made on how it will affect global economic activity ). Some sectors of the economy such as the agriculture, fisheries, forestry, insurance, real estate and tourism are particularly vulnerable to the physical risks of climate due to their direct dependence to the natural environment (Pinkse and Gasbarro, 2016;Winn et al., 2011;Lash and Wellington, 2006;Wellington and Sauer, 2005). For example, luck of snow for a prolonged period could seriously damage the winter -tourism industry (ski resorts, etc.), extreme weather events could damage energy transportation systems, water shortage and temperature extremes could affect water supply and agricultural production (Busch and Hoffman, 2007). In Figure 4 and In this research we are going to focus on two main types of physical risks related to climate change. One involves the risks associated with extreme weather events and the other with precipitation extremes which include heavy rainfall or extreme droughts.

Reputation Risk
Reputation has been given different definitions in academic literature. Fombrun (1996) defines reputation as ''a perceptual representation of a company's past actions and future prospects that describe the firm's overall appeal to all its key constituents when compared to other leading rivals''.  perceive corporate reputation as a social construction of its shareholders while Delgado-García et al.
(2013) consider reputation as "an intangible asset with value enhancing potential, characterized by the imbalance between the length it takes to build it up and its fragility".
Corporate reputation is usually measured via various reputational indexes, which focus mainly on financial performance, quality of management, social and environmental responsibility performance, employee quality and quality of goods and services (Bebbignton et al, 2008). In this sense, corporate reputation is closely related to CSR and can be enhanced through CSR reporting (Brammer and Pavelin, 2006). These authors find that there is a positive relationship between corporate social performance and reputation and that this relationship is stronger in sectors that are associated with salient social and environmental issues.
Delgado- García et al. (2013) examine the effect of corporate reputation on firm risk.
The divide risk into unsystematic, i.e. risk that cannot be captured by overall market trends and include those forces that affect only the particular firm itself and systematic risk i.e. risk that affects the whole market and not just the firm. More specifically, they find that a good reputation reduces a firm's unsystematic risk and total risk but increases systematic risk. They explain these results in terms that a firm, by having a good reputation decreases uncertainty regarding its operational activities and attract more stakeholders. On the other hand, the higher the reputation of the firm, the higher the demands of the stakeholders. This means that in the case of a market downturn, a firm faces higher reputational risks than other firms with not such enhanced reputation, because stakeholders will demand that it keeps the high-quality standards that have helped increase its reputation.

Litigation risk.
Businesses have increased GHGs emissions may be confronted with lawsuits, as with the tobacco companies, pharmaceutical companies, etc. (Lash and Wellington, 2006;Wellington and Sauer, 2005). Along those lines, corporate disclosures regarding climate change practices can provide a sort of ''protection" against potential lawsuits.
Hanley and Hoberg (2012) find that corporate disclosure forms an effective hedging strategy against all types of lawsuits. They claim that for investors, in order to file a lawsuit against a company, two conditions must be met. First, the investors must suffer monetary losses and second, they must e able to prove that the losses suffered where due to misinformation or lack of sufficient information from the part of the firm.
Therefore, enhancing information disclosure reduces the chances of having to suffer economic damages from lawsuits.
Additionally, in their research on the relationship between environmental performance and environmental disclosure, Cho and Patten (2007) find that firms who have worse environmental performance disclose more information regarding their environmental practices than firms with better environmental performance. In this case, environmental disclosure is used as a vehicle to mitigate public scrutiny regarding environmental issues. The same results are also obtained in the study of Cho et al. (2012), where environmental performance is found to be negatively related to the level of disclosure with ''worse performing companies making more extensive disclosures'' So far, we have considered a positive relationship between litigation risk and corporate disclosures. However, academic research has shown that this is not always the case, especially for firms who have suffered severe litigation costs. More specific, Rogers and Van Burskirk (2009) find that firms do not respond to a litigation event by increasing disclosures, rather they reduce the amount of information provided to investors after a litigation event. In the same line, Laux and Stocken (2012) also find that an increase in expected legal damages increases misreporting. This phenomenon is stronger when the firm's management is optimistic about the firm's future prospects, when the costs of litigation are considered low and when the firms has a weak internal control system.

Competitive risk.
As mentioned above, the business operating costs will increase under the pressure of climate legislation. This may adversely affect the competitiveness of companies and thus their profitability and ability to raise capital. For investors, the parameters of the competitive position of the company relative to other companies in the industry and how these are affected by the restrictions in GHGs emissions are particularly important during the financial evaluation of investments (Weber et al., 2010;Busch and Hoffmann, 2007;Wellington and Sauer, 2005).
Companies that have a negative picture regarding climate change (whether it has to do with the policy of the company, its products or its operating procedures) may come up against the consumer shift and investors in markets sensitized to environmental issues (Lash and Wellington, 2006;Wellington and Sauer, 2005;Nikoalou et al., 2015). This is particularly important in industries with high competition, such as car manufacturers and fuel supply companies, where the loyalty of consumers to the name brand is an important feature of the company's value.
In this research, we will examine two dimensions of competitive risk related to climate change. One involves the uncertainly in market signals due to the increases costs of regulation, which affects operational costs of particular economic sectors. The other involves competition risks related to changing consumer behavior. As the negative effects of climate change become more evident, consumers are expected to change their preferences towards products that are ''carbon friendly'' or towards companies that actively manage their GHG emissions.
Building on the above, we are going to create a climate change risk management questionnaire, which we are going to use in order to conduct an extensive survey, during which, climate change experts will assess the difficulty level of identifying the impacts, calculating the financial implications of the impacts, identifying management methods and calculating the cost of management methods related to the climate change risks described above. The questionnaire is presented in Appendix B.

