Investor-Advisor Relationships and Mutual Fund Flows

In my paper, Whom Do You Trust? Investor-Advisor Relationships and Mutual Fund Flows, forthcoming in the Review of Financial Studies, I investigate the role of trust in the asset management industry. While there is plenty of anecdotal and survey evidence which underlines the general importance of trust in finance, academic research has been scarce due to the difficulty of quantifying and measuring trust. In this paper, I use an exogenous shock to the relationships between investors and mutual fund advisory companies (e.g. Fidelity, Wells Fargo, Vanguard, etc.) to try to tease out the effect of trust.

Specifically, I look at funds that change advisors due to mergers & acquisitions at the parent company level. For example, NationsBank acquired (merged with) BankAmerica in 1998. This parent-level transaction caused all investors in the Pacific Horizon family of mutual funds to change advisory firms from BankAmerica to NationsBank (renamed Bank of America). I then examine monthly fund flows around the announcement of these transactions to measure how investors react to this change in advisor. One potential concern is that ownership changes are not exogenous and are correlated with the mutual fund’s past performance (which is also correlated with future fund flows), so I further restrict the event space to those events where the parent companies are predominantly non-investment companies (mostly banks and insurance companies). Mergers and acquisitions of banks and insurance companies are less likely to be driven by performance at a fund advised by a subsidiary advisory firm.

My hypothesis is that mutual fund net flows decline after the announcement of an ownership change at the advisory firm’s parent company level. The intuition for this hypothesis is that as preexisting investor relationships with mutual fund advisory firms break, and are replaced with new relationships, the level of trust is expected to diminish. The reaction by investors, as measured by fund flows, tells us if this disruption in trust is important enough to lead to a termination of the relationship, even in the presence of adjustment costs.

The paper’s main result is that net flows decline by a statistically and economically significant 7% in the year after the announcement date of the ownership change. The decline starts right after the announcement date and then accelerates after the M&A transaction is consummated and the advisory change occurs. To highlight the economic significance of this effect, I juxtapose it with the effect on fund families that were implicated in the 2003-2004 market timing scandals. The outflows during ownership changes in my study are between 33% and 50% of the outflows during those events, and the market-timing scandals were significantly greater shocks to the trust between investors and advisory firms.

I perform a number of additional tests to distinguish the “trust hypothesis” from other explanations for the main result of the paper. I examine whether prior legal controversies at the acquiring parent company have any effect on flows at funds managed by the target firm’s mutual fund subsidiary. The idea for this test is that if fund outflows are due to disruptions of trust, the disruptions should be greater if the new advisory firm has a record that would leave investors doubting whether they can be trusted. Consistent with this hypothesis, I find that prior legal controversies at the acquiring parent company more than doubles the outflows from funds managed by the target.

Furthermore, I decompose total flows into inflows (new purchases of fund shares) and outflows (sales by existing investors), and find that the main results are entirely driven by higher outflows, not lower inflows. This is consistent with existing investors terminating or limiting their investment in the fund around the change in advisor. Also, in order to understand whether the change in flows is due to a change in manager, I control for whether a manager change occurred in the aftermath of the announcement date and find similar results. Many of the M&A events that make up the sample occurred during the peak of the dot-com bubble, an extremely volatile period for returns and fund flows, so I drop this period from my analysis and find similar results. Other checks suggest that the results are robust to different specifications and definitions.

In summary, I find that investors respond to exogenous disruptions to their relationship with the advisory firms that are managing the mutual funds. The average decline in flows is around 7% of fund assets in the year following the announcement date, starting after announcement and accelerating after the closing date of the ownership change. The paper highlights the importance of trust in the asset management industry.

The full paper is available for download here.