ABOUT THE SPEAKER
Geoffrey West - Theorist
Physicist Geoffrey West believes that complex systems from organisms to cities are in many ways governed by simple laws -- laws that can be discovered and analyzed.

Why you should listen

Trained as a theoretical physicist, Geoffrey West has turned his analytical mind toward the inner workings of more concrete things, like ... animals. In a paper for Science in 1997, he and his team uncovered what he sees as a surprisingly universal law of biology — the way in which heart rate, size and energy consumption are related, consistently, across most living animals. (Though not all animals: “There are always going to be people who say, ‘What about the crayfish?’ " he says. “Well, what about it? Every fundamental law has exceptions. But you still need the law or else all you have is observations that don’t make sense.")

A past president of the multidisciplinary Santa Fe Institute (after decades working  in high-energy physics at Los Alamos and Stanford), West now studies the behavior and development of cities. In his newest work, he proposes that one simple number, population, can predict a stunning array of details about any city, from crime rate to economic activity. It's all about the plumbing, he says, the infrastructure that powers growth or dysfunction. His next target for study: corporations.

He says: "Focusing on the differences [between cities] misses the point. Sure, there are differences, but different from what? We’ve found the what."

More profile about the speaker
Geoffrey West | Speaker | TED.com
TEDGlobal 2011

Geoffrey West: The surprising math of cities and corporations

Filmed:
1,583,030 views

Physicist Geoffrey West has found that simple, mathematical laws govern the properties of cities -- that wealth, crime rate, walking speed and many other aspects of a city can be deduced from a single number: the city's population. In this mind-bending talk from TEDGlobal he shows how it works and how similar laws hold for organisms and corporations.
- Theorist
Physicist Geoffrey West believes that complex systems from organisms to cities are in many ways governed by simple laws -- laws that can be discovered and analyzed. Full bio

Double-click the English transcript below to play the video.

