00:00:04
yes I'm interested in understanding how
00:00:07
what we call urban former effects travel
00:00:09
behavior so by urban form I mean all
00:00:12
sorts of characteristics of the built
00:00:14
environment you live in so density is
00:00:16
the amount of jobs were achieved to
00:00:20
residency is a quantity of roads and all
00:00:22
of those characteristics and we're
00:00:25
interested in how this is going to
00:00:26
affect the way you travel and how many
00:00:29
miles you actually do and how many miles
00:00:32
you do especially by car why are we
00:00:34
concerned by that because there are some
00:00:37
social cost associated with using
00:00:39
personal transportation so we know for
00:00:42
each mile that we drive society anchors
00:00:45
a caste or three or four cents maybe
00:00:47
well maybe were slightly more associated
00:00:50
with two different forms of pollution
00:00:52
one is greenhouse gas emissions I a
00:00:55
carbon that fosters climate change
00:00:58
global warming and all of that and the
00:01:00
second one is much more localized small
00:01:04
particulates which typically give asthma
00:01:07
to babies and all people and from time
00:01:08
to time they die
00:01:13
so what we found is that urban density
00:01:17
measured by summing the quantity of jobs
00:01:20
and residence in whatever five miles
00:01:23
around where you live was remind someone
00:01:25
where you live
00:01:27
reduces higher urban when density
00:01:29
reduces your travel so the elasticity is
00:01:32
typically 10% higher well density leads
00:01:36
to reduction in traveling of about one
00:01:38
percent
00:01:43
so the conclusions that surprised us
00:01:45
most is the fact that actually we were
00:01:46
expecting the situation and the results
00:01:48
to be fairly complicated and fairly
00:01:50
subtle what we discovered it actually
00:01:53
reality was for once very simple in the
00:01:56
sense that there's actually one major
00:01:58
characteristic of cities that matters
00:02:00
the moment where we live that matters
00:02:02
that's the was that's the density were
00:02:05
around you everything else is really
00:02:08
well secondary and matters far less and
00:02:10
what we also found is that those results
00:02:12
were extremely robust we were also
00:02:14
expecting like different groups of
00:02:16
residents to react differently depending
00:02:18
on where they lived because we were also
00:02:20
expecting people who like driving to
00:02:22
live in places where it's easier to
00:02:24
drive and these to be really important
00:02:26
in the results actually not at all
00:02:28
everybody seems to be reacting more or
00:02:30
less in the same way or at least we
00:02:32
could not find very much of a difference
00:02:34
across different groups across different
00:02:36
types of environments so basically again
00:02:38
if you were if you bring up what density
00:02:40
by about well 10% you reduce driving by
00:02:42
about 1%
00:02:47
even though we do find some effects but
00:02:50
this 10% higher than CG leading to one
00:02:52
percent less driving is actually a very
00:02:54
small effect so if you want to see what
00:02:57
it means or what to understand what it
00:02:58
means in reality if you were to take the
00:03:00
30 million Americans that live at the
00:03:03
lowest weight density in rural areas and
00:03:06
bring them to nearly the highest rent
00:03:08
density in the country that's a scale of
00:03:12
displace that's a displacement where the
00:03:14
scale of Syria at the moment so it's
00:03:16
massive it's not something that's going
00:03:17
to happen at all but that would lead to
00:03:20
a reduction in driving by whatever four
00:03:22
or five percent some even something that
00:03:24
is still extremely drastic but slightly
00:03:26
less so
00:03:26
so taking again 30 million more
00:03:28
Americans from northern city areas which
00:03:30
instead of shootings end up will
00:03:32
completely bring them to high density
00:03:34
areas but leaving some people into low
00:03:36
density areas the effect on electric
00:03:39
driving would be minimal because who's
00:03:41
that who that are broked to high-density
00:03:43
areas will drive less but those that
00:03:45
actually are left in the low density
00:03:47
areas will drive lots more
00:03:49
and this it's nearly what a perfect
00:03:51
world offset so as a result you won't be
00:03:54
able to achieve will very much in terms
00:03:56
of global warming using urban policies
00:03:59
of that sort what you need to use or to
00:04:02
go after well pollution you need after
00:04:04
well local pollution what you need to
00:04:05
tax for congestion ie the concentration
00:04:09
will the concentration but of course in
00:04:11
some areas where the cities so you need
00:04:13
to make them pay for that and you need
00:04:15
to tax well carbon well emissions like
00:04:18
for instance the province of British
00:04:19
Columbia does in Canada it's a
00:04:21
resounding with success so I know it's
00:04:23
politically a really hard proposition in
00:04:25
this country but that's the one thing
00:04:27
that works everything else will not do
00:04:29
very much
00:04:34
so we have this perception that because
00:04:36
indeed the federal government in this
00:04:38
country is not very effective at doing
00:04:40
very many things that's a lot of the
00:04:42