Carbon Accounting and Reporting
Accounting is used by managers to analyze various aspects of the economic activities of a company. Furthermore, it forms a vehicle of disclosure regarding the financial statements of firms. According to the International Accounting Standards Committee "The objective of financial statements is to provide information about the financial position, performance and changes in financial position of an enterprise that is useful to a wide range of users in making economic decisions". As discussed in the previous sections, climate change poses significant risks that could affect a firm's corporate performance. Climate change regulation, for example, have increased compliance costs for firms operating in carbon intensive sectors either in the form of obtaining emission permits or complying to specific industry standards. Additionally, the increase in extreme weather event or change in precipitation patterns has increased insurance costs for companies that heavily depend on the natural environment. Due to increasing public awareness public awareness regarding the effects of climate change on the physical, economical and social environment, various stakeholders have begun to demand more information regarding the climate change exposure of firms. Information regarding carbon emissions is being made publicly available in the form of business reports, in website and on the media, while public reporting and participation involuntary and mandatory reporting programs at national, regional and local level are being adopted as mainstream climate change strategies worldwide.
As demand for increased disclosure rises, companies have begun to implement business processes that allow them to assess, measure and manage GHG emissions, to develop their GHG accounting capabilities by setting emission baseline's on which future emission trends are calculated and to make provisions for purchasing emission allowances Linnenluecke et al., 2015). Finally, the need for developing climate change strategies have spurred the emergence of firms who conduct climate change related research, for example assessing carbon exposure of companies, firms who are offering management consulting on carbon emissions reduction strategies, or provide carbon offsetting services, through the implementation of CDM projects (Corbera et al., 2009;Dhanda and Hartman, 2011;Schaltegger, and Csutora , 2012).
Carbon Accounting is the estimation, calculation, measurement, monitoring, reporting, validation, verification and auditing of CO 2 emissions. This involves, apart from the general impacts of corporate activities on climate change and vice versa, the GHG emissions released to the atmosphere, the GHG emissions removed from the atmosphere (for example through CCS technology), GHG emission rights and obligations and emission reductions. It also involves the legal financial instruments liked to the above initiatives, including carbon taxes and emissions trading schemes.
Moreover, all the above actions involve climate change impacts and emissions released at global or national level, at corporate, project, or product level, and at supply chain level. Finally, the ultimate purpose of carbon accounting is the production of information that will be used for mandatory or voluntary reporting, disclosure, research , compliance, auditing, as well as for benchmarking and marketing purposes (Ascui and Lovel, 2011).
Carbon accounting has its origins in the conceptual framework of social accounting which examines the role of accounting science not in the narrow boundaries of an organization but a s an enabler in the formation of social processes (Hopwood and Miller, 1994). In this sense accounting can be viewed as '' the mirror with which the public can interpret and understand corporate operations, hence a transparent provision of business data relating to economic, social and environmental issues'' (Ngwakwe, 2012). Academic research has shown that accounting science has begun to engage in sustainability related issues. Nevertheless, this involvement is still weak due to the lack of formalized sustainability reporting standards (Ngwakwe, 2012). The majority of conceptual work on sustainability accounting can be attributed to Gray (1992) who proposed that accounting science should move from its traditional paradigm and incorporate environmental and social aspects of corporate activities. Internal and external auditing regarding the ethical foundation of a firm's activities, reporting of energy consumption or waste management, environmental impact assessment, corporate disclosure and reporting o social and environmental activities, accounting for environmental assets and liabilities are among the aspects of accounting science should take into account Following the same school of thought, Elkington (1997) introduced the term ''triple-bottom line'' (TBL), which emphasizes the necessity that firms report not only on issues regarding their financial performance but also on their social and environmental performance. Finally, financial markets have begun to address sustainability issues, such as responsible investing, or incorporation of sustainability reporting in mainstream financial reports (Hopwood, 2009;Schaltegger and Cscutora, 2012).
The need for accounting and reporting for social and environmental aspects for corporate activities has spurred the development of reporting schemes that aim to create corporate reporting standards. The Global Reporting Initiative (GRI), launched in 1997, was the result of collaboration between the United Nations Environmental Program (UNEP) and Ceres and aims to provide guidelines for the implementation of the TBL reporting framework at corporate level (Linnenluecke et al., 2015). The GRI framework prescribes methodologies and standards, which firms can use to disclose information regarding their economic, environmental and social performance (GRI, 2015), thus contributing to enhanced corporate transparency and accountability. Finally, corporate carbon accounting refers to the physical and monetary accounting of emissions at corporate level, and it also include carbon emissions related to ETS, carbon limits and carbon taxes.
As we have discussed before, climate change is a particularly complex environmental problem because it manifests itself in a spatial and temporal scale that is characterized by a high degree of uncertainty, and because it has widespread environmental, economic and social impacts. Moreover, with the introduction climate change regulation in the form of either command and control regulation or market based regulation, such as ETSs, carbon taxes etc., financial analysts and investors have begun to take into account the value relevance of emission rights and allowances and their effect of financial corporate performance (Schlateger and Csutora, 2012).
In response to carbon trading schemes or in anticipation of future carbon-pricing legislation many companies have begun to develop internal price of carbon schemes.
''Broadly speaking, internal carbon pricing is most frequently used as a shadow price which can be added to future investments and operational costs as a way of hedging against future policy decisions to implement any carbon-pricing mechanisms.'' (The New York Times, 2015). Furthermore, ''carbon pricing systems encourage innovation and help ensure sustained economic competitiveness'' (CDP, 2015). According to the CDP some of the benefits of using internal price of carbon schemes are to mitigate the effect of government legislation regarding carbon emission reductions, to help them comply with government legislation, to avoid transaction costs associated with trading permits in national schemes in favor of factoring in these prices internally, to justify investments that may have smaller margins without a carbon price and to manage risk for future investment.
''In some exceptional circumstances, a carbon price can underpin a more expansive company scheme. In 2012, Microsoft made the decision to use an internal carbon fee which was charged to individual business groups using Microsoft services. The funds from this internal tax were then used to invest in energy-efficiency initiatives, renewable energy and carbon offset projects in order to meet net carbon neutrality targets. Since 2014, 100 percent of Microsoft's energy consumption is sourced or offset due to these projects. In just three years, the company has reduced its emissions by 7.5 million tons of CO 2 and saved more than $10 million on energy costs. This year it expects its scheme to amount to $20 million of internal charges.'' (The New York Times, 2015) Moreover, carbon regulation in the form of carbon taxes and cap-and-trade systems has created new accounting demands (Linnenluecke et al., 2015). Since the introduction of ETSs, there has been an increasing debate on the accounting policies regarding emissions allowances and their treatment either as asset or liabilities, and on the different carbon hedging instruments that have been created (Haupt and Ismer,, 2013). Among the most popular carbon accounting practices adopted by firms are the classification of emission allowance as intangible assets (Lovell et al., 2010;Bebbington and Larringa-Gonzalez, 2008), the initial recognition of emissions allowances at cost (for purchased allowances ) or a nil value (for granted allowances)in the second phase of the EU ETS, a proportion of the initial CO 2 allowances where given for free by governments to companies participating in the EU ETS, while in the third phase of the EU ETS the number of free allowances has been reduced to a great extent-and the measurement of emission allowance either at the initial cost value or through revaluation according to their fair value ( Lovell et al, 2010;Haupt and Ismer, 2013). Therefore, the need for developing credible and reliable accounting methods, at the time of its introduction it gave neither specific guidance on the treatment of emission allowances of non-participants in the EU ETSs, who were also able to acquire allowances for trading or investment purposes , nor on carbon derivatives, and because it didn't set specific measurement models for the emission allowances as liabilities, it was rejected by the majority of stakeholders and finally withdrawn (Lovell et al., 2010;Haupt and Ismer, 2013;Lovell and McKenzie, 2011).
Subsequently, associations for responsible business of investor groups such as the CDP or the CDSB have developed reporting schemes that include performance scores, targets and reporting methodologies regarding the amount of emissions produced, the nature of these emissions, i.e. whether they are the result of the production processes of the firm or have their origin on electricity consumption or they are associated with the supply chain of the firms. Accounting firms have also began to provide advice on issues related to the assessment, accounting reporting and auditing of carbon emissions related information, (KPMG, 2008), since many firms lack the organizational capabilities to develop carbon accounting business processes on their own.
Furthermore, the need for internal and external validation of carbon emissions has also strengthened the role of accounting firms, as there is an increasing demand for carbon assurance services, which enhance the quality of carbon-related information (Moroney et al., 2012).
Regarding the development of assurance services or sustainability reporting ''challenges arise from the requirements of relevance, completeness, reliability, neutrality and understandability because of the broad and complex subject matter in sustainability reports'' (Simnett and Nugent, 2007). Table 5 provides a description of issues related to sustainability assurance.
In responding to the increased stakeholder demand for accurate and reliable information regarding the impacts of corporate activities on climate change, the World Resource Institute (WRI) and the World Business Council for Sustainable Development (WBCSD), created the Greenhouse Gas (GHG) Protocol in 1998. Today, the GHG Protocol constitutes the most complete guide of accounting and reporting standards for business, in order to develop a valid and reliable GHG inventory. As with classical accounting theory, carbon accounting must also use globally accepted principles. Principles can be broadly defined as the general approaches utilized in the recognition and measurement of accounting event. The outcome of an independent verification process, the term is often used interchangeably with the term verification. It is increasingly used to describe the evaluation and assessment services provided by independent accounting and other firms, usually based on specific assurance standards or frameworks.

Level of assurance
Assurance providers often offer two levels: 'reasonable assurance' (i.e. high but not absolute) or 'limited assurance' (i.e. moderate). The higher the level of assurance, the more rigorous the assurance process is, as defined in the standards and procedures used for the specific assurance engagement

Assurance Providers
• Accountancy firms. They are normally connected to global networks; are focused on business; have expertise in financial and non-financial reporting; they have their own systems, controls and audit/assurance procedures (including for climate change/GHG data) • Engineering firms. They normally offer technical certifications and engineering expertise; they understand complex processes and are used to risk-based analysis; they apply a multi-disciplinary approach.
• Sustainability services firms. Their focus and expertise is on sustainability related issues; they are smaller than the others assurance providers' general categories and are usually locally based; they are also often recognized because of their experience with stakeholder issues. The GHG Protocol sets also the organizational boundaries of reporting emissionswhich company entities or activities should be included in the GHG inventory.
According to the protocol, accompany can follow two different approaches in setting up its organizational boundaries. It can use either the control approach, where a company reports 100 percent of the GHG emissions from operations over which it has financial o operation control, and the equity approach, where a company account for the GHG emissions which correspond proportionally to its share of equity in the operation. Furthermore, it sets operational boundaries, which means that it clearly defines the direct and indirect emissions of a company's operational activities. Direct emissions are those who come from sources that are owned or controlled by the company. Indirect emissions are the consequences of the activities of the company but occur at sources owned or controlled by another company. The Protocol uses the term ''scope'' to define the operational boundaries in accounting and reporting GHG emissions. Therefore, scope 1 emissions refer to the direct emissions of the company, scope 2 refers to indirect emissions based on the electricity consumed by the company and scope 3 covers all other indirect emissions. For more information please see Figure   6 and Table 6.

Scope 2: Electricity indirect GHG emissions
Scope 2 accounts for GHG emissions from the generation of purchased electricity consumed by the company. Purchased electricity is defined as electricity that is purchased or otherwise brought into the organizational boundary of the company.
Scope 2 emissions physically occur at the facility where electricity is generated.

Scope 3: Other indirect GHG emissions
Scope 3 is an optional reporting category that allows for the treatment of all other Therefore, in order to achieve consistency over time, historic emission data will have to be recalculated over time. The first step in tracking emissions is to set a baseline year that is a specific year against which ca company's emissions are tracked over time. According to the GHG protocol companies can choose between a target base year and rolling base year. The target base year is used to define a GHG reduction target, for example CO 2 reduction 25% below the emissions target based year 2000. A rolling base year is a base year that changes, usually shifts forward by a certain number of years at regular intervals of time. In this research, we take into account baseline years that are either stable (target-base approach) or rolling, but at a time interval that is more than five years. Taking also into account the need for reducing CO 2 emissions below 1990 emission levels, the reason we chose not to take into account rolling base years with a time interval less than five years is because we believe that companies that calculate their emissions and se emission targets based on the emissions of the previous one or two years, do not set considerable emission reduction targets. Finally, the company must decide when it should recalculate emissions, for example when the company undergoes structural changes or when it decides to use an improve methodology to increase emission data accuracy.
Finally, the GHG protocol provides guidelines on the quantification of project based reductions, carbon offsets and credits. According to the protocol, the following accounting issues must be addressed when a firm calculates the amount of CO 2 emissions avoided by the implementation of a CO 2 reduction project. First, the alternative baseline scenario should be taken into account, i.e. what would have happened with the company's emissions if the project had not been implemented.
Second, the project must clearly reflect additional CO 2 reductions, which means that implementing the project removes more emissions from the atmosphere that would exit if the project had not been implemented. Finally, the accounting methodology used for the calculation of CO 2 emission reductions should avoid double counting of the emissions avoided, that is, if the CO 2 reductions take place to a source not controlled by the firm, the ownership of the source should be clarified to avoid double counting.
Building on the above literature, we are going to create an index that measures the level of corporate commitment regarding carbon accounting. This index will include three dimensions of carbon accounting: Building on the above practices, we are going to create a carbon accounting questionnaire, which we are going to use in order to conduct an extensive survey, during which, climate change experts will assess the difficulty level of implementing of the specific climate change practices described above. The questionnaire is presented in Appendix C.

