Complexity and the Economy

by W Brian Arthur

  exogenous—they were constructed by people with agendas from other times

  sometimes suited to the purposes of other times. We need knowledge of the

  history of economic thought to be fully aware of the associations we make in

  economics and their provenance.

  So what in graduate economics education do we really want? We certainly

  need theory. As a theorist I’m all for theory. But we also need the rich pictures

  [ 168 ] Complexity and the Economy

  given by the study of history and institutions. We need both types of association: the theoretical, quantitative, precise frameworks and the case-based, vivid pictures in their tens of thousands. To teach only theory is equivalent

  to training doctors by teaching only endocrinology and pathology, and not

  the wide diagnostics doctors learn on grand rounds. To operate only with the-

  ory—think of driving a car—makes us beginners. It’s not until we can seam-

  lessly integrate theory and vivid pictures—theory and experience—that we

  become expert. I believe we are currently turning out students who lack those

  pictures. And in doing so, we’re doing them a disservice.

  DO ISSUES OF COGNITION MATTER?

  Perhaps in asking my fellow economists to think about the implications of

  cognition, I am asking for something useful but not necessary—a luxury?

  I don’t believe so. Consider just one example. The Soviet Union in 1990–91

  decided to go capitalist. And from us economists it got much advice. But our

  natural bias, given the current development of economics, was to concentrate

  upon a worthy, but imagined, general-equilibrium outcome where institutions

  would be in place and markets would work smoothly and incentives would be

  correct.

  A cognitive view of economics might have balanced this ideal view with an

  awareness that Russians were not arriving with empty minds to their version

  of capitalism. Not only did they possess old structures both economic and

  political, they harbored from their 70 years of communism and earlier czarist

  past old associations too—of what business means, of how one interacts with

  authorities, of how one organizes if one wants to make money, of what one

  does with economic power and wealth. More enlightened advice would have

  built upon an understanding of how these embedded structures and under-

  standings would play out given the new possibilities. The subsequent history

  of Russia’s experiment with capitalism showed that these matters of cogni-

  tion had great importance.

  Economic agents bring to their actions not just their preferences and

  endowments, but also their understandings—the associations and meanings

  they have derived from their history of previous actions and experiences. In

  many of the small, standard problems of economics, we can ignore this. In the

  larger issues of development and reconstruction, and in constructing an eco-

  nomics for problems of complication and ill-definition, we cannot. We need to

  take cognition seriously.

  cogni t ion [ 169 ]

  REFERENCES

  Arthur, W. B. (1994), “Complexity in Economic Theory: Inductive Reasoning and

  Bounded Rationality,” American Economic Review, 84(2), 406–411.

  Arthur, W. B., J. H. Holland, B. LeBaron, R. Palmer, and P. Tayler (1997), “Asset Pricing under Endogenous Expectations in an Artificial Stock Market,” in

  W. B. Arthur, S. Durlauf, and D. Lane (eds), The Economy as an Evolving Complex System II, Reading, MA. Addison-Wesley.

  Clark, A. (1993), Associative Engines, Cambridge, MA: MIT Press.

  Jaynes, J. (1976), The Origin of Consciousness in the Breakdown of the Bicameral Mind, Boston: Houghton Mifflin Company.

  Laxness, H. (1935), Independent People, New York: Vintage.

  Rust, J. (1987), “Optimal Replacement of GMC Bus Engines: An Empirical Model of Harold Zurcher,” Econometrica, 55(5), 999–1033.

  Sargent, T. J. (1993), Bounded Rationality in Macroeconomics, New York: Oxford University Press.

  [ 170 ] Complexity and the Economy

  CHAPTER 11

  The End of Certainty in Economics

  W. BRIAN ARTHUR

  In the mid-1980s deterministic and rationalistic thinking dominated theoretical economics, yet many economists, including myself, felt profoundly uneasy with it. This essay is a reaction to this narrow form of high-rational thinking and it points out a fundamental indeterminacy in the economy. We act on our forecasts or expectations, and so our expectations create the world—the outcome—our expectations are trying to predict. Without knowledge of others’ expectations, or knowledge that everyone has knowledge of others’ expectations, an agent’s expectations become impossible to formulate logically and so do others’ expectations. Rational expectations thus become dubious and fragile assumptions, and the real world is subject to fundamental uncertainty. This indeterminacy means that economic behavior builds from subjective beliefs; these arise, co-evolve, change, mutually reinforce, mutually negate, and decay.

  I argue that economics should recognize and accept such subjective reasoning as its basic premise.

  This essay was given as a talk at the conference Einstein Meets Magritte, in Brussels, in 1999, and some of its examples repeat ones used elsewhere in this book. The essay appeared in the volume Einstein Meets Magritte, D. Aerts, J. Broekaert, and E. Mathijs (eds.), Kluwer Academic Publishers, Holland, 1999.

