Complexity and the Economy
Page 34
REFERENCES AND NOTES
1. P. Anderson, K. J. Arrow, D. Pines, Eds., The Economy as an Evolving Complex System (Addison-Wesley, Reading, MA, 1988).
2. W. B. Arthur, S. N. Durlauf, D. A. Lane, Eds., The Economy as an Evolving Complex System II (Addison-Wesley, Reading, MA, 1997).
3. W. B. Arthur, Sci. Am. 262, 92 (1990).
4. W. B. Arthur, Increasing Returns and Path Dependence in the Economy (Univ. of Michigan Press, Ann Arbor, Ml, 1994).
5. I have avoided exact definitions of “complexity” and “complex systems.”
Technically, the systems I have described are referred to as adaptive nonlinear networks (J. H. Holland’s term), and typically if they exhibit certain properties that have to do with the multiplicity of potential patterns or with the coher-ence or propagation of substructures, they are said to be “complex.” Definitions vary widely.
6. A. Marshall, Principles of Economics (Macmillan, London, ed. 8, 1920), p. 459.
7. E. Helpman and P. R. Krugman, Market Structure and Foreign Trade (MIT Press, Cambridge, MA, 1985).
8. W. B. Arthur, Econ. J. 99, 116 (1989).
9. W. B. Arthur, Math. Soc. Sci. 19, 235 (1990); P. R. Krugman, J. Polit. Econ. 99, 483 (1991); in ( 2), p. 239; Geography and Trade (MIT Press, Cambridge, MA, 1991).
10. S. N. Durlauf, in ( 2), p. 81; J. Econ. Growth 1, 75 (1996).
11. W. B. Arthur, Am. Econ. Rev. 84, 406 (1994).
comPlexi t y and t He economy [ 187 ]
12. J. Casti, Complexity 1 (no. 5), 7 (1995/96); N. Johnson et al., Physica A 256, 230
(1998); D. Challet and Y.-C. Zhang, ibid. 246, 407 (1997); ibid. 256, 514 (1998).
13. R. E. Lucas, Econometrica 46, 1429 (1978).
14. W. B. Arthur, J. H. Holland, B. LeBaron, R. Palmer, Paul Tayler, in ( 2), p. 15; W.
B. Arthur, Complexity 1 (no. 1), 20 (1995).
15. See K. Lindgren’s classic paper in Artificial Life II, C. G. Langton, C. Taylor, J. D. Farmer, S. Rasmussen, Eds. (Addison-Wesley, Reading, MA, 1991),
p. 295; H. P. Young, Econometrica 61, 57 (1993); L. E. Blume, in (2), p. 425; B.
A. Huberman and N. S. Glance, Proc. Natl. Acad. Sci. U.S.A. 90, 7716 (1993).
16. R. Marimon, E. McGrattan, T. J. Sargent, J. Econ. Dyn. Control 14, 329 (1990); M. Shubik, in ( 2), p. 263; W. A. Brock and P. de Lima, in Handbook of Statistics 12: Finance, G. S. Maddala, H. Rao, H. Vinod, Eds. (North-Holland, Amsterdam, 1995).
17. T. J. Sargent, Bounded Rationality in Macroeconomics (Clarendon Press, Oxford, 1993); D. A. Lane and R. Maxfield, in ( 2), p. 169; V. M. Darley and S. A.
Kauffman, in ( 2), p. 45.
18. D. C. North, in ( 2), p. 223.
19. Y. M. loannides, in ( 2), p. 129; A. Kirman, in ( 2), p. 491; L. Tesfatsion, in ( 2), p. 533.
20. P. Bak, K. Chen, J. Scheinkman, M. Woodford, Ric. Econ. 47, 3 (1993); A. Leijonhufvud, in ( 2), p. 321.
21. R. Axelrod, Am. Pol. Sci. Rev. 80, 1095 (1986); K. Kollman, J. H. Miller, S. E.
Page, in ( 2), p. 461.
