A trio of researchers from NERA Consulting published a white paper entitled, "Money Market Mutual Funds: Stress Testing and the New Regulatory Requirements," which analyzes how money market funds will react in stress scenarios under the new rules. The authors -- Jeremy Berkowitz, Patrick Conroy, and Jordan Milev -- reveal some interesting results. They write, "In the coming months, as money market funds gear toward meeting these requirements, it is increasingly important to be aware of the types of scenarios that the enhanced, SEC-mandated stress tests include and the appropriate econometric and data simulation framework that these funds would need to adopt -- with board oversight -- in order to meet SEC's requirements.... In this article, we demonstrate this framework and evaluate the results from running prescribed stress scenarios for a stylized money market fund to assist funds, their managers, and boards in preparing to implement the rule. The enhanced stress testing requirements may increase demands on money market fund boards to evaluate the comprehensiveness and implementation of the stress testing methodology."
The 9-page report says, "Perhaps the most significant of the 2014 reforms is the enhanced stress testing requirement. This reflects the SEC's concern that money market funds be prepared to deal with particular events such as interest rate changes, higher redemptions, and changes in credit quality of the portfolio. Since the draft rules in 2010, the SEC has required money market fund managers to examine a fund's ability to maintain a stable net asset value per share in the event of such shocks. However, starting in April 2016, money market funds will be required to conduct substantially different, enhanced stress testing. The new prescriptive stress testing regime is designed to minimize the possibility of a failure by forcing money market funds to regularly ask what would happen to their liquidity position under adverse market conditions."
NERA continues, "In particular, the new regulations require funds to test their ability to maintain weekly liquid assets of at least 10% of total assets in response to several scenarios. The hypothetical stress events include: a. increases in the level of short-term interest rates; b. a downgrade or default of particular portfolio security positions; and c. a widening of spreads compared to the indexes to which portfolio securities are tied. Additionally, these stress scenarios must include increases in shareholder redemptions to various levels, and management is encouraged to add any other combinations of events deemed relevant. Recent litigation -- the case of the SEC vs. Ambassador Capital Management in 2014 for example -- has shown that portfolio managers and board members need to be keenly aware of, and fully understand the complex stress testing requirements described in rule 2a-7 of the Investment Company Act of 1940. Failure to implement even one dimension of the prescribed stress scenarios can result in severe penalties for the firm and its officers."
The researchers provide stress testing examples. "We constructed a stylized balance sheet for a hypothetical money market fund, based on the N-Q filing of a major US money market mutual fund in Q4 of 2014. The filing gives detailed information on the portfolio holdings of the fund, including commercial paper, certificates of deposit (CDs), variable rate notes, fixed rate notes, and repurchase agreements ("repos"). We assume that, prior to each of the stress scenarios, the fund holds 1% of its assets in cash and that 41% of total assets are weekly liquid assets. We then subject the portfolio to three of the stress tests required under the new SEC regulations, and report the required measures: weekly liquid assets and principal volatility."
They continue, "The first stress test scenario addresses an increase in the short-term interest rate. We subject the portfolio to a stress test scenario in which the 1-month and 3-month treasury rates increase simultaneously by 30 basis points. We assume this increase in rates continues to hold throughout the stress period of 10 business days. As required under the new rule, we consider various different assumptions about shareholder redemption rates. We assume they increase by 10%, 20%, or 30%, respectively, and remain at that level over the course of the 10-day stress period."
Further, NERA writes, "The second stress test scenario addresses a widening of spreads. In particular, the scenario assumes that credit spreads increase by 1.5%." The paper includes a table which explains the results. "The left side of the table, under the heading "increase in interest rates," shows the stressed levels of weekly liquid assets after the 10-day shock has run its course. The liquidity level declines several percentage points to about 34.7% under the 10% increase in redemption rates and down to 34.4% under the more severe 30% increase. This stress test reveals an important insight about this particular hypothetical money market fund: the interest rate shock and the spread shock scenarios have a profound effect on weekly liquid assets, even for a mild increase in redemption rates of 10%. Additionally, an increase in redemption rates that is three times higher, at 30%, does not change the results substantially. In other words, for this kind of money market portfolio, redemptions have less of an effect on existing shareholders than changes in interest rates. This illustrates the ways in which stress testing allows funds to glean valuable business insight into the dynamics of their portfolio."
They also tell us, "The results indicate that the volatility of NAV would be about 0.8% in response to the interest rate shock and about 1% under the spread shock. Since there is little historical experience with floating NAV funds to put these numbers in context, it is useful to plot the NAV over the course of the 10-day stress event, shown below in Figure 1. When the Reserve Primary Reserve Fund "broke the buck" in 2008, despite having a nominally non-floating NAV, it did so by 3%. In our interest rate shock scenario, the NAV does not drop as steeply, but it does experience a significant decline of nearly 2%. While this is a rather severe decrease, it should be kept in mind that the stress scenario envisioned here is an extreme, overnight move in short rates and in spreads."
They add, "Additionally, we note from the graph that the spread shock results in an even larger decrease in NAV. This is likely due to the fact that the spread shock is a 1.5% increase in spreads, occurring in one day, which would be a historically rare event. Of course, the appropriate parameters of the stress test would depend on many factors. A key part of the effort to design an informative stress test would be to use an appropriate stress-testing range for the relevant parameters. Lastly, it is interesting to note that the decline in NAV occurs almost entirely in the first two days of the stress period, even though redemptions continue to occur at an elevated level for the entire 10 days. This again reflects the fact that, for typical money market portfolios, redemptions have much less of an effect on existing shareholders than the initial change in interest rates or spreads."
The third stress deals with downgrade risk. "The third shock, shown in Panel B, is a major counterparty downgrade. We assume that the counterparty comprises either 2% of the commercial paper held in the portfolio, or 5%. For illustration purposes, we assume that the downgrade results in a 20% reduction in the value of the notes. The results indicate that the reductions in weekly liquid assets are on par with the first two stress scenarios, but the increase in principal volatility is more varied and can be higher. For a high redemption-rate scenario, the variation in NAV can reach 2.5%. Again, assessing whether that is large or small requires an objective view as to what would also be reasonable variation during times of stress."
In conclusion, the NERA paper says, "The SEC's money market fund reforms will undoubtedly change the way that many money market funds operate in a substantive way. The reforms will create new challenges, including enhanced data management, new models for valuing positions and contingencies, and additional report generation and submission capabilities. Risk management procedures will need to be augmented to include evaluation of fund liquidity under complex new stress scenarios in order to ensure compliance with the new rules. Economic and quantitative issues arising from this new framework will need to be dealt with promptly. We expect that forthcoming rules on stress testing from the SEC for broker-dealers will present further challenges, and we will provide an update when those draft rules become available."
Finally, they also outline "Seven Stress-Testing Issues to Consider," writing, "Based on our experience with stress testing methodologies, here is a list of seven questions money market fund boards should be asking when they assess the results presented to them. 1. Do the scenarios cover all SEC-mandated stress testing scenarios? 2. Is management making full use of the testing framework by including other relevant scenarios of business interest? 3. Are the parameters chosen for the stress testing scenarios realistic? 4. How often are the parameters of the approach recalibrated to reflect changes in the portfolio? 5. How effective is the internal validation of the methodology used? 6. What is the margin of error (confidence interval) of the results presented? 7. What is the best way to summarize and present the results of a set of stress tests?" For more information go to NERA.com or download the white paper.
