A Way Out of the Banking Crisis: The Case for a New Style Prime Money Market Fund

The best way out of the bank stability problem revealed by the run on Silicon Valley Bank – but spreading to other banks – may be a new style of prime money market fund (“MMF”) that, unlike existing prime MMFs, has some loss-absorbing capacity and so could offer fixed net asset value (“NAV”).

Recall in the pre-MMF reform days how prime MMFs worked.  They became a haven for cash management by institutional investors because they provided more safety than a bank balance sheet, minimizing credit and interest-rate risk with assets that were diversified, credit-screened, and of short duration.  This minimized risk was basis for fixed NAV.   In turn,  MMF assets were in significant measure short-term funding obligations issued by banks and other financial firms. Thus, the “deposits” that were funneled out of the banking system into prime MMFs were recycled back into the banking system through CDs held by MMFs.  This set-up broke down because MMFs had no loss-absorbing capacity, and under stress (Lehman, etc.), depositors began a run on MMFs.[1]

The 2016 reforms, which eliminated fixed NAV for institutional prime funds, essentially crushed them.  Parties who wanted fixed NAV had to put their money in government MMFs or in the official banking system.  We’ve now seen the problems:  The combination of social media and electronic-funds transfer has dramatically increased the run risk of large depositors from the banking system.  SVB may have created conditions for unusually high depositor correlation but the take up of the Fed’s emergency funding facility suggests that regionals and many other banks experienced significant withdrawals.

Where did this money go?  It looks like most went to government MMFs, the other place for fixed NAV.  Government MMF assets now exceed $5 trillion.  What are the assets now matched against these inflows? We’ve run out of Treasuries and Federal Home Loan Bank issuances, so the money – $2.3 trillion – has flooded into the Fed’s overnight repo repurchase program.
Fed repo assets on MMFs balance sheets thus reflect funds drained out of the banking system; this flow reduces the funds available for private credit intermediation. In effect, large depositors are being given accounts at the Federal Reserve using the government MMFs as an intermediary.  The consequence is a reduction in the availability of credit for small and medium-sized enterprises (“SMEs”), which typically do not have access to the public debt market and thus rely on bank credit.

We have seen a similar story with deposits that have flowed into the money-center banks, the global systemically important banks (“G-SIBs”), presumed to be safe because they are “too big to fail.”  The G-SIBs have no use for this money – meaning no interest in using it to provide more credit.  This is because taking on additional risky assets would require more equity to meet general capital and leverage requirements of the bank regulatory regime.  This would mean raising new equity at a time when doing so is costly for banks. Moreover, unlike the “sending” banks, the G-SIBs have no local knowledge to guide their loan origination.  The result is that the “destination” banks will simply leave the unwanted deposits in their reserve accounts at the Fed,  resulting in more reverse repo.[2]

There are thus two routes for deposit flows out of regional and community banks into undeployed Fed reserve accounts.  One is via government MMFs.  The second is via the redirection of deposit inflows by G-SIBs.  Either way, this is a bad outcome because bank-dependent SMEs will see credit contraction.[3]

But “New-Style prime MMFs” offer a better way.  Their critical features – some loss absorbency and an anti-run structure – mean that they can provide fixed NAV accounts.  As noted above, the principal asset of a pre-reform prime MMF was a portfolio of bank CDs.  These assets are not immediately redeemable, and a diversified portfolio of such CDs means that the MMF is not exposed to massive losses should any single bank issuer default. In consequence, the prime MMF is a much more stable depositor in the banking system than individual large depositors. The nature of the MMF means that such depositors/MMF shareholders have daily liquidity, but their redemptions won’t bring down a systemically important bank or threaten the banking system more generally.  In short, New-Style prime funds would add to rather than subtract from financial stability.

The key is loss absorbency, augmented by an anti-run structure.  Technically, this is not hard to obtain. The Fed had a minimum balance at risk proposal.[4]  I had a two-classes of MMF stock proposal.[5] These approaches would force MMF depositors to internalize losses and also create an explicit “last mover” anti-run dynamic. A complement or alternative, of course, would be sponsor-provided capital.[6]

In the last reform round, such proposals were ruled out because the prevailing short-term rates simply did not provide a margin for profitable operation of such an MMF.  That is how we ended up in the current place, where the only deposit-substitute MMFs are governments.  But that picture has dramatically changed.  I think we will never (or at least not for a long time) go back to that historically unprecedented period of super-low short-term rates.  It was a product of desperation following the Global Financial Crisis and then the Covid shock.  There has been a regime change in monetary policy.  Now we see a positive interest margin that MMF sponsors could use (through fees) to provide a sufficient  equity return.  With yields of 4 percent, MMF depositors ought to be willing to give up a few basis points that could fund a loss-absorbency regime that would enable fixed NAV for New-Style prime funds.

The goal is to make it easier to shift the balance sheets of regional and other banks away from heavy reliance on runnable deposits held by undiversified individual actors to CDs held by MMFs, a more stable source of funding, and to avoid the precipitous shrinking of the banking-system deposit base in response to a shock.  In short, New-Style prime MMFs are the depositors that many banks now need, and we can help address the present banking crisis with their creation.


[1] See generally Jeffrey N. Gordon & Christopher M. Gandia, Money Market Funds Run Risk: Will Floating Net Asset Value Fix the Problem?  2014 Colum. Bus. L. Rev. 313.

[2] This is not just a conjectural outcome. Patrick Bolton et al discuss such effects in the 2020 Covid shock, including deposit flows into Fed reserve accounts, and provide a banking model that incorporates this effect. Patrick Bolton, Ye Li, Neng Wang, & Jinqiang Yang, Dynamic Banking and the Value of Deposits (April 2023), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3624119 .

[3] This is an answer to those who would regard credit contraction from these deposit outflows as a useful augmentation of the Fed’s present monetary policy.  We ought to avoid a monetary policy that disproportionately affects SME borrowers.

[4] Patrick E. McCabe, Marco Cipriani, Michael Holscher & Antoine Martin, The Minimum Balance at Risk: A Proposal to Mitigate the Systemic Risks Posed by Money Market Funds, FRBNY Staff Report No. 564 (July 2012), available at https://www.newyorkfed.org/research/staff_reports/sr564.html. This proposal calls for a 3 percent holdback of redeemed funds for a 30 day period and, in the event of portfolio losses, subordination of the “holdback” amount to the claims of non-redeeming fund shareholders.

[5] Jeffrey N. Gordon, Comment on Money Market Funds for SEC (Aug. 2012), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2133588; Comment Letter to Financial Stability Oversight Council on Money Market Reform (Feb. 2013),  available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2227169. The proposal calls for MMFs to issue two classes of equity, Class A, designed to retain a fixed NAV, and Class B, whose value will vary to cover defaults or declines in the market value of portfolio securities. Institutional prime MMF users would be required buy a Class A/Class B bundle, at a ratio calibrated to absorb all losses, and to maintain at least that ratio in redemptions and purchases of Class A shares.

[6] See, e.g., Samuel Hanson, David S. Scharfstein & Adi Sunderam, An Evaluation of Money Market Reform Proposals 63 IMF Econ. Rev. 984 (Nov. 2015).

This post comes to us from Jeffrey N. Gordon, the Richard Paul Richman Professor of Law at Columbia Law School, co-director of the Millstein Center for Global Markets and Corporate Ownership, and co-director of the Richman Center for Business, Law and Public Policy. He wishes to express appreciation for helpful comments from Patrick Bolton, Howell Jackson, Lev Menand, and Morgan Ricks.