Finding states with faster housing turnover

Brian Landy, CFA | February 11, 2022

This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors. This material does not constitute research.

Mortgage rates continue to climb, raising the importance of finding securities that offer protection against slowing prepayments. Housing turnover is the primary driver of speeds in loans without incentives to refinance, and home-buying rates vary across states. Loans from states with faster turnover should be more valuable in MBS trading below par, but most states’ loans don’t command a pay-up over the TBA price. Investors should be able to buy pools backed by states with faster turnover at little to no pay-up over TBA. Although there is no guarantee that state-by-state turnover differences persist, differences were stable through last year, and a low pay-up minimizes the damage if a state’s turnover speeds do slow.

Home buying has been robust throughout the pandemic as homeowners look for bigger homes to support work-from-home. Remote work also has allowed employees to move farther from the office. But measuring prepayment speeds from turnover is difficult, since the prepayment data does not indicate the reason that a loan paid off. A typical approach is to estimate turnover speeds using discount pools, but few pools have traded at a discount throughout the pandemic.

Another approach is to compare the volume of purchase originations to the number of loans outstanding in each state. This provides a relative indicator of turnover prepayment activity in different states and over time, although it doesn’t necessarily quantify an exact prepayment speed. A purchase could involve a new home, which would not trigger prepayment of an existing loan. A purchase could also involve a home with no mortgage, which again would not trigger a prepayment. But the vast majority of purchases involve existing homes with existing loans. In December, for example, an annualized 6.18 million existing homes sold compared to an annualized 811,000 new homes.

Florida, South Carolina, and Texas appeared to lead the pack last year using this turnover metric (Exhibit 1). The level of purchase activity was estimated using MBS data provided by Fannie Mae, Freddie Mac, and Ginnie Mae. The highest number representing the highest level of turnover activity. There are large differences across states. An estimated 13.7% of loans turned over in the fastest states, while the slowest states—Vermont, Massachusetts, and California—had only 7.7% turnover.

Exhibit 1. Florida and South Carolina led the way in purchase activity

Source: Fannie Mae, Freddie Mac, Ginnie Mae, Amherst Pierpont Securities

This metric is not perfect since it only incorporates information on loans in agency MBS pools. The results can be influenced by the flow of capital to and from other parts of the mortgage market—private-label securitizations, bank portfolios, and even loans paid in cash. But the agency MBS market is large enough that it likely provides a representative sample of the activity in these markets. The numbers seem consistent with reports from other sources. For example, CoreLogic reported that homeowners were migrating from states like California and New York, which are among the slowest in the chart, to states like Texas.and Florida, which are among the fastest.

The relative levels of turnover activity were stable throughout 2021 (Exhibit 2). This compares each states’ purchase activity to its activity in the preceding quarter. The fastest states typically remained the fastest, and the slowest states typically remained the slowest. There is an overall slowing trend evident throughout the year, but that does not alter the relative behavior.

Exhibit 2. The fastest states stayed fast throughout 2021

Source: Fannie Mae, Freddie Mac, Ginnie Mae, Amherst Pierpont Securities

Over time some states are sure to speed up, and others slow down, but this analysis suggests sudden shifts are unlikely. Furthermore, only a few states—New York, Florida, and Texas—produce loans for which investors have to pay more than the TBA price. There should be opportunity to look for pools backed by states like South Carolina, Alabama, and Nevada that could offer faster than average housing turnover and extension protection at little to no pay-up over TBA.