The paper I mentioned in my question "The Complexity Era in Economics" by Holt, Rosser and Colander is uploaded in RG and you can easily read it.  

The three authors Richard P.F. Holt, Barkley Rosser, and David Colander are members of RG. You can see their papers in their personal page (Contribution page in particular) in RG.

Article The Complexity Era in Economics

Like many, I am sceptical of the neoclassical mainstream. And yet! On the one hand, many - maybe all - of its axioms have been cast into doubt by economic experiments; it is powerful but narrow, in some cases acting as a set of blinkers (in particular, lack of historical awareness - think of how badly mainstream economics performed with respect to increasing financial fragility and the global financial crisis); and a corresponding over-emphasis on formalism (mathematical modelling); and so on. Yet, on the other hand, as an applied economist, I appreciate the mainstream commitment to allow theory always to be confronted by and enriched with empirical evidence, which cannot always be said for other, heterodox, schools. Moreover, the neoclassical mainstream is, in practice, much more open and flexible than allowed for in the caricatures that often pass for critique. For example, institutional insights have become mainstream in the past 20 years or so (nobody would now estimate a growth model without taking into account institutional development); insights from behavioural psychology have been absorbed; and the "economics of information" in and of itself almost constitutes a paradigm shift. Finally, my view is that the mainstream will survive particular criticisms and maintain its position/status until we have an alternative body of theory that explains everything encompassed by the current mainstream and more besides. Maybe the mainstream will accomplish this by evolving (or, if you like, by a Hegelian transformation of quantity into quality)? Meanwhile, back to Stata and my day job! Thank you for a thought-provoking comment. Geoff

 It is true that majority of criticism  on neoclassical (or mainstream) economics was focused on the lack of some factors. For example, as Geoffrey Thomas Pugh pointed it, there were many criticisms that denounce that the institutions were not  taken into consideration. Another pattern was that humans were not totally individualistic. There were many patterned criticisms like these. The trouble with them was that there were no alternatives. One accuses the lack of institutions, but did not offer how to analyse the effects of institutions.        

Compared to this kind of criticisms, complexity economics offers much more positive  view points. However, Holt, Rosser and Colander's claim stops at an offer of new tools. They emphasis the newness of complexity idea. They have a strong support from outside of economics, because complexity is becoming a  key concept for physics, biology and engineering. And yet, they have no strong message on the theoretical core of economics. What they observe is the sentence I cited in my question:

• The complexity era has not arrived through a revolution. Instead, it has evolved out of the many strains of neoclassical work, along with works by less orthodox mainstream and heterodox economists.

Their claim is eclectic and easier for everybody to accept than a preach of a theoretical revolution. A new era may come, but major part of present economics will be kept intact. We may expect for a short time that somethings new will come and we will be deceived soon. Our skepticism on the state-of-the-art economics will continue. In my opinion, a more fundamental change in the center core of  economics is necessary.  Holt, Rosser and Colander are not showing this essential point.

Yesterday, I read Wolfram Elsner's paper "The Dichotomy, Inconsistency,   and Peculiar Outmodedness of the `Mainstream` Textbook." It is up in RD and you can read it if you click the window below.

The paper examines how microeconomics textbooks such as Varian are organized. The authors repeat their "standard" explanations in major part of the textbook but add some "new" contents. Oligopoly and strategy of players (game theory) appear after 500 pages.  Behavioral economics comes in Chapter 31 among 39 chapters. Topics like externalities (including commons and network externalities) and assymmetric information are treated in Chapters 35-38.  

In Some Conclusions, Elsner points that

• While the general image of [those million -selling textbooks] ... is one of "established truth," "solidity" and "reliability", "cutting-edge methods" and the like, a closer look at them reveals that their generative pattern is not only one of arbitrariness, simple "additiveness", ..., but also of trivializing new issues, non-integration of new themes or methods, lots of remaining inconsistencies, and thus, in all, of inferior quality.      

This assessment is indicative for about complexity economics. What I worry in Holt, Rosser and Colander's article is that the same treatment will be adopted for complexity economics. Mainstream textbook authors will adopt some of it as new topics and do not change the main part of the theory. They will only add new theory as some of curiosities.  

