Run the five one at a time and they stop looking like a cohort. Dobong needs 11.8% to get level, Gangbuk 5.9%, Geumcheon 5.3%, Nowon 4.4%, and Eunpyeong 0.9% [1][2] (a rounding error beside Jungnang, which has already crossed and now sits 0.4% clear of its own old mark [4]). Average the five and the group is 5.2% under those highs [3], which would be a decent thesis if the highs meant something. They were made by two tenant protection laws that handed renters a renewal right and capped jeonse and monthly rent increases; as jeonse listings thinned and deposits climbed, non-owning households bought where financing was less of a stretch [9]. Call that a price with a cause rather than a value, and note that it is roughly four and a half years old in nominal terms [9].
What is actually moving is the loan ceiling. Last October's tiering set the mortgage cap at 600 million won for homes valued at 1.5 billion or less, 400 million between 1.5 and 2.5 billion, and 200 million above that [5]. These districts still carry plenty of listings under 1 billion won [6], so the cap finances about 60% of a 1 billion won purchase, 16% of a 2.5 billion won one, and 6.7% of a 3 billion won one [5][6]. The tiering is not neutral about geography, and the five below-peak districts, a fifth of the city's 25 [1][8], sit inside the band where borrowed money still does most of the work.
The activity data cuts against the room-to-run story rather than supporting it. Eunpyeong, the closest of the five to a record, is turning resale stock over about 1.67 times as fast as Nowon [7], and Dobong, which carries the largest gap of all, does not appear on July's turnover leaderboard [10]. Distance to a 2021 print is not what buyers are pricing.
This is probably wrong in some direction, but the version I would underwrite is that the cheapest boroughs are now the most policy-sensitive, because both live instruments point at them: the Aug. 13 package eased the marriage penalty in mortgage lending and new-housing applications and raised the ceiling on total household lending growth for young non-owners [8], while the October tiering aimed at exactly the price band those buyers can reach [5]. Nam Hyuk-woo of Woori Bank's real estate research institute reads the setup as durable, saying the conditions for sustained buying by people in their 20s and 30s and by non-homeowners are firmly in place and that current strength should carry through the rest of the year [11]. He may be right through December and still leave the point standing, since his claim is about demand depth and mine is about who is holding the leverage when terms change. What would break my read is Dobong closing its 11.8% on rising transaction counts rather than on the shrinking listing supply that has driven gains since the first half [10][12], with Eunpyeong's turnover holding rather than fading.
