Cities look like networks.

They look like networks.

They're supplied by networks.

And they really are networks, because, and
you heard a little bit of that this morning,

this is a social network, these are people,
and those lines are the connections between them.

This is a city.

Cities are people. What they really are is
a physical manifestation of this, meaning

the interactions between people and the way
people cluster beginning with the individual,

the family, and the various groupings that go on.

It is that kind of structure, and that hierarchy
of structures, that gets manifested in the

physicality of a city, and I'll come back 
to that briefly at the end.

So the question is, do all these networks 
have the same kinds of properties that

we saw in biology?

That is, do cities scale?

That is, is New York a scaled-up Los Angeles, 
which is a scaled-up San Francisco,

which is a scaled-up Santa Fe?

They don't look alike, they look completely 
different.

They have different histories, geography,
cultures. They are built of different things, etc.

Nevertheless, if you believe this viewpoint, 
they're all actually manifestations of this stuff

going on. So maybe there are, in fact,
such systematic underlying constraints

in cities and companies.

Well, this is the first work I got involved in, 
this German colleague, he's a fancy guy,

Dirk Helbing, is a billion dollar, billion Euro 
project to take some of these ideas to

the social sciences is at the ETH in Zurich.

This was a trivial one - we asked about the
number of gas stations as a function of city size

So it's the same plot that we did in biology, 
except the x-axis - the horizontal axis - is

population (that's the proxy for the size
of a city, and what we found is scaling,

scales, systematic scaling.

That is, the number of gas station scales in
a systematic way with the size of a city,

everywhere in Europe.

And I'm just going to jump now to the 
conclusion of this piece, it turns out, what

you notice on this, not only do they all scale, they all
scale pretty much in the same way.

The slope of all those is around 0.8.
The gradient is about 0.8.

This is true around the world.

We've looked at countries around the world, 
it's the same thing.

Not only that, if you look at any infrastructure: 
length of the roads, length of electrical cables,

length of the gas lines, anything you can 
think of to do with infrastructure looks

exactly like this.

They scale, and they scale with the same 
slope, and the slope is less than one,

meaning there is an economy of scale. 
Not surprising, need less gas stations

per capita the bigger the city.

What is surprising: it's systematic and 
it's the same anywhere in the world.

And it's the same whether I look at gas 
stations, road length, electrical line,

whatever you can think of.

Well, that's interesting, this is kind of
like biology,

but what was really interesting was to 
look at socioeconomic quantities,

quantities that have to do with us talking to 
one another, and did not exist on this planet

until we started talking to one another and forming 
communities, maybe ten thousand years ago.

Things like wages, and creative people,
so-called supercreative people like you lot.

Okay, so what's plotted here is the same kind
of graph, on the top is wages per capita in the US.

And this is the number of you lot as a function 
of city size and you see there's quite a bit of

fluctuation, but there's very strong evidence 
for scaling and what is most interesting about

this is that the slope of these lines are bigger
than one.

Systematically, you have more wealth per capita, 
more wages per capita, more fancy people per capita.

you produce more patents per capita,
you have more crime per capita,

all to the same degree anywhere in the world.

So here it is, we just plunked a whole
bunch of things together. These are things

that are seemingly not very connected:
GDP of a city, income of a city, crime in

the city, and number of patents in the city,
and they all fall pretty much on the same line,

indicating that something universal is going on.

Because it's not just universal across the 
metrics, it's universal everywhere,

meaning the same graph is true in Japan, 
as it is in China, as it is in Portugal, as it is in Chile.