blog

Selecting Attractive Pools with Infima’s Solutions

MBS investors are increasingly searching for faster prepayment speeds, as mortgage rates rose above 7% and most of the universe is trading at a steep discount. Because most specified pools were built to protect against fast prepayments, the search for these fast-paying stories is no easy task. Investors can leverage Infima’s solutions to not only find pools with the most attractive prepay profiles but also analyze volatility of their investments under different rate scenarios.

Finding Needles in the Coupon Stack 

To start, we look across the coupon stack to decide where to position in terms of convexity. Negative convexity is a disadvantage to MBS because predicting the embedded prepayment option is complex. Therefore, this convexity proxies the prepayment risk for which investors need to be sufficiently compensated. Infima offers long-term metrics such as option-adjusted convexity. We plot the option-adjusted convexity (OAC) by coupon below, averaged equally between October 7 and November 10. ­ Note that ­convexity accelerates into negative territoryafter FNCL 3.5.(Exhibit 1):

Exhibit 1: OAC significantly worsens after FNCL 3.5 to FNCL 4.0

Sources: Infima Technologies, Inc. 

We then plot the option-adjusted spreads (OAS) to see if valuations reflect the convexity of each coupon (Exhibit 2):

Exhibit 2: FNCL3.0, 3.5 and 4.0 share almost the same OAS  

Sources: Infima Technologies, Inc. 

Fannie 3s and 3.5s offer similar OAS as Fannie 4s as the curve seems to plateau in the middle of the stack. But, because Fannie 3s and 3.5s offer an advantage in OAC, they would be a better investment opportunity. Fannie 4.5s and above have nearly four times the negative convexity, while Fannie 2.5s and below have OAS that are richer by nearly 30 bps. Exhibit 3 below delineates Fannie 3s and Fannie 3.5s with near-zero OACs but over 100 bps of OAS:

Exhibit 3: Coupon Stack shows various rewards for convexity risk 

Sources: Infima Technologies, Inc. 

Checking the value of the FNCL 2.5 butterfly, which measures the negative convexity of the stack around FNCL 2.5, confirms that it is trading with poor carry. Nearly three standard deviations below its three-month average, its carry stands in stark contrast with that of FNCL 3.0 fly, currently 2.25 standard deviations above its three-month average. Accurate predictions of prepayment speeds are critical for security selection, portfolio construction and risk management in Agency MBS markets. Infima’s transformative deep learning technologies set new prediction standards, delivering performance boosting edges to MBS market participants including investors and dealers.

Jinzhao (JZ) Wang

Head of Client Solutions, Infima

Jinzhao Wang is head of client solutions at Infima. Based in New York, her experience spans trading, strategy, and quant modeling for structured products focusing on MBS. Before joining Infima, she traded MBS, co-managing a book of specified pools and TBAs at Mitsubishi Securities. Jinzhao previously served as an investment strategist analyzing CLOs and MBS at Amherst Pierpont Securities. Before that, she worked as a data scientist in academia and start-ups.  

Jinzhao earned an MS in Finance and Management (2016) from the European School of Management and Technology and a BA in Applied Mathematics (2014) from Harvard College.