Land Use-Transport Interactions: Evidence from and Implications for Urban Public Transportation Systems 26 April, 2013 Professor Christopher Zegras Department of Urban Studies & Planning Massachusetts Institute of Technology
Jan 21, 2015
Land Use-Transport Interactions:
Evidence from and Implications
for Urban Public Transportation
Systems
26 April, 2013
Professor Christopher Zegras
Department of Urban Studies & Planning
Massachusetts Institute of Technology
Content
• Built Environment (BE) = f (Transport) and
Transport = f (BE)
– Background and basic theory
• Transport = f (BE)
– theory, evidence, policy implications.
• BE = f (Transport)
– theory, evidence, policy implications.
• Conclusions and Questions
Land Use-Transport Interaction:
Theoretical Framework
Land Use
Land Uses (Activities)
Land, Floor Space
Prices Demand
Transportation
Travel (Activities)
Transportation System
Time
Costs Demand
Connectivity
Spatial
Distribution
Accessibility
The Metropolis in Development
– Two Core Phenomena 1
95
0
19
55
19
60
19
65
19
70
19
75
19
80
19
85
19
90
19
95
20
00
20
05
20
10
20
15
20
20
20
25
20
30
20
35
20
40
20
45
20
50
—
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
Po
pu
lati
on
(M
illio
ns
)
“Less Developed”
Urban
“Developed” Urban
Total World
Source: United Nations, Department of Economic and Social Affairs (DESA)
% Change Population by Census Tract (2000-10)
US
Census
2012
Paris
Angel et al, 2011
Bandung,
Indonesia
Angel et al, 2011
Average Tract Density: 20 US Metro Areas
Angel et al., 2011
World “Suburbanization” Trends
Angel et al., 2011
Transport = f (LU)? Something new?
Meyer, et al, 1965 (from Kain, 1999) Howard’s “Garden City”
11
The Built Environment and Mobility: A
Question of Scale
Scale Refers To Built Environment
Concepts/Indicators
Metropolitan
Urban Structure Overall City Size,
population, gross density,
“skeletal” forms (e.g, radial)
Intra-
Metropolitan
(meso)
Urban Form Dispersion, concentration,
mixes, grain, access
networks
Micro Scale:
(neighborhood)
Urban Design “Internal Texture”, Density,
Mixes of Uses, Street
Networks, etc.
Ingram, 1998, p. 1027.
Urban Density (persons/hectare)
15,000
10,000
5,000
100 200 300 400
Per
Cap
ita C
ar
Km
s
Hong Kong
Sacramento, CA
?
?
xSantiago
13 US Cities
7 Canadian Cities
3 Wealthy Asian Cities
11 European Cities
6 “Developing” Asian Cities
6 Australian Cities
Urban Density (persons/hectare)
15,000
10,000
5,000
100 200 300 400
Per
Cap
ita C
ar
Km
s
Hong Kong
Sacramento, CA
?
?
xSantiago
13 US Cities
7 Canadian Cities
3 Wealthy Asian Cities
11 European Cities
6 “Developing” Asian Cities
6 Australian Cities
Kenworthy & Laube, 1999.
Newman & Kenworthy…
Macro-
Scale
Form &
Function
Bertaud, 2004
14
Micro Scale Built Environment
Crane, 1996
15
Formalizing the Theoretical
Framework
16
Crane’s Trip-Based (Time/Cost-Based)
Framework
Crane, 1996
17
A Trip-Based (Cost-Based)
Framework
Auto Travel
Demand
Indicator
Grid Street
(shorter trips)
Traffic
Calming
(slower trips)
Mixed Uses &
Densification
(one trip, more
purposes,
slower speed
All Three
Car Trips
Increase (for
all modes,
likely)
Decrease Increase or
Decrease
Increase or
Decrease
Vehicle Miles
Traveled
(VMT)
Increase or
Decrease Decrease
Increase or
Decrease
Increase or
Decrease
Car Mode
Choice
Increase or
Decrease Decrease
Increase or
Decrease
Increase or
Decrease
Crane, 1996
18
To Better Understand Possible
Effects…
We need to know
• Elasticities of trip demand with respect to
speed and distance
• Cross-elasticities among modes
– How changes for one mode (eg in distance)
affects demand for other modes
• Differentiate by trip purpose
Net Utility Approach
• Extending beyond Crane…
• The Built Environment influences disutility
and utility
Maat et al, 2005
Stylized Effects of Travel Time
Changes
Maat et al, 2005
Stylized Effects of Mode Changes
Maat et al, 2005
22
Net Utility Framework
• Land uses influence net utility: – Positive utility = activity realization
– Negative utility (disutility) = travel cost
• Extends beyond Crane – Reveals a dual ambiguity of land use’s influences
• Uncertain influence on trip costs (disutility), thus travel
• Uncertain influence on activities (utility), thus travel
• What happens with saved time? A. Invest in going to higher utility destinations
B. Carry out more activities
C. Dedicate more time per activity
– Travel demand increases with?
