The Record · Atlanta residential · Study grade
Atlanta residential, May 2026
A mid-month rate spike erased April's affordability relief in two weeks.
The homes that closed in May were signed in April, and the contract pipeline behind them ran about a fifth thinner than a year earlier.
Key figures
Five figures from this month's ledger, each with its source and as-of date.
The weekly 30-year fixed rose from 6.30 percent on April 30 to 6.53 percent on May 28, the highest print since August 2025.
Source: Freddie Mac via FRED, as of June 11, 2026.
Zillow's home value index for the Atlanta metro was $378,748 in May 2026, down 2.14 percent from May 2025.
Source: Zillow Research, as of October 6, 2026.
Atlanta metro for-sale inventory reached 32,227 in May 2026, up 2.45 percent from May 2025 and the highest reading of 2026 to that point.
Source: Zillow Research, as of October 6, 2026.
Zillow's nowcast put Atlanta metro sales at 8,421 in May 2026, up 3.08 percent from May 2025, an estimate rather than a recorded count.
Source: Zillow Research, as of October 6, 2026.
The share of Atlanta metro listings with a price cut was 26.64 percent in May 2026, below 28.51 percent in May 2025.
Source: Zillow Research, as of October 6, 2026.
30-yr fixed, avg, Freddie Mac
6.44%▲
MoM +0.11 pt · YoY -0.38 pt
30-yr fixed, month-end, Freddie Mac
6.53%▲
MoM +0.23 pt · YoY -0.36 pt
15-yr fixed, avg, Freddie Mac
5.79%▲
MoM +0.11 pt · YoY -0.17 pt
10-yr Treasury, US Treasury
4.48%▲
MoM +0.17 pt · YoY +0.06 pt
Metro home value, Zillow
$379K▼
MoM -0.2% · YoY -2.1%
Median sale, Zillow
$381K▲
MoM +2.0% · YoY +1.7%
Active inventory, Zillow
32,227▲
MoM +6.2% · YoY +2.5%
New listings, Zillow
8,994▲
MoM +8.1% · YoY -3.7%
Months of supply, from Zillow
3.8–
MoM 0.0 mo · YoY 0.0 mo
Days to pending, Zillow
33▼
MoM -4 days · YoY +4 days
Price-cut share, Zillow
27%▲
MoM +1 pt · YoY -2 pt
Metro rent (ZORI), Zillow
$1,829▲
MoM +0.5% · YoY +1.5%
At this month's average 30-year rate, computed on standard amortization, each $100 of monthly principal and interest equals about $15,900 of loan, or about $19,900 of purchase price at 20 percent down, at the 6.44 percent May average.
Part One
The market in context
Part One · I · Rates and affordability
Rates and affordability
The buyer who had not locked by May 14 paid for it. The weekly benchmark survey opened May at 6.30 percent for the week ending April 30 and closed at 6.53 percent on May 28, a climb with no dip a buyer could time. The daily quote a borrower actually saw touched 6.75 percent on May 19, a ten-month high, after the May 12 inflation print pushed the 10-year to 4.67 percent. Freddie Mac's monthly average, the vintage carried on the dashboard, was 6.44 percent, still below last May's 6.82.
SOURCE: FREDDIE MAC PMMS VIA FRED MORTGAGE30US
FREDDIE MAC PMMS VIA FRED · UNITED STATES · 1971-2026 · MONTHLY AVERAGE OF WEEKLYConstruction is The Record's own, the monthly mean of Freddie Mac's weekly survey at a 6.44 percent May reading against the 7.68 percent long-run average. Freddie Mac PMMS via FRED MORTGAGE30US.
30-year fixed
30-year fixed (Freddie Mac): today 6.44%; 2005-07 average 6.21%; long-run norm 7.68%; worst in series 18.45% as of Oct 1981. Below the long-run average.
15-year fixed
15-year fixed (Freddie Mac): today 5.79%; 2005-07 average 5.84%; long-run norm 5.28%; worst in series 8.80% (Dec 1994). Above its long-run norm.
10-year Treasury
10-year Treasury (U.S. Treasury): today 4.48%; 2005-07 average 4.57%; long-run norm 5.81%; worst in series 15.32% (Sep 1981). Below its own long-run norm.
Fed funds
Fed funds (Federal Reserve): today 3.63%; 2005-07 average 4.40%; long-run norm 4.59%; worst in series 19.10% (Jun 1981). Still below the long-run average.
For the transaction
Computed on standard 30-year amortization at May's 6.44 percent average, on the $381,400 metro home with 20 percent down, principal and interest is about $1,917 a month, against about $1,979 at the 6.75 percent daily peak of May 19 (computed, not a lender quote).
How it got here
The 10-year rose from 4.39 percent on May 1 to a 4.67 percent peak on May 19, a ten-month high, and the mortgage followed. Freddie Mac's monthly average rose 11 basis points on the month, from 6.33 percent in April.
The other side
The other side is that spreads held the rate under seven. HousingWire put the mortgage-over-Treasury spread at 1.92 percent on May 16, inside a 1.60 to 1.80 percent historical range, so a friendly spread, not the Fed, kept May's quote where it was.
In the corridor
On the metro median home of $381,400 with 20 percent down, principal and interest was about $1,917 a month at May's 6.44 percent average, computed on standard amortization, up about $186 from February's rate and price.
What to say at the closing table
- To a lender's client
Quote the daily figure, not last week's survey. The number in the paper was 6.53 percent and the number you would have locked on May 19 was closer to 6.75. Run the payment at today's rate, not at an April pre-approval that is already stale.
Sources: Freddie Mac PMMS, weekly releases, May 2026 · U.S. Treasury H.15 via FRED, May 2026 · Mortgage Bankers Association weekly survey, May 2026
Figures and method.
The Record's own affordability measure
Income needed for the typical-home payment, May 2026
Source: The Record own computation, from Census ACS income, Freddie Mac via FRED rate, and Zillow home value
| The computation | Value |
|---|---|
| Atlanta median household income | $92,344 (U.S. Census Bureau, American Community Survey, ACS 2024 1-year, table B19013, Atlanta metro) |
| Zillow typical home value (ZHVI) | $378,748 (Zillow, May 2026) |
| Down payment | 10 percent, leaving a loan of $340,873 |
| 30-year fixed, PMMS average | 6.44% (Freddie Mac via FRED, May 2026) |
| Georgia effective property tax | 0.81% (ratio of ACS 2024 medians, Atlanta metro) |
| Principal and interest, monthly | $2,142 |
| Property tax, monthly | $256 |
| Carrying cost, monthly | $2,398 |
| Carrying cost, annual | $28,776 |
| Share of median income | 31.2% |
The Record's own affordability measure is the share of Atlanta median household income needed for principal and interest on a 30-year loan of 90 percent of the Zillow typical home value at the month's Freddie Mac PMMS average, plus Georgia property tax. All inputs are Class A: Census ACS for income, Freddie Mac via FRED for the rate, and Zillow for the home value.
| Own affordability, share of median income | Reading | Against the measure |
|---|---|---|
| Today, May 2026 | 31.2% | the current reading |
| A year earlier, May 2025 | 35.1% | twelve months before |
| The measure began, October 2024 | 34.5% | the first month on file |
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| 30-year fixed (Freddie Mac) | 6.44% | 6.21% | 18.45% as of Oct 1981 | 7.68% | below the long-run average |
| 15-year fixed (Freddie Mac) | 5.79% | 5.84% | 8.80% (Dec 1994) | 5.28% | above its long-run norm |
| 10-year Treasury (U.S. Treasury) | 4.48% | 4.57% | 15.32% (Sep 1981) | 5.81% | below its own long-run norm |
| Fed funds (Federal Reserve) | 3.63% | 4.40% | 19.10% (Jun 1981) | 4.59% | still below the long-run average |
| Figure | Reading | Source |
|---|---|---|
| 30-year, May 28 | 6.53% | Freddie Mac |
| 10-year Treasury, May 19 | 4.67% | U.S. Treasury |
| 15-year fixed, May | 5.79% | Freddie Mac |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 45 findings for this section.
- The 30-year fixed averaged 6.30 percent for the week ending April 30, up from 6.23 percent and against 6.76 percent a year earlier, the last print before May opened. Freddie Mac, April 30, 2026. Implication: a buyer who locked in the last week of April locked the cycle's low for the spring; every May print came in above it.
- Freddie Mac's April 30 release said purchase applications were running "over 20 percent above a year ago" after several weeks of modest rate declines. Freddie Mac, April 30, 2026. Implication: May opened with real buyer demand queued behind a 6.3 percent rate, which is why the rate spike mid-month mattered so much to volume.
- The 30-year fixed averaged 6.37 percent for the week ending May 7, up from 6.30 percent and against 6.76 percent a year earlier. Freddie Mac, May 7, 2026. Implication: the first May print was 39 basis points below last year, so a buyer comparing against a 2025 pre-approval was still ahead.
- The 15-year fixed averaged 5.72 percent for the week ending May 7, up from 5.64 percent and against 5.89 percent a year earlier. Freddie Mac, May 7, 2026. Implication: the 15-year sat 65 basis points under the 30-year, a gap wide enough to matter for an equity-rich move-down buyer.
- Freddie Mac's May 7 release pointed to "a boost in new-home sales, median new-home prices being down to their lowest level since July 2021, and higher inventory than in recent years" as conditions that "could modestly ease affordability pressures through the spring homebuying season." Freddie Mac, May 7, 2026. Implication: the survey's own economist framed early May as a buyer's window, and the window closed within two weeks.
- The 30-year fixed averaged 6.36 percent for the week ending May 14, down one basis point, against 6.81 percent a year earlier. Freddie Mac, May 14, 2026. Implication: the mid-May survey still read 45 basis points under last year the same week the daily trackers had already started climbing.
- The 15-year fixed averaged 5.71 percent for the week ending May 14, down from 5.72 percent, against 5.92 percent a year earlier. Freddie Mac, May 14, 2026. Implication: the 15-year held under 5.75 percent for the last time in the month.
- Freddie Mac's May 14 release said purchase demand "is softening" but "remains above this time last year." Freddie Mac, May 14, 2026. Implication: the demand read turned from "accelerating" on April 30 to "softening" on May 14, two weeks apart.
- The 30-year fixed jumped 15 basis points to 6.51 percent for the week ending May 21, against 6.86 percent a year earlier. Freddie Mac, May 21, 2026. Implication: the largest weekly move of the month landed in the third week, and a buyer who had not locked by May 14 paid for it.
- The 15-year fixed rose 14 basis points to 5.85 percent for the week ending May 21, against 6.01 percent a year earlier. Freddie Mac, May 21, 2026. Implication: the 15-year's annual discount narrowed to 16 basis points, matched only by the May 28 print.
- Freddie Mac's May 21 release advised that "by shopping around for the best mortgage rate and getting multiple quotes, they can potentially save thousands." Freddie Mac, May 21, 2026. Implication: in a week when lender quotes were moving daily, a second quote was worth more than usual.
- The 30-year fixed averaged 6.53 percent for the week ending May 28, up from 6.51 percent, against 6.89 percent a year earlier, the highest weekly print of the month. Freddie Mac, May 28, 2026. Implication: May closed 23 basis points above where it opened, and the survey of record was still rising into June.
- The 15-year fixed averaged 5.87 percent for the week ending May 28, up from 5.85 percent, against 6.03 percent a year earlier. Freddie Mac, May 28, 2026. Implication: the 15-year ended May 23 basis points above the April 30 print of 5.64 percent, the same move as the 30-year.
- Freddie Mac's May 28 release said "pending home sales have increased three months in a row, indicating there's latent demand and homebuyers are ready to jump back into the market if mortgage rates decline." Freddie Mac, May 28, 2026. Implication: the survey's own economist described the demand as rate-contingent, which is the seller's risk heading into summer.
- The 30-year fixed averaged 6.44 percent for May on the monthly average of the weekly survey, up from 6.33 percent in April and against 6.82 percent in May 2025. Freddie Mac via FRED. Implication: the monthly average rose 11 basis points on the month and sat 38 basis points under a year earlier, a narrower annual discount than April's 40.
- The 15-year fixed averaged 5.79 percent for May on the monthly average, up from 5.68 percent in April and against 5.95 percent in May 2025; the series' 2026 low was February's 5.43 percent. Freddie Mac via FRED. Implication: a refinance into a 15-year that pencilled in February at 5.43 percent cost 36 basis points more by May.
- The four weekly 30-year prints of May were 6.37, 6.36, 6.51 and 6.53 percent; the four 15-year prints were 5.72, 5.71, 5.85 and 5.87 percent. Freddie Mac via FRED, May 28, 2026. Implication: the month split cleanly in two, a flat first half and a 17 basis point step up in the second, so a lock date mattered more than the month's average.
- The 10-year Treasury constant-maturity yield averaged 4.48 percent for May, up from 4.32 percent in April and against 4.42 percent in May 2025. US Treasury via FRED. Implication: the benchmark behind mortgage pricing rose 16 basis points on the month, and the mortgage rate rose 11, so spreads absorbed part of the move.
- The 10-year Treasury closed at 4.39 percent on May 1, 4.59 percent on May 15, 4.67 percent on May 19, 4.56 percent on May 22 and 4.45 percent on May 29. US Treasury via FRED. Implication: the 28 basis point run from May 1 to May 19 is the whole story of the month's mortgage spike, and the 22 basis point retreat into May 29 is why June's first survey print fell.
- The 10-year's May 19 close of 4.67 percent was the highest daily close since January 14, 2025, when the series last stood at 4.78 percent. US Treasury via FRED. Implication: a rate quote on May 19 or 20 was priced off a sixteen-month high in the benchmark, which is the single worst day of the month to have locked.
- The 30-year Treasury closed at 5.18 percent on May 19, the highest daily close in the series since July 2007. US Treasury via FRED, May 19, 2026. Implication: the long end of the curve, which jumbo and portfolio lenders watch, hit a nineteen-year high in the same week the daily mortgage rate peaked.
- The 30-year Treasury averaged 5.03 percent for May, up from 4.91 percent in April; the 2-year averaged 4.00 percent, up from 3.80 percent. US Treasury via FRED. Implication: the 2-year rose 20 basis points in a month with no Fed move, so the market repriced the policy path, not just term premium.
- The 10-year minus 2-year spread stood at 0.54 percentage points on May 18 and 19 and narrowed to 0.43 by May 22. Federal Reserve via FRED. Implication: the curve was steepest at the mortgage-rate peak, which is the usual shape when the long end leads the move.
- Mortgage News Daily's 30-year daily index reached 6.75 percent on May 19, the highest since July 2025, up roughly three-quarters of a point since the Iran war began. Mortgage News Daily, May 19, 2026. Implication: the daily rate a borrower was actually quoted on May 19 ran 24 basis points above that week's Freddie Mac survey.
- Mortgage News Daily attributed the May 19 move to bond-market volatility that pushed 10-year yields to their highest in more than a year, and noted that mortgage bonds outperformed Treasuries because of increased purchases by Fannie Mae and Freddie Mac. Mortgage News Daily, May 19, 2026. Implication: GSE buying of mortgage bonds was cushioning the retail rate, a support that depends on policy rather than on the market.
- The 30-year to 10-year mortgage spread was 1.93 percentage points on May 2 against a historical norm of 1.60 to 1.80, and the author calculated that 2023's worst spreads would have put the 30-year at 7.62 percent. HousingWire, May 2, 2026. Implication: roughly a point of a client's rate in May was being decided by spread behavior, not by the Fed or the Treasury market.
- By May 16 the same analyst put the mortgage spread at 1.92 percentage points, down from 1.96, with the 10-year at 4.596 percent, and held his 2026 forecast ranges at 3.80 to 4.60 percent for the 10-year and 5.75 to 6.75 percent for the 30-year. HousingWire, May 16, 2026. Implication: the daily rate's 6.75 percent print three days later touched the top of that forecast band, so a client's "wait for the dip" case depended on the band holding.
- The same May 16 analysis counted weekly pending sales at 78,006 against 73,523 a year earlier, with purchase applications up 4 percent on the week and 7 percent on the year. HousingWire, May 16, 2026. Implication: national demand was still running 6 percent ahead of last year going into the spike, so the May 21 reversal came off a higher base than 2025's.
- MBA's composite application index fell 4.4 percent for the week ending May 1, with the seasonally adjusted purchase index down 4 percent but 5 percent above a year earlier and the refinance index down 5 percent but 29 percent above a year earlier. Mortgage Bankers Association, May 6, 2026. Implication: the first week of May gave back the late-April surge, so the 21 percent annual purchase gain of mid-April had already faded to 5 percent before rates spiked.
- The average purchase loan size on MBA applications reached a record $467,300 for the week ending May 1, up 1.5 percent on the week and 5.2 percent on the year, against an overall average of $410,900. Haver Analytics, May 6, 2026. Implication: the applicants still in the market were higher-balance borrowers, which fits the corridor's price band better than the metro median.
- MBA's composite index rose 1.7 percent for the week ending May 8, with the purchase index up 4 percent on the week and 7 percent on the year and the refinance index down 1 percent but 28 percent above a year earlier. Mortgage Bankers Association, May 13, 2026. Implication: the second week of May was the demand high-water mark of the month; everything after it declined.
- MBA's 30-year conforming contract rate was 6.46 percent for the week ending May 8, with jumbo at 6.48, FHA at 6.16, the 15-year at 5.83 and the 5/1 ARM at 5.70 percent; the refinance share was 40.8 percent and the ARM share 8.8 percent. The MBA survey, as reported by HousingWire, May 13, 2026. Implication: jumbo priced only 2 basis points over conforming, so a corridor buyer above $832,750 was not paying a meaningful jumbo premium in early May.
- MBA's Joel Kan said on May 13 that purchase applications were "7% ahead of last year's pace, with all loan types showing increases in purchase activity, as potential homebuyers shrugged off the current economic and mortgage rate uncertainties and returned to the market." The MBA survey, as reported by HousingWire, May 13, 2026. Implication: the lender read in mid-May was that buyers had adapted to the rate; the next two weeks tested that.
- MBA's composite index fell 2.3 percent for the week ending May 15, the purchase index fell 4 percent on the week but stayed 8 percent above a year earlier, and the refinance share rose to 41.9 percent. Mortgage Bankers Association, May 20, 2026. Implication: applications hit a five-week low in the week the daily rate hit 6.75 percent, the first clear volume response to the spike.
- MBA's 30-year conforming contract rate rose to 6.56 percent for the week ending May 15 from 6.46, with jumbo at 6.58, FHA at 6.24, the 15-year at 5.93 and the 5/1 ARM at 5.76 percent. The MBA survey, as reported by Mortgage News Daily, May 22, 2026. Implication: the ARM sat about 80 basis points under the fixed rate, which is the spread that starts moving borrowers into adjustable structures.
- The ARM share of applications reached 9.6 percent for the week ending May 15, the highest since October 2025. The MBA survey, as reported by Mortgage News Daily, May 22, 2026. Implication: roughly one in ten applicants chose an adjustable loan to manage the payment, a market-wide response rather than an individual risk signal.
- Joel Kan said applications for the week ending May 15 were "down to the lowest level in five weeks as purchase borrowers pulled back across conventional and government loan types." The MBA survey, as reported by Mortgage News Daily, May 22, 2026. Implication: the pullback was across FHA and VA as well as conventional, so it was a rate effect, not a credit-box effect.
- MBA's composite index fell 8.5 percent for the week ending May 22, the refinance index fell 18 percent, the purchase index slipped 0.4 percent on the week while staying 5 percent above a year earlier, and the refinance share dropped to 37.5 percent. Mortgage Bankers Association, May 27, 2026. Implication: refinance volume collapsed when the rate crossed 6.6 percent, while purchase demand held, so the rate spike hurt lenders' pipelines more than it hurt transactions.
- MBA's 30-year conforming contract rate reached 6.65 percent for the week ending May 22, up 30 basis points over five weeks to the highest since August 2025, with jumbo at 6.68, FHA at 6.31, the 15-year at 5.97 and the 5/1 ARM at 5.81 percent; the average purchase loan size was $473,600. The MBA survey, as reported by HousingWire, May 27, 2026. Implication: the lender-reported rate ended May 14 basis points above the May 21 Freddie Mac survey, which is the gap a client sees between the headline and the quote.
- Joel Kan said on May 27 that "the 30-year fixed rate has increased 30 basis points over the past five weeks to its highest level since August 2025" and that "many borrowers understandably backed away from refinancing last week." The MBA survey, as reported by HousingWire, May 27, 2026. Implication: an owner who had been waiting to refinance a 2023 or 2024 loan lost the May window; the next one depended on the June ceasefire holding.
- The household income needed to afford the median Atlanta home with 20 percent down was $102,374 in May, up from $101,708 in April and down 4.4 percent from $107,092 a year earlier. Zillow Research. Implication: the metro's affordability bar was still 4 percent lower than a year earlier even after May's rate climb, because the rate was lower and the value was lower than in May 2025. (Computed: annual change from the two fetched values.)
- The computed principal-and-interest payment on the metro's May median sale price of $381,400 with 20 percent down at the month's 6.44 percent average rate was $1,917, up from a computed $1,857 in April and $1,731 in February. Computed from Zillow Research median sale price. Implication: the payment on the median home rose $186 in three months, and of May's $60 rise about $22 came from the rate and the rest from the higher median price.
- The monthly mortgage payment on the typical U.S. home was $1,861 in May with 20 percent down, up 1.1 percent from April and still 3.1 percent below May 2025. Zillow Research, June 4, 2026. Implication: nationally the payment was still cheaper than a year ago, so the "rates are up" conversation in May was about the month, not the year.
- Redfin's median monthly housing payment reached $2,637 at a 6.51 percent rate for the four weeks ending May 24, the highest in eleven months, though still 0.9 percent below a year earlier. Redfin, May 28, 2026. Implication: the payment a buyer was underwriting at month's end was the highest since June 2025, which is the figure to put in front of a client comparing against a stale pre-approval.
- Realtor.com® Economic Research recorded mortgage rates climbing from 6.30 to 6.53 percent across May and called it a move that erased "much of April's relief." Realtor.com® Economic Research, June 3, 2026. Implication: the full-month rate path was a straight 23 basis point climb, with no dip a buyer could have timed.
Part One · II · Prices and valuations
Prices and valuations
A slow give-back, and the corridor is the firm part. Zillow's metro home value was $378,748 in May, down about $8,300 a year earlier and about 4 percent below the October 2024 peak of $394,714. Case-Shiller's repeat-sales index for Atlanta, the March print released May 26, read minus 0.46 percent on the year, roughly flat. The corridor held firmer: Milton was up 2.35 percent and Alpharetta flat, while exurban Forsyth gave back 2.74 percent.
SOURCE: ZILLOW RESEARCH ZHVI MID-TIER SA
ZILLOW RESEARCH ZHVI · ATLANTA METRO · 2024-2026 · MID-TIER, SEASONALLY ADJUSTEDConstruction is The Record's own, Zillow's monthly ZHVI from the $394,714 October 2024 peak to $378,748 in May at mid-tier, 20 percent-sampled. Zillow Research ZHVI mid-tier SA, pulled October 6, 2026.
FHFA Atlanta MSAD
FHFA Atlanta MSAD (FHFA): today 374.52; 2005-07 average 177.95; long-run norm 146.30; worst in series 38.87 (1976 Q1). Well above every historical norm.
FHFA U.S.
FHFA U.S. (FHFA): today 719.87; 2005-07 average 366.61; long-run norm 276.52; worst in series 60.04 (1975 Q1). At its series high.
Metro ZHVI
Metro ZHVI (Zillow): today $378,748; worst in series $394,714 (Oct 2024). Below the 2024 peak.
For the transaction
On the $611,786 Forsyth home, the 2.74 percent annual value decline is about $17,000; computed on standard amortization at May's 6.44 percent rate with 20 percent down, principal and interest on that home is about $3,075 a month.
How it got here
The metro index has drifted down from its $394,714 October 2024 peak to $378,748 in May, a slow grind of about 4 percent rather than a step, while the March Case-Shiller read Atlanta at minus 0.46 percent on the year.
The other side
The other side is the repeat-sales and conforming indexes read higher. The FHFA all-transactions report showed Atlanta up 42.7 percent over five years and about 1 percent on the year on the master series, and the national Case-Shiller rose 0.7 percent on the year in March, so a seller can honestly cite either direction.
In the corridor
In the corridor, Fulton's typical value was $416,291, Forsyth $611,786 and Milton $942,205 in May; a Cumming-area seller pricing off a 2025 Forsyth comp is pricing into the county's 2.74 percent annual decline.
What to say at the closing table
- To a seller
Price to the market, not to the memory of 2025. The typical metro home is worth about 2 percent less than a year ago and the value index has drifted down since October 2024. Your price has to clear the market, not match last year's comp.
Sources: Zillow Research ZHVI and median sale, pulled October 6, 2026 · FHFA all-transactions house price index, FHFA master file, pulled October 6, 2026 · S&P Cotality Case-Shiller via FRED, May 26, 2026 (text reference only)
Figures and method.
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| FHFA Atlanta MSAD (FHFA) | 374.52 | 177.95 | 38.87 (1976 Q1) | 146.30 | well above every historical norm |
| FHFA U.S. (FHFA) | 719.87 | 366.61 | 60.04 (1975 Q1) | 276.52 | at its series high |
| Metro ZHVI (Zillow) | $378,748 | -- | $394,714 (Oct 2024) | -- | below the 2024 peak |
| Figure | Reading | Source |
|---|---|---|
| Metro ZHVI, May | $378,748 | Zillow |
| FHFA Atlanta, Q2 2026 | 374.52 | FHFA |
| FHFA Atlanta, five-year | +42.7% | FHFA |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 47 findings for this section.
- The Atlanta metro ZHVI (mid-tier, smoothed, seasonally adjusted) stood at $378,748 for May, down 0.24 percent from $379,648 in April and down 2.14 percent from $387,018 in May 2025. Zillow Research. Implication: the typical metro home lost about $8,300 of value over the year, a number a seller who bought in 2024 or 2025 needs in front of them before pricing. (Computed: changes from the fetched values; the May 17 brief's in-period April endpoint of $381,835 has since been revised to $379,648 in the current file.)
- The metro ZHVI's two-year path runs from a high of $394,714 in October 2024 to $377,428 in August 2026, with May's $378,748 only about $1,300 above that eventual low. Zillow Research. Implication: the metro has drifted down about 4 percent from its 2024 peak in a slow grind rather than a step, which is why list prices have been able to stay above values without a visible crash.
- The metro top-tier ZHVI (65th to 95th percentile) stood at $629,690 for May, down 0.21 percent on the month and 1.19 percent on the year. Zillow Research. Implication: the top tier, which is where most corridor transactions sit, was giving ground at roughly half the pace of the metro median. (Computed: changes from the fetched values.)
- The metro single-family ZHVI stood at $385,563 for May, down 0.24 percent on the month and 2.12 percent on the year. Zillow Research. Implication: detached homes tracked the all-homes index almost exactly, so the softness was not a condo story. (Computed: changes from the fetched values.)
- The metro median list price rose to $404,667 in May, up 1.68 percent from $397,967 in April and 2.27 percent from $395,667 a year earlier. Zillow Research. Implication: sellers were asking 2 percent more than a year ago in a market valued 2 percent lower, a 4 point gap that shows up as price cuts and longer marketing. (Computed: changes from the fetched values.)
- The metro median sale price (not seasonally adjusted) rose to $381,400 in May, up 2.01 percent from $373,900 in April and 1.71 percent from $375,000 in May 2025. Zillow Research. Implication: closed prices were up on the year even as values were down, because the mix of what sold in May skewed larger and newer than the housing stock as a whole. (Computed: changes from the fetched values.)
- The seasonally adjusted metro median sale price was $374,774 in May, up 0.18 percent on the month and 1.62 percent on the year. Zillow Research. Implication: strip out the spring seasonal lift and closed prices were close to flat month to month. (Computed: changes from the fetched values.)
