Companies have been providing information on their interactions with the environment since the 1880s however there has been an upsurge in the amount of information they provide over the last three decades. This has been driven by a variety of legislative, internal, social drivers. These corporate environmental disclosures appear within three key channels and media: via the Internet, in their AnnuaVFinancial Report and in a stand alone Environmental Report. The USA was the fIrst country to introduce statutory requirements relating to the disclosure of environmental information in the AnnuaVFinancial Report - Securities Release 5235 and 5386. Many other countries, including the UK, are now considering making environmental disclosure, in some shape or form, mandatory. Benchmarking studies are frequently applied identify the leading companies in terms of Environmental Reports; companies within the utility sector are repeatedly cited as leaders. However most research and benchmarking studies compare what themes have been disclosed in a single channel or media, for example, emissions, waste, energy, and do not examine the quality of the disclosures or compare those disclosures in the different channels and media. Recently, qualitative characteristics relating to the quality of environmental disclosures have been developed including, inter alia, accessibility, credibility and inclusivity. This thesis therefore aims to (1) assess whether utility companies are also leaders in environmental disclosure via the Internet and in their AnnuaVFinancial Reports, (2) compare the effects of a voluntary and mandatory disclosure requirements on the environmental themes and qualitative characteristics disclosed in the AnnuaVFinancial Reports and Environmental Reports of electric power utilities based in the UK and USA and applies Lasswell's model as an integrating framework for (1) and (2) in order to assess what, if any, inferences can be made from a direct content analysis relating to 'why' and 'with what effect' environmental disclosures are made. Content analysis is applied to meet the three main aims of this thesis. First, a basic appearance or nonappearance form of content analysis is applied on 275 (utility and non-utility) companies to assess whether utility companies are also leaders in environmental disclosure via the Internet. The amount location and type of environmental themes that were disclosed within the AnnuaVFinancial Reports of 100 (utility and non-utility) companies were then identifIed via content analysis to assess whether utility companies are also leaders in environmental disclosure in their Annual Report and Financial Report. A more complex frequency content analysis was applied to assess the environmental themes and the qualitative characteristics disclosed by 20 electric power companies based in the UK and USA in their AnnuaVFinancial Reports and Environmental Reports to evaluate the affect of voluntary and mandatory disclosure requirements. The research shows that no companies are fully exploiting the capabilities of the Internet for environmental disclosure and that no one sector stands out as a leader. The oil, gas, coal and related services sector, not the utility sector, are leaders in terms of environmental disclosure within the AnnuaVFinancial Report, indeed there does not appear to be any consensus within the utility sector regarding the environmental themes that should be disclosed. Mandatory disclosure requirements heavily influence the amount, themes, balance, character and specifIcity of environmental disclosures made by of electric power companies based in the USA who appeared to target their disclosures in the AnnuaVFinancial Report and Environmental Report at different audiences. Electric power companies based in the UK, where there are no mandatory disclosure requirements, disclosed double the amount of information in their Environmental Report than companies based in the USA but disclosed less information in their AnnuaVFinancial Report which was virtually a subset of those in their Environmental Report in terms of the amount, themes, balance, character and specifIcity. It can be concluded from the application of Lasswell's model that environmental disclosures are made due to a combination of drivers, the types and qualitative characteristics of those disclosures depend on the relative importance of those drivers. More importantly the environmental disclosures have little effect due to their lack of credibility, and environmental disclosures, in particular, via the Internet were found to be unduly burdensome in terms of their accessibility. Stakeholders will fmd it difficult to create an informed image until there are generally accepted or mandatory standardised data collection methods, measurements and presentation for collating the environmental information itself and for the verifIcation statement therefore they need to be aware of the limitations of corporate environmental disclosures, be critical of the information and go back to the company for more details on any area of specific interest. Although companies may not want to allow stakeholders to create an informed image, guidelines are presented that can be used by companies to improve the accessibility and credibility of their environmental disclosures.

