The simple version

The government publishes a monthly count of new home sales. A trade association publishes a monthly count of existing home sales. Both are reported as sales, both are seasonally adjusted annual rates, and both get quoted in the same breath.

The obvious difference is the houses: newly built versus previously owned. The difference that actually matters for reading them is timing. One series counts a sale when the contract is signed. The other counts most sales when the deal closes, which is typically weeks later.

This is not an inference. The Census Bureau publishes a short document devoted to the comparison, and it states both definitions and the consequence of the gap between them.

The numbers

• The Census Bureau states its own definition: a sale of the new house occurs with the signing of a sales contract or the acceptance of a deposit, and the house can be at any stage of construction, not yet started, under construction, or already completed (U.S. Census Bureau, New Home Sales vs. Existing Home Sales)
• In the same document, Census states that existing home sales data are provided by the National Association of Realtors and that, according to them, the majority of transactions are reported when the sales contract is closed (U.S. Census Bureau, quoting the National Association of Realtors)
• Census states that most transactions usually involve a mortgage which takes 30 to 60 days to close, so an existing home sale most likely involves a sales contract signed a month or two prior (U.S. Census Bureau)
• Census states that, given the difference in definition, new home sales usually lead existing home sales regarding changes in the residential sales market by a month or two (U.S. Census Bureau)
• Census states that effective with January 2005 the National Association of Realtors created a monthly series to overcome the lagging effect of the existing home sales definition, called Pending Home Sales, based on sales of existing homes where the contract has been signed but the transaction has not been closed (U.S. Census Bureau)
• The National Association of Realtors describes that index in its own words as based on signed real estate contracts for existing single-family homes, condos, and co-ops (National Association of Realtors)
• The association states that Existing-Home Sales for August 2026 will be released on Thursday, September 10, 2026 at 10:00 a.m. Eastern (National Association of Realtors)
• Sales of new single-family houses in July 2026 were at a seasonally adjusted annual rate of 607,000, and the next new residential sales release is scheduled for September 24, 2026 (U.S. Census Bureau and HUD, Monthly New Residential Sales, July 2026, CB26-128, released August 25, 2026)
• For July 2026 the National Association of Realtors reported that existing-home sales edged down 1.7 percent (National Association of Realtors). That figure and the Census figure above are different measurements and are not comparable as levels
• Both series are reported as seasonally adjusted annual rates. Census labels its figure that way in the release, and the association states that seasonally adjusted annualized figures are reported for its monthly and quarterly reports (U.S. Census Bureau; National Association of Realtors)
• An annual rate expresses a month's activity as the pace it would imply over a full year, which is why the figures are far larger than the number of houses sold in any single month (definition)

One transaction, several countable moments

Buying a house is not an event. It is a process with several distinct steps, and more than one of them is a defensible place to say a sale occurred.

A buyer and seller agree and sign a contract. Financing is arranged, inspections happen, an appraisal comes back, and contingencies clear or they do not. Then the transaction closes, money and title change hands, and the house belongs to someone new.

The new home series counts the first of those moments. The existing home series counts the last, for most transactions. Both are documented choices rather than accidents, and neither publisher conceals its method. Census prints its definition in every monthly release and devotes a separate document to explaining how the two series differ.

The consequence is that the two series describe the same market at different points in time. New home sales reflect decisions made recently. Existing home sales largely reflect decisions made weeks earlier that finished in the reference month. When the two disagree, that gap alone can account for it.

Why one leads and one lags

Because contracts precede closings, a series counting contracts responds to changing conditions before a series counting closings does. Census says this directly rather than leaving it to be worked out: given the difference in definition, new home sales usually lead existing home sales regarding changes in the residential sales market by a month or two.

It also supplies the reason. Most transactions involve a mortgage, which Census puts at 30 to 60 days to close, so an existing home sale recorded this month most likely rests on a contract signed a month or two earlier.

That makes a month where the two series move in opposite directions something other than a contradiction. It can describe a market that changed recently, with the closing-based series still completing transactions agreed before the change.

The series built to close the gap

The lag was recognized by the publisher of the existing home series, which built a second series to work around it.

Census records that effective with January 2005 the association created a monthly series to overcome the lagging effect of the existing home sales definition. It is called Pending Home Sales, and it counts existing homes where the contract has been signed but the transaction has not closed. The association describes it in its own words as based on signed real estate contracts for existing single-family homes, condos, and co-ops.

So the full picture is three series rather than two, and reading them together is what separates the question of what buyers decided from the question of what finished. One counts contracts on new construction, one counts contracts on existing homes, and one counts completed sales of existing homes.

A note on revisions, which is the other way these figures move after publication. Census states that the preliminary seasonally adjusted estimate of total sales is revised about 5.0 percent on average, and we have written separately about what a revised base does to a headline percentage. The association revises its seasonal adjustment factors annually, with the February release of revised December data.

The Real Cost lens on which report describes your market

For anyone actually transacting, the timing distinction determines which report is describing the conditions they are living in.

• If you are shopping now, a contract-based series is closer to the conditions you are facing, because it counts decisions made recently
• If you are watching for whether the market has turned, the closing-based series describes deals struck weeks earlier and will register a change later than it happened
• The two series also cover different segments, previously owned homes against newly built ones, so they are not two samples of one population
• Neither national figure describes any local market, and both are estimates subject to revision

None of that is a reason to act on anything. It is the difference between reading a report that describes recent decisions and a report that largely describes decisions made a month or two ago, both published in the same week.

What this means

When housing figures appear in a story, the useful question is not which type of house but which moment got counted. Both publishers document their methods, and the Census Bureau publishes a document comparing them, so the timing explains most apparent contradictions without anyone having to guess.

The broader habit is that any statistic about a process has to pick a moment to count, and that choice determines what the number can see. Two honest measurements of the same activity can disagree because they are watching different steps.

What this is NOT

This is not a prediction of home sales, prices, mortgage rates, or housing activity, and it says nothing about what any upcoming release will show. This is not advice about buying, selling, or timing any housing or financial decision, and it is not advice about any security or fund. This is not a characterization of the housing market as strong or weak. The existing home sales series is published by a trade association and is cited here solely as the source for its own data and its own methods, not for market conditions, forecasts, or affordability. The counting rule for that series is stated here as the Census Bureau states it, which is that the majority of transactions are reported at closing rather than all of them. No share of total sales is claimed, because no government source consulted publishes one. National figures do not describe any local market, and both series are estimates subject to revision. This is not investment or financial advice of any kind.

Sources

• U.S. Census Bureau, New Home Sales vs. Existing Home Sales: https://www.census.gov/construction/pdf/nrs_vs_existing.pdf
• U.S. Census Bureau and U.S. Department of Housing and Urban Development, Monthly New Residential Sales, July 2026 (CB26-128): https://www.census.gov/construction/nrs/pdf/newressales.pdf
• U.S. Census Bureau, New Residential Sales: https://www.census.gov/construction/nrs/index.html
• U.S. Census Bureau, Survey of Construction methodology: https://www.census.gov/construction/soc/methodology.html
• National Association of Realtors, existing home sales: https://www.nar.realtor/research-and-statistics/housing-statistics/existing-home-sales
• National Association of Realtors, existing home sales methodology: https://www.nar.realtor/research-and-statistics/housing-statistics/existing-home-sales/methodology