- The metro median new-construction sale price was $417,637 in May, down from $424,900 in April and 2.2 percent below $426,990 a year earlier. Zillow Research. Implication: the new-build premium over the resale median narrowed to about $36,000 in May, which is the number a builder's incentive has to beat. (Computed: annual change and premium from the fetched values.)
- Zillow's April market report, published May 6, put Atlanta's typical home value at $379,802, up 0.4 percent on the month and down 2.2 percent on the year, with inventory up 3.6 percent and the sales count down 5.7 percent on the year. Zillow Research, May 6, 2026. Implication: this was the Atlanta read on file during May; it showed values falling on the year while the month-to-month figure (raw, not seasonally adjusted) ticked up with the season.
- Zillow's May market report, published June 4, put Atlanta's typical home value at $381,077, up 0.4 percent on the month and down 2.0 percent on the year, with inventory up 0.2 percent and the sales nowcast down 2.4 percent on the year. Zillow Research, June 4, 2026. Implication: the annual decline in Atlanta values narrowed from 2.2 to 2.0 percent in a month. (Later release, published after the period.)
- Fulton County's ZHVI was $416,291 in May, down 0.22 percent on the month and 2.61 percent from $427,464 a year earlier. Zillow Research. Implication: the county that holds Alpharetta, Johns Creek, Milton, Roswell and Sandy Springs tracked the metro's decline almost exactly. (Computed: changes from the fetched values.)
- Forsyth County's ZHVI was $611,786 in May, down 0.39 percent on the month and 2.74 percent from $629,035 a year earlier, the largest monthly decline among the six north-metro counties. Zillow Research. Implication: the highest-value county in the corridor was also the one losing value fastest in May, roughly $17,000 off the typical home over the year. (Computed: changes from the fetched values.)
- Cherokee County's ZHVI was $472,356 in May, down 0.21 percent on the month and 1.04 percent from $477,319 a year earlier, the most resilient of the six counties. Zillow Research. Implication: a seller in Woodstock or Canton had about a point and a half less annual decline to explain than a seller in Forsyth. (Computed: changes from the fetched values.)
- Cobb County's ZHVI was $421,745 in May, down 2.62 percent on the year; Gwinnett's was $406,174, down 2.96 percent; DeKalb's was $336,289, down 2.99 percent. Zillow Research. Implication: the three big inner counties all sat near a 3 percent annual decline, so the corridor's softness was the metro's norm, not an outlier. (Computed: changes from the fetched values.)
- Alpharetta's city ZHVI was $717,500 in May, down 0.30 percent on the month and 0.47 percent from $720,911 a year earlier. Zillow Research. Implication: Alpharetta values were essentially flat on the year while the county around them fell 2.6 percent, which argues against pricing an Alpharetta listing off county-wide comps. (Computed: changes from the fetched values.)
- Johns Creek's city ZHVI was $696,471 in May, down 0.15 percent on the month and up 0.82 percent from $690,797 a year earlier. Zillow Research. Implication: on Zillow's value index Johns Creek was one of the few north-metro cities still up on the year, which cuts against the "Johns Creek down 6 percent" framing that circulated in May. (Computed: changes from the fetched values.)
- Milton's city ZHVI was $942,205 in May, up 0.05 percent on the month and 2.35 percent from $920,544 a year earlier. Zillow Research. Implication: Milton's typical value rose on the year even as its listing prices and price per square foot came in lower, so a Milton seller's problem in May was positioning, not the underlying value. (Computed: changes from the fetched values.)
- Roswell's city ZHVI was $652,061 in May, down 0.09 percent on the month and up 0.91 percent from $646,177 a year earlier. Zillow Research. Implication: Roswell held value on the year, so the price cuts showing up there in May were about asking prices that ran ahead of the index. (Computed: changes from the fetched values.)
- Sandy Springs' city ZHVI was $687,291 in May, up 0.02 percent on the month and 2.76 percent from $668,844 a year earlier, the strongest annual gain among the five corridor cities. Zillow Research. Implication: the inside-the-perimeter end of the corridor was outperforming the outer end, the reverse of the 2021 pattern. (Computed: changes from the fetched values.)
- Brookhaven's ZHVI was $739,738 and Dunwoody's $671,268 in May, each up 2.33 percent on the year, while Cumming's was $597,822, down 2.65 percent, and Suwanee's $598,894, down 2.77 percent. Zillow Research. Implication: the dividing line in May ran between the close-in north cities, which were up, and the Forsyth and Gwinnett exurbs, which were down. (Computed: changes from the fetched values.)
- The city of Atlanta's ZHVI was $383,307 in May, down 3.32 percent on the year; Marietta's was $469,834, down 1.72 percent; Decatur's was $281,858, down 5.19 percent. Zillow Research. Implication: the city proper was falling faster than the corridor, which matters for a client comparing an intown condo against a north-metro house. (Computed: changes from the fetched values.)
- Alpharetta's median list price on Zillow was $714,983 in May, flat on the month and down 8.45 percent from $781,000 a year earlier; Johns Creek's was $722,833, down 3.79 percent; Milton's was $1,234,150, down 5.24 percent. Zillow Research. Implication: asking prices in the three north Fulton cities were down 4 to 8 percent on the year while their values were flat to up, which says the mix of what came to market shifted toward smaller and older homes. (Computed: changes from the fetched values.)
- Roswell's median list price was $681,600 in May, up 2.5 percent on the month and 0.12 percent below a year earlier; Sandy Springs' was $545,500, up 12.5 percent on the month and 11.0 percent on the year. Zillow Research. Implication: Sandy Springs' asking prices jumped in a single month, so a buyer there in May was looking at a different listing mix than in April. (Computed: changes from the fetched values.)
- Alpharetta's seasonally adjusted median sale price on Zillow was $696,872 in May, down 3.27 percent on the month and 4.69 percent on the year; Johns Creek's was $693,540, up 3.61 percent on the year; Milton's was $1,039,606, up 12.41 percent; Roswell's was $634,163, down 5.70 percent; Sandy Springs' was $593,486, down 11.15 percent. Zillow Research. Implication: city-level sale medians swung 5 to 12 percent on the year in both directions on small monthly samples, so a client should be shown the value index and the sale median together, never one alone. (Computed: changes from the fetched values.)
- Redfin's Atlanta metro median sale price was $405,000 in May, up 1.25 percent on the year, with the median list price at $430,000, up 3.2 percent, and the median price per square foot at $190.10, down 0.3 percent. Redfin Data Center, June 2, 2026. Implication: on Redfin's closed-sale data the metro median was up on the year but price per foot was flat, so the gain was size mix rather than appreciation.
- Redfin's county medians for May were Fulton $479,495 (up 3.1 percent on the year), Forsyth $642,500 (up 2.0 percent), Cherokee $495,750 (up 1.2 percent), Cobb $445,000 (down 0.9 percent), Gwinnett $420,000 (down 1.2 percent) and DeKalb $385,000 (up 8.0 percent). Redfin Data Center. Implication: closed medians in Fulton, Forsyth and Cherokee were up on the year even as the value indexes fell, so a seller who priced to last year's closings was not wrong on the number, only on the days it would take.
- Redfin's price per square foot for May was down 2.2 percent on the year in Fulton, down 4.8 percent in Forsyth, down 3.6 percent in Gwinnett and down 1.1 percent in Cobb, while Cherokee's was up 1.1 percent and DeKalb's up 9.3 percent. Redfin Data Center. Implication: Forsyth's price per foot fell almost 5 percent while its median rose 2 percent, the clearest sign in the county data that bigger houses were selling for less per foot.
- Redfin's city medians for May were Alpharetta $785,000 (down 10.8 percent on the year), Johns Creek $733,840 (down 6.6 percent), Milton $1,154,500 (up 4.5 percent), Roswell $618,000 (down 4.9 percent) and Sandy Springs $730,000 (up 0.7 percent). Redfin Data Center. Implication: on 76 Alpharetta and 82 Johns Creek closings the May medians were well below a year earlier, which is the figure a buyer's agent would carry into a negotiation there.
- Redfin's May price per square foot was $262.90 in Alpharetta (down 4.7 percent on the year), $233.06 in Johns Creek (down 4.6 percent), $250.62 in Milton (down 4.9 percent), $237.28 in Roswell (down 5.1 percent) and $253.03 in Sandy Springs (down 0.6 percent). Redfin Data Center. Implication: price per foot fell about 5 percent across four of the five corridor cities in the same month, the one figure that moved together across the corridor and the best single comp adjustment for May.
- Redfin's May median list price was $755,000 in Alpharetta (down 7.9 percent on the year), $669,900 in Johns Creek (down 16.3 percent), $1,350,000 in Milton (up 17.4 percent), $750,000 in Roswell (up 7.1 percent) and $749,000 in Sandy Springs (up 4.0 percent). Redfin Data Center. Implication: Johns Creek's asking prices fell 16 percent on the year on Redfin's listing feed, so a seller there who priced off 2025 listings was pricing against a market that no longer existed.
- Realtor.com® Economic Research's May report put the Atlanta-Sandy Springs-Roswell median list price at $425,000, up 1.2 percent on the year, with price per square foot up 0.4 percent. Realtor.com® Economic Research, June 3, 2026. Implication: Atlanta asking prices were still up on the year in a month when the national median list price fell 2.4 percent, its steepest annual decline in Realtor.com® Economic Research's data. (Later release, published after the period.)
- Realtor.com® Economic Research's April report, published April 30, put the Atlanta median list price at $422,400, up 2.4 percent on the year, with price per square foot down 0.2 percent. Realtor.com® Economic Research, April 30, 2026. Implication: this was the Atlanta asking-price read on file through May, and it showed asking prices rising on the year even as price per foot had stopped.
- The national median list price was $429,500 in May, down 2.4 percent on the year, the seventh straight annual decline and the largest in Realtor.com® Economic Research's data since 2017; price per square foot fell 2.5 percent and declined in 35 of the top 50 metros. Realtor.com® Economic Research, June 3, 2026. Implication: nationally sellers were cutting asking prices at a record pace while Atlanta sellers were still asking more, which left Atlanta listings comparatively expensive for a relocation buyer. (Later release.)
- The S&P Cotality Case-Shiller index for Atlanta showed prices down 0.46 percent on the year in March, up 0.73 percent on the month before seasonal adjustment and down 0.09 percent after it. S&P Dow Jones Indices, May 26, 2026. Implication: the repeat-sales index, which strips out mix, put Atlanta at roughly flat on the year, between Zillow's minus 2 percent and the MLS medians' plus 1 to 2 percent.
- The national Case-Shiller index rose 0.7 percent on the year in March, the 20-city composite 0.8 percent and the 10-city 1.4 percent; Chicago led at 6.1 percent and Seattle trailed at minus 2.5 percent. S&P Dow Jones Indices, May 26, 2026. Implication: Atlanta's minus 0.46 percent was among the declining cities but a small decline next to Seattle, Denver and Tampa.
- S&P's Nicholas Godec said "more than half of the 20 major U.S. housing markets recorded year-over-year price declines in March, reflecting a broadening and deepening housing slowdown." S&P Dow Jones Indices, May 26, 2026. Implication: Atlanta's softness was no longer an outlier by late May; it was the national pattern.
- FHFA's purchase-only index for the Atlanta-Sandy Springs-Roswell metro was down 1.08 percent on the year in the first quarter of 2026, up 1.40 percent on the quarter and up 39.28 percent over five years. FHFA, May 26, 2026. Implication: a seller who bought five years ago still held close to 40 percent appreciation on the conforming-loan index, which is the number that makes a one-year 1 percent dip tolerable.
- FHFA's index for Georgia rose 0.11 percent on the year in the first quarter, ranking 42nd among the states, and the South Atlantic division rose 0.51 percent on the year and 0.50 percent on the quarter. FHFA, May 26, 2026. Implication: Georgia was flat on the year on the conforming index, so Atlanta's decline was being offset by the rest of the state.
- FHFA's national index rose 0.1 percent in March and 1.7 percent on the year, with the seasonally adjusted purchase-only index at 441.5 and 65 of the 100 largest metros posting annual gains. FHFA, May 26, 2026. Implication: Atlanta was one of the 35 large metros with an annual decline, which a relocation buyer from a rising metro will notice.
- The national existing-home median price was $417,700 in April, up 0.9 percent on the year, the 34th straight annual gain; the single-family median was $422,300, up 1.0 percent. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: the national median was still rising on the year, so Atlanta's flat-to-down read was a local story, not a national one. (The association's own April page now redirects to its latest release; two named outlets reporting the May 11 release were used.)
- The median new-home sale price nationally was $422,500 in April, up 8.0 percent on the month and 2.2 percent on the year, with the average at $508,800. U.S. Census Bureau and HUD, May 28, 2026. Implication: the national new-home median bounced back above the resale median in April after March's post-2021 low, so the builder discount window narrowed.
- The Georgia MLS median sale price for the 12-county metro was $417,000 in April, up 4.3 percent from $399,990 in March, against $410,000 a year earlier. The Georgia MLS release, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: on the MLS closings read on file in May, the metro median was up about 1.7 percent on the year, consistent with the Zillow sale median and inconsistent with the value index. (Computed: annual change from the two fetched values; the article's own sign on that change is inconsistent.)
- The Georgia MLS median sale price for May was $418,995, up 1 percent on the month and 2.1 percent on the year. The Georgia MLS release, as reported by Atlanta Agent Magazine, June 15, 2026. Implication: the metro's May closings came in 2 percent above last May's, so the month's record-pace sales were not bought with price concessions at the median. (Later release.)
- The Atlanta REALTORS® Association's March brief, published May 7, put the 11-county single-family median at $418,000, down 1.6 percent on the year, and the average at $525,500, down 0.5 percent. Atlanta REALTORS® Association, May 7, 2026. Implication: the association's own read in May had the median down on the year, the only MLS-based figure of the month that did.
- The association's April brief, published June 3, put the median at $436,000 and the average at $564,500; its May brief, published July 8, put the May median at $440,000 and the average at $553,500. Atlanta REALTORS® Association, June 3, 2026. Implication: the association's May median landed $22,000 above its March median, a seasonal lift a seller should not read as appreciation. (Later releases.)
- The national ZHVI was $368,720 in May, up 0.6 percent on the month and 0.8 percent on the year. Zillow Research, June 4, 2026. Implication: Atlanta's minus 2.0 percent sat almost 3 points below the national value trend. (Later release.)
- The Atlanta Fed's May Beige Book reported that in the Sixth District "home prices remained flat or declined modestly," with the steepest drops in southwest Florida. Federal Reserve Board, June 3, 2026. Implication: the regional Fed's contacts described Georgia's price picture as flat to modestly down, which matches the index data and not the MLS medians. (Later release covering the period.)
Part One · III · The economy and the cycle
The economy and the cycle
The rate cut the market had priced was gone by mid-May. Headline inflation re-accelerated to 3.8 percent in April on the May 12 release, with energy up 17.9 percent on the year, and the 10-year rose to 4.67 percent by May 19. The FOMC held its target at 3.50 to 3.75 percent on April 29, and the effective funds rate sat at 3.63 percent all month. Consumer sentiment fell to 44.8, while the metro jobless rate, as of the May reading, held at 3.2 percent, a point under the nation.
SOURCE: U.S. TREASURY VIA FRED DGS10
U.S. TREASURY VIA FRED · UNITED STATES · MAY 2026 · DAILY CONSTANT-MATURITY CLOSEConstruction is The Record's own, the 10-year's daily close from 4.39 percent on May 1 to a 4.67 percent peak on May 19 against its 5.81 percent long-run average. U.S. Treasury via FRED DGS10, daily.
Unemployment, U.S.
Unemployment, U.S. (BLS): today 4.3%; 2005-07 average 4.77%; long-run norm 5.65%; worst in series 14.8% (Apr 2020). Below its long-run norm.
Unemployment, Atlanta metro
Unemployment, Atlanta metro (BLS): today 3.2%; 2005-07 average 4.80%; long-run norm 5.13%; worst in series 12.3% (Apr 2020). Near a cycle low.
CPI, U.S. year on year
CPI, U.S. year on year (BLS): today 3.8%; long-run norm 2.0% target. Above target since 2021.
Fed funds
Fed funds (Federal Reserve): today 3.63%; 2005-07 average 4.40%; long-run norm 4.59%; worst in series 19.1% (Jun 1981). Held through May.
For the transaction
At a held 3.63 percent funds rate, a corridor buyer should plan principal and interest near $1,917 a month on the $381,400 metro home at 20 percent down and May's 6.44 percent rate, computed on standard amortization.
How it got here
The oil shock did the work: Brent rose to $116.73 on May 18 and the 10-year peaked at 4.67 percent on May 19, then both unwound into month end, which is why the June 4 survey fell.
The other side
The other side is the soft data understated the economy. The Conference Board's confidence index slipped to 93.1 on May 26 and Michigan sentiment hit 44.8, yet the quarter's GDP still grew 1.6 percent, so the mood broke before the hard data did.
In the corridor
Georgia carried the demand: the state's labor force set a record in April and the metro jobless rate held at 3.2 percent in May, a point under the national 4.3 percent, so the corridor's buyer was cautious, not jobless.
What to say at the closing table
- To a lender's client
If the plan only works at 6 percent, it is not a plan. The inflation print on May 12 priced a rate cut out of the market, and the Fed held. Lock, use the seller's money to buy the rate down, and refinance later if the oil shock keeps unwinding.
Sources: Bureau of Labor Statistics CPI and Employment, May 2026 · Federal Reserve FOMC statement, April 29, 2026 · University of Michigan Surveys of Consumers, May 2026
Figures and method.
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| Unemployment, U.S. (BLS) | 4.3% | 4.77% | 14.8% (Apr 2020) | 5.65% | below its long-run norm |
| Unemployment, Atlanta metro (BLS) | 3.2% | 4.80% | 12.3% (Apr 2020) | 5.13% | near a cycle low |
| CPI, U.S. year on year (BLS) | 3.8% | -- | -- | 2.0% target | above target since 2021 |
| Fed funds (Federal Reserve) | 3.63% | 4.40% | 19.1% (Jun 1981) | 4.59% | held through May |
| Figure | Reading | Source |
|---|---|---|
| CPI inflation, April | 3.8% | BLS |
| UMich sentiment, May | 44.8 | University of Michigan via FRED |
| Atlanta metro jobless rate, May | 3.2% | BLS |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 29 findings for this section.
- The Consumer Price Index rose 0.6 percent in April after 0.9 percent in March and 3.8 percent on the year, with core up 0.4 percent on the month and 2.8 percent on the year. Bureau of Labor Statistics, May 12, 2026. Implication: the May 12 print was the single macro event that set the month's rate path; the 10-year rose 21 basis points over the five trading days that followed.
- The energy index rose 3.8 percent in April, "accounting for over forty percent of the monthly all items increase," with gasoline up 5.4 percent on the month and 28.4 percent on the year and energy up 17.9 percent on the year; food rose 0.5 percent on the month and 3.2 percent on the year. Bureau of Labor Statistics, May 12, 2026. Implication: the inflation that moved rates was an oil shock, so the rate relief a buyer hoped for depended on the Strait of Hormuz, not on the Fed.
- The CPI-U index level was 332.407 in April and 333.979 in May, a 0.47 percent monthly rise, against 320.620 in May 2025. Bureau of Labor Statistics via FRED. Implication: on the current series the May print, released June 10, ran about 4.2 percent above a year earlier, so the inflation pressure on rates did not ease in June's first data. (Computed: changes from the fetched index levels; later data noted as such.)
- Nonfarm payrolls rose 115,000 in April and the unemployment rate held at 4.3 percent, with February revised down 23,000 to minus 156,000 and March revised up 7,000 to 185,000. Bureau of Labor Statistics, May 8, 2026. Implication: the labor read on file in May was soft enough to keep a cut in the conversation, which is why the May 12 inflation print was such a reversal.
- Average hourly earnings rose 6 cents to $37.41 in April, up 0.2 percent on the month and 3.6 percent on the year; the participation rate was 61.8 percent and the employment-population ratio 59.1 percent. Bureau of Labor Statistics, May 8, 2026. Implication: wages grew 3.6 percent against 3.8 percent inflation, so the typical buyer's real income fell in the month the payment rose.
- Health care added 37,000 jobs in April, construction 9,000 and the federal government lost 9,000. Bureau of Labor Statistics, May 8, 2026. Implication: construction was still hiring in April even as single-family starts fell, which argues the builder pullback was a demand choice rather than a capacity constraint.
- Zillow's economists noted that April's report left three-month average payroll growth at 48,000 a month, that 188,000 people left the labor force, and that part-time work for economic reasons jumped 445,000 to 4.9 million. Zillow Research, May 8, 2026. Implication: the household survey underneath the headline was weaker than the payroll number, which is the kind of softening that shows up in buyer confidence before it shows up in unemployment.
- The May jobs report, released June 5, showed payrolls up 172,000 and revised April to 179,000 from 115,000 and March to 214,000 from 185,000, with unemployment unchanged at 4.3 percent and wages up 3.4 percent on the year. Zillow Research, June 5, 2026. Implication: April's soft 115,000 was later revised to a solid 179,000, so the "weakening labor market" case that stood during May was revised away a week after the month ended. (Later release and revision, noted as such.)
- Georgia's unemployment rate was 3.5 percent in April, unchanged on the month and 0.8 points below the national rate, with the labor force at a record 5,461,455 and employment at a record 5,271,138. Georgia Department of Labor, May 21, 2026. Implication: Georgia's labor market was at record employment in the month, so the metro's record sales pace had income behind it.
- Georgia's nonfarm jobs were 4,996,000 in April, up 7,100 on the month and 2,700 on the year, with real estate, rental and leasing adding 2,000 jobs and construction losing 1,100. Georgia Department of Labor, May 21, 2026. Implication: statewide job growth on the year was essentially zero, so the record employment was coming from population, not from a hiring boom.
- Georgia Labor Commissioner's release said "more Georgians are working today than at any point in our state's history, with record highs in both our labor force and employment." Georgia Department of Labor, May 21, 2026. Implication: the state's own framing in May was a full-employment economy, which supports the demand side of the corridor market even at a 6.5 percent rate.
- BLS's state release confirmed Georgia's seasonally adjusted unemployment rate at 3.5 percent in April, not significantly different from a year earlier, with nonfarm payrolls at 4,996,000, essentially unchanged on the month and the year; the national rate was 4.3 percent. Bureau of Labor Statistics, May 22, 2026. Implication: Georgia sat 0.8 points below the national unemployment rate, the cushion that keeps a local downturn from becoming a distress cycle.
- The Atlanta-Sandy Springs-Roswell metro unemployment rate (not seasonally adjusted) was 2.8 percent in April and 3.2 percent in May, against 3.2 percent in May 2025. Bureau of Labor Statistics via FRED. Implication: the metro's May unemployment rate matched a year earlier at 3.2 percent, a full point under the nation.
- Atlanta metro nonfarm employment was 3,124,800 in May, down from 3,129,100 in April and up about 9,900, or 0.3 percent, from 3,114,900 in May 2025. Bureau of Labor Statistics via FRED. Implication: the metro added fewer than 10,000 jobs over the year, so the relocation-driven demand story was running on announcements more than on payrolls. (Computed: annual change from the fetched values.)
- Georgia's seasonally adjusted unemployment rate was 3.4 percent in May and statewide nonfarm employment 4,999,000, up 5,200 on the year. Bureau of Labor Statistics via FRED. Implication: the state ticked down a tenth in May and stayed under 3.5 percent through the rate spike. (Later data for the month itself; computed annual change.)
- Advance retail and food services sales were $757.1 billion in April, up 0.5 percent on the month and 4.9 percent on the year, after a 1.6 percent March gain. U.S. Census Bureau, May 14, 2026. Implication: consumers were still spending through the oil shock in April, which gave the Fed no reason to cut and the bond market a reason to sell.
- Real GDP grew at a 1.6 percent annual rate in the first quarter on the second estimate, revised down 0.4 point from the advance estimate, after 0.5 percent in the fourth quarter; the quarter's PCE price index rose 4.5 percent and the core index 4.4 percent. Bureau of Economic Analysis, May 28, 2026. Implication: growth was revised down and inflation was not, the combination that leaves the Fed stuck and mortgage rates high.
- The PCE price index rose 0.4 percent in April and 3.8 percent on the year, with the core index up 0.2 percent and 3.3 percent; disposable income fell 0.1 percent, spending rose 0.5 percent and the saving rate was 2.6 percent. Bureau of Economic Analysis, May 28, 2026. Implication: households spent more than their income grew and saved 2.6 percent, which is thin cover for a down payment or a closing-cost surprise.
- The Conference Board's Consumer Confidence Index slipped to 93.1 in May from a revised 93.8, with the Present Situation Index down 3.2 points to 121.2, the lowest since February, and the Expectations Index up 1.0 to 74.4; median year-ahead inflation expectations held at 6.2 percent. The Conference Board release, as reported by PNC Economics, May 26, 2026. Implication: consumers rated their present situation worse in May and expected 6 percent inflation, a combination that delays big-ticket decisions like a home purchase. (The Conference Board's own May page now redirects to its latest release; a dated bank research note on the release was used.)
- The same note reported that around two-thirds of consumers said they were reducing spending because of rising prices and 61 percent were buying fewer items. The Conference Board release, as reported by PNC Economics, May 26, 2026. Implication: a buyer pool that is cutting back on groceries is a buyer pool that will push harder on price and concessions.
- The University of Michigan's Index of Consumer Sentiment fell to 44.8 in May from 49.8 in April, the lowest reading in the 2025 and 2026 observations pulled, with year-ahead inflation expectations at 4.8 percent. University of Michigan via FRED. Implication: sentiment hit a record low in the same month the metro closed a record number of homes, which says the May closings were April's decisions.
- Brent crude closed at $103.70 on May 6, rose to $116.73 on May 18, and fell to $92.88 by May 29, a 20 percent drop from the month's high. U.S. Energy Information Administration via FRED. Implication: the oil peak on May 18 and the 10-year peak on May 19 were one day apart, and both unwound into month end, which is the entire mechanism behind June's first rate decline. (Computed: the percent drop from the fetched values.)
- The national average price of regular gasoline was $4.452 a gallon the week of May 4, $4.500 the week of May 11, $4.490 the week of May 18 and $4.475 the week of May 25, against $3.12 to $3.17 in the same weeks of 2025. U.S. Energy Information Administration via FRED. Implication: gasoline held above $4.45 all month, the household-budget line that competed most directly with a higher mortgage payment.
- The Atlanta Fed's May Beige Book said "the Sixth District economy expanded at a modest pace," that employment was flat to slightly lower with firms holding headcount or reducing through attrition, that annual wage increases ran 2 to 3 percent, and that "prices and nonlabor costs rose moderately" on fuel, airfare and shipping. Federal Reserve Board, June 3, 2026. Implication: the regional economy was growing but not hiring, and wages at 2 to 3 percent were trailing inflation near 4, which caps how much more a Georgia buyer can stretch. (Later release covering the period.)
- The Atlanta Regional Commission's 2025 estimates, published August 2025 and still the current county figures through May 2026, put the 11-county region's annual gain at 64,400 residents, with Fulton adding 18,800, Gwinnett 15,200, Cherokee 7,100 and Forsyth 6,700, and Forsyth and Cherokee the fastest-growing at 2.4 percent each. The ARC release, as reported by Urbanize Atlanta, August 13, 2025. Implication: the two counties with the metro's fastest inventory growth in May were also its fastest-growing by population, which is the demand that absorbs that inventory over time. (Standing context, not a May 2026 release; the May 17 brief dated these figures to May 2026.)
