Evolutionary Complexity Economics

Transcription

1 Evolutionary Complexity Economics Lecture by Kurt Dopfer, University of St. Gallen Workshop at ETH, Zürich, May, 2015, on Social Norms, Mechanism Designs and Collective Intelligence, Class on Evolution, 27 May, 2015 Address: Sandrainstrasse 21, CH-9010 St.Gallen 1

2 All Sciences seek to observe reality systematically and to make meaningful generalizations about it and back! Economics is no exception! A telescope is an instrument that aids in the observation of remote objects. Mount Wilson Observatory near LA 2

3 Where is the Object of the economy? How can we observe it? A telescope allows us to observe remote objects by collecting electromagnetic radiation, such as visible light. There is no comparable object in economics. The object of the economy is a concept that derives from our imagination about an observed reality we associate with scarcity. E. g. : Economics is the study of individuals operating in the cultural context under conditions of scarcity. Different concepts will lead us to observe different kinds of objects of economic reality. Primitives and ontological axioms may provide us guidance in our search for concepts. 3

4 Primitives. are preconceptions that guide our ontological commitments. Different preconceptions lead to different views about the nature of reality. Primitive 1: One-ness and Many-ness Particles and Ensembles The Universe is not just a Clump Primitive 2: Matter is informed Why matter behaves in some way and not in another Concept of informant agency : Information is not only a Shannon probability measure but also - significantly - a semantic statement about the nature of the behavior of reality. 4

5 Methodological Primitives Physical Primitives Matter and energy can be measured on a quantitative scale Descriptive account on the basis of quantils Informational Primitives Semantic information qualifying physical properties requires Hermeneutics Explanation in terms of qualia 5

6 What are the Questions? Primitive questions for the sciences: 1 How do particles relate to each other and organize into wholes? 2 How do particles and ensemble behave in time and space?.and the primitive questions of economics: 1. How does co-ordination of economic activities come about? Smith s Invisible Hand vs. Walras Auctioneer 2. How does the economy behave in a stationary state/rest at a time, how does it change over time? Circular flow, economic evolution 6

7 Ontological Axioms Answers to the two perennial questions will differ owing to differences in ontological postures about the primitives. The paradigmatic-ontological foundations of a theoretical corpus may be stated in terms of axioms. We apply thus the concept of axioms not to mathematics but to ontology where axioms are statements about the Status of reality considered worth (axio) not to be questioned in their empirical validity. The distinction may be made between (in popular parlance) Mechanistic and Evolutionary Axioms Three axioms suffice for a complete ontological axiomatic. 7

8 Mechanistic Axiomatic Economics as Exemplar The Ontological-Paradigmatic Foundations of Neoclassical Economics (NE) Axiom 1: Particles are physical Particles composed of matter (energy) informed by a law - Nomological analysis. Economic particles are commodities informed by Homo Oeconomicus (HO). Axiom 2: Particles isolate The behavior of an economic particle - a commodity informed by HO - -occurs in isolation, no interaction. Economy is an aggregate of distributed commodities in state space (commodity space). Axiom 3: Particles behave uniformly Operations by HO in commodity space are response functions to exogenous shocks. HO restores equilibrium in state space deterministically. 8

9 9 Information as Law Information lies in the endogenous and exogenous domain of economic theory. Endogenously, information resides as universal law in HO: Decision logic applied perfectly by all agents. The law applies uniformly in all time: Invariance Samuelson s trajectory of maximization of expected utility under constraints x = f (t t o ; x* (t o )). HO is a representative agent. All agents carry perfect decision rule in the same way: Homogeneity Commodities are quantils, representative commodity, q x p, no characteristics.

10 10 Boundary Conditions of Neoclassical Economics Change in the commodity space is brought about by exogenous factors such as preferences, institutions and technology. They furnish side and initial conditions that cause a course along a trajectory: If condition x, then law HO holds yielding x. Paths towards and away from equilibrium: Away from equilibrium: Initial change in exogenous factors causes dis-equilibrium by way of additional opportunities. Towards equilibrium: HO reacts on basis of law equilibrium in commodity space is restored. No process, only logic involved. Cause is exogenous - effect is endogenous Laplacian Demon knows all past and all future

11 Configuring Economics Knowledge governs Operations To perform economic operations requires knowledge. Economic operations include production, consumption and exchange. Theoretical focus may be on: * Knowledge layer: Knowledge for Operations, * Operant layer: Operations based on Knowledge Approaches and schools in economics differ fundamentally owing to different emphasis and interpretation given to these theoretic layers. 11

