The economist who came in from the cold

Seeing the world as a mass of sputtering, erratic and ever-evolving systems can help one to survive even the most volatile of times, says complexity science pioneer W. Brian Arthur

Dr Arthur is a pioneer of complexity science, which looks at how interactions between elements impact the world as opposed to the conventional cause-and-effect approach. PHOTO: SAM CHIN FOR THE STRAITS TIMES

By CHEONG SUK WAI
SENIOR WRITER

YOU can often predict how big an idea is going to be, by how fiercely people resist it at first.

So it was with complexity science pioneer W. Brian Arthur's once-novel "theory of complexity" back in 1979.

Dr Arthur, 66, was at the Nanyang Technological University (NTU) last month to advise on its year-old complexity programme.

Complexity science is essentially a fresh way of thinking about, and coping with, life's most perplexing problems, such as tackling climate change, watching for spikes in financial markets, or strengthening one's immune system against ever-mutating germs.

It sees all the elements in any given system, be it a market, government or company, as individual actors and studies how all these actors might interact and affect the system as a whole.

Take everyday road traffic, said Dr Arthur. Traffic might be smooth-flowing at a given moment, but if just one driver slowed down suddenly, a jam would likely ensue. "Yet, if a Martian were to look at just that one individual, there'd be no way the Martian could see a traffic jam emerging," he said.

For too long, Dr Arthur noted, scientists had been like that Martian zeroing in on that one driver; they have focused studiously on only their subject "like they would a Swiss watch", taking it apart to its smallest elements.

As for most economists, he noted, they had always regarded the economy as a machine that hummed along in equilibrium, with all its parts connected properly and all the people in it behaving rationally.

But to him, even back in 1979, a system was neither a machine nor in a perfect state, because he believed that systems were organic and evolving all the time - adding new layers on top of old ones, with the new changing the functions of the old.

That insight surfaced when he was at the University of California at Berkeley in the 1970s, delving into developmental economics. He did field work in the fledgling economies of Nepal, India, Bangladesh, Kuwait and Syria. "It became clear to me," he said, "that these economies were not perfect machines, and had had to have come about somehow."

His theory, if accurate, meant that most complex systems, for example markets, were ever-evolving and did not always produce the best outcomes, and that few could predict what those outcomes would be.

This was the basis of complexity science, a still-young discipline that Dr Arthur said was really a way of thinking, rather than a set of tools to solve problems.

Such thinking enables people to become more adaptable and so more resilient, however volatile the changes in the global economy, the weather or their personal circumstances.

But where did Dr Arthur's theory leave economists, whose discipline was premised on predictability and utter order; or capitalists, who believed that markets produced only superior outcomes?

"I knew I'd be slammed or even murdered for it," Dr Arthur recalled wryly, adding that he held off publishing his theory for four years for fear of being ostracised by his peers.

Indeed, his European peers were "boggled" by it, while his American counterparts "bitterly resisted the idea that capitalism could produce anything inferior".

For years, he added, no American journal "would touch" his theory. But in 1983, Britain's prestigious The Economic Journal published it, having cautioned him that "we don't expect too many people to notice or cite it".

That article has since garnered almost 6,000 citations from the scientific community, a huge number in academic circles.

So how was it that he thought up something so contrarian?

It began with a trip to Hawaii in 1973, when the US state had just introduced traffic lights. "I began to think: What if Hawaii was run by libertarians led by Milton Friedman, and its drivers were free to choose on which side of the road to drive? If I came back in six months, what would I see? Well, a lot of wreckage!

"But also, if most cars were driving on the right, I would be a fool to drive into traffic. That meant that convention could emerge spontaneously."

Then he got to thinking: What if humankind could innovate such that new energy sources like wind power could make up for, say, a shortfall in coal? He called developments like innovation a "positive feedback mechanism" which could yield "increasing returns".

Take Microsoft, for example, he noted. It initially spent billions of dollars to develop the first Windows disk.

"The second Windows disk cost a fraction of a per cent of what it cost to make the first," he pointed out. "And as more and more people chose to use Windows, the system was out there more and more, which gave Microsoft a bigger cost advantage. So in information technology, the further a thing got, the further an advantage it had."

Today, Dr Arthur's idea of increasing returns has become the way for economies to grow in the Information Age.

"At any time, you're not looking to be perfect; you're looking to be well-positioned," he said, of how best to cope.

In a complex future, then, how should governments govern?

They should, he said, not think of themselves as power station controllers who "pushed buttons and pulled levers", but rather as park rangers who "put up boundaries here, encouraged activities there, made water available and habitats liveable".

