The Impact of Built Environment on the Commuting Distance of Middle/Low-income Tenant Workers in Mega Cities Based on Nonlinear Analysis in Machine Learning

Abstract

The issues of housing and traffic in China's mega cities have become increasingly pressing problems, particularly for middle/low-income tenant workers. These tenants are from less advantaged socioeconomic backgrounds, which has resulted in a significant geographical separation between their workplace and their residence. Although a large number of studies have confirmed that built environment factors have a solid impact on residents’ commuting distance, few studies have investigated the mechanism underlying the nonlinear influence on middle/low-income tenants. This paper aims to provide an in-depth analysis of the key factors and nonlinear influencing mechanism of the built environment on middle/low-income tenant workers’ commuting distance by establishing a gradient-boosting decision tree model, using Beijing as an empirical case. The paper reveals three primary findings: (1) An important nonlinear relationship between the surrounding built environment and peoples’ jobs–housing spatial proximity can be observed for those middle/low-income tenant workers who use slow and public modes of commuting. Specifically, the density of public transport stations, road networks, and workplaces, and the land use mix play a dominant role. (2) A limited effect of built environment factors can be found for the same group of tenant workers who choose cars as their mode of commuting. (3) The differences in self-selected commuting modes have a significant mediating effect on the relationship between the built environment and jobs–housing situation among middle/low-income tenant workers. Given this, effective policy guidance for residents’ travel modes is necessary to optimize the built environment indicators to achieve the best effect. In addition, we should consider giving priority to the matching indicators such as land use mix and resident population density. Another possibility is to strengthen the connection to the public transport stations, which in turn can optimize the walkability in residential environments.

Keywords

Built environment / Jobs–housing balance / Rental housing / Middle/low-income group / Gradient boosting decision tree / Beijing

References

[1] Kain J. Housing segregation, Negro employment, and metropolitan decentralization. Q J Econ, 1968, 82(2): 175-197

[2] Cervero R. Jobs-housing balancing and regional mobility. J Am Plan Assoc, 1989, 55(2): 136-150

[3] Giuliano G. Is jobs housing balance a transportation issue?. Transp Res Rec J Transp Res Board, 1991, 1305: 305-312.

[4] Horner M. Extensions to the concept of excess commuting. Environ Plan A Econ Space, 2002, 34(3): 543-566

[5] Zheng Z, Zhou SH, Deng XD. Exploring both home-based and work-based jobs-housing balance by distance decay effect. J Transp Geogr, 2021, 93: 103043

[6] Blumenberg E, King H. Jobs–housing balance re-re-visited. J Am Plan Assoc, 2021, 87(4): 484-496

[7] Levinson D. Accessibility and the journey to work. J Transp Geogr, 1998, 6(1): 11-21

[8] Zhao PJ, Lv B, De Roo G. Impact of the jobs-housing balance on urban commuting in Beijing in the transformation era. Journey Transp Geogr, 2011, 19(1): 59-69

[9] Cervero R. Transit-based housing in California: evidence on ridership impacts. Transp Policy, 1994, 1(3): 174-183

[10] Boarnet M, Crane R. The influence of land use on travel behavior: specification and estimation strategies. Transp Res Part A Policy Pract, 2001, 35(9): 823-845

[11] Cao XY, Mokhtarian P, Handy S. Neighborhood design and vehicle type choice: evidence from Northern California. Transp Res Part D Transp Environ, 2006, 11(2): 133-145

[12] Cao XY, Schoner J. The influence of light rail transit on transit use: an exploration of station area residents along the Hiawatha line in Minneapolis. Transp Res Part A Policy Pract, 2014, 59(C): 134-143

[13] Valenzuela A, Schweizer L, Robles A. Camionetas: informal travel among immigrants. Transp Res Part A Policy Pract, 2005, 39(10): 895-911

[14] Renne JL, Bennett P. Socioeconomics of urban travel: evidence from the 2009 national household travel survey with implications for sustainability. World Transp Policy Pract, 2014, 20(4): 7-27.

[15] Hu LQ, Schneider R. Different ways to get to the same workplace: how does workplace location relate to commuting by different income groups. Transp Policy, 2017, 59: 106-115

[16] Hu LQ. Racial/ethnic differences in job accessibility effects: explaining employment and commutes in the Los Angeles region. Transp Res Part D Transp Environ, 2019, 76: 56-71

[17] Ha J-H, Lee S, Ko J-H. Unraveling the impact of travel time, cost, and transit burdens on commute mode choice for different income and age groups. Transp Res Part A Policy Pract, 2020, 141: 147-166

[18] Niedzielski MA, Horner MW, Xiao N. Analyzing scale independence in jobs-housing and commute efficiency metrics. Transp Res Part A Policy Pract, 2013, 58: 129-143

[19] Zhou X, Yeh AGO. Understanding the modifiable areal unit problem and identifying appropriate spatial unit in jobs–housing balance and employment self-containment using big data. Transportation, 2021, 48: 1267-1283

[20] Weitz J. Jobs-housing balance, 2003, Chicago, IL: American Planning Association

[21] Qin P, Wang LL. Job opportunities, institutions, and the jobs-housing spatial relationship: case study of Beijing. Transp Policy, 2019, 81: 331-339

[22] Gordon P, Richardson HW, Jun M-J. The commuting paradox evidence from the top twenty. J Am Plan Assoc, 1991, 57(4): 416-420

[23] Peng Z. The jobs-housing balance and urban commuting. Urban Stud, 1997, 34(8): 1215-1235

[24] Anas A, Arnott R, Small KA. Urban spatial structure. J Econ Lit, 1998, 36(3): 1426-1464.

