The lifecycle of railway stations and its effect on property prices: Empirical evidence from Japan

Abstract

This study investigates the nonlinear relationship between real estate prices and the age of railway stations using a hedonic price model and Generalized Additive Models. By drawing parallels between the life cycle of railway stations and human life stages, we examine how property prices evolve in response to the opening, maturation, and eventual decline of stations over time. The analysis focuses on real estate data within a 15-min radius of railway stations in Japan, considering both Tokyo and non-Tokyo regions. The results show distinct price patterns corresponding to different stages in the station’s lifecycle: initial price increases due to capitalization effects from the opening, followed by a decline as stations age, and subsequent price appreciation following redevelopment efforts. The study highlights the importance of station age as a significant factor influencing real estate prices and provides insights into the need for strategic urban redevelopment policies at key lifecycle stages. The findings contribute to the understanding of station-area price dynamics, with implications for urban planning and real estate investment strategies.

Graphical abstract

Keywords

Life cycle / Rail transit / Station area / Real estate / Generalized additive model

1 Introduction

Cities have steadily evolved according to population growth and socioeconomic changes. Cities of all sizes have flourished and declined in stages, and many studies have demonstrated cities’ life cycles (Dyson, 2011; Davis, 2016). Furthermore, Geddes (1915), who compared urban development to organic growth, explained that cities evolve through childhood, adolescence, adulthood, and old age and can be killed, similar to living things.

A city is generally an organization composed of citizens, activities, land, and facilities. Among these, facilities are deliberately arranged to support and enhance various urban activities, making them more efficient than they would be without such infrastructure. Urban facilities include houses, shops, offices, schools, roads, railroads, squares, public institutions, and sewage systems. Together, these elements form the structure of urban space. In other words, an urban space is a collection of interrelated components, and changes in each component can lead to transformations in the city as a whole.

This study focuses on railway stations as key transportation facilities in the cycle of urban change. In modern cities, centers of activity are often aligned with transportation hubs to ensure accessibility. Urban hierarchies are shaped by rail networks, and regional spaces tend to develop around stations. A flexible transportation structure is pursued, in which the railway and road networks function in a complementary manner. In particular, improved accessibility through the expansion of urban rail networks enhances the locational appeal for workers, visitors, and residents, thereby stimulating urban activity. It also helps reduce individual travel costs by encouraging the use of public transportation. These benefits have been shown to contribute to increased real estate prices in areas surrounding railway stations.

Empirical studies―such as those on the Metro in Washington, D.C. (Damm et al., 1980) and the Midway line in Chicago (McDonald et al., 2004)―have shown that expectations regarding new rail infrastructure are capitalized into surrounding property values. As discussed earlier, cities experience life cycles that evolve through stages of growth and decline. If we liken a city to a human body, railway stations function like bones or muscles―essential elements that organically connect different parts through transportation. Accordingly, the area of influence around a station may also follow a recognizable pattern of growth and decline. One of the goals of this study is to empirically demonstrate the life cycle of station areas by analyzing how these areas emerge and transform over time.

The structure of this paper is organized as follows. Section 2 reviews the relevant literature and outlines the research hypotheses. Section 3 examines the study area, focusing on the history of Japanese railways and station-area development. Section 4 describes the empirical strategy employed in the analysis. The findings are presented in Section 5, and finally, Section 6 provides a summary, discussion, and implications for policy.

2 Literature review and research hypothesis

This section surveys relevant literature and establishes the research hypothesis underpinning our study.

2.1 Life cycle approach to urban change

The growth of various types of cities and life cycle approaches to changes in population or spatial structure have been widely explored. Geddes (1915) was the first to conceptualize urban growth as an organic process, describing its evolution through stages of childhood, adolescence, adulthood, and old age. Building on this, Hoyt (1939) applied the life cycle concept to urban decline, while Van den Berg et al. (1982), Dyson (2011), and Davis (2016) divided urban development into stages based on population change. Salvati and Carlucci (2016) further applied this approach to Rome, Italy, empirically categorizing urban change from 1971 to 2011 into four stages: urbanization, suburbanization, slowdown, and reurbanization. Other studies have examined spatial change in residential areas (Lang, 2000) or analyzed urban life cycles across 154 countries based on city size (Cividino et al., 2020).

Most of this research has used life cycle frameworks to examine entire cities, typically using socioeconomic indicators such as population or economic output, as these are the most intuitive proxies for urban growth. However, few studies have applied life cycle approaches to smaller urban units. In contrast, the present study focuses on a more microscopic level: the station area.

