Tracking the unusual rise in homes for sale in the Seattle area housing market
Key Takeaways
- New listings increased at the top. In the rolling 90-day windows ending in February–May 2026, new listings in higher-priced ZIP codes averaged 20.1% above the corresponding windows ending in 2022–2023. In lower-priced ZIP codes, they were 3.9% below.
- The gap stands out nationally. Seattle ranked second among 88 eligible metros for the size of the difference between higher- and lower-priced neighborhoods.
- Inventory rose in both groups. It increased 155.1% in higher-priced ZIP codes and 104.4% in lower-priced ones over the same comparison.
More homes are coming to market in the Seattle area’s most expensive neighborhoods. Available inventory has increased across the region, but it has grown faster in expensive ZIP codes than in cheaper ones. The bigger buildup is on the Eastside and in eastern King County.
That pattern raises an obvious question: are Washington’s tax increases pushing wealthy homeowners to sell? Changes in tech hiring could also be contributing. To understand what is happening, we first need to separate two developments: the shift in new listings within Seattle and the later inventory buildup across the metro relative to other markets.
Inventory is rising broadly, while more sellers come to market at the top
Seattle’s buildup in homes for sale extends across price groups. Inventory more than doubled in both higher- and lower-priced neighborhoods compared with the same period of 2022–2023. Expensive neighborhoods stand out for a different reason: more homes are coming onto the market, even as new listings in cheaper neighborhoods remain below their earlier levels.
I used Redfin data for 102 Seattle-area ZIP codes to compare expensive neighborhoods with cheaper ones. I ranked ZIP codes by their typical sale prices in 2022–2023 and kept those groups fixed. The higher-priced group contains 20 ZIP codes; the lower-priced group contains 57. Six ZIP codes around major tech job centers are tracked separately, with the remaining ZIP codes forming a middle-priced group.
The comparison uses February–May 2026 and the same months of 2022–2023. Matching the calendar months helps account for the usual spring selling season. New-listing and sales counts cover overlapping 90-day windows ending in each month, rather than individual months.
New listings in the higher-priced group averaged about 2,290 per rolling 90-day window in February–May 2026, up 20.1% from about 1,910 in the baseline. In the lower-priced group, they slipped 3.9%, from about 6,100 to 5,860.
The ratio of new listings in expensive neighborhoods to those in cheaper neighborhoods was therefore 24.9% above its baseline. Seattle ranked second among 88 metros for the size of the difference between the two groups’ new-listings growth rates. It also ranked first among the 11 markets in the tech-metro comparison, which includes San Jose, San Francisco, Austin and Boston.
The pattern is unusual by historical standards, too. Repeating the comparison across other metros for each year from 2019 through 2024 produced 492 comparisons. In fewer than 1% did new listings in expensive neighborhoods outpace those in cheaper neighborhoods by as much as they did in Seattle in February–May 2026. Seattle itself never recorded a difference that large during those years.
More homes are waiting for buyers across price groups
The increase in available homes was broad, even though the increase in new listings was concentrated at the top. Higher-priced ZIP codes accounted for about 22% of the additional inventory. Lower-priced ZIP codes accounted for about 47%, while middle-priced and tech-center ZIP codes accounted for the remaining 31%.
More homes can remain for sale because new listings increase, sales slow, or both. Expensive neighborhoods experienced both: new listings rose while completed sales fell. In cheaper neighborhoods, inventory accumulated despite fewer new listings, alongside a larger decline in completed sales.
The table compares February–May 2026 averages with the corresponding averages in 2022–2023. Inventory counts homes still for sale at the end of each period. New listings and sales cover rolling 90-day windows.
| Measure | Higher-priced ZIPs | Lower-priced ZIPs |
|---|---|---|
| New listings | +20.1% | −3.9% |
| Inventory | +155.1% | +104.4% |
| Months’ supply | +175.4% | +154.9% |
| Completed sales | −11.2% | −23.8% |
Inventory in higher-priced ZIP codes rose from an average of about 540 homes to 1,370. In lower-priced ZIP codes, it rose from about 1,680 to 3,430. Expensive neighborhoods had the faster percentage increase, but cheaper neighborhoods added more homes in absolute numbers.
Months’ supply measures how long the available homes would last at the recent sales pace if no more homes were listed. It increased from 1.13 to 3.10 months in the higher-priced group and from 1.03 to 2.64 months in the lower-priced group.
Buyers have more choice, and sellers face more competition across both groups. The unusual increase in new listings helps explain why inventory grew faster in expensive neighborhoods. But the broader buildup also reflects homes taking longer to find buyers; it cannot be attributed to more sellers entering the market alone.