00:16
Cities are the crucible of civilization.
0
1000
3000
00:19
They have been expanding,
1
4000
2000
00:21
urbanization has been expanding,
2
6000
2000
00:23
at an exponential rate in the last 200 years
3
8000
2000
00:25
so that by the second part of this century,
4
10000
3000
00:28
the planet will be completely dominated
5
13000
2000
00:30
by cities.
6
15000
3000
00:33
Cities are the origins of global warming,
7
18000
3000
00:36
impact on the environment,
8
21000
2000
00:38
health, pollution, disease,
9
23000
3000
00:41
finance,
10
26000
2000
00:43
economies, energy --
11
28000
3000
00:46
they're all problems
12
31000
2000
00:48
that are confronted by having cities.
13
33000
2000
00:50
That's where all these problems come from.
14
35000
2000
00:52
And the tsunami of problems that we feel we're facing
15
37000
3000
00:55
in terms of sustainability questions
16
40000
2000
00:57
are actually a reflection
17
42000
2000
00:59
of the exponential increase
18
44000
2000
01:01
in urbanization across the planet.
19
46000
3000
01:04
Here's some numbers.
20
49000
2000
01:06
Two hundred years ago, the United States
21
51000
2000
01:08
was less than a few percent urbanized.
22
53000
2000
01:10
It's now more than 82 percent.
23
55000
2000
01:12
The planet has crossed the halfway mark a few years ago.
24
57000
3000
01:15
China's building 300 new cities
25
60000
2000
01:17
in the next 20 years.
26
62000
2000
01:19
Now listen to this:
27
64000
2000
01:21
Every week for the foreseeable future,
28
66000
3000
01:24
until 2050,
29
69000
2000
01:26
every week more than a million people
30
71000
2000
01:28
are being added to our cities.
31
73000
2000
01:30
This is going to affect everything.
32
75000
2000
01:32
Everybody in this room, if you stay alive,
33
77000
2000
01:34
is going to be affected
34
79000
2000
01:36
by what's happening in cities
35
81000
2000
01:38
in this extraordinary phenomenon.
36
83000
2000
01:40
However, cities,
37
85000
3000
01:43
despite having this negative aspect to them,
38
88000
3000
01:46
are also the solution.
39
91000
2000
01:48
Because cities are the vacuum cleaners and the magnets
40
93000
4000
01:52
that have sucked up creative people,
41
97000
2000
01:54
creating ideas, innovation,
42
99000
2000
01:56
wealth and so on.
43
101000
2000
01:58
So we have this kind of dual nature.
44
103000
2000
02:00
And so there's an urgent need
45
105000
3000
02:03
for a scientific theory of cities.
46
108000
4000
02:07
Now these are my comrades in arms.
47
112000
3000
02:10
This work has been done with an extraordinary group of people,
48
115000
2000
02:12
and they've done all the work,
49
117000
2000
02:14
and I'm the great bullshitter
50
119000
2000
02:16
that tries to bring it all together.
51
121000
2000
02:18
(Laughter)
52
123000
2000
02:20
So here's the problem: This is what we all want.
53
125000
2000
02:22
The 10 billion people on the planet in 2050
54
127000
3000
02:25
want to live in places like this,
55
130000
2000
02:27
having things like this,
56
132000
2000
02:29
doing things like this,
57
134000
2000
02:31
with economies that are growing like this,
58
136000
3000
02:34
not realizing that entropy
59
139000
2000
02:36
produces things like this,
60
141000
2000
02:38
this, this
61
143000
4000
02:42
and this.
62
147000
2000
02:44
And the question is:
63
149000
2000
02:46
Is that what Edinburgh and London and New York
64
151000
2000
02:48
are going to look like in 2050,
65
153000
2000
02:50
or is it going to be this?
66
155000
2000
02:52
That's the question.
67
157000
2000
02:54
I must say, many of the indicators
68
159000
2000
02:56
look like this is what it's going to look like,
69
161000
3000
02:59
but let's talk about it.
70
164000
3000
03:02
So my provocative statement
71
167000
3000
03:05
is that we desperately need a serious scientific theory of cities.
72
170000
3000
03:08
And scientific theory means quantifiable --
73
173000
3000
03:11
relying on underlying generic principles
74
176000
3000
03:14
that can be made into a predictive framework.
75
179000
2000
03:16
That's the quest.
76
181000
2000
03:18
Is that conceivable?
77
183000
2000
03:20
Are there universal laws?
78
185000
2000
03:22
So here's two questions
79
187000
2000
03:24
that I have in my head when I think about this problem.
80
189000
2000
03:26
The first is:
81
191000
2000
03:28
Are cities part of biology?
82
193000
2000
03:30
