Transcript Population Density across the City: The Case of 1900 Manhattan

Population Density
across the City: The
Case of 1900
Manhattan
Jason Barr
Rutgers University, Newark
November 6, 2013
Motivation
• Little work in economics has explored how population bursts
determined land use patterns within the city during period of
rapid urbanization.
• Land use today is a direct product of decisions made over a
century ago.
• Buildings are durable by nature and expensive to tear down.
• Zoning regulations have tended to “lock-in” uses that were in
place circa 1916.
• Landmark preservation also serves similar purpose.
• Aim to test theories of determinants of urban spatial structure
before land use controls.
Current Residential
Buildings Built
before 1917*
*Includes conversions to resid.
Landmark Districts in Manhattan Today
Urban Density in the 19th C.
• Average densities within cities remained
around 125-150 people per acre, despite
rapid urbanization.
• Land annexations kept pace with
population growth.
• However, within cities, very unequal
distribution of populations.
Urban Density in the 19th C
NYC Density Today (2010)
Borough
Note: All densities
are people per acre
2010 Pop
Density
Bronx
1,385,108
51.5
Brooklyn
2,504,700
55.9
Manhattan
1,585,873
112.6
Queens
Staten Island
Pelham, NY
NYC
2,230,722
468,730
12,396
8,175133
32.0
12.4
8.8
42.3
Research Questions
• What role did the pre-European ecology and
topography play?
• How did local amenities, including access to
public transportation, and parks affect density?
• How did location to Broadway, downtown and
rivers affect density?
• Why the Lower East Side?
• Building scale and the problem of tenement
reform.
Basic Theory: Density
Pop
Density 
Acre
 Pop  House   House 


 /

 # Fam  Acre   # Fam 
 Housing Per Acre
 Fam size 
 Housing per Family



 Structural Density (supply)
 Fam size 
 Housing Density (demand)



Basic Theory: Density
• Positively related to amount of housing per acre (e.g.,
more floors per acre, means more people per acre.)
• A higher price for housing will increase structural density.
• Negatively related to housing per person. The greater the
demand for larger houses per family, the lower the
density at a location.
• A higher price for housing will reduce H/Family, and thus increase
density.
• The greater the price of housing per square foot, the
more density, cet. par.
Theory: Demand
• Agents choose housing quantity subject to budge constraint:
Y=Z+P*H+TX.
• TX= cost of commuting distance X
• Assume there is a quantity of neighborhood amenity, A, that
will affect price of housing (e.g. parks, access to
transportation).
• Assume A is negatively related to distance
• Given spatial equilibrium condition:
• Price of housing falls from center to compensate for commuting
costs.
• Price of housing falls from center to compensate for loss of
amenity.
•  Density falls from center
Theory: Income & Density
• Income can affect density in negative or positive way
depending on how higher income people value amenity.
• If rich have high valuation for amenity, then they will
outbid poor to live closer to the center.
• This will bid up price of housing, increasing structural
density, but rich will also use their income to buy more
housing.
• ”Central Park Effect”: High Structural Density but Low
Housing Density.
• Standard relationship between income and distance to
center can be inverted (similar to Paris or Madrid).
Theory: Supply
• Housing reformers decried the fixed plot sizes of
25’ x 100’ as being bad for health and promoting
excess density.
• They argued that building tenements on larger
plots would reduce building density.
• This would not be true if there were constant
returns to scale with building.
Theory: Supply
• Building housing with capital (K)--# floors, and Land (L):
 
  PK L  rK  L
• If we assume L is exogenously determined, then structural
density—floors per acre (FAR):
 P 
K 

 r 
•  if CRS,
1
1 
K  P 
FAR   

L  i 

1
L
1
1
Manhattan
• Amenities (hypothesized affect on density):
•
•
•
•
Access to public transportation (the Els). (+)
Access to parks (+)
Access to high ground (+)
Closeness to “paisanos” (+)
• Disamenities:
• Poor drainage (+ or -)
• Location-based (for jobs, retail or recreation):
• Distance to downtown (+)
• Distance to Broadway (+)
• Distance the rivers (+ or -).
• Supply: Block size (-)
Mannahatta
• Pre-European ecology could affect later settlement
patterns in a few ways.
• Early lock-in: If early development creates local
economies of scale and fixed-cost “lock-in”, then
attractive locations in the 17th century are likely to
predict density in the early 20th century.
• Locations on Manhattan favorable to early agriculture might be a
predictor of later density.
• Location of Oak Tulip trees is important because they grow on land
good for agriculture (on slopes, not too wet, not too dry).
• Health and disease:
• low-lying wetlands were breeding grounds for disease; higher
elevations were more salubrious.
• Sewage removal was more likely to be worse in low lying areas.
Mannahatta
Sanderson (2009)
Mannahatta
Mannahatta
Els Map
Density at the Block Level
Immigration Patterns
Germany
Ireland
Italy
Russia
0.6
0.5
0.4
0.3
0.2
0.1
0
-
2.00
4.00
6.00
8.00
MA of density vs distance from City Hall
Results w/o Demographics
The Lower East Side
The Lower East Side
foreign
Extreme Density and Enthicity
Russian/Pol
Aus. Hung.
1300
1200
1100
1000
900
800
-
0.200
0.400
0.600
0.800
1.000
Density vs HHI
Density vs Ethnicity in LES
1300
1200
1100
1000
900
800
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Extreme Density and Family Size
1300
1200
1100
1000
900
800
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
Density vs Fraction of Pop. With 6 or more people per
Household on Block
Economies of Scale Results