Talk:Complexity economics

From Wikipedia, the free encyclopedia
Jump to: navigation, search


Eric Beinhocker claims to have invented this term in his 2006 book The Origin of Wealth. It has the same connotation in his book. A highly dynamic, constantly evolving system, in contrast to the closed system of neo-classical economics.

I do not think think the presentation of neoclassical economics on this page is fair.

Claims that are questionable:

1. Micro and Macro Are Separate - Modern macro is based on microfoundations. Many would consider macro to now be an applied micro field.

2. No method for endogenous growth - Endogenous Growth Theory is a mainstream movement in neoclassical economics pioneered by Paul Romer in the late 80's.

3. All knowing / perfectly rational agents - fields of behavioral economics and information economics specifically avoid these assumptions —The preceding unsigned comment was added by (talk) 04:02:38, August 19, 2007 (UTC)

As I read it, Beinhocker criticized neoclassical economics as an obsolete paradigm that cannot be fixed with adjustments, corrections or extensions. He claims that although there are advances like mentioned above, they do not fix the underlying problem: the theories are based on classical physics. It doesn't help that behavioral economics avoids the assumption of perfectly rational agents, if the rest of economics still continues to make the same assumption. Claims 1 and 2 are questionable, though; the only question is whether these theories are used and taught in schools. Again, it doesn't help if the modern cutting edge is realistic when the basic courses still include only the old theories. At least the basic economics I have been taught is purely classical, macro and micro separate and endogenous growth theory not mentioned. Actually, in the table, what is compared appears to be the original form of traditional economics, not its current state. --Vuo 09:22, 19 August 2007 (UTC)
I'm currently studying economics at a graduate level, when I read this article I feel like it is being compared to what economics was 30-40 years ago, and that the writers might be a bit out of touch with what is going on today. --Rob.s.brit 00:55, 16 January 2010 (UTC)
            Transfinancial Economics would be of interest.....

No doubt Compexity Economics would appreciate the research, and development project of Transfinancial Economics. Much work on the latter still needs to be done.

R.Searle —Preceding unsigned comment added by (talk) 11:05, 1 September 2008 (UTC)

German hyperinflation[edit]

These edits added material on German hyperinflation which appear to be original research. The reference to I believe this page of a book doesn't really help verify the material. It's possible complexity economics could be applied to hyperinflation, but it needs a source from someone doing it. CRETOG8(t/c) 09:07, 20 August 2010 (UTC)

= I agree with CRETOG8(t/c) ! However I did some studies about the German Hyperinflation as an example for various hyperinflations, and I rejected the Traditional Economics.--Fox1942 (talk) 13:49, 7 September 2010 (UTC)

POV/UNDUE/CHEERLEADING problems with the article[edit]

The "Behavior of complex systems" and "Comparison with traditional economics" sections are fine. The lede and the Introduction section are strewn with junk:

  • It is one of the four C's of a new paradigm surfacing in the field of economics. - according to whom? This "new paradigm" has been surfacing for about twenty years now and still hasn't really surfaced. Also, I really hate the word "paradigm" and it should be on Wikipedia's list of words to avoid.
  • "characteristics to describe a dynamic system such as emergence, sensitive dependence on initial conditions, red queen behavior..." - huh?
  • Then it's alright for awhile, then we get: "In the light of the new concepts introduced, economic systems shall no more be considered as "naturally" inclined to achieve equilibrium states. " which is just bad writing. "Shall no more"? Is that a dictate from the heavens? "In light of"?
  • Traditional economic models have been constructed by allowing only a very small number of degrees of freedom, in order to simplify models. For example, the relation of unemployment and inflation is traditionally considered to be a simple function with one degree of freedom, allowing for very little entropy. - this is mumbo jumbo. I mean there's some actual content there but it's dressed up in mumbo jumbo.
  • As to the practicability of theoretical instruments, there is also a crucial difference to allow for: traditional economics was conceived before computers had been invented. - so? You gonna throw out all the physics, chemistry, math and biology that predates computers?
  • For example, Pareto's law can be demonstrated to arise spontaneously. - another huh? I think it's talking about Zipf's law (which "traditional" economic models can also generate, uh, "spontaneously").

