NAHB Housing Market Index Explained: What Builder Confidence Actually Measures

The NAHB/Wells Fargo Housing Market Index (HMI) is a monthly measure of builder sentiment in the market for newly built single-family homes. It is based on a survey that asks single-family builders to rate current sales, expected sales over the next six months, and the traffic of prospective buyers. Their answers are turned into an index that runs from 0 to 100.

In its October 19, 2026 release, the National Association of Home Builders (NAHB) reported that the HMI fell three points to 29 in October, the lowest reading since June 2012. All three components declined: current sales conditions fell to 33, sales expectations for the next six months to 32, and prospective buyer traffic to 18. (Source: NAHB, "Builder Sentiment Falls on Rising Rates," October 19, 2026.)

A reading of 29 is easy to misread. It does not mean home sales fell 29%, that the market is at 29% of normal, or that 29% of builders are confident. It is an index level built from survey responses, and its meaning comes from how the index is constructed. NAHB says readings above 50 indicate that more builders view conditions as good than poor, while lower readings signify less optimism.

This article explains what the HMI measures, the three questions behind it, how each component is scored, what a reading of 50 means, how NAHB combines the components, and why builder sentiment should be read alongside, not in place of, measured housing activity such as starts, sales, inventory, and prices. It contains no forecast.

About the data

The HMI is a survey-based sentiment measure. It is not a count of homes sold, started, permitted, or completed, and it is not a price or inventory measure.

What the Housing Market Index actually measures

NAHB describes the HMI as designed to gauge and track the pulse of the single-family housing market. It is based on a monthly survey of single-family builders. (Source: NAHB, Housing Market Index page.) NAHB notes that it has conducted the survey for more than 40 years, and its first HMI reading dates to January 1985. (Sources: NAHB, October 19, 2026 release; Housing Market Index page.)

What the survey captures is perception. Builders are asked how they see conditions in their markets: whether sales are good, fair, or poor, what they expect over the next six months, and how much buyer traffic they are seeing. The HMI summarizes those judgments.

That sets the HMI apart from most of the housing data covered in other Metro Analytics Insights:

• It is not a transaction count. It does not record how many homes were sold or went under contract.
• It is not a construction statistic. It does not count units permitted, started, or completed.
• It is not a price measure. It does not report sale prices or price changes.
• It is not an inventory measure. It does not count homes for sale.
• It is not a mortgage-rate index.

NAHB describes how its builder panel is drawn: the panel is stratified by region of the country and size of builder, and it is refreshed annually to support response rates and balanced participation. (Source: NAHB, Housing Market Index page.) NAHB's current HMI page and October 2026 release do not state a current monthly respondent count, so this article does not estimate one.

The HMI is therefore best understood as a structured reading of how single-family builders view their own markets.

The three questions behind the HMI

Each month, builders rate three conditions in the market for new single-family homes. (Source: NAHB, Housing Market Index page.)

The three questions behind the HMI

Component	What builders rate	Response options
Present sales	Current sales of new single-family homes	Good, fair, or poor
Expected sales	Sales of single-family homes expected over the next six months	Good, fair, or poor
Buyer traffic	Traffic of prospective buyers of new single-family homes	High to very high, average, or low to very low

Swipe horizontally to view all columns →

Source: NAHB, Housing Market Index page and October 19, 2026 release.

Present sales

33

How do builders rate current sales of new single-family homes?

Response scale: Good / Fair / Poor

Six-month expectations

32

How do builders rate expected sales over the next six months?

Response scale: Good / Fair / Poor

Buyer traffic

18

How do builders rate traffic of prospective buyers?

Response scale: High to very high / Average / Low to very low

All three are survey-based component indices.

The questions cover different time frames and different aspects of the market. Present sales describe how builders see current conditions. Expected sales describe their outlook over roughly the next half year. Buyer traffic describes how many prospective buyers they are seeing, which is a different question from how many are actually buying.

The response scales also differ. Sales conditions are rated good, fair, or poor. Traffic is rated high to very high, average, or low to very low. Each question produces its own component index, and NAHB publishes all three alongside the headline.

How each HMI component is scored

NAHB converts the responses to each question into a component index using a simple formula. (Source: NAHB, Housing Market Index page.)