Environmental management and financial performance
Firms devote considerable resources on the implementation of environmental management practices, in an attempt to control for the negative impacts of their operational activities on the natural environment. Therefore, it is essential that we examine the effect of these actions on the financial performance of firms and whether they add costs that are irretrievable, and can only worsen a firm's financial performance, or whether they can enhance financial performance, for example through the adoption of energy efficient technology that reduces overall production costs or through the development of differentiated eco-friendly products for which customers are willing to pay premium prices (Porter and van der Linde,1995;Ruf et al., 2001).
Many researchers have the relationship between environmental management and corporate performance. According to Dangelico and Pontrandolfo (2015), a firm's ability to successfully implement environmental actions, especially those focused on energy efficiency and pollution reduction, has a positive effect on the firm's market performance, measured in terms of access to new markets, increased market share, increased competitive and increased customers willingness to pay a premium price for products. On the other hand, a firm's image performance, measured through improved reputation, improved regulatory compliance and better innovation, is found to be positively related to the capabilities of the firm to implement environmental actions focused on the design and development of eco-friendly products and the elimination of hazardous input materials in the production process (Dangelico and Pontrandolfo, 2015).
Furthermore, Gallego-Alvarez et al. (2014) in examining the relationship between environmental and financial performance, find that this relationship becomes stronger in times of economic crisis for companies in carbon intensive sectors. Therefore, in order to enhance stakeholder relations and achieve higher economic benefits, the authors suggest that firms should invest in sustainable projects even in the times of economic crisis. Boiral et al. (2012) in their study on Canadian manufacturing firms confirm the positive relationship between corporate commitment to reduce GHG emissions and enhanced financial performance. Eccles et al. (2013), in examining the relationship between corporate sustainability processes and corporate performance for US companies, find that companies that have voluntarily adopted sustainability policies from 1993 to 2009 exhibit higher corporate performance (both in terms of stock market as well as accounting performance) than firms, with same organizational characteristics, who have not adopted sustainability policies. Additionally, Al-Najjar and Anfimiadou (2012) and Sinkin et al. (2008) also confirm that eco-efficient firms exhibit a higher market value than those lacking environmental strategies. Furthermore, Nakao et al. (2007) have shown that a firm's environmental performance has a positive significant impact on its financial performance and vice versa, and estimate that increasing environmental management scores could lead to an increase in a firm's intangible assets in the long-run.
Finally, Lannelongue et al. (2015) suggest that, in order to gain a better understanding of the effect of environmental management on financial performance, we should not focus on examining separately the effect of environmental management or environmental performance on financial performance. They find that only when we take into account environmental management productivity, which they define as the ratio between environmental management and environmental performance we can gain a more comprehensive understanding of the relationship between environmental management and financial performance.
However, not all studies find a positive relationship between environmental management and corporate performance. For example, Busch and Hoffmann (2011) find that carbon management is negatively related to financial performance. However, they find that reduced carbon intensity of business activities is positively related to financial performance. Cong and Freedman (2011), also find that there is no relationship between good governance and good pollution performance while they find that there is a positive relationship between pollution disclosure and corporate governance. Finally, Dobler et al. (2014) find that there is a negative association between environmental performance and environmental risk and that active risk management does not contribute significantly to environmental performance.
Along those lines, environmental disclosures have also been examined as having a positive significant impact on corporate performance. Prior empirical research has examined the relationship between firms' financial performance and their environmental disclosure practices. Most studies show a positive relationship between financial performance and environmental disclosures (Freedman and Jaggi, 2005;Liu and Anbumzhi, 2009;Nakao et al., 2007;Richardson and Welker, 2001;Brammer and Pavelin, 2006;Clarkson et al,.2008; Gallego-Alvarez 2010; Galani et al., 2011;Zhang et al., 2008). More specifically, firms with higher sustainability scores perform better than firms with lows sustainability scores (Eccles et al., 2013). On the other hand, Ghoul et al. (2011) show that firms with better corporate social performance (CSR) scores, especially those associated with environmental policies, have access to financing equity at a lower cost than firms with low CSR scores. Finally, Yadav et al. (2015) also confirm that firms with repeated high "green" rankings show significantly higher firm value than firms with either reduced or unchanged rankings across time.
However, the inverse relationship between financial performance and volume of information issued has also been detected by Freedman and Jaggi (1982) (Table 7), Climate Change Risk management (Table 8) and Carbon Accounting (Table 9). Using the above critical factors, we developed three questionnaires, which were sent to three different target groups.
Respondents were asked to rate each critical factor under a seven-point Likert scale, to indicate the difficulty level of implementing the respective climate change practice.
The questionnaires are presented in Appendix A, Appendix B and Appendix C. After our data was collected, they were assessed for validity and reliability by using exploratory factor analysis. The methodology used is extensively discussed in the next section.
Based on the results of the factor analysis eight corporate indexes for measuring the level of commitment (or management effort) regarding the implementation of climate change strategies, the identification and assessment of climate change risks and the adoption of carbon accounting practices, were developed. Three indexes are related to climate change strategy, three to risk climate change risk management and two to carbon accounting: On the contrary, Nishitani and Kokubu (2012) show that GHG emission reduction enhances firm value. However, this stands only for firms, for which strong market disciple, imposed by stockholders or investors, forces them to reduce GHG emissions. Finally, Busch and Hoffman (2011) show that a high carbon intensity of corporate activities negatively affects a firm's corporate performance. Taking as a reference the above authors, we pose the following hypothesis: H 4 : A company's level of commitment to reduce GHG emissions is positively related to its financial performance.   assessed are those belonging to the Financial Times 500 Global database. We will examine the climate change practices of these firms for the years 2012-2015.
The assessment of the selected companies will be conducted through a predefined scoring methodology that is described in this section of the study. After the extracting the relevant scores for its company, we will create eight variables in accordance with the indexes developed that will measure the level of corporate commitment regarding Practices. We will then examine the relationship between the above indexes and corporate financial performance.

Development of survey instruments
Based on the literature review, we developed a set of critical factors for Climate Change Strategy (Table 7), Climate Change Risk management (Table 8) and Carbon Accounting ( were developed under these factors. Using the above critical factors, we developed three questionnaires, which were sent to three different target groups. Respondents were asked to rate each critical factor under a seven-point Likert scale (1=Very Easy ... 7=Very Difficult), to indicate the difficulty level of implementing the respective climate change practice described in Table 7, Table 8 and Table 9. The questionnaires are presented in Appendix A, Appendix B and Appendix C. Replacement of carbon-based products by non-carbon based products. STRATEGY_ITEM_17 Implementation of end-use energy efficiency processes (e.g. turning down heating and cooling during non-working hours, reduce non-necessary travel).

STRATEGY_ITEM_18
Optimization of current business processes in order to reduce CO 2 emissions. STRATEGY_ITEM_19 Control of non-CO 2 gas emissions (e.g. CH4, H2O). STRATEGY_ITEM_20

Carbon Compensation Strategies
Participating in emissions trading schemes. STRATEGY_ITEM_21 Creating project-based carbon credits. STRATEGY_ITEM_22 Purchasing carbon credits. STRATEGY_ITEM_23 Identify impacts from changes in temperature and precipitation extremes. RISK_ITEM_13 Calculate financial implications from changes in temperature and precipitation extremes. RISK_ITEM_14 Identify management methods in order to address impacts from changes in temperature and precipitation extremes.

RISK_ITEM_15
Calculate the cost of management methods needed to address impacts from changes in temperature and precipitation extremes.

Reputational Risks
Identify potential impacts from negative reputation, related to climate change. RISK_ITEM_17 Calculate potential financial implications from negative reputation, related to climate change. RISK_ITEM_18 Identify management methods in order to address impacts from negative reputation, related to climate change.

RISK_ITEM_19
Calculate the cost of management methods needed to address impacts from negative reputation, related to climate change.

Uncertainty in Market signals
Identify potential impacts from climate change competitive risks regarding uncertainty in market signals.

RISK_ITEM_21
Calculate potential financial implications from climate change competitive risks regarding uncertainty in market signals.

RISK_ITEM_22
Identify management methods in order to address climate change competitive risks regarding uncertainty in market signals.

RISK_ITEM_23
Calculate the cost of management methods needed to address climate change competitive risks regarding uncertainty in market signals.

Changing Consumer Behavior
Identify potential impacts from climate change competitive risks related to changing consumer behavior.

RISK_ITEM_25
Calculate potential financial implications from climate change competitive risks related to changing consumer behavior.

RISK_ITEM_26
Identify management methods in order to address climate change competitive risks related to changing consumer behavior.

RISK_ITEM_27
Calculate the cost of management methods needed to address climate change competitive risks related to changing consumer behavior.