Is London a great big whale?
83
195000
2000
03:32
Is Edinburgh a horse?
84
197000
2000
03:34
Is Microsoft a great big anthill?
85
199000
2000
03:36
What do we learn from that?
86
201000
2000
03:38
We use them metaphorically --
87
203000
2000
03:40
the DNA of a company, the metabolism of a city, and so on --
88
205000
2000
03:42
is that just bullshit, metaphorical bullshit,
89
207000
3000
03:45
or is there serious substance to it?
90
210000
3000
03:48
And if that is the case,
91
213000
2000
03:50
how come that it's very hard to kill a city?
92
215000
2000
03:52
You could drop an atom bomb on a city,
93
217000
2000
03:54
and 30 years later it's surviving.
94
219000
2000
03:56
Very few cities fail.
95
221000
3000
03:59
All companies die, all companies.
96
224000
3000
04:02
And if you have a serious theory, you should be able to predict
97
227000
2000
04:04
when Google is going to go bust.
98
229000
3000
04:07
So is that just another version
99
232000
3000
04:10
of this?
100
235000
2000
04:12
Well we understand this very well.
101
237000
2000
04:14
That is, you ask any generic question about this --
102
239000
2000
04:16
how many trees of a given size,
103
241000
2000
04:18
how many branches of a given size does a tree have,
104
243000
2000
04:20
how many leaves,
105
245000
2000
04:22
what is the energy flowing through each branch,
106
247000
2000
04:24
what is the size of the canopy,
107
249000
2000
04:26
what is its growth, what is its mortality?
108
251000
2000
04:28
We have a mathematical framework
109
253000
2000
04:30
based on generic universal principles
110
255000
3000
04:33
that can answer those questions.
111
258000
2000
04:35
And the idea is can we do the same for this?
112
260000
4000
04:40
So the route in is recognizing
113
265000
3000
04:43
one of the most extraordinary things about life,
114
268000
2000
04:45
is that it is scalable,
115
270000
2000
04:47
it works over an extraordinary range.
116
272000
2000
04:49
This is just a tiny range actually:
117
274000
2000
04:51
It's us mammals;
118
276000
2000
04:53
we're one of these.
119
278000
2000
04:55
The same principles, the same dynamics,
120
280000
2000
04:57
the same organization is at work
121
282000
2000
04:59
in all of these, including us,
122
284000
2000
05:01
and it can scale over a range of 100 million in size.
123
286000
3000
05:04
And that is one of the main reasons
124
289000
3000
05:07
life is so resilient and robust --
125
292000
2000
05:09
scalability.
126
294000
2000
05:11
We're going to discuss that in a moment more.
127
296000
3000
05:14
But you know, at a local level,
128
299000
2000
05:16
you scale; everybody in this room is scaled.
129
301000
2000
05:18
That's called growth.
130
303000
2000
05:20
Here's how you grew.
131
305000
2000
05:22
Rat, that's a rat -- could have been you.
132
307000
2000
05:24
We're all pretty much the same.
133
309000
3000
05:27
And you see, you're very familiar with this.
134
312000
2000
05:29
You grow very quickly and then you stop.
135
314000
2000
05:31
And that line there
136
316000
2000
05:33
is a prediction from the same theory,
137
318000
2000
05:35
based on the same principles,
138
320000
2000
05:37
that describes that forest.
139
322000
2000
05:39
And here it is for the growth of a rat,
140
324000
2000
05:41
and those points on there are data points.
141
326000
2000
05:43
This is just the weight versus the age.
142
328000
2000
05:45
And you see, it stops growing.
143
330000
2000
05:47
Very, very good for biology --
144
332000
2000
05:49
also one of the reasons for its great resilience.
145
334000
2000
05:51
Very, very bad
146
336000
2000
05:53
for economies and companies and cities
147
338000
2000
05:55
in our present paradigm.
148
340000
2000
05:57
This is what we believe.
149
342000
2000
05:59
This is what our whole economy
150
344000
2000
06:01
is thrusting upon us,
151
346000
2000
06:03
particularly illustrated in that left-hand corner:
152
348000
3000
06:06
hockey sticks.
153
351000
2000
06:08
This is a bunch of software companies --
154
353000
2000
06:10
and what it is is their revenue versus their age --
155
355000
2000
06:12
all zooming away,
156
357000
2000
06:14
and everybody making millions and billions of dollars.
157
359000
2000
06:16
Okay, so how do we understand this?
158
361000
3000
06:19
So let's first talk about biology.
159
364000
3000
06:22
This is explicitly showing you
160