– A and B
– Consistent with…. constant travel time budgets (e.g., Schafer, 2000).
TB = f (BE)?
Empirical Challenges: Unclear
pathways of effects Transport-Efficient
Neighborhood
Transport-Efficient
Behavior
Transport-Efficient
Preferences Spatial cognition, etc…
A “Macro-Level” Example
Netherlands
Policy Land Use Behavior
(Schawen et al, 2004)
National-Level Planning Policies Netherlands
• 1970s-1980s – “concentrated decentralization”
• 1980s – “compact urban growth”
– with urban renewal subsidies
• 1990s – “A-B-C location policy”
• A: centrally located sites
• B: outside CBDs, but still public transport connected
• C: highway-oriented sites
• Challenge: growth in service/office sector
• Retail policy
• Overall: mixed success – Primarily guiding residential and retail development
Schwanen et al, 2004.
Netherlands: Estimated Effects?
• Data
– Travel
• One-day travel survey (NTS)
• Male/female Head of Household
– Land Uses
• Macro: urban structure (mono-, poly-centric)
• Meso: degree of urbanization
• Travel Effects
– Mode Choice
– Distance and time
Schwanen et al, 2004.
Netherlands: Conclusions & Recs
Schwanen et al, 2004.
“Micro-Scale” Effects
Boston, Jinan
Land Use in Boston Work Trip
Mode Choice Model
Zhang, 2004
Micro-level Example: BE and BRT Pedestrian
Catchment Area (PCA) in Jinan China (Jiang et al, 2012)
Arterial- Edge Corridor
(Jingshi St.)
1
(Jiang 2010)
Integrated- Boulevard Corridor
(Lishan Rd.)
2
(Jiang 2010)
Below- Expressway Corridor
(Beiyuan St.)
3
(Jiang 2010)
Approach
𝐷𝐼𝑆𝑇𝑖 = 𝑓(𝑇𝑀𝑖 ,𝑇𝑅𝑖 , 𝑆𝑖 ,𝐶𝑖 ;𝛽) + 𝜀𝑖
• Station area user survey
• Built Environment Analysis
• Regression
CORRIDOR WALKABILITY
A BRT Users’ Perspective
29% 33% 33%
26% 26% 28%
18%
24% 26%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Crossing is safe. Crossing is easy. Walking on sidewalksis safe.
Arterial-edge(n=464)
Integrated-boulevard(n=356)
Below-expressway(n=946)
Unsafe crossing, poor signals…
(Jiang 2010)
Distance… (Jiang 2010)
(Jiang 2010)
CORRIDOR WALKABILITY
A BRT Users’ Perspective
69%
47% 45% 50%
33%
24%
38% 35%
27%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Pavement is good. Streets are clean. Few blockages are onsidewalks.
Arterial-edge(n=464)
Integrated-boulevard(n=356)
Below-expressway(n=946)
CORRIDOR WALKABILITY
A BRT Users’ Perspective
48%
42%
70%
58%
39%
49%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Trees on sidewalks makewalking comfortable.
Facilities along streetsmeet my demand.