The bigger buildup is on the Eastside
The 20 higher-priced ZIP codes are not one market. Fourteen cover parts of Bellevue, Kirkland, Mercer Island, Sammamish and eastern King County. The other six include four Seattle ZIP codes and two covering parts of Bothell and Woodinville.
The Eastside and eastern King County group had the larger increase in both new listings and inventory, measured against lower-priced ZIP codes. Its months’ supply increased 16.0% relative to the lower-priced group. In the other six higher-priced ZIP codes, months’ supply fell 8.5% relative to that same comparison group.
Two comparisons show different timelines
Within the Seattle metro area, are new listings increasing faster in expensive neighborhoods than in cheaper ones? Across metros, is Seattle accumulating more homes for sale than we would expect based on how similar housing markets are behaving?
Within Seattle, the shift toward more new listings in expensive ZIP codes was already visible in January 2024. That month, the ratio of new listings in expensive neighborhoods to those in cheaper neighborhoods was about 9% above its January 2022–2023 average. The difference reached 23.6% in the first half of 2025, eased to 10.2% in the second half, then returned to 24.9% in February–May 2026. Each comparison uses the same calendar months of 2022–2023 as its baseline.
To examine the broader metro, I built a benchmark from a weighted mix of other metro areas that closely tracked Seattle’s active listings and months’ supply in 2022–2023. I call that benchmark “synthetic Seattle.” It provides a comparison for how Seattle’s housing market might have evolved if it had continued to follow those other markets.
Seattle had fewer active listings than this benchmark throughout 2024 and into mid-2025. The gap then narrowed, and Seattle moved above it later in 2025. By May 2026, Seattle had about 12.5% more active listings than its synthetic comparison, and months’ supply was 23.6% higher.
The timing matters. New listings in expensive Seattle-area neighborhoods began pulling ahead of cheaper ones in 2024. The broader metro’s buildup in active listings relative to other housing markets emerged later in 2025.
Where tax policy fits into the picture
Washington added a 2.9-percentage-point capital-gains surcharge above $1 million and raised its top estate-tax rate to 35% in 2025. The legislation was introduced in April and signed in May. In March 2026, the state enacted a 9.9% tax on income above $1 million, starting in 2028. It also reversed its estate-tax rate increases for deaths from July 2026 onward.
The metro-wide buildup relative to synthetic Seattle emerged after the 2025 legislation. That timing is consistent with taxes potentially contributing to the change. It does not establish that they caused it.
The distinction matters. The earlier ZIP-level listing shift makes it difficult to attribute that entire shift to the 2025 legislation. It does not explain away the later divergence between Seattle and its synthetic comparison.
Real estate sales are exempt from Washington’s capital-gains tax. That exemption could make it easier for wealthy homeowners to sell and leave the state to avoid higher taxes on future income or investment gains. The housing effect would come through their decision to relocate. This analysis does not track where sellers move or why they sell, so it cannot establish whether that is happening.
Layoffs and changes in immigration could also be contributing
The region’s largest tech employers announced substantial layoffs around the time Seattle began accumulating more homes for sale relative to its synthetic comparison. Microsoft announced Washington job cuts affecting 1,985 workers in May 2025, another 305 in June and 830 in July—more than 3,100 positions across those three rounds. Amazon subsequently announced approximately 14,000 corporate job cuts worldwide in October 2025 and another 16,000 in January 2026. Amazon’s figures are global, not Seattle-area totals.
Slower hiring and layoffs can affect the housing market in different ways. Slower hiring means fewer people moving into the region for work, reducing the flow of potential buyers and renters. Layoffs could push existing residents to leave if they cannot find another job locally, prompting some homeowners to sell. Both could contribute to the inventory buildup that emerged later in 2025 and continued into early 2026.
For workers whose immigration status depends on employment, losing a job can create additional pressure to relocate. If they cannot secure another qualifying job or another way to remain legally in the country, they may have to leave. A homeowner in that position might put a property on the market. A worker whose H-1B status cannot be extended could face a similar decision.
To examine one part of this story, I counted H-1B labor-condition applications certified by the Labor Department during October 2025–June 2026 and compared them with the same months a year earlier. Employers use these applications as part of the process of sponsoring H-1B workers.
For primary worksites in Bellevue, Redmond and Kirkland, certified applications fell 7.6%. Microsoft’s fell 24.1%, while applications from three major Amazon entities rose 22.4%. In Seattle city, applications rose 4.8%.