Legal Risk Management Legal Risks
Identify potential impacts from climate change legal risks. RISK_ITEM_29 Calculate potential financial implications from climate change legal risks. RISK_ITEM_30 Identify management methods in order to address climate change legal risks. RISK_ITEM_31 Calculate the cost of management methods needed to address climate change legal risks. RISK_ITEM_32

ACCOUNT_ITEM_06
Set an emissions baseline year. ACCOUNT_ITEM_07 Use an established methodology to calculate emissions. ACCOUNT_ITEM_08

Reporting and Verification of Carbon Emissions
Have CO 2 emissions verified through an internal auditing process.

ACCOUNT_ITEM_09
Have CO 2 emissions verified through an independent external auditing process.

ACCOUNT_ITEM_10
Achieve Reasonable Assurance Verification. ACCOUNT_ITEM_11 Report CO 2 emissions to voluntary GHG reporting programs.

ACCOUNT_ITEM_12
Report CO 2 emissions to mandatory GHG reporting programs.   Table 10 for climate change strategy, Table 11 for Climate Change Risk Management and Table 12 for Carbon Accounting and Reporting.   1,20% 1 At least 1 year but less than 3 years 5,20% 5 At least 3 years but less than 5 years 10,20% 11 At least 5 years but less than 10 years 54,20% 56 10 years or more 29,20% 30 Total 100,00% 103

Data Descriptive Statistics
The descriptive statistics of our critical climate change strategy factors are presented in Table 13. As we can observe, the skewness and kurtosis values are among the range of +/-1 which indicate that are data are close to a normal distribution. Furthermore, regarding the climate change governance factors, we see that STRATEGY_ITEM_03, which corresponds to the executive management compensation being linked to climate change performance targets, has the highest mean among the rest of the items.
Regarding, the climate change risk management factor items, these has been assess generally at the same level of difficultly by our survey respondents. As far as carbon reduction targets is concerned, targets related to absolute CO 2 emission reductions generally receive a higher mean of implementation difficulty than intensity targets.
Regarding Carbon Reduction Technologies, Carbon Capture and Storage of CO 2 has by far the highest mean difficulty than all the other factors, which can easily be explained if we take into account the fact that it is the newest carbon reduction technology and that it is considered to be at a demonstration stage. Finally, regarding carbon compensation strategies, participation in ETS and creation of project-based carbon credit have higher mean difficulty values than purchasing carbon credits. The descriptive statistics of our critical factors related to climate change risk management are presented in Table 14. As we can observe, the skewness and kurtosis values are also among the range of +/-1 which indicate that are data are close to a normal distribution. Regarding the risk factor related to regulation risk management we can observe that the items corresponding to command and control regulation generally present higher mean difficulty values that items included in the market based regulation factor. This result can be explained if we take into account the fact that command and control regulation is more difficult to comply with, since it does not offer any sort of means of compensating for GHG emissions as it is with market based regulation, where companies can buy carbon credits in ETS market.
Furthermore, we can observe that physical risk management related to changes in temperature and precipitation extremes present higher mean difficulty values than physical risk management related to extreme weather events. This can be explained perhaps by the fact that traditionally companies were using insurance mechanisms to protect against extreme weather events, therefore they do not assess it as especially difficult since it something that it is being implemented for many years. On the other hand, changes in precipitation extremes, reduced water supplies, droughts or extreme heat waves are physical risks that have more recently begun to affect operational activities. Therefore, it is expected to receive high mean average values.
Moreover, we can observe high mean difficulty values for reputational risk management as well as for competition risk management, which is to be expected since these types of risks are very difficult to measure and since they are susceptible to sudden changes, especially for reputational risks and changing consumer behavior.
Finally, regarding legal risk, we can see that the factors that receive the higher mean difficulty values are those associated with the costs of identifying potential financial implications and the costs of implementing management methods to address litigation risks. This can be explained by the high degree of uncertainty regarding the cost of having to deal with lawsuits for issues, such as climate change, for which there is no clear legislation in place. The descriptive statistics of our critical carbon accounting factors are presented in  and subsequently the quality of our weights, we have to measure the quality of our questionnaires.
The quality of a questionnaire is generally measured by conducting validity and reliability analysis. According to Hair et al. (1995) validity is the degree to which a measure accurately represents what it is supposed to. Reliability analysis indicates the extent to which repeating the same survey procedure yields the same results (Carmines and Zeller, 1979).

Validity assessment
Three types of validity are generally used in assessing the accuracy of a questionnaire: content validity, criterion validity and construct validity. A measurement construct is considered to have content validity when ''there is a general agreement that the measure has items that covers all aspects of the variable being measured" (Wee and Quazi, 2005). It is not based on the calculation of numerical factors but is subjectively judged by the researchers. Regarding this study, since the selection of the initial items in the critical factors of the questionnaire was based on extensive literature review, we consider it to have sufficient content validity. Criterion validity, also known as predictive validity assesses the extent to which a particular construct measure predicts or is related to other construct measures. When using criterion validity, correlations between the constructs and one established criterion are calculated and if these correlations are found to be significant then we can consider the constructs to have criterion-related validity. Finally, a construct measure is considered to have construct validity when it measures the theoretical construct that it was designed to measure.
When examining construct validity, we focus on the convergence between the items included in an original construct measure and how these items can be separated and reorganized in new constructs that can accurately measure different theoretical aspects of the original construct (Forza, 2002).
The most common method in examining construct validity is the use of factor analysis (Nunnaly, 1978;Wee and Quazi, 2005;Saraph et al., 1989). Factor analysis is generally used in the field of social sciences in order to interpret self-reporting questionnaires (Williams et al, 2010;Byrant et al, 1999). It is generally used to reduce a large number of variables into a smaller set of variables and it explores underlying dimensions between the variables that allow us to interpret the theoretical basis of our variables more accurately (Williams et al, 2010). There are two types of factor analysis, exploratory factor analysis and confirmatory factor analysis. Exploratory factor analysis -as its name suggests-is exploratory by nature (Williams et al, 2010).
That is, the researcher has no previous information on how the variables should be organized and examined. Of course, content validity can help us to broadly define sets of variables, however it cannot be used on its own when we want to develop and validate a "new" theory. On the contrary, confirmatory factor analysis is used to test an established theory and to examine the degree to which it can be applied to different context, for example different target population or in different time frames.
In this research, we are going to use exploratory factor analysis since our questionnaires have not been previously validated by other researchers. Thus, exploratory factor analysis will allow us to better understand the structure of our variables and to measure the accuracy of our survey results. Finally, it will help us to reorganize our initial critical factors, and their relevant items, into a new set of factors, which will make the original information obtained more manageable (Field, 2013).
Regarding the procedure of conducting exploratory factor analysis, we are going to adopt the one described by Field (2013). First, initial checks will be made regarding the appropriateness of the data for the factor analysis. Then, for the main analysis, the factor extraction and rotation methods will be examined the results of the analysis will be presented. Finally, the factors developed in the previous step will be assessed for reliability by using the internal consistency method. This procedure will be repeated separately for each questionnaire.
Regarding the appropriateness of the data, the following criteria must be met in order to be able to conduct factor analysis: 1. Sample size. The size of our sample must be sufficient in order to conduct factor analysis. There are several rules regarding the number of observations needed in order to conduct factor analysis. Hair et al. (1995) suggest that the observations must be at least five times as many as there are variables. Nunnally (1978) recommends that observations must be 10 times the number of initial measurement items included in the questionnaire. Comrey and Lee (2013) suggest samples sizes: 100 as poor, 300, as good and 1000 or more as excellent. Furthermore, Guadagnoli and Velicer (1988) claim that if a factor has four or more loadings greater than 0.6, then the factor is reliable regardless of the number of observations while MacCallum, Widaman, Zhang, and Hong (1999) find that with all communalities above 0.6, samples less than 100 are adequate. Finally, the adequacy of sampling is determined by Kaiser-Meyer-Olkin 's (KMO) Measure of Sampling Adequacy. The KMO test ranges from 0 to 1, with 0 denoting diffusion in the pattern of correlations between the variable items, thus making factor analysis inappropriate for the analysis of the data and 1 denoting patterns of correlations that are ''relative compact'', thus allowing factor analysis to produce ''distinct and reliable'' factors (Field, 2013). More specifically, KMO values around 0.6 are considered mediocre, values around 0.7 middling, and values around 0.8 or more meritorious (Field, 2013).

Correlations between variables.
In order to proceed with the extraction of factors, we must check for correlations between the variables.
Correlations have to be high (above 0.3), but not too high ( for example above 0.8). In both cases, the underlying variables must be removed from the analysis since they could potentially reduce the validity of the structure measurements. Correlations can be examined by looking at the correlation matrix of the variables. As mentioned above, correlations must be greater than 0.3, in order for factor analysis to be able to extract reliable factors (Hair et al., 1995). Also, examining the significance of Barllett's test of sphericity can help us assess the overall significance of the correlation matrix i.e. if p<=0.000, then Barllett's test is significant, which means that the correlations between variables are overall significantly different from zero (Field, 2013 can be based on the number of eigenvalues than are above 1. "This criterion is based on the idea that the eigenvalues represent the amount of variation explained by a factor and than an eigenvalue of 1 represents a substantial amount of variation" (Field, 2013).
The scree test is a diagram than shows each eigenvalue (Y-axis) against the factor with which it is associated (X-axis). According to the scree test, the number of factors extracted can be determined by the point where the slope of the curve of the diagram changes dramatically (Field, 2013). Finally, regarding the cumulative percent of variance explained, there is no fixed threshold which can determine the number of factors that should be extracted (Williams et al, 2010). In social sciences, the explained variance usually varies between 0.5 to 0.6 (Hair et al.,1995).
Finally, after the factors have been extracted we can ''calculate the degree to which variables load on these factors" (Field, 2013) or as it is commonly known the ''factor loadings''. In examining the factors extracted, some of the variables will have higher factor loadings on a specific factor and smaller loadings on other factors. Regarding the value of factor loadings, values of 0.45 or above are considered to be high enough to be included in a factor (Hair et al., 1995). However, the fact that the same variable has factor loadings in different factors often makes the interpretation of the factor analysis result difficult (Field, 2013). For that reason, before extracting the factors, we must decide the rotation methodology that we will use in our analysis. The rotation of the factors helps us to better interpret the results of the factor analysis. Varimax rotation is usually selected when we want the factors to be uncorrelated and Oblique rotation is used when we want our factors to be correlated. In this study, we will use the varimax rotation.