Research Review article of leading microeconomic textbooks

Dear colleague, hello Yoshinori,

i fully agree with you that complexity economics, properly (i.e., consequently) conceived with all its implications, is a paradigm of its own and, thus, a scientific revolution vis-á-vis the neoclassial mainstream.

i have written some stuff about that. but, as you cited my argument from the mainstream textbooks review, the alleged "flexibility" and "evolutionary change" of the mainstream is largely misunderstood, e.g., by colander & co (i know all three of them and appreciate these colleagues in general).

there is some more flexibility and openness in research, but no openness in public statements, policy advice, textbook writitng and mass class education (mainly undergraduate). the world is becoming ever more neoliberal and the mainstream continues to play its active role here.

so we can no longer leave the mainstream appearing in the soft glow of a pink light, as it is strategically positioned to play its role. its individual representatives often enough reinterpret even their most interesting research results in the sense of the current public atmosphere of the ruling powers and their media. so we have to say that it is, in fact, schizophrenic, inconsistent, and lacks honesty and often enough quality.

best wihes,

wolfram.

Hi Yoshinori,

I like how you shake things up! You have given me some reading to do. But, before you throw out neoclassical theory, just remember that perfect competition has never been suggested as "real"; instead it is a means of seeing implications - a sort of if/then approach. Classical economics has been much maligned because it is viewed as a explanation for economic behavior. It is not and never was. It was a call to action to avoid mercantilist mistakes. (For instance, no Classical economist thought we automatically had full employment! That's why they had so many business cycle theories.)

No doubt economics is complex. But models provide insight only. After all, a map ain't Colorado.

Keep attacking!

     Pete

Dear Wolfram Elsner,

I thank you for your quick reaction. I have also read your paper:

Elsner, W. 2012 Theory of Institutional Change Revisited: The Institutional Dichotomy, Its Dynamic, and Its Policy Implications in a More Formal Analysis. Journal of Economics Issues 46(1): 1-44.

There are many points in common with my paper below: Evolutionary Economics in the 21st Century: A Manifesto, 2004.

Your are emphasizing relevance of game theory to evolutionary economics, whereas I am much more concerned in the structure of  human "rational" behavior. 

I know Rosser, too. He came to Japan at the occasion of International Conference of Japan Association for Evolutionary Economics, which was held March 26 and 27 in Tokyo. I started to read his and colleagues' article after that encounter.

Article Evolutionary Economics in the 21st Century: A Manifesto

Hi again, Yoshinori

One more thing: we needn't and shouldn't throw out the rationality assumption just because people are ignorant or stupid. People don't knowingly make mistakes (unless impaired). That's why they don't walk out of windows. Dogs and birds are rational in that way too. Behavioral economists get a big kick out of finding some "irrational" behavior. No. Just ignorant, or improperly tested. Read my "In Defense of the Rationality Assumption" to see what I mean.

It's ok to eat more M&Ms if there are more colors. Don't worry you're still rational!

   Pete

Yoshinori,

will read your paper! thanks.

Pete,

under real-world economic complexity, i still may ask, what if they were rational. but i also have to ask: how rational can the most rational individual really be under certain conditions? in any supermarket, i would need a brain as large as our galaxy to calculate and order quickly all commodity-bundles that are avaliable.

we usually start with rational choice in game theory, but as soon as i have a large and mobile population with heterogeneous agents, agents do face radical uncertainty, and it helps liitle, i am afraid, to calculate what would be first best here (if i could calculate that at all).

thus, if computation, we probably need non-determinstic agent-based computational economics ...

best,

wolfram.

Hello, Peter V. Bias.

We have exchanged some posts in the question page: Return to Keynes, is it sufficient? (2)

https://www.researchgate.net/post/Return_to_Keynes_is_it_sufficient_2?_tpcectx=profile_answers

I have read your paper with your colleagues: In Defense of the Rationality Assumption. 

In the Summery, citing Winston Churchill, you conclude:

• [R]ational choice theory may well be the worst social science methodology ever invented, except for all the others. We should be open to the insights provided by rational choice theory without embracing the approach with religious fervor. The approach can be and has bee extremely useful. It should not be discarded in favor of other approaches except in extreme modeling situations.

You believe there is no better methodology than the rational choice theory. Is it true? You are confusing with rationality and maximization.

As computing complexity theory has revealed we are surrounded by NP-hard problems and we cannot solve them in a appropriate span of time even if we use computers.