Arterial-edge(n=464)
Integrated-boulevard(n=356)
Below-expressway(n=946)
Walk next to trees… Arterial-Edge Corridor
Walk under trees… Integrated-Boulevard Corridor
Walk without trees… Below-Expressway Corridor
475
647
582
329
501 459
0
100
200
300
400
500
600
700
Avg Walking Distance
Avg Straight-line Distance
(m)
Detour
Factor 1.59 1.36 1.33
CORRIDOR WALKABILITY
Directness
Walking
distance
Straight-line
distance
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%0
15
0
30
0
45
0
60
0
75
0
90
0
10
50
12
00
13
50
15
00
16
50
18
00
19
50
21
00
22
50
24
00
25
50
27
00
28
50
30
00
31
50
33
00
34
50
36
00
37
50
39
00
Pe
rce
nta
ge o
f B
RT
rid
ers
Access/Egress Walking Distance (m)
Terminal Station
Transfer Station
Typical Station
Station Function vs. Access/Egress Walking Distance
Walking Distance (m) Typical Station Transfer Station Terminal Station
Mean 547 587 1365
Median 435 458 1311
Maximum 2738 2067 5114
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%0
15
0
30
0
45
0
60
0
75
0
90
0
10
50
12
00
13
50
15
00
16
50
18
00
19
50
21
00
22
50
24
00
25
50
27
00
28
50
30
00
31
50
33
00
34
50
36
00
37
50
39
00
Pe
rce
nta
ge o
f B
RT
rid
ers
Access/Egress Walking Distance (m)
Arterial-Edge
Integrated-Boulevard
Below-Expressway
Corridor Type vs. Access/Egress Walking Distance (non-terminal stations only)
Walking Distance (m) Arterial-Edge Integrated-Boulevard Below-Expressway
Mean 475 649 580
Median 412 520 458
Maximum 1635 2023 2738
Potentially confounding factors
Trip Maker
• Age
• Gender
• Car Ownership
• Household Income
• Occupation
• Frequent BRT User or not
Trip
• Purpose
• Time
• Alternative Mode Availability
• In Group or not
System
• Level of Service
• Transit Fare
Station Context
• Station Function (terminal, transfer?)
• Distance to City Center
• Density Gradient
• Connectivity (Feeder road length)
• Level of Feeder-bus Service
No need control because BRT riders are granted free transfer
between BRT lines and thus using the same system per se.
Catchment Area Density Gradient: Hill/ Valley/ Flat
Hill Pattern (convex) Valley Pattern (concave)
BRT
BRT
Station 3 Station 8
STATION CONTEXT
Source: http://jinan.edushi.com/
E(Walk Distance)
= 600
+ 150 *(Integrated_Boulevard_Corridor)
+ 400 *(Terminal_Station)
- 100 *(Transfer_Station)
- 150 *(Density_Hill)
+ 150 *(Density_Valley)
+ 50 *(Distance_to_Center in km)
Radial Distance Guidelines for Pedestrian Zones around
BRT Stations AND RRT Stations
Radial Distance (meters) Corridor Type Terminal Station Non-terminal Station
BRT Arterial-Edge 600-1000 300-600
BRT Integrated-Boulevard 1000-1500 600-1000
BRT Below-Express 800-1200 400-800
RRT Underground 1200 700-900
RRT Elevated 1300 800-1000
Jiang et al, 2012; Zhao & Deng, 2013
E(Walk Distance)
= 900*(Underground typical sta.)
+ 300 *(Terminal_Station)
+ 100 *(Elevated Station)
- 100 *(if Transfer station)
+ 10 *(Distance_to_Center in km)
Terminal station presents a unique opportunity
for large transit-oriented development…
RECOMMENDATIONS
(Jiang 2010)
This probably will NOT work…
(Jiang 2010)
RECOMMENDATIONS
Make crossing safer…
(Jiang 2010)
Put more trees and stores along the sidewalk
in an appropriate way… (Jiang 2010)
T = f (BE)
An Example Policy Implication
A “Demand Side” Example:
Location Efficient Mortgage
• Also known as “Smart Commute
Mortgage”
• Basic Theory:
– Driving less increases household disposable
income
– Can qualify for better mortgage characteristics
(higher mortgage-to-income qualifying ratio)
– Basically attempt to capitalize on the location-
transport cost trade-off
Decision Process
1. Household relocating (potentially in the
market)
2. Interested in buying (in the market)
3. Attracted to “location efficient” areas
4. Qualified to buy
5. Interested in LEM
Hypothetical Example
Item Without LEM With LEM
Applicant Income
(per month)
$2,100 $2,100
Available for down
payment
$6,000 $6,000
Housing to Income
Ratio Limit
28% 28%
Transport Savings
(per month)
n.a. $653
Mortgage Available $76,000 $115,611
Major Risks…
• LEM has the effect of reducing the down
payment as share of property value
• Assumes household will
– Reduce vehicle ownership
– Reduce transport expenses
“Testing the Rhetoric” • Basic hypothesis
– Location efficiency reduces mortgage risk
• How to test?