Validity assessment for the Climate Change Strategy Questionnaire
Regarding our sample size (n=188), we adopt Hair's rule for the number of observations to be five times the number of variables in the initial structure.
Furthermore, as it is shown in Table 16, the factor loadings of the variables are all above 0.6, which, according to Guadagnoli and Velicer (1988), means that the factors extracted are reliable regardless of sample size. Moreover, the Kaiser-Meyer-Olkin Measure of Sampling Adequacy is 0,869 which indicates high sampling adequacy.
Regarding the correlations between the variables, Table 17 shows that the majority of correlations are above 0.3 and at the same time not above 0.6 which means that factor analysis will be able to extract reliable factors. Finally, Barllett's test of sphericity is significant (p<=0.000) and the Determinant is above the critical value of 0.00001 (it is 0.000453 to be precise), which indicate the absence of multicollinearity between the variables.
In Table 16 we can see the results of the factor analysis. Factor loadings for the variable items are above 0.6 which means that the factors created are highly reliable. In Table 22 we can see the factors as they have been constructed after the analysis. In specific, the first factor combines the three initial critical factors, namely, Climate

Change Governance, Climate Change Risk Management Integration and Carbon
Reduction Targets, as well as two items from the Carbon Reduction Strategies factor, Replacement of carbon -based products by non-carbon based products, Implementation of end-use energy efficiency process and Optimization of current business processes. We can observe that the items included in this factor are highly associated with the managerial dimension of climate change since they include organizing, target setting and process and product development dimensions regarding climate change. Therefore, we are going to name this factor Climate Change Organizational Management Strategies. The second factor includes the carbon reduction strategies that are not included in the previous factor. They include the implementation of technologies that reduce carbon emissions either through increased energy efficiency (Items 12 to 15) or through control of GHG emissions (items 16 and 20). We are going to name this factor Carbon Reduction Technology Strategies.
Finally, the last factor remains the same as the initial critical factor named Carbon Compensation Strategies.

Validity assessment for the Risk Management Questionnaire
The sample size (n=168) for this questionnaire is also in accordance to Hair's rule.
Moreover, the Kaiser-Meyer-Olkin Measure of Sampling Adequacy is 0,843 which indicates high sampling adequacy. Regarding the correlations between the variables, Table 19 shows that the majority of correlations are above 0.3 and at the same time not above 0.6 which means that factor analysis will be able to extract reliable factors.
Finally, Barllett's test of sphericity is significant (p<=0.000) and the Determinant value is 0.000271 (> 0.00001) which indicate the absence of multicollinearity between the variables.
In Table 18 we can see the results of the factor analysis. Factor loadings for the variable items are above 0.6 which means that the factors created are highly reliable. In Table 23 we can see the factors as they have been constructed after the analysis. In specific, the first factor consists of the initial Regulation Risk Management factor and the Competition Risk Management factor regarding Uncertainty in Markets Signals.
This means that -according to the survey respondents -risk management regarding changing market conditions is considered to be closely related to changes in regulation or that it is highly influenced by changes in regulation. Therefore, we will name this factor Regulation -induced Risk Management. On the other hand, changing consumer behavior, which is the other factor in Competition Risk Management, is included in the second factor along with Reputation Risk, which means that risk management regarding changes in consumer preferences is highly affected by the reputation of the company or the sector in which it operates. We will name this variable construct   In Table 20 we can see the results of the factor analysis. Factor loadings for the variable items are above 0.6 which means that the factors created are highly reliable. In Table 24 we can see the factors as they have been constructed after the analysis. In specific, the first variable construct consists of the initial Identification and Calculation of Carbon Emissions Regulation critical factor except from ACCOUNT_ITEM_05, regarding Scope 3 emissions calculation, which is included in the second factor.

Reputation -induced Risk
Furthermore, items 10, 11 and 13 regarding external verification of carbon emissions, reasonable assurance verification and mandatory reporting are also included in the first factor. The second variable structure includes the initial Internal Price of Carbon critical factor and the scope 3 emissions calculation item. If we look at the descriptive statistics of the initial variables (Table 15) we can observe that the variables in the second factor have much higher mean values than the rest of the variables, which means that are generally more difficult to implement than the items included in the first factor. Therefore, we will name the final variable constructs as Basic Carbon Accounting and Reporting Practices and Advanced Carbon Accounting Practices, for factors one and two respectively. Finally, regarding ACCOUNT_ITEM_09 and ACCOUNT_ITEM_12, since they have low factor loadings in comparison to the other variables, we decided not to include them in the final variable structures. Reliability indicates the level of dependency, stability, predictability, consistency and accuracy of a measurement instrument (Kerlinger, 1986). Reliability is estimated with different methods. According to Nunnaly (1978) and Carmines and Zeller (1979)   The internal consistency method mathematically calculates the reliability of the measurement instrument. "It assesses the equivalence, homogeneity and intercorrelation of the items used in a measure. This means that the items of a measure should hang together as a set and should be capable of independently measuring the same construct." (Forza, 2002 ). Internal consistency of a structure is measured with the Cronbach alpha coefficient (Cronbach, 1951). Cronbach's alpha calculates the average inter-item correlation among the items included in the measurement instrument. It is therefore related both to the level of correlation between the items as well as the number of items included in the construct. According to Nunnally (1978), the construct's Cronbach alpha coefficient should be over 0,800 in order for the construct to yield reliable results. Also, the item-related Cronbach alpha coefficients should be lower than the construct's total Cronbach alpha. A higher item-related Cronbach alpha coefficient means that if the specific item is removed from the construct, then the reliability of the construct will increase.
In the previous section we assessed the validity of the questionnaires used in order to calculate the weights that will be used in our corporate indexes. In doing so, we employed factor analysis and developed new variable constructs. In this section, the factors developed will be assessed for reliability by using the internal consistency method. Internal consistency for the climate change strategy, risk management and carbon accounting is depicted in Table 22, Table 23 and Table 24 respectively. As we can observe, the item-related Cronbach alpha coefficients of each variable construct are lower than the construct's total Cronbach alpha. Therefore, we can consider the factors extracted to be reliable.   In order to measure the level of commitment of each company related to the aspects of climate change management described above we are going to use the mean average values of the items included in each index as weights. The result will be the development of weighted corporate climate change indexes that will accurately and reliably measure a company's level of commitment regarding specific climate change practices. The companies that will be assessed will be Oil and Gas Companies and Banks included in the Financial Times Global 500 data set for the years 2012-2015.
The assessment was based on the CDP climate change reports of the companies. As we have discussed in previous section of this study, the CDP is a global not-for-profit organization, which provides a global disclosure system through which thousands of firms around the world report, manage and share environmental information.
Furthermore it holds the largest and most comprehensive collection globally of primary corporate climate change information (CDP, 2015). Reports were assess for the years 2012-2015.
In the following section we will describe the scoring methodology used in measuring the level of commitment of each company for each respective index.

Scoring Methodology
The scoring methodology used in this study was based on a twostep procedure described below. First, we conducted a content analysis of the CDP reports for each company by using the items included in the climate change indexes described above.
''Content analysis has been defined as a systematic, replicable technique for compressing many words of text into fewer content categories based on explicit rules of coding (Berelson, 1952;GAO, 1996;Krippendorff, 1980;Weber, 1990). Content analysis enables researchers to sift through large volumes of data with relative ease in a systematic fashion (GAO, 1996). It can be a useful technique for allowing us to discover and describe the focus of individual, group, institutional, or social attention (Weber, 1990).'' (Stemler, 2001). The method of content analysis has been used by many authors in studying environmental management and environmental disclosures (Freedman and Jaggi, 2005;Clarkson et al., 2008;Brammer and Pavelin, 2008;Prado-Lorenzo et al., 2009;Tagesson et al., 2009;Gallego-Alvarez, 2010;Haque and Deegan, 2010;Galani et al., 2011;Martínez-Ferrero et al., 2013). We will also apply this methodology in order to analyze information provided in the CDP reports of Oil For example, figure 8 is a snapshot of the CDP report of HESS CORPORATION, one of the largest oil and gas produces in the U.S. In this case items related to climate change market-based regulation will receive the value of 1 for ″Identify impacts from market-based regulation″, ″Calculate financial implications from market-based regulation″, ″Identify management methods in order to address impacts from marketbased regulation″, ″Calculate the cost of management methods needed to address impacts from market-based regulation″. In another example, figure 9 is a snapshot of the CDP report of TOTAL, the largest oil producer in France. In this case items related to climate change market-based regulation will receive the value of 1 for ″Identify impacts from market-based regulation″, and ″Identify management methods in order to address impacts from market-based regulation″, and 0 for ″Calculate financial implications from market-based regulation″ and ″Calculate the cost of management methods needed to address impacts from market-based regulation″. After content analysis was conducted for all the sample companies for the years 2012-2015, the data were weighted with the respective mean average value. The reason why we decided to weight the content analysis data is based on the fact that not all climate change practices have the same level of difficulty to be implemented. For example, identifying the impact of marketbased regulation has a mean difficulty value of 4.54 which is lower than the mean difficulty value of calculation the financial implications of market-based regulation (4.77). Therefore, in order to gain a better understanding of the level of corporate commitment regarding climate change practices, we will weigh the data obtained by the content analysis according to the mean average values.
In Table 25 and Table 26 and we provide an example, which explains the reason why we decided to weigh the results of content analysis. risks, but focuses on identifying impact and management solutions as well as calculating the cost for these items. If we were to employ simple content analysis, then corporate level of commitment for both companies would be the same ie. 8. However, when we apply weights to the content analysis data then we observe that the level of corporate commitment for company B (37.75) is higher than that of company A (33.58).

