Résumé : The financial crisis of 2007-09 has raised questions concerning orthodox ideas of how financial markets operate and how financial institution firms (and their decision makers) behave in such market settings. The crisis has revealed the need for new thinking in these areas. This paper outlines a grounded theory of the fund management firm and investment decision making at the level of individuals and teams. The theory comprises the role of external and internal firm contexts on individual fund managers and teams and the funds they manage. It includes the properties of FM firm organisational contexts comprising order, creativity, knowledge, coherence and matching; and the relative peer group strengths of these properties. It includes the properties of individual and team contexts and their strengths. These contexts and their properties all purposefully interacted as collective and integrated FM organisational means (or dynamic system) to help individuals and teams to reduce the complexity of new information flows, to make sense of information, to avoid their own negative behaviour, to exploit the behaviour of others, to take investment decisions, to create diversified portfolios, and to produce FM financial performance (at the level of funds managed, and the firm overall). The FM value creation process, the continuous interaction (and periodic disclosure) with investee companies and with stock and information markets, and ongoing investment decision making at individual and team level, also led to cumulative FM learning and strategic choices about the perceived drivers of fund and market outcomes. The interaction and learning created new knowledge intensive contexts or priors for FM investment decision behaviour which then became drivers of subsequent investment decisions. The ongoing investment interactions, the longer term knowledge creating interactions, and the FM responses revealed the dynamic elements to FM investment behaviour. The FM contextual elements and dynamic interactions were common empirical patterns across case FMs. Variation in context and properties were used to tentatively explain, in part, FMs type or style. Variation in strength (weaknesses) of contextual properties (within a fund type or peer group) were used to tentatively explain, in part, differences in performance. A major function of the grounded theory may be to provide a new conceptual tool to fully investigate the reasons for such variety in FM styles. In addition it may provide a new tool to fully investigate and explain the reasons for pervasive and systematic FM underperformance rather than to explain the more unusual cases of success. The research was conducted through case interview field work in 20 large international FMs during 2004-2011 and use was made of archival sources. A grounded theory approach was employed in processing the data. The results were discussed relative to relevant literature and to previous grounded theory of FM. Areas for further research were identified.