2.2 Changes in real estate prices in station areas over time

This study adopts a life cycle approach to examine the processes of growth and decline in station areas, as reflected by changes in real estate prices. The opening of a new railway station tends to increase property values in its vicinity, largely due to enhanced transportation accessibility and improved locational appeal. In particular, changes in surrounding land use and greater connectivity contribute to rising demand. Therefore, the price of real estate in a station area can serve as an important indicator of a station’s influence.

Previous research has employed time series methods to analyze property value changes in station areas. For example, McDonald et al. (2004) examined the capitalization effect of a rapid transit line in Chicago, USA, covering a timeline from 10 years before to 5 years after opening. They found that the effect began before opening, intensified as the opening approached, and gradually declined afterward. Mathur (2020) used regression analysis to estimate housing price trends between 2007 and 2018 in areas with new stations or railway extension projects in California, showing an initial rise followed by a decline over time. Similarly, Singhal and Tyagi (2021) analyzed commercial property prices in station areas of Delhi from 2000 to 2008, demonstrating a comparable pattern of change.

As mentioned earlier, many studies have investigated the impact of improved transportation accessibility through the opening of railway stations or the extension of railway lines―on nearby real estate prices, typically using 10–20 years of data. One reason for this limited timeframe is the lack of digital tools in the 20th century, which made it difficult to construct spatiotemporal datasets from long historical periods. As a result, few empirical studies have explored a timeframe long enough to capture the full life cycle of a station area.

This study seeks to address that gap by analyzing changes spanning approximately 100 years―comparable to a human lifespan―to identify the life cycle of station areas. Rather than tracking temporal change in a few specific locations, we focus on the age of each station. Using a life cycle approach, the study examines real estate price patterns around stations across Japan, ranging from newly established stations to those over 100 years old.

2.3 Research hypothesis, purpose and framework

To address the research gaps identified in previous studies, this paper proposes the following hypotheses.

1) The property prices in areas around railway stations are likely to exhibit cyclical patterns of rises and falls.

2) This cycle is expected to differ between Tokyo, the nation’s capital, and other regions.

Based on empirical validation, this study aims to reveal the lifecycle of station-area developments in relation to rail infrastructure. The findings are interpreted within the historical context of Japan’s station-area development and used to explain regional differences in these cyclical patterns. By analyzing the full span from station openings to aging, the study seeks to offer new insights into the long-term dynamics of railway station areas. The research framework and hypotheses are summarized in Fig. 1.

3 Institutional backgrounds

Japan’s first railway was inaugurated in 1872, connecting Shimbashi Station in Tokyo with Yokohama Station. Subsequently, the nationwide railway network was established through the 1906–1907 Railway Nationalization Act. With the introduction of the Shinkansen in 1964, high-speed rail became a hallmark of Japan’s transport system. In 1987, the privatization of the national railway marked a structural transformation.

The Japanese government strategically expanded its railway system to bolster economic growth and improve transportation infrastructure (Ministry of Land, Infrastructure, Transport and Tourism, 2020). High-speed rail development prioritized Tokyo as a central hub, later extending to integrate other regions. By the 1990s, the foundational Shinkansen framework was completed, though network expansion stagnated until 2004 (Lin and Xie, 2020; Liu et al., 2020). The economic benefits of Shinkansen growth were most prominent in Tokyo, where income levels increased significantly, contrasting with minimal impacts on regional economies (Yoo et al., 2023). Consequently, cities distant from Shinkansen lines lagged in growth. Real estate trends paralleled these economic developments. Following the collapse of Japan’s housing bubble in the 1990s, property markets remained subdued nationwide but rebounded in Tokyo during the late 2010s (Peng, 2022). The United Bank of Switzerland AG (2020) highlighted Tokyo’s distinctive real estate dynamics, noting an annual price increase of 5% since 2014.

With its extensive railway history and diverse network, Japan serves as an ideal subject for analyzing stations aged from newly constructed to over a century old. This study examines the varying life cycles of railway station areas, comparing Tokyo’s distinctive market with other regions to understand regional differences. Figure 2 depicts the railway network across Japan, with a particular focus on the Tokyo region.

4 Empirical strategy

This section outlines the empirical approach used to estimate the lifecycle of railway station areas. Section 4.1 describes the physical boundaries of station areas based on previous research. Section 4.2 discusses the data used in the analysis and provides descriptive statistics. Section 4.3 explains the proposed methodological framework and compares it with existing approaches.