Scoring Methodology for the Banking Sector
At this point we would like to clarify a bit on the scoring methodology used when assessing the climate change practices of companies belonging to the banking sector.
Since the banking sector is a services sector, we did not assess climate change practices that are directly related to a bank's operational activities. What we mean is, for example, regarding the ''Energy source switching, from fossil fuels to renewable energy sources'', what we did assess was whether the bank invests on renewable energy sources or encourage its clients to invest in these types of carbon reduction technologies, such as combined heat and power or expand its business to the natural gas industry. Direct emissions reduction strategies from a bank's operational activities are not included in the Carbon Reduction Technologies Strategies rather in the Organization Management Strategies Index under the items: Implementation of enduse energy efficiency processes and Optimization of current business processes in order to reduce CO 2 emissions. Therefore, the Banking strategies for reducing CO 2 emissions were focus on developing and expanding green banking activities, which basically means promoting on-line banking services to reduce paper consumption or the reducing unnecessary traveling by using online communication media or by installing energy efficient equipment for lighting premises etc. Therefore, as we also going to see in Table 27 and Table 28, the maximum score for banks regarding carbon reduction strategies is very low in comparison to this of oil and gas companies.
Moreover, regarding Carbon Compensation Strategies, which include emissions trading and creating or purchasing project-based carbon credits, a score would be given when the bank either facilitated the buy and selling of carbon credits for its clients or when it took into its financial assessment of a company potential liabilities which could arise due to the lack of compliance to the requirements of emissions permits. This also the same, regarding regulation-induced risk management. Corporate commitment regarding regulation risk management did involved the direct effect of climate change regulation on the bank itself, as banks do not belong in carbon intensive sectors, but rather the indirect effects through increased compliance costs for their customers, which could potential alter their risk assessment profile, for example in loan granting etc. Physical Risk Management has also to do with the risk profile of a bank's clients and how vulnerable their business is to the natural adverse effects of climate change. So basically, a high score for a bank in the Physical Risk Management Index would mean that the bank devotes a considerable amount of resources in identifying and assessing climate change physical risks related to their clients' nature of business activities. Moreover, regarding a bank's level of corporate commitment related to reputation-induced climate change risks, this is defined in the strict borders of the bank as a company and how it is viewed by the public, for example whether it invests on carbon-friendly technologies etc. Finally, for carbon accounting, the scores produces are also strictly related to the activities of the bank and not its clients.

Analysis of the scoring results
Building on the above, we are going to use the weighted content analysis methodology in order to assess the corporate commitment of our sample companies. Table 27 presents the descriptive statistics for the oil and gas companies and Table 28 presents the descriptive statistics for the banks, for all the climate change indexes. According to above descriptive statistics, the level of corporate commitment regarding climate change practices varies to a great extent between companies in both sectors. Regarding carbon reduction technology strategies, we can observe that the level of corporate commitment for companies that belong to the oil and gas sector is much higher than that for the companies that belong to the banking sector. This is also the same for regulation-induced climate change risk management, carbon compensation strategies and advanced carbon accounting practices. These results can be easily explained if we take into account the fact that oil and gas companies are directly affected by climate change regulation since they belong to the highest carbon intensity business sector, which explains the regulation-induced climate change risk management effect. Also the fact that in certain countries, such as the European Union, participation in carbon trading schemes (which is included in the carbon compensation index) is mandatory, also explains the high level of corporate commitment regarding carbon compensation strategies.
In accordance with the above, corporate commitment regarding advanced carbon accounting practices is also much higher for the oil and gas companies. This particular climate change index, measures the level of commitment of a company to account for CO2 emissions across its value chain (Scope 3 emissions) as well as its ability to develop and implement internal price of carbon schemes. Therefore, since oil and gas companies are directly affected by climate change regulation and since the purpose of climate change regulation is to put a price on carbon emissions, it is essential for companies in the oil and gas sector to incorporate the cost of carbon emissions in their business activities.  is a global not-for-profit organization, which provides a global disclosure system through which thousands of firms around the world report, manage and share environmental information. Furthermore it holds the largest and most comprehensive collection globally of primary corporate climate change information (CDP, 2015).
For the purpose of this study, we obtained the GHG emissions from the CDP's data on emissions. Both scope 1 and scope 2 emissions were used in this study. Scope 1 GHG emissions are those who come directly from sources that are owned or controlled by the reporting firm. Scope 2 GHG emissions are a consequence of the activities of the reporting firm, but occur at sources owned or controlled by another firm (CDP, 2014).
In order to test our two hypotheses we have selected, as the target population, firms in the Financial Times Global 500 data set. The economic sectors selected to undertake this research were Oil and Gas Companies and Banks. A sample of 275 firms was chosen and data was selected for the years 2012-2015. Firms were selected on the basis of having both a disclosure score rating and providing data on their emissions according to the CDP. Accounting data were selected from the financial and annual reports of our sample firms.

Dependent Variables
Firm financial performance is the depended variable in our regression model. In Tobin's q (Nishitani and Kokubu, 2012;Busch and Hoffman, 2011;Wang et al., 2014), share/stock price ratio (Al-Najjar and Anfimiadou, 2012) and standardized cumulative abnormal returns (Yadav et al., 2015) have been used to measure the corporate performance of a firm. Freedman and Jaggi (2005) Following the research of both Hoffman, (2011), Clarkson et al. (2008) and Nakao et al., (2007), we are going to use two accounting measures, ROA and ROE, and one market performance measure, Tobin's q, to examine a firm's corporate performance. We will use ROA and ROE because they are the most commonly used measure regarding the efficiency of tangible assets (Busch and Hoffman, 2011).
Additionally, Tobin's q is considered as an indicator of intangible assets (Busch and Hoffman, 2011), since it reflects market expectations, reputational effects and financial risks (Wang et al., 2014;Busch and Hoffman, 2011). We define ROA as the net profits of a firm divided by its total assets, ROE as the Stakeholder's Equity divided by total assets and Tobin's q as the sum of market value, book value of long-term debt and net current liabilities divided by the book value of total assets.

Independent Variables
In this study, we are going to use nine different independent variables, the climate change corporate commitment indexes as they have been described in previous sections of this study and carbon reduction variation. Previous research has examined different measures for environmental performance. Nakao et al., (2007)  variation between years to measure the environmental performance of firms. Finally, Busch and Hoffman (2011) have used the natural logarithm of carbon intensity (carbon emissions divided by sales), divided by median of the sample firms. In our study, we will also use emission variation between years to measure the climate change performance of firms.
With regard to measuring a firm's leverage, Busch and Hoffman, (2011) Brammer and Pavelin (2008), Stanny and Ely (2008) and Clarkson et al. (2008) have used a debt/asset ratio. In this study, we will also use a debt/asset ratio. Regarding capital intensity, Wang et al., (2014) have used a capital expenditure to sales ratio, and Nakao et al. (2007) have used sales to total assets ratio as a proxy variable for capital intensity. Following Nakao et al. (2007), sales to total assets ratio as a proxy variable for capital intensity.

Regression analysis
Multiple regression analysis was conducted in order to test the relationship between corporate commitment regarding climate change strategy, climate change risk management, carbon accounting, GHG reduction and corporate financial performance. Size, calculated as the natural logarithm of total assets, Leverage, expressed as the ratio of debt to total assets and Capital Intensity, represented by the ratio sales to total assets were used as control variables.
Our models can be written as:   Finally, Reputation and Physical Risk Management also exhibit a high degree of variation ranging from 0,00 to 87,46 and 12,47 to 100,00 respectively. there is no significant correlation between the carbon reductions and ROA and Carbon Compensation Strategies with the financial performance indicators.

