Changing Residential and Employment Distributions in Guangzhou

1
Changing Residential and Employment Distributions
in Guangzhou

• An Analysis of Commuting Flows

2
Tremendous transformation of China cities

• Gradual weakening of the work-unit system
• Dismantling of work-unit compounds for real
  estate development
• Working and living within a work-unit compound
  increasingly less common

3

• Land re-commodification and the (re)emergence of
  the urban rent gradient
• Re-emergence of the CBD
• Spatial differentiation of land use in response
  to differential rent
• Local government as an active agent in urban
  restructuring
• Development zones and industrial suburbanization
• Central city redevelopment and showcase projects
  by municipal governments

4

• Phenomenal growth in the real estate industry
• Massive real estate developments, especially on
  the urban fringe (in the form of xiaoqu or
  largely gated residential estates)
• Equally massive investments in transport
  infrastructures urban motorways, metro systems

5

• Significantly affects where people live and where
  they work
• Hence the pattern of commuting
• Time and distance (and hence cost) of commute
• Pattern of the commuting flows over the city, and
  implications for traffic congestions
• Means of commuting

6
Data and Methodology

• Two household surveys in Guangzhou
• 2001 1500 households
• 2005 1203 households
• Multi-level probability proportionate to size
  sampling
• To ensure spatial distribution of the sample
  approximates the spatial distribution of the
  population
• Level 1 Urban Districts (qu)
• Level 2 Streets or Sub-districts (jiedao)

7

• Residence/workplace recorded at street
  (???)/postal zone levels
• Both retrospective residential and job histories
  dated back to 1980 available
• Commute distance calculated using Geographic
  Information Systems (GIS)

8
Guangzhou

• Inner Core Yuexiu, Liwan and Dongshan and the
  northern strip of Haizhu
• Inner Suburbs The rest of Haizhu, Tianhe,
  Fancun, southern Baiyun and part of Huangpu
• Outer Suburbs Panyu and the rest of Baiyun and
  Huangpu

9
Changing Distributions of Residence
10
Changing Distributions of Employment
11
Changing Mode of Transport
12
Results

• Mean commute distance increased from 4.6km in
  2001 to 5.7 km in 2005
• Mean commute time from 26 min to 28 min
• gt mean commute speed from 10.4km/h to 12.2 km/h
• Probably a result of the commencement of the
  Metro System in 1999 and subsequent extensions in
  recent years

13
Pattern of Commute, Guangzhou 2001
14
Pattern of Commute, Guangzhou 2005
15
Commuting Pattern

• Despite the longer average commute distance, long
  distance suburb-central city commute actually
  declined
• people lived in inner suburb and commute to
  central city from 46.3 to 26.6
• people lived in outer suburb and commute to
  central city 47.8 to 18.2

16
Factors affecting commute distance Equations
without Location Dummies
17
Gender

• 2001 eqtn gender takes on the wrong sign but
  highly significant
• Lingering influence of work units in assignment
  of housing
• Gender inequality in housing assignment within a
  work unit
• Inclusion of the residential location dummies
  renders it insignificant.
• 2005 eqtn correct sign, although
  insignificant increasing resemblance of western
  norm

18
Income

• Positive and Significant in both 2001 2005
  eqtns.
• Income elasticity increased over time
• Increasing validity of the Alonso-Muth model
  (greater tendency for high income households to
  move to the suburbs) ?
• Inclusion of the residential location dummies
  significantly reduces the income effect in 2005
• Increasingly, income has an effect on commute
  distance via its effect on residential location.

19
(No Transcript)
20
Occupational status

• Relation between occupational status and commute
  distance weak in 2001
• A clear association is discerned in 2005
• Association strengthened significantly with
  inclusion of residential location dummies
• High-status workers more likely to occupy inner
  locations, after controlling for income

21
(No Transcript)
22
Industry and enterprise type

• Workers in govt and Party organizations had
  shorter commutes than those in other work units,
  but the difference was smaller in 2005, compared
  with 2001
• Govt and Party organizations only began to
  earnestly implement housing reforms after 2000
• Workers in manufacturing had shorter commutes in
  2005
• Out-migration of manufacturing jobs have not
  brought about increase in commuting distance

23
Conclusions

• Rapid suburbanization of both residence and jobs
  in Guangzhou
• Intra-zone traffic today dominates the commuting
  scene in both central city and suburbs
• Gender and income effects increasingly resemble
  those found in the West
• But lingering influence of the work unit system
  is still apparent.