Skip to content

Complexity, Spontaneous Order, blah, blah, blah…and Wow

UPDATE: Thanks to the commenters who confirmed the “hostile reactions” thesis while disavowing hostility :>)… By the way, I am surprised nobody has yet mentioned that blogs, Twitter, Facebook, etc. are superb examples of Spontaneous Order.

I was surprised by hostile reactions to mentioning complexity on the Ivory Coast coup debate. Of course, I dish out hostility like water myself, so it’s only fair that I got accused of mindlessly mumbling complexity to sound trendy.

Regardless of exactly what language you use and whom you give credit for what ideas, I think we can all agree that Complex Adaptive Systemic Emergence of Spontaneous Order General Equilibrium in Development (acronym? don’t go there) incorporates the following ideas, all of which I think are really exciting.

  1. Nobody designed it.
  2. The old idea that complex phenomena imply a designer (an old argument for the existence of God) doesn’t survive. Surprisingly complex phenomena can emerge without design, like Dawkins’ “Blind Watchmaker” in evolution. The same in development.

    Corollaries: nobody needs to direct it. nobody needs to even understand it.

  3. Surprisingly simple behaviors and rules can result in complex phenomena.
  4. A lot of the complexity of nature  (to keep using the evolution metaphor) results from the very simple principle of Natural Selection. The analogue in economics is the Invisible Hand, which is also very simple (the latter inspired the former, by the way).

  5. Spontaneous order is not automatically good.
  6. The Mafia is a spontaneous order, case closed. In economics, we get the good “Invisible Hand” outcomes when private returns equal social returns (which, uh, is not true in the Mafia spontaneous order).

  7. Actions can have unintended consequences.
  8. In neoclassical economics, one example is the Theory of The Second Best. If there is one part of the economy where private returns do not equal social returns, then correcting a different part of the economy to make private returns equal social returns could actually make things worse rather than better.

    It should not be too hard to think of Finance examples from the recent Crash.  Could somebody please suggest  something like how going more “free market” in  financial deregulation when the government was implicitly bearing a lot of the risk….made bad things happen?

  9. What do you mean by “actions”?
  10. In development, the trend is to think of more and more things as the undesigned, unplanned outcomes of some spontaneous order…institutions, politics, cultural values, social networks, and so on. So there is nobody left to stand outside the whole system and pull a lever to move everything.

  11. Partial equilibrium analysis still works.
  12. While (1) through (5) apply to the whole system, you can still simplify by isolating a particular part where you can usually link actions to consequences (and then cross your fingers that the general equilibrium effects don’t cancel out or reverse the partial equilibrium prediction).

    So I still feel confident saying that price controls will lead to long lines and are a bad idea, that expropriating private property will decrease investment, and that letting you choose for yourself is usually better than having the Central Authorities tell you what to do.

This entry was posted in Economics principles. Bookmark the permalink. Follow any comments here with the RSS feed for this post. Both comments and trackbacks are currently closed.


  1. When you talk of the Invisible Hand, do you mean in the senses that Adam Smith actually used it? Or in the made-up sense of modern economics which deludes itself it is using Adam Smith’s metaphor?

    Posted January 19, 2011 at 1:20 am | Permalink
  2. Jacob AG wrote:

    “…letting you choose for yourself is usually better than having the Central Authorities tell you what to do”

    Aren’t price controls, long lines, and the Central Authority all parts of spontaneous order equilibrium? Isn’t that a big part of studying political economy?

    Posted January 19, 2011 at 2:26 am | Permalink
  3. William Easterly wrote:

    Lorenzo, yes I mean the latter self-deluded kind, also known as good economics.

    Jacob, you have captured the irreducible tension between (5) and (6). On one hand, the policy recommender should try to understand the political economy of existing distortions (5). On the other hand, the recommender should recommend what are good economics principles in a partial equilbrium sense (6). This is why policy recommendation itself is more like an art than a science.

    Posted January 19, 2011 at 9:01 am | Permalink
  4. 1) “Regardless of exactly what language you use and whom you give credit for what ideas…” Definition & attribution are fairly important elements of the scientific process.

    2) “Surprisingly complex phenomena can emerge without design…” Distributed design is not the same as no design. In human societies (as opposed to cellular automata) agency is present everywhere. People think, decide, and act. As a consequence, things happen.

    3) “Spontaneous order” is a term that is more confusing than edifying. Development is driven by initiatives undertaken in the face of mundane but ubiquitous economic phenomena such as non-zero transactions costs, asymmetric information, intrinsic uncertainty. (Ref. Hayek on the importance of local information.)