12 Theoretical Vocabulary Knowledge is rule information actualized in a carrier Rule is a deductive format for operations Rule denotes a semantic content Carrier is physical locus of rule actualization Information about a rule is a potential that can be actualized by/in a carrier Information can be transmitted It is a process component Knowledge is a rule actualized by/in a carrier The carrier is said to entertain a knowledge or rule base or stock Information: Dis-embodied rule Note: Requires physical medium. e.g. paper Knowledge: Embodied rule Note: Actual entity, not medium is carrier. 12

13 Different Layers - Different Questions In NC knowledge is fixed (HO) or exogenous (ceteris paribus clause): Operant layer only. In contrast, Evolutionary Economics (EE) focuses on the knowledge layer. EE program addresses two research issues: 1. Mereology: Coordination of rules Emergent knowledge structure defined as rule complementarities 2. Causality: Evolution of rules Continual change of knowledge structure Statics and dynamics of operations and distribution in the commodity space are analyzed in relation to the structure and evolution of knowledge. 13

14 Unified Knowledge-based Theory The challenge is to combine the two generic questions into a single, integrated theoretical framework. F. de Saussure in his Course Linguistique (1916) has tackled this issue in linguistics, and his students have proposed this diagram: C A B D The horizontal axis (AB) represents the synchronic analysis, i. e. logic of the structure of language at a time and the vertical axis (CD) the diachronic analysis addressing its change over time. Big Question: What is the elementary unit that combines explanatory principles of both axes? 14

15 Generic Neologism for the New Canon The neologism generic denotes knowledge that meets the requirements stated. A generic rule is * Heterogeneous - Synchronic axis * Evolves - Diachronic axis. The neologism introduced is equivalent to the term nomological in the old canon. If generic is employed it stands for a fixed class, say for immutable genus in Linnaeus taxonomy. To date, one uses generic still this way when referring to generics, say generic drugs or washing powder. In an evolutionary perspective classes evolve. In modern biology, generic rules are biological rules, i. e. genes, in EE they are economic rules. Note: No general term exists in economics analogous to gene in biology; only special terms, e. g. Nelson-Winter routines. The term economic rules - or simply rules - is aimed at closing this gap. 15

16 Analogies require Ontology Generic properties of rules cannot be derived at the theoretical plane by way of simple analogy to biological rules, such as genes. The general properties common to both kinds of rules must be stated explicitly at the more general level of ontology. Analogies make sense only when the generalontological core - which allows us to infer from phenomena of one discipline to those of others - is made explicit. Example: If talk is on economic genes it must be justified why gene may serve as transdisciplinary term. 16

17 Evolutionary (Generic) Axioms Axioms underpinning analysis of complex evolving systems Axiom 1: Bimodality Any existence is composed of both Matter/Energy and Information Axiom 2: Association Particles have a Propensity to Associate Ensembles have Formative Power (e.g.. Order Parameter) Emergence of structured ensembles (Wholes) Axiom 3: Process Information actualized in structured Ensembles evolves over time Information is retained in meta-stable state for recurrent operations. These axioms provide a complete ontological underpinning of EE. 17

18 Economy as Generic Network An economic network defines connections among economic particles. Following axiom 1 bimodality - we get a Composed of two layers: Generic Network Deep Layer: Connections of Rules Semantic Pattern Surface Layer: Connections of Rule Carriers Physical Rule Pattern Methodologically, the deep layer consists of invisible variables. The surface layer of actualized rules consists of observables. Operations are performed on the physical layer of the generic network. Operant variables are observable and, together with generic surface variables they define the domain of quantitative analysis. 18

19 .. and where shall we go from here? Let us turn now to the particular kind of species that asks this kind of questions This species also constructs artificial agents.. but it is itself not artificial but natural equipped with subjective (subject-related) characteristics. 19

20 Homo Sapiens: A Rule Maker Economic agent belongs to evolved species of Homo Sapiens (HS). HS has the biological pre-disposition to make and use cultural rules. Implication: We cannot infer directly from biological invariants to behaviour. Operating under scarcity HS becomes Homo Sapiens Oeconomicus (HSO) Preferences, social behaviour, decision making, technologies, institutions, etc. all embody distinct cultural rules actualized on the basis of HSO s biological predisposition. HSO is a heterogeneous agent: This species carries different rules of a kind at a time and which evolve over time. 20