4.1 Defining the scope of station area

To examine the life cycle of station areas, this study analyzed all railway stations across Japan rather than limiting the focus to specific cases. Given the considerable variation in socioeconomic conditions across regions, analyzing the entire network allows for more generalized conclusions and helps minimize regional biases.

Since the analysis is grounded in real estate prices surrounding stations, clearly defining the spatial boundaries of station areas was necessary. Prior empirical studies have demonstrated that a station’s accessibility significantly influences nearby property values within a certain range. For example, Wen et al. (2018) found that the price premium extended up to 2000 m in Hangzhou. Debrezion et al. (2007) identified an impact range of 250–500 m, while Lewis-Workman and Brod (1997) reported a range from 600 to 1600 m. A review of these studies suggests that the station influence typically ranges between 250 and 2000 m.

This study utilized real estate transaction data provided by Japan’s Ministry of Land, Infrastructure, Transport and Tourism (MLIT). Although the dataset does not disclose exact property addresses, it includes the walking time to the nearest station. Accordingly, walking time was used as a proxy for physical distance in the analysis. To support this approach, Bohannon and Andrews (2011), in a meta-analysis of 41 studies on walking speed, found that healthy adults aged 20 to 50 generally walk at a speed of 80–90 m per minute. Based on this and similar findings, the station area in this study was defined as a 15-min walking radius, equivalent to approximately 1200 to 1300 m.

4.2 Data

Japan has a total of 9169 railway stations (Japan Geographic Data Center, 2024). From this set, stations with relatively short travel distances―such as tram stations with lower passenger volumes than heavy rail―were excluded. Additionally, stations lacking transaction data for any part of the period from 2009 to 2022 were removed from the analysis. As a result, the final dataset includes 5098 stations. The study utilized a total of 484,420 real estate transaction records collected between 2009 and 2022, based on data provided by the Ministry of Land, Infrastructure, Transport and Tourism (MLIT). For analytical purposes, the dataset was divided into two groups: Tokyo and non-Tokyo regions. A total of 404,793 observations were used for the Tokyo model, and 79,627 observations were included in the model for non-Tokyo regions.

4.2.1 Constant price

The dependent variable in this study is the transaction price per square meter (in units of 10,000 JPY), adjusted to constant 2009 prices. To account for time-series fluctuations over the 14-year study period, the data were converted using a deflation approach consistent with prior studies (e.g., Damm et al., 1980; Cao and Lou, 2018). The real estate price index used for this adjustment is published monthly by Japan’s Ministry of Land, Infrastructure, Transport and Tourism (MLIT). The formula used to convert nominal prices to constant 2009 prices is provided in Eq. (1).

\begin{array}{}\text{(1)}& {\text{Price}}_{\text{constant}}=\frac{{\text{Price}}_{\text{current}}}{{\text{Price}}_{\text{base}}}\times 100.\end{array}

Price_constant is the constant price of the transaction price (JPY 10,000) per m² used as the dependent variable. Price_current is the actual transaction price, and Price_base is the real estate sales index at the time of transaction.

4.2.2 Attributes data of real estate

Factor determinants such as real estate structural characteristics, neighborhood environment, and regional attributes have long been recognized as key influencers of property values (Chalermpong, 2007; Armstrong and Rodriguez, 2006; Chen et al., 1998). Based on these factors, the independent variables in this study were selected for their relevance to existing price estimation frameworks. Individual property characteristics included land shape, building orientation, zoning type, building structure, road width, land area, floor area ratio (FAR), building coverage ratio (BCR), and frontage width. Variables related to the nearby station area included the age of the nearest station and walking time to the station (in minutes). All variables were obtained from Japan’s Ministry of Land, Infrastructure, Transport and Tourism (MLIT) and processed for analysis. The empirical analysis was conducted using R version 4.3.2 and Stata version 16.

Each of the independent variables used in this study has been validated in previous literature. For instance, land shape was employed as a physical variable in the price estimation model of Glumac et al. (2019). Similarly, variables such as land area (Zhu et al., 2022; Helbich et al., 2014), zoning (Robert et al., 2022), floor area ratio and building coverage ratio (Hong et al., 2020), and accessibility to railway stations (Pickett and Perrett, 1984; Geng et al., 2015; Wang, 2017) have been widely used.