• Present sales and expected sales: (percent "good" − percent "poor" + 100) ÷ 2
• Buyer traffic: (percent "high/very high" − percent "low/very low" + 100) ÷ 2

In plain terms, the formula takes the share of builders giving the positive answer, subtracts the share giving the negative answer, and rescales the result to run from 0 to 100. The middle response, "fair" for sales or "average" for traffic, is not counted as positive or negative. It sits in between.

NAHB lists how the index responds to shifts in answers. Each percentage point of builders moving from poor to good raises the index by 1 point, while a shift from fair to good raises it by half a point. Shifts in the other direction lower it by the same amounts. (Same source.)

The resulting scale has clear endpoints:

• 0: every builder gives the negative answer.
• 100: every builder gives the positive answer.
• 50: the share giving the positive answer equals the share giving the negative answer.

(Same source.)

ILLUSTRATIVE ONLY: hypothetical responses, not NAHB data. If 20% of builders rated present sales "good," 40% "fair," and 40% "poor," the component would be (20 − 40 + 100) ÷ 2 = 40. If the shares were instead 30% good, 40% fair, and 30% poor, the component would be exactly 50, even though only 30% said conditions were good.

That second case is the key to reading the scale. A component of 50 does not mean half of builders are positive. It means positive and negative answers are balanced.

What an HMI reading of 50 actually means

Fifty is the HMI's key reference point, and it is easy to describe loosely. For each component index, 50 is the midpoint of a diffusion index: it is where the share of builders giving the positive response equals the share giving the negative response. (Source: NAHB, Housing Market Index page.)

Component index scale

0

Negative response extreme

50

Positive-response share = negative-response share

100

Positive response extreme

This exact 50-point balance interpretation applies to each individual HMI component index. The headline HMI is a weighted composite.

For an individual component, the 50-point midpoint has that exact positive-share-equals-negative-share interpretation. The headline HMI, however, is a weighted composite of the three component indices. NAHB describes a headline reading above 50 as indicating stronger, more positive builder confidence. Its October press release says any number over 50 indicates that more builders view conditions as good than poor, and its HMI page says a reading above 50 indicates that the majority of builders feel confident about the current and near-term outlook, while lower readings signify less optimism. (Sources: NAHB, October 19, 2026 release; Housing Market Index page.)

Because the headline HMI is a weighted composite, a composite reading of 50 should not be read as a literal count showing that all positive responses across the survey exactly equal all negative responses.

What 50 does not mean is that 50% of builders are optimistic. Because neutral answers ("fair" or "average") also exist, a component reading of 50 is consistent with many different combinations, such as 10% positive and 10% negative with 80% neutral, or 40% positive and 40% negative with 20% neutral. The threshold is about balance between positive and negative answers, not the size of the positive share.

For a component, a reading below 50 means negative answers outnumber positive ones. October's headline of 29 is a weighted index level well below the 50 reference point. For context on the full range, NAHB reports that the HMI's all-time low was 8, in January 2009, and its all-time high was 90, in November 2020. (Source: NAHB, Housing Market Index page.)

The 50 threshold describes builder sentiment. It is not a dividing line between a healthy and an unhealthy housing market.

How NAHB combines the three components

The headline HMI is a weighted average of the three component indices. NAHB publishes the weights. (Source: NAHB, Housing Market Index page.)

59.20%

27.22%

13.58%

Present sales59.20%

Buyer traffic27.22%

Expected sales13.58%

NAHB publishes the seasonally adjusted composite HMI; rounded displayed component readings should not be used to reconstruct the headline.

Present sales carry the largest weight, followed by buyer traffic, with expected sales weighted least. NAHB's public methodology lists the weights but does not explain on that page how they were chosen, and this article does not speculate. The larger weight on present sales describes how the headline index is constructed. It is not a statement about which condition matters most in economic terms.

NAHB publishes the headline HMI as a seasonally adjusted index. (Source: NAHB, October 19, 2026 release.) Its public methodology does not set out every step of how seasonal adjustment interacts with the component weights.

For that reason, readers should not try to reconstruct the headline by multiplying the published, rounded component readings by the weights. The published composite reflects NAHB's full calculation, which may involve seasonal adjustment, unrounded values, or other steps not shown in the rounded readings. NAHB's published figure of 29 is the authoritative October reading.

Why HMI 29 does not mean the market fell 29%

29

Builder sentiment index

×Not a 29% decline in home sales

×Not 29% of normal

×Not 29% of builders confident

Index level, not percentage change.