I am not opposing the rationality itself. We behave as rationally as possible, but we have no infinite computing capability to maximize our utility or many other targets we encounter in the production site and others.

Instead of maximization, I am proposing to consider human purposeful behavior as "rule-based behavior," "routine behavior" or "patterned behavior". (In Japanese, I say "teikei kōdō".) Wolfram Elsner calls it "patters of behavior."

These behaviors are highly predictable and usable for the construction of new theory. When there is a big gap between our capability and the difficulty of a task, it is better to appeal to a patterned behavior. For this account, see Heiner (1983) The Origin of Predictable Behavior, American Economic Review 73(4): 560-595.

We have a repertory of patterned behaviors or as Elsner put it "set of patterns of behavior." We observe the result and we let our behavior evolve. Rational choice works at this level. This is the rationality which works in a real world and the very foundation of evolutionary economics.

I posed my present question, because I felt that Holt, Rosser and Colander's paper cannot contribute to the further development of economic theory. Complexity will be added to the old core. Textbook writers can boast of the knowledge of cutting edge methodology but continue to spread confused knowledge.

Conference Paper In Defense of the Rationality Assumption

Yoshinori and Wolfram,

I'm more or less in agreement with you both, that you can be overwhelmed with choices. My point was only that we make mistakes and resort to rules of thumb and so on, not because we're irrational but because we're overwhelmed or uncertain. So rationality in modeling as a first-blush assumption still makes sense to me. For instance, notice that  rules of thumb are reasonable, not clearly irrational.

I agree fully that agent-based models are the way to go, not because we knowingly make mistakes (what I would consider irrational), but because that's where all of the economics is. The aggregations methods we use are, I believe, faulty. I'm much more concerned about this than the rationality question.

No arguments here, mostly semantics I guess.

Wonderful discussion!

     Pete 

Peter and Wolfram,

 we arrived at an essential point of our discussion. Peter wrote:

•  My point was only that we make mistakes and resort to rules of thumb and so on, not because we're irrational but because we're overwhelmed or uncertain. So rationality in modeling as a first-blush assumption still makes sense to me.

 I (and probably Wolfram) have never insisted that we are irrational. I know many economists who have claimed that we are irrational and therefore neoclassical economics should be overthrown. I have never argued like that.

 In some cases, maybe we are irrational. For example, we observe many people who kills their friends driven by their strong emotion. It is not important for economics. In overwhelming majority of occasions, economic life is conducted by a rational, calculated judgment. However, economic behaviors are not based on maximization principle, because it is simply impossible for majority of cases.

Peter says that we resort to "rules of thumb", because we are overwhelmed or uncertain. We are thinking in the same way. However, the main point starts from here. If our behavior is not based on maximization principle, how is it organized? Why does it take a form of "rules of thumb" behavior? And how does our rational calculation work?

Peter thinks without any justification that "rationality in modeling as a first-blush assumption still makes sense." Here, Peter identifies "rationality" with maximization. This is a gross confusion. Maximization is impossible! Then, does maximization work as an approximation? Peter thinks that it does as a "first-blush." No! Rationality works in a different way. And this is related to the core framework of economics.  

Many economists think like Peter. They suppose that human behavior is approximated by maximization principle. Have they ever shown or proved that? I doubt if they have ever thought about it. They simply assume that, probably because they do not think on this question seriously. 

Evolutionary economics thinks that human behavior is better described as rule-based behavior. I have discussed about it long and I do not repeat it. I ask readers to read Section 6 of my paper Evolutionary Economics in the 21st Century: A Manifesto (the same paper I have cited in an answer to Wolfram).

Section 6 comprises 6 subsections:

• 6.1 Maximization principle
• 6.2 Expected utility and uncertainty
• 6.3 Rule-based behaviors
• 6.4 An illustration from daily life
• 6.5 Mode of selection of behaviors
• 6.6 True level of choice

I have discussed how our rationality works. It works through evolutionary selection. Mainstream economics does not distinguish two levels of choice. One is a choice inside of a rule-based behavior. It is a selection of parameters which define a behavior concretely. The other is the selection of behaviors themselves. Our behavior becomes more rational and effective through this evolutionary process.

As a conclusion, I believe I can say that rationality works in a totally different way than that assumed by the maximization principle.