- ARC's next estimates, published September 1, 2026, put the following year's gain at 53,690, or 1.0 percent, down from 1.2 percent, with Fulton leading at 1.9 percent and Forsyth second at 1.5 percent, and the region's building permits in 2025 at 24,015, down 18.5 percent. Atlanta Regional Commission, September 1, 2026. Implication: population growth slowed and permits fell by nearly a fifth in the year that included May, so both the demand and the supply pipelines were decelerating together. (Later release.)
- Hisense announced on April 15 that it was moving its U.S. headquarters from Suwanee to Windward Parkway in Alpharetta, with its president calling the move "a defining moment for Hisense USA." Global Atlanta, April 15, 2026. Implication: the relocation that stood as the corridor's newest corporate anchor during May was an intra-metro move from Gwinnett to north Fulton, not new jobs to the region. (April event; the square footage and headcount quoted in the May briefs are not on the fetched page.)
- The 2-year Treasury averaged 4.00 percent in May, up from 3.80 percent in April and 3.54 percent in January, while the 10-year to 2-year spread ended May at 0.47 points. US Treasury via FRED. Implication: the market priced out roughly two quarter-point cuts between January and May, which is the repricing a lender's client felt as a higher quote.
- No Providentia cycle snapshot was recorded for May 2026; no phase is asserted here, and the macro prints above stand in for it. The Record's own convention; no external source. Implication: the cycle read for May rests on the divergence between the metro's record closings and the national contract reversal, not on a composite score.
Part Two
The local evidence
Part Two · IV · Inventory and supply
Inventory and supply
The buyer's leverage is slower absorption, not a seller flood. Zillow's metro for-sale count was 32,227 in May, up 6.23 percent on the month and 2.45 percent on the year, the highest reading of 2026 to that point. New listings rose to 8,994 but came in 3.66 percent below a year earlier, so the build was homes sitting, not sellers rushing in. Redfin's feed disagreed on direction, counting 26,672 active, down 15.2 percent, which a professional should name.
SOURCE: ZILLOW INVENTORY OVER SALES NOWCAST
ZILLOW INVENTORY AND SALES NOWCAST · ATLANTA METRO · 2026 · INVENTORY OVER THE SALES NOWCASTConstruction is The Record's own, metro inventory over the sales nowcast at 3.83 months in May against 3.80 in April and 3.90 in March, a seasonally smoothed monthly rate. A Record computation from Zillow for-sale inventory and the sales nowcast.
Active inventory, metro
Active inventory, metro (Zillow): today 32,227; worst in series 34,140 (Aug 2025). Below the 2025 reading.
New listings, metro
New listings, metro (Zillow): today 8,994; worst in series 9,336 (May 2025). Fewer than a year ago.
Active listings, metro
Active listings, metro (Realtor.com® Economic Research): today 27,670; long-run norm 21,292; worst in series 32,397 (Jul 2016 start). Up 2.6% on the year.
New-home supply, U.S.
New-home supply, U.S. (Census): today 9.4; 2005-07 average 6.42; long-run norm 6.21; worst in series 12.20 (Jan 2009). A buyer's market in new build.
For the transaction
A buyer with about 1,900 more homes to choose from than in April gains room; on the $381,400 metro home at May's 6.44 percent rate, each 1 percent of price conceded is about $23 a month of principal and interest, computed on standard amortization.
How it got here
The metro count climbed from its 30,336 April low to 32,227 in May, while new listings at 8,994 stayed below last May's 9,336, so slower absorption, not new supply, did the building.
The other side
The other side is it is barely a flood. Redfin's May feed showed metro active listings down 15.2 percent on the year, the opposite of Zillow's count, and national new listings on the two feeds were roughly flat, so the market had largely rebuilt its inventory by 2025.
In the corridor
Supply growth was uneven: Alpharetta active listings were up 22.99 percent on the year to 428, Johns Creek up 26.59 percent and Roswell up 28.00 percent, while the city of Atlanta tightened 8.43 percent.
What to say at the closing table
- To a buyer
More choice and fewer rivals is leverage, not a reason to overpay. Metro listings were up about 2 percent on the year in May and new listings came in below last May's. You have more to choose from and the seller has more competition, so negotiate the price down.
Sources: Zillow Research for-sale inventory, pulled October 6, 2026 · Redfin Data Center metro tracker, May 2026 · Census and HUD new residential sales, May 28, 2026
Figures and method.
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| Active inventory, metro (Zillow) | 32,227 | -- | 34,140 (Aug 2025) | -- | below the 2025 reading |
| New listings, metro (Zillow) | 8,994 | -- | 9,336 (May 2025) | -- | fewer than a year ago |
| Active listings, metro (Realtor.com® Economic Research) | 27,670 | -- | 32,397 (Jul 2016 start) | 21,292 | up 2.6% on the year |
| New-home supply, U.S. (Census) | 9.4 | 6.42 | 12.20 (Jan 2009) | 6.21 | a buyer's market in new build |
| Figure | Reading | Source |
|---|---|---|
| Active inventory, May | 32,227 | Zillow |
| New listings, May | 8,994 | Zillow |
| Months of supply, May | 3.83 | Zillow |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 33 findings for this section.
- Atlanta metro active for-sale inventory reached 32,227 in May, up 6.23 percent from 30,336 in April and 2.45 percent from 31,457 in May 2025, the highest reading of 2026 to that point. Zillow Research. Implication: a buyer had about 1,900 more homes to choose from than in April and 770 more than a year earlier, enough to keep negotiating leverage even in the month's record sales pace. (Computed: changes from the fetched values.)
- Metro inventory's two-year path shows May 2026's 32,227 still below the 34,140 peak of August 2025 and above every reading from October 2024 through March 2025. Zillow Research. Implication: the metro was about 6 percent below its 2025 inventory peak in May, so the "inventory is back" story had already happened; May was a seasonal climb, not a new high.
- Metro new listings rose to 8,994 in May, up 8.13 percent from 8,318 in April but 3.66 percent below 9,336 in May 2025. Zillow Research. Implication: fewer owners listed than last May, so the inventory build came from slower absorption of what was already listed, not from a seller rush. (Computed: changes from the fetched values.)
- Months of supply computed from metro inventory against the sales-count nowcast was 3.83 in May, against 3.80 in April and 3.90 in March. Computed from Zillow Research inventory and sales count nowcast. Implication: supply and sales rose together, leaving the balance where it had been since March, under four months, which is still a market where a well-priced home moves.
- Fulton County's active inventory was 6,906 in May, up 6.44 percent on the month but 3.52 percent below a year earlier; DeKalb's was 3,837, down 11.53 percent on the year. Zillow Research. Implication: the two inner counties had less for sale than a year ago while the metro had more, so the inventory growth was happening in the outer ring. (Computed: changes from the fetched values.)
- Forsyth County's active inventory was 1,427 in May, up 15.08 percent on the month and 16.21 percent on the year; Cherokee's was 1,653, up 8.82 percent on the month and 24.10 percent on the year. Zillow Research. Implication: the two fastest-growing north counties also had the fastest inventory growth, so a Forsyth or Cherokee seller faced a quarter more competition than a year earlier. (Computed: changes from the fetched values.)
- Cobb County's active inventory was 3,557 in May, up 6.98 percent on the year, and Gwinnett's was 4,092, up 8.17 percent. Zillow Research. Implication: the big suburban counties were adding supply at a mid-single-digit pace, between the inner counties' declines and the exurbs' double-digit gains. (Computed: changes from the fetched values.)
- Alpharetta's active inventory was 428 in May, up 15.99 percent on the month and 22.99 percent from 348 a year earlier; Johns Creek's was 319, up 26.59 percent on the year; Roswell's was 480, up 28.00 percent; Milton's was 227, up 10.19 percent; Sandy Springs' was 587, up 4.82 percent. Zillow Research. Implication: Roswell, Johns Creek and Alpharetta each had roughly a quarter more homes for sale than a year earlier, the corridor's own version of the outer-ring inventory build. (Computed: changes from the fetched values.)
- Cumming's active inventory was 1,125 in May, up 15.86 percent on the year; Woodstock's was 612, up 26.71 percent; Canton's was 650, up 25.97 percent; Marietta's was 1,421, up 12.33 percent. Zillow Research. Implication: the Forsyth and Cherokee county seats carried the same quarter-higher inventory as the north Fulton cities. (Computed: changes from the fetched values.)
- The city of Atlanta's active inventory was 4,594 in May, down 8.43 percent on the year, and Decatur's was 1,144, down 17.99 percent. Zillow Research. Implication: intown supply was shrinking while suburban supply grew, the opposite of the pandemic pattern. (Computed: changes from the fetched values.)
- New listings in May were 161 in Alpharetta (up 12.59 percent on the year), 116 in Johns Creek (flat), 72 in Milton (down 8.86 percent), 183 in Roswell (up 11.59 percent) and 178 in Sandy Springs (down 2.73 percent). Zillow Research. Implication: Alpharetta and Roswell owners were listing more than a year ago while Milton owners were listing less, which is consistent with Milton's firmer value index. (Computed: changes from the fetched values.)
- Redfin's Atlanta metro feed counted 26,672 active listings in May, down 15.2 percent on the year, with 9,679 new listings, down 13.3 percent, and 3.7 months of supply. Redfin Data Center. Implication: Redfin's listing feed showed shrinking supply in the same month Zillow's showed growth, so the two most-cited public inventory counts for Atlanta disagreed on direction and a professional should name which one is being used.
- Redfin's county feeds for May showed Fulton inventory at 5,597 (down 17.9 percent on the year), Forsyth 1,359 (down 4.8 percent), Cherokee 1,378 (up 1.2 percent), Cobb 2,734 (down 8.3 percent), Gwinnett 3,531 (down 5.4 percent) and DeKalb 2,742 (down 31.6 percent). Redfin Data Center. Implication: even on Redfin's shrinking feed, Cherokee was the one north-metro county with more for sale than a year earlier.
- Redfin's months of supply for May were 4.5 in Fulton, 4.3 in Forsyth, 3.2 in Cherokee, 2.8 in Cobb, 3.8 in Gwinnett and 3.5 in DeKalb. Redfin Data Center. Implication: Fulton and Forsyth carried a month and a half more supply than Cobb, which is where a buyer's leverage was greatest in the corridor.
- Redfin's city feeds for May showed Alpharetta inventory at 331 (up 5.1 percent on the year) with 4.4 months of supply, Johns Creek at 237 (up 23.4 percent) with 2.9 months, Milton at 185 (down 8.0 percent) with 4.0 months, Roswell at 323 (up 13.3 percent) with 2.7 months and Sandy Springs at 420 (down 8.9 percent) with 3.2 months. Redfin Data Center. Implication: Alpharetta and Milton sat above four months of supply while Roswell and Johns Creek sat under three, so the same corridor held both a buyer's and a seller's market in May.
- Redfin's new listings for May were 151 in Alpharetta (up 12.7 percent on the year), 108 in Johns Creek (down 13.6 percent), 77 in Milton (down 13.5 percent), 158 in Roswell (up 1.9 percent) and 197 in Sandy Springs (up 9.4 percent). Redfin Data Center. Implication: Johns Creek and Milton owners held back listings in May on both public feeds, the one corridor pattern Zillow and Redfin agreed on.
- Realtor.com® Economic Research's May report put Atlanta's active listing count up 2.6 percent on the year with new listings down 6.1 percent and median days on market up 3 days. Realtor.com® Economic Research, June 3, 2026. Implication: on the third public feed Atlanta's new listings fell on the year, which with Zillow's minus 3.7 percent makes two of three feeds agreeing that fewer owners listed. (Later release.)
- Realtor.com® Economic Research's April report, on file through May, put Atlanta's active listings up 4.3 percent on the year, new listings down 4.1 percent and days on market up 3 days. Realtor.com® Economic Research, April 30, 2026. Implication: Atlanta's inventory growth slowed from 4.3 to 2.6 percent between the April and May reports, the same deceleration Zillow's feed showed.
- National active listings on Realtor.com® Economic Research were 1,058,693 in May, up 5.6 percent on the month and 2.2 percent on the year, 11.6 percent below the 2017 to 2019 norm, with the South up just 0.3 percent on the year. Realtor.com® Economic Research, June 3, 2026. Implication: the South's inventory growth had effectively stopped by May, so the Sun Belt supply wave of 2024 and 2025 was over. (Later release.)
- National new listings on Realtor.com® Economic Research were 474,976 in May, up 2.1 percent on the year and the highest May since 2022, while the median time on market was 52 days, one day slower than a year earlier. Realtor.com® Economic Research, June 3, 2026. Implication: the national seller came back in May; the Atlanta seller did not. (Later release.)
- Zillow counted 1.36 million homes for sale nationally in May, up 4.6 percent on the month and 1 percent on the year, the 30th straight month of annual growth, with new listings at 422,956, down 4.1 percent on the year. Zillow Research, June 4, 2026. Implication: Zillow's economists said weekly data "suggests it could flatline in the next four weeks," so the inventory tailwind buyers had enjoyed since 2023 was ending. (Later release.)
- Zillow's April report, published May 6, counted 1.3 million homes for sale nationally, up 3.7 percent on the year, with new listings at 426,356, up 2.1 percent on the year and 10.7 percent on the month. Zillow Research, May 6, 2026. Implication: on the read available in May, new listings were growing faster than sales for the first time in 2026, which Zillow called a pause in the spring rebound.
- Housing starts ran at a 1,465,000 annual rate in April, down 2.8 percent from a revised 1,507,000 in March and up 4.6 percent from April 2025; single-family starts fell 9.0 percent to 930,000 while 5-plus-unit starts were 529,000. U.S. Census Bureau and HUD, May 21, 2026. Implication: builders pulled back on single-family groundbreaking in the month resale inventory was climbing, which thins the new-build pipeline competing with corridor resale listings in late 2026.
- Building permits rose 5.8 percent to a 1,442,000 annual rate in April, 0.2 percent below April 2025, while single-family permits fell 2.6 percent to 872,000. U.S. Census Bureau and HUD, May 21, 2026. Implication: the permit rise was multifamily; single-family permits fell for a second month, so the rental pipeline kept growing while the for-sale pipeline shrank.
- In the South, April permits ran at 756,000 (518,000 single-family) and starts at 723,000 (548,000 single-family); national completions were 1,449,000, 903,000 of them single-family. U.S. Census Bureau and HUD, May 21, 2026. Implication: the South accounted for roughly half of national starts and nearly 60 percent of single-family starts, so the national single-family pullback was largely a Southern pullback.
- New single-family home sales ran at a 622,000 annual rate in April, down 6.2 percent on the month and 11.3 percent on the year, with 489,000 new homes for sale, a 9.4-month supply. U.S. Census Bureau and HUD, May 28, 2026. Implication: a nine-month supply of new homes is a buyer's market by any definition, and it is where a corridor buyer should look for incentives.
- New-home sales in the South ran at a 370,000 annual rate in April, down 9.8 percent on the month and 14.7 percent on the year. U.S. Census Bureau and HUD, May 28, 2026. Implication: Southern builders sold 15 percent fewer homes than a year earlier, which is why their incentives were widening into May.
- Zillow's raw count of metro Atlanta new-construction sales was 1,160 in May, up from 1,113 in April and 8.7 percent below 1,271 in May 2025. Zillow Research. Implication: new-build closings in the metro were down about 9 percent on the year while resale closings hit a record, so the spring's buyers chose resale. (Computed: annual change from the fetched values.)
- The Atlanta REALTORS® Association's March brief, published May 7, counted 17,723 active listings across the 11-county area, up 5.1 percent on the year, 9,161 new listings, flat on the year, and 4.0 months of supply. Atlanta REALTORS® Association, May 7, 2026. Implication: the association's supply read on file during May was four months, the same balance as Zillow's computed figure.
- The association's April brief counted 19,224 active listings, 9,683 new listings and 4.4 months of supply; its May brief counted 19,715 active listings, 8,833 new listings and 4.5 months of supply. Atlanta REALTORS® Association, June 3, 2026. Implication: on the association's own count, May's new listings fell 850 below April's and supply crept to four and a half months. (Later releases.)
- Redfin's national count of active listings was 1,493,419 for the four weeks ending May 17, up 1.4 percent on the year, with new listings down 0.5 percent on the year. Redfin, May 21, 2026. Implication: national supply growth had slowed to about 1 percent by mid-May, so the "more choice every month" era for buyers was ending.
- Redfin's new listings rose 0.2 percent on the week for the four weeks ending May 24, the first weekly increase after four declines, and active listings were up 0.8 percent on the year. Redfin, May 28, 2026. Implication: seller participation ticked up at the end of May just as buyer demand fell, which is the setup that widens negotiating room in June.
- The Atlanta Fed's May Beige Book reported that in the Sixth District "existing inventory shrank as delistings rose" and that builders saw "sharp traffic declines and growing buyer hesitancy," with one builder saying it "can't give a house away." Federal Reserve Board, June 3, 2026. Implication: the regional Fed's contacts described sellers pulling listings and builders unable to move spec homes even with incentives, the two supply stories a corridor buyer can use. (Later release covering the period.)
Part Two · V · Buyer and seller activity
Buyer and seller activity
The homes that closed were signed before the rate spike. Zillow's sales nowcast reached 8,421 in May, the most in the October 2024 to August 2026 window, and the typical metro home went pending in 33 days. But the contract pipeline on file ran 21.8 percent below a year earlier on Georgia MLS April signings, and the share of homes selling above list fell about eight points across the five corridor cities, to 19.77 percent in Alpharetta.
SOURCE: ZILLOW RESEARCH SHARE SOLD ABOVE LIST, CITY
ZILLOW RESEARCH · ATLANTA CORRIDOR CITIES · 2025-2026 · SHARE SOLD ABOVE LIST, SMOOTHEDConstruction is The Record's own, Zillow's share of city homes selling above list at 19.77 percent in Alpharetta against 27.19 a year earlier, each city paired at the smoothed rate. Zillow Research share of homes sold above list, city files.
Sales-count nowcast, metro
Sales-count nowcast, metro (Zillow): today 8,421. Most since Oct 2024.
Median days to pending, metro
Median days to pending, metro (Zillow): today 33; worst in series 61 (Jan 2026). Faster than winter.
Mean sale-to-list, metro
Mean sale-to-list, metro (Zillow): today 98.30%; worst in series 97.48% (Jan 2026). Within two points of list.
Median days on market, metro
Median days on market, metro (Realtor.com® Economic Research): today 50; long-run norm 51; worst in series 72 (Jan 2017). Near its long-run norm.
For the transaction
A mean sale at 98.30 percent of a $404,667 list price is about a $6,900 give-back; computed on standard amortization at May's 6.44 percent rate, that give-back buys about $36 a month of principal and interest at 20 percent down.
How it got here
Contracts written before the May 19 spike were already down: Georgia MLS April signings ran 21.8 percent below a year earlier, and the share of corridor homes selling above list fell about eight points, to 19.77 percent in Alpharetta and 29.02 percent in Johns Creek.
The other side
The other side is that what sold still cleared. The mean sale closed at 98.30 percent of list in May, within two points, and corridor days to pending stayed fast: Alpharetta went pending in 22 days in May, against 17 a year earlier.
In the corridor
By May the price-cut share had risen in four of the five cities: Alpharetta 25.66 percent of listings, Roswell 26.92 percent and Johns Creek 22.53 percent, each above a year earlier, while Redfin put the metro share at 32.6 percent.
What to say at the closing table
- To a seller
Price to the April comps now; the buyer's agent has read the same pipeline. May closed the most homes in two years on the nowcast, but those buyers signed in April, and contract signings on file ran about a fifth below last year. Price to sell on day one; the market is not rewarding a test.
Sources: Zillow Research sales count and days to pending, pulled October 6, 2026 · Redfin Data Center metro tracker, May 2026 · Georgia MLS, as reported by Atlanta Agent Magazine, May 8, 2026
Figures and method.
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| Sales-count nowcast, metro (Zillow) | 8,421 | -- | -- | -- | most since Oct 2024 |
| Median days to pending, metro (Zillow) | 33 | -- | 61 (Jan 2026) | -- | faster than winter |
| Mean sale-to-list, metro (Zillow) | 98.30% | -- | 97.48% (Jan 2026) | -- | within two points of list |
| Median days on market, metro (Realtor.com® Economic Research) | 50 | -- | 72 (Jan 2017) | 51 | near its long-run norm |
| Figure | Reading | Source |
|---|---|---|
| Sales nowcast, May | 8,421 | Zillow nowcast |
| Median days on market, May | 50 | Realtor.com® Economic Research via FRED |
| Mean sale-to-list, May | 98.30% | Zillow |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 48 findings for this section.
- Zillow's sales-count nowcast for the Atlanta metro reached 8,421 in May, up 5.42 percent from 7,988 in April and 3.08 percent from 8,169 in May 2025, the highest reading in the series from October 2024 through August 2026. Zillow Research. Implication: the metro closed more homes in May than in any month of the prior two years, so the rate spike did not reach the closing table in May; what closed had gone under contract in April. (Computed: changes from the fetched values; the nowcast is an estimate, not a recorded count.)
- Median days to pending in the metro fell to 33 in May from 37 in April, against 29 in May 2025; the series' fastest reading in the October 2024 to August 2026 window was that 29, and its slowest was 61 in January 2026. Zillow Research. Implication: a correctly priced metro listing went pending in about a month, four days slower than last May and four weeks faster than in January.
- Mean days to pending in the metro was 69 in May, down from 76 in April, against 59 a year earlier. Zillow Research. Implication: the mean ran more than twice the median, so a long tail of stale listings sat behind the fast ones, and ten days longer than last year.
- The metro's mean sale-to-list ratio was 98.30 percent in May, up from 98.08 percent in April and against 98.51 percent in May 2025; the series' 2026 low was January's 97.48 percent. Zillow Research. Implication: the average closing came in 1.7 percent under list, a smaller discount than in winter but a larger one than a year earlier, so a seller should expect to give up about $6,500 on a $380,000 list. (Computed: the dollar figure from the ratio and the median list price.)
- The metro's median sale-to-list ratio was 99.25 percent in May, up from 99.05 percent in April and against 99.67 percent a year earlier. Zillow Research. Implication: the typical home closed within a point of list, so the discounting was concentrated in the stale tail, not spread across every sale.
- The share of metro listings with a price cut rose to 26.64 percent in May from 25.50 percent in April, still below 28.51 percent in May 2025. Zillow Research. Implication: roughly one listing in four had cut its price, fewer than last May, which says sellers were pricing closer to the market up front this year.
- The share of metro homes sold above list was 21.81 percent in May, up from 20.11 percent in April and against 23.47 percent in May 2025. Zillow Research. Implication: about one closing in five still went over asking, so a well-presented listing in the right pocket could still draw competition.
- Zillow's Market Heat Index for the Atlanta metro read 48 in May, down from 50 in April and against 47 in May 2025, on Zillow's 0 to 100 scale. Zillow Research. Implication: Zillow's own composite put Atlanta a shade below its 50 midpoint in May, within a point of where it sat a year earlier.
- Median days to pending in May was 22 in Alpharetta (down from 38 in April, against 17 a year earlier), 21 in Johns Creek (against 17), 27 in Milton (against 19), 19 in Roswell (against 16) and 24 in Sandy Springs (against 21). Zillow Research. Implication: every corridor city went pending faster than the metro's 33 days, and Alpharetta's time to pending nearly halved from April, which is the clearest sign the corridor's spring arrived in May.
- Median days to pending in May was 36 in Cumming (against 25 a year earlier), 28 in Woodstock (against 17), 33 in Canton (against 27), 28 in Marietta (against 18) and 38 in the city of Atlanta (against 36). Zillow Research. Implication: the Forsyth and Cherokee county seats were ten to eleven days slower than a year earlier, the widest slowdown in the north metro.
- The mean sale-to-list ratio in May was 98.61 percent in Alpharetta (against 99.13 a year earlier), 98.55 in Johns Creek (against 99.51), 99.18 in Milton (against 98.43), 99.22 in Roswell (against 99.92) and 98.35 in Sandy Springs (against 99.18). Zillow Research. Implication: Johns Creek sellers gave up about a point more off list than a year earlier while Milton sellers gave up less, the reverse of what the asking-price data implied.
- The share of listings with a price cut in May was 25.66 percent in Alpharetta (against 27.64 a year earlier), 22.53 in Johns Creek (against 18.77), 25.30 in Milton (against 22.77), 26.92 in Roswell (against 23.34) and 24.09 in Sandy Springs (against 22.45). Zillow Research. Implication: four of the five corridor cities had more price cuts than a year earlier, with Roswell and Johns Creek up more than 3 points, so overpricing was being corrected after listing rather than before.
- The share of homes sold above list in May was 19.77 percent in Alpharetta (against 27.19 a year earlier), 29.02 in Johns Creek (against 37.17), 19.71 in Milton (against 27.22), 29.21 in Roswell (against 37.64) and 24.51 in Sandy Springs (against 27.84). Zillow Research. Implication: bidding above asking fell about 8 points in every corridor city on the year, the single clearest measure that the multiple-offer market of 2025 had cooled.
- Dunwoody's mean sale-to-list ratio was 100.74 percent in May with 42.15 percent of homes sold above list, while Cumming's was 98.13 percent with 13.20 percent above list. Zillow Research. Implication: the hottest and coolest pockets of the north metro were forty minutes apart, so a client's experience depended on the submarket, not the metro headline.
- Redfin's Atlanta metro feed counted 7,142 homes sold in May, down 6.6 percent on the year, with 8,020 pending sales, down 8.0 percent, a median of 43 days on market, two days slower than a year earlier. Redfin Data Center. Implication: Redfin's MLS-based closings were down on the year in the same month Zillow's nowcast hit a record, so the "busiest month on record" claim rests on one estimate and should be stated that way.
- Redfin's Atlanta metro average sale-to-list ratio was 98.41 percent in May, with 22.1 percent of homes sold above list (down 1.3 points on the year), 32.6 percent of listings with a price drop (up 2.1 points) and 24.6 percent of listings off market within two weeks. Redfin Data Center. Implication: one listing in three cut its price on Redfin's count, but one in four was gone inside two weeks, which is the two-speed market a seller has to pick a lane in.
- Redfin's county feeds for May counted 1,247 homes sold in Fulton (down 11.4 percent on the year), 316 in Forsyth (down 11.7 percent), 426 in Cherokee (up 5.4 percent), 963 in Cobb (up 1.2 percent), 932 in Gwinnett (down 12.6 percent) and 783 in DeKalb (down 7.0 percent). Redfin Data Center. Implication: Fulton and Forsyth closings fell more than a tenth on the year while Cherokee's rose, so the county with the most resilient values was also the one still gaining volume.
- Redfin's pending sales in May were down 8.0 percent on the year in Fulton and up 2.6 percent in Forsyth and 7.3 percent in Cherokee; median days on market were 42 in Fulton (up 1), 38 in Forsyth (up 3) and 34 in Cherokee (up 4). Redfin Data Center. Implication: the two outer counties were still signing more contracts than a year earlier in May, which argues that the metro's contract slump was a Fulton, Gwinnett and DeKalb story.
- Redfin's May price-drop shares were 31.1 percent in Fulton, 35.3 percent in Forsyth, 33.9 percent in Cherokee and 32.6 percent in Cobb, with the share sold above list at 24.4 percent in Fulton, 17.4 percent in Forsyth and 20.0 percent in Cherokee. Redfin Data Center. Implication: Forsyth had the most price drops and the fewest over-asking sales of the north counties, consistent with its 4.8 percent drop in price per foot.
- Redfin's city feeds for May counted 76 homes sold in Alpharetta (down 10.6 percent on the year), 82 in Johns Creek (down 21.9 percent), 46 in Milton (down 20.7 percent), 120 in Roswell (down 5.5 percent) and 131 in Sandy Springs (up 4.8 percent). Redfin Data Center. Implication: Johns Creek and Milton closed a fifth fewer homes than a year earlier, so the corridor's record-pace headline did not apply to its two highest-priced cities.