These figures show a decline in certified applications on the Eastside, led by Microsoft. They do not measure visa approvals, rejection rates or the number of workers arriving or leaving. Applications can cover new employment, extensions and job changes, and one application can cover multiple positions. We therefore cannot tell from these counts whether fewer newcomers arrived, more existing workers departed, or employers simply needed fewer applications.
Despite these caveats, the housing channels are plausible: fewer incoming workers could reduce demand, while departures could add homes for sale. Although applications from three major Amazon entities increased, that increase offset less than half of Microsoft’s decline. Applications from other Eastside employers combined also fell, leaving the total down 7.6%. These figures leave slower hiring and changes in immigration as possible contributors, but do not establish how much they contributed to the increase in listings.
| Employer group | Oct. 2024– June 2025 | Oct. 2025– June 2026 | Change | Percent change |
|---|---|---|---|---|
| All Eastside employers | 12,858 | 11,887 | −971 | −7.6% |
| Microsoft | 4,577 | 3,473 | −1,104 | −24.1% |
| Three Amazon entities | 2,200 | 2,693 | +493 | +22.4% |
| Other employers | 6,081 | 5,721 | −360 | −5.9% |
New construction appears unlikely to be the main driver
The construction data offer little support for a building surge as the main explanation for rising inventory. Across the Seattle–Tacoma–Bellevue metro, permits authorized 7,784 housing units in February–May 2026, down 4.9% from the average for the same months of 2022–2023. Authorizations for one-unit homes fell 7.9%, while those for units in buildings with two or more homes fell 3.7%. Housing completions also declined in Seattle metro.
Bellevue’s recent development activity does not overturn that picture. Projects entering construction and zoning changes allowing more housing describe potential future supply.
Some newly built condos, townhomes and detached houses could still have contributed. Homes completed in early 2026 may have been permitted years earlier, and unsold new homes can accumulate even as construction slows. But the evidence does not point to a surge in construction driving the inventory buildup.
More homes for sale across the region put downward pressure on prices
Seattle’s inventory buildup extends well beyond its most expensive neighborhoods. Comparing February–May 2026 with the same months of 2022–2023, higher-priced ZIP codes accounted for about 22% of the additional homes available for sale. Lower-priced ZIP codes accounted for about 47%, while middle-priced and tech-center ZIP codes accounted for the remaining 31%.
Expensive neighborhoods stand out because more homes are coming onto the market. But the accumulation of unsold homes is broad. Inventory grew faster in percentage terms in expensive ZIP codes, while cheaper neighborhoods added more homes in absolute numbers. Months’ supply rose from 1.13 to 3.10 in higher-priced neighborhoods and from 1.03 to 2.64 in lower-priced ones. Across the broader metro, active listings moved above Seattle’s synthetic comparison later in 2025. The accumulation in housing inventory puts downward pressure on prices.
The construction evidence offers little support for a building surge as the main explanation. Taxes, layoffs and immigration restrictions remain plausible contributors, but this analysis cannot establish how much each contributed.
About the analysis
Data. The main comparison uses Redfin’s ZIP-code data for all residential property types from January 2022 through May 2026. New listings and completed sales cover rolling 90-day periods, dated by the month they end. Inventory counts homes still available for sale at the end of each period. Because the windows overlap, I average observations rather than add them.
Which ZIP codes are included. A ZIP code must have a monthly record throughout the study period, remain assigned to the same metro and have at least 18 months of valid sale prices in 2022–2023. For each measure, I retain only ZIP codes with complete data for its required inputs. Coverage can therefore differ across measures.
Price groups. Within each metro, I ranked ZIP codes by the median of their reported median sale prices in 2022–2023. The top fifth form the higher-priced group, the bottom 60% the lower-priced group, and the rest the middle group. In Seattle, six ZIP codes around major tech job centers—98007, 98008, 98052, 98053, 98109 and 98121—were set aside before ranking and are tracked separately. The remaining groups contain 20 higher-priced, 57 lower-priced and 19 middle-priced ZIP codes. Groups stay fixed throughout the analysis.
Measuring change. Each group’s change compares its February–May 2026 average with its average for the same months of 2022–2023. Matching calendar months helps account for the usual spring selling season. The relative gap divides the higher-priced group’s growth factor by the lower-priced group’s growth factor, then subtracts one. For example, a 20.1% increase in higher-priced ZIP codes and a 3.9% decline in lower-priced ones imply that the ratio between the groups rose about 24.9%. Months’ supply is group inventory divided by the monthly sales pace—rolling 90-day sales divided by three—then averaged over the comparison months.