22. I thank J. Casti and S. Durlauf for comments on this article.
[ 188 ] Complexity and the Economy
AN HISTORICAL FOOTNOTE
The table below is an entry from my research notebook from November 5,
1979. I include it because it has been much cited in the last few years, and
because the thoughts here were the basis from which the papers in this volume arose.
In 1979 I was working at the International Institute for Applied Systems
Analysis in Austria. I had been reading a lot of biology, especially the work
of Jacques Monod and Francois Jacob, and much of the work of the Brussels
and Stuttgart groups on self-organizing systems. Given these ideas, and my
own predilections, it gradually became clear to me that economics would be
different in the future—it would be based on different principles. I wrote
what I thought these principles would be in my notebook. The table includes
thoughts on demography, which was one of my chief interests at that time,
and is reproduced in whole (complete with haphazard punctuation).
ECONOMICS OLD AND NEW
Old
New
Decreasing returns
Much use of increasing returns
Based on marginality (neoclassical)
Other principles possible (e.g. accounting principles)
Maximizing principles (profit motive)
Order principles
Preferences given
Formation of preferences becomes central
Individuals selfish
individuals not necessarily selfish
Society as backdrop
Institutions come to the fore as a main decider of
possibilities, order and structure
(continued)
Old
New
Technology given or selected on
Technology initially fluid, then tends to set
economic basis
Essentially deterministic, forecastible
Non-deterministic. Unforecastible because of
fluctuations and strange attractors (and suchlike)
Based on 19th-century physics
Based on biology (structure, pattern,
(equilibrium, stability, deterministic
self-organization, life cycle)
dynamics)
Time. Treated not at all (Debreu)
Time. Becomes central
Treated superficially (growth theory)
Closely tied to age
Age. Very little done
Individual life comes to center. Age begets time.
Emphasis on quantities, prices and
Emphasis on structure, pattern, and function (of
equilibrium
location, technology, institutions, and possibilities)
Elements are quantities and prices
Elements are patterns and possibilities. Compatible
structures carry out some functions in each society
(cf. anthropology)
Language: 19th-century math and game Language more qualitative. Game theory recognized theory, and fixed point topology
for its qualitative uses. Other qualitative
mathematics useful
Generations not really seen
Generational turnover becomes central. Membership
in economy changing and age-structure of
population changing. Generations “carry” their
experiences
Heavy use of indices
Focus on individual life. People separate and
People identical
different. Continual switching between aggregate
and the individual. Welfare indices different and
used as rough measure. Individual lifetime seen as
measure
No real dynamics in sense that
Economy is always on the edge of time. It rushes
everything is at equilibrium. Cf. Ball
forward, structures constantly coalescing,
on string in circular motion. No real
decomposing, changing. All this due to
change happening: just dynamic
externalities, increasing returns, transactions
suspension
costs, structural exclusion, leading to jerky motion
[ 190 ] An Historical Footnote
Old
New
If only there were no externalities and
Externalities and differences become driving force.
all had equal abilities we’d reach
No Nirvana. System constantly unfolding.
Nirvana
Most questions unanswerable. Unified
Questions remain hard to answer. But assumptions
system incompatible
clearly spelled out.
“Hypotheses testable” (Samuelson)
Models are fitted to data (as in [exploratory data
Assumes laws exist
analysis]) A fit is a fit is a fit. No laws really
possible. Laws change.
Sees subject as structurally simple
Sees subject as inherently complex
Economics as soft physics
Economics as high complexity science.
Exchange and resources drive economy
Externalities, differences, ordering principles,
compatibility, mind-set, family, possible lifecycle
and increasing returns drive institutions, society
and economy.
an Hi stor ic al f ootnot e [ 191 ]
[ 192 ] An Historical Footnote
OTHER PAPERS ON COMPLEXITY AND THE
ECONOMY BY W. BRIAN ARTHUR
The following papers, not included in this volume, may be of interest to the
reader.1
“Complexity, the Santa Fe Approach, and Nonequilibrium Economics,” in
History of Economic Ideas, 8, 2, 149–166, 2010.