- Redfin's May pending sales were 102 in Alpharetta (up 1.0 percent on the year), 108 in Johns Creek (down 10 percent), 68 in Milton (flat), 131 in Roswell (down 3.7 percent) and 148 in Sandy Springs (up 12.1 percent). Redfin Data Center. Implication: Sandy Springs was the one corridor city signing meaningfully more contracts than a year earlier, which matches its value index lead.
- Redfin's median days on market in May was 24 in Alpharetta (four faster than a year earlier), 27 in Johns Creek (two faster), 33 in Milton (seven slower), 28 in Roswell (one slower) and 33 in Sandy Springs (two faster). Redfin Data Center. Implication: Milton was the one corridor city where homes sat longer than last year, a week longer, so a Milton seller's pricing margin for error was the thinnest.
- Redfin's May price-drop shares were 29.3 percent in Alpharetta, 37.6 percent in Johns Creek, 27.6 percent in Milton, 35.0 percent in Roswell and 34.5 percent in Sandy Springs, with sale-to-list averages between 98.7 and 99.2 percent. Redfin Data Center. Implication: more than a third of Johns Creek, Roswell and Sandy Springs listings cut their price in May even though closings came within a point and a half of list, so the cut was the path to the close, not a sign of distress.
- Redfin's city of Atlanta feed counted 602 homes sold in May, down 8.6 percent on the year, 672 pending sales, down 5.8 percent, a median of 48 days on market and 5.3 months of supply; Marietta's median days on market was 51, eleven days slower than a year earlier. Redfin Data Center. Implication: intown Atlanta carried five months of supply and seven weeks to sell, a slower market than any corridor city.
- Georgia MLS counted 5,114 closed sales across its 12-county metro area in April, up 1.1 percent on the month and 0.6 percent below April 2025's 5,146. The Georgia MLS release, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: the MLS closings read on file in May had the metro flat on the year, not the 5 percent decline some May 5 coverage reported.
- Georgia MLS counted 4,810 homes going under contract in April, down 1.8 percent on the month and 21.8 percent below April 2025's 6,147. The Georgia MLS release, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: the metro's contract pipeline was more than a fifth thinner than a year earlier going into May, the number that warned June's closings would not hold May's pace. (The source is Georgia MLS; several May briefs attributed the figure to FMLS.)
- Georgia MLS's John Ryan said on May 5 that "the data shows a market in transition" with "a clear shift away from the urgency that defined recent years, as buyers are taking more time, weighing their options and approaching decisions with greater caution," and on May 8 that "as inventory builds, the market is giving buyers more leverage." As reported by Atlanta News First, May 5, 2026. Implication: the MLS's own framing in May was buyer leverage, which is the posture a listing agent should have assumed in pricing conversations.
- Georgia MLS counted 5,590 closed sales in May, up 7.6 percent on the month and 0.6 percent on the year, and 4,512 homes going under contract, down 6.2 percent on the month and 27.9 percent below May 2025's 6,259. The Georgia MLS release, as reported by Atlanta Agent Magazine, June 15, 2026. Implication: May's closings were flat on the year while contract signings fell 28 percent, the widest gap between the two in the series, which is the clearest single warning about the summer pipeline. (Later release.)
- The Atlanta REALTORS® Association's March brief counted 4,670 single-family sales across 11 counties, up 4.0 percent on the year and 29.2 percent on the month. Atlanta REALTORS® Association, May 7, 2026. Implication: the association's read on file in May had closings up on the year, the only metro closings figure of the spring that did.
- The association's April brief counted 4,892 residential sales with 19 average days on market; its May brief counted 5,323 sales with 18 average days on market and 22 cumulative days. Atlanta REALTORS® Association, June 3, 2026. Implication: on the association's measure a May listing averaged under three weeks to contract, the fastest reading of the spring. (Later releases.)
- National existing-home sales ran at a 4.02 million annual rate in April, up 0.2 percent on the month and flat on the year, with 1.47 million homes for sale, up 5.8 percent on the month, and 4.4 months of supply. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: the national resale market was running at the same pace as a year earlier, so Atlanta's record May was ahead of the country.
- The median existing home nationally sold in 32 days in April, down from 41 in March and against 29 a year earlier; first-time buyers were 33 percent of sales, cash buyers 25 percent, investors and second-home buyers 16 percent and distressed sales 2 percent. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: one buyer in four nationally paid cash, which is the competition a financed corridor buyer faced on a well-priced home.
- Existing-home sales in the South ran at a 1.87 million annual rate in April, up 0.5 percent on the month and 2.7 percent on the year, 46.5 percent of national sales. The National Association of REALTORS® release, as reported by Haver Analytics, May 11, 2026. Implication: the South was the only region with annual sales growth, so the Sun Belt was carrying the national number.
- NAR's Lawrence Yun said of April that "mortgage rates are lower from a year ago, and average income growth is outpacing home price gains." The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: the affordability case was being made on the year-over-year comparison, which the mid-May rate spike then eroded.
- The Pending Home Sales Index rose 1.4 percent in April and 3.2 percent on the year; the South fell 0.7 percent on the month but was up 4.7 percent on the year, the largest annual gain of any region, while the Northeast rose 6.6 percent on the month. National Association of REALTORS® (archived), May 19, 2026. Implication: contract activity in the South was already slipping on the month before the rate spike, so the May decline had a running start.
- Yun said on May 19 that "buyers are coming out with cautious optimism despite increasing economic uncertainty and a slight rise in mortgage rates" and that "unless supply meaningfully increases, home price growth could outpace wage growth and further erode the homeownership rate." National Association of REALTORS® (archived), May 19, 2026. Implication: the association's chief economist framed supply, not rates, as the binding constraint, a point that does not fit Atlanta's four months of supply.
- Redfin's four-week pending sales were up 7.7 percent on the year for the period ending May 3, the highest level since September 2022, with the median sale price at $394,803, up 1.9 percent, and 26.4 percent of homes sold above list. Redfin, May 7, 2026. Implication: national demand peaked in the first week of May on the back of April's rate dip.
- Redfin's four-week pending sales were up 9.6 percent on the year for the period ending May 10 at 346,104, with the median sale price at $397,740, up 2.2 percent, 42 median days on market and 26.6 percent sold above list. Redfin, May 14, 2026. Implication: the annual gain widened to nearly 10 percent in the second week of May, the strongest national demand read of 2026.
- Redfin's four-week pending sales fell 1.1 percent on the week for the period ending May 17, the first decline since early April, while still 7.8 percent above a year earlier; the median sale price was $398,653, up 2.2 percent, median days on market 41, three slower than a year earlier, and 27.1 percent sold above list. Redfin, May 21, 2026. Implication: the demand reversal landed within days of the daily rate's 6.75 percent print, which makes the May 19 spike the pivot of the spring.
- In the same May 21 report Redfin listed Atlanta among the metros with annual pending-sales declines, at minus 1.1 percent. Redfin, May 21, 2026. Implication: Atlanta's contracts were down on the year in a week the nation's were up nearly 8 percent, the national-to-local gap in one line.
- Redfin's four-week pending sales fell 1.5 percent on the week for the period ending May 24, the second straight decline, with the annual gain narrowing to 4.7 percent, median days on market at 40 and 27.5 percent sold above list; Atlanta's pending sales were down 3.1 percent on the year. Redfin, May 28, 2026. Implication: Atlanta's annual contract decline tripled in a week, from 1.1 to 3.1 percent, as the rate spike reached the local pipeline.
- Redfin wrote on May 28 that "house hunters are backing off mainly because mortgage rates are rising, with the daily average hitting a 10-month high of 6.75%," and on May 21 that "higher mortgage rates are scaring off some buyers, but that's opening the door for others." Redfin, May 28, 2026. Implication: the publisher's own read was that the buyers who stayed in gained room to negotiate, which is the buyer-side talking point for late May.
- Realtor.com® Economic Research's count of listings in pending status was up 4.3 percent on the year in May, the sixth straight annual gain, with the flow of new contract signings up 2.6 percent; the share of listings with a price cut was 17.5 percent, down 1.6 points. Realtor.com® Economic Research, June 3, 2026. Implication: Realtor.com® Economic Research's economists described sellers as "pricing to sell rather than pricing to test the market," which is the posture that kept national contracts rising. (Later release.)
- Atlanta's price-reduced share on Realtor.com® Economic Research was 20.4 percent in May, down 2.9 points on the year, after 19.5 percent in April, down 1.4 points. Realtor.com® Economic Research, June 3, 2026. Implication: on Realtor.com® Economic Research's feed Atlanta sellers cut prices less often than a year earlier, the same direction as Zillow's feed and the opposite of Redfin's.
- Realtor.com® Economic Research reported that contract cancellations in 2026 "have remained lower than in the past several years" through May. Realtor.com® Economic Research, June 3, 2026. Implication: the rate spike showed up in sentiment and in rates but not yet in deals falling apart, which is the reassurance a nervous seller under contract needed in late May. (Later release.)
- Nationally, Zillow's typical listing went pending in 18 days in May, one day slower than a year earlier, with 341,929 homes sold on the nowcast, down 2.9 percent on the year, and 23.9 percent of listings carrying a price cut. Zillow Research, June 4, 2026. Implication: the metro's 33 days to pending was nearly double the national 18, a gap that has held through the spring. (Later release.)
- Zillow's April report, published May 6, counted 323,631 national sales on the nowcast, down 0.4 percent on the year and up 7.9 percent on the month, with newly pending listings up 7.1 percent on the year and 23.5 percent of listings with a price cut. Zillow Research, May 6, 2026. Implication: the April read on file in May showed demand recovering but not yet above 2025, which was the baseline the rate spike then hit.
- The Atlanta Fed's May Beige Book reported that in the Sixth District "home sales improved slightly, primarily driven by aggressive price reductions and incentives." Federal Reserve Board, June 3, 2026. Implication: the regional Fed's contacts credited the spring's sales to seller concessions, not to buyer strength. (Later release covering the period.)
Part Three
The carrying cost and credit
Part Three · VI · Investors, rental, and the carrying cost
Investors, rental, and the carrying cost
For the owner who already holds, the yield quietly improved. The metro Zillow rent index was $1,829 in May, up about 1.5 percent on the year, while values fell about 2 percent; single-family rent at $2,278 outran apartment rent five to one. Owning the median home cost $1,917 a month on principal and interest, about $88 more than the $1,829 rent. Georgia still led the country for investor share, but 59.1 percent of Atlanta rental listings offered a concession.
SOURCE: ZILLOW ZORI; RECORD P&I MODEL
ZILLOW RENT AND COMPUTED P&I · ATLANTA METRO · MAY 2026 · 20 PERCENT DOWN AT 6.44 PERCENTConstruction is The Record's own, Zillow's metro rent of $1,829 against a $1,917 owner payment at 20 percent down and the 6.44 percent May average, principal and interest only. Zillow ZORI against a Record payment model, principal and interest only.
Metro rent, ZORI
Metro rent, ZORI (Zillow): today $1,829. Up about 1.5% on the year.
Debt-service share of income
Debt-service share of income (Federal Reserve): today 5.83%; 2005-07 average 8.21%; long-run norm 6.63%; worst in series 8.95% (Oct 2007). Far below the bubble.
Homeownership rate, U.S.
Homeownership rate, U.S. (Census): today 65.0%; 2005-07 average 68.60%; long-run norm 65.28% as of 1965; worst in series 69.2% (Apr 2004). Near its long-run norm.
Single-family rent, metro
Single-family rent, metro (Zillow): today $2,278. House rent outran apartment rent.
For the transaction
On a $400,000 corridor rental, computed on standard amortization at May's 6.44 percent rate with a quarter down, principal and interest is about $1,886 a month against the $1,829 metro rent, so a buy-to-rent at this rate runs negative before taxes and upkeep.
How it got here
Rents rose from about $1,803 a year earlier to $1,829 in May while the value index fell about 2 percent on the year, the whole of the rent-it-out case, though single-family rent at $2,278 carried it, not apartments.
The other side
The other side is that apartments are soft. Rental Beast put Atlanta's concession rate at 61.2 percent in the first quarter of 2026 and Zillow at 59.1 percent in its April report, so the effective rent a landlord collected sat below the headline.
In the corridor
A sub-$400,000 rental in Gwinnett or Fulton competes with a small landlord: county rents were up 2.36 percent in Fulton and 2.57 percent in DeKalb on the year while corridor city rents were flat to the dollar.
What to say at the closing table
- To an investor
If you already own, the math moved your way; if you are buying, it did not. Atlanta rents rose about 2 percent while values slipped about 2, so a held rental's yield improved on its own. The competition below $400,000 is a small landlord with cash, and three listings in five offer a concession.
Sources: Zillow Research ZORI and rental report, May 2026 · Federal Reserve MDSP via FRED, May 2026 · Census Housing Vacancies and Homeownership, 2026
Figures and method.
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| Metro rent, ZORI (Zillow) | $1,829 | -- | -- | -- | up about 1.5% on the year |
| Debt-service share of income (Federal Reserve) | 5.83% | 8.21% | 8.95% (Oct 2007) | 6.63% | far below the bubble |
| Homeownership rate, U.S. (Census) | 65.0% | 68.60% | 69.2% (Apr 2004) | 65.28% as of 1965 | near its long-run norm |
| Single-family rent, metro (Zillow) | $2,278 | -- | -- | -- | house rent outran apartment rent |
| Figure | Reading | Source |
|---|---|---|
| Metro rent (ZORI), May | $1,829 | Zillow |
| Atlanta rental concessions | 59.1% | Zillow |
| Figure | Reading | Source |
|---|---|---|
| Own the median home | $1,917 | On P&I alone, about $88 more a month than renting it (computed). |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 22 findings for this section.
- The Zillow Observed Rent Index for the Atlanta metro (smoothed, not seasonally adjusted) was $1,828.90 in May, up 0.52 percent from $1,819.52 in April and 1.46 percent from $1,802.50 in May 2025. Zillow Research. Implication: metro rent rose about $26 a month over the year, a pace well under inflation, so a landlord's real rent was still falling. (Computed: changes from the fetched values.)
- The same rent series ran from a low of $1,764.56 in December 2024 to $1,852.99 in August 2026, and its largest monthly gain in that window was February 2025's 0.66 percent, so May's 0.52 percent was the third-largest, not the largest. Zillow Research. Implication: May's rent increase was a normal spring step, not a regime change, which matters for an investor underwriting rent growth. (Computed: monthly changes from the fetched values.)
- The seasonally adjusted metro ZORI was $1,833.07 in May, up 0.38 percent on the month and 1.45 percent on the year. Zillow Research. Implication: strip out the season and metro rent growth was running at about a 1.5 percent annual pace, the same number on both series. (Computed: changes from the fetched values.)
- The metro single-family ZORI was $2,278.30 in May, up 0.42 percent on the month and 2.30 percent on the year, while the multifamily ZORI was $1,649.03, up 0.24 percent on the month and 0.47 percent on the year. Zillow Research. Implication: house rents were growing five times faster than apartment rents in the metro, so the apartment construction wave was holding down the multifamily number and not the single-family one. (Computed: changes from the fetched values.)
- County rent indexes for May were Fulton $1,911.66 (up 2.36 percent on the year), DeKalb $1,747.68 (up 2.57 percent), Cherokee $2,042.94 (up 0.97 percent), Cobb $1,743.69 (up 0.84 percent), Forsyth $2,234.65 (up 0.65 percent) and Gwinnett $1,816.73 (down 0.68 percent). Zillow Research. Implication: the inner counties' rents were growing while the outer ones were flat, so an investor in Forsyth or Gwinnett was underwriting a rent that had not moved in a year. (Computed: changes from the fetched values.)
- City rent indexes for May were Johns Creek $2,539.94 (flat on the year), Milton $2,228.28 (up 0.25 percent), Alpharetta $2,131.52 (down 0.07 percent), Roswell $1,706.02 (up 0.11 percent) and Sandy Springs $1,691.12 (down 0.31 percent). Zillow Research. Implication: corridor rents were flat to the dollar on the year in all five cities, so a corridor rental's yield was being set entirely by purchase price. (Computed: changes from the fetched values.)
- The city of Atlanta's rent index was $1,915.42 in May, up 2.81 percent on the year, and Woodstock's was $2,105.14, up 1.75 percent, while Cumming's was $2,226.40, down 0.21 percent. Zillow Research. Implication: intown rent was growing fastest in the metro, which is where a yield-focused investor's rent growth case was strongest. (Computed: changes from the fetched values.)
- Zillow's April rental report, published May 27, found 59.1 percent of Atlanta rental listings offering a concession, up 4.9 points on the year, against 39.8 percent nationally, with the typical Atlanta rent at $1,825 and the income needed to afford it at $72,995. Zillow Research, May 27, 2026. Implication: three Atlanta rental listings in five were offering free rent or waived fees, so the effective rent a landlord collected was meaningfully below the headline.
- The same report put the national rental vacancy rate at 7.3 percent, up from 5.6 percent in 2021, and named Denver (68.3 percent), Charlotte (66.6 percent) and Dallas (64.2 percent) as the metros with the most concessions. Zillow Research, May 27, 2026. Implication: Atlanta's 59 percent concession share put it among the Sun Belt construction-boom metros, which is the comparison set for an investor's rent assumptions.
- Zillow's May rental report, published June 18, put the typical Atlanta rent at $1,840, up 1.5 percent on the year, with 85.0 percent of listings affordable to a median-income household, 3.2 percent listed under $1,000 and 58.8 percent offering a concession. Zillow Research, June 18, 2026. Implication: Atlanta was one of the most affordable large rental markets in the country in May, which weakens the "rents will force renters to buy" argument. (Later release.)
- Nationally 74 percent of rental listings were affordable to a median-income household in May, the highest for the month in Zillow's data, with the typical rent at $1,951, up 2 percent on the year, and 39.6 percent of listings offering a concession. Zillow Research, June 18, 2026. Implication: the national renter had more room than at any point since 2021, so the buy-versus-rent math was not being forced by rent. (Later release.)
- Zillow's April and May market reports put Atlanta's typical rent at $1,825 (up 0.3 percent on the month, 1.2 percent on the year) and then $1,840 (up 0.6 percent on the month, 1.5 percent on the year). Zillow Research, May 6, 2026. Implication: rent growth in the metro accelerated slightly through the spring, from 1.2 to 1.5 percent, the opposite direction from values.
- Rental Beast's first-quarter Atlanta report put the concession rate at 61.2 percent, up 15.5 percent from the fourth quarter, against 41.8 percent nationally, with single-family rents near $1,600, down 3 percent, and one-bedroom rents at $1,464, up 4.9 percent. Rental Beast. Implication: two independent trackers put Atlanta's concession share near 60 percent in the spring, so an investor's pro forma should carry at least a month of free rent per lease.
- The computed principal-and-interest payment on the metro median home, $1,917 at May's rate and price, exceeded the metro's $1,829 rent by about $88, wider than April's roughly $38 gap. Computed from Zillow Research median sale price and ZORI. Implication: on principal and interest alone renting the median home was cheaper than owning it for a second straight month, and adding taxes, insurance and HOA widens that further.
- Individual investors and second-home buyers were 16 percent of national existing-home purchases in April, down from 18 percent in March and against 15 percent a year earlier; cash buyers were 25 percent. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: small investors were still one buyer in six nationally, so the institutional retreat did not take the investor bid out of the market.
- The American Economic Liberties Project's April 22 report put institutional ownership at about 72,000 single-family rentals in metro Atlanta, nearly 30 percent of the metro's single-family rental market, more than double second-place Phoenix, with over 3,000 build-to-rent units delivered in 2024. American Economic Liberties Project, April 22, 2026. Implication: no metro has more to gain in owner-occupant supply from an institutional wind-down than Atlanta, and none has more rental supply competing with a small landlord.
- The same report found that institutional investors "may have been responsible for as much as 25% of the decline in homeownership in Atlanta after the Great Financial Crisis" and that nationally they "hold under 5% of national SFR market share." American Economic Liberties Project, April 22, 2026. Implication: Atlanta's institutional share runs about six times the national figure, which is why federal and state investor rules land harder here.
- Executive Order 14376 of January 20, 2026 directed HUD, FHFA, VA, USDA and GSA to stop "providing for, approving, insuring, guaranteeing, securitizing, or facilitating the acquisition by a large institutional investor of a single-family home," with an exception for build-to-rent communities "planned, permitted, financed, and constructed as rental communities." The White House, January 20, 2026. Implication: the order that stood through May cut off federal financing channels for institutional resale purchases but left Atlanta's build-to-rent pipeline untouched.
- The Census Bureau's quarterly vacancy release restated the first-quarter 2026 homeownership rate at 65.3 percent, the rental vacancy rate at 7.3 percent and the homeowner vacancy rate at 1.1 percent, with the South's rental vacancy rate the nation's highest at 9.5 percent in the second quarter. U.S. Census Bureau, July 28, 2026. Implication: nearly one Southern rental in ten sat empty, which is the vacancy an Atlanta investor should carry in a pro forma, not the 5 percent of the 2021 playbook. (Later release restating the first-quarter figures that stood during May.)
- The CPI shelter index rose 0.6 percent in April and 3.3 percent on the year, with rent of primary residence up 0.5 percent on the month and 2.8 percent on the year and owners' equivalent rent up 0.5 percent and 3.3 percent. Bureau of Labor Statistics, May 12, 2026. Implication: the government's rent measure was still running two points above Atlanta's market rent growth, because CPI rent lags the market by a year or more.
- Zillow's economists noted after the April CPI that owners' equivalent rent rose 0.53 percent against their 0.44 percent forecast and that "housing is still expected to be an overall disinflationary influence on consumer prices in 2026." Zillow Research, May 12, 2026. Implication: the shelter surprise in April CPI was part of what pushed the 10-year up in mid-May, so Atlanta's flat rents were not what the bond market was looking at.
- The national average price of regular gasoline reached $4.50 a gallon the week of May 11, up from $3.12 the same week of 2025. U.S. Energy Information Administration via FRED, May 11, 2026. Implication: a corridor commuter's fuel cost was up roughly $1.40 a gallon on the year in May, a carrying cost that competes with the mortgage payment in a household budget.
Part Three · VII · Credit, distress, and policy
Credit, distress, and policy
A slow leak to source from, not a wave to wait for. The single-family delinquency rate was 1.86 percent, near its series low against an 11.48 percent peak in 2010, and the 90-plus transition rate ticked to 1.5 percent from 1.4. Distressed sales were still 2 percent of national closings. Georgia changed the carrying-cost rules in the owner's favor: the May 11 signing cut the income tax to 4.99 percent and capped homestead assessments at inflation.
SOURCE: FED DRSFRMACBS VIA FRED; NY FED HHDC
FEDERAL RESERVE AND NY FED · UNITED STATES · 2026 · DELINQUENCY RATES, PERCENTConstruction is The Record's own, today's 1.86 percent single-family delinquency against the 11.48 percent all-bank peak of January 2010. Federal Reserve DRSFRMACBS via FRED and the NY Fed household-debt report.
Single-family delinquency
Single-family delinquency (Federal Reserve): today 1.86%; 2005-07 average 1.94%; long-run norm 3.72%; worst in series 11.48% (Jan 2010). Near the series low.
Debt in delinquency, U.S.
Debt in delinquency, U.S. (Federal Reserve): today 4.8%. Rising from a low base.
90-plus transition rate
90-plus transition rate (Federal Reserve): today 1.5%. Up a tenth on the quarter.
Atlanta unemployment, May 2026
Atlanta unemployment, May 2026 (BLS): today 3.20%; 2005-07 average 4.80%; long-run norm 5.13%; worst in series 12.30% (Apr 2020). Below long-run norm.
For the transaction
Computed on standard amortization at May's 6.44 percent rate, principal and interest on a $400,000 corridor loan is about $2,512 a month; the May 11 cut in the state income tax to 4.99 percent trims a Georgia buyer's qualifying math by a few hundred dollars a year.
How it got here
Distress rose from a low base, not a cliff: the single-family delinquency rate was 1.86 percent against an 11.48 percent peak in 2010, and the share of all debt in delinquency was 4.8 percent at the end of March.
The other side
The other side is that the credit box did not tighten. The Fed's April loan-officer survey, released May 4, reported standards basically unchanged, and home-equity balances rose to $446 billion in the quarter, so owners were tapping equity rather than selling.
In the corridor
Georgia's policy moved in the owner's favor in May: the 4.99 percent income tax rate took effect for 2026, the homestead cap limits Fulton and Forsyth assessment growth to inflation, and the state issued $500 refunds from May 4.
What to say at the closing table
- To a lender's client
If a file is struggling, it is the payment, not the credit box. Banks did not tighten mortgage standards in the first quarter of 2026 on the Fed's own survey, and delinquencies are rising only from 1.4 to 1.5 percent. If a file is tight, the issue is the borrower's numbers at the new rate.
Sources: Federal Reserve DRSFRMACBS and SLOOS, May 2026 · Federal Reserve Bank of New York household debt, May 2026 · Office of the Governor of Georgia, May 11, 2026
Figures and method.
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| Single-family delinquency (Federal Reserve) | 1.86% | 1.94% | 11.48% (Jan 2010) | 3.72% | near the series low |
| Debt in delinquency, U.S. (Federal Reserve) | 4.8% | -- | -- | -- | rising from a low base |
| 90-plus transition rate (Federal Reserve) | 1.5% | -- | -- | -- | up a tenth on the quarter |
| Atlanta unemployment, May 2026 (BLS) | 3.20% | 4.80% | 12.30% (Apr 2020) | 5.13% | below long-run norm |
| Figure | Reading | Source |
|---|---|---|
| U.S. delinquency, Q2 | 1.86% | Federal Reserve |
| 90-plus transition rate, Q1 | 1.5% | Federal Reserve |
| Georgia income tax rate, 2026 | 4.99% | Georgia governor |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 25 findings for this section.
- The FOMC held the federal funds target at 3.50 to 3.75 percent on April 29 by an 8 to 4 vote, with Stephen Miran preferring a quarter-point cut and Beth Hammack, Neel Kashkari and Lorie Logan opposing the statement's easing bias. Federal Reserve Board, April 29, 2026. Implication: the policy rate that stood through May was set by a committee split three ways, which is why the market repriced the path so sharply when inflation printed.
- The April 29 statement said "inflation is elevated, in part reflecting the recent increase in global energy prices," that "job gains have remained low, on average," and that "economic activity has been expanding at a solid pace." Federal Reserve Board, April 29, 2026. Implication: the Fed named energy as the inflation driver, so the May oil spike went straight to the rate outlook.
- The minutes of that meeting, released May 20, recorded that most participants expected cuts to become appropriate once "disinflation is firmly back on track," while a majority judged that "some policy firming would likely become appropriate if inflation were to continue to run persistently above 2 percent." Federal Reserve Board, May 20, 2026. Implication: the minutes put a hike on the table in writing, and they landed the day after the daily mortgage rate peaked.
- The minutes also noted that "the ongoing deceleration in housing services prices was likely to continue to be a source of disinflationary pressure." Federal Reserve Board, May 20, 2026. Implication: the Fed was counting on flat rents like Atlanta's to pull inflation down, which is the one housing input working in a borrower's favor.
- The effective federal funds rate averaged 3.63 percent in May, unchanged in practice since February. Federal Reserve via FRED. Implication: the short rate did not move in May; every basis point of the mortgage move came from the bond market.
- The Federal Reserve Board named Jerome Powell chair pro tempore on May 15 following the expiration of his term as chair, to serve "until Warsh is sworn in as the new chair." Federal Reserve Board, May 15, 2026. Implication: for a week in May the Fed had a caretaker chair, during the exact window the 10-year hit its high.