    4) “I still feel confident saying that price controls will lead to long lines and are a bad idea, that expropriating private property will decrease investment, and that letting you choose for yourself is usually better…” This is good, reassuring material from a course in the principles of microeconomics (prices matter!) but, as you suggest, it doesn’t follow at all from the foregoing.

    Overall: Better to redirect to complexity as a guide to learning rather than as an implicit justification for inaction: (@ithorpe)

    Posted January 19, 2011 at 9:08 am | Permalink
  5. Ian wrote:

    What a co-incidence that we both chose to blog on this topic on the same day. Spontaneous order I suppose!

    My take on it is that complexity is all around us. It’s not new, even if talking about it is, and that while it can make development more difficult and unpredictable, there are still simple, practical approaches you can use when running government or aid programmes that don’t necessarily require a deep knowledge of economics or complexity theory.

    Here’s the post in case you are interested:
    Who’s afraid of complexity in aid?

    Posted January 19, 2011 at 9:27 am | Permalink
  6. Pablo Abitbol wrote:

    In this context, I am still trying to figure out the soundness of this argument by Goodin:

    “Insofar as the social world is subject to evolutionary pressures, we might want to apply design principles to reshape the selection criteria and social reward structures according to which some innovations succeed and others fail.”*

    Which, by the way, reminds me of this nice paper by Vanberg.

    * Robert E. Goodin (1996) Institutions and Their Design; in Goodin, R.E. ed.: The Theory of Institutional Design. Cambridge: Cambridge University Press, p. 29.

    Posted January 19, 2011 at 10:41 am | Permalink
  7. I don’t get it. When I don’t know, I know that I don’t know, and I rely on “best practices” because that’s all there is. That’s how doctors prescribe aspirin because it doesn’t much matter that until now they don’t know how it works.

    Complexity “theory” tries to do the impossible: explain the lack of knowledge in the sense that we can’t even use Occam’s Razor to choose a “reigning” theory, because, by definition, complexity means there can’t be a reigning theory. So, what do you have? A lot of mathematics and “theory” to explain a wide error band. Shakespeare got it right: Much ado about nothing.

    Posted January 19, 2011 at 10:47 am | Permalink
  8. Ana wrote:

    It’s like, how do you learn to be so maddeningly confidently bloviating in the blog, which is so much the opposite of PhD training and the intellectual endeavour as a whole. “The Mafia is spontaneous order. Case closed.” OMG. What human interactions would qualify as “deliberate order” if that were indeed a reasoned statement?

    Posted January 19, 2011 at 12:20 pm | Permalink
  9. Dan Kyba wrote:

    North describes the ever-changing world and our imperfect perception of it as: ‘non-ergodic’.

    See: North, Douglass “Understanding the Process of Economic Change” p. 22.

    Posted January 19, 2011 at 12:29 pm | Permalink
  10. Reuben wrote:

    “1. Nobody designed it.” Well, maybe you can take a (deeper) look at Asian development strategies.

    2. “I was surprised by hostile reactions to mentioning complexity (…)”. The reactions (hostile or not) were about the simplistic use of the words “complexity theory”, without really arguing from complexity theory. You might as well say that Development is caused by the laws of quantum mechanics; we learn as much from the latter as from your new research program.

    3. I don’t have a problem against a research program on complexity in the social sciences, and in particular in understanding the process of economic development. I would just like to see it.


    Posted January 19, 2011 at 12:35 pm | Permalink
  11. Merijn Knibbe wrote:

    Way back, in the tens and twenties and thirties, and fourties and fifties – governments all over the globe had an agenda: roads, the elctricity grid, housing, sewer systems. They did a pretty good job. Well, not everywhere (read: Marquez: El amor en los tiempos de colera), for some reason or another. At the same time, companies did a good job, turning out increasingly sophisticated cars and light bulbs which were used to drive over these roads and to light these houses. At the same time, households did a good job, learning to drive, buying cars and lightbulbs – you get it. The government and the households and the companies made plans to do this. Nobody planned the entire thing, everybody just stumbled ahead on the way paved by the others – but it isn’t spontaneous either. Well, that’s to crash – quite a lot of our present health system was entirely planned – using roads and cars and light bulbs and whatever.

