Earlier this week, staffers from the Federal Reserve Bank of Boston and the Federal Reserve Bank of New York published a paper on "The Money Market Mutual Fund Liquidity Facility," which reviews the Fed's emergency support measures from March 2020. Authors Ken Anadu, Marco Cipriani, Ryan Craver, and Gabriele La Spada explain, "In this article, we discuss the run on prime money market funds (MMFs) that occurred in March 2020, at the onset of the COVID-19 pandemic, and describe the Money Market Mutual Fund Liquidity Facility (MMLF), which the Federal Reserve established in response to it. We show that the MMLF, like a similarly structured Federal Reserve facility established during the 2008 financial crisis, was an important tool in stemming investor outflows from MMFs and restoring calm in short-term funding markets. The usage of the facility was higher by funds that suffered larger outflows. After the facility's introduction, outflows from prime MMFs decreased more for those funds that had a larger share of illiquid securities. Importantly, following the introduction of the MMLF, interest rates on MMLF-ineligible securities decreased at a slower rate than those on MMLF-eligible securities, even after controlling for credit risk."
They write, "In March 2020, at the onset of the Covid-19 pandemic, investors redeemed their shares en masse from dollar-denominated prime money market funds (MMFs). The large redemptions occurred both in U.S. MMFs registered with the Securities and Exchange Commission (SEC) and governed by its Rule 2a-7 ('domestic' funds) and in dollar-denominated MMFs domiciled in Europe and governed by European rules ('offshore' funds). In percentage terms relative to the size of the industry, the run was remarkably similar to that experienced by MMFs in September 2008, during the Global Financial Crisis, notwithstanding the starkly different natures of the shocks that precipitated the runs. As was the case in 2008, the 2020 run amplified strains in the short-term funding markets, a key source of liquidity for businesses, as rates on several money market securities increased steeply." (Note: The study uses a smaller subset of money fund assets than Crane Data's or the SEC's collections, which make the March 2020 outflows appear larger than they appear in our data. We show the March 2020 asset outflows to be smaller on a percentage basis than those in September 2008.)
The paper tells us, "In mid-March, the Federal Reserve, with approval of the Secretary of the Treasury, established the Money Market Mutual Fund Liquidity Facility (MMLF) to assist MMFs in meeting heightened investor redemptions, stabilize the U.S. short-term funding markets, and support credit provision to the real economy. Under the facility, which was similar in its structure and purpose to the Asset Backed Commercial Paper Money Market Mutual Fund Liquidity Facility (AMLF) established in 2008, the Board of Governors authorized the Federal Reserve Bank of Boston to make non-recourse loans to eligible banks to facilitate the purchase of eligible assets from domestic prime, single state, or other tax-exempt MMFs."
It continues, "In this paper, we discuss the March 2020 run on MMFs, describe the MMLF's design and operations, and assess its effectiveness in stemming fund outflows and calming money market rates. First, we discuss the different reasons that led investors to run and document the dislocations in money market rates that accompanied the run. As shown in Cipriani and La Spada (2020) and Li et al. (2020), institutional investors ran more from funds for which the imposition of redemption gates and liquidity fees -- introduced by the 2014 SEC reform -- was more likely due to lower levels of 'weekly liquid assets' (WLA) in their portfolios. The outflows of retail investors, in contrast, were unrelated to fund-level liquidity and reflected other factors, including contagion from the behavior of institutional investors within the same fund family."
Anadu, et. al., state, "Second, we describe the MMLF's structure and compare it with that of the AMLF, highlighting similarities and differences. Both facilities used banks as a conduit to provide liquidity to domestic prime (and, for the MMLF, also tax-exempt) MMFs. A material difference, however, is that the AMLF only facilitated banks' purchases of ABCP from MMFs, whereas the MMLF made loans against a broader set of assets. Third, we describe the usage of the facility. We show that the MMLF was used more by funds that suffered larger outflows, and that funds sold securities with longer maturities, consistent with their incentive to boost their liquidity positions and especially their WLA."
They comment, "Finally, we identify the effect of the MMLF on investor flows by showing that, after the facility's introduction, outflows from prime MMFs decreased more for those funds that were eligible to participate in the MMLF program (i.e., domestic ones), that had a larger share of illiquid securities, and whose investors were more concerned about the funds' liquidity (i.e., institutional ones). Moreover, we show that, after the introduction of the MMLF, the rates of MMLF-ineligible securities declined more slowly than those of MMLF-eligible securities, even after controlling for credit risk. Overall, our analysis shows that, as the AMLF had been in 2008, the MMLF was an important tool in stabilizing prime-MMF flows and short-term funding markets at large."
Describing the "The Money Market Mutual Fund Liquidity Facility (MMLF)" in more detail, the paper tells us, "On March 18, 2020, the Federal Reserve, with approval of the Secretary of the Treasury and $10 bn of credit protection from the Exchange Stabilization Fund, announced the introduction of the MMLF to provide liquidity to MMFs. To do this, the Federal Reserve had to address two challenges. First, the Federal Reserve needed to protect itself from credit risk, for example by offering loans only against high-quality collateral. Second, lending to MMFs is problematic, as it would have increased their leverage, amplifying any losses for the shareholders and increasing their incentive to run. The Federal Reserve faced the same challenges in 2008, when it set up the AMLF in response to the MMF run triggered by Lehman Brothers' default. Although the type of shock was different, it was natural to design the 2020 facility based on its 2008 predecessor."
It explains, "Through the MMLF, which was established under the authority of Section 13(3) of the Federal Reserve Act, the Federal Reserve Bank of Boston made non‐recourse loans to eligible borrowers, taking as collateral eligible assets purchased by the borrowers from eligible MMFs. The eligible borrowers were U.S. depository institutions, U.S. bank holding companies (parent companies incorporated in the U.S. or their U.S. broker-dealer subsidiaries), and U.S. branches and agencies of foreign banks. Eligible collateral was limited to U.S. Treasuries and fully-guaranteed agencies, Government Sponsored Enterprise (GSE) securities, highly-rated CP (including ABCP), negotiable CDs, and short-term municipal debt (including VRDNs that met certain criteria). Eligible funds were limited to domestic prime, single state, or other tax-exempt MMFs."
The paper concludes, "In March 2020, as the Covid-19 pandemic hit the U.S. and Europe, prime MMFs suffered very large investor outflows, of similar percentage magnitude to those experienced in 2008 during the Great Financial Crisis. The Federal Reserve established the MMLF in order to assist 'money market funds in meeting demands for redemptions by households and other investors, enhancing overall market functioning and credit provision to the broader economy' (Federal Reserve Press Release, 3/18/2020)."
It reiterates, "Through the MMLF, the Federal Reserve Bank of Boston made non‐recourse loans to eligible borrowers, taking as collateral eligible assets purchased by the borrowers from eligible MMFs. The facility, which was similar to the AMLF established in 2008, absorbed $58 billion of prime-MMF assets. With the facility's assistance, MMFs sold their most illiquid securities, thereby boosting their liquidity positions while meeting redemptions. In the aftermath of the MMLF's inception, outflows from prime funds abated, and the strains in the broader short-term funding markets subsided."
Finally, the paper adds, "Because the MMLF was established in the midst of a financial crisis and a rapidly changing economic outlook, it is difficult to directly estimate its impact. Nonetheless, we provide evidence that the facility directly helped stem the outflows from prime MMFs and contributed to the easing in money market rates. Because of its positive effect on secondary markets, the facility also had a beneficial impact on offshore prime MMFs, which it did not directly target. By helping prime MMFs meet redemptions and reducing their outflows, the facility improved overall market functioning and supported credit provision to the real economy."