- Kevin Warsh took the oath of office as chairman on May 22, after Senate confirmation as a Board member on May 12 and as chairman on May 13; his term as chairman ends May 21, 2030 and the FOMC unanimously selected him as its chairman the same day. Federal Reserve Board, May 22, 2026. Implication: the June 16 to 17 meeting would be the new chair's first, the catalyst a lender's client should have been watching rather than the calendar.
- Governor Stephen Miran, the lone vote for a cut on April 29, submitted his resignation from the Board on May 14, effective when or shortly before his successor is sworn in. Federal Reserve Board, May 14, 2026. Implication: the committee's only dissent in favor of easing was on its way out in May, which tilted the June vote count toward holding.
- The April Senior Loan Officer Opinion Survey, published May 4, reported banks "left standards basically unchanged over the first quarter for most RRE loan categories" and saw "basically unchanged or weaker demand, on net, for all RRE loan categories," with "modest net shares of banks" reporting stronger demand for home equity lines of credit. Federal Reserve Board, May 4, 2026. Implication: the credit box did not tighten in the first quarter on the Fed's own survey, so a May denial was about the borrower, not the bank; the HELOC demand rise says equity-rich owners were tapping rather than selling.
- Total household debt was $18.8 trillion at the end of the first quarter, up $18 billion, with mortgage balances at $13.19 trillion, up $21 billion, and $530 billion of new mortgage originations in the quarter. Federal Reserve Bank of New York. Implication: mortgage balances barely grew in the quarter, which is what a lock-in market looks like in the aggregate.
- Home equity line balances rose $12 billion to $446 billion in the first quarter, the sixteenth straight quarterly increase. Federal Reserve Bank of New York. Implication: owners were borrowing against equity rather than moving, four years running, which is the supply-side reason corridor listings stayed scarce in Milton and Johns Creek.
- The share of outstanding debt in some stage of delinquency was 4.8 percent at the end of March, the annualized rate of mortgage balances transitioning into 90-plus-day delinquency rose to 1.5 percent from 1.4 percent, and about 59,000 consumers had a new foreclosure notation added in the quarter. Federal Reserve Bank of New York. Implication: mortgage distress was rising from a very low base, not surging; 59,000 new foreclosures nationally in a quarter is small against the 1.47 million homes listed for sale in April.
- Distressed sales were 2 percent of national existing-home sales in April, flat on the month and the year. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: there was no distressed-inventory channel feeding the market in May, so a buyer waiting for foreclosures was waiting for something the data did not show.
- The refinance share of mortgage applications moved from 40.8 percent for the week ending May 8 to 41.9 percent for the week ending May 15 and 37.5 percent for the week ending May 22, while the ARM share went from 8.8 to 9.6 to 9.4 percent. HousingWire, May 13, 2026. Implication: borrowers answered the spike by shifting structure toward adjustable loans and by abandoning refinances, not by leaving the purchase market.
- MBA's FHA contract rate ran 30 basis points under conforming at 6.16 percent for the week ending May 8 and 34 under at 6.31 percent for the week ending May 22, while jumbo ran 2 to 3 basis points over conforming. HousingWire, May 13, 2026. Implication: an FHA borrower had a third of a point of rate advantage over conventional in May, and a jumbo borrower paid almost no premium, which widened the financing options at both ends of the corridor's price range.
- The 2026 conforming loan limit for one-unit properties stood at $832,750, up $26,250 from 2025, with the high-cost ceiling at $1,249,125, set by a 3.26 percent rise in the FHFA index from the third quarter of 2024 to the third quarter of 2025. FHFA, November 25, 2025. Implication: Redfin's May medians in Alpharetta ($785,000), Johns Creek ($733,840), Roswell ($618,000) and Sandy Springs ($730,000) all sat inside the conforming limit with 20 percent down; Milton's $1,154,500 did not.
- Governor Kemp signed SB 33 on May 11, which the Governor's office described as enacting "broad property tax reform establishing a new Local Homestead Option Sales Tax (LHOST) for homeowner relief," alongside HB 463, which lowers the state income tax rate from 5.19 to 4.99 percent effective January 1, 2026. Office of the Governor of Georgia, May 11, 2026. Implication: the property-tax reform the May briefs dated to May 5 was signed May 11, and the income-tax cut in the same signing lowers a Georgia buyer's qualifying math by a few hundred dollars a year.
- Georgia Recorder reported that SB 33 "caps property tax assessments at the rate of inflation and allows cities and counties to increase sales taxes to give homeowners additional property tax relief," that the bill was House Speaker Jon Burns' priority, and that the Governor said "this bill delivers on that issue." Georgia Recorder, May 11, 2026. Implication: a corridor owner's assessed value growth is capped at inflation going forward, which improves the hold case for anyone who bought before 2022 and expects to stay.
- The same report quoted Rep. Scott Holcomb that "complying with the Constitution is not optional or discretionary," signaling a constitutional challenge on the revenue-bill origination question. Georgia Recorder, May 11, 2026. Implication: the assessment cap carried litigation risk from the day it was signed, so a seller should not price a tax benefit that a court could unwind.
- The May 11 signing also included HB 445, expanding appeal rights for tangible personal property reassessments, and HB 1209, a limited sales-tax exemption for construction materials near state convention facilities. Office of the Governor of Georgia, May 11, 2026. Implication: neither bill touches residential closings directly, but the appeal-rights expansion signals the General Assembly's direction on assessment disputes.
- Georgia began issuing special income tax refunds under HB 1000 on May 4, $250 for single filers, $375 for heads of household and $500 for joint filers, returning more than $1 billion to taxpayers who filed 2024 and 2025 returns. Office of the Governor of Georgia, May 4, 2026. Implication: a Georgia couple received $500 in May, a small but dated addition to a closing-cost cushion.
- Governor Kemp suspended Georgia's motor fuel tax of 33.3 cents a gallon on gasoline and 37.3 cents on diesel from May 20 through June 3, citing Memorial Day travel. Office of the Governor of Georgia, May 15, 2026. Implication: the state was spending its own revenue to offset the oil shock that was pushing mortgage rates up, a reminder that the energy channel was hitting Georgia households twice.
- Georgia's net tax revenues for April were almost $3.95 billion, up 5.9 percent on the year, with individual income tax up 10.1 percent and net sales tax up 8.5 percent. Office of the Governor of Georgia, May 7, 2026. Implication: the state's fiscal position going into the tax cuts was strong, which lowers the odds the homestead cap gets walked back for revenue reasons.
- Initial unemployment claims in Georgia were 18,336 in April, up 867 on the month and down 2,700 on the year. Georgia Department of Labor, May 21, 2026. Implication: layoffs in Georgia were running below last year, so there was no local income shock behind mortgage distress in the spring.
- National initial jobless claims were 199,000 for the week ending May 2, 212,000 for May 9, 210,000 for May 16, 212,000 for May 23 and 225,000 for May 30; continuing claims ranged from 1,771,000 to 1,786,000. U.S. Department of Labor via FRED. Implication: claims drifted up through May but stayed near the lowest readings of the year, which is why lenders were not tightening on employment risk.
The month
What changed
The month · VIII · What changed since last month
What changed since last month
One clear reversal, and a spring that ran on April's decisions. The weekly survey moved from 6.30 percent on April 30 to 6.53 percent on May 28, the monthly average from 6.33 to 6.44 percent, and the 10-year's daily high from 4.42 percent in April to 4.67 percent on May 19. The local series did not turn with it: value drifted to $378,748, inventory built to 32,227, and the metro closed the most homes since October 2024 on the nowcast, while the contract pipeline on file ran 21.8 percent below a year earlier.
SOURCE: FREDDIE MAC PMMS VIA FRED MORTGAGE30US
FREDDIE MAC PMMS VIA FRED · UNITED STATES · 2026 · WEEKLY SURVEY, PERCENTConstruction is The Record's own, Freddie Mac's survey from 6.30 percent on April 30 to 6.53 percent on May 28 against a 7.68 percent long-run average. Freddie Mac PMMS via FRED MORTGAGE30US.
30-year weekly
30-year weekly (Freddie Mac): today 6.53%; 2005-07 average 6.21%; long-run norm 7.68%; worst in series 18.45% (Oct 1981). The month's high print.
10-year Treasury
10-year Treasury (U.S. Treasury): today 4.67%; 2005-07 average 4.57%; long-run norm 5.81%; worst in series 15.32% (Sep 1981). Highest since early 2025.
New-home supply, U.S.
New-home supply, U.S. (Census): today 9.4; 2005-07 average 6.42; long-run norm 6.21; worst in series 12.20 (Jan 2009). Builders carry the overhang.
For the transaction
The 23 basis point rise in the survey from April 30 to May 28 is about $46 a month on the $305,120 metro loan at 20 percent down, computed on standard amortization, before the 6.75 percent daily peak.
How it got here
The policy direction reversed: the April minutes released May 20 put a hike in writing, Kevin Warsh was sworn in as chair on May 22, and the daily quote peaked at 6.75 percent on May 19 before easing to 6.53 percent by month end.
The other side
The other side is that the local series held through the month. Zillow's metro value fell only from $379,648 in April to $378,748 and inventory rose to 32,227, so the May rate shock had not yet reached the recorded local data by month-end on May 31.
In the corridor
The corridor firmed as the metro softened: Alpharetta's value held near flat on the year while Forsyth gave back 2.74 percent, and Alpharetta went pending in 22 days against 17 a year earlier.
What to say at the closing table
- To a buyer
The month ended with the oil shock unwinding, and that is the opening. The survey closed May at its high of 6.53 percent, but Brent crude had already fallen 20 percent from its May 18 peak and the June 4 print eased. The window is the inventory, up on the year; the rate is why it matters now.
Sources: Freddie Mac PMMS and U.S. Treasury via FRED, May 2026 · Zillow Research, pulled October 6, 2026 · NAR existing-home sales, May 11, 2026
Figures and method.
| Measure (publisher) | Today | 2005-07 | Worst in series (date) | Long-run norm | What it says |
|---|---|---|---|---|---|
| 30-year weekly (Freddie Mac) | 6.53% | 6.21% | 18.45% (Oct 1981) | 7.68% | the month's high print |
| 10-year Treasury (U.S. Treasury) | 4.67% | 4.57% | 15.32% (Sep 1981) | 5.81% | highest since early 2025 |
| New-home supply, U.S. (Census) | 9.4 | 6.42 | 12.20 (Jan 2009) | 6.21 | builders carry the overhang |
| Figure | Reading | Source |
|---|---|---|
| The monthly survey change | +23 bps | Freddie Mac |
| 10-year, May 19 | 4.67% | U.S. Treasury |
| New-home months of supply | 9.4 | Census |
Dollar lines use standard 30-year amortization at the month's 6.44 percent average rate, 20 percent down; each line shows its inputs and is not a lender quote.
All 15 findings for this section.
- The 30-year fixed's monthly average rose from 6.33 percent in April to 6.44 percent in May, and the weekly survey went from 6.30 percent on April 30 to 6.53 percent on May 28. Freddie Mac via FRED. Implication: April's relief was fully reversed; May's last print was the highest since August 2025.
- The daily 30-year rate's high moved from 6.46 percent on April 2 to 6.75 percent on May 19, per the two publishers that track it. Realtor.com® Economic Research, April 30, 2026. Implication: the intra-month peak rose 29 basis points from April to May, a bigger move than the surveys show.
- The 10-year Treasury's monthly average rose from 4.32 to 4.48 percent and its daily high went from 4.42 percent (April 29) to 4.67 percent (May 19). US Treasury via FRED. Implication: the benchmark set a sixteen-month high in May that it had not approached in April.
- The metro ZHVI slipped from $379,648 to $378,748, and its annual decline narrowed from 2.2 percent in Zillow's April report to 2.0 percent in its May report. Zillow Research. Implication: values kept drifting down in dollars while the annual comparison got slightly easier. On Zillow's current series the annual decline had narrowed every month since October 2025, so May continued a deceleration rather than starting one.
- Metro inventory rose from 30,336 to 32,227 and new listings from 8,318 to 8,994, while the sales nowcast rose from 7,988 to 8,421 and days to pending fell from 37 to 33. Zillow Research. Implication: supply, sales and speed all rose together in May, which is what a seasonal peak looks like, not a turn.
- Georgia MLS contract signings went from 21.8 percent below a year earlier in April to 27.9 percent below in May, while closings went from 0.6 percent below to 0.6 percent above. Georgia MLS, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: the gap between what closed and what went under contract widened by six points in a month, the clearest change in the local data. (The May figure is a later release.)
- Redfin's national four-week pending sales went from up 9.6 percent on the year (period ending May 10) to up 4.7 percent (period ending May 24), and Atlanta's from down 1.1 percent to down 3.1 percent. Redfin, May 14, 2026. Implication: national demand halved its annual gain in two weeks, and Atlanta went from slightly negative to clearly negative.
- Headline CPI inflation went from 3.3 percent in March to 3.8 percent in April on the May 12 release, and the Michigan sentiment index went from 49.8 to 44.8. Bureau of Labor Statistics, May 12, 2026. Implication: inflation re-accelerated and sentiment collapsed in the same month, which flipped the rate conversation from "when do cuts come" to "is a hike possible."
- The Federal Reserve went from Jerome Powell as chair through May 15, to Powell as chair pro tempore, to Kevin Warsh sworn in on May 22. Federal Reserve Board, May 22, 2026. Implication: the June meeting became a new chair's first, which added a policy unknown on top of the oil unknown.
- Georgia went from a pending property-tax bill in April to a signed assessment cap and a 4.99 percent income tax rate on May 11, plus a gas-tax suspension from May 20. Office of the Governor of Georgia, May 15, 2026. Implication: the state's carrying-cost rules changed in the owner's favor in May while the financing cost moved against the buyer.
- Metro rent rose from $1,819.52 to $1,828.90 while the computed payment on the median home rose from about $1,857 to $1,917, widening renting's principal-and-interest edge from about $38 to $88. Computed from Zillow Research ZORI and median sale price. Implication: the carrying-cost math moved further from ownership for a second month, driven by the rate rather than by price.
- The MBA ARM share went from 8.8 percent in the week ending May 8 to 9.6 percent in the week ending May 15, the highest since October 2025, and the refinance index fell 18 percent in the week ending May 22. HousingWire, May 13, 2026. Implication: borrower behavior changed inside the month: more adjustable loans, far fewer refinances, roughly the same number of purchases.
- Brent crude went from $103.70 on May 6 to $116.73 on May 18 and back to $92.88 on May 29, and the 10-year went from 4.67 percent on May 19 to 4.45 percent on May 29. EIA and US Treasury via FRED. Implication: the month ended with the oil shock unwinding, so the June rate story opened in the buyer's favor even though the May survey closed at its high.
- Alpharetta's median days to pending fell from 38 in April to 22 in May, and Sandy Springs' from 30 to 24, while Cumming's held at 36. Zillow Research. Implication: the corridor's spring arrived in May in north Fulton and did not arrive in Forsyth.
- The first June survey print fell to 6.48 percent for the week ending June 4, the first weekly decline since the one-basis-point dip of May 14, before rising again to 6.52 percent on June 11. Freddie Mac via FRED, June 11, 2026. Implication: the May 28 print of 6.53 percent was the top of this episode, which a client who locked in May could not have known. (Later data, noted as such.)
The close · IX · The playbook
The playbook
Five seller-side moves
- Price to the April comps on day one; the metro value index has drifted to $378,748, about 2 percent under a year earlier.
- Expect the cut to be the path to the close; about 26.64 percent of metro listings had already reduced by May.
- In Alpharetta and Johns Creek, price ahead of the market: the share selling above list fell about eight points on the year, to 19.77 and 29.02 percent.
- Quote the net after a typical give-back, not the list price, on a market where the mean sale closed at 98.30 percent of list.
- If the buyer waited past May 14, use their higher payment as the reason to price to sell, not to test.
Five buyer-side moves
- Quote the daily rate near 6.75, not the 6.53 weekly survey, when you budget principal and interest.
- Use the choice: metro active listings reached 32,227 in May, about 1,900 more than April and 770 more than a year earlier.
- Ask the builder for the discount; new-home sales in the South ran about 15 percent below a year earlier with a 9.4-month supply in April.
- If an adjustable loan fits, you are in the mainstream: the ARM share reached 9.6 percent in mid-May, the most since October 2025.
- Lead the corridor negotiation with price per foot, down about 5 percent on the year in four of the five cities.
Each tied to a figure on this page
Each move above cites a figure carried in the sections: the $378,748 value index, the 26.64 percent price-cut share, the 32,227 active count, the 9.6 percent ARM share, and the corridor's price-per-foot decline.
The close · X · What to watch
What to watch
- The June 16 to 17 FOMC meeting, Kevin Warsh's debut as chair, after the April minutes put a hike in writing (Federal Reserve).
- The Freddie Mac survey after the month, which fell to 6.48 percent for June 4 before rising to 6.52 (Freddie Mac via FRED).
- Brent crude, which fell to $92.88 by May 29 from its $116.73 high on May 18; the June rate path tracks it (EIA via FRED).
- The Georgia homestead assessment cap signed May 11, which faces a constitutional challenge on the revenue-origination question (Georgia Recorder).
- The May existing-home and Case-Shiller releases due in June, which will show if the contract decline reached closings (NAR; S&P Dow Jones).
The close · Sources
Sources and ledger
Corrected October 2026: an earlier version called May's narrowing the first sign of deceleration. On Zillow's current series the decline had narrowed every month since October 2025.
Corrected October 2026: the chart "The 10-year Treasury through May, daily close" now plots every observation on a true time scale; its low label is revised to 4.4% (2026-05-06), from 4.4% (2026-05-01).
All sources and the full finding ledger.
Freshness and proxy block. Rate and Treasury lines are May monthly and daily figures from Freddie Mac and the U.S. Treasury via FRED. Zillow value, inventory, rent, and activity series are May readings from the public files. Redfin metro, county, and city figures are the May market tracker. Case-Shiller is the March repeat-sales print released May 26, and FHFA is the first-quarter report released May 26. Figures published after the month are labeled.
I Rates and affordability
- The 30-year fixed averaged 6.30 percent for the week ending April 30, up from 6.23 percent and against 6.76 percent a year earlier, the last print before May opened. Freddie Mac, April 30, 2026. Implication: a buyer who locked in the last week of April locked the cycle's low for the spring; every May print came in above it.
- Freddie Mac's April 30 release said purchase applications were running "over 20 percent above a year ago" after several weeks of modest rate declines. Freddie Mac, April 30, 2026. Implication: May opened with real buyer demand queued behind a 6.3 percent rate, which is why the rate spike mid-month mattered so much to volume.
- The 30-year fixed averaged 6.37 percent for the week ending May 7, up from 6.30 percent and against 6.76 percent a year earlier. Freddie Mac, May 7, 2026. Implication: the first May print was 39 basis points below last year, so a buyer comparing against a 2025 pre-approval was still ahead.
- The 15-year fixed averaged 5.72 percent for the week ending May 7, up from 5.64 percent and against 5.89 percent a year earlier. Freddie Mac, May 7, 2026. Implication: the 15-year sat 65 basis points under the 30-year, a gap wide enough to matter for an equity-rich move-down buyer.
- Freddie Mac's May 7 release pointed to "a boost in new-home sales, median new-home prices being down to their lowest level since July 2021, and higher inventory than in recent years" as conditions that "could modestly ease affordability pressures through the spring homebuying season." Freddie Mac, May 7, 2026. Implication: the survey's own economist framed early May as a buyer's window, and the window closed within two weeks.
- The 30-year fixed averaged 6.36 percent for the week ending May 14, down one basis point, against 6.81 percent a year earlier. Freddie Mac, May 14, 2026. Implication: the mid-May survey still read 45 basis points under last year the same week the daily trackers had already started climbing.
- The 15-year fixed averaged 5.71 percent for the week ending May 14, down from 5.72 percent, against 5.92 percent a year earlier. Freddie Mac, May 14, 2026. Implication: the 15-year held under 5.75 percent for the last time in the month.
- Freddie Mac's May 14 release said purchase demand "is softening" but "remains above this time last year." Freddie Mac, May 14, 2026. Implication: the demand read turned from "accelerating" on April 30 to "softening" on May 14, two weeks apart.
- The 30-year fixed jumped 15 basis points to 6.51 percent for the week ending May 21, against 6.86 percent a year earlier. Freddie Mac, May 21, 2026. Implication: the largest weekly move of the month landed in the third week, and a buyer who had not locked by May 14 paid for it.
- The 15-year fixed rose 14 basis points to 5.85 percent for the week ending May 21, against 6.01 percent a year earlier. Freddie Mac, May 21, 2026. Implication: the 15-year's annual discount narrowed to 16 basis points, matched only by the May 28 print.
- Freddie Mac's May 21 release advised that "by shopping around for the best mortgage rate and getting multiple quotes, they can potentially save thousands." Freddie Mac, May 21, 2026. Implication: in a week when lender quotes were moving daily, a second quote was worth more than usual.
- The 30-year fixed averaged 6.53 percent for the week ending May 28, up from 6.51 percent, against 6.89 percent a year earlier, the highest weekly print of the month. Freddie Mac, May 28, 2026. Implication: May closed 23 basis points above where it opened, and the survey of record was still rising into June.
- The 15-year fixed averaged 5.87 percent for the week ending May 28, up from 5.85 percent, against 6.03 percent a year earlier. Freddie Mac, May 28, 2026. Implication: the 15-year ended May 23 basis points above the April 30 print of 5.64 percent, the same move as the 30-year.
- Freddie Mac's May 28 release said "pending home sales have increased three months in a row, indicating there's latent demand and homebuyers are ready to jump back into the market if mortgage rates decline." Freddie Mac, May 28, 2026. Implication: the survey's own economist described the demand as rate-contingent, which is the seller's risk heading into summer.
- The 30-year fixed averaged 6.44 percent for May on the monthly average of the weekly survey, up from 6.33 percent in April and against 6.82 percent in May 2025. Freddie Mac via FRED. Implication: the monthly average rose 11 basis points on the month and sat 38 basis points under a year earlier, a narrower annual discount than April's 40.
- The 15-year fixed averaged 5.79 percent for May on the monthly average, up from 5.68 percent in April and against 5.95 percent in May 2025; the series' 2026 low was February's 5.43 percent. Freddie Mac via FRED. Implication: a refinance into a 15-year that pencilled in February at 5.43 percent cost 36 basis points more by May.
- The four weekly 30-year prints of May were 6.37, 6.36, 6.51 and 6.53 percent; the four 15-year prints were 5.72, 5.71, 5.85 and 5.87 percent. Freddie Mac via FRED, May 28, 2026. Implication: the month split cleanly in two, a flat first half and a 17 basis point step up in the second, so a lock date mattered more than the month's average.
- The 10-year Treasury constant-maturity yield averaged 4.48 percent for May, up from 4.32 percent in April and against 4.42 percent in May 2025. US Treasury via FRED. Implication: the benchmark behind mortgage pricing rose 16 basis points on the month, and the mortgage rate rose 11, so spreads absorbed part of the move.
- The 10-year Treasury closed at 4.39 percent on May 1, 4.59 percent on May 15, 4.67 percent on May 19, 4.56 percent on May 22 and 4.45 percent on May 29. US Treasury via FRED. Implication: the 28 basis point run from May 1 to May 19 is the whole story of the month's mortgage spike, and the 22 basis point retreat into May 29 is why June's first survey print fell.
- The 10-year's May 19 close of 4.67 percent was the highest daily close since January 14, 2025, when the series last stood at 4.78 percent. US Treasury via FRED. Implication: a rate quote on May 19 or 20 was priced off a sixteen-month high in the benchmark, which is the single worst day of the month to have locked.
- The 30-year Treasury closed at 5.18 percent on May 19, the highest daily close in the series since July 2007. US Treasury via FRED, May 19, 2026. Implication: the long end of the curve, which jumbo and portfolio lenders watch, hit a nineteen-year high in the same week the daily mortgage rate peaked.
- The 30-year Treasury averaged 5.03 percent for May, up from 4.91 percent in April; the 2-year averaged 4.00 percent, up from 3.80 percent. US Treasury via FRED. Implication: the 2-year rose 20 basis points in a month with no Fed move, so the market repriced the policy path, not just term premium.
- The 10-year minus 2-year spread stood at 0.54 percentage points on May 18 and 19 and narrowed to 0.43 by May 22. Federal Reserve via FRED. Implication: the curve was steepest at the mortgage-rate peak, which is the usual shape when the long end leads the move.
- Mortgage News Daily's 30-year daily index reached 6.75 percent on May 19, the highest since July 2025, up roughly three-quarters of a point since the Iran war began. Mortgage News Daily, May 19, 2026. Implication: the daily rate a borrower was actually quoted on May 19 ran 24 basis points above that week's Freddie Mac survey.
- Mortgage News Daily attributed the May 19 move to bond-market volatility that pushed 10-year yields to their highest in more than a year, and noted that mortgage bonds outperformed Treasuries because of increased purchases by Fannie Mae and Freddie Mac. Mortgage News Daily, May 19, 2026. Implication: GSE buying of mortgage bonds was cushioning the retail rate, a support that depends on policy rather than on the market.
- The 30-year to 10-year mortgage spread was 1.93 percentage points on May 2 against a historical norm of 1.60 to 1.80, and the author calculated that 2023's worst spreads would have put the 30-year at 7.62 percent. HousingWire, May 2, 2026. Implication: roughly a point of a client's rate in May was being decided by spread behavior, not by the Fed or the Treasury market.
- By May 16 the same analyst put the mortgage spread at 1.92 percentage points, down from 1.96, with the 10-year at 4.596 percent, and held his 2026 forecast ranges at 3.80 to 4.60 percent for the 10-year and 5.75 to 6.75 percent for the 30-year. HousingWire, May 16, 2026. Implication: the daily rate's 6.75 percent print three days later touched the top of that forecast band, so a client's "wait for the dip" case depended on the band holding.
- The same May 16 analysis counted weekly pending sales at 78,006 against 73,523 a year earlier, with purchase applications up 4 percent on the week and 7 percent on the year. HousingWire, May 16, 2026. Implication: national demand was still running 6 percent ahead of last year going into the spike, so the May 21 reversal came off a higher base than 2025's.
- MBA's composite application index fell 4.4 percent for the week ending May 1, with the seasonally adjusted purchase index down 4 percent but 5 percent above a year earlier and the refinance index down 5 percent but 29 percent above a year earlier. Mortgage Bankers Association, May 6, 2026. Implication: the first week of May gave back the late-April surge, so the 21 percent annual purchase gain of mid-April had already faded to 5 percent before rates spiked.
- The average purchase loan size on MBA applications reached a record $467,300 for the week ending May 1, up 1.5 percent on the week and 5.2 percent on the year, against an overall average of $410,900. Haver Analytics, May 6, 2026. Implication: the applicants still in the market were higher-balance borrowers, which fits the corridor's price band better than the metro median.
- MBA's composite index rose 1.7 percent for the week ending May 8, with the purchase index up 4 percent on the week and 7 percent on the year and the refinance index down 1 percent but 28 percent above a year earlier. Mortgage Bankers Association, May 13, 2026. Implication: the second week of May was the demand high-water mark of the month; everything after it declined.
- MBA's 30-year conforming contract rate was 6.46 percent for the week ending May 8, with jumbo at 6.48, FHA at 6.16, the 15-year at 5.83 and the 5/1 ARM at 5.70 percent; the refinance share was 40.8 percent and the ARM share 8.8 percent. The MBA survey, as reported by HousingWire, May 13, 2026. Implication: jumbo priced only 2 basis points over conforming, so a corridor buyer above $832,750 was not paying a meaningful jumbo premium in early May.
- MBA's Joel Kan said on May 13 that purchase applications were "7% ahead of last year's pace, with all loan types showing increases in purchase activity, as potential homebuyers shrugged off the current economic and mortgage rate uncertainties and returned to the market." The MBA survey, as reported by HousingWire, May 13, 2026. Implication: the lender read in mid-May was that buyers had adapted to the rate; the next two weeks tested that.