Regression Analysis
In order to estimate the regression model, several statistical assumptions of the regression analysis were used. Regarding normality, we applied the Kolmogorov-Smirnov test which showed us that the variables generally follow a normal distribution (Table 30). However, for some of the variables that do not show normal distribution, according to Gallego-Alvarez (2012) and Lumley et al. (2002), the absence of a normal distribution does not reduce the validity of the model. In order to alleviate heteroscedasticity problems, we transformed some of our variables into a logarithm.
Regarding autocorrelation, we conducted the Durbin-Watson's test. Our models exhibit values around 2 (Table 32, Table 34, Table 36 and Table 38) which reflects the absence of autocorrelation in the residuals. Finally, in the case of multicollinearity, the values obtained in tolerance have to be high (0< tolerance<1) and the values obtained in the variance-inflation factors (VIF) have to be low (VIF<10). Our model presents tolerances between 0,579 and 0,954, and VIFs between 1,069 and 1,330, indicating the absence of multicollinearity (Table 33, Table 35, Table 37, Table 38).               Regarding the explanatory power of our models, the R 2 have values ranging from 0,283 to 0,515 for a confidence level of 99% (p < 0.01). The explanatory power of our models is similar to other studies. Specifically, Lannelonge et al. (2015), Gallego-Alvarez et al. (2012), and Busch and Hoffman (2011), obtain values between 0.16 and 0,43 for the R 2 while Wang et al. (2014), Nishitani and Kokubu (2012) and Nakao et al. (2007) obtain R 2 values of 0,411, 0,104 and 0,411 respectively.
The results obtained from the estimation of our proposed models using the ordinary least squares methodology are presented in Table 32, respectively. Reputation and Physical Risk Management also exhibit a high degree of variation ranging from 0,00 to 86,78 and 00,00 to 100,00 respectively. The Pearson's correlation matrix for the regression variable is presented in Table 40 It is worth mentioning that the correlations between carbon reduction and Tobin's q are very high. Correlations are also very high for organizational management strategies, ROA and ROE and for carbon technology reduction strategies and Tobin's q.
Regarding the climate change risk management indexes, the highest correlations are detected between physical risk management, ROA and Tobin' q while there is also a strong correlation between core business accounting and reporting and ROA. *. Correlation is significant at the 0.05 level (2-tailed).

Regression Analysis
In order to estimate the regression model, several statistical assumptions of the regression analysis were used. Regarding normality, we applied the Kolmogorov-Smirnov test which showed us that the variables generally follow a normal distribution (Table 41). However, for some of the variables that do not show normal distribution, according to Gallego-Alvarez (2012) and Lumley et al. (2002), the absence of a normal distribution does not reduce the validity of the model. In order to alleviate heteroscedasticity problems, we transformed some of our variables into a logarithm.
Regarding autocorrelation, we conducted the Durbin-Watson's test. Our models exhibit values around 2 (Table 42, Table 44, Table 46 and Table 48), which reflects the absence of autocorrelation in the residuals. Finally, in the case of multicollinearity, the values obtained in tolerance have to be high (0< tolerance<1) and the values obtained in the variance-inflation factors (VIF) have to be low (VIF<10). Our model presents tolerances between 0,674 and 0,915, and VIFs between 1,092 and 1,527, indicating the absence of multicollinearity (Table 43, Table 45, Table 47, Table 48).               The results obtained from the estimation of our proposed models using the ordinary least squares methodology are presented in Table 42,   (Table 53, Table 55, Table 57, Table 58).         The results obtained from the estimation of our proposed models using the ordinary least squares methodology are presented in Table 52, Table 54, Table 56 and Table 58 The models estimated determine the relationship between financial performance, climate change strategy, climate change risk management, carbon accounting and GHG reduction of business activities. A positive and significant relationship is found between carbon reduction and ROA for a significance level of 90% while a positive and significant relationship is also detected between carbon reduction and Tobin's q for a

Regression Analysis
In order to estimate the regression model, several statistical assumptions of the regression analysis were used. Regarding normality, we applied the Kolmogorov-Smirnov test which showed us that the variables generally follow a normal distribution (Table 60). However, for some of the variables that do not show normal distribution, according to Gallego-Alvarez (2012) and Lumley et al. (2002), the absence of a normal distribution does not reduce the validity of the model. Regarding autocorrelation, we conducted the Durbin-Watson's test. Our models exhibit values around 2 (Table 62,   Table 64, Table 66 and Table 68 ), which reflects the absence of autocorrelation in the residuals. Finally, the values obtained for tolerances are between 0,650 and 0,967, and values obtained for VIFs are between 1,036 and 1,175, indicating the absence of multicollinearity (Table 63, Table 65, Table 67, Table 68)         Regarding the explanatory power of our models, in models S1 -S4, R1 -R4 and A1 -A3, where ROA is the dependent variable, the R 2 has a value ranging between 0,310 and 0,565 for a confidence level of 99% (p < 0.01). The explanatory power of our models is relatively high to similar studies. The results obtained from the estimation of our proposed models using the ordinary least squares methodology are presented in Table 62, Table 64, Table 66 and Table 68.
A positive and significant relationship is detected among carbon reduction and all financial performance indexes, with Tobin's q having the strongest relation to carbon reduction (0,389). Positive significant relationship between ROE and carbon reduction is detected for a confidence level of 90%. This allows as to accept hypotheses H 4 .
Moreover, a positive and significant relationship is detected between all organizational management, carbon reduction technologies and all financial performance indexes.
Contrary to the period 2012-2013, a positive but no significant relationship is detected between carbon compensation strategies and financial performance.
Regarding carbon reduction technologies, the strongest relationship is observed for Finally, it is noticeable, that carbon compensation strategies, which include participation in emission trading schemes and buying and selling carbon offsets, are not significantly related to financial performance of companies for the period 2012-2013 while they become significantly associated to financial performance in the period 2014-2015. This can be explained by the fact that in 2012 the second phase of the EU ETS ended. As we have discussed above, during phases 1 and 2 a large number of carbon emission allowances were initially granted to companies participating in the EU ETS for free. However, from 2013 and afterwards, the number of initial free allocation allowance decreased, which added an additional cost to carbon intensive firms. Furthermore, the costs of no compliance to the emissions allowed for each company increased while the amount of emissions allowed for each company has been steadily decreasing from year to year. As a result, companies have begun to place greater attention to carbon trading schemes and emission allowances which explains the difference between the significance levels regarding the relationship between financial performance and corporate commitment in implementing carbon compensation strategies.
Regarding corporate commitment in climate change risk management, we can observe an increase in the significance level between regulation-induced risk management and financial performance from the period 2012-2013 to the period 2014-2015 which can be attributed to the fact that climate change regulation has increased during the last years, with the introduction of carbon taxes and the establishment of carbon trading schemes, whose rules, are becoming more strict from year to year. Reputation remains a highly significant risk related to corporate performance in both time periods.
However, as we can observe in Models R4, R8 and R12 its significance in relation to the other types of risk management has decreased from the period 2012-2013 to the period 2014-2015. On the other hand, the significance of physical risk management has also increased from the first to the second period. This means that companies in the oil and gas sector have begun to treat climate change implications at a more substantial level and not purely as a CSR issue. That is not to claim that reputation risk management does not contribute to increased financial performance, but rather that attention has been shifted towards more direct risk management issues, such as adhering to climate change regulation and managing physical implications of climate change.
Finally, regarding corporate commitment on carbon accounting, a positive and significant relationship between Core Business Carbon Accounting and Reporting, Advanced Carbon Accounting Practices and financial performance is detected at both time periods. Nevertheless, there is a tendency for this relationship to become more significant in the period 2014-2015 in terms of ROA. On the other hand, relationship between the carbon accounting indexes and Tobin's q is found to be highly significant in both time periods. This is closely linked to the increased demand for disclosure regarding carbon emissions of companies or carbon intensity of business activities.
The fact that the relationship between carbon accounting and market performance was of higher significance than this between accounting performance and carbon accounting in the first time period, means that it was the market's demand (or the stakeholders' demand), who began to value the importance of corporate commitment of carbon accounting earlier than the firms themselves, and who eventually demanded for more information regarding corporate carbon accounting practices.

Discussion of Results in the Banking Sector
Climate change strategy corporate commitment, organizational management strategies and carbon reduction strategies are found to be positively and significantly correlated to financial performance in both time periods. Furthermore, we can observe that the significance level increased from the first to the second time period. This is in accordance with the general trend to invest both in carbon friendly technologies, to include sustainability criteria in the financing of projects and to develop business processes that reduce energy consumption, such as promoting e-banking services.
Moreover, in the transition from the first to the second time period we observe an increase in the significance level of organizational management strategies related to both ROA and Tobin's q. As these two particular financial indexes reflect stakeholder attitudes towards corporate performance (with ROA taking expressing shareholders' interests and Tobin's q reflecting market expectations on corporate performance), we can claim that there is an increasing demand from stakeholders to demand higher corporate commitment regarding climate change strategies. Furthermore, the fact that the carbon reduction technology index has the strongest connection to Tobin's q at both time periods, means that the market generally rewards companies who invest in innovation technologies or finance technological innovation projects.
Regarding carbon compensation strategies, their relationship to corporate performance is found to be positive and significant in the period 2012-2013 but insignificant in the period 2014-2015. Carbon compensation strategies include participation in emissions trading schemes, creating project-based carbon credits and purchasing carbon credits. These practices are of course not directly related to the operational activities of banks, since banks are not considered to be a carbon intensive sector. Therefore, the assessment of this index refers to the services provided by banks to other companies related to the specific climate change strategies. If we examine this index form this point of view, we can interpret the high significance level between financial performance and carbon compensation strategies during the first time period as well as the lack of significance during the second time period.
During the first time period, the provision of services regarding emissions trading or buying and selling emission allowances could have possibly been considered as a competitive advantage for banks who decided to provide these types of services, therefore enhancing their financial performance. However, as years passed from the introduction of emissions trading schemes, the provision of related services gradually seized to be a competitive advantage since it became a mainstream service provided by the majority of financial institutions. As a result, the relationship between carbon compensation strategies and financial performance in the second time period seized to be significant.
In examining the trends between corporate commitment in climate change risk management and financial performance, we observe that the effect of regulation risk management on financial performance slightly decreases, while on the other hand the effect of physical risk management on financial performance increases. The decrease in the significant level of regulation risk management can be explained by taking into account the fact that, as years pass, climate change regulation becomes more clear and substantial, making it easier for financial institutions to incorporate climate change performance criteria into their investment decisions. Therefore, as we have discussed above about the relationship between carbon compensation strategies and financial performance, in the same line of thought, the effect of corporate commitment in regulation risk management also decreases. On the other hand, the fact that the relationship between physical risk management and financial performance becomes stronger can be explained by the fact that as years pass, the incurrence of extreme weather events and changes in precipitation patterns becomes more evident, thereby affecting the business activities and profitability of the majority of firms. For that reason, devoting management effort in assessing the exposure of their clients to the physical risks of climate change will probably have a strong effect on the financial performance of banks. As far as reputational risk management is concerned, we can observe that its significance increases from the 2012-2013 period to the 2014-2015 period, in relation to ROE and Tobin's q which also enhances our claim that corporate stakeholders have begun to pay more attentions to the corporate practices of financial institutions regarding climate change management.
Finally, regarding the relationship between carbon accounting and financial performance, this becomes significantly stronger from the first to the second time period for all financial performance indicators. As discussed above, this confirms the increased demand form stakeholders for more information regarding the corporate climate change practices of firms, which explains the strong relationship between core business carbon accounting and reporting and both ROA, ROE and Tobin's q.
Furthermore, it highlights the need for incorporating advance carbon accounting practices, including the development of internal price of carbon schemes in order for financial institutions to accurately measure the carbon profile of a potential client and to assess its exposure to the adverse effects of climate change.