“The Structure of Invention,” Research Policy, 36, 2, March 2007.
“Agent-Based Modeling and Out-Of-Equilibrium Economics,” in Handbook of
Computational Economics, Vol. 2, K. Judd and L. Tesfatsion, eds, Elsevier/
North-Holland, 2006.
“Time Series Properties of an Artificial Stock Market,” with B. LeBaron and
R. Palmer, Journal of Economic Dynamics and Control, 23, 1487–1516, 1999.
“Beyond Rational Expectations: Indeterminacy in Economic and Financial
Markets,” in Frontiers of the New Institutional Economics, 291–303, J. N.
Drobak and J. V. Nye (eds.), Academic Press, San Diego, Ca, 1997.
“Artificial Economic Life: A Simple Model of a Stockmarket,” with R. Palmer,
J. Holland, B. LeBaron, and P. Tayler, Physica D, 75, 264–274, 1994.
“Economic Agents That Behave like Human Agents,” Journal of Evolutionary
Economics, 3, 1–22, 1993.
“Why Do Things Become More Complex?” Essay in Scientific American, May 1993.
“Learning and Adaptation in the Economy,” Santa Fe Institute Paper 92-07-
038, 1992.
“Designing Economic Agents That Act like Human Agents: A Behavioral
Approach to Bounded Rationality,” American Economic Review (A.E.A.
Papers and Proc.) 81, 353–359, 1991.
“The Economy as a Complex System,” in Complex Systems, D. Stein (ed.), Wiley, New York, 1989.
1. So may some of my papers on positive feedback (often seen as part of complexity).
These appeared in the earlier collection, Increasing Returns and Path Dependence in the Economy, W. B. Arthur, University of Michigan Press, 1994.
INDEX
academic theorists, compared to market
hypothetical patterns, frameworks
traders, 40
and associations, 164
active predictors, 35, 185
imposing meaning on problem
active repertoire, 126, 127f
situations, 165
adaptation, 92, 154
inhabiting a world they must
adaptive complex system, 37
cognitively interpret, 93
adaptive nonlinear networks, 92–93
inside information, 110
adjacent probable, 128
interacting through impersonal
adopters, 71n4, 80–82
markets, 94
adoption, 71–72, 72t
introducing heterogeneity, 43
adoption process, 75, 80
keeping strategies in mind, 115
agent-based models, xv, 24, 39, 95, 99,
learning from each other, 98
102, 112–113, 115
learning which hypotheses work, 33,
agent homogeneity, 53, 101
49, 165
agents, 2–3. See also economic agents
making sense of their problems, 93
acting as market statisticians, 96
not knowing, 5
adapting forecasts, 42
optimizing allocation, 47
affecting other players’ well-being, 110
recognizing different states of the
arriving inductively at a
market, 49
homogeneity, 53
remembering what happened
basing actions on expectations, 95
before, 50
choosing between technologies, 70
showing uncertainty in choice, 16
entertaining more than one market
subjective beliefs, 32
hypothesis, 60
taking over part of the machinery of a
evolutionary selection of via
system, 108
wealth, 66
“tested” for survival, 7
expectations, 42–43, 49, 78
testing forecasting parameters, 52
exploring their way forward, 7
agent space, emergence of structure
facing not a problem but a
in, 99
situation, 159
aggregate patterns, 3, 99
finding out what works, 115
“AI-complete” problems, 116
forecasting outcomes, 178
aircraft designers, knowing causes of
gaming the criteria, 108
failure, 110
generating contingent actions or
air transport, problem of worldwide
rules, 114
spread of infections, 142
algorithm
asymmetric information, 107
arriving at complicated circuits, 125
