Tuesday, February 12, 2013

Are Money Market Funds the Banking System's Achilles Heel?

The systematic effect of money market funds on the wider financial system is a topic of hot debate, with finance ministers, regulators, and standard setting bodies all over the globe weighing in. The New York Federal Reserve Bank is the latest to do so, asking "does money market fund intermediation make the banking system inherently unstable?" The New York Fed's latest paper, "Money Market Funds Intermediation, Bank Instability, and Contagion," addresses the question by comparing two market structures – direct finance, where investors deposit directly into the banks, and money market fund intermediation, where the relationship between investors and banks is intermediated through money market funds. The paper concludes that, because money market funds are themselves subject to runs, the intermediation they provide to investors makes the financial system more fragile, and acts as a source of contagion when a run occurs.

In this paper, we show that intermediation through MMFs allows investors to limit their exposure to a given bank (i.e., reap gains from diversification). However, since MMFs are themselves subject to runs from their own investors, a banking system intermediated through MMFs is more unstable than one in which investors interact directly with banks. A mechanism through which instability can arise in an MMF-intermediated financial system is the release of private information on bank assets, which is aggregated by MMFs and could lead them to withdraw en masse from a bank. In addition, we show that MMF intermediation can also be a channel of contagion among banking institutions.

By drawing a contrast between an environment of direct finance and one of money market fund intermediation, the New York Fed demonstrates that money market funds represent a fundamental fragility, or an "Achilles heel," in the financial system.

Under direct finance, unexpected withdrawals cause bank bankruptcy only if the amount withdrawn is large enough to force the bank into liquidation. In contrast, with MMF intermediation, when a fraction of investors unexpectedly redeem from the MMF, their actions represent a (noisy) signal on the state of the world for the MMF. If this signal is strong enough, the MMF will run the bank, withdrawing all its funds and causing bankruptcy even if the fraction of the unexpected redemptions was small enough that bankruptcy would not have occurred under direct finance. The instability of MMF intermediation stems from the fact that the negative information content of an unexpected redemption from an intermediary such as an MMF amplifies the effect of the redemptions themselves. Because of this, an economy intermediated by MMFs is generically more unstable than a direct finance structure.

The amplification mechanism is possible given that MMFs are subject to run-like redemptions because they offer investors demandable liabilities in order to satisfy their liquidity needs. When an [sic] MMF experiences large unexpected redemptions, it runs the bank to protect all its investors, and not just those initiating the redemptions. Because of the bank’s fixed promise, the MMF, receiving negative information on the bank’s assets, obtains a higher payoff for its investors if it runs than if it does not.

This latest paper adds more fuel to the debate over the need for additional controls or structural reforms over the money market industry, and will no doubt draw a chorus of rebuttals from money market industry players.