Volunteer Marek (talk) 20:32, 7 April 2011 (UTC)

>Traditional economic models have been constructed by allowing only a very small number of degrees of freedom, in order to simplify models. For example, the relation of unemployment and inflation is traditionally considered to be a simple function with one degree of freedom, allowing for very little entropy. - this is mumbo jumbo. I mean there's some actual content there but it's dressed up in mumbo jumbo.
No it isn't and you should be less of a besserwisser. Degrees of freedom and entropy are well-defined and exact concepts. This was one of Beinhocker's central claims: the information content of a traditional economic model is too low for accurate realism. Spherical horses and so on. Declaring an argument you dispute "mumbo jumbo" doesn't make it invalid. --vuo (talk) 22:51, 21 November 2011 (UTC)
You should feel welcome to fix the problems if you think you can - I agree about paradigm and the writing style. My impression is that the main thrust of the method is pretty simple: program the agents according to their tendencies (rational or otherwise), program the environment as accurately as possible, put it into a program, and see what arises - somewhat similar to the Sim games except without godlike powers. W. Brian Arthur's page Complexity Economics might be a good place to start. Regarding II | (t - c) 23:10, 7 April 2011 (UTC)
Ok, but how is that different from Agent-based model?Volunteer Marek (talk) 23:28, 7 April 2011 (UTC)
That is about a model, while this page is about a field of study. I actually don't know a ton about the subject - Rosser's 1999 article in JEP is probably a good start (also discusses the Four Cs). II | (t - c) 06:43, 8 April 2011 (UTC)
That article notes that the 4C terminology was used by John Horgan to ridicule this latest fad. (I am beginning to wonder whether this page is a trolling hoax....) Kiefer.Wolfowitz  (Discussion) 14:29, 13 April 2011 (UTC)

Radical nonsensectomy[edit]

Further discussion here. Rinconsoleao (talk) 14:14, 11 April 2011 (UTC)

Economic history?[edit]

I removed a lot of junk. There is plenty of good work done on economic history by intelligent economists who did not fail their qualifying examinations even at the New School. The article should focus on the most reliable sources, and not on promoting a Manichean view of economics, with distorting unconscious grotesques of mainstream economics.  Kiefer.Wolfowitz  (Discussion) 16:06, 8 April 2011 (UTC)

Some useful improvements but I think your comment above might be a little problematic. I think you might be making the mistake of confusing economic history with history of economic thought or methodology. This is, I think a very common but fundamental mistake, made by even those who have passed ECON7065 Advanced Mainstream Economics with flying colours! Best wishes :) (Msrasnw (talk) 14:02, 11 April 2011 (UTC))
I wrote "economic history" as I intended. An economist claiming to describe economic dynamics who ignores history is about as useful as a time-series statistician who ignores data, and we have enough of those already! (The mighty Peter Whittle has confessed that time-series statistics has an unusually high theory/data ratio!)
Best regards,  Kiefer.Wolfowitz  (Discussion) 18:34, 17 April 2011 (UTC)
Jolly sorry Mr Wolfowitz. My reading was that you did not like all the, to my mind strange, methodology / het stuff that seemed to be involved here that you had tidied from this article. I had not seen how your improvements related suggestions for more understanding and use of economic history. But I only undertook a superficial look at the article before making my overhasty "accusation". Sorry again and best wishes (Msrasnw (talk) 19:40, 17 April 2011 (UTC))
Don't worry about it! If I were in your shoes, I would have bet that I really meant "history of economic thought" too! (It would make sense!) I write directly, per Strunk & White, but you should know that usually I am having fun as I write. (If I am irritated, truly, I have trouble hiding that!) On this page, I shake my head at some shenanigans (sometimes even my own!), but how can stay being upset with youthful exuberance on behalf of economics worthy of Ben and Jerry's (whose test asks whether a new flavor be weird enough!)?
I found my way to mathematics partly via the writings of the "lifelong but wayward Marxist" Richard M. Goodwin (and Ian Steedman & Ulrich Krause and Leontieff, etc.), also, so I have considerable sympathy for the research & personalities associated with "complexity econonomics".
Sincerely,  Kiefer.Wolfowitz  (Discussion) 20:45, 17 April 2011 (UTC)

