Guide

NAHB Housing Market Index explained

Harbor Development’s land-acquisition team added a Phoenix parcel in March when the NAHB Housing Market Index (HMI) printed 42 — up five points from January — even though Census building permits were still falling month-over-month. Six weeks later, single-family permits stabilized and a national builder exercised its option on the adjacent phase. The HMI did not predict the exact permit number; it signaled that builders were seeing better foot traffic and six-month sales expectations before hard authorization data caught up.

The HMI is a monthly diffusion index published jointly by the National Association of Home Builders (NAHB) and Wells Fargo. It surveys single-family home builders on current sales conditions, sales expectations for the next six months, and prospective buyer traffic. Because it releases around the 15th of each month — roughly two weeks before Census permits and starts — it is one of the earliest soft reads on the for-sale housing cycle. This guide covers survey scope, component math, regional tables, how HMI relates to mortgage rates and new home sales, a Harbor Development sentiment read worked example, an indicator decision table, pitfalls, and an investor checklist.

What the NAHB Housing Market Index measures

The HMI is not a count of homes sold or permits issued. It is a sentiment survey: NAHB asks member builders whether conditions in three areas are “good,” “fair,” or “poor” compared with the prior month. Responses are converted into diffusion indexes where values above 50 mean more builders report improving conditions than worsening ones, and values below 50 mean the opposite.

The survey covers single-family home builders only. It does not include multifamily apartment developers, remodelers, or land developers who never build spec homes. That scope matters when you pair HMI with Census data: multifamily permit waves can move headline construction numbers while HMI stays flat because for-sale builders and rental developers face different financing and demand curves.

NAHB typically surveys a few hundred builders each month. Participation is voluntary and weighted toward association members, so the sample skews toward established regional and national builders rather than tiny custom-home shops. The index is still widely followed because those builders account for a large share of spec inventory and because public homebuilder earnings management teams reference HMI trends on conference calls.

The three HMI components

Each release publishes three sub-indexes plus a headline composite. Understanding the split prevents over-reading a single number:

Present single-family sales — how builders rate current sales of newly built homes. This component is the most coincident: it reflects contracts signed and closings in progress. A jump here often confirms that rate-sensitive buyers returned after a mortgage rate dip rather than predicting a distant future.
Sales expectations for the next six months — the forward-looking piece. Builders factor in their backlog, planned community openings, incentive budgets, and rate outlook. This sub-index often turns before present sales when builders anticipate better spring selling seasons or fear incentive wars ahead.
Traffic of prospective buyers — foot traffic through model homes and online lead quality. Traffic can improve while sales lag if buyers are browsing but waiting for rates or prices to move. Harbor Development watches traffic leads expectations leads present sales as a rough chain at cycle inflection points.

The headline HMI is a weighted average of the three components (historically equal weights). A headline print of 50 does not mean “neutral housing market” in absolute activity terms — it means an even split between improving and worsening sentiment among surveyed builders. Activity levels can still be low in a deep downturn even when sentiment rebounds from 35 to 45.

Diffusion index math and the 50 breakeven

Like ISM PMI and regional Fed manufacturing surveys, the HMI uses diffusion index arithmetic:

HMI = % reporting “good” + 0.5 × % reporting “fair”

The “poor” share is implicit: if 20% say good, 50% say fair, and 30% say poor, the index is 20 + 0.5 × 50 = 45. Values above 50 indicate net improving sentiment; below 50 indicate net deterioration. The scale is bounded between 0 and 100 but rarely hits extremes except in 2008-style collapses or post-pandemic booms.

Month-over-month changes of two to three points are often within survey noise; sustained moves of five or more points over two or three months usually reflect real shifts in builder planning. Compare component moves: if traffic rises but expectations fall, builders may see browsers but not confident enough to raise prices or cut incentives.

Release calendar and lead-lag relationships

NAHB releases the HMI on or around the 15th of each month at 10:00 a.m. ET, covering conditions during the survey window in the first half of that month. Census housing starts and permits follow on the 17th or 18th (8:30 a.m. ET). That two-day gap is small for traders but meaningful for land investors: HMI is the last major soft signal before hard authorization counts confirm or contradict builder optimism.

Empirically, HMI leads single-family permits at turning points by one to three months, not at every data point. During steady expansions, permits and HMI move together. During rate shocks, HMI often collapses first as traffic dries up; builders delay pulling permits until they clear inventory with incentives. During recoveries, traffic and expectations can rebound while permits lag because labor, lot supply, and municipal processing slow authorization.

