A plain-English guide to Existing Home Sales as a macro signal.

Tuesday's analysis piece is the live example. NAR Existing Home Sales printed 3.98 million SAAR in August, a 2.0 percent month-on-month drop, with months-supply climbing to 4.9 (source: NAR release, September 22) and median price up 1.6 percent year-on-year. The tape moved almost not at all: DXY 100.60 (+17 pips), 10Y 4.968 (+0.5bp). This reference piece explains what the report actually contains, why the quantity component leads the price component, and how to read a soft housing print inside a hawkish-Fed regime.

What the report is

The National Association of Realtors publishes Existing Home Sales monthly, typically on the third or fourth Tuesday of the following month at 10:00 AM ET. The headline number is a seasonally adjusted annual rate (SAAR) of closed transactions on single-family homes, condos, and co-ops. The report also carries median price, months-supply of inventory, regional breakdown (Northeast, Midwest, South, West), and a first-time buyer share. Coverage is broad: NAR estimates the series covers roughly 90 percent of the US residential resale market.

Existing home sales are distinct from New Home Sales (a Census Bureau series that measures contracts signed on newly-built homes; released later in the month) and Pending Home Sales (a NAR forward indicator based on contracts signed but not closed). Existing sales lag by a month or two because they measure closings; pending sales lead by roughly six weeks. When the two diverge, pending is usually the earlier signal.

Why quantity leads price

Homes are illiquid, price-sticky assets. Sellers anchor to their acquisition price or to comparable recent sales; buyers face financing costs that move with the mortgage-rate curve. When mortgage rates rise or affordability compresses, sellers pull listings before they cut prices, and buyers walk before they overpay. The quantity dimension therefore adjusts first, months-supply builds, and only later (typically two to three quarters lagged) does the median price follow. That is the specific reason Tuesday's 4.9-months-supply read is more informative than the 1.6 percent year-on-year median-price rise: the supply build is the leading signal; the price rise is the trailing story.

The three data points that carry the macro read

1. The SAAR level. The absolute pace tells you where the housing dashboard is versus its cycle range. Sub-4 million SAAR is a five-decade low ex-COVID; 5 to 6 million is a typical late-cycle range; 6+ million was the 2005 peak. Tuesday's 3.98M print is inside the "recession-adjacent" range even without other recession signals.
2. Months-supply of inventory. This is the inventory count divided by the current sales pace. Under 3 months is a seller's market; 4 to 6 is balanced; over 6 is a buyer's market. Tuesday's 4.9 is balanced-tilting-buyer, but the trajectory (climbing every month for six consecutive prints) is the specific signal that the buyer-market is forming.
3. The regional bifurcation. Northeast and Midwest weakness with a firm West is a specific labour-market-plus-mortgage-rate pattern; South weakness reads through to Sunbelt migration reversal. Tuesday's -4/-3.1/-1.6/0 split across Northeast/Midwest/South/West is the second consecutive month of that specific pattern.

What a soft print does not signal (yet)

Existing Home Sales is a demand-side quantity indicator, not an inflation indicator. It does not touch Owner's Equivalent Rent (which is imputed from BLS survey data rather than actual home prices), so a soft print does not undermine the Fed's inflation-first reaction function. That is the reason Tuesday's soft print produced almost no cross-asset move: the FOMC's SEP built around Core PCE staying above 3 percent through 2027, and a housing volume slowdown does not touch that projection.

Where the signal does matter is in the medium-term growth dashboard. A sustained sub-4M pace for four to six months typically translates to a 0.3 to 0.5 percentage-point drag on quarterly real GDP through the residential-investment component. That is not a session-mover; it is a quarterly-anchor mover, and it feeds into the December SEP more than the October FOMC.

The three specific errors to avoid when reading Existing Home Sales

• Reading price as the leading signal. Median price is the trailing signal. Months-supply is the leading signal. When the two diverge, follow supply.
• Confusing existing with new. Existing sales measure resale volume; new sales measure builder inventory turnover. Divergences between the two carry information about the specific channel through which mortgage rates transmit (existing = homeowner behaviour; new = builder incentives). Tuesday's Existing print pairs with the New Home Sales release Thursday; the two together are more informative than either alone.
• Reading a first-time-buyer-share shift as noise. The first-time-buyer share is the specific indicator of new-household formation entering the housing market; a sub-25 percent share for four consecutive months is a leading indicator of demand-side weakness. Watch it as a component even when the headline is unchanged.

Related reading

• Tuesday's analysis piece: the live example this framework describes.
• Week after an FOMC hike: the specific policy backdrop that muted Tuesday's tape reaction.
• CPI components: the shelter series inside CPI that Existing Home Sales does not directly touch.