00:00:04
I work on urbanization in developing
00:00:06
countries and in my recent research I
00:00:09
have looked at the spatial patterns of
00:00:11
urbanization in India trying to
00:00:13
understand the economic implications of
00:00:16
different city structures and more
00:00:18
specifically I've looked at one
00:00:20
particular property of urban footprints
00:00:22
which is their geometry if you look at
00:00:24
urban areas from a plane for example you
00:00:27
can see that cities become in different
00:00:28
shapes some have outlines that are
00:00:31
roughly circular others appear more
00:00:33
fragmented and yet others are
00:00:35
constrained by their geographies for
00:00:37
instance cities on islands or peninsulas
00:00:39
end up having even more irregular shapes
00:00:41
so my question was what influence if any
00:00:44
does a cities shape have on the location
00:00:47
choices of consumers and firms do
00:00:50
consumers and firms benefit in terms of
00:00:53
welfare or productivity from locating in
00:00:55
cities with particular shapes now why
00:00:59
would we expect City shape to matter at
00:01:00
all from an economic standpoint
00:01:02
primarily because as urban planners have
00:01:05
long recognized City shape is one of the
00:01:08
determinants of urban commuting
00:01:10
efficiency along with more widely
00:01:12
studied factors such as infrastructure
00:01:14
or travel demand and the reason is that
00:01:17
more compact more circular city
00:01:20
geometries are conducive to shorter
00:01:22
within city trips to give you a sense of
00:01:25
how this works this is just coming from
00:01:27
the properties of geometric shapes think
00:01:29
of a circle and now imagine that holding
00:01:32
area constant you start tweaking the
00:01:34
circle turning it into some irregular
00:01:36
shape as the shape of this polygon
00:01:39
departs from that of a circle becoming
00:01:41
less and less compact the average
00:01:43
distance between points within this
00:01:46
polygon will tend to increase and the
00:01:48
same thing happens to cities as they
00:01:50
expand in space
00:01:55
a good real-world example of these types
00:01:58
of interactions is the comparison
00:02:00
between kolkata and bangalore of the top
00:02:03
cities in india kolkata is the least
00:02:05
compact one it kind of looks like an
00:02:07
upside down giraffe head with a long
00:02:09
north-south tentacle so to say bangalore
00:02:13
on the other hand is more like a
00:02:14
pentagon it's the most compact one i
00:02:17
calculate that all else being equal if
00:02:19
kolkata had the same compact shape that
00:02:22
bangalore has the average within city
00:02:25
trip proxied by the average distance
00:02:27
between points within the city would be
00:02:30
shorter by six point two kilometers and
00:02:32
this is not trivial if you think that
00:02:34
the average commuting speed in the top
00:02:35
cities in india is around 15 kilometers
00:02:38
per hour so what i did was to look at
00:02:41
the outlines of over 450 cities in india
00:02:45
based on how they appear in nighttime
00:02:47
satellite imagery if you look at a
00:02:49
picture of the earth at night taken from
00:02:51
a satellite urban areas are clearly
00:02:54
visible as patches of light essentially
00:02:57
so for each city for each of these
00:03:00
patches of light i computed a number of
00:03:02
indicators for geometric shape that are
00:03:05
basically telling us the extent to which
00:03:07
a city's shape departs from that of a
00:03:10
circle and I did this for multiple years
00:03:12
to have a sense of the dynamic evolution
00:03:15
of City shapes over time and finally I
00:03:18
linked City shape to economic outcomes
00:03:20
measured at the city level such as
00:03:22
population aggregate look sorry average
00:03:25
local wages and average housing rents
00:03:31
so I find that households locating in
00:03:34
more compact cities are better off first
00:03:37
more compact cities attract larger
00:03:40
populations second households that are
00:03:42
located in more compact cities are
00:03:44
implicitly paying a premium in terms of
00:03:47
higher housing rents net of local wages
00:03:49
in order to live in cities with better
00:03:52
geometries so this tells us that
00:03:54
essentially consumers are valuing City
00:03:58
compactness as an urban amenity that
00:04:00
they are willing to implicitly pay for
00:04:02
in this implied premium that they are
00:04:05
paying is actually quite large i
00:04:06
estimate that households are trading off
00:04:08
implicitly five percent of their income
00:04:10
in order to live in cities that are more
00:04:13
compact by one standard deviation now
00:04:16
for the average City a one standard
00:04:18