- MBA's composite index fell 2.3 percent for the week ending May 15, the purchase index fell 4 percent on the week but stayed 8 percent above a year earlier, and the refinance share rose to 41.9 percent. Mortgage Bankers Association, May 20, 2026. Implication: applications hit a five-week low in the week the daily rate hit 6.75 percent, the first clear volume response to the spike.
- MBA's 30-year conforming contract rate rose to 6.56 percent for the week ending May 15 from 6.46, with jumbo at 6.58, FHA at 6.24, the 15-year at 5.93 and the 5/1 ARM at 5.76 percent. The MBA survey, as reported by Mortgage News Daily, May 22, 2026. Implication: the ARM sat about 80 basis points under the fixed rate, which is the spread that starts moving borrowers into adjustable structures.
- The ARM share of applications reached 9.6 percent for the week ending May 15, the highest since October 2025. The MBA survey, as reported by Mortgage News Daily, May 22, 2026. Implication: roughly one in ten applicants chose an adjustable loan to manage the payment, a market-wide response rather than an individual risk signal.
- Joel Kan said applications for the week ending May 15 were "down to the lowest level in five weeks as purchase borrowers pulled back across conventional and government loan types." The MBA survey, as reported by Mortgage News Daily, May 22, 2026. Implication: the pullback was across FHA and VA as well as conventional, so it was a rate effect, not a credit-box effect.
- MBA's composite index fell 8.5 percent for the week ending May 22, the refinance index fell 18 percent, the purchase index slipped 0.4 percent on the week while staying 5 percent above a year earlier, and the refinance share dropped to 37.5 percent. Mortgage Bankers Association, May 27, 2026. Implication: refinance volume collapsed when the rate crossed 6.6 percent, while purchase demand held, so the rate spike hurt lenders' pipelines more than it hurt transactions.
- MBA's 30-year conforming contract rate reached 6.65 percent for the week ending May 22, up 30 basis points over five weeks to the highest since August 2025, with jumbo at 6.68, FHA at 6.31, the 15-year at 5.97 and the 5/1 ARM at 5.81 percent; the average purchase loan size was $473,600. The MBA survey, as reported by HousingWire, May 27, 2026. Implication: the lender-reported rate ended May 14 basis points above the May 21 Freddie Mac survey, which is the gap a client sees between the headline and the quote.
- Joel Kan said on May 27 that "the 30-year fixed rate has increased 30 basis points over the past five weeks to its highest level since August 2025" and that "many borrowers understandably backed away from refinancing last week." The MBA survey, as reported by HousingWire, May 27, 2026. Implication: an owner who had been waiting to refinance a 2023 or 2024 loan lost the May window; the next one depended on the June ceasefire holding.
- The household income needed to afford the median Atlanta home with 20 percent down was $102,374 in May, up from $101,708 in April and down 4.4 percent from $107,092 a year earlier. Zillow Research. Implication: the metro's affordability bar was still 4 percent lower than a year earlier even after May's rate climb, because the rate was lower and the value was lower than in May 2025. (Computed: annual change from the two fetched values.)
- The computed principal-and-interest payment on the metro's May median sale price of $381,400 with 20 percent down at the month's 6.44 percent average rate was $1,917, up from a computed $1,857 in April and $1,731 in February. Computed from Zillow Research median sale price. Implication: the payment on the median home rose $186 in three months, and of May's $60 rise about $22 came from the rate and the rest from the higher median price.
- The monthly mortgage payment on the typical U.S. home was $1,861 in May with 20 percent down, up 1.1 percent from April and still 3.1 percent below May 2025. Zillow Research, June 4, 2026. Implication: nationally the payment was still cheaper than a year ago, so the "rates are up" conversation in May was about the month, not the year.
- Redfin's median monthly housing payment reached $2,637 at a 6.51 percent rate for the four weeks ending May 24, the highest in eleven months, though still 0.9 percent below a year earlier. Redfin, May 28, 2026. Implication: the payment a buyer was underwriting at month's end was the highest since June 2025, which is the figure to put in front of a client comparing against a stale pre-approval.
- Realtor.com® Economic Research recorded mortgage rates climbing from 6.30 to 6.53 percent across May and called it a move that erased "much of April's relief." Realtor.com® Economic Research, June 3, 2026. Implication: the full-month rate path was a straight 23 basis point climb, with no dip a buyer could have timed.
II Prices and valuations
- The Atlanta metro ZHVI (mid-tier, smoothed, seasonally adjusted) stood at $378,748 for May, down 0.24 percent from $379,648 in April and down 2.14 percent from $387,018 in May 2025. Zillow Research. Implication: the typical metro home lost about $8,300 of value over the year, a number a seller who bought in 2024 or 2025 needs in front of them before pricing. (Computed: changes from the fetched values; the May 17 brief's in-period April endpoint of $381,835 has since been revised to $379,648 in the current file.)
- The metro ZHVI's two-year path runs from a high of $394,714 in October 2024 to $377,428 in August 2026, with May's $378,748 only about $1,300 above that eventual low. Zillow Research. Implication: the metro has drifted down about 4 percent from its 2024 peak in a slow grind rather than a step, which is why list prices have been able to stay above values without a visible crash.
- The metro top-tier ZHVI (65th to 95th percentile) stood at $629,690 for May, down 0.21 percent on the month and 1.19 percent on the year. Zillow Research. Implication: the top tier, which is where most corridor transactions sit, was giving ground at roughly half the pace of the metro median. (Computed: changes from the fetched values.)
- The metro single-family ZHVI stood at $385,563 for May, down 0.24 percent on the month and 2.12 percent on the year. Zillow Research. Implication: detached homes tracked the all-homes index almost exactly, so the softness was not a condo story. (Computed: changes from the fetched values.)
- The metro median list price rose to $404,667 in May, up 1.68 percent from $397,967 in April and 2.27 percent from $395,667 a year earlier. Zillow Research. Implication: sellers were asking 2 percent more than a year ago in a market valued 2 percent lower, a 4 point gap that shows up as price cuts and longer marketing. (Computed: changes from the fetched values.)
- The metro median sale price (not seasonally adjusted) rose to $381,400 in May, up 2.01 percent from $373,900 in April and 1.71 percent from $375,000 in May 2025. Zillow Research. Implication: closed prices were up on the year even as values were down, because the mix of what sold in May skewed larger and newer than the housing stock as a whole. (Computed: changes from the fetched values.)
- The seasonally adjusted metro median sale price was $374,774 in May, up 0.18 percent on the month and 1.62 percent on the year. Zillow Research. Implication: strip out the spring seasonal lift and closed prices were close to flat month to month. (Computed: changes from the fetched values.)
- The metro median new-construction sale price was $417,637 in May, down from $424,900 in April and 2.2 percent below $426,990 a year earlier. Zillow Research. Implication: the new-build premium over the resale median narrowed to about $36,000 in May, which is the number a builder's incentive has to beat. (Computed: annual change and premium from the fetched values.)
- Zillow's April market report, published May 6, put Atlanta's typical home value at $379,802, up 0.4 percent on the month and down 2.2 percent on the year, with inventory up 3.6 percent and the sales count down 5.7 percent on the year. Zillow Research, May 6, 2026. Implication: this was the Atlanta read on file during May; it showed values falling on the year while the month-to-month figure (raw, not seasonally adjusted) ticked up with the season.
- Zillow's May market report, published June 4, put Atlanta's typical home value at $381,077, up 0.4 percent on the month and down 2.0 percent on the year, with inventory up 0.2 percent and the sales nowcast down 2.4 percent on the year. Zillow Research, June 4, 2026. Implication: the annual decline in Atlanta values narrowed from 2.2 to 2.0 percent in a month. (Later release, published after the period.)
- Fulton County's ZHVI was $416,291 in May, down 0.22 percent on the month and 2.61 percent from $427,464 a year earlier. Zillow Research. Implication: the county that holds Alpharetta, Johns Creek, Milton, Roswell and Sandy Springs tracked the metro's decline almost exactly. (Computed: changes from the fetched values.)
- Forsyth County's ZHVI was $611,786 in May, down 0.39 percent on the month and 2.74 percent from $629,035 a year earlier, the largest monthly decline among the six north-metro counties. Zillow Research. Implication: the highest-value county in the corridor was also the one losing value fastest in May, roughly $17,000 off the typical home over the year. (Computed: changes from the fetched values.)
- Cherokee County's ZHVI was $472,356 in May, down 0.21 percent on the month and 1.04 percent from $477,319 a year earlier, the most resilient of the six counties. Zillow Research. Implication: a seller in Woodstock or Canton had about a point and a half less annual decline to explain than a seller in Forsyth. (Computed: changes from the fetched values.)
- Cobb County's ZHVI was $421,745 in May, down 2.62 percent on the year; Gwinnett's was $406,174, down 2.96 percent; DeKalb's was $336,289, down 2.99 percent. Zillow Research. Implication: the three big inner counties all sat near a 3 percent annual decline, so the corridor's softness was the metro's norm, not an outlier. (Computed: changes from the fetched values.)
- Alpharetta's city ZHVI was $717,500 in May, down 0.30 percent on the month and 0.47 percent from $720,911 a year earlier. Zillow Research. Implication: Alpharetta values were essentially flat on the year while the county around them fell 2.6 percent, which argues against pricing an Alpharetta listing off county-wide comps. (Computed: changes from the fetched values.)
- Johns Creek's city ZHVI was $696,471 in May, down 0.15 percent on the month and up 0.82 percent from $690,797 a year earlier. Zillow Research. Implication: on Zillow's value index Johns Creek was one of the few north-metro cities still up on the year, which cuts against the "Johns Creek down 6 percent" framing that circulated in May. (Computed: changes from the fetched values.)
- Milton's city ZHVI was $942,205 in May, up 0.05 percent on the month and 2.35 percent from $920,544 a year earlier. Zillow Research. Implication: Milton's typical value rose on the year even as its listing prices and price per square foot came in lower, so a Milton seller's problem in May was positioning, not the underlying value. (Computed: changes from the fetched values.)
- Roswell's city ZHVI was $652,061 in May, down 0.09 percent on the month and up 0.91 percent from $646,177 a year earlier. Zillow Research. Implication: Roswell held value on the year, so the price cuts showing up there in May were about asking prices that ran ahead of the index. (Computed: changes from the fetched values.)
- Sandy Springs' city ZHVI was $687,291 in May, up 0.02 percent on the month and 2.76 percent from $668,844 a year earlier, the strongest annual gain among the five corridor cities. Zillow Research. Implication: the inside-the-perimeter end of the corridor was outperforming the outer end, the reverse of the 2021 pattern. (Computed: changes from the fetched values.)
- Brookhaven's ZHVI was $739,738 and Dunwoody's $671,268 in May, each up 2.33 percent on the year, while Cumming's was $597,822, down 2.65 percent, and Suwanee's $598,894, down 2.77 percent. Zillow Research. Implication: the dividing line in May ran between the close-in north cities, which were up, and the Forsyth and Gwinnett exurbs, which were down. (Computed: changes from the fetched values.)
- The city of Atlanta's ZHVI was $383,307 in May, down 3.32 percent on the year; Marietta's was $469,834, down 1.72 percent; Decatur's was $281,858, down 5.19 percent. Zillow Research. Implication: the city proper was falling faster than the corridor, which matters for a client comparing an intown condo against a north-metro house. (Computed: changes from the fetched values.)
- Alpharetta's median list price on Zillow was $714,983 in May, flat on the month and down 8.45 percent from $781,000 a year earlier; Johns Creek's was $722,833, down 3.79 percent; Milton's was $1,234,150, down 5.24 percent. Zillow Research. Implication: asking prices in the three north Fulton cities were down 4 to 8 percent on the year while their values were flat to up, which says the mix of what came to market shifted toward smaller and older homes. (Computed: changes from the fetched values.)
- Roswell's median list price was $681,600 in May, up 2.5 percent on the month and 0.12 percent below a year earlier; Sandy Springs' was $545,500, up 12.5 percent on the month and 11.0 percent on the year. Zillow Research. Implication: Sandy Springs' asking prices jumped in a single month, so a buyer there in May was looking at a different listing mix than in April. (Computed: changes from the fetched values.)
- Alpharetta's seasonally adjusted median sale price on Zillow was $696,872 in May, down 3.27 percent on the month and 4.69 percent on the year; Johns Creek's was $693,540, up 3.61 percent on the year; Milton's was $1,039,606, up 12.41 percent; Roswell's was $634,163, down 5.70 percent; Sandy Springs' was $593,486, down 11.15 percent. Zillow Research. Implication: city-level sale medians swung 5 to 12 percent on the year in both directions on small monthly samples, so a client should be shown the value index and the sale median together, never one alone. (Computed: changes from the fetched values.)
- Redfin's Atlanta metro median sale price was $405,000 in May, up 1.25 percent on the year, with the median list price at $430,000, up 3.2 percent, and the median price per square foot at $190.10, down 0.3 percent. Redfin Data Center, June 2, 2026. Implication: on Redfin's closed-sale data the metro median was up on the year but price per foot was flat, so the gain was size mix rather than appreciation.
- Redfin's county medians for May were Fulton $479,495 (up 3.1 percent on the year), Forsyth $642,500 (up 2.0 percent), Cherokee $495,750 (up 1.2 percent), Cobb $445,000 (down 0.9 percent), Gwinnett $420,000 (down 1.2 percent) and DeKalb $385,000 (up 8.0 percent). Redfin Data Center. Implication: closed medians in Fulton, Forsyth and Cherokee were up on the year even as the value indexes fell, so a seller who priced to last year's closings was not wrong on the number, only on the days it would take.
- Redfin's price per square foot for May was down 2.2 percent on the year in Fulton, down 4.8 percent in Forsyth, down 3.6 percent in Gwinnett and down 1.1 percent in Cobb, while Cherokee's was up 1.1 percent and DeKalb's up 9.3 percent. Redfin Data Center. Implication: Forsyth's price per foot fell almost 5 percent while its median rose 2 percent, the clearest sign in the county data that bigger houses were selling for less per foot.
- Redfin's city medians for May were Alpharetta $785,000 (down 10.8 percent on the year), Johns Creek $733,840 (down 6.6 percent), Milton $1,154,500 (up 4.5 percent), Roswell $618,000 (down 4.9 percent) and Sandy Springs $730,000 (up 0.7 percent). Redfin Data Center. Implication: on 76 Alpharetta and 82 Johns Creek closings the May medians were well below a year earlier, which is the figure a buyer's agent would carry into a negotiation there.
- Redfin's May price per square foot was $262.90 in Alpharetta (down 4.7 percent on the year), $233.06 in Johns Creek (down 4.6 percent), $250.62 in Milton (down 4.9 percent), $237.28 in Roswell (down 5.1 percent) and $253.03 in Sandy Springs (down 0.6 percent). Redfin Data Center. Implication: price per foot fell about 5 percent across four of the five corridor cities in the same month, the one figure that moved together across the corridor and the best single comp adjustment for May.
- Redfin's May median list price was $755,000 in Alpharetta (down 7.9 percent on the year), $669,900 in Johns Creek (down 16.3 percent), $1,350,000 in Milton (up 17.4 percent), $750,000 in Roswell (up 7.1 percent) and $749,000 in Sandy Springs (up 4.0 percent). Redfin Data Center. Implication: Johns Creek's asking prices fell 16 percent on the year on Redfin's listing feed, so a seller there who priced off 2025 listings was pricing against a market that no longer existed.
- Realtor.com® Economic Research's May report put the Atlanta-Sandy Springs-Roswell median list price at $425,000, up 1.2 percent on the year, with price per square foot up 0.4 percent. Realtor.com® Economic Research, June 3, 2026. Implication: Atlanta asking prices were still up on the year in a month when the national median list price fell 2.4 percent, its steepest annual decline in Realtor.com® Economic Research's data. (Later release, published after the period.)
- Realtor.com® Economic Research's April report, published April 30, put the Atlanta median list price at $422,400, up 2.4 percent on the year, with price per square foot down 0.2 percent. Realtor.com® Economic Research, April 30, 2026. Implication: this was the Atlanta asking-price read on file through May, and it showed asking prices rising on the year even as price per foot had stopped.
- The national median list price was $429,500 in May, down 2.4 percent on the year, the seventh straight annual decline and the largest in Realtor.com® Economic Research's data since 2017; price per square foot fell 2.5 percent and declined in 35 of the top 50 metros. Realtor.com® Economic Research, June 3, 2026. Implication: nationally sellers were cutting asking prices at a record pace while Atlanta sellers were still asking more, which left Atlanta listings comparatively expensive for a relocation buyer. (Later release.)
- The S&P Cotality Case-Shiller index for Atlanta showed prices down 0.46 percent on the year in March, up 0.73 percent on the month before seasonal adjustment and down 0.09 percent after it. S&P Dow Jones Indices, May 26, 2026. Implication: the repeat-sales index, which strips out mix, put Atlanta at roughly flat on the year, between Zillow's minus 2 percent and the MLS medians' plus 1 to 2 percent.
- The national Case-Shiller index rose 0.7 percent on the year in March, the 20-city composite 0.8 percent and the 10-city 1.4 percent; Chicago led at 6.1 percent and Seattle trailed at minus 2.5 percent. S&P Dow Jones Indices, May 26, 2026. Implication: Atlanta's minus 0.46 percent was among the declining cities but a small decline next to Seattle, Denver and Tampa.
- S&P's Nicholas Godec said "more than half of the 20 major U.S. housing markets recorded year-over-year price declines in March, reflecting a broadening and deepening housing slowdown." S&P Dow Jones Indices, May 26, 2026. Implication: Atlanta's softness was no longer an outlier by late May; it was the national pattern.
- FHFA's purchase-only index for the Atlanta-Sandy Springs-Roswell metro was down 1.08 percent on the year in the first quarter of 2026, up 1.40 percent on the quarter and up 39.28 percent over five years. FHFA, May 26, 2026. Implication: a seller who bought five years ago still held close to 40 percent appreciation on the conforming-loan index, which is the number that makes a one-year 1 percent dip tolerable.
- FHFA's index for Georgia rose 0.11 percent on the year in the first quarter, ranking 42nd among the states, and the South Atlantic division rose 0.51 percent on the year and 0.50 percent on the quarter. FHFA, May 26, 2026. Implication: Georgia was flat on the year on the conforming index, so Atlanta's decline was being offset by the rest of the state.
- FHFA's national index rose 0.1 percent in March and 1.7 percent on the year, with the seasonally adjusted purchase-only index at 441.5 and 65 of the 100 largest metros posting annual gains. FHFA, May 26, 2026. Implication: Atlanta was one of the 35 large metros with an annual decline, which a relocation buyer from a rising metro will notice.
- The national existing-home median price was $417,700 in April, up 0.9 percent on the year, the 34th straight annual gain; the single-family median was $422,300, up 1.0 percent. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: the national median was still rising on the year, so Atlanta's flat-to-down read was a local story, not a national one. (The association's own April page now redirects to its latest release; two named outlets reporting the May 11 release were used.)
- The median new-home sale price nationally was $422,500 in April, up 8.0 percent on the month and 2.2 percent on the year, with the average at $508,800. U.S. Census Bureau and HUD, May 28, 2026. Implication: the national new-home median bounced back above the resale median in April after March's post-2021 low, so the builder discount window narrowed.
- The Georgia MLS median sale price for the 12-county metro was $417,000 in April, up 4.3 percent from $399,990 in March, against $410,000 a year earlier. The Georgia MLS release, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: on the MLS closings read on file in May, the metro median was up about 1.7 percent on the year, consistent with the Zillow sale median and inconsistent with the value index. (Computed: annual change from the two fetched values; the article's own sign on that change is inconsistent.)
- The Georgia MLS median sale price for May was $418,995, up 1 percent on the month and 2.1 percent on the year. The Georgia MLS release, as reported by Atlanta Agent Magazine, June 15, 2026. Implication: the metro's May closings came in 2 percent above last May's, so the month's record-pace sales were not bought with price concessions at the median. (Later release.)
- The Atlanta REALTORS® Association's March brief, published May 7, put the 11-county single-family median at $418,000, down 1.6 percent on the year, and the average at $525,500, down 0.5 percent. Atlanta REALTORS® Association, May 7, 2026. Implication: the association's own read in May had the median down on the year, the only MLS-based figure of the month that did.
- The association's April brief, published June 3, put the median at $436,000 and the average at $564,500; its May brief, published July 8, put the May median at $440,000 and the average at $553,500. Atlanta REALTORS® Association, June 3, 2026. Implication: the association's May median landed $22,000 above its March median, a seasonal lift a seller should not read as appreciation. (Later releases.)
- The national ZHVI was $368,720 in May, up 0.6 percent on the month and 0.8 percent on the year. Zillow Research, June 4, 2026. Implication: Atlanta's minus 2.0 percent sat almost 3 points below the national value trend. (Later release.)
- The Atlanta Fed's May Beige Book reported that in the Sixth District "home prices remained flat or declined modestly," with the steepest drops in southwest Florida. Federal Reserve Board, June 3, 2026. Implication: the regional Fed's contacts described Georgia's price picture as flat to modestly down, which matches the index data and not the MLS medians. (Later release covering the period.)
VII The economy and the cycle
- The Consumer Price Index rose 0.6 percent in April after 0.9 percent in March and 3.8 percent on the year, with core up 0.4 percent on the month and 2.8 percent on the year. Bureau of Labor Statistics, May 12, 2026. Implication: the May 12 print was the single macro event that set the month's rate path; the 10-year rose 21 basis points over the five trading days that followed.
- The energy index rose 3.8 percent in April, "accounting for over forty percent of the monthly all items increase," with gasoline up 5.4 percent on the month and 28.4 percent on the year and energy up 17.9 percent on the year; food rose 0.5 percent on the month and 3.2 percent on the year. Bureau of Labor Statistics, May 12, 2026. Implication: the inflation that moved rates was an oil shock, so the rate relief a buyer hoped for depended on the Strait of Hormuz, not on the Fed.
- The CPI-U index level was 332.407 in April and 333.979 in May, a 0.47 percent monthly rise, against 320.620 in May 2025. Bureau of Labor Statistics via FRED. Implication: on the current series the May print, released June 10, ran about 4.2 percent above a year earlier, so the inflation pressure on rates did not ease in June's first data. (Computed: changes from the fetched index levels; later data noted as such.)
- Nonfarm payrolls rose 115,000 in April and the unemployment rate held at 4.3 percent, with February revised down 23,000 to minus 156,000 and March revised up 7,000 to 185,000. Bureau of Labor Statistics, May 8, 2026. Implication: the labor read on file in May was soft enough to keep a cut in the conversation, which is why the May 12 inflation print was such a reversal.
- Average hourly earnings rose 6 cents to $37.41 in April, up 0.2 percent on the month and 3.6 percent on the year; the participation rate was 61.8 percent and the employment-population ratio 59.1 percent. Bureau of Labor Statistics, May 8, 2026. Implication: wages grew 3.6 percent against 3.8 percent inflation, so the typical buyer's real income fell in the month the payment rose.
- Health care added 37,000 jobs in April, construction 9,000 and the federal government lost 9,000. Bureau of Labor Statistics, May 8, 2026. Implication: construction was still hiring in April even as single-family starts fell, which argues the builder pullback was a demand choice rather than a capacity constraint.
- Zillow's economists noted that April's report left three-month average payroll growth at 48,000 a month, that 188,000 people left the labor force, and that part-time work for economic reasons jumped 445,000 to 4.9 million. Zillow Research, May 8, 2026. Implication: the household survey underneath the headline was weaker than the payroll number, which is the kind of softening that shows up in buyer confidence before it shows up in unemployment.
- The May jobs report, released June 5, showed payrolls up 172,000 and revised April to 179,000 from 115,000 and March to 214,000 from 185,000, with unemployment unchanged at 4.3 percent and wages up 3.4 percent on the year. Zillow Research, June 5, 2026. Implication: April's soft 115,000 was later revised to a solid 179,000, so the "weakening labor market" case that stood during May was revised away a week after the month ended. (Later release and revision, noted as such.)
- Georgia's unemployment rate was 3.5 percent in April, unchanged on the month and 0.8 points below the national rate, with the labor force at a record 5,461,455 and employment at a record 5,271,138. Georgia Department of Labor, May 21, 2026. Implication: Georgia's labor market was at record employment in the month, so the metro's record sales pace had income behind it.
- Georgia's nonfarm jobs were 4,996,000 in April, up 7,100 on the month and 2,700 on the year, with real estate, rental and leasing adding 2,000 jobs and construction losing 1,100. Georgia Department of Labor, May 21, 2026. Implication: statewide job growth on the year was essentially zero, so the record employment was coming from population, not from a hiring boom.
- Georgia Labor Commissioner's release said "more Georgians are working today than at any point in our state's history, with record highs in both our labor force and employment." Georgia Department of Labor, May 21, 2026. Implication: the state's own framing in May was a full-employment economy, which supports the demand side of the corridor market even at a 6.5 percent rate.
- BLS's state release confirmed Georgia's seasonally adjusted unemployment rate at 3.5 percent in April, not significantly different from a year earlier, with nonfarm payrolls at 4,996,000, essentially unchanged on the month and the year; the national rate was 4.3 percent. Bureau of Labor Statistics, May 22, 2026. Implication: Georgia sat 0.8 points below the national unemployment rate, the cushion that keeps a local downturn from becoming a distress cycle.
- The Atlanta-Sandy Springs-Roswell metro unemployment rate (not seasonally adjusted) was 2.8 percent in April and 3.2 percent in May, against 3.2 percent in May 2025. Bureau of Labor Statistics via FRED. Implication: the metro's May unemployment rate matched a year earlier at 3.2 percent, a full point under the nation.
- Atlanta metro nonfarm employment was 3,124,800 in May, down from 3,129,100 in April and up about 9,900, or 0.3 percent, from 3,114,900 in May 2025. Bureau of Labor Statistics via FRED. Implication: the metro added fewer than 10,000 jobs over the year, so the relocation-driven demand story was running on announcements more than on payrolls. (Computed: annual change from the fetched values.)
- Georgia's seasonally adjusted unemployment rate was 3.4 percent in May and statewide nonfarm employment 4,999,000, up 5,200 on the year. Bureau of Labor Statistics via FRED. Implication: the state ticked down a tenth in May and stayed under 3.5 percent through the rate spike. (Later data for the month itself; computed annual change.)
- Advance retail and food services sales were $757.1 billion in April, up 0.5 percent on the month and 4.9 percent on the year, after a 1.6 percent March gain. U.S. Census Bureau, May 14, 2026. Implication: consumers were still spending through the oil shock in April, which gave the Fed no reason to cut and the bond market a reason to sell.
- Real GDP grew at a 1.6 percent annual rate in the first quarter on the second estimate, revised down 0.4 point from the advance estimate, after 0.5 percent in the fourth quarter; the quarter's PCE price index rose 4.5 percent and the core index 4.4 percent. Bureau of Economic Analysis, May 28, 2026. Implication: growth was revised down and inflation was not, the combination that leaves the Fed stuck and mortgage rates high.
- The PCE price index rose 0.4 percent in April and 3.8 percent on the year, with the core index up 0.2 percent and 3.3 percent; disposable income fell 0.1 percent, spending rose 0.5 percent and the saving rate was 2.6 percent. Bureau of Economic Analysis, May 28, 2026. Implication: households spent more than their income grew and saved 2.6 percent, which is thin cover for a down payment or a closing-cost surprise.
- The Conference Board's Consumer Confidence Index slipped to 93.1 in May from a revised 93.8, with the Present Situation Index down 3.2 points to 121.2, the lowest since February, and the Expectations Index up 1.0 to 74.4; median year-ahead inflation expectations held at 6.2 percent. The Conference Board release, as reported by PNC Economics, May 26, 2026. Implication: consumers rated their present situation worse in May and expected 6 percent inflation, a combination that delays big-ticket decisions like a home purchase. (The Conference Board's own May page now redirects to its latest release; a dated bank research note on the release was used.)
