U.S. March New Home Sales Bounce Back from January Lull but Show No Broader Acceleration
Core Conclusion
March new home sales data confirm the housing market remains range-bound at depressed levels, with no evidence of a sustained recovery. The 7.4% month-over-month bounce to 682k units merely reverses January's weakness and keeps the three-month moving average at 633k, below the 655k recorded a year ago. Any acceleration signal from late 2025 has fully reversed. Residential investment in 1Q GDP should see only a marginal upward revision of roughly 20 basis points—insufficient to alter the macro narrative or justify bullish positioning in housing-exposed assets.
Sales Bounce Is a Statistical Artifact, Not a Trend Inflection
February and March sequential gains simply recovered from January's outsized 19.9% decline. The three-month moving average of 633k remains below both the 2025 year-end level (709k in December) and the year-ago average. By regional composition, the March uptick was concentrated in the Northeast (27k from 15k) and South (441k from 397k), while the Midwest and West declined—a geographically uneven pattern inconsistent with broad-based demand improvement.
Investment implication: Single-month new home sales data are volatile and should be viewed through the three-month average lens. The current 633k trend suggests no cyclical acceleration; positioning for a housing recovery based on this print alone is premature.
Price Declines and Elevated Inventory Continue to Depress Builder Sentiment
Median new home prices fell 6.2% year-over-year to $387,400; average prices declined 1.2%. Months' supply stood at 3.1, marginally above February's 3.0, reflecting persistent inventory overhang. Completed units accounted for 56% of March sales, while units not yet started made up only 8%—near multi-decade lows. This composition signals builders lack confidence to begin speculative construction, consistent with subdued housing starts (weather-boosted March jump aside).
Investment implication: Falling prices and high supply months constrain builder margins. Until new home sales sustain above 700k for multiple months, housing starts are unlikely to gain meaningful upward momentum.
Residential Investment GDP Impact: Negligible
The BEA's initial 1Q GDP estimate assumed new home sales at 600k for both February and March. Actual prints of 635k and 682k imply a roughly 20 basis point upward revision to residential investment, which was initially reported at -8.0% (quarter-over-quarter annualized). Thursday's construction spending data will provide additional input.
Investment implication: A 20bp GDP adjustment is macro-irrelevant. The housing sector remains a modest drag on growth, not a driver. Markets should focus on Thursday's construction spending release for confirmation, but the weak structural narrative persists.
Key Risks
- Prolonged elevated interest rates: Mortgage rates staying higher for longer would further suppress buyer affordability and builder starts, prolonging the current stagnation.
- Existing home sales dysfunction: The low supply of existing listings (not captured by new home data) keeps total single-family sales at depressed levels—new homes at 15.8% of all single-family sales, well above historical norms, reflects this distortion, not new home strength.
- Macroeconomic downside: An economic slowdown or labor market deterioration would disproportionately hit discretionary new home purchases, deepening the current bottom.
Trade Implications
The housing sector lacks structural catalysts for upside. The modest 20bp GDP revision from this data is noise, not a signal. Second-half 2025 sales may see a minor pickup, but upside remains contingent on a rate path that is currently uncertain. Rate-sensitive assets (homebuilders, building products, regional banks with mortgage exposure) should be approached with caution. Thursday's construction spending release provides the next near-term data point for direction confirmation. No tactical re-rating is warranted from the March new home sales release.