deviation improvement in compactness
00:04:20
implies that your average round-trip
00:04:22
commute decreases by 720 meters these
00:04:27
are potential commutes estimated based
00:04:29
on the distribution of points within the
00:04:32
city these are not observed commutes
00:04:34
which unfortunately I can't see in my
00:04:35
data on the other hand firms are not
00:04:39
affected in their productivity by City
00:04:41
shape compact compact cities are just as
00:04:44
productive and compactness appears to be
00:04:47
a pure consumption amenity now you may
00:04:50
worry that compact cities are
00:04:52
systematically different from others in
00:04:54
some dimension other than geometry what
00:04:56
if these more compact cities are also
00:04:58
more successful more developed cities
00:05:01
that consumers like for reasons that are
00:05:03
just correlated with geometry but have
00:05:05
nothing to do with it to address the
00:05:09
fact that these confounding factors
00:05:10
might be biasing my results in my
00:05:13
analysis I'm not simply making a crude
00:05:16
comparison between compact and non
00:05:18
compact cities what I do is look at what
00:05:21
happens when a given City becomes less
00:05:24
compact as a result of hitting some
00:05:26
topographic obstacle like a mountain or
00:05:29
a lake as it expands in space and what I
00:05:31
find is that when a city ends up with
00:05:34
the worst geometry because of hitting
00:05:37
some geographic constraints its
00:05:39
population growth slows down and housing
00:05:42
rents net of local wages tend to
00:05:44
decrease
00:05:48
well India's predicted to urbanize very
00:05:51
rapidly in the next decades and local
00:05:54
policymakers need options to accommodate
00:05:57
this urban growth in space and my
00:05:59
findings suggest that accommodating this
00:06:01
urban expansion in a more or less
00:06:03
compact way can have potentially
00:06:06
relevant implications for household
00:06:07
welfare and for urbanization patterns
00:06:09
now clearly not all cities can be
00:06:12
circular due to topography and in fact
00:06:15
at a descriptive level one of the
00:06:17
patterns that I find is that cities tend
00:06:19
to become less and less compact over
00:06:21
time as they expand but there is room
00:06:23
for land use regulations to promote more
00:06:27
or less compact development and as an
00:06:29
example of this in my work I consider a
00:06:32
regulatory tool that is considered quite
00:06:34
controversial in India right now
00:06:36
vertical limits restrictions on building
00:06:39
heights that in India are considered
00:06:41
very strict relative to international
00:06:44
standards and a lot of commentators and
00:06:46
local policymakers are concerned that
00:06:48
these stringent vertical limits are
00:06:50
inducing sprawl and making cities more
00:06:52
and more spread out what I find is that
00:06:55
cities that have more restrictive
00:06:57
building height regulations end up
00:07:00
having footprints that are not only
00:07:01
larger but also less compact relative to
00:07:04
what their topography and predicted
00:07:06
growth would suggest so relaxing these
00:07:09
constraints would be a way at least to
00:07:11
slow down the deterioration in City
00:07:13
shape that city growth entails and a
00:07:17
second policy implication could be
00:07:18
related to investments in urban
00:07:20
infrastructure one of my findings is
00:07:23
that the negative consequences of bad
00:07:26
city geometry are more pronounced in
00:07:28
cities that have a poor road network so
00:07:31
providing road infrastructure might be a
00:07:33
way to compensate consumers for the poor
00:07:37
shape of the city that they live in
00:07:42
well so far in my research I have looked
00:07:45
at City level aggregate outcomes and I
00:07:47
have compared the outlines of cities
00:07:49
looking at cross city comparisons what
00:07:53
I'd like to do next is take this to a
00:07:54
more disaggregated level and look at
00:07:56
what happens within cities look at the
00:07:58
internal structure of cities as a
00:08:00
follow-up to this project it would be
00:08:03
important to understand who bears the
00:08:05
costs of longer commutes in cities that
00:08:08
are not compact what categories of
00:08:10
consumers are affected the most perhaps
00:08:12
low-income households that don't have
00:08:14
access to individual means of
00:08:16
transportation what consequences if any
00:08:19
does City shape have on residential
00:08:22
patterns and in particular residential
00:08:24
segregation by socioeconomic status are
00:08:26
some kinds of City shapes more conducive
00:08:29
to segregation and others more conducive
00:08:31
to integration
00:08:50
you