  The story of the sciences in the 20th century is one of a steady loss of cer-

  tainty. Much of what was real and machine-like and objective and determi-

  nate at the start of the century, by mid-century was a phantom, unpredictable, subjective, and indeterminate. What had defined science at the start of the century—its power to predict, its clear subject/object distinction—no longer

  defines it at the end. Science after science has lost its innocence. Science after science has grown up.

  What then of economics? Is economics a science? Well yes, I believe so. For sure it is a body of well-reasoned knowledge. Yet until the last few years it has maintained its certainty, it has escaped any loss of innocence. And so we must ask: is its object of study, the economy, inherently free of uncertainties and indeterminacies? Or is economics in the process of losing its innocence and

  thereby joining the other sciences of this century?

  I believe the latter. In fact, there are indications everywhere these days in

  economics that the discipline is losing its rigid sense of determinism, that the long dominance of positivist thinking is weakening, and that economics is

  opening itself to a less mechanistic, more organic approach. In this talk I want to show my own version of this loss of certainty. I want to argue that there are major pockets of uncertainty in the economy. I want to show that the clear

  subject/object distinction in the economics often blurs. I want to show that

  the economy is not a gigantic machine, but a construct of its agents. These are not “anomalies” to be feared, they are natural properties of the economy, and

  if we accept them, we will have a stronger, not a weaker science.

  Let me start from the beginning. The fundamental ideas in economics stem

  from the thinking of the 18th century, in particular from the thinking of the

  English and Scottish Enlightenment. In 1733, at the height of the intoxica-

  tion of enlightenment thinking, Alexander Pope condensed its essence in one

  stanza of his Essay on Man:

  All Nature is but Art unknown to Thee

  All Chance, Direction, which thou canst not see

  All Discord, Harmony, not understood
>
  All partial Evil, universal Good:

  And, spite of Pride, in erring Reason’s spite

  One truth is clear, “Whatever is, is right. ”

  In this context “Art” means artifice. It means technique or mechanism. And

  so, all the intricate wonders we see in nature, says Pope, are in fact a gigantic machine, an artifice like the mechanical automata figures of his time. All that looks unkiltered really has direction behind it. All that looks complex and dis-cordant, like the movements of planets before Kepler’s and Newton’s times,

  has a hidden simplicity. All that affects each of God’s creations adversely,

  in some unspoken way works to the good of the whole. Quoting Socrates,

  “Whatever is, is right.”

  These were not merely the ideas of Pope. They were the ideas that filled the

  intellectual air when Adam Smith was growing up. Smith went on to enshrine

  them in The Wealth of Nations, that magnificent work that uncovered the hidden simplicity behind the traffickings of traders and manufactories and

  butchers and bakers. The economy was indeed Art, and its principles were now

  unhidden. The selfish interests of the individual were guided as by an invisible

  [ 172 ] Complexity and the Economy

  hand to the common interest of all. Whatever was, was right. Two centuries later, the philosopher of science, Jacob Bronowski, was to comment glumly

  that economics never recovered from the fatally rational structure imposed on

  it in the 18th century. But we inherited more than Smith’s rational structure.

  Deep in some recess of our minds, we inherited the thinking that the econ-

  omy is but Art, a gigantic machine, that if we merely understood its parts, we could predict the whole. Certainly when I was studying economics in Berkeley

  25 years ago, many economists hoped (as I did) that a Grand Unified Theory

  of economics was possible. From the axioms of rational human behavior, a

  theory of the consumer could be constructed. From this and a correspond-

  ing theory of the firm a consistent microeconomics could be constructed.

  From this, somehow, an aggregate theory of the economy, macroeconomics,

  could be constructed. All this would constitute a Grand Unified Theory of the

  economy.

  There have always been two embarrassments to this hope of constructing

  a theory of the economy from its reductionist parts. One is that the economy

  relies on human beings, not on orderly machine components. Human beings

  with all their caprices and emotions and foibles. The second embarrassment

  is technology. Technology destroys the neatness because it keeps the econ-

  omy changing. Human behavior was finessed in economics by the device of

  Economic Man, that perfectly rational being who reasons perfectly deduc-

  tively on well-defined problems. Technology change was not so much finessed

  as ignored, or treated as exogenous. And so to make an orderly, predictive

  theory possible, Economic Man (the subject) needs to operate on well-defined

  Problems (the object). There should be no blurring of agent and problem. The

  well-defined Problems should have well-defined Solutions. And the solutions

  would comprise the building blocks for the next aggregated level of the theory.

  This approach works. But it runs into difficulties when problems start to

  involve more than one decision maker and any degree of complication. Then

  heroic assumptions must be made. Otherwise well-definedness unravels,

  agent and problem become blurred, and pockets of uncertainty start to bulge.