Because the HMI is reported as a number between 0 and 100, it is often mistaken for a percentage. A reading of 29 does not mean:

• home sales or new-home sales fell 29%
• sales are at 29% of normal
• house prices fell 29%
• housing starts fell 29%, or construction will fall 29%
• 29% of builders are confident, or 71% are pessimistic
• buyer traffic fell 29%
• the housing market is "29% healthy"

The HMI is an index level. October's 29 is a weighted index level well below the 50 reference point, built from three component readings that were each below 50.

Changes in the HMI are measured in index points, not percentages. The correct description of October is that the HMI fell three points to 29, from 32 in September. Describing it as a percentage decline in builder sentiment would apply percentage arithmetic to a diffusion index, which is not how NAHB reports the measure.

The same applies to the components. Buyer traffic at 18 is an index level, not an 18% decline in traffic. Current sales conditions at 33 are an index level, not a 33% fall in sales.

What the latest October 2026 HMI shows

Composite HMI29-3 points

Present sales33-2 points

Six-month expectations32-5 points

Prospective buyer traffic18-5 points

Source: NAHB, October 19, 2026 release. Changes are in index points, not percentages.

All three components declined. NAHB reported that the composite reading was the lowest since June 2012. (Same source.)

NAHB's explanation of the decline should be kept separate from the measured index. NAHB attributed October's weakness to rising mortgage rates, energy and material costs, and worsening labor and lot conditions. NAHB Chief Economist Robert Dietz said higher mortgage rates, elevated construction costs, and weaker consumer confidence were keeping more buyers on the sidelines. (Same source.) Those are NAHB's interpretations. The HMI itself records how builders rated conditions, not why.

Price cuts and incentives are separate survey findings

NAHB's monthly releases often report additional findings from the same survey, and October's included three:

• 43% of builders reported cutting prices in October, which NAHB said was the highest share during the 2020–2026 period.
• The average price cut was 6%, unchanged for the seventh consecutive month.
• 65% of builders reported using sales incentives, the 19th straight month in which at least 60% of builders did so.

(Source: NAHB, October 19, 2026 release.)

Separate survey findings — not HMI component scores

43%of surveyed builders reported cutting prices

6%average reported price reduction

65%of surveyed builders reported using incentives

These numbers are measured differently from the HMI. The 43% and 65% are shares of surveyed builders reporting a practice. The 6% is the average size of the price reductions reported. None of them is an index level, and none of them is one of the three HMI components. They are not inputs to the formula that produced October's 29.

They also have a specific scope:

• 43% is the share of surveyed builders that cut prices, not the share of new homes sold with a price cut.
• 6% is the average reported cut, not the cut every builder made, and not a decline in new-home prices overall.
• 65% is the share of surveyed builders using incentives, not the share of new homes sold with incentives.

Read that way, the findings add useful detail on how surveyed builders were responding to market conditions, alongside the HMI's summary of how they rated those conditions. They should not be mixed with the index readings of 29, 33, 32, and 18 as if they were the same kind of number.

Builder sentiment vs. housing starts

Sentiment

HMI

Builder sentiment

Measured activity

Housing starts

Construction beginning

New-home sales

Separate actual sales measure

Housing inventory

For-sale stock

FHFA HPI

House-price change

Sentiment and measured market activity are different things.

The HMI and housing starts both concern single-family construction, but they measure different things.

• The HMI records how builders rate market conditions.
• Housing starts measure when construction actually begins on new homes, as estimated by the Census Bureau and HUD.

The two are related. Builders who see weak sales and traffic may be less inclined to start new homes, and NAHB's HMI page presents the index alongside single-family starts. NAHB also cites research associating its survey with housing starts. It points to a 1994 peer-reviewed article finding that the NAHB survey helped predict housing market variables like starts, and says NAHB economists later found the HMI retained essentially the same predictive ability. (Source: NAHB, Housing Market Index page.)

That is a historical statistical association reported by NAHB, not a rule. A given HMI reading does not determine how many homes will be started, and sentiment and starts can diverge for extended periods. Builders may continue projects already planned or financed, or hold back despite improving sentiment.

For how construction activity itself is measured, see how permits, housing starts, and completions measure construction activity.

Builder sentiment vs. new-home sales

Two of the HMI's components ask about sales of new single-family homes, so it is natural to treat the index as a sales measure. It is not.

The HMI asks builders how they rate current sales and how they expect sales to go. It does not count homes sold. Actual sales of newly built homes are measured separately, in the government's new residential sales statistics.

