Evolving land value effects of BRT and MRT: Evidence from Jakarta’s mobility transition

Abstract

Bus rapid transit (BRT) has been widely adopted in emerging economies for its affordability and incremental implementation potential. Yet, many cities are now starting to implement urban rail as a higher-quality mass-transit alternative. This raises the question of the role of existing BRT networks once rail arrives, particularly regarding their land-value effects. This paper examines how BRT-related land value uplift (LVU) evolves after rail begins operation, using Jakarta as a case study. The study analyses residential land values around Transjakarta BRT and MRT Jakarta stations for 2017 (pre-rail) and 2021 (post-rail) using Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR). The findings reveal that (1) proximity to Transjakarta stations was associated with uplift in 2017, particularly in South Jakarta; (2) by 2021, BRT proximity penalties were reported citywide, especially in the north and east, but also in Central Jakarta; and (3) proximity to MRT Jakarta stations was associated with consistent uplift in 2021, with strongest effects at upgraded interchange nodes in South Jakarta. The timing and spatial coherence of these patterns are consistent with a rail-led substitution mechanism in which urban policy attention and developer actions concentrate willingness-to-pay near rail, while stand-alone BRT corridors increasingly reflect proximity penalties in prices. Policy recommendations include strategic co-location and integration of BRT-MRT stations, mitigating BRT proximity effects with context-sensitive station design, and timely transit-oriented development (TOD) and land value capture (LVC) at integrated hubs to harness value where market signals are strongest.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.