Building direction was categorized as either south-facing or non-south-facing. This distinction is particularly meaningful in East Asian countries such as Japan, South Korea, and China, where the four-season climate creates strong preferences for south-facing buildings, which stay cooler in summer and warmer in winter, thus reducing energy costs. Lu (2018) found that homes facing south were approximately 14% more expensive than those facing other directions. Summary statistics for all variables used in the analysis are presented in Table 1.

4.2.3 Variables with nonlinear relationships

In this study, the variables of station age and travel time to the nearest station were examined in light of prior research that identified a nonlinear relationship between station proximity and real estate prices (Bajic, 1983; Chen et al., 2022; Tordai and Munkacsy, 2022). While previous studies have frequently used the age of the property itself as a determinant of price (Bailey, 2022; Zhu et al., 2022), fewer have considered the age of the nearest railway station as a predictive factor.

Building on the literature reviewed in Section 2.2, which highlights the temporal evolution of station-area real estate prices, this study hypothesizes a nonlinear relationship between station age and property values. To explore this, scatter plot analyses were conducted to examine the relationships between (1) travel time to the station and property price, and (2) station age and property price (Figs. 3 and 4).

The results suggest that these relationships are better described by nonlinear models than by simple linear ones. Specifically, the relationship between travel time and price follows a log-linear or quadratic trend, consistent with prior studies. In contrast, the relationship between station age and price exhibits a more complex, nonlinear pattern. Based on these findings, it is appropriate to apply modeling techniques capable of capturing nonlinear effects.

4.3 Empirical model

The hedonic price model has been used in many studies to predict real estate prices. In the case of the hedonic price model, a linear function is assumed between the real estate price and real estate characteristic variables. In theory, the price function is a concept that includes relationships with many price determinants; thus, accurately determining its form is almost impossible (Mason and Quigley, 1996). Therefore, the form of the price function is an empirical problem rather than a theoretical problem, and researchers are attempting to use nonparametric models to find the price function based entirely on data.

This study aimed to conduct an analysis assuming nonlinearity, which is the change in nearby real estate prices according to the age of the station. Therefore, the analysis was conducted using the generalized additive model (GAM), a nonparametric model based on the existing hedonic price model. The linear regression model of the existing hedonic price model is shown in Eq. (2).

In the GAM, the linear combination ß_jX_ij is replaced with a nonlinear function f_j(x_ij) to reflect the nonlinear relationship between the independent variable and the dependent variable in Eq. (2). Therefore, GAM is expressed as Eq. (3).

\begin{array}{}\text{(2)}& P_i={\beta }_0+{\beta }_1{X}_{i1}+{\beta }_2{X}_{i2}+{\beta }_3{X}_{i3}+\mathrm{...}+{\beta }_p{X}_{ip}+\text{e}\mathrm{,}\end{array}

where P_i = Dependent variable, real estate price;

X_i = Independent variables, factors that can affect real estate prices;

ß_i = Coefficient value of independent variable;

ß₀ = Constant term.

\begin{array}{}\text{(3)}& \begin{array}{ll}{P}_i\hfill & ={\beta }_0+f_1(x_{i1})+f_2(x_{i2})+\dots +f_p(x_{ip})+\text{e}\hfill \\ \hfill & ={\beta }_0+{\displaystyle \sum _{j=1}^p}{f}_j\left(x_{ij}\right)+\text{e}\mathrm{.}\hfill \end{array}\end{array}

In other words, the function f_j is calculated for each explanatory variable x_j and summed. The function f_j can be calculated using various methods, including smoothing spline, local regression, and polynomial regression.

Many studies have used GAM to estimate housing prices by applying a hedonic approach. Pace (1998) demonstrated that the price prediction power of GAM is superior to that of the multinomial regression model, and Bao and Wan (2004) used the smoothing spline method to predict housing prices in Hong Kong and demonstrated that the model prediction power was better than that of existing methods. As a result, in this study, the variables that required confirmation of nonlinear relationships (time to the station, construction age of the station) were constructed as nonlinear models, and the other independent variables were constructed as linear models to form the final model.

5 Results

5.1 Summary of regression results

This study employed a hedonic price model using a generalized additive model (GAM) to examine the nonlinear life cycle of station areas based on station age. Guided by theoretical considerations, the analysis utilized real estate transaction data within a 15-min walking radius of railway stations across Japan. To account for inflation over the 14-year period, transaction prices were converted to constant 2009 values. The dependent variable was defined as the transaction price per square meter, allowing for price standardization across properties of varying sizes.