solutions have to be local and you are
00:04:45
whatever Meyers for America and all the
00:04:47
solutions have to be coming from City
00:04:49
Hall's and from Wheeling Meijer to some
00:04:52
extent but global problems they require
00:04:55
global solutions they require federal
00:04:57
interventions we are not going to solve
00:05:00
all global warming by having cities
00:05:02
practicing smart growth that will only
00:05:04
achieve very very little and some of it
00:05:06
like imposing green bells if anything
00:05:09
might be just counterproductive
00:05:14
so what sets us about what I eat
00:05:16
scenario where there's been a lot of
00:05:17
research so what he's setting us about
00:05:19
is first we use extensive data for
00:05:22
travel instead of looking at the tiny
00:05:24
area and the number of cars we actually
00:05:27
rely on a big survey done by the
00:05:29
department of transportation with nearly
00:05:31
a million trips so that's a very large
00:05:33
gauge survey then we use extensive
00:05:35
measures of the local environments where
00:05:38
people reside so what we've done was
00:05:40
actually used satellite imagery and grid
00:05:43
every seeing on a one kilometer by one
00:05:45
kilometer basis both the description of
00:05:47
the physical environment but also using
00:05:49
a vast array of data are coming from the
00:05:51
census coming from employments and so on
00:05:55
and so forth and greeting all that to
00:05:57
know exactly what there is around where
00:06:01
people leave in terms of number of jobs
00:06:03
of number of residents how richest
00:06:04
residents are how educated they are and
00:06:07
this and that so we use vastly more data
00:06:09
also to describe the environment and the
00:06:11
last thing we did it's not because you
00:06:13
observe a negative correlation between
00:06:15
how dance and REI is and how little
00:06:18
people drive that is that this
00:06:21
relationship is causal this relationship
00:06:24
may may just be completely spurious
00:06:26
because again people that like to drive
00:06:28
may want to live in low-density
00:06:30
environment so what we've done was to
00:06:32
design a new methodology to address that
00:06:34
question looking at the fact that people
00:06:36
when you in some cases they move what
00:06:40
will somewhere where car went mm they
00:06:42
moved to neighborhoods that were very
00:06:44
highly well comparable in terms of all
00:06:46
characteristics but some of those
00:06:47
neighborhoods between 2000 and 2010 got
00:06:50
a lot more population and not everybody
00:06:53
moved out or move Dean so we look at
00:06:55
people how they responded to well
00:06:58
differences in the amount of population
00:07:01
went at the guards for zoos that have
00:07:03
stayed for a long time so that allows us
00:07:05
to go sorry to be with technical here
00:07:07
but but dr. Rosen's good to go with
00:07:09
ruvie selection problem and actually
00:07:11
make some causal statements where we can
00:07:13
actually claim that the local
00:07:15
environment has an effect on the amount
00:07:17
of driving one that you do but this
00:07:19
effect again is more
00:07:24
so what's next is I want to think more
00:07:27
about what congestion because we think
00:07:29
of congestion as a big problem it's true
00:07:32
that to where traffic jams are for
00:07:34
everyone to see is everywhere
00:07:36
particularly in developing countries
00:07:38
where people keep well buying cars with
00:07:40
economic growth and the situation gets
00:07:43
worse at the same time we have no
00:07:45
serious idea about the economic cost of
00:07:47
that congestion and for that I want to
00:07:49
be able to use new data that's now made
00:07:52
available through through web mapping
00:07:54
with services like Google Maps where you
00:07:56
can actually get some travel times from
00:07:59
anywhere to anywhere else in real time
00:08:02
knowing about the number one of
00:08:03
travelers you can actually estimate the
00:08:05
true cost of congestion the other thing
00:08:08
I want to look at well more is urban
00:08:11
land use because we tend to think about
00:08:13
land use as a use so it's something that
00:08:15
could be one use one day maybe another
00:08:17
use another day so really key is when we
00:08:20
build stuff it's for a very very long
00:08:22
time and very often that stuff is very
00:08:25
pal it's very well tacky weather to the
00:08:29
needs of a particular user or set of
00:08:31
users so when we think about land use we
00:08:36
can see land use moving from agriculture
00:08:39
to residential and VC's for everyone to
00:08:42
see but what's happening to all
00:08:44
manufacturing it's usually not places we
00:08:46
go to so some of it gets reconverted
00:08:49
whenever it's really really central very
00:08:52
valuable land but there's a lot of
00:08:53
suburban land in declining cities it
00:08:56
looks like nothing is happening so we'd
00:08:58
like to be able to know actually how
00:09:00
much changes well there is in land use
00:09:03
well over time among the more
00:09:06
residential slash manufacturing slash
00:09:09
with different types of development and
00:09:11
we will also like to know if things
00:09:13
revert back with to nature well from
00:09:15
time to time our perception is not but
00:09:18
again this is something that we don't
00:09:19
know