– “Efficient” locations should be negatively correlated with
mortgage default rates, ceteris paribus
• Data
– 8,000 mortgages from 1,000 census tracts in Chicago
• Analytic Approach
– Probability of Default = f (Sociodemographic and other
controls, location efficient characteristics)
• Findings
– Location factors have no influence on default rates
Blackman, 2002; Blackman & Krupnick, 2001
LEM: Interpretations &
Implications Possible Explanations
• Savings not large enough to influence
– Counter-factual (location inefficient location) is
inaccurate
– VMT and ownership model wrong
• Or, real estate market already capitalizing
financial benefits.
– i.e., value already “captured”
Implications
• Might still have other benefits
• But, must be weighed relative to costs
Land Use = f (Transport)?
Muller, 2004
Rail Transit Effects (Baum-Snow & Kahn, 2000)
Aims
1. How new rail transit attracts commute
trips to transit
2. Which demographic groups benefit most
from rail improvements
3. Rail transit influence on land values
Approach
• Case Studies
– Expansions
• Boston, Chicago
– Comprehensive New Networks
• Atlanta, Washington, DC
– Incremental Expansion
• Portland, OR
Possible Rail Transit Effects
• Existing Residents Switch to Rail
• New Residents Move into Transit Tracts
• Property Values Increase
Data
• Census Tract Data
• Public Use Microdata Sample (PUMS)
– 1% sample, micro data
• Constructed Transit Coverages to represent system changes (1980-1990)
– Show declines in mean tract distance from transit (all cities): 5 km to 3 km
Analytical Approach
• Transit Use: 3 models 1. Use = f (Tract Distance)
2. Change in use = f (Change in Tract Distance)
3. Change in use = f (Change in Tract Distance, Migration)
• Transit Capitalization – “Hedonic” home price capitalization
– Change in home price = f (change in distance)
• Transit Beneficiaries – Change in Distance to Transit = f (demographics)
Results: Transit Use
• There is some Tiebout migration of transit users to tracts – i.e., “self-selection”
– Migration rates are higher in tracts with increased transit access
• Induced transit-oriented development
• Also, transit-shifting by existing residents – In fact, most mode shift due to this effect
• Overall effects… – Small 1.4% increase in transit with a 2 km decrease
in distance to transit (from 3 to 1 km)
Results: Transit Capitalization &
User Groups
• 3 km to 1 km decrease in transit distance
increases rents by $19/month, house
value by $5,000
– More gain in travel time savings: $1,200/year
• College educated and home-owners more
likely to be in census tracts closer to
transit
Relative Suburban Benefits from
Rail Transit
Baum-Snow & Kahn, 2005.
Public
Tra
nsit U
se b
y D
eca
de f
or
16 C
itie
s
tha
t E
xpand
ed R
ail
Tra
nsit (
1970
-20
00
)
Some Problems with Baum-Snow & Kahn
• City fixed effects – Transit markets/service very local
• Ignore other investments/policies occurring at same time – E.g., highway investments
– And their expansionary effects
• Rail transit almost certainly retains central city vitality – Not captured in their model
– No employment effects captured in model
• Commute trips only
• Possible issues with using census tract…
See, e.g., Voith, 2005.
Bus Rapid Transit Effects
Transmilenio Case
~Current Network
114 Stations; 84 Kms; 1263 vehicles; 27 km/h; 200K peak hour passengers
83 Feeder routes; 516 feeder buses
Hidalgo, 2006.
Calle 13 – Av. Caracas
Vehicles
Graftieaux, 2005.
Stations
Graftieaux, 2005.
Transmilenio BRT: Land Effects?
