General comments

Please leave comments here on groups of papers or the conference as a whole.

5 comment

  • Stephen I. Ternyik says:

    As we are inquiring into the economic philosophy of complexities, the late Cartoonist Rube Goldberg (1883-1970) came into my mind; he designed many complex machineries to solve simple tasks. In my opinion, our greatest challenge, concerning economic complexity, comes from exponential knowledge automation (e.g. in finance and accounting), without having reviewed our accounts of society, i.e. we are using simple accounting principles for increasing complex economic tasks.

  • mishail mokiy says:

    Dear colleagues. Many thanks to the conference organizers for what they have raised such an important topic.
    Economic reality is way different from economic theory. That is clear to economists. Today, it is obvious that none of the dominant paradigms of economic development serves the goals of mankind. No doubt it is related to the fact that economic relations are a complex structure.
    In the articles of the conference carried out a critical analysis of existing approaches to solving the problems of complexity of economic science and economic relations. Of course such works is necessary. This is stated in comments by colleagues.
    However, I want to quote «Essay on the Nature and Significance of Economic Science», Robbins L , 1932. « The efforts of economists during the last hundred and fifty years have resulted in the establishment of a body of generalisations whose substantial accuracy and importance are open to question only by the ignorant or the perverse. But they have achieved no unanimity concerning the ultimate nature of the common subject-matter of these generalisations………We all talk about the same things, but we have not yet agreed what it is we are talking about.». Не said that 85 years ago, but the situation has not changed.
    In his time Einstein wrote: “The wording of the problem is often more essential than its solution which may only be a matter of mathematical or experimental art. Raising new questions, developing new opportunities, considering old problems from a new angle require creative imagination and reflect the real success in science” [Einstein, A. (1936). Physics and reality. In Author, 1956, Out of My Later Years (pp. 59-95), Secaucus,NJ].
    In my opinion, adequate answers to the questions in the Theoretical Background Statement of the conference is possible only when solving the ontological problems of economic science. In other words, we need to go back to find the answer to the main question – what is “economic relations?” It is necessary to change the worldview of economists. Of course, that the formation of the worldview is the task of philosophy.
    This approach will allow first answer to the question “how and why there is a difficulty of the economy?” and only then define the problem of the interrelationship of Economics, sociology, politics, Finance and more.
    Failure to understand what we are studying does not allow us to apply suitable methods to the object of the study which creates an epistemological problem. In any case, it leads to a wrong setting of objectives, finding wrong solutions which hampers the achievement of goals.
    I am very glad that formed a community of scientists who see the connection between economic science and philosophy. Economic relations is not only about money and profit. They can’t be investigated using outdated models of supply and demand and other models of the mainstream.
    I really hope that the next conference will be to discuss the philosophical foundations of ontology economic relations
    I wish you academic success to all participants
    With respect and hope for further cooperation
    Professor Mokiy M. S.

  • Yoshinori Shiozawa says:

    As it is the last day of the forum, I would like to present my summery of comments on the forum.

    I have read all three papers cited in the references of “Theoretical Background Statement” as my preparation. I have read about ten papers that were made public in the forum and commented for seven of them. Probably the organizer of this forum will make a synthetic report on the result or achievement of this forum. My comment is a kind of general impression that I felt during reading above papers.

    As I have written many years ago (first in 1994, reproduced in 1997 in my book Consequences of Complexity [in Japanese]), we have to distinguish in economics three different levels or aspects of complexity:
    (1) complexity for the economic agents,
    (2) complexity of the economic system, and
    (3) complexity that conditions the development of economics as science.

    The first complexity is related to our human capabilities. H. A. Simon thematized this in the topic of bounded rationality. Simon included in this concept two meanings for our capabilities: limited capacity in information obtaining and limited capacity for rational computation and reasoning. It would be better to add the third capacity as our bounded capability i.e. limited capacity for actions. If we accept that humans are bounded in these three capabilities, we have to attack two problems:
    (A) How are we behaving with these bounded capabilities in a complex situation as economy?
    (B) What kind of scheme should we adopt as our analysis?