367000
2000
06:24
how things scale,
161
369000
2000
06:26
and this is a truly remarkable graph.
162
371000
2000
06:28
What is plotted here is metabolic rate --
163
373000
3000
06:31
how much energy you need per day to stay alive --
164
376000
3000
06:34
versus your weight, your mass,
165
379000
2000
06:36
for all of us bunch of organisms.
166
381000
3000
06:39
And it's plotted in this funny way by going up by factors of 10,
167
384000
3000
06:42
otherwise you couldn't get everything on the graph.
168
387000
2000
06:44
And what you see if you plot it
169
389000
2000
06:46
in this slightly curious way
170
391000
2000
06:48
is that everybody lies on the same line.
171
393000
3000
06:51
Despite the fact that this is the most complex and diverse system
172
396000
3000
06:54
in the universe,
173
399000
3000
06:57
there's an extraordinary simplicity
174
402000
2000
06:59
being expressed by this.
175
404000
2000
07:01
It's particularly astonishing
176
406000
3000
07:04
because each one of these organisms,
177
409000
2000
07:06
each subsystem, each cell type, each gene,
178
411000
2000
07:08
has evolved in its own unique environmental niche
179
413000
4000
07:12
with its own unique history.
180
417000
3000
07:15
And yet, despite all of that Darwinian evolution
181
420000
3000
07:18
and natural selection,
182
423000
2000
07:20
they've been constrained to lie on a line.
183
425000
2000
07:22
Something else is going on.
184
427000
2000
07:24
Before I talk about that,
185
429000
2000
07:26
I've written down at the bottom there
186
431000
2000
07:28
the slope of this curve, this straight line.
187
433000
2000
07:30
It's three-quarters, roughly,
188
435000
2000
07:32
which is less than one -- and we call that sublinear.
189
437000
3000
07:35
And here's the point of that.
190
440000
2000
07:37
It says that, if it were linear,
191
442000
3000
07:40
the steepest slope,
192
445000
2000
07:42
then doubling the size
193
447000
2000
07:44
you would require double the amount of energy.
194
449000
2000
07:46
But it's sublinear, and what that translates into
195
451000
3000
07:49
is that, if you double the size of the organism,
196
454000
2000
07:51
you actually only need 75 percent more energy.
197
456000
3000
07:54
So a wonderful thing about all of biology
198
459000
2000
07:56
is that it expresses an extraordinary economy of scale.
199
461000
3000
07:59
The bigger you are systematically,
200
464000
2000
08:01
according to very well-defined rules,
201
466000
2000
08:03
less energy per capita.
202
468000
3000
08:06
Now any physiological variable you can think of,
203
471000
3000
08:09
any life history event you can think of,
204
474000
2000
08:11
if you plot it this way, looks like this.
205
476000
3000
08:14
There is an extraordinary regularity.
206
479000
2000
08:16
So you tell me the size of a mammal,
207
481000
2000
08:18
I can tell you at the 90 percent level everything about it
208
483000
3000
08:21
in terms of its physiology, life history, etc.
209
486000
4000
08:25
And the reason for this is because of networks.
210
490000
3000
08:28
All of life is controlled by networks --
211
493000
3000
08:31
from the intracellular through the multicellular
212
496000
2000
08:33
through the ecosystem level.
213
498000
2000
08:35
And you're very familiar with these networks.
214
500000
3000
08:39
That's a little thing that lives inside an elephant.
215
504000
3000
08:42
And here's the summary of what I'm saying.
216
507000
3000
08:45
If you take those networks,
217
510000
2000
08:47
this idea of networks,
218
512000
2000
08:49
and you apply universal principles,
219
514000
2000
08:51
mathematizable, universal principles,
220
516000
2000
08:53
all of these scalings
221
518000
2000
08:55
and all of these constraints follow,
222
520000
3000
08:58
including the description of the forest,
223
523000
2000
09:00
the description of your circulatory system,
224
525000
2000
09:02
the description within cells.
225
527000
2000
09:04
One of the things I did not stress in that introduction
226
529000
3000
09:07
was that, systematically, the pace of life
227
532000
3000
09:10
decreases as you get bigger.
228
535000
2000
09:12
Heart rates are slower; you live longer;
229
537000
3000
09:15
diffusion of oxygen and resources
230
540000
2000
09:17
across membranes is slower, etc.
231
542000
2000
09:19
The question is: Is any of this true
232
544000
2000
