What can chaos theory teach us about management and governance? A lot, as it turns out. Decentralization and self-organization in business, tech, and government are trends we often laud, and have even claimed that they will help societies avoid catastrophes by "collapsing upwards". But how does it work in practice, and how does it compare to traditional hierarchy in getting things done?
In a nutshell, decentralizing is a way to be as big as a dinosaur and as nimble as a cat at the same time. Consider a swarm of bees: it can effectively be an animal twenty feet wide, a hundred feet long, with a thousand eyes and sophisticated complex behavior--bigger and smarter than most dinosaurs--but it can turn on a dime (in several directions at once, no less!) and is unburdened by the metabolic overhead of a single huge body. Also because of its distributed nature, and the multiple redundancy of its many separate bodies, it is much harder to kill. Activists with cell phones have learned this lesson; some corporations are beginning to learn it as well. Giants such as BP and Toyota have flattened the hierarchies in their management structures to become more innovative and nimble while remaining large companies, and it is proving effective.
Both academic and practical research has been done on changing management structures to be less hierarchical and more flexible. The ICoSS Project at the London School of Economics studies how complexity theory (also known as chaos theory or systems theory) can change organizational and management structure. The New England Complex Systems Institute, who have said "the inability of conventional hierarchical control and the need to understand distributed control, self-organization and networks is increasingly apparent", have a book and even teach a course on the subject. Perhaps most widely known is the Society for Organizational Learning, founded by Peter Senge; his book The Fifth Discipline is a classic explication of systems-thinking for management.
More recently, The Economist mentioned a book (more of a pamphlet, really) that has been put out by Gerard Fairtlough, former CEO of Shell Chemicals UK and founder of biotech firm Celltech. Fairtlough's book, called The Three Ways of Getting Things Done, is a great intro for those looking to dip their toes into the water, and describes traditional hierarchy plus two proven-viable alternatives: "heterarchy" and "responsible autonomy". Heterarchical systems share power--for example, a board that votes to decide issues, or different branches of government that have checks and balances through separation and overlap of power. Responsible autonomy is purer self-organization--i.e. it has no inherent structure. It distinguishes itself from anarchy by holding decision-makers responsible for the outcomes of their decisions.
Even organizations with rigidly hierarchical governance-structures can do a lot to flatten their informal channels of communication and influence (which all management theory admits are as important--sometimes even more important--than an organization's formal structures.) For instance, the Society for Organizational Learning and other groups often recommend institutionalizing After-Action Reviews in companies. After-Action Reviews were first used by the US Army in the 1970's and spread to the business world in the 1990's; they encourage feedback up and down levels of hierarchy by creating temporary hierarchy-free times when criticisms and suggestions can be aired by everyone involved in a project. They are credited as being a useful tool in transforming top-down authoritative culture into both-ways collaborative culture, even when official management structures remain the same. This is especially useful because changing a company's culture is often one of the hardest things to do, especially in giant hide-bound firms.
Another example of alternative corporate structures is the co-op. A long-time favorite of labor justice activists, the co-op structure makes all workers owners, and usually requires corporate governance to be a democratic system rather than an autocratic system. Most co-ops flounder not because it is a bad structure, but because the personalities most likely to create a co-op are generally the least likely to be business-savvy. Some co-ops have been huge successes, perhaps most notably The Co-Operative Group in the UK. It runs a bank, over a thousand grocery & convenience stores, insurance, internet service, a travel agency, and many other businesses; it has been described by The Guardian as "Britain's biggest funeral business and its largest commercial farming operation".
Do you have examples of large successful non-hierarchical businesses (or large businesses that are becoming more competitive by flattening their hierarchies)? We'd love to hear about them.
Semco springs to mind
I've heard the Basque Mondragon Co-operative Federation is one of the larger cooperative structures around, although I haven't read too much on their structure. Seems like it is in part a flattened Hierarchy. For the purpose of seeking additional examples of flattened Hierarchy it is best to keep the scope somewhat general... But, I would like to go ahead and ask: what is success? Financial competitiveness is what first come to mind; however, there are many ways to measure that alone, and there are many other important measuring sticks - such as individual/workers/human rights/respect, and health.
Here are two links that talk about The MCF, and (in part) its structure:
More about coops from the University of Wisconsin's Center for Cooperatives:
John Lewis, a national Sears-like retailer in the UK is a co-op of sorts: http://www.johnlewispartnership.co.uk/ - they've got a decent write up of their structures and philosophy on their site.
There is of course the famous Mondragon Corporation Cooperative, which is a hierarchically but democratically organized cooperative, and the first industrial group of basque country:
This topic certainly needs mention of Visa/Mastercard, a trillion-dollar entity with virtually no central management. Its founder, Dee Hock, coined the term "chaordic" to describe this kind of organizational structure. Learn more at:
Also: Alcoholics Anonymous, Overeaters Anonymous, and similar groups.
Jer, are complexity theory, chaos theory and systems theory really the same thing?
Jon: Complexity theory and chaos theory are exactly the same thing. Systems theory more or less is too, but the people doing it tend to study the interactions of multiple-component systems, whereas chaos theory also talks about the complex behavior of single things.
David: Yes, how silly of me! I totally should have mentioned VISA.
Others: thanks for the references! Keep 'em coming...
I don't mean to debate Mr. Faludi's mastery of the topic under discussion, but I understand chaos theory to be a (rather small) subset of complexity theory.
Specifically, while complexity theory attempts to understand systems whose behavior is governed by nonlinear relationships between components, chaos theory is concerned only with those nonlinear systems whose behavior is sensitive to initial conditions.
Not all nonlinear systems are sensitive in this way. Consider for example an ecosystem in which the total population is far below the carrying capacity, comprised of several different species that interact in different and unspecified ways. One can imagine that for a large range of initial populations (species A has initial population a, B has b, etc.), the populations will stabilize into a roughly constant proportion after a long time. Such a system would be stabilizing (as opposed to being chaotic) despite being nonlinear (complex).
One can also imagine however that if we started the total population _above_ the carrying capacity (for example), the long-term or "steady state" behavior could be strongly dependent on what proportion of species we started off with. Such a problem would be formally defined as chaotic.
This is how I understand the terms; please correct me if this understanding is erroneous. (More detailed information, including formal definitions, can be found in the relevant Wikipedia entries and the textbooks they reference.)
As for "systems theory", I have heard the phrase used to refer to a great many things, from a philosophical framework to a theory for optimizing manufacturing processes in operations research. I feel unqualified to comment on any of these.
I hope this has been more instructive than pedantic, and that my example has made clear the critical importance of understanding complex and chaotic systems to our current ecological and demographic predicament.
Also, many thanks to Mr. Faludi for these awesome posts and insights into the phenomenally complex workings of the business community. In the spirit of answering the question that was actually asked, I'd like to offer up Gore Technologies as a fairly famous (mentioned, if I recall correctly, in Malcolm Gladwell's _The Tipping Point_) example of an impressively effective company operating without central management.
Happy holidays, everyone.
Six, you're technically right about the difference between chaos & complexity. I may be a bit behind in my jargon, but I don't think anybody cares when talking about the fields in general--I think people only care when classifying a particular system, as you did. ...But I'm not a chaos or complexity or systems theorist, so maybe they do care.
Anyway, thanks for the tip on Gore Technologies.