What Do Organisms, Crowded Cities, and Corporations Have in Common?

Detail from cover of Scale

What do you, your town, and your employer all have in common? Scalability. According to physicist Geoffrey West, there are mathematical principles that govern the growth and longevity of complex organisms, crowded cities, and even corporations. West’s new book Scale: The Universal Laws of Growth, Innovation, Sustainability, and the Pace of Life in Organisms, Cities, Economies, and Companies, introduces readers to this hidden, fascinating world.

In the following interview, West explains the difference between complicated and complex, what the rise of Donald Trump suggests about the state of the world, and why the company you work for could be living on borrowed time.

SIGNATURE: You’re known as “the dean of complexity theory”. A lot of our readers may not understand what complexity means and why you study it.

GEOFFREY WEST: Science has progressed, at least beginning with the physical sciences, by always being what’s called reductionistic: that is, reducing things to their primary elements, whether they’re electrons, atoms, molecules, or genes and so forth. That has been enormously successful, but one of the things that we’ve begun to appreciate more and more — especially in the last 50 years, certainly the last 20 — is that kind of paradigm has extraordinary limitations.

When you try to build up from these fundamental elements to the collective whole, you discover that the whole is much greater than, behaves differently than, and is structured differently from the sum of its parts. What you recognize in parallel with that is almost all of the major issues that we face on the planet in a tsunami of challenges and crises — everything from climate change and the question of stability in markets to potential questions about risk and how we deal with things like cancer, and the encroaching threat of global urbanization — are what we call complex. They’re not easily, or even potentially, reduced to the sum of their parts.

For example, organisms like you and I are much more than the sums of our cells or genes. A city is much more than the sum of all its people or roads and businesses. Furthermore, all of these are not just conglomerations of all of these things: They’re all highly dependent upon one another. There’s what we call express emergent properties: New things emerge as you build these systems up, whether they are economies, climates, cities, or our bodies.

With the näiveté that came out of a physics paradigm adopted by the other sciences, the idea became to reduce everything to its fundamental elements, and then we’ll build up to understand the whole. What has emerged through the failure of this paradigm to deal with complex systems is the realization that we have to think more systemically and holistically. I’ve just been rambling on. Does that help?

SIG: Yes, it’s better than just saying, “Things are complicated”.

GW: Well, now listen, that’s a very important statement. It isn’t that things are complicated, because complicated is to be distinguished from complex. Let me give you a couple of examples.

One of the great successes in the development of science was understanding the motion of the planets. Newton’s laws of gravity and motion and so on set the template of everything we do. Newton’s laws explained all of that, which was fantastic. This reductionistic and simplistic — that is to say it’s not complex — explanation was very profound. Your cellphone wouldn’t work if we didn’t understand all of that with great accuracy, since we use satellites to send our messages. To do that in detail, to understand it in the detail that it is needed to make technology like this work, we need to put in all of the various little corrections that occur to Newton’s laws due to things like the atmosphere and things like satellites that may have deviated slightly because of an asteroid and so forth, and all of that gets complicated. That’s complicated.

The fundamentals are deeply understood. We know how to do these things to extraordinary accuracy even though they’re complicated, but the system is simple because we have the equations and understand them. Those equations are expressed in a very parsimonious, mathematical form. That extends the paradigm. That distinguishes it from complex. If you think of your body, it’s impossible to imagine that there’s a parsimonious set of equations that will describe in detail the way your body works. This is sort of tongue in cheek, but there probably aren’t even infinite equations that would do that. Or, let’s put it this way: You would probably require an infinite number of equations and infinite computing power for the degree of accuracy that we’ve developed for understanding mechanical motions such as planets going around the sun.

The other example would be the building of an airplane. We understand in detail and in great depth the physics and material sciences of flight — the engineering and math involved. In that sense, an airplane is a simple system: We can describe it with a relatively small set of equations or algorithms. It’s extremely complicated, potentially, if you try to build a Boeing 787. It is very complicated, but nothing in principle will stop you from doing it. I think there are two great big manuals at Boeing that tells you how to build the 787, the Dreamliner. It’s how they carry it out. You cannot imagine manuals for knowing how your body, or New York City, or the stock market works.

When we go to these complex systems — when you think along the lines I’ve elaborated on — you conclude that to get a quantitative, predictive understanding of them is, in principle, impossible if you insist on getting into great detail. This is where I come in.

