A new working paper from the SEC's Division of Economic and Risk Analysis (DERA) examines, "The Effect of NAV Flotation on the Management of Prime Money Fund Portfolios." Written by the U.S. Securities and Exchange Commission's Su Li and Wei Liu, along with the University of Pennsylvania's David Musto, the paper's "Abstract" explains, "The October, 2016 money-fund reform obliged institutional prime funds to start floating their Net Asset Values (NAVs), whereas retail prime funds could continue with stable NAVs. We use this contrast to assess the effect of NAV flotation on fund management. We find that institutional funds reduced their interest-rate, liquidity and credit risk around the reform, contributing to lower yields and NAVs, and less volatile NAVs, but by all measures, their relative risk taking was approximately back to where it was well before the reform by the end of 2017."
It continues, "Prime funds adjusted to their shrinkage largely through transaction sizes, but also somewhat through the number of securities per issuer they invested in, and to a small extent through the number of issuers they invested in. Transaction-size shrinkage harms transaction quality, though the quality of a fund's transactions relates more strongly to the size of its fund complex than to the size of the transactions themselves."
The Working Paper says, "In October, 2016, the final piece of money-fund reform took effect. This ended a long process of reflection, design, commentary and redesign that dated largely to the money-fund turmoil following the September, 2008 bankruptcy of Lehman Brothers. It also began a new era for money funds, one where large sectors of the industry would start marking their portfolios to market, i.e. would 'float' their Net Asset Values (NAVs).... In this paper we document the effect of this flotation on funds' portfolios by contrasting the portfolio management of institutional and retail funds, and by tracking this contrast from the year before the onset of the reform through the year after."
The authors tell us, "Our main finding is that, once the effects specific to the transition subside after a few months, NAV flotation does not affect risk-taking by money funds. Tracking the complete set of relevant money funds that operate from the beginning of 2015 through the end of 2017, we find that the relative risk taking of funds that become floating-NAV funds in 2016, compared to those that remain stable-NAV funds, is approximately the same at the end as it was at the beginning. This is apparent along all the key dimensions of portfolio risk: interest-rate risk, liquidity risk (measured two different ways) and credit risk (measured two different ways). We also find that both the relative level and relative volatility of the NAVs of the floating- vs. stable-NAV funds is approximately the same at the end as at the beginning. There is an effect on risk-taking and NAVs right around October 2016, but this effect is transitory and no longer evident by December 2017."
They write, "Besides the onset of flotation, funds also had to adapt to asset shrinkage across the reform. We investigate this adaptation, and find that the biggest effect was on transaction size. Funds shrank the principal amount of their purchases, while to a smaller extent they shrank the number of securities they purchased per issuer, and to an even smaller extent shrank the number of issuers they invested with. We find that smaller transactions execute at worse prices, i.e. lower yields, but that the transaction quality of a trade relates more strongly to the size of the complex of the fund making the trade than to the size of the trade itself."
Li, Liu and Musto state, "Among the targets of the reform was the resilience of money funds to market stress, and in particular to the risk of panic withdrawals. Money funds are not banks, but they share some features of banks that expose them to the risk of panics that resemble bank runs. Money funds compete with banks for accounts, and indeed were designed for this competition in the 1970s, when they offered.... This potential for a run became reality after the bankruptcy of Lehman Brothers."
The paper adds, "The details of what happened just after Lehman Brothers are key to the subsequent reform.... Reflection on this episode has identified two key reasons why these share classes were particularly vulnerable. One is that institutional accounts are larger and overseen by professionals, so there is a larger reward from redeeming at prices above market value, and a better chance that the investor will spot the opportunity and move quickly. The other is that prime funds invest in asset classes where credit risk, such as with Lehman, can depress market values. So the stable $1 quoted NAV imparted relatively more bank-run risk to these share classes, and this helps explain why the reform singled out institutional prime, along with institutional tax-free, share classes for the floating NAV, while allowing the retail share classes to continue, subject to limits, with the stable $1 quoted NAV."
The DERA work asks, "How should this flotation affect institutional funds, relative to retail funds? Should one expect more risk-taking, or less? There are really two questions here: how should the impending onset of flotation affect institutional vs retail funds, and how should the shift to flotation affect the funds in the longer run? Regarding the impending onset, it was widely reported in the days leading up to the October 14th, 2016 effective date of the reform that prime funds, and especially institutional prime funds, were scaling back risk out of concern for what might transpire when the day came. The analysis in this paper can shed light on the shape and extent of this anticipatory effect, but the focus is more on the latter question, the effect of flotation once the effects specific to the transition have played out."
It comments, "Flotation could encourage more risk-taking, by the logic of bank-run theory. That is, flotation means that redeeming when the NAV is below $1 no longer affects the wealth of those who don't redeem, so the incentive to join a run is weaker, so the threat posed by low NAVs to fund stability is diminished, so the managers could be less averse to risking low NAVs. On the other hand, the floating NAVs could encourage less risk taking, to the extent that the money-fund investor base liked money funds the way they were, when NAVs didn't fluctuate, and so will vote with their feet for funds that minimize fluctuation."
The authors also tell us, "Our analysis of the reform has three parts. The first part takes the highest-level view, looking at the transition of the industry through the reform, and documenting which choices were made for which funds, and how the relative yields and NAVs of money funds evolved. This analysis pays particular attention to those money funds that remained prime funds over our sample period, 2015 through 2017, and focuses on the evolution of the difference between retail and institutional funds. The second part uses data on the securities purchased by the funds to explore the relative risk taking of institutional vs. retail funds, and the implications of this risk-taking for NAVs. The third part asks how funds have handled their asset shrinkage through the reform."
Finally, the piece concludes, "This paper documents the effect on money-fund portfolio management of the NAV flotation introduced for institutional, but not retail, prime funds by the October, 2016 reform. This analysis focuses primarily on those funds that remained prime funds from January, 2015 to December, 2017. Our main finding is that the net effect, over the three years, is small. There was an effect in the months around and just after the reform took effect, generally in the direction of less risk-taking by institutional funds, but this difference faded through 2017, leaving the relative risk-taking of institutional and retail funds about where it started, on all the dimensions we explore. These dimensions include the interest-rate, liquidity and credit risk of the portfolios, and the volatility of the funds' NAVs. The funds' gross yields follow a similar pattern, with institutional funds earning less than retail funds during the period of reduced risk-taking, but earning the same by the end."
It adds, "The prime category shrank during the sample period, in particular the institutional prime category. Many funds ceased being prime funds, most often by converting to government funds or liquidating. Despite these departures, the funds that remained also shrank. This shrinkage may have played a role in the small magnitude of the effect of reform on institutional-fund management, since the investors most averse to NAV volatility would intuitively be the ones most likely to have left those funds. While the reform does not appear to have ultimately affected risk-taking, funds still did have to adapt to the shrinkage.... These influences of transaction, fund and complex size on transaction quality may help explain which funds persisted as prime money funds."