atomic power, problem of disposal of
behind the computation, 8
waste, 142
building a library of functionalities,
attractor, in the bar problem, 37
124
autocorrelated volatility, 57n14
comment on, 131
autopoiesis, 20, 120n2
for the formation of the economy, 19,
avalanches
137
of change, 15, 59, 122, 181, 186
working best where needs are ordered,
of collapse, 130, 130f
129
Axelrod, Robert, 111
algorithmic model, rewriting in equation
Axtell, Robert, xxi, xxiii, 2n3
form, 11n15
algorithmic parse trees, 150
Bailey, James, xxiv, 11n16
alife approach, 102
Bak, Per, xvi
Allen, Peter, xxi, xxiii
Baker, W., 99
allocation
bar problem. See El Farol bar problem
within the economy, 22
Barrett, Chris, xxiii
mathematical analysis of, 23
base of a system, collapse near, 151
problems, 22, 70
behavior. See also exploitive behavior
process, 75, 84
adjusting to appear virtuous, 108
in the three regimes, 73–75
assuming to make models realistic,
alternating current, 81
6n8
alternative expectations, 53
depending on exogenous
alternative explorations, 54
characteristics, 100
alternative technologies, 78
economic, 171
Anderson, Phil, 89, 101
human, 173
anthropomorphic market, 176
inductive, 30–38, 164, 185
anticipation, by adaptive nonlinear
intelligent, 6, 44–46, 64
networks, 92
behavioral economics, ix, 6
anticipative models, 93
behavioral noise-trader literature, 41
arbitrage pricing model,
43–46
Beinhocker, Eric, xxiii, 2n2, 8n11
Arrow, Kenneth J., x–xii, xv, xxiii, 30,
belief-models, each agent tracking, 33
89, 107
beliefs. See subjective beliefs
Arthur, W. Brian, 79, 90, 193
Berra, Yogi, 35n3
Arthur-Ermoliev-Kaniovski theorem,
binary decision diagrams (BDDs), 132
80, 86
biological organisms, built from
artificial agents, 179
modules, 155
artificial evolution contexts, 147
biologists, doubts about linkage between
artificial intelligence, 116
evolution and complexity, 145
artificial markets, 66
biology
artificial stock market, 179–181
intermediate structures in, 122
artificial traders, 180
as theoretical but not mathematical,
assemblies, of technologies, 18
21
asset composition, 47
black box, getting from problem to
asset markets
solution, 159
models of, 47
Blume, Larry, xxiii, 61, 98, 101
reflexive nature of, 62
Boldrin, Michele, 90
asset pricing, 11, 42, 96, 176
bounded rationality, 6n9, 31
associations, cognitive, 160, 163, 168
Braudel, Fernand, 101
associative engines, brains as, 162
Bray models, 61
[ 196 ] Index
breakdown, in engineering designs, 110
as black box of economics, 158–169
British railways, narrow gauge, 81n9
implications of, 169
Brock, William, 90, 96
between the problem and the
Bronk, Richard, xx, xxiii, 6
action, 160
Bronowski, Jacob, 173
cognitive foundations, 93–94
Brownian motion, 6, 7, 16–17
cognitive process, modeling, 164–166
bubbles, 13, 40, 43, 54, 177. See also
cognitive reasoning, mirroring actual, 49
price bubbles
cognitive science, 6, 160
building blocks, 120, 126
Colander, David, xxii, xxiii, 24, 167
build-out of technology, modeling, 121
collapses, near the base of the
bull-market uptrend, prolonged, 60
dependency hierarchy, 147
Buss, Leo, 98
collective technology, 18, 20
combination, as the key driving force of
California, freeing of its energy
formation, 18n25
market, 104
combinatorial chemistry, 124
Cambrian explosion, 148, 152
combinatorial evolution, 119
Campbell’s law, 108
combinatorics, for an 8-bit adder, 125