    Posted January 19, 2011 at 1:03 pm | Permalink
  12. Sticking to applications of complexity theory in the social or human sciences (the notion of a “spontaneous order” is an older and more profound topic), one should consult Ben Ramalingam’s ODI paper at: . Ben and colleagues make a sustained attempt to usefully apply complexity theory to the problems of development aid–but I fear with little success. One can always reformulate some bits of old wisdom (openness, restraint, humbleness, courage) in terms of the jargon of some new faddish theory, but that is hardly a distinctive contribution of the theory. As Ben notes, there has for some time been a craze in organizational theory and business management to apply the buzz and jargon of complexity theory but with little if any results that are new or distinctive. Interconnectedness! Nonlinearity! Sensitivity to initial conditions! Unintended consequences! Adaptive agents! Wow!…
    One could go on but I might try to cut to the chase and indicate why theories that may give some insights when applied to physical systems (e.g., “self-organizing” sand piles) and insect societies may rather “miss the boat” when applied to human affairs.
    The mistake in applying complexity theory to human relationships such as the education, management, development aid, and helping in general is that the basic problem is NOT that the human “systems” are complex, “messy,” nonlinear, etc. The basic problem, across the whole range of the human helping relationships (like aid) between what might be called the “helper” and the “doer,” is that success lies in achieving more autonomy on the part of the doers, and autonomy is precisely the sort of thing that cannot be externally supplied or provided by the would-be helpers. This is the fundamental conundrum of all human helping relations, and it is the basic reason, not complexity, why engineering approaches and the like don’t work. Thus the application of complexity theory to development aid–as if the basic problem with aid was the complexity of the systems–is “unhelpful” from the get go. I could go on but this is only a comment; one could write a whole book on the topic.

    Posted January 19, 2011 at 1:21 pm | Permalink
  13. William Easterly wrote:


    You are being coy, you did write a whole book on the topic! (and I like the book a lot!) Please feel free to add a link to your book in the comments section!


    Posted January 19, 2011 at 1:27 pm | Permalink
  14. Troy Camplin wrote:

    Actually, the Mafia is an organization, not a spontaneous order. It’s a strong-bond network, while spontaneous orders are weak-bond networks of individuals and organizations.

    I talk about self-organizing complex systems quite a bit at

    Posted January 19, 2011 at 1:38 pm | Permalink
  15. Efi Pylarinou wrote:

    “Industrial accidents and financial disasters have a lot in common.”
    A great read On complex, tightly coupled systems, as nuclear plants and financial systems.


    Posted January 19, 2011 at 2:28 pm | Permalink
  16. Ok, the link to the blurb for my book, Helping People Help Themselves: From the World Bank to an Alternative Philosophy of Development Assistance (U. of Michigan Press, 2005) is: . There is also a South Asian edition from Tulika Press.

    Posted January 19, 2011 at 3:48 pm | Permalink
  17. Perhaps obviously, I share Bill’s enthusiasm for this topic – although I don’t necessarily agree with his final point – I have blogged about the politics of self-organisation here:

    David, I wanted to say thanks for the comments on the ODI paper I led on. I’m a big fan of your work and have learned a tremendous amount from your writings over the years. However, I can’t accept your summary as an Official Truth :)

    I do agree that there is a fundamental problem of autonomy of the recipient, but this doesn’t mean that complexity science concepts don’t have something to add to our thinking.

    There are growing examples of the use of the complexity sciences to look at domestic policy issues in developed and developing countries alike, from education to healthcare to community resilience and economic growth. Researchers are using it to look at the dynamics of disasters, at supply chains, at processes of innovation – the list goes on and on.

    Not all of this can be dismissed as the work of management consultant ‘jargonauts’ – a fair amount is being undertaken by serious thinkers and practitioners…

    To pick just a few examples, Elinor Ostrom has written on complexity in institutions and in the context of resilient communities (and is using our aforementioned paper in her teaching at Indiana); Alan Savory has used complex systems approaches to establish one of the very few successful anti-desertification programmes (; the Bank of England’s director of financial stability has used it to rethink the nature of risk in financial markets (see forthcoming post). There are many others.

    This work does seem to have direct relevance for – indeed, some is focused on – the challenges we face in the aid sector.

    I certainly don’t think that complexity is at the heart of all of the problems facing the sector, and went to some length to say so in the 2008 paper.

    But I do think that there are some entrenched and longstanding problems in aid which can be thought about differently using ideas from the complexity sciences.