There is, potentially, an in-between place where we may not have a theory for the way the system works in infinite detail, but my work, and the work of my colleagues, has shown that you generally can get what physics calls a “coarse grain description”: an understanding of the idealized, average behavior of these systems, and maybe even more than that — how they deviate from various things and so forth. That’s what the scaling laws represent. They show that if you tell me the size of a mammal — its mass, how much it weighs. I can tell you to 80 or 90 percent accuracy how much food it needs to eat a day, how fast its heart beats, how long it is going to live, the length and radius of the fifth branch of its circulatory system, the flow rate of blood of a typical capillary, the structure of its respiratory system, how long it needs to sleep and so on.

You can answer all of these kinds of systems for the average idealized mammal of a given size, and it will be correct to an 85 or 90 percent level. I can predict the metabolism of an elephant, for example, but give me a specific element and I won’t be able to tell you exactly what the metabolic rate is. By metabolic rate, I mean that in colloquial terms: How much it is going to need to eat each day. To bring it closer to home, one can roughly predict the lifespan of a mammal of a given size, and in particular where this life span of a hundred years come from. I can tell you what the parameters are that control that, but I won’t be able to tell you detail how long you’re going to live.

SIG: You’ve taken some of things you’ve applied to animals and cities, and now you’re applying them to companies. How can you look at a company and infer some of these same things?

GW: It is very important to recognize that this work is inspired by data. The first thing is to look at the data. If you think of cities, the idea that New York is a scaled-up Los Angeles, which is a scaled-up Chicago, which is a scaled-up Santa Fe, where I live.They look completely different: different geographies, cultures, et cetera, but all cities have things in common: roads, businesses, people living in them, all of these obvious things.

You might wonder if there’s a similar dynamic at work. The only way to answer that, a priori, is to look at the data. You look at these various metrics that have been measured about cities, from the lengths of its roads and the number of patents they produce and so on, and you ask how that scales with city size. As what was seen in biology, just as an elephant doesn’t look much like a whale, they are scaled-up versions of one another, in this coarse-grained sense. So is New York and Los Angeles. They follow scaling laws.

There are two things I want to add to that. The first is that this is from the data. As a physicist, one tries to understand where all of this comes from. A theory was developed, originally in a biological context that was then adapted to a city context, to understand where this scaling came from. Not just the scaling in general, the fact that they scale, but quantitatively, how it scales. For example, in animals, every time you double the size, you save about 25 percent in its metabolism. That quarter dominates all of life. It was developing a theory, an understanding, of not just the scaling, but why a quarter? Why not a tenth, or a third?

The theory was based on seeing all of these kinds of systems, and in that case, organisms, as networks. We’re an agglomeration of many different networks transporting energy, and information, and resources, and so forth: everything from your circulatory system to your neural system. Going to companies, the same paradigm was followed. That is, we had to get the data — the number of employees they have, their sales, assets, profits — and see if there was any regularity in the way they scale. You look at say, sales and profits, and you plot that versus the size of the company using the number of employees. We found that there was scaling.

There was a much greater spread in the data among companies than there was among cities or organisms, but that’s predictable. There’s more spread because many companies haven’t been around for long: They’re not very old. One of the things we discovered was that when we got hold of these big data sets was that the expected lifespan of an average publicly traded company in the United States was about 10 years. When you’ve mastered gestation and succeeded in going public, you can only expect to last 10 years before you disappear again. That is very long for these companies to equilibrate and start to evolve, so it’s not surprising to see a lot of spread among the data.

SIG: Based on your observations, what is a common red flag for a company in trouble?

GW: That’s very hard to answer, and that work is very much in progress. The usual life cycle of a company is that it grows rapidly when it starts, like us. The other thing that reflects is a greater diversity in its product space, typically: lots of ideas, fluidity and excitement, a relatively small bureaucracy, and so on. As that company grows and becomes more established, market forces typically narrow its product space. It goes from being multidimensional to unidimensional in terms of the number of products. Ones that sell well get reinforced, and the ones that don’t — even if they are great ideas — have to be put on a shelf and forgotten. As the company grows, this space narrows and they become less diverse. At the same time, because they’re growing, they need to increase their administration and bureaucracy because they have to be sure they’re abiding by the tax codes and their offices are being cleaned and all of the rest of the usual stuff. Because of the addition of the bureaucracy, more and more rules and laws govern that, which tends to dampen the innovation of the company.

