Abstract
<title>Abstract</title> <p> Density instruments are usually evaluated through price and quantity effects. This paper instead asks how transferable development rights (TDR) may reshape a second moment: urban residual dependence. Using 22,020 commercial and mixed-use transactions across Korea's seven metropolitan cities from 2018–2025, we measure nominal development saturation as realized FAR relative to a base use-zone FAR benchmark, retaining observations above one because lawful bonuses, legacy permissions, and denominator incompleteness can place realized density above that benchmark. In Seoul, where binding density appears most pronounced, a multi-matrix spatial-error model identifies a small saturation-organized residual-dependence component beyond geography. The pooled estimate is λs = 0.0667 (p = 0.001), while a year-block graph excluding cross-year links gives λs = 0.0548 (p = 0.012), about 82% of the pooled magnitude and sign-stable across k = 4–16. The channel does not improve rolling temporal out-of-sample covariance prediction and is therefore treated as an in-sample structural descriptor. Under a volume-conserving TDR counterfactual holding built stock fixed, metropolitan correlation changes are numerically negligible, while formed saturation edges gain about 0.014 and severed edges lose about 0.012. Holding the transferred floor area fixed and varying only the designation geometry across six candidate designations, the per-edge magnitude is nearly invariant while the share of the saturation network reassigned ranges from 19.5% to 44.2%, so volume prices the reassignment and geometry determines its extent. The framework is an ex ante policy-design diagnostic, not a causal ex post evaluation. Edge-group monitoring is preferable to citywide averages. <bold>JEL classification:</bold> R52, R31, C21 </p>