[25] Downs A. Still stuck in traffic: coping with peak-hour traffic congestion, 2004, Washington, DC: Brookings Institution Press

[26] Schafer A. Regularities in travel demand: an international perspective. J Transp Stat, 2000, 3(3): 1-31

[27] Painter G, Liu CY, Zhuang D. Immigrants and the spatial mismatch hypothesis: employment outcomes among immigrant youth in Los Angeles. Urban Stud, 2007, 44(13): 2627-2649

[28] Sultana S. Job/housing imbalance and commuting time in the Atlanta metropolitan area: exploration of causes of longer commuting time. Urban Geogr, 2002, 23(8): 728-749

[29] Jin J, Paulsen K. Does accessibility matter? Understanding the effect of job accessibility on labour market outcomes. Urban Stud, 2018, 55(1): 91-115

[30] Cui BE, Boisjoly G, El-Geneidy A, Levinson D. Accessibility and the journey to work through the lens of equity. Transp Geogr, 2019, 74: 269-277

[31] Hanson S, Pratt G. Spatial dimensions of the gender division of labor in a local labor market. Urban Geogr, 1988, 9(2): 180-202

[32] Raphael S, Stoll MA. Job sprawl and the suburbanization of poverty, 2010, Washington, DC: Metropolitan Policy Program at Brookings

[33] Hu L, Wang L. Housing location choices of the poor: does access to jobs matter?. Hous Stud, 2019, 34(10): 1721-1745

[34] Ewing R, Cervero R. Travel and the built environment: a meta-analysis. J Am Plan Assoc, 2010, 76(3): 265-294

[35] Jin J. The effects of labor market spatial structure and the built environment on commuting behavior: considering spatial effects and self-selection. Cities, 2019, 95: 102392

[36] Yang LY, Ding C, Ju Y, Yu B. Driving as a commuting travel mode choice of car owners in urban China: roles of the built environment. Cities, 2021, 112: 103114

[37] Tian L, Yao ZH, Fan CJ, Zhou L. A systems approach to enabling affordable housing for migrants through upgrading Chengzhongcun: a case of Xiamen. Cities, 2020, 105: 102186

[38] Li C, Li J, Chan X. The crowding out effect of housing price rise on residents' consumption in China. Stat Res, 2014, 12: 32-40

[39] Sanchez TW, Shen Q, Peng ZR. Transit mobility, jobs access and low-income labour participation in US metropolitan areas. Urban Stud, 2004, 41(7): 1313-1331

[40] Benner C, Karner A. Low-wage jobs-housing fit: identifying locations of affordable housing shortages. Urban Geogr, 2016, 37(6): 883-903

[41] Ouyang W, Wang BY, Tian L, Niu XY. Spatial deprivation of urban public services in migrant enclaves under the context of a rapidly urbanizing China: an evaluation based on suburban Shanghai. Cities, 2017, 60: 436-445

[42] Ding C, Cao X, Nass P. Applying gradient boosting decision trees to examine non-linear effects of the built environment on driving distance in Oslo. Transp Res Part A Policy Pract, 2018, 110: 107-117

[43] Tao T, Wang JY, Cao XY. Exploring the non-linear associations between spatial attributes and walking distance to transit. J Transp Geogr, 2020, 82: 102560

[44] Zhang WJ, Zhao YJ, Cao XY, Lu DM, C YW, Nonlinear effect of accessibility on car ownership in Beijing: pedestrian-scale neighborhood planning. Transp Res Part D Transp Environ, 2020, 86: 102445

[45] Ding C, Cao XY, Liu C. How does the station-area built environment influence Metrorail ridership? Using gradient boosting decision trees to identify non-linear thresholds. J Transp Geogr, 2019, 77: 70-78

[46] Shao QF, Zhang WJ, Cao XY, Yang JW, Yin J. Threshold and moderating effects of land use on metro ridership in Shenzhen: implications for TOD planning. J Transp Geogr, 2020, 89: 102878

[47] Yang JW, Su PR, Cao XY. On the importance of Shenzhen metro transit to land development and threshold effect. Transp Policy, 2020, 99: 1-11

[48] Yang JW, Cao XY, Zhou YF. Elaborating non-linear associations and synergies of subway access and land uses with urban vitality in Shenzhen. Transp Res Part A Policy Pract, 2021, 144: 74-88

[49] Zhang Y, Chai YW. The spatio-temporal activity pattern of the middle and the low-income residents in Beijing China. Sci Geogr Sin, 2011, 31(9): 1056-1064

[50] Xia XX, Lin KX, Ding Y, Dong XL, Sun HJ, Hu BB. Research on the coupling coordination relationships between urban function mixing degree and urbanization development level based on information entropy. Int J Environ Res Public Health, 2021, 18(1): 242

[51] Saha D, Alluri P, Gan A. Prioritizing highway safety manual’ s crash prediction variables using boosted regression trees. Accid Anal Prev, 2015, 79: 133-144

[52] Friedman J. Greedy function approximation: a gradient boosting machine. Ann Stat, 2001, 29(5): 1189-1232

[53] Zhang YR, Haghani A. A gradient boosting method to improve travel time prediction. Transp Res Part C Emerg Technol, 2015, 58: 308-324

[54] Yang SY, Wu JP, Du YM, He YQ, Chen X. Ensemble learning for short-term traffic prediction based on gradient boosting machine. J Sens, 2017, 2024: 1-15.

Funding

Beijing Outstanding Young Scientist Program(JJWZYJH01201910003010)

National Natural Science Foundation of China(52108056)