You see this phenomenon that the company slows in growth and becomes more vulnerable to changes in the market. Companies often go on for a long time in this condition, but often, any significant change in the market typically leads to their demise. They either get bought up, or they liquidate, or they make a decision that it’s going to be hard to compete and they sell out and become acquired by another company. I think the signal, and this is what we’re still working on to quantify it, is something to do with the metrics of diversity and innovation.

The other killing negative feedback loop is that as this phenomenon of slowing growth and becoming more vulnerable develops, when a mini crisis comes one of the first things a company does is cut back on their research and development, or its equivalent. The innovative part gets pushed back on the back burner, with the idea that it isn’t needed now and the important thing is to continue to produce products; that they’ll return to that in a couple of years after they’ve recovered. That typically does not happen. Typically, the company disappears.

That’s a very typical life history of many companies. What we’ve been thinking about is trying to develop metrics relative to performances, where one can measure and have a metric, or signal, that quantitatively would signal this impending doom, so to speak. It is kind of astonishing how hard it is for companies to recognize that they’re in trouble. When they’re very large, it’s the proverbial problem that you can’t turn around the battleship.

SIG: Then they can become “too big to fail”?

GW: Yes, they’re too big, and then we run into this situation of “too big to fail” in certain financial institutions. By the way, before I was working on this, I was very enthusiastic about “too big to fail” in 2008. Once I worked on this, I wasn’t so sure. It is good that companies fail, by the way. You don’t need those dinosaurs hanging around. We don’t care about the companies, we care about the people, everyone from the CEO. That’s who we care about. We care about producing new ideas, creating wealth, and innovating. When you have an institution oppressing that, either consciously or unconsciously — and I think a lot of this is unconscious, actually — then get rid of it.

SIG: There was an interview you did in which you spoke about mavericks and company size. I’d love to hear more.

GW: One of the driving questions that go the into thinking about the extension of biology to cities and companies was why is it that almost all companies fail but cities go on, regardless? Most social institutions do. What is that dynamic? You can drop bombs on a city and, 25 years later, they’re fine — typically speaking. But all you need is a fluctuation in the stock market and you lose a TWA or Montgomery Ward. We lost General Motors, effectively. They were artificially resurrected.

It doesn’t take a hell of a lot to kill a company, but it takes something truly extraordinary to kill a modern city. Why is that? I described a little bit of the company dynamic. The city dynamic is in marked contrast to that. The great thing about cities is that they’re really not top down. There’s an administration, obviously, but they’re facilitators. They don’t really control the city. Quite the contrary: They’ve got this marvelous, open ended — I used the term free market. It doesn’t matter what it is.

In a great city, anything goes, at least in principle. New York, London, and other great cities have this characteristic that they somehow encourage maverick, slightly deviant behavior that in some cases lead to criminality but in many cases is exactly the source of its great creativity, success, and attractiveness. A city, as it grows, goes from being a small town like Santa Fe, for example, that doesn’t really have a breadth of industry, to large, like New York City, which has an extraordinary breadth of different kinds of employment and jobs. As cities grow, that opens up more and more. Cities become multi-dimensional, in contrast to companies, which become more unidimensional.

One of the fantastic things about a city is that it tolerates weirdness: people pushing the boundaries of the arts and culture, science, business. A great city’s administration facilitates that in one way or another. This is in complete contrast to companies. They want uniformity and efficiency. There’s this culture that almost inevitably develops of suppressing crazy people wandering down the hallways of a major company. Not even Google, which likes to project that image.

One of my private peeves about Google is that it likes to project this image that everyone’s a little bit crazy coming out of the basement at MIT, but in fact, it’s surprisingly uniform. In that sense, it’s not very diverse. Google has done very well, but nevertheless, as it grows, I think that’s going to be one of its biggest battles and challenges. Microsoft went that way, and there are even rumbles that Microsoft will not survive.

At any stage in a major company’s lifetime it seems impossible. In my own lifetime, who would have believed that there wouldn’t be a TWA, or even a Montgomery Ward, or Sears, which is going down the tubes.

SIG: You said that as a business gets in trouble, it cuts its research and development, and successful companies become less tolerant of eccentricity and maverick thinking. Can something be drawn from these two things regarding innovation, conservative business practices, and whether there’s a balance between these two things that ought to be addressed?