Contributions to the inventory increase. The breakdown of additional inventory uses 99 ZIP codes with complete inventory records: 20 higher-priced, 55 lower-priced, 18 middle-priced and six tech-center ZIP codes. Each group’s contribution is its increase in average inventory divided by the total increase across these ZIP codes. These shares describe the observed increase from the 2022–2023 baseline, not Seattle’s excess inventory relative to its synthetic comparison.
Comparison with other metros. Candidate comparison metros must have at least 40 eligible ZIP codes and sufficient usable data for the measure being compared. Other Washington metros and designated metros crossing Washington’s border are excluded. For new listings, metros are ranked by the size of the relative gap in either direction. Seattle ranked second among 88 metros, including Seattle. It ranked third after scaling each metro’s gap by the size of its baseline deviations in 2022–2023.
Comparison with history. I repeated the calculation for eligible metros in each year from 2019 through 2024, keeping the price groups fixed. Each historical year uses February–May observations and a two-year baseline positioned the same distance back as the current comparison. This requires complete data back to 2015, so coverage can differ by measure and metro. For new listings, Seattle retains all 20 higher-priced and 57 lower-priced ZIP codes. The historical reference covers 82 other metros over six years, or 492 metro-years. The reported share counts comparisons with an equally large or larger positive gap. It describes the pattern’s rarity, not whether a policy caused it.
Month-by-month timing. Monthly ZIP figures compare the ratio between higher- and lower-priced activity with the geometric average of that ratio in the same calendar month of 2022–2023. The calculation uses logarithms, then converts the result into a percentage. Period summaries instead compare average levels, so the two calculations need not match exactly.
Synthetic Seattle. I built a separate weighted comparison for each metro-wide measure, fitting Seattle’s monthly path in January 2022–December 2023. Each metro’s series is first divided by its own average during that period. Weights are nonnegative, sum to one and remain fixed after fitting. The resulting benchmark is then scaled to Seattle’s baseline average. This follows the synthetic-control approach developed by Abadie, Diamond and Hainmueller (2010).
The broad comparison pool contains 78 eligible donor metros. Eligibility requires median monthly home sales of at least 1,000 in 2015–2020 and complete required data; other Washington and designated cross-border metros are excluded. The charts also show a sensitivity comparison using 15 tech-focused donor metros. A positive gap means Seattle’s actual value exceeds its fitted benchmark. These comparisons show when Seattle diverged, not why.
H-1B applications. I used the Labor Department’s labor-condition application files, retaining cases with an exact “Certified” status and removing duplicate case numbers across releases. Comparisons use decision dates covering October–June of fiscal 2026 and the same months of fiscal 2025. Each case is assigned to its primary worksite city, with Bellevue, Redmond and Kirkland defining the Eastside for this comparison. “Amazon” combines Amazon.com Services, Amazon Development Center U.S. and Amazon Web Services.
Primary worksites provide a consistent comparison because the fiscal 2026 supplemental-worksite file covers only 77% of certified cases nationally. Including supplemental worksites produces an 11.2% Eastside decline and roughly unchanged Seattle-city counts, but that comparison has incomplete recent coverage. Applications can cover multiple positions, new employment, extensions and job changes. They do not count unique workers, visa approvals, arrivals or departures.
Construction. Permit figures come from the U.S. Census Bureau’s Building Permits Survey for the Seattle–Tacoma–Bellevue metro, an area larger than Redfin’s Seattle metro. They compare housing units authorized in February–May 2026 with the average for the same months of 2022–2023. Permits describe authorized construction, not completed homes or homes listed for sale.
Limits. Price groups describe neighborhoods, not households, so they do not identify wealthy residents or taxpayers. Completed sales do not measure buyer demand directly. The analysis does not observe why owners listed their homes or where they moved. Historical comparisons apply price groups defined using later, 2022–2023 information. Rolling windows overlap, historical comparisons share economic shocks, and many measures and comparisons were examined. The 2022–2023 synthetic fitting period also follows Washington’s original capital-gains-tax legislation, so it cannot measure that law’s total effect. These results describe housing-market patterns; they do not estimate the effect of any single policy.
Sources
- Redfin Data Center
- Labor Department LCA disclosure files
- Washington capital-gains surcharge
- Washington income-tax information
- Washington estate-tax legislation and rate reversal
- Bellevue middle-housing code changes
- Kirkland council adoption of middle-housing changes
- U.S. Census Bureau Building Permits Survey
- Microsoft Washington layoff notices (GeekWire, May 2025)
- Microsoft June and July 2025 Washington layoff notices (KIRO 7)
- Amazon’s October 2025 and January 2026 corporate job cuts (AP via Boston.com)
- Abadie, Diamond and Hainmueller (2010), synthetic control method