    In my impression, arguments related to this aspect of complexity were rare. This may reveal a defect of our discussion. A rare exception was Cedrini and Fontana’s mention on Ronald Heiner (1983)’s C-D gap theory in The Origin of Predictable Behavior. This kind of study is crucial for the complexity arguments but it was the most neglected topic in the forum. (See also my paper: The Microfoundations of Evolutionary Economics, in my Contribution page in ResearchGate)

    The second complexity is that of the economic system. Many papers argued this aspect but most of the discussions were comparisons between two analytical tools or frameworks like linear vs. non-linear, statics vs. dynamics, closed vs. open, and equilibrium vs. non-equilibrium/disequilibrium. Some authors argued as if the introduction of new ideas like emergence, chaos, attractors, herd-like behavior, and others solves the difficult problem we face. Those people who like to herald new ideas were optimistic in the future development of complexity economic, but I was rather disappointed because I know those new ideas help to elucidate only a special and very restricted point of the problem.

    The third complexity provides the reason why philosophical interventions were necessary for complexity questions in economics. But, in my impression, there were a few papers which suggest a deep reflection on this theme. Complexity of the economy exceeds in some aspects our capacity of theory making. The question we must pose is how to cope with this difficult situation.

    Agent-Based Simulation (ABS) is a new method which became available for us and opens us a new possibility. It became possible to incorporate behaviors of heterogeneous agents’. ABS made it possible to escape from equilibrium framework that was a straight-jacket of economics for a long time. However, ABS is not yet a ripe as a method of scientific research. (See my paper A Guided Tour of the Backside of Agent-Based Simulation) The success of ABS depends on what problems we pose and what we aim for the future development of the economics.

    There were ample discussion about mainstream and heterodox economics, but few hints were given on how the huge complex network as big as national and global economy works. This must be the core question of the complexity arguments in economics but few suggestions were made. Paradoxically, complexity arguments do not helps us to think this main question. In the comment that I cited in a comment on Greg Hill’s paper, I argued that many economists including heterodox ones are still a victim of rationality bias. They believe that economy works because human agents behave if not completely rational but sufficiently intelligent way. We are still captives of Arrow-Debreu type general equilibrium theory. Hill is trying to escape as far as possible but he sees no firm direction to orient himself.

    To escape from rationality bias, it is necessary to show how the huge network of exchange and production works with minimum of rationality. In the next post, I will show a rough description on it.

  • Yoshinori Shiozawa says:

    How economy works with minimum of rationality.

    To know how the economy works without relying too much on our rationality, we should abandon equilibrium framework. This is the main reason why many of economic theories must rely on rationality.

    Economy is a system of sequences. Analyses should be formulated sequentially. The second thing we should take care of is to distinguish price adjustment and quantity adjustment. General equilibrium theory is totally confused on this point. It thinks that prices and quantities are adjusted at the same time. No! Price adjustments and quantity adjustments are normally independent and they interact only at some special cases (for example, when the labor supply approaches its limit, production its capacity, etc.).

    (A) Price adjustment process
    Each firm announces their product prices and fixes them for certain lapse of time, e.g. a day, a week, or a year. Even in the case in which it re-chooses prices each day, the firm may not change them and the same prices may continue during many days. The same applies to weekly or annual adjustments. At any rate, firms fix their product prices for a while.

    How do firms fix their prices? The most preferred method is full-cost pricing. Firms have their own markup rates (m). We should ask how these markup rates are determined, but I do not enter in this detail. Firms can calculate production costs (c) of their products. The prices (p) are given by a simple formula

        p = (1+m) c.            (1)

    This is a very simple story. No extreme rationality is required.

    Next question is what happens in the economy as a whole. If all firms repeat full-cost pricing, this process approaches to a limit. Let us assume, for simplicity, that all firms re-fixe prices at the same time (simultaneous pricing). Then the price adjustment process is given by the following formula:

      p(t+1) = (1+m) {w a0 + A p(t)}.      (2)

    Formula (1) is concerned with particular price and cost. Formula (2) describes the change of the system of prices. Here, a0 denotes the labor input coefficients vector and A denotes the material input coefficients matrix. Markup rates are here supposed to be identical for all products, but it is easy to re-adjust this assumption by replacing m by a diagonal matrix M composed of different markup rates.