  Let me show you what I mean in the context of a typical microeconomic

  situation in modern economics. (I have chosen it from the late-1970s litera-

  ture on industrial organization.1) Consider this problem: We have a circle that we might think of as a 24-hour clock. A number of firms, say 20 airline companies, have to decide in which time slot of this clock their planes will take off in, for example from La Guardia Airport to go to Washington. Of course the

  different airlines have different preferences when to take off. They know their preferences and are going to book suitable takeoff slots. The choices will be

  1. The example is based on the approach used in “Sequential Location among Firms with Foresight,” E. C. Prescott and M, Visscher, Bell Journal of Economics, 8, 2, 378–393, 1977.

  t He end of certain t y in economics [ 173 ]

  made once and for all. There is a trade-off (in every decent economic problem there is always a trade-off) between where they really want to take off versus not being too close to other airlines’ choices of their time slots. So, given the airlines’ preferences, which time slots will they choose? This is the problem.

  We might feel uneasy about saying much with certainty here. But I want

  to show you the modern version of the Enlightenment approach, where we

  find the Harmony of a solution within the Discord of the situation. This High

  Modern approach is called rational expectations. I will first spell it out, then shine a bright light of realism on it, so that it starts to unravel and pockets of uncertainty appear. Let’s go ahead.

  In the rational expectations approach, we begin by supposing we know the

  order in which the airlines will submit their choices. Now imagine airline number 20 reasons like this: knowing where the first 19 airlines are going to be, I will know where I want to be. So regardless of any arbitrary choice of the first 19 airlines, I will know which time-slot to choose. This is an easy problem for me as the 20th. What about airline number 19? Well, airline number 19, when

  choosing, will know the arbitrarily chosen positions of the previous 18 airlines and can figure what it should do, given that the 20th will choose an optimal

  position given the positions of the 18 other airlines and 19’s choice. What

  about the 18th? Well, the 18th, knowing where the previous 17 will be, arbi-

  trarily can solve the problem of selecting an optimal placement knowing what

  the 19th will do, given that the 19th makes his optimal choice, given what the 20th will do as a result of 19’s choice. Getting complicated? Yes. But you can work the whole logic in reverse order by backward deduction, or more properly

  by dynamic programming, and deduce how all 20 firms will place themselves.

  Notice the properties of this procedure. The problem is well defined, by

  making it sequential and assuming the firms use logical backward deduc-

  tion. The solution is precise and clean in a mathematical sense. The problem

  becomes a mathematical one. (Indeed all such problems become mathemati-

  cal. Economics in turn becomes mathematics.) Another property that we nor-

  mally have in this kind of problem is that the individual act comes to good of the whole, that is, partial evil is universal good. It is not quite true in this case, but nevertheless this is a generic property that often holds in economics.

  But the Solution comes with a lot of fine print. Airlines must know exactly

  their preferences. Not only that, they must know the preferences of all other

  airlines. Further they must know that every other airline accurately knows

  the preferences of every other airline. They also must know that every airline knows that every airline knows the preferences of every other airline, and so

  on in an infinite regress. Also, each airline must be rational enough to work

  out the solution. Further, each airline must believe that every other airline

  is rational and will use perfect rationality to work out the solution. Further, each airline must know in an infinite regress that every other airline is using this rational way to work out the prob
lem, because if one of these airlines

  [ 174 ] Complexity and the Economy

  messes up, it messes the solution up for every other airline. Further, the optimal placement of each airline using this backward deduction must be unique.

  If any link of this network of requirements breaks, the solution ceases to exist.

  In the spirit of being in Belgium, my comment on this is: “C’est magnifique,

  mais ce n’est pas la guerre.”2

  This type of multi-agent choice problem is pervasive in economics. So let us

  take this solution approach seriously. What if we are airline number 3 and we

  feel uncertain as to what airline number 17 is going to do? As airline number

  3, we might say: I don’t think the people of airline number 17 are that super

  bright, and I’m not sure whether they are going to solve this problem by this

  rational method. If they don’t work it out in this way then I am not sure what my optimal choice would be as the third chooser in the process. This is sufficient to upset the situation. But worse, airline number 3 may communicate

  its uncertainty to other airlines and they may no longer rely on number 3 or

  number 17. The entire solution is starting to unravel. In fact the Solution as created is a function of airlines’ expectations or predictions of what other airlines are going to do. So the problem is that if I am a representative airline I am trying to figure out what my expectations ought to be—I am trying to predict

  a world that is created by the expectations of myself and everybody else. There is a self-referential loop here. The outcome each airline is trying to predict depends on the predictions it and others might form. In other words, predictions are forming a world those predictions are trying to forecast. Barring some coordinating device, by which an airline can logically determine the predictions of others (such as the tortured solution-reasoning above), there is no logical way it can determine its prediction.

  There is a logical indeterminacy. So in the economy, people are creating a

  world that forms from their predictions, but if they try to form these expec-

  tations in a perfectly logical deductive way, they get into a self-referential loop. There is a logical hole in standard economic thinking. Our forecasts