    And as Einstein said, you can’t solve problems using the same kind of thinking that created them…

    Posted January 19, 2011 at 6:57 pm | Permalink
  18. Ben, I do appreciate that your uses of complexity theory have been guarded and (as one can see from my book) I am certainly a great fan of eclecticism and interdisciplinary thinking. If anyone comes to some insights through complexity theory (as I also have, e.g., the series-parallel interplay between “exploitation versus exploration”), then that is great–even though other routes may also have been available. Some of the best thinkers in development such as Hirschman (who wrote the Foreword to my book) and more recently Ostrom are vigorously interdisciplinary (which is why Elinor has used complexity theory along with game theory and much else in her work). They exemplify using whatever part of the human sciences that is best adapted to analyzing a problem, rather than trying to fit every problem to the Procrustean bed of one’s academic credentials–which would be the occupational disease of economists if it were not their occupation. In that spirit, it is fine that you are mining complexity theory for insights about development aid.
    My problems lie in how seemingly every advance in the natural sciences is turned into a fad, usually first in management theory, which is then used to avoid looking at deeper persistent sources of dysfunctionality. In business enterprises, management sits astride huge organizations based on the employment relation, but then constantly tries to escape the resulting dysfunctionality by surfing the latest fads popularized from the natural sciences. Similarly, we see the large development aid bureaucracies that are deeply failing for structural reasons but constantly grasping for the latest fad-theories to explain “why it wasn’t working as expected” and to provide rhetorical cover for their “new ways” of doing development assistance.
    In short, my message is: eclectic interdisciplinary approaches to development, Yes; new popsci cover stories for the failures of the development aid bureaucracies, No.

    Posted January 19, 2011 at 9:25 pm | Permalink
  19. Roger wrote:

    Langton’s Ant is an example of how order and structure can emmerge from very simple basic rule, watch it here:

    Posted January 20, 2011 at 11:19 am | Permalink
  20. rick davies wrote:

    Wrong: “Nobody designed it”
    Right: There was no _central_ planner (either a God or a Central Committee)

    As far as I understand, in biological evolution and in human social systems, there is still a lot of _localised_ planning and responding by agents.

    The important point is that none have comprehensive knowledge of the overall system, and most are concerned with locally maximising their welfare.

    Interetsing question: How much, if at all, will increased knowledge of the whole system e.g. via greater aid transparency, affect aid actors local behavior?

    Posted January 20, 2011 at 4:45 pm | Permalink
  21. Jacob Oost wrote:

    I don’t like point number one. Or at least, I don’t like bringing evolution into it. Spontaneous order of the kind Hayek was talking about involves humans making decisions. Evolutionary theory of the kind Dawkins is talking about involves a purely naturalistic process. They are quite different.

    Posted January 21, 2011 at 3:23 am | Permalink

8 Trackbacks

  1. […] This post was mentioned on Twitter by Mark Thoma, Ryan Briggs and others. Ryan Briggs said: .@bill_easterly slows down, simplifies, and says something thoughtful and cogent about complexity and development: […]

  2. By Les liens du matin (79) « Rationalité Limitée on January 19, 2011 at 2:12 am

    […] « Complexity, Spontaneous Order, blah, blah, blah… and Wow » – William […]

  3. […] the original post: Complexity, Spontaneous Order, blah, blah, blah…and Wow AKPC_IDS += "9841,";Popularity: 50% […]

  4. By Mes liens du matin (5) – Parallèles et Paradoxes on January 20, 2011 at 3:19 am

    […] Complexity, Spontaneous Order, blah, blah, blah…and Wow par William Easterly sur Aid Watchers. […]

  5. By Is aid really so complex? — Evolving Economics on January 20, 2011 at 8:29 am

    […] this third point matters in the current debate. Complexity is not the major problem. As stated by David Ellerman in response to Easterly’s update on where the complexity debate is at: The mistake in […]

  6. By Recomendaciones « intelib on January 20, 2011 at 4:17 pm

    […] Recomendaciones Complexity, Spontaneous Order, blah, blah, blah…and Wow, by William Easterly […]

  7. By Some Links on January 21, 2011 at 11:19 am

    […] Bill Easterly ponders complexity and spontaneous order. […]

  8. […] This post was mentioned on Twitter by christhamrin, Jesse Walker. Jesse Walker said: "seemingly every advance in the natural sciences is turned into a fad, usually first in management theory" […]

  • About Aid Watch

    The Aid Watch blog is a project of New York University's Development Research Institute (DRI). This blog is principally written by William Easterly, author of "The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics" and "The White Man's Burden: Why the West's Efforts to Aid the Rest Have Done So Much Ill and So Little Good," and Professor of Economics at NYU. It is co-written by Laura Freschi and by occasional guest bloggers. Our work is based on the idea that more aid will reach the poor the more people are watching aid.

    "Conscience is the inner voice that warns us somebody may be looking." - H.L. Mencken

  • Archives