![Page 1: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/1.jpg)
Challenge the future
DelftUniversity ofTechnology
Negative Externalities of Structural Vacant OfficesPhilip Koppels and Hilde Remøy
![Page 2: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/2.jpg)
2/27
Introduction
• High levels of (structural) vacancy• Concentration in specific areas and buildings• Social security & criminality
• Negative externalities related to structural vacant buildings?
• Positive externalities are often assumed• Landmark buildings• Public space: proximity to parks & water bodies
• Difficult to measure and quantify: spatially lagged variables• Function of distance
Negative Externalities and Structural Vacancy
![Page 3: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/3.jpg)
3/27
IntroductionNegative Externalities and Structural Vacancy
Source:Thibodeau (1990)
![Page 4: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/4.jpg)
4/27
Introduction
• Isolating the effect:• Market conditions: rental adjustment equation• Omitted variables: proxy for neighborhood quality
Negative Externalities and Structural Vacancy
![Page 5: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/5.jpg)
5/27
Structural Vacancy
• “Structural” nature: three consecutive years• Visibility of vacancy: at least 50% of the building is vacant
Operationalization
![Page 6: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/6.jpg)
6/27
2003
Number of Structural vacant buildings = 1
![Page 7: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/7.jpg)
7/27
2004
Number of Structural vacant buildings = 8
![Page 8: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/8.jpg)
8/27
2005
Number of Structural vacant buildings = 17
![Page 9: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/9.jpg)
9/27
2006
Number of Structural vacant buildings = 34
![Page 10: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/10.jpg)
10/27
2007
Number of Structural vacant buildings = 32
![Page 11: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/11.jpg)
11/27
2008
Number of Structural vacant buildings = 29
![Page 12: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/12.jpg)
12/27
2009
Number of Structural vacant buildings = 38
![Page 13: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/13.jpg)
13/27
2010
Number of Structural vacant buildings = 54
![Page 14: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/14.jpg)
14/27
Research Method
Data sample:152 buildings334 transactions2003-2010
Hedonic Pricing
![Page 15: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/15.jpg)
15/27
Model SpecificationSpatial Segmentation
Discrete spatial Heterogeneity
12 submarkets
Modelled by random effects
![Page 16: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/16.jpg)
16/27
Model Specification
Rent
X coordinate
Discrete Spatial Heterogeneity
Submarket A Submarket B Submarket C
![Page 17: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/17.jpg)
17/27
Model SpecificationOmitted Variables
Nested model: Submarkets Postcode area’s
Buildings
Modelled by random effects
![Page 18: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/18.jpg)
18/27
Model SpecificationGeneral Economic Trend (City-wide)
Nominal Rent Level
![Page 19: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/19.jpg)
19/27
Sub-market Specific Economic Trends
Nominal Rent Level
Model Specification
![Page 20: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/20.jpg)
40.000 m² 500 m²
Building size (GFA)
office buildings
structural vacant office buildings
50 m r=500m
450 m
300 m
![Page 21: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/21.jpg)
21/27
Weight factor
Spatially Lagged VariableModel Specification
Distance0 R=250m. R=500m. R=750m.
1
R=150m.
How to measure distance:As the crow-fliesAlong the road network
![Page 22: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/22.jpg)
22/27
Distance
Weight factor
0 R=250m. R=500m. R=750m.
1
R=150m.
Spatially Lagged VariableModel Specification
![Page 23: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/23.jpg)
50 m r=500m
450 m
300 m
1
1 10,000450
1
1 10,000300
1
1 1,00050
22.22 33.33 20 75.55
Impact depends on:DistanceBuilding size
![Page 24: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/24.jpg)
24/29
Negative Externalities of Structural Vacant Offices
• Best performing model assumes:• Constant distance relationship with threshold distance at
250m.• Distance measured: as the crow-flies
0R=250m.
1
Distance
• Impact: independent of the size of the structural vacant building
![Page 25: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/25.jpg)
25/29
Solution for Fixed Effects
EffectTransformation
Estimate
Std. Err. Pr > |t|
Intercept 51.167 0.2667 <.0001Building age Logarithmic -0.06412 0.01621 0.0001Travel time to highway Logarithmic 0.01296 0.02816 0.6457Intercity location (r=250m): Yes 0.03733 0.05951 0.5310Intercity location (r=250m): No - - -Distance tram/metro stop Logarithmic -0.01702 0.02837 0.5491Daily amenities Logarithmic 0.02676 0.02033 0.1901
Employment industry Logarithmic -0.000010.000027 0.5967
Employment F&B Logarithmic 0.03777 0.02738 0.1704Inverse parking ratio Logarithmic 0.1154 0.03405 0.0008Garage: No -0.09504 0.03125 0.0025Garage: Yes - - -Number of floors Logarithmic 0.05236 0.02718 0.0549Façade: Low/Standard -0.03755 0.02961 0.2056Façade: High - - -Company logo: No -0.02808 0.02359 0.2348Company logo: Yes - - -Size of entrance as % of total LFA Logarithmic 0.02787 0.01406 0.0482Numb. Structural Vacant offices (r=250m) -0.02332 0.01203 0.0537
![Page 26: Challenge the future Delft University of Technology Negative Externalities of Structural Vacant Offices Philip Koppels and Hilde Remøy](https://reader036.vdocument.in/reader036/viewer/2022081603/5697bf741a28abf838c7f78b/html5/thumbnails/26.jpg)
26/27
Conclusion & Discussion
• Base model seems plausible• Significant relationship structural vacant offices and rent
prices nearby buildings• Vicious circle?• Collective action
• Some issues to consider• Proxy• Weight matrix specification• Submarket dummies
Negative Externalities and Structural Vacancy