CONCLUSION
Climate change has been globally acknowledged as a major source of physical, economic and social risk. The mandatory GHGsreduction targets that have been set for the industrialized countries under the auspices of the Kyoto Protocol have given rise to the adoption of climate change mitigation policies at national and international level. These include putting a price on carbon emissions through setting process and product standards for industry, carbon taxes and the establishment of carbon trading programs such as the EU ETS. As a result companies are expected to face increased costs in their production processes both directly for the GHG-intensive sectors and indirectly by passing carbonrelated cost through the supply chain, thereby affecting the profitability of the majority of firms.
These increased risks that companies face have also caught the attention of various stakeholders, such as institutional investors, banks, accounting firms, governmental agencies, NGOs and consumers who have been demanding information on companies' part, regarding their corporate climate change practices. The aim of this study was to examine the various climate change management practices adopted by firms and how these practices affect their financial performance. In order to do so, we examined three dimensions of climate change management: Strategy, Risk Management and Carbon Accounting and Reporting. Our claim was that devoting corporate resources in the implementation of climate change strategies, in the identification, assessment and management of climate change risks and in the accurate and precise accounting and reporting of carbon emissions enhances the financial performance of firms. Furthermore, we proposed that reducing CO 2 emissions also enhances corporate financial performance.
In order to prove the validity of our claims we developed eight corporate indexes for In order the measure the level of corporate commitment regarding its climate change practices we decided to weight the corporate indexes according to the level of difficulty in implementing the climate change practice included in each respective index. In order to assign proper weighs to our indexes we developed and validated critical factors of climate change strategy, climate change risk management and carbon accounting by conducting separate surveys for climate change strategy, risk management and carbon accounting and reporting. After collecting our data, we conducted validity assessment, by using exploratory factor analysis, and reliability assessment, by using the internal consistency method, of our survey instruments.
According to the results of this process, we will developed the respective climate change indexes described above and assigned relevant weights according to the mean average values of the variables included in each index.
After this procedure we tested the indexes in two different business sectors: the Oil and Gas sector and the Banking sector for the years 2012-2015. The assessment of the selected companies was conducted through a predefined scoring methodology that was based on weighted content analysis. After the extracting the relevant scores for its company, we separated our initial sample into two sub-sample, one for the years 2012-2013 and one for the years 2014-2015. We then examined the relationship between the respective indexes and corporate financial performance of our sample firms for the two different time periods, by using linear regression analysis. The financial performance indicators used in this research were ROA, ROE and Tobin's q.
Regarding the oil and gas companies, for the first times period, a positive but no significant relationship was found between carbon reductions and ROA while a positive and significant relationship was found between ROE and Tobin's Q.
Moreover, a positive and significant relationship was detected between organizational management strategies, carbon reduction technology strategies and financial performance measured by both ROA, ROE and Tobin's q while a positive but no significant relationship was observed between carbon compensation strategies and the financial performance measures. Regarding the relationship between corporate commitment in climate change risk management, a positive and significant relationship was detected between all the risk management indexes and the financial performance measures, with the exception of physical risk management and ROE, where no significant relationship was detected. Furthermore, the strongest relationship was detected between reputation risk management and financial performance.
Additionally, all risk management indexes had the strongest connection to the Tobin's q index. Finally, regarding the carbon accounting indexes, they were positively and significantly related to the financial performance indexes, with the strongest relationship detected between the Tobin's q index.
Contrary to the results in the time period 2012-2013, in the second time period a positive and significant relationship was found between carbon reductions and ROA while a positive and significant relationship was also detected between carbon reductions, ROE and Tobin's Q. Moreover, a positive and significant relationship was detected between organizational management strategies, carbon reduction technology strategies and financial performance measured by both ROA, ROE and Tobin's q . In contrast to the previous time period a positive but significant relationship was observed between carbon compensation strategies and the financial performance measures. Regarding the relationship between corporate commitment in climate change risk management, a positive and significant relationship was detected between all the risk management indexes and the financial performance measures. Physical risk management has the strongest influence related to the other risk management indexes in terms of ROA, while regulation and reputational risk management influence financial performance to a higher degree than physical risk management.
Moreover, regarding Tobin's q all types of risk management contribute equally to enhanced financial performance. Finally, regarding the carbon accounting indexes, they were all found to be positively and significantly related to the financial performance indexes, with the strongest relationship detected for the Tobin's q index.
Regarding the banking sector, for the period 2012-2013, a positive and significant relationship was found between carbon reduction, ROA and Tobin's q while a negative but no significant relationship was detected between carbon reduction and ROE. Moreover, a positive and significant relationship was detected between all climate change strategy indexes and the financial performance indexes. Regarding carbon reduction technology strategies, their effect is stronger to financial performance when examined in terms of Tobin's q. All risk management indexes were found to be positively and significantly related to the indexes measuring financial performance. Regulation and physical risk management have the strongest influence on ROA, while reputational risk management has the strongest influence on Tobin's q. Finally, regarding the carbon accounting indexes, the core business carbon accounting and reporting index was positively and significantly related to ROA and Tobin's q, while it was positively but no significantly related to ROE, while the index measuring advance carbon accounting practices was positively and significantly related only to ROA, while it was positively but no significantly related to Tobin's q and negatively but no significantly related to ROE.
Regarding the period 2014-2015, a positive and significant relationship was detected among carbon reduction and all financial performance indexes. Moreover, a positive and significant relationship was detected between organizational management, carbon reduction technologies and all financial performance indexes. Contrary to the period 2012-2013, a positive but no significant relationship was detected between carbon compensation strategies and financial performance. Regarding carbon reduction technologies, the strongest relationship is observed for Tobin's q. All risk management indexes are found to be positively and significantly related to the indexes measuring financial performance, with regulation and physical risk management having the strongest influence on ROA, while reputational risk management having the strongest influence on ROE and Tobin's q. Finally, regarding the carbon accounting indexes, they are all positively and significantly related to all financial performance indexes.
Summarizing, our results show that devoting resources in the implementation of climate change corporate practices enhances corporate performance of firms, and this relationship tends to be even stronger as years pass. Furthermore, we detect a strong steady relationship between carbon reduction technologies and Tobin's q, which reinforces Porter's win-win hypothesis that innovation enhances corporate performance. Our research has a number of limitations. First, in order to weight our corporate climate change indexes, we used survey data based on the opinions of climate change experts. However, although our sample population was formed by experts who work in large multinational corporations, which were also the target population of our oil and gas companies and our banks, the data collected were based on the opinion of experts who work in multiple business sectors. In conducting our survey, we had to compromise with experts working in different sectors and not specific on the oil and gas or the banking sector, due to the fact that our target population was very limited. If we were to conduct our survey only on experts in the specific two sectors, we would have been unable to gather enough data to be able to perform validity and reliability analysis of our questionnaire. We believe that in the future we will be able to locate more experts of each respective sector in order to have more precise responses regarding the difficulty level of implementing climate change practices. Furthermore, future research could also examine other economic sectors, and do comparative research between carbonintensive and non-carbon intensive sectors, as we also did in this study.
Additionally, in the factor analysis regarding climate change risk management, an additional factor, namely legal risk management was extracted. However, in the analysis of the climate change practices of our sample firms, we did not examine corporate commitment regarding legal risk management. The reason we decided not to examine legal risks is because we couldn't find reliable data regarding corporate practices in this particular field. Future research could examine the relationship between climate change legal risks related and financial performance, especially as climate change regulation evolves and becomes more precise and strict. Finally, our research was focused on large multinational companies and on their corporate climate change practices. Future research should also focus on examining the climate change practices of smaller firms and the capabilities they have in implementing climate change strategies, identifying risks and tracking down their carbon emissions.