    Let (1+m) A be productive as a matrix, then it is easy to show that this process converges to a price vector p that is the solution of equation (2) when p(t+1) and p(t) are replaced by the same vector p. Note that the number of commodities (product goods and services) can be very extremely great. It may be of the order of ten to hundred million.

    In a real economy, the price adjustement is made sporadically by each firms. Even in that case, the price adjustment process converges to (2) when each firm repeats re-adjusting prices infinite times. See my old paper:

    Non-Simultaneous Mark-up Pricing Processes

    (B) Quantity adjustment process
    I have to cite results which are not published in English. I am now preparing English book with Kazuhisa Taniguchi and Masashi Morioka. It will be titled Microfoundations of evolutionary economics. The first chapter of the book is my paper with the same titel. This is a part of the present project:

    Four chapters for two books

    The microfoundations book will comprise two chapters of mine and three others of Taniguchi and Morioka. Taniguchi and Morioka each have written a book in Japanese, in 1997 and 2005 respectively. Unfortunately, these important results were not published in English. That is why we are now endeavoring to write a new English book.

    The idea is simple. Suppose a given system of prices including wage rates is given and suppose each firm can produce their product at a positive markup rate. This rate may not be satisfactory one for a firm. The firm may change the price of the product. However, it does not change the price everyday. We are more concerned what happens at the given prices. Even if the prices are fixed, the demand for products changes everyday (even every hours). How can a firm that has only a limited capability to predict the today’s or tomorrow’s demand? They may predict what will be the future demands. However, it is inevitable that the prediction is not always inexact. Thus, the firms are obliged to keep product stock as inventory. Both makers and sales shops do the same thing. Adjustment frequency depends much upon various conditions. However, let us assume that all firms adjust their production volume per day and pursue to keep their product stock at the plausible level.

    Of course, this level depends on the demand fluctuation and prediction. We can assume various complicated calculations to “optimize” the stock level. However, Taniguchi and Morioka discovered that firms can catch up the slow change of everyday demand flows when they use average of actual sales for more than 5 days.

    Please note that each firm uses various inputs. So the inventory adjustment affects the flows of other makers which provide the materials and parts. Taniguchi and Morioka found that the whole network of this economy wide adjustment process can follow the slow change of final (consumer and public) demand flows. If the demand remains constant, the discrepancies from the stable level of stocks diminish to half each about 8 period (8 times of adjustment, say 8 days). This means that the whole economic system can follow slow movements of demands without any previous predictions.

    On the other hand, if firms use predictive numbers by calculating tendencies, the system become unstable and almost all occasions the adjustment process diverges.

    (C) Functions of prices and choice of production techniques
    Now we can argue what the functions of prices are. Neoclassical economics assumes that economic adjustment relies totally on the prices. This is the reason why GE model has to assume perfect rationality both for consumers and producers. In reality, price mechanism works very differently from what neoclassical economists imagine. Prices are important parameter by which to determine which production technique is better than others. This is the problem of choice of production techniques.

    Production techniques are represented by their input coefficients. Thus each production technique u is expressed by a couple of a0(u) and a(u) where a0(u) is the labor input coefficient and a(u) = (a1(u), a2(u), … , aN(u)) is the vector of material input coefficients vector.

    A firm may possess many different production techniques to produce the same products. They may represent different processes with different machines. They may also represent improved methods of treating intermediate products. Work organization may change production techniques. However, if a set of wage and prices is given, the firm can choose a best production technique that gives the least unit cost of production.

    Even if each firm possesses many different production techniques for each product, the economy can choose best set of techniques. Economy has a mechanism that warrants this. The mechanism was first found by Paul Samuelson and named at first substitution theorem and renamed soon after to non-substitution theorem. I do not use this name, because I believe that it is better called minimal price theorem.

    Samuelson proved the minimal price theorem only for two-commodity case and it was Koopmans and Arrow who respectively proved the theorem for three-commodity case and for any number of commodities. Samuelson emphasized that this minimal price theorem holds only when production techniques are of simple production type (meaning that only one product can have a net positive output) and unique factor of production. This is true as a theorem. But to emphasize too much this fact has a special negative significance. If the minimal price theorem is valid, it is evident that the economy is the world that classical economists imagined. There is a set of natural prices and change of demand proportions does not affect their natural prices.

