Economics as a Complex System

1
Economics as a Complex System

• Alessandro Cappellini
• November 19th 2004
• cappellini_at_econ.unito.it

2

• In fact, Adam Smiths invisible hand (1776) is
  a classical example of unexpected features that
  complex systems exhibit whenever their
  participants reach a critical mass and
  self-organize, in this case into a centralized
  competitive market.
• (Lewis L. Smith 2002)

3
Agenda

• Economics as a Complex System (Network)
• Complex System (Lewis L. Smith)
• Karl Popper
• Mainstream and Complex System approach
• Product Function - Keynes
• Firm as Complex System
• Macro Meso Micro
• Firms as Complex Systems 1/2
• Firms as Complex Systems 2/2
• Alfred Marshall
• Remarks

4
Economics as a Complex System (Network)

• Quantitative associations between monetary
  associations and variables.
• Learning and experimentation forge new connection
  between network elements.
• Joseph Schumpeter innovation and growth are to
  be paralleled to the extension of new credit.
• If everything is fully connected to everything
  else (perfect knowledge) then there is no real
  distinction between elements and systems.
• Complex system are not fully connect but
  partially, and have actual potential connection
  with other system.

5
Complex System (Lewis L. Smith)

• Properties
• A critical mass of varied participants with some
  characteristics in common (heterogeneous
  agents).
• Access of this participants to some local
  information.
• A set of rules for their interaction.
• Exhibit surprising emergent properties.
• Ability of self-organization depend on positive
  feedback.
• Spend more time in dis-equilibrium than in
  equilibrium.
• In economics and markets instability and
  disequilibrium are normal, stability and
  equilibrium are not.

6
Karl Popper

• Warning on historicism (trying to predict the
  course of history in context that involve open
  system and evolutionary change).
• Modern economists say that
• Theory and history should be separated
• But
• Historical trends can be separated by theoretical
  propositions, concerning logical tendencies
  towards never existing long run equilibrium state.

7
Mainstream and Complex System approach

• from individual (psycological system) to the
  whole economy (macroeconomic system).
• from logical proposition to represent tendencies
  observed to underlying principles that govern
  behaviour of system.
• Value of economics
• Veblen (1898) to explain and to predict in
  historical contexts.
• Marshall to address historical events over short
  period.

8
Product Function - Keynes

• Production function is a trivial association of
  input and output.
• Data and process are correlated but separated in
  quantitative and qualitative domains.
• Data represent only a manifestations of
  historical process.
• Keynes theory was general because it could
  encompass the historical experience of systems as
  reflected in quantitative data.
• Aspects of process that investigate
  macro-instability.
• There isnt supply/demand dichotomy at the
  aggregate level, but only a network of
  connections for expenditures and income for good
  and services.

9
Firm as Complex System

• Firm is a governance structure (Williamson).
• In the case of the firm, full information (full
  connectedness) allows economists to treat the
  firm as a single decision-making entity that can
  optimize in a way that permits a supply curve to
  be derived from its cost structure.
• This kind of economics as about over-connected,
  simplistic systems that, by definition, are not
  representations of reality.
• Network structure flow and asset values.
• Connections that are definably economic are
  those that have a monetary value attached.

10
Macro Meso Micro

• One of the limitations of Keyness analysis is
  that the macroeconomic level of inquiry is
  inappropriate to capture the complexity of
  processes in an economic system what we observe
  are the stock and flow of funds consequences of
  complex interactions. The elements and
  connections that make up the network structure of
  the economy and how this changes are, rendered
  invisible by aggregations of value flows and
  asset valuations. At the same time, a wholly
  microeconomic perspective that tries to build up
  from individual elements and connections is not
  very helpful either because so much behaviour at
  that level constitutes component connections
  between elements embedded in higher level network
  structures. Meso-level analysis, that
  identifies the generic rules that govern the
  behaviour of complex systems, is most appropriate
  approach. There are many examples of complex
  network structures in the economy that are
  governed by generic rules that are meso in
  character. The one that we wish to focus on here
  is the firm.

11
Firms as Complex Systems 1/2

• Firms are complex adaptive systems with connected
  network structures that evolve over time not
  occur across the total space of connective
  possibilities but evolves sequentially from a
  pre-existing network base then unique connective
  structures emerge.
• The profitability of a firm ultimately depends
  upon making the right internal and external
  network connections and knowing how to break
  connections that no longer perform a useful
  function without, at the same time, destroying
  connections that still deliver value.

12
Firms as Complex Systems 2/2

• Thus, a supposedly suboptimal configuration of
  inputs and outputs may allow the firm to survive
  simply because the neglect of optimizing
  calculations and adjustments allows the firm to
  concentrate on, for example, learning by doing
  and experimentation in the product development
  and marketing activities that are crucial to
  generate revenue flows.
• Not directly related to production but to other
  processes going on as time passes, the
  environment changes, requiring new products and
  processes, but much of the network structure of
  the firm is committed to earlier strategies.
  Thus, this relates to the emergence of external
  disconnections as time passes.

13
Alfred Marshall

• All inputs can be varied in the long period, he
  recognized that the complex nature of the firm
  and its environment diminished the usefulness of
  constrained optimization theory.
• Partial constrained optimization should be used
  implies that network connections are restricted
  and incomplete.

14
Remarks

• It is possible to obtain simple representations
  of their growth and decline by theorizing about
  the flows of funds that parallel the processes
  that such systems enact.
• The creation, growth and destruction of firms are
  emergent processes that can be understood
  analytically and empirically investigated. How we
  construct emergent models must be guided by
  multi-agent simulations that are subject to the
  operation of stylized rules that are calibrated
  on actual rules we observe in the real world.

15

• Does our market converge to the rational
  expectations equilibrium of the academic theory
  or does it show some other behavior? What we
  found to our surprise was that two different
  regimes emerged. One, which we called the
  rational expectations regime, held sway when we
  started our agents off with sets of predictive
  hypotheses close to rational expectations.
  But there was a second regime, which we called
  the complex regime, and it prevailed in a much
  wider set of circumstances. We found that if we
  started our agents with hypotheses a little
  removed from rational expectations, or
  alternatively, if we allowed them to come up with
  hypotheses at a slightly faster rate then before,
  the behavior of the market changed. Subsets of
  mutually reinforcing predictions emerged .
• (W. Brian Arthur, The End of Certainty in
  Economics Talk given at the Conference Einstein
  Meets Magritte, Free University of Brussels, July
  1994. )

16
Bibliography

• Based on
• John Foster Why is Economics not a Complex
  Systems Science? presented at International J.A.
  Schumpeter Society Conference University of
  Bocconi, Milan, 9-12 June 2004.
• W. Brian Arthur, The End of Certainty in
  Economics, Talk given at the Conference Einstein
  Meets Magritte, Free University of Brussels, July
  1994. Appeared in Einstein Meets Magritte, D.
  Aerts, J. Broekaert, E. Mathijs, eds. 1999,
  Kluwer Academic Publishers, Holland. Reprinted in
  The Biology of Business, J.H. Clippinger, ed.,
  1999, Jossey-Bass Publishers.
• John Foster From Simplistic to Complex Systems in
  Economics, presented at Economics for the Future
  Conference Cambridge (UK) 17-19th September 2003
• Lewis L. Smith, Economies and markets as complex
  systems Looking at them this way may provide
  fresh insight, 2002