09:21
for cities and companies?
233
546000
3000
09:24
So is London a scaled up Birmingham,
234
549000
3000
09:27
which is a scaled up Brighton, etc., etc.?
235
552000
3000
09:30
Is New York a scaled up San Francisco,
236
555000
2000
09:32
which is a scaled up Santa Fe?
237
557000
2000
09:34
Don't know. We will discuss that.
238
559000
2000
09:36
But they are networks,
239
561000
2000
09:38
and the most important network of cities
240
563000
2000
09:40
is you.
241
565000
2000
09:42
Cities are just a physical manifestation
242
567000
3000
09:45
of your interactions,
243
570000
2000
09:47
our interactions,
244
572000
2000
09:49
and the clustering and grouping of individuals.
245
574000
2000
09:51
Here's just a symbolic picture of that.
246
576000
3000
09:54
And here's scaling of cities.
247
579000
2000
09:56
This shows that in this very simple example,
248
581000
3000
09:59
which happens to be a mundane example
249
584000
2000
10:01
of number of petrol stations
250
586000
2000
10:03
as a function of size --
251
588000
2000
10:05
plotted in the same way as the biology --
252
590000
2000
10:07
you see exactly the same kind of thing.
253
592000
2000
10:09
There is a scaling.
254
594000
2000
10:11
That is that the number of petrol stations in the city
255
596000
4000
10:15
is now given to you
256
600000
2000
10:17
when you tell me its size.
257
602000
2000
10:19
The slope of that is less than linear.
258
604000
3000
10:22
There is an economy of scale.
259
607000
2000
10:24
Less petrol stations per capita the bigger you are -- not surprising.
260
609000
3000
10:27
But here's what's surprising.
261
612000
2000
10:29
It scales in the same way everywhere.
262
614000
2000
10:31
This is just European countries,
263
616000
2000
10:33
but you do it in Japan or China or Colombia,
264
618000
3000
10:36
always the same
265
621000
2000
10:38
with the same kind of economy of scale
266
623000
2000
10:40
to the same degree.
267
625000
2000
10:42
And any infrastructure you look at --
268
627000
3000
10:45
whether it's the length of roads, length of electrical lines --
269
630000
3000
10:48
anything you look at
270
633000
2000
10:50
has the same economy of scale scaling in the same way.
271
635000
3000
10:53
It's an integrated system
272
638000
2000
10:55
that has evolved despite all the planning and so on.
273
640000
3000
10:58
But even more surprising
274
643000
2000
11:00
is if you look at socio-economic quantities,
275
645000
2000
11:02
quantities that have no analog in biology,
276
647000
3000
11:05
that have evolved when we started forming communities
277
650000
3000
11:08
eight to 10,000 years ago.
278
653000
2000
11:10
The top one is wages as a function of size
279
655000
2000
11:12
plotted in the same way.
280
657000
2000
11:14
And the bottom one is you lot --
281
659000
2000
11:16
super-creatives plotted in the same way.
282
661000
3000
11:19
And what you see
283
664000
2000
11:21
is a scaling phenomenon.
284
666000
2000
11:23
But most important in this,
285
668000
2000
11:25
the exponent, the analog to that three-quarters
286
670000
2000
11:27
for the metabolic rate,
287
672000
2000
11:29
is bigger than one -- it's about 1.15 to 1.2.
288
674000
2000
11:31
Here it is,
289
676000
2000
11:33
which says that the bigger you are
290
678000
3000
11:36
the more you have per capita, unlike biology --
291
681000
3000
11:39
higher wages, more super-creative people per capita as you get bigger,
292
684000
4000
11:43
more patents per capita, more crime per capita.
293
688000
3000
11:46
And we've looked at everything:
294
691000
2000
11:48
more AIDS cases, flu, etc.
295
693000
3000
11:51
And here, they're all plotted together.
296
696000
2000
11:53
Just to show you what we plotted,
297
698000
2000
11:55
here is income, GDP --
298
700000
3000
11:58
GDP of the city --
299
703000
2000
12:00
crime and patents all on one graph.
300
705000
2000
12:02
And you can see, they all follow the same line.
301
707000
2000
12:04
And here's the statement.
302
709000
2000
12:06
If you double the size of a city from 100,000 to 200,000,
303
711000
3000
12:09
from a million to two million, 10 to 20 million,
304
714000
2000
12:11
it doesn't matter,
305
716000
2000
12:13
then systematically
306
718000
2000
12:15
you get a 15 percent increase
307
720000
2000
12:17
in wages, wealth, number of AIDS cases,
308
722000
2000
12:19
number of police,
309
724000
2000
