|Over the last decade, sustainability has truly moved from niche to mainstream when it comes to attracting attention among investors and policymakers. The number of funds investing under environmental, social and governance considerations has surged, and with that, the call for transparency is stronger than ever. Corporate statements and reports are therefore a valuable resource as they represent a wealth of information regarding companies’ operations. There are two main purposes of this thesis: The first is to create a tool that captures ESG-related disclosures in annual 10-K reports of underlying companies in mutual funds. Secondly, to see if disclosure relates to sustainability performance, represented by the score in the Morningstar sustainability rating (MSR). The sample consists of 118 US mutual funds, observed over a three-year timeframe, from 2016 to 2018. The first research question examines if the level of disclosure in underlying companies can predict sustainability performance of funds. Our results indicate that there is a relationship between the level of disclosure in underlying firms and sustainability performance for the following investment categories: US large cap blend, US large cap growth, US large cap value, US mid cap, and finance. For sector-specific categories such as healthcare, consumer goods and services and technology, no significant relationship is found. The explanatory power of textual disclosure score on sustainability performance of funds is limited but the model shows potential for more precise predictions for certain investment categories. Estimates appear to be less accurate for more volatile funds for which the difference between MSR and ESG disclosure score is larger. We also find that “green labelled” funds in our sample have better sustainability performance than conventional funds, while we find no difference in the disclosure score. Lastly, despite the increasing amount of sustainable investing, our data does not suggest an increasing trend of ESG-disclosures in 10-K filings over the sample period.