Mainstream economics[edit]

A user included a table describing something called "traditional economics". The user seems to imply that "traditional economics" is influential today. But what the user calls traditional economics sounds like economic theory from the 1920s or so... it has nothing to do with mainstream economics today. Here is an expansion of the table to reflect mainstream economics today:

Complexity Economics "Traditional" Economics (1920?) Mainstream economics today
Dynamic Open, dynamic, non-linear systems, far from equilibrium Closed, static, linear systems in equilibrium Dynamic systems close enough to trend to permit linear approximation
Agents Modelled individually; use inductive rules of thumb to make decisions; have incomplete information; are subject to errors and biases; learn to adapt over time; heterogeneous agents Modelled collectively; use complex deductive calculations to make decisions; have complete information; make no errors and have no biases; have no need for learning or adaptation (are already perfect), mostly homogeneous agents Modelled individually; perform complex calculations; frequently have incomplete information (e.g. Stiglitz; Akerlof; Spence);

frequently learn (e.g. Sargent; Honkapohja); heterogeneity considered when relevant (e.g. Benabou; Heathcote, Storesletten, Violante)

Networks Explicitly model bi-lateral interactions between individual agents; networks of relationships change over time Assume agents only interact indirectly through market mechanisms (e.g. auctions) Mostly assume agents only interact indirectly through market mechanisms
Emergence No distinction between micro/macro economics; macro patterns are emergent result of micro level behaviours and interactions. Micro-and macroeconomics remain separate disciplines Macroeconomics results from aggregating many microeconomic decisions
Evolution The evolutionary process of differentiation, selection and amplification provides the system with novelty and is responsible for its growth in order and complexity[vague] No mechanism for endogenously creating novelty, or growth in order and complexity Lots of work on how firm selection affects productivity (e.g. Hopenhayn; Melitz; Caballero, Hammour)
Technology Technology fluid, endogenous to the system[vague] Technology as given or selected on economic basis Lots of work on technological innovation through research and development (e.g. Aghion, Howitt)
Preferences Formulation of preferences becomes central[vague]; individuals not necessarily selfish Preferences given; Individuals selfish Most work takes preferences as given, assumes individuals selfish; exceptions in contexts, like families, where this matters (e.g. Chiappori)
Origins from Physical Sciences Based on Biology (structure, pattern, self-organized, life cycle) Based on 19th-century physics (equilibrium, stability, deterministic dynamics) Based on mathematics of expectational difference equations (extends control and filtering theories from engineering to take into account forward-looking behavior)
Elements Patterns and Possibilities[vague] Price and Quantity Mostly prices and quantities

The table in the text is extremely misleading, because it sets up a "straw man" with virtually no relation to mainstream economics today. And even withing mainstream economics there is lots of work that goes farther than my summary of the most common assumptions. The table should be therefore be deleted, or moved to a section on "misunderstanding of mainstream economics by self-styled complexity theorists". Rinconsoleao (talk) 10:56, 13 April 2011 (UTC)

This is just junk that is beneath our attention and discussion. It has been added and re-added by one anonymous IP editor who has made no attempt to respond to criticism or seek consensus.
I believe that everybody has agreed that the topics are important, and that even the research mentioned is worthy of inclusion in a proper WP article. However, there is no excuse for POV-pushing and further failure to adapt to WP community norms. This is supposed to be an encyclopedia rather than an (interesting and perhaps often correct) blog.
The discussion of the history of mainstream economics is just wrong. Linear models became mainstream after linear programming and activity analysis arrived, and they were always regarded as polyhedral approximations to complicated sets and functions.
The discussion of hierarchies and networks is old hat, and was discussed in the 1950s by Oscar Morgenstern and Albert O. Hirschman. I challenge anybody to find a mainstream economist who has ever written a word against Hirschman!
Ditto with Ken Arrow and learning by doing and information economics. Arrow and Sen have written some of the best essays on rationality and economics, and who is more mainstream than Arrow and Sen?  Kiefer.Wolfowitz  (Discussion) 12:18, 13 April 2011 (UTC)
Clearly the original table was fatally flawed. But I think that a table could make a good addition to the article. It shouldn't include the "traditional" column (which frankly looks older than 1920s to me), and should remove most of the overlap between the remaining columns. Of course the vague rows (esp. the last) need to be removed or rewritten, but it would be good to see at a glance how this differs from the mainstream. CRGreathouse (t | c) 13:59, 13 April 2011 (UTC)
Let me repeat that a good table would be most welcome. Just add it here on the talk page (trying to avoid original research and maintaining reliability and NPOV) and get feedback before adding it to the article.  Kiefer.Wolfowitz  (Discussion) 18:47, 17 April 2011 (UTC)