Above all, he said, leaders today had to govern with "a fairly light touch" by enabling what "seemed inevitable" and occasionally steering people away from "boxed-in thinking".

He then stressed, for good measure: "It's not that you don't need governments or that you can't rely on markets. But you don't want governments to plan everything. It's about allowing for the natural unfolding of systems."

That said, he cautioned governments everywhere against pandering to what people thought mattered most. "In too many countries," he mused, "we're looking to make people comfortable, but not looking to make them highly adaptable and, more importantly, having an inner toughness.

"You'd find a lot of resilience in people who've been through extraordinary difficulties, such as Mr Lee Kuan Yew's generation, who had to deal with the Japanese, the British and so on."

Grasping complexity science can be frustrating at first, especially because it only hints at how one should approach problems, and so it offers no solutions.

During this interview, he allowed that, perhaps, he had been able to see systems differently, thanks to the several disciplines he had absorbed in his lifetime.

Calling himself "intensely curious", the Irish-born thinker trained first as an electrical engineer at Queen's University, Belfast, and then received master's degrees in mathematics and economics from the University of Michigan and University of California at Berkeley before securing his PhD in operations research, also at Berkeley.

His theory got a real leg-up at Stanford University in 1987, when his mentor, Nobel economics laureate Kenneth Arrow, got him to take part in a 10-day conference he was organising at the new Santa Fe Institute in New Mexico with 19 other thinkers to mull over the issue of how strong the foundation of economics was. That question was posed by Citibank's then chairman John Reed, after his bank lost billions in loans to Latin American countries.

After that gathering, Dr Arthur was chosen to lead the institute's first research project, on Mr Reed's query, which Citibank funded.

Dr Arthur recalled: "To me, it was like a Cinderella story. I had been shut out of economics to quite a degree for some years, and then suddenly, I was invited to the ball."

At the time, some of the physicists at the conference had just set up the institute to nurture complexity science.

Dr Arthur's earlier ideas squared well with those of the institute's founders, who were physicists. "Physicists embrace uncertainty because of their work," he said, "and they knew well how complicated systems were not predictable and could 'lock into' inferior outcomes."

The founding of the institute coincided with the wider availability of computers, which meant that the experts could now examine, say, interactions in a flock of geese as computers enabled them to write a program for each bird, and track their interactions easily.

In 1990, Dr Arthur published his "increasing returns" theory in The Scientific American magazine, which got American scientists, if not economists, abuzz.

It also garnered him the Schumpeter Prize in economics that year.

By then, he was teaching at Stanford, having become the university's youngest endowed chair in economics and population studies in 1983, at the age of 37.

After leaving Stanford in 1996, he held various other appointments and fellowships at the institute and in other institutions, including at the legendary US technology incubator Palo Alto Research Centre (better known as Parc). Former Xerox Corporation chief and Parc director John Seely Brown once said: "Hundreds of millions of dollars slosh around Silicon Valley every day, based on Brian's ideas."

Dr Arthur is now an external research don at the institute, as well as a visiting fellow at Parc's Intelligent Systems Lab.

What is his biggest wish for complexity science?

He said: "That it were not called complexity science, because people could see it as a cult! It's actually what everyone needs to build his inner toughness for the most difficult of times."

Dr W. Brian Arthur on:

The hardest thing for advanced economies to do today

You would think that it's getting people to adapt, but what is harder to do is to construct a society where people are adaptable, and not just adapting.

What was strange about the US presidential election

That neither Mr Barack Obama nor Mr Mitt Romney really asked anything of anybody. Mr Obama did say, 'We're going to ask rich people to pay more taxes' but he wasn't like John F.Kennedy, who said 'ask not what your country can do for you - ask what you can do for your country.'

What complexity science is - and is not

It tells us how heavily interconnected networks work, but it does not show people how to think - that's up to the education system or the informed public to do so. Complexity scientists are interested in understanding the world, not designing it.

How relevent complexity thinking is

There is a certain idea that you should only do science if it's relevant to public policy, so you should study complex systems only if it could improve the traffic system in Singapore. But if I want to work on malaria, it doesn't mean that I want to see more malaria in the world.

Why the quest for perfection or equilibrium is essentially flawed

The standard way of looking at the world is all in categories or boxes which, when put together, forms a system that we like to think is in some way optimal or perfect. But if it's not, we think that we can push it here and prod it back into some degree of perfection and we think such a system will last for some time and maybe change into another system that we can perfect very rapidly.

Opinion, The Straits Times, Friday, December 28 2012, Pg A30