Final Report Summary - DYNAMIC DELEGATION (Implications of the Dynamic Nature of Portfolio Delegation)
The overarching objective of this project is to improve understanding of different aspects of delegation. We strive to better understand incentives that emerge in a dynamic delegation relation and associated implication. We are interested in advancing knowledge on how these incentives impact trading strategies and how these strategies influence asset price dynamics, as well as improving understanding of why delegation contracts are structured as they are. We combine theoretical modeling, which we relate to patterns observed in the data, with new empirical evidence that highlights implications of delegation.
We theoretically analyze the effects of increased share of delegated capital for trading strategies and equilibrium prices. We show that due to the convex flow-performance relation when the share of delegated capital is sufficiently high funds with identical incentives employ heterogeneous strategies, where a subset of funds lever up and trade against the rest of the fund population. Our model links increases in delegated portfolio management to increased amounts of borrowing and lending in the economy. While we show that increased delegation is associated with cross sectional dispersion in fund returns, we further predict that on average funds will outperform the market in recessions and underperform in expansions.
We establish a clear causal relation in the data between media attention and mutual fund flows, whereby appearance in a Wall Street Journal “Category Kings" ranking list, on a single day, leads to more than a 30 percent increase in quarterly capital flows. We show that funds near the list cutoff a month prior to anticipated publication of the lists, and only these funds, strategically increase pre-publication tracking error volatility relative to peer funds, in an attempt to make it to the list and benefit from the extra flows. Fund families take advantage of appearance of their funds in the list by increasing subsequent advertising.
We analyze a multi-agent contracting setting where managers have “keeping up with the Joneses” (KUJ) preferences, and care in part about how their pay compares to that of others. We show that in stark contrast to the standard contracting setting without externalities, contractual pay can be increasing in peers’ output, providing a simple rational for the lack of negative relative performance sensitivity in observed managerial contracts. We contrast optimal contract characteristics, managerial effort and efficiency, when there is a single principal employing multiple managers to when there are multiple principals. When a single principal can commit to a public contract, the optimal contract hedges risk of managers’ relative wage without sacrificing efficiency. While output is unchanged, hedging makes contracts appear inefficient, as performance seems inadequately benchmarked. With multiple principals, or the principal is unable to commit, efficiency is undermined. Managers are more productive, but average wages increase even more, reducing firm profits. When principals manage teams, it is efficient to have compensation contracts disclosed within teams when KUJ effects are weak, but kept secret when KUJ effects are sufficiently strong. When contracts are disclosed externally to managers on other teams, effort is reduced and expected compensation increases. Consequently, when KUJ preferences are mild external disclosure increases firm profits in equilibrium. However, when KUJ preferences are strong incentives may collapse leading to lower profits.
We analyze how forcing fund managers to commit part of their personal wealth to the fund they manage affects fund allocations, depending on the type of restrictions imposed on trading in their personal account.
Using a unique dataset we constructed for this project, containing information on income of fund managers, we provide novel evidence on the dynamic interaction between fund families and the fund managers who manage the assets. Prior literature has mostly been silent on this, bundling managers with the funds they run. We uncover three key findings, each challenging common perceptions. First, sensitivity of pay to manager-level assets under management, elasticity of 0.15 is considerably lower than the fixed fraction of assets under management that funds typically charge investors. Second, there is surprisingly weak sensitivity of pay to performance, even after accounting for potential non-linearity in pay to performance sensitivity. Third, firm-level characteristics, typically ignored in the literature, add substantial explanatory power for manager compensation: sensitivity of manager compensation to firm revenue is comparable to that of manager revenue. Firm profit matters considerably: firms with higher profits pay significantly more. At the same time, profitability lowers the sensitivity of compensation to manager revenues and increases that to performance.
Our evidence underscores that managers’ compensation can not, and should not, be evaluated in isolation. Instead, to understand agency frictions associated with delegation a more holistic perspective is in order, acknowledging compensation externalities within fund families. Furthermore, our evidence highlights the need to study skill complementarities between fund managers and fund complexes.
We theoretically analyze the effects of increased share of delegated capital for trading strategies and equilibrium prices. We show that due to the convex flow-performance relation when the share of delegated capital is sufficiently high funds with identical incentives employ heterogeneous strategies, where a subset of funds lever up and trade against the rest of the fund population. Our model links increases in delegated portfolio management to increased amounts of borrowing and lending in the economy. While we show that increased delegation is associated with cross sectional dispersion in fund returns, we further predict that on average funds will outperform the market in recessions and underperform in expansions.
We establish a clear causal relation in the data between media attention and mutual fund flows, whereby appearance in a Wall Street Journal “Category Kings" ranking list, on a single day, leads to more than a 30 percent increase in quarterly capital flows. We show that funds near the list cutoff a month prior to anticipated publication of the lists, and only these funds, strategically increase pre-publication tracking error volatility relative to peer funds, in an attempt to make it to the list and benefit from the extra flows. Fund families take advantage of appearance of their funds in the list by increasing subsequent advertising.
We analyze a multi-agent contracting setting where managers have “keeping up with the Joneses” (KUJ) preferences, and care in part about how their pay compares to that of others. We show that in stark contrast to the standard contracting setting without externalities, contractual pay can be increasing in peers’ output, providing a simple rational for the lack of negative relative performance sensitivity in observed managerial contracts. We contrast optimal contract characteristics, managerial effort and efficiency, when there is a single principal employing multiple managers to when there are multiple principals. When a single principal can commit to a public contract, the optimal contract hedges risk of managers’ relative wage without sacrificing efficiency. While output is unchanged, hedging makes contracts appear inefficient, as performance seems inadequately benchmarked. With multiple principals, or the principal is unable to commit, efficiency is undermined. Managers are more productive, but average wages increase even more, reducing firm profits. When principals manage teams, it is efficient to have compensation contracts disclosed within teams when KUJ effects are weak, but kept secret when KUJ effects are sufficiently strong. When contracts are disclosed externally to managers on other teams, effort is reduced and expected compensation increases. Consequently, when KUJ preferences are mild external disclosure increases firm profits in equilibrium. However, when KUJ preferences are strong incentives may collapse leading to lower profits.
We analyze how forcing fund managers to commit part of their personal wealth to the fund they manage affects fund allocations, depending on the type of restrictions imposed on trading in their personal account.
Using a unique dataset we constructed for this project, containing information on income of fund managers, we provide novel evidence on the dynamic interaction between fund families and the fund managers who manage the assets. Prior literature has mostly been silent on this, bundling managers with the funds they run. We uncover three key findings, each challenging common perceptions. First, sensitivity of pay to manager-level assets under management, elasticity of 0.15 is considerably lower than the fixed fraction of assets under management that funds typically charge investors. Second, there is surprisingly weak sensitivity of pay to performance, even after accounting for potential non-linearity in pay to performance sensitivity. Third, firm-level characteristics, typically ignored in the literature, add substantial explanatory power for manager compensation: sensitivity of manager compensation to firm revenue is comparable to that of manager revenue. Firm profit matters considerably: firms with higher profits pay significantly more. At the same time, profitability lowers the sensitivity of compensation to manager revenues and increases that to performance.
Our evidence underscores that managers’ compensation can not, and should not, be evaluated in isolation. Instead, to understand agency frictions associated with delegation a more holistic perspective is in order, acknowledging compensation externalities within fund families. Furthermore, our evidence highlights the need to study skill complementarities between fund managers and fund complexes.