21 Products as Artefacts Monkeys make tools by way of sensory template, e. g. using a stone for nut cracking. HSO can use symbols based on cognition prior to perception of physical object. HSO produces objects based on Symbolic rules 60,000 years old ostrich eggs with engraved patterns, Diepkloof, South Africa. - Picture by courtesy of Pierre-Jean Texier In a cave in Blombos ornamented shells used as decorative objects dating back about 77,000 years. HSO creates esthetic surplus. Monkeys don t. 21

22 Rule Taxonomy: Rules relate to two kinds of carrier: Subject (HSO): Behavioral Rules (Internal = Cognitive) Object (Artefacts): Organizational Rules Social and Technical (Artefacts = Artificially Created Environment) Rules tell the semantic story of the evolution of complex economic systems. 22

23 Example: Traffic A traffic system is a composite set of rules for operations of locomotion. Operation: Driving Intentional locomotion Carriers of Rules: Driver: Vehicle: Cognitive Rules Behavioral Rules Technical Organizational Rules Combined with Social Organizational Rules Infrastructure: Social Organizational Rules Combined with Technical Organizational Rules Historic-Generic Context (contingency) of traffic system: * Metastable rule system for recurrent operations * Evolution of rule system 23

24 Rule Trajectory How do rules originate and how do they become effective in a system? Rules have a life cycle. Its generic dynamic can be stated in terms of a threephase trajectory: Phase 1: Origination of rule Phase 2: Selective Rule Adoption as Diffusion Phase 3: Retention of rule Macaques washing sweet potatoes, Koshima, Japan, photo by Frans de Waal, reproduced with permission 24

25 HSO s Rule Trajectory unpacked I Origination Sub-phase 1 Ideation: Creation of novel idea i.e. rule invention Sub-phase 2 Search, discovery and recognition process; internal selection, inside the firm II Adoption Sub-phase 1 First actualisation i.e. rule innovation; rule as seed for macroscopic adoption Sub-phase 2 External (Darwinian) selection, out there in market; Path dependence,; Complexity III Retention Sub-phase 1 Routines selectively retained for recurrent operations; meta-stability of actualisation process Sub-phase 2 Existing regime as breeding ground for novel rule(s); link to phase I 25

26 MESO: One and Many A Rule: A Population: Structure Component Process Component Meso: Physical actualization of a rule along a population trajectory as component of structure. Macro is composed of many meso. Meso is a rule and its population of micro. Micro is a member of a meso. Macro nests meso that nests micro Structure and process of the economic macro system are made of many meso components. 26

27 Macro - Deep Layer Structure of Meso Rule Complementarities A New Look at Structure: Mereology Upward and Downward Complementarity Upward: Adam Smith s Division of Labor Downward: Assembling Parts into a New Whole Modularity: N-Decomposability Plasticity c d a j n e c = Component = a,... e,..., k b a j 1 a 2 = c j i j = Upward complementarity Structural modularity i = Downward complementarity Process modularity a i 1 a a i 2 a i m 27

28 Uni-directionality of Evolution Histonomic approach: Time-asymmetry and double-contingency Basis for Decision Making in Economic Policy Past: Contextual Determinism; single exit (blue) Future: Contextual Autonomy; finality (white) X N max N min 28 x = Index for Economic Evolution/Development Contingency given Natural Constraints (N): N max = Ecological carrying capacity N min = Subsistence limit of human species; biological constraints Historical time

29 The Complex Evolving Economy Integrate the theoretical building blocks into an edifice! An economy is a multi-agent system based on a population of rules, a structure of rules, and a process of rules. X j (r 1 c ) Micro F(r c ) Meso r c Surface structure Macro Deep structure F(r a ) r a r b F(r b ) X j (r 3 a ) X j (r 2 b ) 29

30 THANK YOU FOR YOUR ATTENTION! 30

31 Selected Bibliography Dopfer, K., Potts, J. (2004), Evolutionary realism: a new ontology for economics, Journal of Economic Methodology, 11:2, Dopfer, K. (2004), The economic agent as rule maker and rule user: Homo Sapiens Oeconomicus, Journal of Evolutionary Economics, 14: Dopfer, K., Foster; J., Potts, J. (2004), Micro meso macro, Journal of Evolutionary Economics, 14: Dopfer, K. (2015 forthcoming), Evolutionary Economics, Handbook of the History of Economic Analysis, Volume II, Schools of Thought in Economics, editors Gilbert Faccarello and Heinz D. Kurz, Cheltenham: Edward Elgar. 31