Prior to the regression analysis, multicollinearity was assessed to ensure the reliability of the explanatory variables. The variance inflation factor (VIF) values for all independent variables were below 10, confirming that multicollinearity was not a concern (Table 2).

The estimation results of the final model are presented in Table 3. The interpretation below focuses on the linear fit of control variables, as the main variable of interest―station age―is discussed in detail later. For land shape, in Tokyo, all categories except for Irregular Quadrilateral were associated with lower property prices. Conversely, in non-Tokyo regions, these shapes were linked to higher prices. Regarding building direction, south-facing properties were priced lower than non-south-facing ones in Tokyo, whereas the opposite trend was observed outside of Tokyo. For zoning, in both regions, all types except commercial zones were associated with lower prices, suggesting that commercial zoning has the strongest positive effect on property value. As for building structure, Steel Reinforced Concrete (SRC) was found to have the most favorable impact on prices across both regions. Among the continuous variables, adjacent road width and floor area ratio (FAR) had positive effects on prices, while land area, building coverage ratio (BCR), and frontage width were negatively associated with property values.

Figures 5 and 6 illustrate the nonlinear effects of station age and travel time to the nearest station, estimated using the generalized additive model (GAM). In both figures, the vertical axis represents the smoothing function of each variable, indicating the relative change in real estate prices. These visualizations allow for a clearer understanding of how these two factors influence property values in a nonlinear manner, capturing more complex patterns than a simple linear regression could reveal.

Figure 5 shows the nonlinear relationship between station age and nearby real estate prices. Both Tokyo and non-Tokyo regions display cyclical fluctuations, although the specific patterns and turning points differ by region. A detailed interpretation of each regional cycle is provided in Section 5.2.

Figure 6 visualizes the relationship between travel time to the station and property prices. The observed pattern is consistent with findings from previous studies, such as Bajic (1983), which reported a logarithmic relationship―property values tend to decrease as travel time from the station increases.

5.2 Empirical results of the life cycle

Figures 7 and 8 expand on the nonlinear relationship between property prices and station age presented in Section 5.1, offering detailed interpretations for each segment of the life cycle. Figure 7 illustrates the station-area life cycle in Tokyo, which displays four notable inflection points―evidence of a cyclical pattern of appreciation and decline.

The first phase extends from the station’s opening to approximately 27.7 years of age. During this period, property prices increase due to the capitalization of expectations surrounding the new station. This is consistent with prior findings: McDonald et al. (2004) reported price increases for about five years post-opening, while Mathur (2020) observed upward trends lasting up to ten years.

The second phase is characterized by a mild decline in the price premium between 27.7 and 39.6 years. Mathur (2020) also found that the initial price gains from new stations tend to diminish over time. In Japan, although there is no legally defined redevelopment timeline, urban renewal efforts often occur to address infrastructure aging and restore urban vitality.

The third phase―between approximately 39.6 and 54.5 years―typically involves redevelopment. In practice, redevelopment often occurs 30–40 years after opening. For instance, Koga Station in Fukuoka and Hiroshima Station underwent redevelopment roughly 30 years after their inauguration. Such efforts usually improve both the station building and surrounding urban services.

Finally, the fourth phase sees a second wave of price appreciation peaking around 54.5 years, largely driven by the positive impacts of redevelopment. Numerous empirical studies support the notion that redevelopment has a substantial influence on property values in station areas.

Numerous studies have shown that redevelopment projects near railway stations lead to increases in surrounding real estate prices. For example, Grimes and Young (2013) conducted a comparative analysis of property prices before and after the redevelopment of the Western Line in Auckland, New Zealand. They observed price increases both at the time of the redevelopment announcement and after its completion.

Similarly, Koster (2017) analyzed property prices in Amsterdam and nearby areas, comparing stations that underwent different types of redevelopment. These included improvements in accessibility (e.g., adding new lines) and simple structural replacements without enhanced access. In both cases, property prices increased following redevelopment.

However, this upward trend does not persist indefinitely. In the analysis, property prices began to decline again, reaching a local minimum at approximately 71.3 years. This decline likely reflects the need for a second round of redevelopment, as about 30 years have passed since the first. In this study, stations that were permanently closed and lost their function were excluded from the analysis. As a result, the dataset includes many aging stations that still operate but are in need of redevelopment. For instance, Hiroshima Station underwent a second redevelopment 37 years after the first, and Shibuya Station in Tokyo―one of the most prominent cases―experienced large-scale commercial, office, and hotel redevelopment approximately 40 years after its initial renewal.