Pair HMI with new home sales (Census, monthly, mid-month) and Case-Shiller price indexes (lagged two months) for a fuller demand picture. HMI is forward sentiment; new home sales are contracted demand; Case-Shiller measures closed resale prices that feed shelter CPI with long lags.

Regional HMI tables and national vs local reads

NAHB publishes regional indexes for the Northeast, Midwest, South, and West. National HMI can mask divergences: Sun Belt builders may report strong traffic while Northeast builders cite high insurance costs and slow migration inflows. Harbor Development maps each land bank to a Census division and tracks regional HMI against local permit tables rather than national headlines alone.

Regional HMI is especially useful when comparing builder exposure. A portfolio heavy in Florida and Texas homebuilders should weight South and West regional prints; Midwest-centric builders need the Midwest line. National equity indexes for homebuilders (XHB, ITB) react to headline HMI, but fundamentals diverge when regions desynchronize.

Harbor Development sentiment read worked example

Harbor’s monthly housing memo uses a simple rule set after each HMI release:

Record headline HMI and all three components — note month-over-month and three-month average. A headline below 40 for three months triggers defensive land-sale mode; above 45 for two months after being below 40 triggers selective lot acquisitions.
Check traffic vs expectations gap — traffic up 6 points, expectations flat: Harbor scouts model-home traffic in Phoenix and Jacksonville before committing capital; traffic confirms demand, expectations confirm builders have not yet priced it in.
Cross-check 30-year mortgage rate — if HMI rises while rates are still elevated, the move may be incentive-driven (buydowns, rate locks). Harbor reads builder earnings call language on incentive spend before assuming margin expansion.
Wait for permits confirmation — HMI alone does not close land deals. Harbor requires single-family permits SAAR to stabilize or rise within two monthly Census releases after a sustained HMI upturn.
Log one paragraph for the IC — e.g. “HMI 42 (+5 m/m). Traffic +7, expectations +4, present sales +3. South regional HMI 48 leads nation. Maintain Phoenix option; hold Jacksonville until permits confirm.”

Harbor treats HMI as an early warning and positioning tool, not a standalone trade signal. Sentiment without authorization and sales confirmation has led to false starts after temporary rate dips.

Indicator decision table

Question you have	Best indicator	Why
Earliest builder sentiment before hard data?	NAHB HMI (traffic + expectations)	Monthly survey; releases before Census permits
Authorized future construction units?	Building permits (single-family)	Hard Census count; lags sentiment at turns
Contracts signed on new homes?	New home sales SAAR	Census contract data; can cancel before start
Affordability constraint on demand?	30-year mortgage rate + median price	Payment-to-income drives traffic and sales
Apartment supply pipeline?	Multifamily (5+) permits	HMI excludes multifamily developers
Closed resale price momentum?	Case-Shiller HPI	Repeat sales; lags contracts by months
Macro recession signal from housing?	Single-family starts + permits trend	NBER-weighted; HMI is soft and volatile
Builder earnings pre-announcement read?	HMI present sales + traffic	Correlates with order growth commentary

Common pitfalls

Treating HMI as a volume index — it measures sentiment diffusion, not homes built or sold.
Ignoring the single-family scope — multifamily booms do not appear in HMI.
Trading one-month spikes — use three-month averages; survey noise is real.
Assuming HMI above 50 means a strong market — 52 can still accompany historically low activity levels.
Skipping component splits — headline moves driven only by traffic differ from expectations-led turns.
Overlooking incentives — rising HMI during heavy buydown campaigns may not imply margin recovery.
Applying national HMI to every metro — read regional tables for geographic portfolios.
Expecting HMI to lead permits every month — lead-lag appears at inflection points, not in steady states.

Investor checklist

On release day (~15th of month, 10:00 a.m. ET), record headline HMI and three components.
Compare surprise to consensus; note month-over-month and year-over-year changes.
Calculate three-month moving average of headline and traffic sub-index.
Read Northeast, Midwest, South, and West regional prints for geographic exposure.
Cross-check 30-year mortgage rate trend and weekly MBA purchase applications if available.
Two days later, compare Census single-family permits to HMI direction for confirmation or divergence.
Map to holdings: homebuilders, building materials, mortgage REITs, regional banks.
Log whether move is traffic-led (early) or present-sales-led (confirmation).

Key takeaways

The NAHB HMI is a monthly builder sentiment diffusion index for single-family for-sale housing.
Three components — present sales, six-month expectations, buyer traffic — tell different parts of the demand story.
Values above 50 mean net improving sentiment among surveyed builders, not necessarily high activity.
HMI often leads single-family permits at cycle turns by one to three months but confirms with hard data during steady periods.
Harbor Development uses regional HMI plus traffic-expectations gaps to time land options before permit stabilization.