12:21
anything you can think of.
310
726000
2000
12:23
It goes up by 15 percent,
311
728000
2000
12:25
and you have a 15 percent savings
312
730000
3000
12:28
on the infrastructure.
313
733000
3000
12:31
This, no doubt, is the reason
314
736000
3000
12:34
why a million people a week are gathering in cities.
315
739000
3000
12:37
Because they think that all those wonderful things --
316
742000
3000
12:40
like creative people, wealth, income --
317
745000
2000
12:42
is what attracts them,
318
747000
2000
12:44
forgetting about the ugly and the bad.
319
749000
2000
12:46
What is the reason for this?
320
751000
2000
12:48
Well I don't have time to tell you about all the mathematics,
321
753000
3000
12:51
but underlying this is the social networks,
322
756000
3000
12:54
because this is a universal phenomenon.
323
759000
3000
12:57
This 15 percent rule
324
762000
3000
13:00
is true
325
765000
2000
13:02
no matter where you are on the planet --
326
767000
2000
13:04
Japan, Chile,
327
769000
2000
13:06
Portugal, Scotland, doesn't matter.
328
771000
3000
13:09
Always, all the data shows it's the same,
329
774000
3000
13:12
despite the fact that these cities have evolved independently.
330
777000
3000
13:15
Something universal is going on.
331
780000
2000
13:17
The universality, to repeat, is us --
332
782000
3000
13:20
that we are the city.
333
785000
2000
13:22
And it is our interactions and the clustering of those interactions.
334
787000
3000
13:25
So there it is, I've said it again.
335
790000
2000
13:27
So if it is those networks and their mathematical structure,
336
792000
3000
13:30
unlike biology, which had sublinear scaling,
337
795000
3000
13:33
economies of scale,
338
798000
2000
13:35
you had the slowing of the pace of life
339
800000
2000
13:37
as you get bigger.
340
802000
2000
13:39
If it's social networks with super-linear scaling --
341
804000
2000
13:41
more per capita --
342
806000
2000
13:43
then the theory says
343
808000
2000
13:45
that you increase the pace of life.
344
810000
2000
13:47
The bigger you are, life gets faster.
345
812000
2000
13:49
On the left is the heart rate showing biology.
346
814000
2000
13:51
On the right is the speed of walking
347
816000
2000
13:53
in a bunch of European cities,
348
818000
2000
13:55
showing that increase.
349
820000
2000
13:57
Lastly, I want to talk about growth.
350
822000
3000
14:00
This is what we had in biology, just to repeat.
351
825000
3000
14:03
Economies of scale
352
828000
3000
14:06
gave rise to this sigmoidal behavior.
353
831000
3000
14:09
You grow fast and then stop --
354
834000
3000
14:12
part of our resilience.
355
837000
2000
14:14
That would be bad for economies and cities.
356
839000
3000
14:17
And indeed, one of the wonderful things about the theory
357
842000
2000
14:19
is that if you have super-linear scaling
358
844000
3000
14:22
from wealth creation and innovation,
359
847000
2000
14:24
then indeed you get, from the same theory,
360
849000
3000
14:27
a beautiful rising exponential curve -- lovely.
361
852000
2000
14:29
And in fact, if you compare it to data,
362
854000
2000
14:31
it fits very well
363
856000
2000
14:33
with the development of cities and economies.
364
858000
2000
14:35
But it has a terrible catch,
365
860000
2000
14:37
and the catch
366
862000
2000
14:39
is that this system is destined to collapse.
367
864000
3000
14:42
And it's destined to collapse for many reasons --
368
867000
2000
14:44
kind of Malthusian reasons -- that you run out of resources.
369
869000
3000
14:47
And how do you avoid that? Well we've done it before.
370
872000
3000
14:50
What we do is,
371
875000
2000
14:52
as we grow and we approach the collapse,
372
877000
3000
14:55
a major innovation takes place
373
880000
3000
14:58
and we start over again,
374
883000
2000
15:00
and we start over again as we approach the next one, and so on.
375
885000
3000
15:03
So there's this continuous cycle of innovation
376
888000
2000
15:05
that is necessary
377
890000
2000
15:07
in order to sustain growth and avoid collapse.
378
892000
3000
15:10
The catch, however, to this
379
895000
2000
15:12
is that you have to innovate
380
897000
2000
15:14
faster and faster and faster.
381
899000
3000
15:17
So the image
382
902000
2000
15:19
is that we're not only on a treadmill that's going faster,
383
904000
3000
15:22
but we have to change the treadmill faster and faster.