Article Evolutionary Economics in the 21st Century: A Manifesto

I have uploaded a working paper which include the discussion above. I wrote in the answer (post number 12) as follows:

• Peter thinks without any justification that "rationality in modeling as a first-blush assumption still makes sense." Here, Peter identifies "rationality" with maximization. This is a gross confusion. Maximization is impossible! Then, does maximization work as an approximation? Peter thinks that it does as a "first-blush." No! Rationality works in a different way. And this is related to the core framework of economics.  

I have discussed long why we should construct an economy which does not rely too much on human rationality. Theories and models based on rational expectation hypothesis is criticized as well.

Some parts are very similar to  "Evolutionary Economics in the 21st Century: A Manifesto" that I have cited above, but you will find many new arguments. 

Any comments are welcome. If some can give a comment before the end of this month (May, 2016), I may have time to revise the relevant part.

Thank you in advance.

Working Paper Microfoundations of Evolutionary Economics

I came across a paper that is probably giving a good example that paradigm change does not occur even if a theory accumulated many anomalies and internal contradictions.

Please see the following question and my answers, in particular, Joel Hellier's two posts on June 17th 2016 and my responses to those.

https://www.researchgate.net/post/What_are_the_pragmatic_problems_with_Heckscher-Ohlin_model

Dear Geoffrey, Peter and all readers,

In the post #2 in this question page, Geoffrey Thomas Pugh argued that despite of all defects of neoclassical economics, it better than other heterodox economics. He my be right. Peter V. Bias, in the post # 5, contended that "models give insight only. After all a map ain't Colorado." You may not be a specialist in international economics, but how do you think the present state of international economics? We have full series of empirical works that refute HOS/HOV models and yet many people prefer to conserve their model. Doesn't this teach that we need a radical change in the international trade theory?

Please read

What are the pragmatic problems with Heckscher-Ohlin model?

https://www.researchgate.net/post/What_are_the_pragmatic_problems_with_Heckscher-Ohlin_model

Who still supports Hechscher-Ohlin model

https://www.researchgate.net/post/Who_still_supports_Hechscher-Ohlin_model

Professor Shiozawa, I would like to make just a humble contribution, namely to recall what professors Nelson and Winter said in their seminal work of 1982 about rationality and the optimization of objective functions: 

"[P]ossession of a complete, clearly defined objective function is not a necessary condition for business operation in the real world; all that is required is a procedure for determining the action to be taken. [...] this proposition has an important corollary: the imputation of such an objective function to the firm is not a sine qua non of effective theory construction."

In fact, there is an implicit assumption here: that the only way to treat rationality in mathematical or formal terms is to make it equivalent to the algorithm of optimization, along with the use of calculus. Nelson and Winter continues:

"Most economists would , we suspect, readily concede the inadequacy of the conceptualization of the firm as a rational actor when the task is to explain particular decisions by particular large firms".

But this is not a departure from the optimization paradigm: economists have actually incorporated information costs and expectations to their decision algorithms. The point of evolutionary economics, as displayed by Nelson and Winter, is that

"There are many other ways to represent motivational influences in a theoretical model; our own models illustrate only a few of the possibilities. In particular [...] the plausible assumption that making money (in some sense) is a dominant business motivation need not be represented as profit or present value or market value maximization . The choice of those specific representations is easily understood as a response to demands for definiteness , precision, and internal consistency. But the source of those demands is not to be found in the realities of business."

The authors suggest a hypothesis for this formal necessity:

"The concern that orthodoxy has lavished on the question of objectives is a reflection of the logical imperatives of its own normative structure- and also, as we have suggested, of its aspiration to reach broad normative conclusions on the efficacy of market mechanisms."

All in all, I find it fascinating that, although there is a great body of evolutionary literature that has come after 1982, there is a wide range of topics in which An Evolutionary Theory of Economic Change is still useful to shed light to.

Dear Arturo,

Yes, Nelson and Winter are the real originator of the rejuvenation of evolutionary economics. My two papers

• Evolutionary Economics in the 21st Century: A Manifesto
• Microfoundations of Evolutionary Economics

(that you can see in posts #12 and #13) are written under the strong influence of their works.

Optimization became an economic paradigm by two main reasons. One is the popular under standing that economic effort is to seek the best result. This is a right opinion as far as it is possible to find that maximum. However, in a real life, in many (or perhaps in majority of occasions)  it is not possible to do that. Objective function itself is often very vague. Another reason is you can do an easy work with optimization. If you define your objective function and constraints, you get a mathematical question. You can pretend doing something economical and mathematical.       