The two can diverge. A builder can rate sales conditions as poor while continuing to sell homes, perhaps more slowly than hoped or with price cuts and incentives. Sentiment can also improve before measured sales change materially. The HMI describes how builders feel about sales. A sales statistic describes how many sales occurred.

The same distinction applies to buyer traffic. The traffic component reflects builders' ratings of prospective-buyer traffic. It is not a measure of signed contracts or completed sales.

Builder sentiment vs. inventory and home prices

Three measures that are often discussed together answer different questions:

Three measures that are often discussed together

Measure	Question it answers
NAHB/Wells Fargo HMI	How do single-family builders rate current sales, expected sales, and buyer traffic?
Housing inventory	How much for-sale stock is available, under a given source's methodology?
FHFA House Price Index	How are house prices changing for repeat transactions on the same properties?

Swipe horizontally to view all columns →

Sources: NAHB, Housing Market Index page; FHFA House Price Index methodology.

The HMI is not inventory. It does not count homes for sale or months of supply. A low HMI does not show that inventory is high, and a high HMI does not show that it is low. See what housing inventory and months of supply actually measure.

The HMI is not a price measure. A low reading does not prove that home prices are falling, and a rising reading does not prove that they are rising. Price cuts reported by surveyed builders, discussed above, are a separate survey finding, not a price index. For price change, see how the FHFA House Price Index measures price change.

How to compare HMI readings correctly

HMI comparisons are only meaningful when the measurement is held constant. Before comparing two readings, check:

• The same index: the composite HMI with the composite, or a component with the same component.
• The same time basis: monthly readings with monthly readings, or moving averages with moving averages.
• Index points vs. percentages: changes in the HMI are expressed in points. Avoid converting them to percentage changes.
• National vs. regional: NAHB reports regional HMI readings as three-month moving averages, while the national headline is a single month.
• Survey sentiment vs. measured activity: an HMI reading should not be compared directly with a count of starts, sales, or homes for sale.

The regional figures show why the time basis matters. In its October release, NAHB reported three-month moving-average regional readings of 38 in the Northeast (down 1 point), 40 in the Midwest (down 3), 30 in the South (down 1), and 26 in the West (down 2). (Source: NAHB, October 19, 2026 release.) Those values average three months of data. They are not single-month October readings, and they should not be set directly against the national monthly reading of 29 as if they shared the same basis. They describe regional builder sentiment. They do not rank regional markets.

Regional HMI — three-month moving averages

Northeast

38

-1 point

Midwest

40

-3 points

South

30

-1 point

West

26

-2 points

Three-month moving averages, not single-month October readings. Not directly comparable to the national monthly composite.

For local housing context, you can explore Metro Analytics market pages. The Metro Analytics methodology explains how the measures on those pages are defined and sourced.

What HMI can—and cannot—tell you

Used carefully, the HMI can help describe:

• Builder sentiment: whether single-family builders, on balance, view market conditions positively or negatively.
• Current sales perceptions: how builders rate present sales of new single-family homes.
• Six-month expectations: how builders expect sales to develop over roughly the next half year.
• Buyer-traffic perceptions: how builders rate the traffic of prospective buyers.
• Changes in sentiment over time: when the same index and time basis are compared.

On its own, the HMI cannot tell you:

• the actual number of homes sold
• the actual number of housing starts or permits
• home prices or price changes
• how much inventory is available
• total housing demand
• mortgage rates
• future home sales or future starts
• investment returns, or whether a market is "good" or "bad"

The HMI is a measure of perception. It is most informative when read alongside measured activity, such as starts, sales, inventory, and prices, rather than as a substitute for them. It is not a forecast, and it is not a recommendation about any market.

The takeaway

The NAHB/Wells Fargo Housing Market Index is a monthly survey-based measure of single-family builder sentiment. It combines three component indices, for present sales, six-month sales expectations, and prospective buyer traffic, using fixed weights to produce a seasonally adjusted headline on a 0 to 100 scale. For each component, a reading of 50 marks the point where positive and negative assessments balance. It does not mean half of builders are optimistic.

October 2026's reading of 29, down three points, is an index level. It is not a 29% decline in sales, prices, construction, or demand, and not the share of builders who are confident. The HMI measures how builders see the market. It is not housing starts, new-home sales, inventory, or a price index, and it should be read alongside those measures rather than in place of them.

To see local housing context for a specific area, explore Metro Analytics Markets, or view the Methodology to see how each measure is built.