The Federal Reserve Bank of New York published a "Staff Report," entitled, "Sophisticated and Unsophisticated Runs," which analyzes money fund outflows during last March's sudden coronavirus lockdown. Authors Marco Cipriani and Gabriele La Spada tell us, "In March 2020, at the beginning of the Covid-19 pandemic, investors redeemed en mass from prime money market funds (MMFs). At the height of the run, cumulative redemptions from US-dollar prime MMFs, both onshore and offshore, were 22% of industry's total net assets (TNA) as of the end of 2019. Outflows were significant for both institutional (32%) and retail investors (11%). This is the second time the industry has suffered a run over the last 20 years: in 2008, after Lehman bankruptcy, redemption pressures of similar magnitude buffeted the industry. In both cases, the Federal Reserve intervened to stem the outflows with the establishment of emergency lending facilities."
They explain, "In this paper, we use the 2020 run to characterize the behavior of sophisticated and unsophisticated MMF investors. The 2014 SEC reform separated prime MMFs by investor type, separating funds catering only to retail (unsophisticated) investors from those catering to institutional (sophisticated) ones. We show that the behavior of these two classes of investors were dramatically different during the crisis, which we view as a result of their different level of sophistication."
The report continues, "Institutional investors left those prime funds whose liquid assets were closer to the regulatory thresholds for gates and fees on redemptions. Therefore, their decision to run was based on an assessment of the likelihood that their access to liquidity might be impaired. This is true both for institutional onshore funds and for institutional offshore funds."
It says, "We identify the impact of gates and fees in two ways. First, since gates and fees can be imposed only if a fund's weekly liquid assets (WLA) drop below a threshold, we instrument funds' WLA with their average values in 2019Q4; our instrument allows us to control for reverse causality issues due to a fund's liquidity management during the run. We find that a 10 percentage-point decrease in 2019Q4 WLA leads to an increase in daily outflows of 1.1 percentage points in onshore institutional prime MMFs."
The report states, "Second, we exploit the fact that some offshore institutional prime MMFs are exempt from the imposition of gates or fees. During the run, such funds experienced daily outflows that were 1.3 percentage points smaller than those suffered by offshore institutional prime funds that can impose gates and fees."
It comments, "No such relationship between the likelihood of gates or fees and investor flows exists in retail funds. Instead, retail investors left those prime MMFs belonging to families that also catered to institutional prime investors and suffered heavier institutional redemptions. For example, during the run, retail prime MMFs in families also offering institutional funds experienced daily outflows that were 1.7 percentage points larger. In other words, (unsophisticated) retail investors seem to base their decision by following the behavior of (sophisticated) institutional investors within the same fund family."
The report also tells us, "The imposition of gates or fees is one of the main regulatory changes of the prime MMF industry introduced by the SEC in its 2014 reform. At that time, some practitioners, policy makers, and academics feared that the new regulation would make runs more likely by giving an incentive for investors to leave a fund preemptively ahead of a gate or fee being imposed. For instance, Cipriani et al. (2014) show in a simple Diamond and Dybvig (1983) model that such an incentive is present and that preemptive runs would occur. In this paper, we show that such a theoretical mechanism is indeed consistent with investors' behavior; however, it does require some level of sophistication for investors to run preemptively, and, as a result, we do not observe preemptive runs in retail prime MMFs. The unintended consequence of redemption gates and fees, however, can still propagate to retail funds via within-family spillovers from institutional to retail investors."
It adds, "Finally, investors who left the prime MMF sector largely moved their money to government funds, consistently with past episodes of industry dislocation. This is true both of retail and of institutional investors. Prime MMFs belonging to families more specialized in government MMFs experienced larger outflows. For example, a 10 percentage-point increase in the share of government MMFs in a family's MMF business in 2019Q4 leads to daily outflows during the run that are larger by 0.24 percentage points in onshore institutional prime funds and 0.20 percentage points in onshore retail prime funds. We obtain similar results when we compare outflows from prime funds between families that also offer government funds and those that do not. This result suggests that, for both institutional and retail investors, there are lower switching costs in fund families that are relatively more specialized in government MMFs, which could lead to higher flow volatility during periods of stress."
The report also comments, "Several recent papers have studied episodes of severe dislocation in the MMF industry. The 2008 run on prime MMFs was described in Baba et al. (2009), Brady et al. (2012), and Kacperczyk and Schnabl (2013), among others. Cipriani and La Spada (2017) analyze investors' behavior in response to the 2014 SEC reform of the MMF industry; they show that the imposition of a system of gates and fees on onshore prime MMFs (along with the floating NAV requirement for institutional ones) led to outflows of more than a trillion dollars from prime MMFs. Schmidt et al. (2016) study the relationship between fund flows during the 2008 run and investor sophistication, focusing on the externalities caused by the presence of investors with different levels of sophistication (e.g., institutional and retail ones) within the same fund. We follow their interpretation of the differences in institutional versus retail behavior as reflecting different levels of investor sophistication."
Lastly, the paper's "Introduction" says, "The 2020 Covid-19 run is also studied by Li et al. (2020). They provide similar evidence to ours on the impact of gates and fees on investor flows in US institutional prime MMFs, while differing in the identification strategy; we discuss the difference in our empirical analysis. Casavecchia et al. (2020) document the March 2020 run focusing on institutional prime MMFs and their floating NAV feature.... The remainder of the paper is as follows: Section 2 describes onshore and offshore MMFs and our dataset; Section 4 describes the run by (sophisticated) institutional investors; Section 5 describes the run by (unsophisticated) retail investors; and Section 6 shows the effect of family specialization in government funds on run behavior."