Figure 8 illustrates the station-area life cycle for regions outside Tokyo. Property prices rise during the first 14 years following station opening, reflecting capitalization effects. This is followed by a decline until around 29 years, after which prices increase again due to redevelopment efforts.

From approximately 29 to 43 years, prices rise once more but subsequently decline significantly as stations and surrounding areas undergo pronounced aging, reaching a low point around 68 years. A moderate price increase is observed from 68 to 85 years, likely reflecting a second wave of redevelopment, but a gradual decline resumes thereafter. While the overall pattern shares certain features with the Tokyo case―such as the timing of redevelopment and the cyclical nature of price changes―some distinctions are also evident. A detailed comparative discussion is provided in Section 5.3.

5.3 Comparison Between Tokyo and non-Tokyo regions

This section compares the station-area life cycles of Tokyo and non-Tokyo regions. Figure 9 overlays the nonlinear trends from Figs. 7 and 8 to facilitate direct comparison. In a generalized additive model (GAM), the smoothing function captures the nonlinear effect of an independent variable on the dependent variable. However, since these functions reflect relative changes within each model, their y-axis values are not directly comparable. Consequently, differences in y-values between Tokyo and non-Tokyo curves should be interpreted as relative rather than absolute differences.

In the graph, the blue line represents Tokyo and the red line represents non-Tokyo regions. Key inflection points along each curve are marked, with the station’s age (in years) indicated above each point. One notable difference lies in the duration of the initial price increase. In Tokyo, prices rise for up to 27.7 years after a station’s opening, nearly twice as long as the 14-year period observed in non-Tokyo areas. This extended capitalization effect likely reflects the stronger and more sustained demand in Tokyo’s high-value real estate market. As a result, the timing of redevelopment is also delayed in Tokyo compared to other regions.

When comparing the peak effects of the first redevelopment, Tokyo reaches its highest price point at 54.5 years. In contrast, non-Tokyo areas reach a peak that is comparable to the initial price premium following station opening. In Tokyo, the price premium after opening diminishes more gradually, and the uplift from redevelopment is more pronounced.

Following the second redevelopment phase, both regions experience a decline in property prices due to aging. However, Tokyo exhibits continued price growth after this point, while non-Tokyo areas show only a modest increase before entering a sustained decline.

This divergence is likely attributable to the larger investment scale required for station-area redevelopment in Tokyo, as well as the city’s broader economic base and more dynamic real estate market. These factors make Tokyo more responsive to redevelopment efforts. Moreover, the probability of actual redevelopment taking place is significantly higher in Tokyo than in other regions. As a result, Tokyo’s station-area prices conclude with an upward trajectory, whereas those in non-Tokyo areas enter a long-term downward trend after multiple inflection points.

6 Conclusion

This study highlights a cyclical and nonlinear relationship between real estate prices and the age of railway stations, drawing a parallel to the human life cycle. Just as individuals pass through stages of growth, maturity, and aging, station areas exhibit similar patterns―initial price appreciation, stabilization, and eventual decline.

Using a Generalized Additive Model (GAM) within a hedonic price framework, this study empirically demonstrates how both station age and walking time to the station affect property values in complex, nonlinear ways. For example, newly opened stations contribute to increased accessibility and rising prices, resembling a growth phase. As stations age, prices stabilize, reflecting maturity. Eventually, prices decline, signaling a phase of deterioration―though redevelopment may offer a temporary revival, akin to interventions in human aging.

These lifecycle patterns were particularly pronounced in Tokyo, where redevelopment effects were most evident between 30 and 40 years after a station’s opening. In non-Tokyo regions, the cycles were shorter, reflecting different market dynamics and redevelopment practices. These findings underscore the importance of incorporating station lifecycle stages into urban development and investment planning, especially in aging rail networks across cities.

However, it is important to acknowledge that property values surrounding railway stations are significantly influenced by local economic conditions and region-specific real estate policies. Although this study aimed to reduce regional biases by modeling Tokyo separately due to its distinct property market, and aggregating other areas into a unified model, the analysis could not fully control for each region’s unique economic circumstances and local real estate policies. This constitutes a clear limitation of the current study. Future research could benefit from developing models that explicitly account for these regional economic and policy variations, enabling a more precise and nuanced understanding of the cyclical patterns in property values around railway stations.