the place of...[edit]

This edit added a section on "The Place Of Complexity Economics Among Contemporary Trends In Economics". I think the classification schemes (or whatever you want to call them) of Davis are kinda interesting, but all in all, this section seems a lot of buck for little bang. Also, Davis' schemes are just an idea, not some overall generally accepted practice, so they shouldn't be given a huge weight. Can these ideas be whittled down to just a note on the place of complexity economics, without weighing all those other fields as well? CRETOG8(t/c) 00:13, 14 April 2011 (UTC)

John B. Davis classifies complexity economics as heterodox in practice, but within the sociological mainstream of economics.[1][2]

  1. ^ Davis, John B. "The Turn in Economics: Neoclassical Dominance to Mainstream Pluralism?" Journal of Institutional Economics, V. 2, N. 1 (2006): 1-20
  2. ^ Davis, John B. "The Turn in Recent Economics and Return of Orthodoxy", Cambridge Journal of Economics, V. 32, N. 3: 349-366
Following the above comment, I removed that sentence.
It is a weakness of this Economics Project that it is run by people obsessed with classifying economists as belonging to different "schools" and "ideologies" and grossly over-emphasizing heterodox economics. Maybe such persons should form a project on the whining sociology of economics, and leave economics to people interested in economics.  Kiefer.Wolfowitz  (Discussion) 08:51, 14 April 2011 (UTC)
Strange. The above comment does not justify such removal. Kiefer Wolfowitz is the one who introduced the classification of complexity economics as heterodox [1]. Kiefer Wolfowitz talks on the economics project talk page as if he understands wikipedia policy, but the above is a personal attack, as well as a non sequitur.
Davis is not alone in his view about divisions within the mainstream and the relationship of traditional heterodox schools to those divisions. In particular, Colander, Holt, and Rosser, Jr. have expressed such a view in the introduction to a book of interviews that they edited.
Complexity economics is not one of the traditional heterodox schools or fields. I think it is important in Davis' view that the heterodox, including complexity economics in his view, rejects the principles of orthodox economics listed in the table, with the exception of formal modeling.
If the content of the article were expanded to the size it deserves, the table would appear to be a more reasonable proportion in the article. (talk) 09:44, 14 April 2011 (UTC) RLV
The Economics project has discussed the role of so-called "Austrian economics" in a forth-right manner. Look at the New Palgrave article of the comments of Stigler and Solow, and see whether the recently economics project is guilty of the same flaws that Stigler and Solow criticized in the NP.  Kiefer.Wolfowitz  (Discussion) 09:55, 14 April 2011 (UTC)
To avoid distractions about "heterodoxy", I have removed the classification of this as "heterodox".  Kiefer.Wolfowitz  (Discussion) 09:58, 14 April 2011 (UTC)


The IP editor's claim that traditional economics was based on linear models, unlike complexity economics would benefit from a reality check---of the writings on Richard Murphy Goodman, Ulrich Krause (Ian Steedman), Davis's fascination with Sraffa, etc. On the contrary, these guys have pushed the Perron-Frobenius theorem as though Leontiev models were useful for modern economics.  Kiefer.Wolfowitz  (Discussion) 22:07, 15 April 2011 (UTC)


I am pleased by the use of page references and quotations, especially given questions raised by other editors too. This is a very helpful development.  Kiefer.Wolfowitz  (Discussion) 21:59, 17 April 2011 (UTC)