- The same note reported that around two-thirds of consumers said they were reducing spending because of rising prices and 61 percent were buying fewer items. The Conference Board release, as reported by PNC Economics, May 26, 2026. Implication: a buyer pool that is cutting back on groceries is a buyer pool that will push harder on price and concessions.
- The University of Michigan's Index of Consumer Sentiment fell to 44.8 in May from 49.8 in April, the lowest reading in the 2025 and 2026 observations pulled, with year-ahead inflation expectations at 4.8 percent. University of Michigan via FRED. Implication: sentiment hit a record low in the same month the metro closed a record number of homes, which says the May closings were April's decisions.
- Brent crude closed at $103.70 on May 6, rose to $116.73 on May 18, and fell to $92.88 by May 29, a 20 percent drop from the month's high. U.S. Energy Information Administration via FRED. Implication: the oil peak on May 18 and the 10-year peak on May 19 were one day apart, and both unwound into month end, which is the entire mechanism behind June's first rate decline. (Computed: the percent drop from the fetched values.)
- The national average price of regular gasoline was $4.452 a gallon the week of May 4, $4.500 the week of May 11, $4.490 the week of May 18 and $4.475 the week of May 25, against $3.12 to $3.17 in the same weeks of 2025. U.S. Energy Information Administration via FRED. Implication: gasoline held above $4.45 all month, the household-budget line that competed most directly with a higher mortgage payment.
- The Atlanta Fed's May Beige Book said "the Sixth District economy expanded at a modest pace," that employment was flat to slightly lower with firms holding headcount or reducing through attrition, that annual wage increases ran 2 to 3 percent, and that "prices and nonlabor costs rose moderately" on fuel, airfare and shipping. Federal Reserve Board, June 3, 2026. Implication: the regional economy was growing but not hiring, and wages at 2 to 3 percent were trailing inflation near 4, which caps how much more a Georgia buyer can stretch. (Later release covering the period.)
- The Atlanta Regional Commission's 2025 estimates, published August 2025 and still the current county figures through May 2026, put the 11-county region's annual gain at 64,400 residents, with Fulton adding 18,800, Gwinnett 15,200, Cherokee 7,100 and Forsyth 6,700, and Forsyth and Cherokee the fastest-growing at 2.4 percent each. The ARC release, as reported by Urbanize Atlanta, August 13, 2025. Implication: the two counties with the metro's fastest inventory growth in May were also its fastest-growing by population, which is the demand that absorbs that inventory over time. (Standing context, not a May 2026 release; the May 17 brief dated these figures to May 2026.)
- ARC's next estimates, published September 1, 2026, put the following year's gain at 53,690, or 1.0 percent, down from 1.2 percent, with Fulton leading at 1.9 percent and Forsyth second at 1.5 percent, and the region's building permits in 2025 at 24,015, down 18.5 percent. Atlanta Regional Commission, September 1, 2026. Implication: population growth slowed and permits fell by nearly a fifth in the year that included May, so both the demand and the supply pipelines were decelerating together. (Later release.)
- Hisense announced on April 15 that it was moving its U.S. headquarters from Suwanee to Windward Parkway in Alpharetta, with its president calling the move "a defining moment for Hisense USA." Global Atlanta, April 15, 2026. Implication: the relocation that stood as the corridor's newest corporate anchor during May was an intra-metro move from Gwinnett to north Fulton, not new jobs to the region. (April event; the square footage and headcount quoted in the May briefs are not on the fetched page.)
- The 2-year Treasury averaged 4.00 percent in May, up from 3.80 percent in April and 3.54 percent in January, while the 10-year to 2-year spread ended May at 0.47 points. US Treasury via FRED. Implication: the market priced out roughly two quarter-point cuts between January and May, which is the repricing a lender's client felt as a higher quote.
- No Providentia cycle snapshot was recorded for May 2026; no phase is asserted here, and the macro prints above stand in for it. The Record's own convention; no external source. Implication: the cycle read for May rests on the divergence between the metro's record closings and the national contract reversal, not on a composite score.
III Inventory and supply
- Atlanta metro active for-sale inventory reached 32,227 in May, up 6.23 percent from 30,336 in April and 2.45 percent from 31,457 in May 2025, the highest reading of 2026 to that point. Zillow Research. Implication: a buyer had about 1,900 more homes to choose from than in April and 770 more than a year earlier, enough to keep negotiating leverage even in the month's record sales pace. (Computed: changes from the fetched values.)
- Metro inventory's two-year path shows May 2026's 32,227 still below the 34,140 peak of August 2025 and above every reading from October 2024 through March 2025. Zillow Research. Implication: the metro was about 6 percent below its 2025 inventory peak in May, so the "inventory is back" story had already happened; May was a seasonal climb, not a new high.
- Metro new listings rose to 8,994 in May, up 8.13 percent from 8,318 in April but 3.66 percent below 9,336 in May 2025. Zillow Research. Implication: fewer owners listed than last May, so the inventory build came from slower absorption of what was already listed, not from a seller rush. (Computed: changes from the fetched values.)
- Months of supply computed from metro inventory against the sales-count nowcast was 3.83 in May, against 3.80 in April and 3.90 in March. Computed from Zillow Research inventory and sales count nowcast. Implication: supply and sales rose together, leaving the balance where it had been since March, under four months, which is still a market where a well-priced home moves.
- Fulton County's active inventory was 6,906 in May, up 6.44 percent on the month but 3.52 percent below a year earlier; DeKalb's was 3,837, down 11.53 percent on the year. Zillow Research. Implication: the two inner counties had less for sale than a year ago while the metro had more, so the inventory growth was happening in the outer ring. (Computed: changes from the fetched values.)
- Forsyth County's active inventory was 1,427 in May, up 15.08 percent on the month and 16.21 percent on the year; Cherokee's was 1,653, up 8.82 percent on the month and 24.10 percent on the year. Zillow Research. Implication: the two fastest-growing north counties also had the fastest inventory growth, so a Forsyth or Cherokee seller faced a quarter more competition than a year earlier. (Computed: changes from the fetched values.)
- Cobb County's active inventory was 3,557 in May, up 6.98 percent on the year, and Gwinnett's was 4,092, up 8.17 percent. Zillow Research. Implication: the big suburban counties were adding supply at a mid-single-digit pace, between the inner counties' declines and the exurbs' double-digit gains. (Computed: changes from the fetched values.)
- Alpharetta's active inventory was 428 in May, up 15.99 percent on the month and 22.99 percent from 348 a year earlier; Johns Creek's was 319, up 26.59 percent on the year; Roswell's was 480, up 28.00 percent; Milton's was 227, up 10.19 percent; Sandy Springs' was 587, up 4.82 percent. Zillow Research. Implication: Roswell, Johns Creek and Alpharetta each had roughly a quarter more homes for sale than a year earlier, the corridor's own version of the outer-ring inventory build. (Computed: changes from the fetched values.)
- Cumming's active inventory was 1,125 in May, up 15.86 percent on the year; Woodstock's was 612, up 26.71 percent; Canton's was 650, up 25.97 percent; Marietta's was 1,421, up 12.33 percent. Zillow Research. Implication: the Forsyth and Cherokee county seats carried the same quarter-higher inventory as the north Fulton cities. (Computed: changes from the fetched values.)
- The city of Atlanta's active inventory was 4,594 in May, down 8.43 percent on the year, and Decatur's was 1,144, down 17.99 percent. Zillow Research. Implication: intown supply was shrinking while suburban supply grew, the opposite of the pandemic pattern. (Computed: changes from the fetched values.)
- New listings in May were 161 in Alpharetta (up 12.59 percent on the year), 116 in Johns Creek (flat), 72 in Milton (down 8.86 percent), 183 in Roswell (up 11.59 percent) and 178 in Sandy Springs (down 2.73 percent). Zillow Research. Implication: Alpharetta and Roswell owners were listing more than a year ago while Milton owners were listing less, which is consistent with Milton's firmer value index. (Computed: changes from the fetched values.)
- Redfin's Atlanta metro feed counted 26,672 active listings in May, down 15.2 percent on the year, with 9,679 new listings, down 13.3 percent, and 3.7 months of supply. Redfin Data Center. Implication: Redfin's listing feed showed shrinking supply in the same month Zillow's showed growth, so the two most-cited public inventory counts for Atlanta disagreed on direction and a professional should name which one is being used.
- Redfin's county feeds for May showed Fulton inventory at 5,597 (down 17.9 percent on the year), Forsyth 1,359 (down 4.8 percent), Cherokee 1,378 (up 1.2 percent), Cobb 2,734 (down 8.3 percent), Gwinnett 3,531 (down 5.4 percent) and DeKalb 2,742 (down 31.6 percent). Redfin Data Center. Implication: even on Redfin's shrinking feed, Cherokee was the one north-metro county with more for sale than a year earlier.
- Redfin's months of supply for May were 4.5 in Fulton, 4.3 in Forsyth, 3.2 in Cherokee, 2.8 in Cobb, 3.8 in Gwinnett and 3.5 in DeKalb. Redfin Data Center. Implication: Fulton and Forsyth carried a month and a half more supply than Cobb, which is where a buyer's leverage was greatest in the corridor.
- Redfin's city feeds for May showed Alpharetta inventory at 331 (up 5.1 percent on the year) with 4.4 months of supply, Johns Creek at 237 (up 23.4 percent) with 2.9 months, Milton at 185 (down 8.0 percent) with 4.0 months, Roswell at 323 (up 13.3 percent) with 2.7 months and Sandy Springs at 420 (down 8.9 percent) with 3.2 months. Redfin Data Center. Implication: Alpharetta and Milton sat above four months of supply while Roswell and Johns Creek sat under three, so the same corridor held both a buyer's and a seller's market in May.
- Redfin's new listings for May were 151 in Alpharetta (up 12.7 percent on the year), 108 in Johns Creek (down 13.6 percent), 77 in Milton (down 13.5 percent), 158 in Roswell (up 1.9 percent) and 197 in Sandy Springs (up 9.4 percent). Redfin Data Center. Implication: Johns Creek and Milton owners held back listings in May on both public feeds, the one corridor pattern Zillow and Redfin agreed on.
- Realtor.com® Economic Research's May report put Atlanta's active listing count up 2.6 percent on the year with new listings down 6.1 percent and median days on market up 3 days. Realtor.com® Economic Research, June 3, 2026. Implication: on the third public feed Atlanta's new listings fell on the year, which with Zillow's minus 3.7 percent makes two of three feeds agreeing that fewer owners listed. (Later release.)
- Realtor.com® Economic Research's April report, on file through May, put Atlanta's active listings up 4.3 percent on the year, new listings down 4.1 percent and days on market up 3 days. Realtor.com® Economic Research, April 30, 2026. Implication: Atlanta's inventory growth slowed from 4.3 to 2.6 percent between the April and May reports, the same deceleration Zillow's feed showed.
- National active listings on Realtor.com® Economic Research were 1,058,693 in May, up 5.6 percent on the month and 2.2 percent on the year, 11.6 percent below the 2017 to 2019 norm, with the South up just 0.3 percent on the year. Realtor.com® Economic Research, June 3, 2026. Implication: the South's inventory growth had effectively stopped by May, so the Sun Belt supply wave of 2024 and 2025 was over. (Later release.)
- National new listings on Realtor.com® Economic Research were 474,976 in May, up 2.1 percent on the year and the highest May since 2022, while the median time on market was 52 days, one day slower than a year earlier. Realtor.com® Economic Research, June 3, 2026. Implication: the national seller came back in May; the Atlanta seller did not. (Later release.)
- Zillow counted 1.36 million homes for sale nationally in May, up 4.6 percent on the month and 1 percent on the year, the 30th straight month of annual growth, with new listings at 422,956, down 4.1 percent on the year. Zillow Research, June 4, 2026. Implication: Zillow's economists said weekly data "suggests it could flatline in the next four weeks," so the inventory tailwind buyers had enjoyed since 2023 was ending. (Later release.)
- Zillow's April report, published May 6, counted 1.3 million homes for sale nationally, up 3.7 percent on the year, with new listings at 426,356, up 2.1 percent on the year and 10.7 percent on the month. Zillow Research, May 6, 2026. Implication: on the read available in May, new listings were growing faster than sales for the first time in 2026, which Zillow called a pause in the spring rebound.
- Housing starts ran at a 1,465,000 annual rate in April, down 2.8 percent from a revised 1,507,000 in March and up 4.6 percent from April 2025; single-family starts fell 9.0 percent to 930,000 while 5-plus-unit starts were 529,000. U.S. Census Bureau and HUD, May 21, 2026. Implication: builders pulled back on single-family groundbreaking in the month resale inventory was climbing, which thins the new-build pipeline competing with corridor resale listings in late 2026.
- Building permits rose 5.8 percent to a 1,442,000 annual rate in April, 0.2 percent below April 2025, while single-family permits fell 2.6 percent to 872,000. U.S. Census Bureau and HUD, May 21, 2026. Implication: the permit rise was multifamily; single-family permits fell for a second month, so the rental pipeline kept growing while the for-sale pipeline shrank.
- In the South, April permits ran at 756,000 (518,000 single-family) and starts at 723,000 (548,000 single-family); national completions were 1,449,000, 903,000 of them single-family. U.S. Census Bureau and HUD, May 21, 2026. Implication: the South accounted for roughly half of national starts and nearly 60 percent of single-family starts, so the national single-family pullback was largely a Southern pullback.
- New single-family home sales ran at a 622,000 annual rate in April, down 6.2 percent on the month and 11.3 percent on the year, with 489,000 new homes for sale, a 9.4-month supply. U.S. Census Bureau and HUD, May 28, 2026. Implication: a nine-month supply of new homes is a buyer's market by any definition, and it is where a corridor buyer should look for incentives.
- New-home sales in the South ran at a 370,000 annual rate in April, down 9.8 percent on the month and 14.7 percent on the year. U.S. Census Bureau and HUD, May 28, 2026. Implication: Southern builders sold 15 percent fewer homes than a year earlier, which is why their incentives were widening into May.
- Zillow's raw count of metro Atlanta new-construction sales was 1,160 in May, up from 1,113 in April and 8.7 percent below 1,271 in May 2025. Zillow Research. Implication: new-build closings in the metro were down about 9 percent on the year while resale closings hit a record, so the spring's buyers chose resale. (Computed: annual change from the fetched values.)
- The Atlanta REALTORS® Association's March brief, published May 7, counted 17,723 active listings across the 11-county area, up 5.1 percent on the year, 9,161 new listings, flat on the year, and 4.0 months of supply. Atlanta REALTORS® Association, May 7, 2026. Implication: the association's supply read on file during May was four months, the same balance as Zillow's computed figure.
- The association's April brief counted 19,224 active listings, 9,683 new listings and 4.4 months of supply; its May brief counted 19,715 active listings, 8,833 new listings and 4.5 months of supply. Atlanta REALTORS® Association, June 3, 2026. Implication: on the association's own count, May's new listings fell 850 below April's and supply crept to four and a half months. (Later releases.)
- Redfin's national count of active listings was 1,493,419 for the four weeks ending May 17, up 1.4 percent on the year, with new listings down 0.5 percent on the year. Redfin, May 21, 2026. Implication: national supply growth had slowed to about 1 percent by mid-May, so the "more choice every month" era for buyers was ending.
- Redfin's new listings rose 0.2 percent on the week for the four weeks ending May 24, the first weekly increase after four declines, and active listings were up 0.8 percent on the year. Redfin, May 28, 2026. Implication: seller participation ticked up at the end of May just as buyer demand fell, which is the setup that widens negotiating room in June.
- The Atlanta Fed's May Beige Book reported that in the Sixth District "existing inventory shrank as delistings rose" and that builders saw "sharp traffic declines and growing buyer hesitancy," with one builder saying it "can't give a house away." Federal Reserve Board, June 3, 2026. Implication: the regional Fed's contacts described sellers pulling listings and builders unable to move spec homes even with incentives, the two supply stories a corridor buyer can use. (Later release covering the period.)
IV Buyer and seller activity
- Zillow's sales-count nowcast for the Atlanta metro reached 8,421 in May, up 5.42 percent from 7,988 in April and 3.08 percent from 8,169 in May 2025, the highest reading in the series from October 2024 through August 2026. Zillow Research. Implication: the metro closed more homes in May than in any month of the prior two years, so the rate spike did not reach the closing table in May; what closed had gone under contract in April. (Computed: changes from the fetched values; the nowcast is an estimate, not a recorded count.)
- Median days to pending in the metro fell to 33 in May from 37 in April, against 29 in May 2025; the series' fastest reading in the October 2024 to August 2026 window was that 29, and its slowest was 61 in January 2026. Zillow Research. Implication: a correctly priced metro listing went pending in about a month, four days slower than last May and four weeks faster than in January.
- Mean days to pending in the metro was 69 in May, down from 76 in April, against 59 a year earlier. Zillow Research. Implication: the mean ran more than twice the median, so a long tail of stale listings sat behind the fast ones, and ten days longer than last year.
- The metro's mean sale-to-list ratio was 98.30 percent in May, up from 98.08 percent in April and against 98.51 percent in May 2025; the series' 2026 low was January's 97.48 percent. Zillow Research. Implication: the average closing came in 1.7 percent under list, a smaller discount than in winter but a larger one than a year earlier, so a seller should expect to give up about $6,500 on a $380,000 list. (Computed: the dollar figure from the ratio and the median list price.)
- The metro's median sale-to-list ratio was 99.25 percent in May, up from 99.05 percent in April and against 99.67 percent a year earlier. Zillow Research. Implication: the typical home closed within a point of list, so the discounting was concentrated in the stale tail, not spread across every sale.
- The share of metro listings with a price cut rose to 26.64 percent in May from 25.50 percent in April, still below 28.51 percent in May 2025. Zillow Research. Implication: roughly one listing in four had cut its price, fewer than last May, which says sellers were pricing closer to the market up front this year.
- The share of metro homes sold above list was 21.81 percent in May, up from 20.11 percent in April and against 23.47 percent in May 2025. Zillow Research. Implication: about one closing in five still went over asking, so a well-presented listing in the right pocket could still draw competition.
- Zillow's Market Heat Index for the Atlanta metro read 48 in May, down from 50 in April and against 47 in May 2025, on Zillow's 0 to 100 scale. Zillow Research. Implication: Zillow's own composite put Atlanta a shade below its 50 midpoint in May, within a point of where it sat a year earlier.
- Median days to pending in May was 22 in Alpharetta (down from 38 in April, against 17 a year earlier), 21 in Johns Creek (against 17), 27 in Milton (against 19), 19 in Roswell (against 16) and 24 in Sandy Springs (against 21). Zillow Research. Implication: every corridor city went pending faster than the metro's 33 days, and Alpharetta's time to pending nearly halved from April, which is the clearest sign the corridor's spring arrived in May.
- Median days to pending in May was 36 in Cumming (against 25 a year earlier), 28 in Woodstock (against 17), 33 in Canton (against 27), 28 in Marietta (against 18) and 38 in the city of Atlanta (against 36). Zillow Research. Implication: the Forsyth and Cherokee county seats were ten to eleven days slower than a year earlier, the widest slowdown in the north metro.
- The mean sale-to-list ratio in May was 98.61 percent in Alpharetta (against 99.13 a year earlier), 98.55 in Johns Creek (against 99.51), 99.18 in Milton (against 98.43), 99.22 in Roswell (against 99.92) and 98.35 in Sandy Springs (against 99.18). Zillow Research. Implication: Johns Creek sellers gave up about a point more off list than a year earlier while Milton sellers gave up less, the reverse of what the asking-price data implied.
- The share of listings with a price cut in May was 25.66 percent in Alpharetta (against 27.64 a year earlier), 22.53 in Johns Creek (against 18.77), 25.30 in Milton (against 22.77), 26.92 in Roswell (against 23.34) and 24.09 in Sandy Springs (against 22.45). Zillow Research. Implication: four of the five corridor cities had more price cuts than a year earlier, with Roswell and Johns Creek up more than 3 points, so overpricing was being corrected after listing rather than before.
- The share of homes sold above list in May was 19.77 percent in Alpharetta (against 27.19 a year earlier), 29.02 in Johns Creek (against 37.17), 19.71 in Milton (against 27.22), 29.21 in Roswell (against 37.64) and 24.51 in Sandy Springs (against 27.84). Zillow Research. Implication: bidding above asking fell about 8 points in every corridor city on the year, the single clearest measure that the multiple-offer market of 2025 had cooled.
- Dunwoody's mean sale-to-list ratio was 100.74 percent in May with 42.15 percent of homes sold above list, while Cumming's was 98.13 percent with 13.20 percent above list. Zillow Research. Implication: the hottest and coolest pockets of the north metro were forty minutes apart, so a client's experience depended on the submarket, not the metro headline.
- Redfin's Atlanta metro feed counted 7,142 homes sold in May, down 6.6 percent on the year, with 8,020 pending sales, down 8.0 percent, a median of 43 days on market, two days slower than a year earlier. Redfin Data Center. Implication: Redfin's MLS-based closings were down on the year in the same month Zillow's nowcast hit a record, so the "busiest month on record" claim rests on one estimate and should be stated that way.
- Redfin's Atlanta metro average sale-to-list ratio was 98.41 percent in May, with 22.1 percent of homes sold above list (down 1.3 points on the year), 32.6 percent of listings with a price drop (up 2.1 points) and 24.6 percent of listings off market within two weeks. Redfin Data Center. Implication: one listing in three cut its price on Redfin's count, but one in four was gone inside two weeks, which is the two-speed market a seller has to pick a lane in.
- Redfin's county feeds for May counted 1,247 homes sold in Fulton (down 11.4 percent on the year), 316 in Forsyth (down 11.7 percent), 426 in Cherokee (up 5.4 percent), 963 in Cobb (up 1.2 percent), 932 in Gwinnett (down 12.6 percent) and 783 in DeKalb (down 7.0 percent). Redfin Data Center. Implication: Fulton and Forsyth closings fell more than a tenth on the year while Cherokee's rose, so the county with the most resilient values was also the one still gaining volume.
- Redfin's pending sales in May were down 8.0 percent on the year in Fulton and up 2.6 percent in Forsyth and 7.3 percent in Cherokee; median days on market were 42 in Fulton (up 1), 38 in Forsyth (up 3) and 34 in Cherokee (up 4). Redfin Data Center. Implication: the two outer counties were still signing more contracts than a year earlier in May, which argues that the metro's contract slump was a Fulton, Gwinnett and DeKalb story.
- Redfin's May price-drop shares were 31.1 percent in Fulton, 35.3 percent in Forsyth, 33.9 percent in Cherokee and 32.6 percent in Cobb, with the share sold above list at 24.4 percent in Fulton, 17.4 percent in Forsyth and 20.0 percent in Cherokee. Redfin Data Center. Implication: Forsyth had the most price drops and the fewest over-asking sales of the north counties, consistent with its 4.8 percent drop in price per foot.
- Redfin's city feeds for May counted 76 homes sold in Alpharetta (down 10.6 percent on the year), 82 in Johns Creek (down 21.9 percent), 46 in Milton (down 20.7 percent), 120 in Roswell (down 5.5 percent) and 131 in Sandy Springs (up 4.8 percent). Redfin Data Center. Implication: Johns Creek and Milton closed a fifth fewer homes than a year earlier, so the corridor's record-pace headline did not apply to its two highest-priced cities.
- Redfin's May pending sales were 102 in Alpharetta (up 1.0 percent on the year), 108 in Johns Creek (down 10 percent), 68 in Milton (flat), 131 in Roswell (down 3.7 percent) and 148 in Sandy Springs (up 12.1 percent). Redfin Data Center. Implication: Sandy Springs was the one corridor city signing meaningfully more contracts than a year earlier, which matches its value index lead.
- Redfin's median days on market in May was 24 in Alpharetta (four faster than a year earlier), 27 in Johns Creek (two faster), 33 in Milton (seven slower), 28 in Roswell (one slower) and 33 in Sandy Springs (two faster). Redfin Data Center. Implication: Milton was the one corridor city where homes sat longer than last year, a week longer, so a Milton seller's pricing margin for error was the thinnest.
- Redfin's May price-drop shares were 29.3 percent in Alpharetta, 37.6 percent in Johns Creek, 27.6 percent in Milton, 35.0 percent in Roswell and 34.5 percent in Sandy Springs, with sale-to-list averages between 98.7 and 99.2 percent. Redfin Data Center. Implication: more than a third of Johns Creek, Roswell and Sandy Springs listings cut their price in May even though closings came within a point and a half of list, so the cut was the path to the close, not a sign of distress.
- Redfin's city of Atlanta feed counted 602 homes sold in May, down 8.6 percent on the year, 672 pending sales, down 5.8 percent, a median of 48 days on market and 5.3 months of supply; Marietta's median days on market was 51, eleven days slower than a year earlier. Redfin Data Center. Implication: intown Atlanta carried five months of supply and seven weeks to sell, a slower market than any corridor city.
- Georgia MLS counted 5,114 closed sales across its 12-county metro area in April, up 1.1 percent on the month and 0.6 percent below April 2025's 5,146. The Georgia MLS release, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: the MLS closings read on file in May had the metro flat on the year, not the 5 percent decline some May 5 coverage reported.
- Georgia MLS counted 4,810 homes going under contract in April, down 1.8 percent on the month and 21.8 percent below April 2025's 6,147. The Georgia MLS release, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: the metro's contract pipeline was more than a fifth thinner than a year earlier going into May, the number that warned June's closings would not hold May's pace. (The source is Georgia MLS; several May briefs attributed the figure to FMLS.)
- Georgia MLS's John Ryan said on May 5 that "the data shows a market in transition" with "a clear shift away from the urgency that defined recent years, as buyers are taking more time, weighing their options and approaching decisions with greater caution," and on May 8 that "as inventory builds, the market is giving buyers more leverage." As reported by Atlanta News First, May 5, 2026. Implication: the MLS's own framing in May was buyer leverage, which is the posture a listing agent should have assumed in pricing conversations.
- Georgia MLS counted 5,590 closed sales in May, up 7.6 percent on the month and 0.6 percent on the year, and 4,512 homes going under contract, down 6.2 percent on the month and 27.9 percent below May 2025's 6,259. The Georgia MLS release, as reported by Atlanta Agent Magazine, June 15, 2026. Implication: May's closings were flat on the year while contract signings fell 28 percent, the widest gap between the two in the series, which is the clearest single warning about the summer pipeline. (Later release.)
- The Atlanta REALTORS® Association's March brief counted 4,670 single-family sales across 11 counties, up 4.0 percent on the year and 29.2 percent on the month. Atlanta REALTORS® Association, May 7, 2026. Implication: the association's read on file in May had closings up on the year, the only metro closings figure of the spring that did.
- The association's April brief counted 4,892 residential sales with 19 average days on market; its May brief counted 5,323 sales with 18 average days on market and 22 cumulative days. Atlanta REALTORS® Association, June 3, 2026. Implication: on the association's measure a May listing averaged under three weeks to contract, the fastest reading of the spring. (Later releases.)
- National existing-home sales ran at a 4.02 million annual rate in April, up 0.2 percent on the month and flat on the year, with 1.47 million homes for sale, up 5.8 percent on the month, and 4.4 months of supply. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: the national resale market was running at the same pace as a year earlier, so Atlanta's record May was ahead of the country.
- The median existing home nationally sold in 32 days in April, down from 41 in March and against 29 a year earlier; first-time buyers were 33 percent of sales, cash buyers 25 percent, investors and second-home buyers 16 percent and distressed sales 2 percent. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: one buyer in four nationally paid cash, which is the competition a financed corridor buyer faced on a well-priced home.