The cyclical nature of station-area real estate prices highlights the importance of timely and targeted urban planning and redevelopment policies. Just as well-timed healthcare interventions can improve human quality of life, strategic investments in station redevelopment can revitalize aging infrastructure and extend the functional lifespan of station areas. Policymakers should incorporate station age as a critical variable in urban regeneration strategies, ensuring that redevelopment efforts align with key inflection points in the station lifecycle to maximize their impact. In Tokyo, where redevelopment yields more substantial and immediate price effects, more comprehensive interventions may be warranted. In contrast, lighter and more cost-effective strategies may be suitable for non-Tokyo regions. Moreover, enhancing accessibility and improving transportation infrastructure can help mitigate the price decline phase, offering long-term benefits for both property markets and broader urban development.

References

[1] Armstrong, R.J., Rodríguez, D.A., 2006. An evaluation of the accessibility benefits of commuter rail in eastern Massachusetts using spatial hedonic price functions. Transportation 33 (1), 21–43.

[2] Bailey, J.R., Lauria, D., Lindquist, W.B., Mittnik, S., Rachev, S.T., 2022. Hedonic models of real estate prices: GAM models; environmental and sex-offender-proximity factors. J. Risk Financ. Manag. 15 (12), 601.

[3] Bajic, V., 1983. The effects in Metropolitan Toronto. Urban Stud. 20 (2), 147–158.

[4] Bao, H.X., Wan, A.T., 2004. On the use of spline smoothing in estimating hedonic housing price models: empirical evidence using Hong Kong data. Real Estate Econ. 32 (3), 487–507.

[5] Bohannon, R.W., Andrews, A.W., 2011. Normal walking speed: a descriptive meta-analysis. Physiotherapy 97 (3), 182–189.

[6] Cao, X., Lou, S., 2018. When and how much did the green Line LRT increase single-family housing values in St. Paul, Minnesota? J. Plann. Educ. Res. 38 (4), 427–436.

[7] Chalermpong, S., 2007. Rail transit and residential land use in developing countries: hedonic study of residential property prices in Bangkok. Transp. Res. Rec. 2038 (1), 111–119.

[8] Chen, H., Rufolo, A., Dueker, K.J., 1998. Measuring the impact of light rail systems on single-family home values: a hedonic approach with geographic information system application. Transp. Res. Rec. 1617 (1), 38–43.

[9] Chen, K., Lin, H., Liao, L., Lu, Y., Chen, Y.J., Lin, Z., Teng, L., Weng, A., Fu, T., 2022. Nonlinear rail accessibility and road spatial pattern effects on house prices. Sustainability 14 (8), 4700.

[10] Cividino, S., Halbac-Cotoara-Zamfir, R., Salvati, L., 2020. Revisiting the “city life cycle”: global urbanization and implications for regional development. Sustainability 12 (3), 1151.

[11] Damm, D., Lerman, S.R., Lerner-Lam, E., Young, J., 1980. Response of urban real estate values in anticipation of the Washington Metro. J. Transport Econ. Pol. 14 (3), 315–336.

[12] Davis, D.E., 2016. Reflections on the relations between development and urbanization: past trajectories and future challenges. Int. J. Unity Sci. 20 (1), 1–14.

[13] Debrezion, G., Pels, E., Rietveld, P., 2007. The impact of railway stations on residential and commercial property value: a meta-analysis. J. R. Estate Finance Econ. 35 (2), 161–180.

[14] Dyson, T., 2011. The role of the demographic transition in the process of urbanization. Popul. Dev. Rev. 37, 34–54.

[15] Geddes, P., 1915. Cities in Evolution. Willams and Norgate, London.

[16] Geng, B., Bao, H., Liang, Y., 2015. A study of the effect of a high-speed rail station on spatial variations in housing price based on the hedonic model. Habitat Int. 49, 333–339.

[17] Glumac, B., Herrera-Gomez, M., Licheron, J., 2019. A hedonic urban land price index. Land Use Policy 81, 802–812.

[18] Grimes, A., Young, C., 2013. Spatial effects of urban rail upgrades. J. Transport Geogr. 30, 1–6.

[19] Helbich, M., Brunauer, W., Vaz, E., Nijkamp, P., 2014. Spatial heterogeneity in hedonic house price models: the case of Austria. Urban Stud. 51 (2), 390–411.

