within a 5- year horizon of financing. This means, although the property tax increment
can finance more investment than desired in open space, that public investment may not
be socially optimal.
Panel B and C illustrates the location effect. Although both the net social value of
open space and property tax increment demonstrate similar trends with respect to the area
of open space, the peak-value size of open space is different in three spatial locations. As
we can see, the peak-value size is 400 acres for the open space located to the right and
300 acres for located in the upper community relative to community center, which
implies the maximum net social value of open space could be different for different
locations. In our simulation, the maximum net social value of open space that can be
reached is the highest with the central location. Also, tax increment curve is different in
three locations. When the community park is located to the right of community center,
the maximum capacity of tax increment financing is around 1130 acres, while it drops to
1010 acres when open space is located in the upper community.
These changes in both the financing capacity of property tax increment and the
net social value caused by varying locations can be attributed to the effect of community
shape on the externality of open space. Preserving open space in a community can be
considered as producing an amenity field, analog to the physical gravity or magnetic
field, in which each location is associated with an amenity generated by that open space.
When open space is located in community center, most, if not all, of its positive
externality is captured in the value of the land within community boundaries (or more
community land are covered by open space amenity). But when open space is not
centrally located, it is very likely that relatively less positive externality is captured into
28