I have uploaded a working paper on the theme how neoclassical economics came to the world. From an interanilst point of view, I explained that neoclassical economics immerged in an effort to solve indeterminate terms of trade problems. In spite of young John Mill's intension, the situation he studied was a pure exchange situatation and this marked a turn from economic of prodcuion to economics of exchange. I believe that it is necessary to return to classical theory of value in order to really reconstruct what Keynes wanted to express by the term "effective demand."

Mill's Reversion and an Origin of Neoclassical Revolution

https://www.researchgate.net/publication/305810200_Mill%27s_Reversion_and_an_Origin_of_Neoclassical_Revolution

Working Paper An Origin of the Neoclassical Revolution: Mill's "Reversion"...

Additionally notice that the results that are being provided by the evolutionary models are not just mirroring the neoclassical ones. The contemporary (21st Century) efforts are moving, along the distinct subtopics, in at least 4 directions:

A) Evolutionary results that replicate from radically different microfoundations a few well-established neiclassical findings.

B)Totally fresh new evolutionary results which contradict many previous neoclassical ones and are gradually being tested against empirical work and _at least in Europe- capturing attention of students, policy makers, business people and technical people, math, computer science etc.

C)A continuous dialogue between formal and appreciative theorizing which keeps our feet on earth. The multidisciplinary dialogue among evolutionary neoschumpetetian people and other humanistic and natural sciences is fluid and solid.

D) Perhaps I miss a rigurous mathematical but teachable text book at undergraduate and Graduate levels. But its appearance is just a question of time.Wait and see.

Dear Prof. Yoshinori Shiozawa,

"the complexity era."

The complexity era has not arrived through a revolution. Instead, it has evolved out of the many strains of neoclassical work, along with works by less orthodox mainstream and heterodox economists.

This contention contradicts the standard theory of scientific revolutions. It sounds like a claim that a scientific revolution is possible without a revolution.

In my view the complexity problems arise not only as economic problems.

They are related to the need of extensive interdisciplinary research and the use of sophisticated tools of mathematics, physics, computer sciences, big data analysis, philosophy, metaphysics, etc. This research is going on, best exemplified by the new findings of chaotic behaviour which changes the paradigm of mainstream economics.

I am unfamiliar with the standard theory of scientific revolutions. Who is going to announce it and by what authority?

Mandelbrot died without any outstanding splendour. Feigenbaum will probably not get any Nobel. The new interdisciplinary science of chaos

was making news some years ago, mostly in mass media, less so among peers. If we look at the academic teaching, who teaches and what is being taught I do not see scholars able to announce even a tiny bit of scientific revolution. They do not understand physics and maths to such a degree which could convince themselves and convert them to a new paradigm.

The theory of chaos, speaking in words of its dynamics, does not expand at an exponential rate, it is rather treated like a kind of fad.

I have no doubt that

the introduction of complexity is a new paradigm for economics and deserves the name revolution.

It offers absolutely new insight into the behaviour of non-linear chaotic systems which can be found in real economies and real liquid markets.

Times are different than they used to be. I am afraid that today even Einstein would have problems with gaining fame and a good job. What he would say if the interviewer demanded a home made atomic bomb as a token of his genius, not the equation.

In general, I agree with Peter Griepink, that science like art is evolutionary. We can talk about some kind of revolution only if someone is able to see beyond a single discipline, to see more and to use more other tools and still be able to keep the broadest possible picture. Specifically, the term complexity I relate to two phenomena: multidimensionality (only topology can handle it) and the impossibility of modelling with too many variables.

The theory of chaos has found a way out of it. This is new mathematics.

Applied to real time markets, it works. When it shows, only then you can realise that there is another type of complexity "inside" the system and the play of determinism versus uncertainty.

What makes this art really astounding is the pip-to-pip precision of fractal expansions, the working of Feigenbaum constants, the importance of some other constants derived during research, the never fading inner memory, etc. etc.

@Francisco Fatas-Villafranca, Isabel Almudi, Peter Griepink, Paul Koszarny

This is an eruption of all interesting arguements and commememts. Please wait for two weeks or so. I am now fully occupied by works which are in competition with time.