Data in Action

Tapping into mortgage origination data for capital market signals

Inflation has crept into nearly every facet of life -- from gasoline to avocados, prices are substantially higher now compared with a year ago, and financial markets are on high alert. As the Federal Reserve and other central banks begin to remove monetary policy accommodation, asset valuations have come under pressure. While real estate has not yet followed stocks and bonds lower, it is attracting scrutiny as many pundits expect affordability issues to eventually crimp demand.

A key input to determining housing affordability is the interest rate at which a potential buyer can borrow money to finance their home purchase. Transparency into this variable is critical for market participants operating across the life cycle of a mortgage loan. Originators can benefit from understanding how their lending practices compare to the broader industry. Servicers can more efficiently adapt to changes in lending conditions. And investors can access more timely and representative measures of residential home loan borrowing activity to fine tune prepayment or relative value modeling. The ICE U.S. Residential Mortgage Rate Lock Indices use a data-driven methodology based on loan applications processed by ICE Mortgage Technology to track the average rate at which new residential home loans are locked each day. These benchmarks can serve as valuable indicators of where actual lending is occurring. The ICE U.S. Residential Mortgage Rate Lock Composite Index, which aggregates rate lock activity for all types of U.S. residential loans, has risen nearly 250 bps year-to-date, and is sitting at a multi-year high.

The ICE family of residential mortgage rate lock indices contains over 80 indices that allow for detailed analysis of relationships and trends. These sub-indices range from loan program, to loan term, to loan purpose, with additional sub-indices produced for borrower attributes such as Loan to Value (LTV) and credit score (FICO). Over the past 18 months, the trajectory of interest rates has crept higher, yet the pace of increase has differed from instrument to instrument. For example, using the ICE U.S. Residential Mortgage Rate Lock Index, we can examine the path taken by 30-Year Fixed New Purchase Mortgage Rate Locks for Conforming loans and compare this to movement in Jumbo loans. Interestingly, over the past 18 months, the relationship between those two loan programs (i.e. the Jumbo-Conforming basis) has grown from plus or minus a few basis points to as much as negative 44 bps. All else equal, this would imply that while mortgage borrowing has become more expensive, Jumbo loans have seen a more muted rise relative to comparable Conforming loans.

Source: ICE Data Services

Identifying explanatory drivers for changing relationships such as the Jumbo-Conforming basis, can be difficult. However, data from the ICE U.S. Residential Mortgage Rate Lock Index Series can shed light on the attributes of the loans processed by ICE Mortgage Technology. Here, we can examine the Loan to Value (LTV) ratio and Credit Score (FICO) of the ICE U.S. Jumbo 30-Year Fixed New Purchase Mortgage Rate Lock Index from a loan weighted average perspective. In this case, data shows that the LTV of Jumbo loans has decreased over time, with LTV currently running below average and FICO has increased, with FICO approximately at average. Other data that can be a signal for investors is the points & fees paid by borrowers, calculated as the difference between APR and the locked rate. In our Jumbo 30-Year Fixed New Purchase Mortgage example, it appears that lenders may be capturing more upfront fees in exchange for a better rate.

Source: ICE Data Services

Demand from investors can certainly influence rates at which originators are willing to lend. Traditionally, mortgage investors, either in whole loan form or via securitizations, have been buyside firms such as asset managers, insurance companies, and hedge funds. Yet since the financial crisis, the Federal Reserve (“the Fed”) has become one of the largest investors in mortgage-backed securities (MBS), and now owns more than $2.7 trillion. Recently, the Fed embarked on a quantitative tightening campaign by allowing its holdings to run off through paydowns, eliminating a large source of demand for Agency MBS. The other side of the coin is supply, which has seen a significant slowdown as rates have risen. According to SIFMA data, both Agency and Non-Agency issuance is running at a slower pace this year compared to the same time last year. But while Agency issuance this year is basically in-line with 2020, Non-Agency issuance is substantially below. The feedback loop from supply/demand technicals (i.e. who is buying/selling, how much are they buying/selling, and at what levels they wish to buy/sell) can potentially explain changes in rate lock activity.

Source: SIMFA

The fast and furious jump in mortgage rates this year has manifested in some eye-catching secondary market movements where Agency MBS are traded. Investors in this segment are closely watching the “current coupon” - essentially the MBS concept most closely tied to what is happening in the primary market. Although not a one for one relationship, this concept can help approximate the rate at which originators are willing to make mortgage loans at a given time, and as a function of that, what supply looks like in the pipeline, and where to expect the most hedging activity to occur.

Examining data from ICE’s Evaluated Pricing service, we can see that prior to the start of 2022, the entire liquid coupon stack of 30 Year Agency MBS, 2.5% through 5.5%, were being priced by the marketplace at a premium over par. Over the past six months, prices for those same MBS have almost all fallen below par, indicating a significant portion of the universe now priced at a discount. Par is an important pivot point for investors because it is an inflection point for the impact of prepayments on returns. With discount prices, faster prepayments would lead to higher returns. Understanding how pricing trends may be unfolding in the secondary can help inform decision making for primary market participants and vice versa.

Source: ICE Data Services

Although Agency securitizations make up the lion’s share of the MBS universe, Private Label, or Non-Agency paper, is an equally important segment. With relatively more liquidity and transparency, in some cases Agency MBS can be used as benchmarks for gauging valuation of comparable Non-Agency MBS. Based on ICE Evaluated Pricing data, the relationship or basis between Agency and Non-Agency passthroughs priced relative to TBAs (To Be Announced), currently ranges from -2.4 points to -4.4 points on average. These pricing spreads change over time and are similarly influenced by supply/demand technicals as well as fundamentals.

Source: ICE Data Services

The connection between primary and secondary mortgage markets is a critical one that players up and down the loan life cycle are keen to tap into. Through its various product offerings, ICE can assist investment decision makers to quantify signals from the origination world and translate them into actionable insights for security selection, risk management, benchmarking, and modeling. Our newest foray in this dynamic space are the ICE Mortgage Rate Lock Index Futures which offer market participants a more precise hedging tool for residential mortgage interest rate risk. We are excited to continue to innovate and leverage ICE’s end-to-end mortgage workflow capabilities for capital markets use cases.