[20] Hong, J., Choi, H., Kim, W., 2020. A house price valuation based on the random forest approach: the mass appraisal of residential property in South Korea. Int. J. Strat. Property Manag. 24 (3), 140–152.

[21] Hoyt, H., 1939. The Structure and Growth of Residential Neighborhoods in American Cities. U.S. Government Printing Office.

[22] Koster, J.T., 2017. Assessing the Impact of Redevelopment of Railway Stations on House Prices. University of Groningen (Masters Thesis).

[23] Lang, R.E., 2000. Did neighborhood life-cycle theory cause urban decline? Housing Policy Debate 11 (1), 1–6.

[24] Lewis-Workman, S., Brod, D., 1997. Measuring the neighborhood benefits of rail transit accessibility. Transp. Res. Rec. 1576 (1), 147–153.

[25] Lin, J.-J., Xie, Z.-X., 2020. The associations of newly launched high-speed rail stations with industrial gentrification. J. Transport Geogr. 83, 102662.

[26] Liu, S., Wan, Y., Zhang, A., 2020. Does China’s high-speed rail development lead to regional disparities? A network perspective. Transport. Res. Pol. Pract. 138, 299–321.

[27] Lu, J., 2018. The value of a south-facing orientation: a hedonic pricing analysis of the Shanghai housing market. Habitat Int. 81, 24–32.

[28] Mason, C., Quigley, J.M., 1996. Non-parametric hedonic housing prices. Hous. Stud. 11 (3), 373–385.

[29] Mathur, S., 2020. Impact of transit stations on house prices across entire price spectrum: a quantile regression approach. Land Use Policy 99, 104828.

[30] McMillen, D.P., McDonald, J., 2004. Reaction of house prices to a new rapid transit Line: chicago’s midway line, 1983–1999. Real Estate Econ. 32 (3), 463–486.

[31] Ministry of Land, Infrastructure, Transport and Tourism, 2020. White Paper on Land, Infrastructure, Transport and Tourism in Japan (Tokyo).

[32] Pace, K., 1998. Appraisal using generalized additive models. J. R. Estate Res. 15 (1), 77–99.

[33] Peng, T.-C., Wang, C.-C., 2022. The application of machine learning approaches on real-time apartment prices in the Tokyo metropolitan area. Soc. Sci. Jpn. J. 25 (1), 3–28.

[34] Pickett, M.W., Perrett, K.E., 1984. The effect of the tyne and wear metro on residential property values. (TRRL Supplementary Report, No. 825), Crowthorne: Department of Transport.

[35] Robert, J.G., Zahirovic-Herbert, V., 2022. The influence of privately initiated rezoning on housing prices. Int. J. Hous. Mark. Anal. 15 (3), 592–612.

[36] Salvati, L., Carlucci, M., 2016. In-between stability and subtle changes: urban growth, population structure, and the city life cycle in Rome. Popul. Space Place 22 (3), 216–227.

[37] Singhal, S., Tyagi, Y., 2021. Analyzing the influence of Metro stations on commercial property values in Delhi: a hedonic approach. Real Estate Manag. Valuat. 29 (4), 10–22.

[38] Tordai, D., Munkácsy, A., 2022. The benefit of rail: estimating the impact of station accessibility on residential property prices. Buildings 12 (10), 1760.

[39] van den Berg, L., Drewett, R., Klaassen, L.H., Rossi, A., Vijverberg, C.H., 1982. Urban Europe: a Study of Growth and Decline. Pergamon Press, Oxford.

[40] Wang, X., 2017. Subway capitalization effect in Beijing: theory and evidence on the variation of the subway proximity premium. Pap. Reg. Sci. 96 (3), 495–518.

[41] Wen, H., Gui, Z., Tian, C., Xiao, Y., Fang, L., 2018. Subway opening, traffic accessibility, and housing prices: a quantile hedonic analysis in Hangzhou, China. Sustainability 10 (7), 2254.

[42] Yoo, S., Kumagai, J., Hong, S., Kawasaki, K., Zhang, B., Managi, S., 2023. Economic and air pollution disparities: insights from transportation infrastructure expansion. Transport. Res. Transport Environ. 125, 103981.

[43] Zhu, Z., Zhu, Y., Liu, R., Zhang, L., Yuan, J., 2022. Examining the effect of urban rail transit on property prices from the perspective of sustainable development: evidence from Xuzhou, China. Buildings 12 (10), 1760.

RIGHTS & PERMISSIONS

2095-2635/2026 Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.