Rodriguez and Targa (2004) Approach
• Estimate Effects on Property Values – Hedonic Model
• Rental Properties – Feb-Apr, 2002
– Field visits and newspaper adds
– All properties for rent
– 494 multifamily residential properties
• Dependent variable – Asking price
• Influencing variables (of interest) – Accessibility (local and regional)
1.5 km
buffer
Accessibility: How Measured? • Local
– Shortest walking time on road network from location of each property to closest BRT
• Regional
– Line-haul travel time from closest BRT station to Financial District
– Line-haul travel time from closest BRT station to Financial District Downtown
– Weighted index of travel time to all BRT stations • Weighted by the number of passengers travelling
between each pairs
Other Variables
• Proximity effects
– Straight line distance to corridor
– To capture possible negative externalities
• Control variables
– Apartment: Size, # bedrooms, age, etc.
– Location: buffer with spatial average of zone
attributes
• Crime, socioeconomic, demographic, land uses,
etc.
Results
• Elasticity of rent with respect to BRT stop dist. – -0.16 to -0.22
• Every five minutes from BRT stop, rent declines by US$15
• Elasticity of rent with respect to BRT Corridor – 0.19 to 0.21
• Every 100 meters from corridor, rent goes up by US$77
Comparing Results
• Results (in terms of % change in property value)
fairly comparable to
– Los Angeles Blue Line
– DC WMATA
• Slightly lower than San Diego (LRT) and UK
Tramlink (Manchester)
• Estimated absolute premium (annualizing rents)
– US$440-650 per 100 meters
– Roughly Double the Baum-Snow & Kahn Effect
(measured from 3 to 1 km change)
Other Notes and Commentary
• No apparent Regional Accessibility Benefit
• Short time frame of analysis may mean
conservative estimate
• Cross-sectional analysis
• Corridor effect might be confounded
– By other traffic
• But, station effects might also be confounded
– E.g., urban recovery
• Residential land only
Urban Recovery
Hidalgo, 2006.
Commercial Development
Hidalgo, 2006.
Commercial Development
Hidalgo, 2006.
BE = f (Transport)
An Example Policy Implication
Chicago: Hedonic Model, CTA
Station Access
p = f (I, N, T)
where:
p is the property sales price;
I is a vector of attributes of the improvements on the parcel, such as number of bathrooms, number
of floors, and age, etc.;
N is a vector of attributes of the neighborhood, such as quality of public facilities and services
(including schools) and socioeconomic characteristics; and,
T is a combined vector of attributes of the transportation-related locational accessibility of the
parcel, such as proximity to transportation services (including transit), relative accessibility to
opportunities across the broader metropolitan area, etc.
Variation in Elasticity of Property Value with
Respect to Walking Time Based on Properties’
Walk Times to CTA Station
Land-Based Finance Mechanisms
Derived from Lari et al, 2009
Rail Transit Value Capture
Potential: Chicago, Lisbon
Zegras et al 2013b
94
Transit = f (BE): Summary
• Consider the geographical scale of analysis/intervention – Generally, theory implies same types of effects, operating at
different scales
• Theoretically, impacts are ambiguous
• Complexity of LUT relationships increases with society’s complexities – Time routines, age, family cycle, etc.
– Keep in mind the type of potential activities (e.g., trip purpose) and related spatial and temporal constraints
• Simple consideration: BE influence on walk influence to station access
BE = f (Transit): In Summary • Public Transit, in right conditions, will influence
urban form
• Land Value effects are consistently seen
• Institutionality is barrier to land value capture
(LVC)
– Including poor transport finance pictures
• LVC not a panacea
• Realistic amount to raise, will be modest, in most
cases
• Ex-ante system in place (before build/expand)
BRT Centre May Webinar
Cost Efficiency under Negotiated Performance-Based
Contracts and Benchmarking – Are there gains through
Competitive Tendering in the absence of an Incumbent
Public Monopolist?
Friday, May 24th at 4pm Sydney, Australia time (UTC+10)
Presented by Professor David Hensher
Institute of Transport and Logistics Studies
The University of Sydney
References • Angel, S., J. Parent, D. Civco, A. Blei (2011) Making Room for a Planet of Cities, Policy Focus
Report, Lincoln Institute of Land Policy.
• Baum-Snow, N. and M. Kahn (2000) The effects of new public projects to expand urban rail
transit. Journal of Public Economics, Vol. 77, pp. 241-263.
• Bertaud, A. (2004) The spatial organization of cities: Deliberate outcome or unforeseen
consequence? May: http://alain-
bertaud.com/images/AB_The_spatial_organization_of_cities_Version_3.pdf
• Blackman, A. (2002) Testing the Rhetoric. Regulation (Spring): 34-38.
