Placing the production of investment returns: an economic geography of asset management in public pension plans

Traditionally, asset owners such as pension funds and sovereign wealth funds have relied almost exclusively on financial intermediaries to invest their assets. However, since the 2008 financial crisis the idea of bypassing the industry by insourcing investment functions has gained momentum amongst practitioners and academics alike. Dixon and Monk first reviewed this grassroots movement in a publication titled "frontier finance" (2014). Their global survey revealed mixed empirical results suggesting that while the model had generated a marked interest, it had failed to be widely implemented. Since the organisations they investigated were most often found outside established financial agglomerations, their results pointed to the enduring dominance of international financial centres (IFCs) in the provision of investment return production. This doctoral thesis proposes to revaluate their conceptual proposition in the context of state and local pension plans in the United States. By unveiling and critically assessing the provision of investment services amongst public pension funds, this thesis contributes to scholarship on the necessary foundations to safeguard the procurement of decent retirement income in the future. To meet its objective, the thesis proceeds inductively using mixed-methods including original quantitative and qualitative data. These include an econometric analysis of a unique panel dataset on 21 state pension plans and semi-structured interviews with 37 investment professionals employed by a large frontier pension plan. Triangulated with additional close-dialogue and secondary data analysis, these insights are ultimately assembled into a comprehensive theory of insourcing for public pension funds. The findings contribute to economic and financial geography scholarship on institutional investors and global financial market investments. In light of the empirical findings, it is argued that the production of institutional investment returns is underpinned by two concomitant and seemingly contradictory processes: the increasing agglomeration of financial institutions around IFCs and the unprecedented virtualisation of their tasks and functions. The argument is sustained in reference to the different combination of technological and human capital required to manage distinct type of financial market investments. On one hand the thesis shows the declining relevance of co-locating decision-making with financial agglomerations to produce investment returns on public markets covering proximate economies (domestic and developed). As these markets have become increasingly virtualised, liquid and efficient, they require less intensive and specialised investment processes thus lending themselves well to investment at-a-distance. Since public markets remain the backbone of asset owners’ portfolios, it is suggested that insourcing holds an untapped potential. On the other hand, the thesis emphasises the enduring relevance of specialised clusters providing tasks and functions that make intensive use of human capital and non-commodified information. As such, private market investments and more generally investments that call upon highly specialised decision-making remain territorially contained and thus less well suited to investment-at-a-distance. Notwithstanding marginal limits, the thesis demonstrates that Dixon and Monk's (2014) frontier finance, and in a larger sense O’Brien's (1992) thesis on the "end of geography", are increasingly legitimate theoretical concepts to describe asset owners' production of investment returns.