GW: That’s a very difficult question. Many of the things that I’m saying, I’m sure, have been recognized by other people. What I’m trying to do is bring science to it so it can be quantified and understood; to see it as part of the natural process. That’s a huge challenge. The point is that all of these things have similar dynamics underlying them that have to do with networking and the structure of networks. It’s in our DNA, if you like. It’s very hard to overcome that.

You have to, as you said, find a balance between having an efficient company that produces a first-rate product, and being sufficiently flexible that you encourage new ideas. That turns out to be a huge challenge, especially because there is a continual build-up of bureaucracy, administration, rules, and laws. Some of my primary areas of research is trying to quantify the build-up of rules and laws in societies, both on a federal, state, and local level, as well as universities. I’m probably much older than you. My God, administrations of university labs used to be tiny 50 years ago, relatively speaking. They have truly grown exponentially, and now all of us feel this tremendous burden of constraints.

That’s a necessary component of running a large system, and balancing it with flexibility is crucial, which takes me back to this idea of coarse grain and the scaling laws. They tell you what a company, city, or organism should be given its size — the ones that are measurable, that is. As I said earlier, there are deviations and fluctuations from that, and one of the questions is how big those fluctuations can be. The brilliance of a highly successful, long-lasting company is to take advantage of how much flexibility there is to be able to deviate from these scaling laws. That’s one way of looking at this.

Another challenge is the question of global sustainability. Buying into this idea of the planet as a whole, and resetting the clock and reinventing ourselves, whether as a whole society, a business, a city, or even an individual. We are continually getting into these loops, and the challenge is to recognize that when it is happening. The sad thing that’s happening globally is that, while there is some recognition of this truth in terms of our political leadership worldwide, there’s no great voice for resetting the clock, other than, amazingly, Donald Trump — God help us. I’m thinking of writing an article that Donald Trump has come too early, and is the wrong person!

SIG: What is it that you see Donald Trump saying that needed to be said?

GW: The fact is that he, somehow, has picked up in the zeitgeist — or the zeitgeist has picked up in him — this idea that we’re heading for something that is not good. There’s great fear. It is interesting that it has been picked up, because if you look at it in absolute terms, the United States is doing extraordinarily well. There’s a very high quality of life. This is just a guess, but I’d say 80 percent of the people who voted for Trump, their standard of living is extraordinary compared to their parents and grandparents, yet they have somehow felt that something bad is happening.

Most of it is to do with the fact that workers are no longer guaranteed a 5 to 10 percent raise every year, and that hasn’t happened for a long time. I suspect that’s one of the major drivers, because I believe greed is a driving dynamic of what’s going on. Trump has somehow been thrust into this position, but he has it all wrong, of course. Not to recognize global warming and not to understand even anything about they dynamics of globalization, to try to cut back seriously on science — these are all cynical, negative responses. We need truly bold, visionary leadership in the opposite direction of these thing. His personality type — and I can’t stand him — it’s that kind of presence: you need a Roosevelt, or a Winston Churchill, that can galvanize and inspire people. Some of us thought that when Obama was elected he would be that.

SIG: Looking at what we’ve learned about the resilience of modern cities, can we apply any of it to ourselves and the world at large? Can we all apply these lessons?

GW: I certainly hope so, but we need to somehow bring together different ways of thinking about this, and to understand the interrelation of all of these things that are seen as completely independent — or dealt with independently — and to see the interconnectedness and integration. We need to start to think in some of these terms. No way am I advocating that one shouldn’t maintain the present way of thinking about these things, which is highly stove-piped, but we need to develop a bigger umbrella way of thinking as a competent and supplement to all of that and apply it to multiple levels — maybe the global level now because we’re reaching a state of emergency. It may be too late.

We should apply it to our financial markets, corporate structures, and individual level: to think of one’s own life, psychology, and social interactions in these terms. I actually had a section in the book, I took it out, which was along those lines. If producing ideas and understanding is what this enormous enterprise is all about, creating a good life and a meaningful one for seven-plus billion people on this planet, then we’ve got to come to terms with and understand these dynamics, and start to exercise a much bigger picture way of looking at it. The way that we structure our thinking is limited in its horizons, both in space and particularly in time. There needs to be much more long-term thinking, both structurally and intellectually.