To combat climate change multiple initiatives have been launched to steer the financial market towards a more sustainable and resilient path. For example the Montreal Pledge that have committed over 120 investors to measure and disclose their carbon footprints of their portfolios. ISS-Ethix Climate Solution provides climate change related services to investors. In order to evaluate companies’ sustainability ISS-Ethix Climate Solution estimates companies’ direct and indirect greenhouse gas emissions. To simplify these estimations, the emissions from corporations are divided into three scopes, where scope 1 and 2 cover the emissions from the combustion of fuels used in the company and electricity generation. Scope 3 corresponds to all other emissions generated upstream and downstream the companies’ supply chain. The aim of this study was to help ISS-Ethix Climate Solution to develop a model that estimates the indirect scope-3-emission intensity for companies in the automobile sector, fossil fuel sector and utility sector. The first objective was to examine if the variations within the sectors could be explained and categorized. To carry this out each sector was defined and their emission sources identified. The emissions could be explained and categorized for the automobile sector and fossil fuel sector. However, the emissions for the utility sector could only partly be explained and categorized. The second objective was to examine which parameters and subcategories were relevant for estimating the emissions. Two methods were investigated to carry out the second objective; correlation analysis and the average-data method. No correlations could be found between any of the sectors and the selected parameters. The estimated emissions using the average-data method were verified to the companies reported emissions. For the automobile and the fossil fuel companies the estimated emissions followed the same trend as the reported data. However, no trend could be found for the utility companies. Estimating emissions using the average-data method requires a certain corporation structure. The method can be used for corporations with a specific output, but does not suit corporations with a more complex structure. The largest limitation with the models was the information shortages from the corporations. Therefore increased transparency from the companies is a necessity in order to develop the models. / För att minska klimatförändringen har ett flertal initiativ lanserats för att göra finanssektorn mer hållbar. Tillexempel Montreal förbindelsen som har fått över 120 investerare att mäta och publicera klimatutsläppen i sina aktieportföljer. Företaget ISS- Ethix Climate Solution erbjuder klimatrelaterade tjänster för investerare. För att värdera hur hållbart ett företag är estimerar ISS-Ethix deras direkta och indirekta utsläpp av växthusgaser. För att förenkla dessa estimeringar är utsläppen indelade i tre så kallade scopes (områden), där scope 1 och 2 motsvarar emissionerna som genereras av att företaget förbränner fossila bränslen och deras elanvändning. Scope 3 motsvarar alla utsläpp som sker uppströms och nedströms företagens leverantörskedja. Syftet med denna studie var att hjälpa ISS-Ethix Climate Solution att utveckla en modell som estimerade scope 3 utsläppen från företag inom fordonssektorn, fossila- bränslen-sektorn och energisektorn. Det första målet var att undersöka om variationerna inom sektorerna kunde förklaras och kategoriseras. Detta utfördes genom att varje sektor först definierades och utsläppskällorna identifierades. Emissionerna kunde förklaras och kategoriseras för fordonssektorn och fossila-bränslen-sektorn. Däremot kunde utsläppen från energisektorn bara delvis förklaras och kategoriseras. Det andra målet var att undersöka vilka parametrar och sub-kategorier som var viktiga för att estimera sektorernas emissioner. För att göra detta undersöktes två olika metoder; korrelationsanalys och medelvärdesmetoden. Inga korrelationer kunde hittas mellan någon av sektorerna och de undersökta parametrarna. De estimerade emissionerna när medelvärdesmetoden användes, verifierades mot företagens självrapporterade utsläpp. För fordonssektorn och fossila-bränslen-sektorn följde de estimerade och rapporterade utsläppen samma trend. Däremot påträffades ingen trend för energibolagen. Att estimera växthusgasutsläpp med hjälp av en medelvärdesmetod kräver en viss typ av företagsstruktur. Metoden kan användas för företag med en specifik produkt, men är inte lämplig för företag med en mer komplex struktur. Modellernas största begränsning var informationsbristen från företagen. Därför behövs mer transparens från företagen för att kunna utveckla modellerna.