- Existing-home sales in the South ran at a 1.87 million annual rate in April, up 0.5 percent on the month and 2.7 percent on the year, 46.5 percent of national sales. The National Association of REALTORS® release, as reported by Haver Analytics, May 11, 2026. Implication: the South was the only region with annual sales growth, so the Sun Belt was carrying the national number.
- NAR's Lawrence Yun said of April that "mortgage rates are lower from a year ago, and average income growth is outpacing home price gains." The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: the affordability case was being made on the year-over-year comparison, which the mid-May rate spike then eroded.
- The Pending Home Sales Index rose 1.4 percent in April and 3.2 percent on the year; the South fell 0.7 percent on the month but was up 4.7 percent on the year, the largest annual gain of any region, while the Northeast rose 6.6 percent on the month. National Association of REALTORS® (archived), May 19, 2026. Implication: contract activity in the South was already slipping on the month before the rate spike, so the May decline had a running start.
- Yun said on May 19 that "buyers are coming out with cautious optimism despite increasing economic uncertainty and a slight rise in mortgage rates" and that "unless supply meaningfully increases, home price growth could outpace wage growth and further erode the homeownership rate." National Association of REALTORS® (archived), May 19, 2026. Implication: the association's chief economist framed supply, not rates, as the binding constraint, a point that does not fit Atlanta's four months of supply.
- Redfin's four-week pending sales were up 7.7 percent on the year for the period ending May 3, the highest level since September 2022, with the median sale price at $394,803, up 1.9 percent, and 26.4 percent of homes sold above list. Redfin, May 7, 2026. Implication: national demand peaked in the first week of May on the back of April's rate dip.
- Redfin's four-week pending sales were up 9.6 percent on the year for the period ending May 10 at 346,104, with the median sale price at $397,740, up 2.2 percent, 42 median days on market and 26.6 percent sold above list. Redfin, May 14, 2026. Implication: the annual gain widened to nearly 10 percent in the second week of May, the strongest national demand read of 2026.
- Redfin's four-week pending sales fell 1.1 percent on the week for the period ending May 17, the first decline since early April, while still 7.8 percent above a year earlier; the median sale price was $398,653, up 2.2 percent, median days on market 41, three slower than a year earlier, and 27.1 percent sold above list. Redfin, May 21, 2026. Implication: the demand reversal landed within days of the daily rate's 6.75 percent print, which makes the May 19 spike the pivot of the spring.
- In the same May 21 report Redfin listed Atlanta among the metros with annual pending-sales declines, at minus 1.1 percent. Redfin, May 21, 2026. Implication: Atlanta's contracts were down on the year in a week the nation's were up nearly 8 percent, the national-to-local gap in one line.
- Redfin's four-week pending sales fell 1.5 percent on the week for the period ending May 24, the second straight decline, with the annual gain narrowing to 4.7 percent, median days on market at 40 and 27.5 percent sold above list; Atlanta's pending sales were down 3.1 percent on the year. Redfin, May 28, 2026. Implication: Atlanta's annual contract decline tripled in a week, from 1.1 to 3.1 percent, as the rate spike reached the local pipeline.
- Redfin wrote on May 28 that "house hunters are backing off mainly because mortgage rates are rising, with the daily average hitting a 10-month high of 6.75%," and on May 21 that "higher mortgage rates are scaring off some buyers, but that's opening the door for others." Redfin, May 28, 2026. Implication: the publisher's own read was that the buyers who stayed in gained room to negotiate, which is the buyer-side talking point for late May.
- Realtor.com® Economic Research's count of listings in pending status was up 4.3 percent on the year in May, the sixth straight annual gain, with the flow of new contract signings up 2.6 percent; the share of listings with a price cut was 17.5 percent, down 1.6 points. Realtor.com® Economic Research, June 3, 2026. Implication: Realtor.com® Economic Research's economists described sellers as "pricing to sell rather than pricing to test the market," which is the posture that kept national contracts rising. (Later release.)
- Atlanta's price-reduced share on Realtor.com® Economic Research was 20.4 percent in May, down 2.9 points on the year, after 19.5 percent in April, down 1.4 points. Realtor.com® Economic Research, June 3, 2026. Implication: on Realtor.com® Economic Research's feed Atlanta sellers cut prices less often than a year earlier, the same direction as Zillow's feed and the opposite of Redfin's.
- Realtor.com® Economic Research reported that contract cancellations in 2026 "have remained lower than in the past several years" through May. Realtor.com® Economic Research, June 3, 2026. Implication: the rate spike showed up in sentiment and in rates but not yet in deals falling apart, which is the reassurance a nervous seller under contract needed in late May. (Later release.)
- Nationally, Zillow's typical listing went pending in 18 days in May, one day slower than a year earlier, with 341,929 homes sold on the nowcast, down 2.9 percent on the year, and 23.9 percent of listings carrying a price cut. Zillow Research, June 4, 2026. Implication: the metro's 33 days to pending was nearly double the national 18, a gap that has held through the spring. (Later release.)
- Zillow's April report, published May 6, counted 323,631 national sales on the nowcast, down 0.4 percent on the year and up 7.9 percent on the month, with newly pending listings up 7.1 percent on the year and 23.5 percent of listings with a price cut. Zillow Research, May 6, 2026. Implication: the April read on file in May showed demand recovering but not yet above 2025, which was the baseline the rate spike then hit.
- The Atlanta Fed's May Beige Book reported that in the Sixth District "home sales improved slightly, primarily driven by aggressive price reductions and incentives." Federal Reserve Board, June 3, 2026. Implication: the regional Fed's contacts credited the spring's sales to seller concessions, not to buyer strength. (Later release covering the period.)
V Investors, rental, and the carrying cost
- The Zillow Observed Rent Index for the Atlanta metro (smoothed, not seasonally adjusted) was $1,828.90 in May, up 0.52 percent from $1,819.52 in April and 1.46 percent from $1,802.50 in May 2025. Zillow Research. Implication: metro rent rose about $26 a month over the year, a pace well under inflation, so a landlord's real rent was still falling. (Computed: changes from the fetched values.)
- The same rent series ran from a low of $1,764.56 in December 2024 to $1,852.99 in August 2026, and its largest monthly gain in that window was February 2025's 0.66 percent, so May's 0.52 percent was the third-largest, not the largest. Zillow Research. Implication: May's rent increase was a normal spring step, not a regime change, which matters for an investor underwriting rent growth. (Computed: monthly changes from the fetched values.)
- The seasonally adjusted metro ZORI was $1,833.07 in May, up 0.38 percent on the month and 1.45 percent on the year. Zillow Research. Implication: strip out the season and metro rent growth was running at about a 1.5 percent annual pace, the same number on both series. (Computed: changes from the fetched values.)
- The metro single-family ZORI was $2,278.30 in May, up 0.42 percent on the month and 2.30 percent on the year, while the multifamily ZORI was $1,649.03, up 0.24 percent on the month and 0.47 percent on the year. Zillow Research. Implication: house rents were growing five times faster than apartment rents in the metro, so the apartment construction wave was holding down the multifamily number and not the single-family one. (Computed: changes from the fetched values.)
- County rent indexes for May were Fulton $1,911.66 (up 2.36 percent on the year), DeKalb $1,747.68 (up 2.57 percent), Cherokee $2,042.94 (up 0.97 percent), Cobb $1,743.69 (up 0.84 percent), Forsyth $2,234.65 (up 0.65 percent) and Gwinnett $1,816.73 (down 0.68 percent). Zillow Research. Implication: the inner counties' rents were growing while the outer ones were flat, so an investor in Forsyth or Gwinnett was underwriting a rent that had not moved in a year. (Computed: changes from the fetched values.)
- City rent indexes for May were Johns Creek $2,539.94 (flat on the year), Milton $2,228.28 (up 0.25 percent), Alpharetta $2,131.52 (down 0.07 percent), Roswell $1,706.02 (up 0.11 percent) and Sandy Springs $1,691.12 (down 0.31 percent). Zillow Research. Implication: corridor rents were flat to the dollar on the year in all five cities, so a corridor rental's yield was being set entirely by purchase price. (Computed: changes from the fetched values.)
- The city of Atlanta's rent index was $1,915.42 in May, up 2.81 percent on the year, and Woodstock's was $2,105.14, up 1.75 percent, while Cumming's was $2,226.40, down 0.21 percent. Zillow Research. Implication: intown rent was growing fastest in the metro, which is where a yield-focused investor's rent growth case was strongest. (Computed: changes from the fetched values.)
- Zillow's April rental report, published May 27, found 59.1 percent of Atlanta rental listings offering a concession, up 4.9 points on the year, against 39.8 percent nationally, with the typical Atlanta rent at $1,825 and the income needed to afford it at $72,995. Zillow Research, May 27, 2026. Implication: three Atlanta rental listings in five were offering free rent or waived fees, so the effective rent a landlord collected was meaningfully below the headline.
- The same report put the national rental vacancy rate at 7.3 percent, up from 5.6 percent in 2021, and named Denver (68.3 percent), Charlotte (66.6 percent) and Dallas (64.2 percent) as the metros with the most concessions. Zillow Research, May 27, 2026. Implication: Atlanta's 59 percent concession share put it among the Sun Belt construction-boom metros, which is the comparison set for an investor's rent assumptions.
- Zillow's May rental report, published June 18, put the typical Atlanta rent at $1,840, up 1.5 percent on the year, with 85.0 percent of listings affordable to a median-income household, 3.2 percent listed under $1,000 and 58.8 percent offering a concession. Zillow Research, June 18, 2026. Implication: Atlanta was one of the most affordable large rental markets in the country in May, which weakens the "rents will force renters to buy" argument. (Later release.)
- Nationally 74 percent of rental listings were affordable to a median-income household in May, the highest for the month in Zillow's data, with the typical rent at $1,951, up 2 percent on the year, and 39.6 percent of listings offering a concession. Zillow Research, June 18, 2026. Implication: the national renter had more room than at any point since 2021, so the buy-versus-rent math was not being forced by rent. (Later release.)
- Zillow's April and May market reports put Atlanta's typical rent at $1,825 (up 0.3 percent on the month, 1.2 percent on the year) and then $1,840 (up 0.6 percent on the month, 1.5 percent on the year). Zillow Research, May 6, 2026. Implication: rent growth in the metro accelerated slightly through the spring, from 1.2 to 1.5 percent, the opposite direction from values.
- Rental Beast's first-quarter Atlanta report put the concession rate at 61.2 percent, up 15.5 percent from the fourth quarter, against 41.8 percent nationally, with single-family rents near $1,600, down 3 percent, and one-bedroom rents at $1,464, up 4.9 percent. Rental Beast. Implication: two independent trackers put Atlanta's concession share near 60 percent in the spring, so an investor's pro forma should carry at least a month of free rent per lease.
- The computed principal-and-interest payment on the metro median home, $1,917 at May's rate and price, exceeded the metro's $1,829 rent by about $88, wider than April's roughly $38 gap. Computed from Zillow Research median sale price and ZORI. Implication: on principal and interest alone renting the median home was cheaper than owning it for a second straight month, and adding taxes, insurance and HOA widens that further.
- Individual investors and second-home buyers were 16 percent of national existing-home purchases in April, down from 18 percent in March and against 15 percent a year earlier; cash buyers were 25 percent. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: small investors were still one buyer in six nationally, so the institutional retreat did not take the investor bid out of the market.
- The American Economic Liberties Project's April 22 report put institutional ownership at about 72,000 single-family rentals in metro Atlanta, nearly 30 percent of the metro's single-family rental market, more than double second-place Phoenix, with over 3,000 build-to-rent units delivered in 2024. American Economic Liberties Project, April 22, 2026. Implication: no metro has more to gain in owner-occupant supply from an institutional wind-down than Atlanta, and none has more rental supply competing with a small landlord.
- The same report found that institutional investors "may have been responsible for as much as 25% of the decline in homeownership in Atlanta after the Great Financial Crisis" and that nationally they "hold under 5% of national SFR market share." American Economic Liberties Project, April 22, 2026. Implication: Atlanta's institutional share runs about six times the national figure, which is why federal and state investor rules land harder here.
- Executive Order 14376 of January 20, 2026 directed HUD, FHFA, VA, USDA and GSA to stop "providing for, approving, insuring, guaranteeing, securitizing, or facilitating the acquisition by a large institutional investor of a single-family home," with an exception for build-to-rent communities "planned, permitted, financed, and constructed as rental communities." The White House, January 20, 2026. Implication: the order that stood through May cut off federal financing channels for institutional resale purchases but left Atlanta's build-to-rent pipeline untouched.
- The Census Bureau's quarterly vacancy release restated the first-quarter 2026 homeownership rate at 65.3 percent, the rental vacancy rate at 7.3 percent and the homeowner vacancy rate at 1.1 percent, with the South's rental vacancy rate the nation's highest at 9.5 percent in the second quarter. U.S. Census Bureau, July 28, 2026. Implication: nearly one Southern rental in ten sat empty, which is the vacancy an Atlanta investor should carry in a pro forma, not the 5 percent of the 2021 playbook. (Later release restating the first-quarter figures that stood during May.)
- The CPI shelter index rose 0.6 percent in April and 3.3 percent on the year, with rent of primary residence up 0.5 percent on the month and 2.8 percent on the year and owners' equivalent rent up 0.5 percent and 3.3 percent. Bureau of Labor Statistics, May 12, 2026. Implication: the government's rent measure was still running two points above Atlanta's market rent growth, because CPI rent lags the market by a year or more.
- Zillow's economists noted after the April CPI that owners' equivalent rent rose 0.53 percent against their 0.44 percent forecast and that "housing is still expected to be an overall disinflationary influence on consumer prices in 2026." Zillow Research, May 12, 2026. Implication: the shelter surprise in April CPI was part of what pushed the 10-year up in mid-May, so Atlanta's flat rents were not what the bond market was looking at.
- The national average price of regular gasoline reached $4.50 a gallon the week of May 11, up from $3.12 the same week of 2025. U.S. Energy Information Administration via FRED, May 11, 2026. Implication: a corridor commuter's fuel cost was up roughly $1.40 a gallon on the year in May, a carrying cost that competes with the mortgage payment in a household budget.
VI Credit, distress, and policy
- The FOMC held the federal funds target at 3.50 to 3.75 percent on April 29 by an 8 to 4 vote, with Stephen Miran preferring a quarter-point cut and Beth Hammack, Neel Kashkari and Lorie Logan opposing the statement's easing bias. Federal Reserve Board, April 29, 2026. Implication: the policy rate that stood through May was set by a committee split three ways, which is why the market repriced the path so sharply when inflation printed.
- The April 29 statement said "inflation is elevated, in part reflecting the recent increase in global energy prices," that "job gains have remained low, on average," and that "economic activity has been expanding at a solid pace." Federal Reserve Board, April 29, 2026. Implication: the Fed named energy as the inflation driver, so the May oil spike went straight to the rate outlook.
- The minutes of that meeting, released May 20, recorded that most participants expected cuts to become appropriate once "disinflation is firmly back on track," while a majority judged that "some policy firming would likely become appropriate if inflation were to continue to run persistently above 2 percent." Federal Reserve Board, May 20, 2026. Implication: the minutes put a hike on the table in writing, and they landed the day after the daily mortgage rate peaked.
- The minutes also noted that "the ongoing deceleration in housing services prices was likely to continue to be a source of disinflationary pressure." Federal Reserve Board, May 20, 2026. Implication: the Fed was counting on flat rents like Atlanta's to pull inflation down, which is the one housing input working in a borrower's favor.
- The effective federal funds rate averaged 3.63 percent in May, unchanged in practice since February. Federal Reserve via FRED. Implication: the short rate did not move in May; every basis point of the mortgage move came from the bond market.
- The Federal Reserve Board named Jerome Powell chair pro tempore on May 15 following the expiration of his term as chair, to serve "until Warsh is sworn in as the new chair." Federal Reserve Board, May 15, 2026. Implication: for a week in May the Fed had a caretaker chair, during the exact window the 10-year hit its high.
- Kevin Warsh took the oath of office as chairman on May 22, after Senate confirmation as a Board member on May 12 and as chairman on May 13; his term as chairman ends May 21, 2030 and the FOMC unanimously selected him as its chairman the same day. Federal Reserve Board, May 22, 2026. Implication: the June 16 to 17 meeting would be the new chair's first, the catalyst a lender's client should have been watching rather than the calendar.
- Governor Stephen Miran, the lone vote for a cut on April 29, submitted his resignation from the Board on May 14, effective when or shortly before his successor is sworn in. Federal Reserve Board, May 14, 2026. Implication: the committee's only dissent in favor of easing was on its way out in May, which tilted the June vote count toward holding.
- The April Senior Loan Officer Opinion Survey, published May 4, reported banks "left standards basically unchanged over the first quarter for most RRE loan categories" and saw "basically unchanged or weaker demand, on net, for all RRE loan categories," with "modest net shares of banks" reporting stronger demand for home equity lines of credit. Federal Reserve Board, May 4, 2026. Implication: the credit box did not tighten in the first quarter on the Fed's own survey, so a May denial was about the borrower, not the bank; the HELOC demand rise says equity-rich owners were tapping rather than selling.
- Total household debt was $18.8 trillion at the end of the first quarter, up $18 billion, with mortgage balances at $13.19 trillion, up $21 billion, and $530 billion of new mortgage originations in the quarter. Federal Reserve Bank of New York. Implication: mortgage balances barely grew in the quarter, which is what a lock-in market looks like in the aggregate.
- Home equity line balances rose $12 billion to $446 billion in the first quarter, the sixteenth straight quarterly increase. Federal Reserve Bank of New York. Implication: owners were borrowing against equity rather than moving, four years running, which is the supply-side reason corridor listings stayed scarce in Milton and Johns Creek.
- The share of outstanding debt in some stage of delinquency was 4.8 percent at the end of March, the annualized rate of mortgage balances transitioning into 90-plus-day delinquency rose to 1.5 percent from 1.4 percent, and about 59,000 consumers had a new foreclosure notation added in the quarter. Federal Reserve Bank of New York. Implication: mortgage distress was rising from a very low base, not surging; 59,000 new foreclosures nationally in a quarter is small against the 1.47 million homes listed for sale in April.
- Distressed sales were 2 percent of national existing-home sales in April, flat on the month and the year. The National Association of REALTORS® release, as reported by HousingWire, May 11, 2026. Implication: there was no distressed-inventory channel feeding the market in May, so a buyer waiting for foreclosures was waiting for something the data did not show.
- The refinance share of mortgage applications moved from 40.8 percent for the week ending May 8 to 41.9 percent for the week ending May 15 and 37.5 percent for the week ending May 22, while the ARM share went from 8.8 to 9.6 to 9.4 percent. HousingWire, May 13, 2026. Implication: borrowers answered the spike by shifting structure toward adjustable loans and by abandoning refinances, not by leaving the purchase market.
- MBA's FHA contract rate ran 30 basis points under conforming at 6.16 percent for the week ending May 8 and 34 under at 6.31 percent for the week ending May 22, while jumbo ran 2 to 3 basis points over conforming. HousingWire, May 13, 2026. Implication: an FHA borrower had a third of a point of rate advantage over conventional in May, and a jumbo borrower paid almost no premium, which widened the financing options at both ends of the corridor's price range.
- The 2026 conforming loan limit for one-unit properties stood at $832,750, up $26,250 from 2025, with the high-cost ceiling at $1,249,125, set by a 3.26 percent rise in the FHFA index from the third quarter of 2024 to the third quarter of 2025. FHFA, November 25, 2025. Implication: Redfin's May medians in Alpharetta ($785,000), Johns Creek ($733,840), Roswell ($618,000) and Sandy Springs ($730,000) all sat inside the conforming limit with 20 percent down; Milton's $1,154,500 did not.
- Governor Kemp signed SB 33 on May 11, which the Governor's office described as enacting "broad property tax reform establishing a new Local Homestead Option Sales Tax (LHOST) for homeowner relief," alongside HB 463, which lowers the state income tax rate from 5.19 to 4.99 percent effective January 1, 2026. Office of the Governor of Georgia, May 11, 2026. Implication: the property-tax reform the May briefs dated to May 5 was signed May 11, and the income-tax cut in the same signing lowers a Georgia buyer's qualifying math by a few hundred dollars a year.
- Georgia Recorder reported that SB 33 "caps property tax assessments at the rate of inflation and allows cities and counties to increase sales taxes to give homeowners additional property tax relief," that the bill was House Speaker Jon Burns' priority, and that the Governor said "this bill delivers on that issue." Georgia Recorder, May 11, 2026. Implication: a corridor owner's assessed value growth is capped at inflation going forward, which improves the hold case for anyone who bought before 2022 and expects to stay.
- The same report quoted Rep. Scott Holcomb that "complying with the Constitution is not optional or discretionary," signaling a constitutional challenge on the revenue-bill origination question. Georgia Recorder, May 11, 2026. Implication: the assessment cap carried litigation risk from the day it was signed, so a seller should not price a tax benefit that a court could unwind.
- The May 11 signing also included HB 445, expanding appeal rights for tangible personal property reassessments, and HB 1209, a limited sales-tax exemption for construction materials near state convention facilities. Office of the Governor of Georgia, May 11, 2026. Implication: neither bill touches residential closings directly, but the appeal-rights expansion signals the General Assembly's direction on assessment disputes.
- Georgia began issuing special income tax refunds under HB 1000 on May 4, $250 for single filers, $375 for heads of household and $500 for joint filers, returning more than $1 billion to taxpayers who filed 2024 and 2025 returns. Office of the Governor of Georgia, May 4, 2026. Implication: a Georgia couple received $500 in May, a small but dated addition to a closing-cost cushion.
- Governor Kemp suspended Georgia's motor fuel tax of 33.3 cents a gallon on gasoline and 37.3 cents on diesel from May 20 through June 3, citing Memorial Day travel. Office of the Governor of Georgia, May 15, 2026. Implication: the state was spending its own revenue to offset the oil shock that was pushing mortgage rates up, a reminder that the energy channel was hitting Georgia households twice.
- Georgia's net tax revenues for April were almost $3.95 billion, up 5.9 percent on the year, with individual income tax up 10.1 percent and net sales tax up 8.5 percent. Office of the Governor of Georgia, May 7, 2026. Implication: the state's fiscal position going into the tax cuts was strong, which lowers the odds the homestead cap gets walked back for revenue reasons.
- Initial unemployment claims in Georgia were 18,336 in April, up 867 on the month and down 2,700 on the year. Georgia Department of Labor, May 21, 2026. Implication: layoffs in Georgia were running below last year, so there was no local income shock behind mortgage distress in the spring.
- National initial jobless claims were 199,000 for the week ending May 2, 212,000 for May 9, 210,000 for May 16, 212,000 for May 23 and 225,000 for May 30; continuing claims ranged from 1,771,000 to 1,786,000. U.S. Department of Labor via FRED. Implication: claims drifted up through May but stayed near the lowest readings of the year, which is why lenders were not tightening on employment risk.
VIII What changed since last month
- The 30-year fixed's monthly average rose from 6.33 percent in April to 6.44 percent in May, and the weekly survey went from 6.30 percent on April 30 to 6.53 percent on May 28. Freddie Mac via FRED. Implication: April's relief was fully reversed; May's last print was the highest since August 2025.
- The daily 30-year rate's high moved from 6.46 percent on April 2 to 6.75 percent on May 19, per the two publishers that track it. Realtor.com® Economic Research, April 30, 2026. Implication: the intra-month peak rose 29 basis points from April to May, a bigger move than the surveys show.
- The 10-year Treasury's monthly average rose from 4.32 to 4.48 percent and its daily high went from 4.42 percent (April 29) to 4.67 percent (May 19). US Treasury via FRED. Implication: the benchmark set a sixteen-month high in May that it had not approached in April.
- The metro ZHVI slipped from $379,648 to $378,748, and its annual decline narrowed from 2.2 percent in Zillow's April report to 2.0 percent in its May report. Zillow Research. Implication: values kept drifting down in dollars while the annual comparison got slightly easier. On Zillow's current series the annual decline had narrowed every month since October 2025, so May continued a deceleration rather than starting one.
- Metro inventory rose from 30,336 to 32,227 and new listings from 8,318 to 8,994, while the sales nowcast rose from 7,988 to 8,421 and days to pending fell from 37 to 33. Zillow Research. Implication: supply, sales and speed all rose together in May, which is what a seasonal peak looks like, not a turn.
- Georgia MLS contract signings went from 21.8 percent below a year earlier in April to 27.9 percent below in May, while closings went from 0.6 percent below to 0.6 percent above. Georgia MLS, as reported by Atlanta Agent Magazine, May 8, 2026. Implication: the gap between what closed and what went under contract widened by six points in a month, the clearest change in the local data. (The May figure is a later release.)
- Redfin's national four-week pending sales went from up 9.6 percent on the year (period ending May 10) to up 4.7 percent (period ending May 24), and Atlanta's from down 1.1 percent to down 3.1 percent. Redfin, May 14, 2026. Implication: national demand halved its annual gain in two weeks, and Atlanta went from slightly negative to clearly negative.
- Headline CPI inflation went from 3.3 percent in March to 3.8 percent in April on the May 12 release, and the Michigan sentiment index went from 49.8 to 44.8. Bureau of Labor Statistics, May 12, 2026. Implication: inflation re-accelerated and sentiment collapsed in the same month, which flipped the rate conversation from "when do cuts come" to "is a hike possible."
- The Federal Reserve went from Jerome Powell as chair through May 15, to Powell as chair pro tempore, to Kevin Warsh sworn in on May 22. Federal Reserve Board, May 22, 2026. Implication: the June meeting became a new chair's first, which added a policy unknown on top of the oil unknown.
- Georgia went from a pending property-tax bill in April to a signed assessment cap and a 4.99 percent income tax rate on May 11, plus a gas-tax suspension from May 20. Office of the Governor of Georgia, May 15, 2026. Implication: the state's carrying-cost rules changed in the owner's favor in May while the financing cost moved against the buyer.
- Metro rent rose from $1,819.52 to $1,828.90 while the computed payment on the median home rose from about $1,857 to $1,917, widening renting's principal-and-interest edge from about $38 to $88. Computed from Zillow Research ZORI and median sale price. Implication: the carrying-cost math moved further from ownership for a second month, driven by the rate rather than by price.
- The MBA ARM share went from 8.8 percent in the week ending May 8 to 9.6 percent in the week ending May 15, the highest since October 2025, and the refinance index fell 18 percent in the week ending May 22. HousingWire, May 13, 2026. Implication: borrower behavior changed inside the month: more adjustable loans, far fewer refinances, roughly the same number of purchases.
- Brent crude went from $103.70 on May 6 to $116.73 on May 18 and back to $92.88 on May 29, and the 10-year went from 4.67 percent on May 19 to 4.45 percent on May 29. EIA and US Treasury via FRED. Implication: the month ended with the oil shock unwinding, so the June rate story opened in the buyer's favor even though the May survey closed at its high.
- Alpharetta's median days to pending fell from 38 in April to 22 in May, and Sandy Springs' from 30 to 24, while Cumming's held at 36. Zillow Research. Implication: the corridor's spring arrived in May in north Fulton and did not arrive in Forsyth.
- The first June survey print fell to 6.48 percent for the week ending June 4, the first weekly decline since the one-basis-point dip of May 14, before rising again to 6.52 percent on June 11. Freddie Mac via FRED, June 11, 2026. Implication: the May 28 print of 6.53 percent was the top of this episode, which a client who locked in May could not have known. (Later data, noted as such.)
Data credits. Source: U.S. Census Bureau, American Community Survey · Source: U.S. Bureau of Labor Statistics · Source: Federal Reserve · Source: Freddie Mac via FRED · Source: FHFA · Data Provided by Zillow Group · Realtor.com® Economic Research
The Record · Atlanta residential · May 2026 · Study grade · compiled October 6, 2026
Educational, not professional advice. Nothing on this page is affiliated with or endorsed by its publisher.

