Why 2026 foreclosure gains are not a housing crash signal

Foreclosure data

One of the things I’ve stressed when I talk at events this year is that we have had many recessions post-WWII but only one foreclosure crisis. That foreclosure crisis started with a massive credit boom from 2002-2005, and then a credit bust. That credit bust pushed foreclosures up, according to New York Fed data, in 2005, 2006, 2007 and 2008. Then, the Great Recession happened. As you can see, none of that is happening now — we aren’t even back to 2019 levels yet, and it’s August 2026.

Here is how the Fed tracks the data: New foreclosures. Number of individuals with foreclosures first appearing on their credit report during the past 3 months. Based on foreclosure information provided by lenders (account level foreclosure information) as well as through public records.

This is key to what I will present next, because I can explain why housing inventory was down year over year, even though for 3.5 years now headlines were showing big percentage increases in foreclosure data.

Inventory

When you don’t have a lot of distressed sellers in the mix, we just deal with the normal supply and demand equilibrium for housing; a surge of actual foreclosures in 2026 would have easily put the inventory data much higher in 2026.

Keep it simple: housing demand is up 2.4% year to date and new listings didn’t explode, so inventory growth slowed and declined only slightly year over year in the last existing home sales report this week. 

We track inventory differently than the NAR; we have no contract data in our inventory, so these are the homes available for sale. Inventory is up 0.78% fron the previous week.

• Weekly inventory change (July 31-Aug. 7): Inventory fell from 872,932 to 865,709
• Same week last year (Aug. 1-Aug. 8): Inventory fell from 865,600 to 859,050

New listings data is key

When you have a massive buildup in foreclosure data, as we saw from 2005-2008, you will get a surge of new listings data. These aren’t sellers that will be buyers; these are distressed sellers in the mix. Not to mention, after a significant high-LTV credit boom and bust, a ton of people were underwater: In 2010, over 23% of homes were underwater. The run-up in foreclosure data from 2005-2008 was going to be a problem, because the higher the percentage of underwater homes, the more likely a foreclosure will happen. In contrast, people with a lot of equity can sell and prevent that foreclosure.

In addition, our new listing data isn’t surging. From 2013-2019, the normal for our new listings has been 80,000-100,000 per week during the seasonal peak months — and we haven’t had any normal years since 2020. New listings have picked up over the last two years, but it’s mostly been the traditional seller-as-buyer. This explains why inventory growth has been low this year.

Here is last week’s new listings data for the past two years:

• 2026: 67,301
• 2025:  66,341

Some context for those who believe the new listings data resembles the housing bubble years: during that time, new listings ranged from 250,000 to 400,000 per week for several years. Let me repeat that: 250,000-400,000 per week for years. New listings data today isn’t even back to normal levels, with foreclosure data not back to 2019 levels.

Conclusion

Once the foreclosure data starts to pick up beyond a normal trend — and it will with a job-loss recession — then you need to wait for it to be reflected in the new listings data. The entire process, from start to finish, might take 9-18 months; in some cases, many years. Understanding the data means you can properly track and talk about foreclosures and the relationship of foreclosures to inventory.