    Neoclassical economists emphasized that minimal price theorem is not valid when capital goods are used, because either (1) they are a factor of production or (2) they make production techniques joint production type. Morishima also emphasized in a similar way. However, capital goods are produced goods and are not primary factor of production. Using fixed capital machines are treated as byproducts but as Sraffa has shown in Pat Two of his book of 1960, we can treat durable capital goods just as variants of production techniques of simple production type. As a conclusion, the minimal price theory holds for wide enough situations. Thus it is not necessary that prices change when the demand composition changes. The economy can adjust itself only by changing production scales.

    For these arguments, see two of my papers:
    (a) The Revival of Classical Theory of Values
    published as a chapter in Yokokawa et al. (Eds.) The rejuvenation of political economy, Routledge.

    (b) Durable Capital Goods and Their Valuation

    (D) Total assessment
    In step (A), we have confirmed that price adjustment process using full-cost pricing converges. In step (B), the stock adjustment process also converges when firms use past averages for a plausibly number of periods. As a result, we can say that economy (be it nation wide or world wide) is stable in the sense that ordinary process of price and quantity adjustments converges or follows the demand flows.

    In a real economy, there are all kinds of complications. Some firms cannot change their production volumes everyday. For example, in the case of high furnace, it is difficult to change the production level of iron frequently. It is necessary to keep the production level at much more constant level. Even in such firms, they change the level of production volume per period using all kind of techniques. For example, the high furnace production can adjusted within certain range by using oil injection technique.

    When there is a big shock either in production (e.g. production stop by an accident) or in demand (e.g. sudden financial collapse and rapid decrease of consumption as a result), the shock will propagate from one sector to whole economy. This propagated shock decreases if the particular original shock stops.

    We have assumed above that input coefficients are constant. This assumption is unnecessary. If there is technical evolution by productivity improvements, the firm may change its product price and demand flow may change as a result. Processes (A) and (B) can well follow this change. In fact, in a real economy, there is constant change of input coefficients. Firms cannot distinguish this or that change of prices or demand comes from which reasons. However, there is no need to know such things. These changes are included among many other random fluctuations but the firms can follow those changes without knowing. Thus the whole picture that processes (A) and (B) provide is quite dynamic and flexible.

    Thus we get a very plausible picture how market economy works and re-adjust itself. The conclusions we can derive from this are three:

    (1) There is no need of high level of rationality for the economic agents.

    (2) Free market economy is provided with a self adjustment mechanism which is very different for that supposed by general equilibrium theory.

    (3) The adjustment process can well adapt its-self to various random fluctuations. It reveals the dynamic and flexible character of the market economy.

    We may sum up that this adjustment process parable is much more realistic than Arrow-Debreu general equilibrium model and teach us how the market economy really works.

  • Ping Chen, Fudan Univeristy says:

    The deadline is past. I may leave some feedback to the organizer.
    I should first apologize that my participation is too late for family reasons. The progress made by this forum was beyond my expectations.
    A few comments for future development.
    (1) The academic level of a forum mainly was determined by keynote speakers. If the leading discussant could start with a systematic review of complexity science or brief introduction of main issues and competing schools of thoughts in complexity economics, later discussion would be much easier to focus on fundamental progress or major unsolved issues. If the organizer provided a list of major literature in this field could help new comers and graduate students.
    (2) Contributed papers could organize into several groups and organize some dialogue among competing authors. For example, there are several authors working on CAS/ABS approach, if they could exchange ideas, we may see better progress. Some authors started from empirical problem, some started from theoretical/philosophical problem, they may also have a fruitful dialogue.
    (3) The organizer/editor may play a more active role during the process. For example, some author raised very FUNDAMENTAL questions, the organizer could invite relevant EXPERT to answer his/her question. I do not think current level of this forum reflect the state of art progress in complexity economics.

    Sorry for being late in general discussion.
    Ping Chen, 19:25, Dec.1, 2017 (US Central Time)