Comment on “Construction and Opening Effects of a Large-Scale Bridge Project on Real Estate Prices: Evidence From the Mumbai Trans-Harbor Link”

Abstract

Transport infrastructure improves market access and can thereby reshape productivity, commuting patterns, land values, and spatial inequality. Evaluating its economic impact remains a central policy challenge, especially in developing-country cities. Based on a selective literature review on the evaluation of transport infrastructure, Ashida et al. (2026) provide an empirical study of the Mumbai Trans-Harbor Link (MTHL), a major bridge project that warrants careful evaluation given its potential to reshape commuting patterns, urban congestion, and land markets.

Their results suggest that the implementation of the MTHL was capitalized into land values in Mumbai City District near the bridge, beginning around its construction phase. Using Annual Statement of Rates (ASR) data, Ashida et al. find that benchmark land prices in 2025 rose more for parcels located within the closest band to the new bridge link compared to parcels in more distant control districts within Mumbai City. This pattern is present in both an event study with a continuous treatment and in a piecewise linear function model. When suburban areas are included, the estimated effects are attenuated with strong nonlinearity in the effect across districts farther from the MTHL, possibly due to its interaction with the existing urban transport network.

To interpret these findings, one useful framework is the Tiebout-Oates capitalization hypothesis. In this view, local public goods are capitalized into property values through individuals' sorting into their preferred community by “voting with their feet”. The MTHL can therefore be interpreted as a local infrastructure that enhances market access and raises households' willingness to pay for nearby properties. This approach contrasts with related studies, such as Donaldson and Hornbeck (2016) and Heblich et al. (2020), which emphasize general equilibrium forces—productivity and commuting behavior—that respond to lower trade or transport costs and, in turn, reshape rental values of land over space.

Ashida et al.'s main contribution lies in the construction and use of parcel-level ASR data by property type to approximate land prices in a setting where transaction-level data are scarce. Their paper offers an important example of how administrative benchmark data can be used to study the local effects of large-scale infrastructure projects. In addition, analyzing the mechanism behind the estimated effect through the lens of the capitalization hypothesis, combined with the unique data, could potentially be a novel contribution of the paper. In this comment, I would like to highlight methodological issues and unique institutional contexts that are worth further discussion.

First, although the compilation of land price data through web scraping and geocoding is commendable, it is important to recognize that ASR values serve as administrative benchmarks used for stamp duty and registration fees rather than actual market values. These values are informed by past transactions and tend to lag behind or underestimate prevailing market prices, particularly in rapidly growing neighborhoods. This limitation is not unique to India: fewer than a third of countries maintain a comprehensive private plots registry (Awasthi et al. 2021), and international evidence shows similar discrepancies between administrative assessments and true market values. Against this backdrop, the key empirical question, therefore, is whether it is reasonable to expect the effects of the MTHL that opened in 2024 to be reflected in the 2025 ASR revision. Ashida et al. (2026) address this concern by explicitly discussing the data limitation and conducting additional robustness checks that account for the measurement error and lagged effects. These exercises strengthen the interpretation that the estimated changes reflect capitalization effects.

Second, the mechanism behind the rapid capitalization effect of the MTHL deserves further elaboration. The Tiebout-Oates theory may provide a useful starting point for explaining the spatial distribution of ASR values. However, the MTHL is not a conventional local public good. Railway and metro stations often generate a localized land value premium because stations become focal points for residential and commercial development. Bridges, by contrast, primarily function as transit corridors and their immediate neighborhoods do not necessarily experience concentrated development. The significant effect on residential properties (Ashida et al.'s Table 4) suggests that the MTHL may already be influencing housing demand and commuting patterns. Within Mumbai City District, this effect may be consistent with a Tiebout-style proximity premium, in which households sort toward parcels near the bridge and bid up land values nearby. Once Mumbai's suburban areas are included, however, the effect becomes nonlinear and weaker with distance from the MTHL, which may point to a different mechanism—improved market access across the broader transport network rather than proximity to the bridge itself. To further substantiate this interpretation, future research could incorporate alternative data sources on commuting flows and mobility patterns—such as mobile phone location data. Such evidence would provide more direct evidence on capitalization driven by expected accessibility, actual commuting improvements, and housing demand, strengthening the interpretation of these short-run effects.

Beyond these issues of measurement and interpretation, two institutional features specific to the MTHL further complicate the evaluation of urban infrastructure in metropolitan areas. The first concerns multiple, overlapping treatments arising from simultaneous infrastructure investments and industrial policy interventions. During the analysis period, the Mumbai Bay area was undergoing rapid modernization, with the MTHL construction coinciding with the construction of new Metro Lines, a new airport as well as the establishment of economic zones. Because the bridge forms part of a broader transport network connecting these emerging focal points, demand for residential and industrial properties may reflect not only the direct value of the MTHL, but also anticipated access to the new metro, airport, and economic zones. Ashida et al. (2026) undertook a robustness check that excludes parcels located near other major transport infrastructure. Nevertheless, with only 1 year of post-treatment data and an absence of frequent market price data, disentangling these channels remains difficult.

The second issue concerns the toll structure of the MTHL. Ashida et al. note that MTHL operates as a toll road, primarily serving private vehicles and high-speed transit. Local news reports suggest that toll charges – perceived as relatively high—have sparked debate, particularly regarding its financial burden for low-income commuters. In such a context, infrastructure-driven capitalization may disproportionately benefit higher-income households. From a policy perspective, it would be valuable to explore the project's impact at the individual level, examining how willingness to pay varies across commuters and creates distributional consequences (see Tsivanidis (2026)). Future research that integrates spatial equilibrium models with commuter-level data would provide deeper insights into how infrastructure investments reshape commuting behavior and urban inequality.

Overall, Ashida et al. (2026) make an important contribution by bringing new parcel-level land price data to the evaluation of an infrastructure project in developing countries. The paper also sets out an important methodological agenda for future works—most notably the challenge of evaluating multiple urban infrastructure interventions that overlap both spatially and temporally. Recent literature has begun to explicitly incorporate multimodal transport networks in evaluating the economic effects of infrastructure investments (Fuchs and Wong 2026). Yet estimating the causal impact of an individual project requires richer spatial and temporal data. Such analysis also calls for empirical or structural methods capable of disentangling the project's direct effect from complementary investments and broader network changes. In this sense, Ashida et al. (2026) offer a useful benchmark for future studies of infrastructure capitalization in rapidly transforming urban economies.

The author declares no conflicts of interest.