• Crane, R. (1996) On form versus function: Will the new urbanism reduce traffic, or increase it?
Journal of Planning Education and Research, Vol. 15, pp. 117-126.
• Geurs, K.T. and B. van Wee (2004) Accessibility Evaluation of Land-Use and Transport
Strategies: Review and Research Directions. Journal of Transport Geography Vol. 12: 127-140.
• IBI Group. 2000. Greenhouse Gas Emissions from Urban Travel: Tool for Evaluating
Neighborhood Sustainability. Healthy Housing and Communities Series Research Report,
prepared for Canada Mortgage and Housing Corporation and Natural Resources Canada,
February.
• Graftieux, P. (2005). World Bank, Personal communication.
• Hidalgo, D. (2006). EMBARQ, Personal communication.
• Ingram, G. (1998) Patterns of Metropolitan Development: What Have We Learned? Urban
Studies, Vol. 35, No. 7, June, pp. 1019-1035.
• Jiang, Y. (2010). CSTC, personal communication.
• Jiang, Y., C. Zegras, Mehndiratta, S. (2012). Walk the line: station context, corridor type and bus
rapid transit walk access in Jinan, China.” Journal of Transport Geography, 20(1), 1–14.
References (cont’d)
• Kain, J. (1999) The Urban Transportation Problem: A Reexamination and Update. Essays in
Transportation Economics and Policy. Brookings.
• Kenworthy, P. and F. Laube (1999) Patterns of automobile dependence in cities: an international
overview of key physical and economic dimensions with some implications for urban policy.
Transportation Research A, Vol. 33, pp. 691-723.
• Lari, A., Levinson, D., Zhao, Z., Iacono, M., Aultman, S. Das, K.V., Junge, J., Larson, K.,
Scharenbroich, M. (2009) Value Capture for Transportation Finance: Technical Research Report.
Minneapolis: The Center for Transportation Studies, University of Minnesota
• Maat, K., B. van Wee, D. Stead (2005) Land use and travel behaviour: expected effects from the
perspective of utility theory and activity-based theories. Environment and Planning B: Planning
and Design, Vol. 32, pp. 33-46.
• McNally, M. and A. Kulkarni. (1997) Assessment of Influence of Land Use-Transportation System
on Travel Behavior. Transportation Research Record 1607, pp. 105-115.
• Muller, Peter O. Transportation and Urban Form: Stages in the Spatial Evolution of the American
Metropolis. Chapter 3 in The Geography of Urban Transportation, 59-85. S. Hanson, ed. 3rd
edition, Guildford Press, 2004
References (cont’d) • Rodríguez, D. and Targa, F. (2004) Value of Accessibility to Bogotá’s Bus Rapid Transit System.
Transport Reviews, Vol. 24, No. 5 (September): 587-610.
• Schwanen, T., Dijst, M. and Dieleman, F. (2004) Policies for Urban Form and their Impact on
Travel: The Netherlands Experience. Urban Studies Vol. 41, No. 3: 579-603.
• US Census Bureau (2012) Patterns of Metropolitan and Micropolitan Population Change: 2000 to
2010, Census Special Reports, September.
• Voith, R. (2005) Comment on Effects of Urban Rail Transit Expansions: Evidence from Sixteen
Cities, 1970–2000 (Baum-Snow and Kahn). Brookings-Wharton Papers on Urban Affairs: 198-
206.
• Zegras, C., S. Jiang, C. Grillo (2013a) Sustaining Mass Transit through Land Value Taxation?
Prospects for Chicago, Draft Paper prepared for Lincoln Institute of Land Policy.
• Zegras, C., S. Jiang, C. Grillo, L. Martinez (2013b) Capture the Value to Finance Transit
Systems? A Comparative Assessment of Chicago and Lisbon, Draft.
• Zhang, M. (2004) The Role of Land Use in Travel Mode Choice: Evidence from Boston and Hong
Kong. Journal of the American Planning Association, Vol. 70, No. 3, Summer, pp. 344-360.
• Zhao, J. and Deng, W. (2013) Relationship of Walk Access Distance to Rapid Rail Transit
Stations with Personal Characteristics and Station Context. Journal of Urban Planning and
Development (forthcoming).