384
907000
3000
15:25
We have to accelerate on a continuous basis.
385
910000
3000
15:28
And the question is: Can we, as socio-economic beings,
386
913000
3000
15:31
avoid a heart attack?
387
916000
3000
15:34
So lastly, I'm going to finish up in this last minute or two
388
919000
3000
15:37
asking about companies.
389
922000
2000
15:39
See companies, they scale.
390
924000
2000
15:41
The top one, in fact, is Walmart on the right.
391
926000
2000
15:43
It's the same plot.
392
928000
2000
15:45
This happens to be income and assets
393
930000
2000
15:47
versus the size of the company as denoted by its number of employees.
394
932000
2000
15:49
We could use sales, anything you like.
395
934000
3000
15:52
There it is: after some little fluctuations at the beginning,
396
937000
3000
15:55
when companies are innovating,
397
940000
2000
15:57
they scale beautifully.
398
942000
2000
15:59
And we've looked at 23,000 companies
399
944000
3000
16:02
in the United States, may I say.
400
947000
2000
16:04
And I'm only showing you a little bit of this.
401
949000
3000
16:07
What is astonishing about companies
402
952000
2000
16:09
is that they scale sublinearly
403
954000
3000
16:12
like biology,
404
957000
2000
16:14
indicating that they're dominated,
405
959000
2000
16:16
not by super-linear
406
961000
2000
16:18
innovation and ideas;
407
963000
3000
16:21
they become dominated
408
966000
2000
16:23
by economies of scale.
409
968000
2000
16:25
In that interpretation,
410
970000
2000
16:27
by bureaucracy and administration,
411
972000
2000
16:29
and they do it beautifully, may I say.
412
974000
2000
16:31
So if you tell me the size of some company, some small company,
413
976000
3000
16:34
I could have predicted the size of Walmart.
414
979000
3000
16:37
If it has this sublinear scaling,
415
982000
2000
16:39
the theory says
416
984000
2000
16:41
we should have sigmoidal growth.
417
986000
3000
16:44
There's Walmart. Doesn't look very sigmoidal.
418
989000
2000
16:46
That's what we like, hockey sticks.
419
991000
3000
16:49
But you notice, I've cheated,
420
994000
2000
16:51
because I've only gone up to '94.
421
996000
2000
16:53
Let's go up to 2008.
422
998000
2000
16:55
That red line is from the theory.
423
1000000
3000
16:58
So if I'd have done this in 1994,
424
1003000
2000
17:00
I could have predicted what Walmart would be now.
425
1005000
3000
17:03
And then this is repeated
426
1008000
2000
17:05
across the entire spectrum of companies.
427
1010000
2000
17:07
There they are. That's 23,000 companies.
428
1012000
3000
17:10
They all start looking like hockey sticks,
429
1015000
2000
17:12
they all bend over,
430
1017000
2000
17:14
and they all die like you and me.
431
1019000
2000
17:16
Thank you.
432
1021000
2000
17:18
(Applause)
433
1023000
9000

▲Back to top

ABOUT THE SPEAKER
Geoffrey West - Theorist
Physicist Geoffrey West believes that complex systems from organisms to cities are in many ways governed by simple laws -- laws that can be discovered and analyzed.

Why you should listen

Trained as a theoretical physicist, Geoffrey West has turned his analytical mind toward the inner workings of more concrete things, like ... animals. In a paper for Science in 1997, he and his team uncovered what he sees as a surprisingly universal law of biology — the way in which heart rate, size and energy consumption are related, consistently, across most living animals. (Though not all animals: “There are always going to be people who say, ‘What about the crayfish?’ " he says. “Well, what about it? Every fundamental law has exceptions. But you still need the law or else all you have is observations that don’t make sense.")

A past president of the multidisciplinary Santa Fe Institute (after decades working  in high-energy physics at Los Alamos and Stanford), West now studies the behavior and development of cities. In his newest work, he proposes that one simple number, population, can predict a stunning array of details about any city, from crime rate to economic activity. It's all about the plumbing, he says, the infrastructure that powers growth or dysfunction. His next target for study: corporations.

He says: "Focusing on the differences [between cities] misses the point. Sure, there are differences, but different from what? We’ve found the what."

More profile about the speaker
Geoffrey West | Speaker | TED.com