Invest1 distinct publisher3 min readUpdated
Realhouse's read of registered transactions shows 55.5% of Seoul apartment sales this year closed below 1 billion won, with 1,717 deals above 3 billion pulling the citywide average up.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
Realhouse, a real estate data firm, pulled Seoul apartment transactions from the Ministry of Land, Infrastructure and Transport's official price disclosure system and found the median sale between February and early August this year closed at 916 million won, against an average of 1.17 billion won [1][2][3][4]. The 258.5 million won gap between those two numbers is the story: the average describes a market that most buyers in Seoul were never in [5].
Of the 37,246 sales in the sample, 20,676, or 55.5%, went for less than 1 billion won, roughly $722,000 [2][7]. Move the line to 1.5 billion won and 28,765 deals, 77.2% of the total, sit below it [8]. Sales at 2 billion won or more numbered 4,444, or 11.9%, and just 1,717, or 4.6%, cleared 3 billion won [9]. Realhouse attributes the entire median-to-average spread to that thin top layer [6]. Put differently, the median transaction was 78.3% of the mean [1].
The volume sits where the cheap stock is. Nowon District logged the most deals at 4,453, or 12.0%, followed by Gangseo at 2,500, Guro at 2,452 and Seongbuk at 2,245 [10]. All three of the top-volume districts had medians below the citywide 916 million won, with Nowon at 600 million [11]. Nowon, Dobong and Gangbuk together produced 6,961 deals, 18.7% of the market, exceeding the 4,752 recorded across Gangnam, Seocho and Songpa by 2,209 [12].
Gangnam managed only 1,442 sales, 3.9% of the total, at a median of 2.66 billion won, more than four times Nowon's [13]. Only three districts posted medians above 2 billion won: Gangnam, Seocho at 2.4525 billion and Yongsan at 2.06 billion, together accounting for 3,167 deals or 8.5% of volume [15]. The district spread ran to 2.12 billion won, or 4.9 times, from Gangnam's high to Dobong's 540 million won low [14]. Concentration at the top is near total: Gangnam (561), Seocho (439), Songpa (434) and Yongsan (140) accounted for 91.7% of all 3 billion won-plus deals [17], and 12 of Seoul's 25 districts recorded none at all [18]. Roughly 39% of Gangnam's own transactions were in that bracket [3], while the three Gangnam-area districts, 12.8% of volume, held 83.5% of the ultra-high-priced trades [2][4].
Volume recovery has not translated into price. Nowon's share of record-high-price deals was 2.9%, the lowest in the city, and Dobong's 3.0%, against a citywide average of 10.5%, or 3,929 deals [16].
One methodology note worth keeping. KB Kookmin Bank's July median for Seoul apartments was 1.27583 billion won, 359.83 million above the transaction-based median, because KB estimates market values across the entire housing stock rather than counting closed deals [19]. The Realhouse sample also includes 76 sub-20-square-metre units that sold for under 100 million won, about 0.2% of transactions [20][5].
What to watch: whether the record-high-price share in the outer districts moves off 3%, which would mean the mid-tier is repricing rather than just transacting; and whether the 12 districts with zero 3 billion won sales stay at zero. Anyone underwriting Seoul on the 1.17 billion won average is underwriting 4.6% of the market.
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
Realhouse, a real estate data firm, analyzed transaction records from the Ministry of Land, Infrastructure and Transport's official price disclosure system.
There were 37,246 Seoul apartment sales reported between February and early August this year.
The median transaction price for Seoul apartments during the period was 916 million won.
The average sale price over the same period was 1.17 billion won.
The median was 258.5 million won lower than the average sale price.
Realhouse attributed the median-average gap to the 1,717 ultra-high-priced deals of 3 billion won or more, which distorted the citywide average upward.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Registry-grounded but single-vendor
The numbers rest on Ministry of Land, Infrastructure and Transport registered-transaction disclosures, are internally consistent (band counts reconcile to the 37,246 total, the 258.5 million won gap matches the stated median and average), and the source discloses two material methodology points: inclusion of 76 ultra-small sub-100 million won units and the reason KB Kookmin Bank's appraisal-based median runs 359.83 million won higher. It falls short of high evidence because everything comes from one commercial firm's unpublished cut, no per-district table or 'record-high-price' definition is provided, and there is no prior-period baseline against which the recovery framing could be tested.
No adoption signal in scope
This is a housing-market statistics story; the supplied source reports transaction counts and prices, not the release, deployment, or uptake of any product, standard, or dataset by users. Transaction volume is the subject under study, not evidence that the Realhouse analysis or its methodology has been taken up by other analysts, agencies, or market participants, and no such uptake is reported. No adoption observations can be recorded without inferring facts the source does not supply.
Numbers sober, one interpretive stretch
The quantitative claims are stated conservatively and track the underlying counts, and the piece actively deflates the inflated citywide average rather than amplifying it. The overstatement is interpretive: the opening line reads the sub-1 billion won majority as 'pointing to a market driven by owner-occupier buyers' without any buyer-type, mortgage, or registration data, and 'transactions recovered' is asserted with no prior-period comparison in the source. Small positive gap rather than large, because the headline arithmetic itself is not exaggerated.
Commercial data vendor is the sole source
Every figure originates with Realhouse, a commercial real estate data firm whose visibility and product credibility benefit from a widely quoted finding that its transaction-based median is the 'real' number while a rival benchmark from KB Kookmin Bank overstates prices by 359.83 million won. That is a legitimate methodological argument, but it is also a competitive one, and no independent replication or dissenting analyst appears in the cluster. Moderate rather than high, because the underlying inputs are government-disclosed transactions that others can in principle re-derive.
Solid arithmetic, thin corroboration
Confidence is supported by the official registry base and by figures that reconcile internally across bands, districts and the median-average spread. It is held down by structural thinness: one publisher, one vendor, a partial-year February-to-early-August window presented as 'this year', an undefined record-high-price metric, and no adoption or independent-replication evidence to triangulate against.
invest
Korea's 3-million-won emergency loans jump 39.2% while banks squeeze everything else1 distinct publisher
invest
Korea's brokerages added 402 staff in a quarter while mortgages headed to 5.78%1 distinct publisher
invest
Korea's COFIX climbs a fourth month, pushing variable mortgage ceilings to 5.89%1 distinct publisher
invest
Seoul's 73,000-home promise now runs through the Gangnam greenbelt1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
en.sedaily.com
1 article · August 19, 2026