Mind, Reciprocity and Markets in the Laboratory
Vernon L. Smith
Interdisciplinary Center for Economic Science And Mercatus Center
George Mason University
Fruitful social science must be very largely a study of what is not. Hayek (1983)
The deep structure of human behavior falls into three interdependent categories: the internal order
of mind; the external order of social exchange; and the extended order of markets. Each forms a
complex self ordering system governed by endogenous rules reflecting individual and species
experience, and biological /cultural evolution.
Most of our operating knowledge we do not remember learning because the learning was
autonomic, imitative and formed by practice. By age four we were accomplished in a native
language, capable of algorithmic inflection and syntactical construction. For English we added ed to
convert verbs into their past tense form and s to obtain the plural of nouns, with some of the
irregular exceptions yet to be memorized. Such learning required no instructions from mom or
teachers; only listening, to initialize the circuitry with which we were born. Our vision circuitry had
already learned to interpret flat retinal signals in perspective, so that we experienced the visual
world as three dimensional, making us competent navigators, but necessarily vulnerable to optical
illusions. Between age four and five (if unimpaired by autism) we learned to ‘mind read,’ to infer
what someone must believe or think based on their actions or words, that a playmate will believe
falsely that a toy is in the box if she did not see it being removed. Subsequently, on a flexible
timetable, we learned ever more sophisticated social exchange behavior, enabling us to enjoy
increasing gains from this trade, with the exception of the universal cross-cultural 3-4% of people
who are socio-paths and who account for a large percentage of prison populations.
In 1952 Hayek, perhaps the 20th century’s most versatile and deep intellect, articulated a model of
perception subsequently corroborated by neurobiology, that experience is not formed, as we might
think, by our receipt of signals reflecting fixed attributes of external phenomena. Rather, perception
involves an interaction between current stimuli, and our past experience of similar conditions.
Mental categories are formed dynamically out of the relative frequency of concurrence between
memory and current experience. All perception is memory: what is stored are external stimuli
modified by processing systems conditioned by past experiences.
A peculiarity of mind is its penchant for anthropomorphizing the order it sees in society. If human
structures are well designed someone in the distant past must have designed them. Individuals,
committees and legislatures make rules and laws, and the mind sees designers at work. But what
is unseen is the process that ignores, alters in use, and/or substitutes emergent norms for portions
of the rules, a process that no one monitors because such detailed histories are not recorded. If
something is functioning, the mind sees it as an invention of a comprehending mind, not the
cumulative state of countless trial-and-error modifications of countless myopic minds. Contrary to
brain research findings, the mind always believes it is in control, leading the neuroscientist,
Gazzinega, to wonder why the brain fools the mind into thinking it is in control.
That the brain learns and accomplishes so much without resort to self-conscious direction is
economical in that the brain decentralizes most of its activity to off-line processes, thereby
economizing on scarce attending resources that are reserved for intervention in response to the
demand for novelty, creativity and reasoning. Thus standard economic theory as learned and
taught is based on reasoning; it derives rational constructivist models of interactive decision
ranging from two person bargaining games to many person markets. In such analyses the models
require information not given to any one mind in the worlds they represent; individual payoffs are
common knowledge, and also that everyone is self-interested and competent to analyze choice
consequences. The assumed conditions on interactive structure (the institution) and the payoff
environment are easily reproduced in the experimental laboratory, with subject choices motivated
by real money. What do we learn of human social and market behavior from these studies?
In two person games, people exhibit patterns of trust and trustworthiness, even when matched
anonymously, that enable many to achieve payoff levels in excess of the narrowly self-interested
outcomes predicted by theory, a super rational result of our biological and cultural adaptation. For
example, in each session of a trust game experiment, 12 people are randomly and anonymously
assigned to six pairs and to the role of player 1 (P1) or 2 (P2). Each pair faces the following move
consequences: P1 can choose an equal split of $20 between P1, P2, or pass to P2, in which case
the prize increases to $30; then P2 chooses between the alternatives $12 for P1 and $18 for P2, or
$5 for P1 and $25 for P2. Theory predicts that since P1 will expect that P2 will go for the $25, P1
will choose not to pass to P2. Hence, the equilibrium outcome is $10 for each, wastefully leaving
$10 out of the potential $30 prize, on the table. Many such experiments, including some that vary
the payoff distribution within this trust structure (and the stakes are varied up to $50), yield the
following results: about half of the P1s pass to P2, and two-thirds to three-quarters of the P2s
respond with the cooperative outcome, ($12, $18) in the above example. The rest defect, yielding
($5, $25). In punishment versions of the game, if P2 defects, play returns to P1 who at a personal
cost can punish P2 with a lower payoff, e.g. in the above example P1 now chooses between
defection ($5, $25), and (0, 0). Contrary to theoretical predictions, up to half of the defectors get
punished in a single play of games of this type.
Why so much cooperation and the use of costly punishment options in single play games? Many
convergent studies suggest that it is reciprocity: when P1 passes to P2, it enables both to increase
their reward, affording P2 the opportunity to reciprocate the favor by choosing to cooperate.
Humans are ceaseless social exchangers, always trading favors, gifts and services among
associates. Bilateral exchanges of food for help are observed even among those in chimp and
monkey communities. In humans this imperative is so strong that it survives in laboratory
experiments where people will never know with whom they are interacting.
Subjects from the same population pools used in the above experiments have also participated in
hundreds of laboratory bidding, auctioning and posted pricing markets with well-defined property
rights. These market experiments show that human subjects behave very self-interestedly, tending
to maximize the gains from trade in environments where information on value and cost is dispersed
privately among the traders. This research illuminates the findings of Adam Smith and F. Hayek
that markets enable thousands even millions of people to specialize, and create wealth by
cooperating through self ordering pricing systems without any one having to tell them what to do.
Remarkably, each of us functions in two worlds: one of personal exchange governed by self-policed
norms of reciprocity (positive and negative) in which there is much intentional cooperation
generating gains from such trades; another of impersonal exchange through markets governed by
constantly evolving cultural rules invented by no one person, in which, unintentionally, and without
awareness, we cooperate by pursuing our own self defined interests. Each world is a complex self-ordering
system to which we, our parents, our parents’ parents, etc have adapted and contributed
to its evolution.
No doubt market exchange evolved out of social exchange, the latter enabling specialization long
before this evolution would bring forth monetary exchange. As noted by Hayek these two worlds
are governed by distinctive rule systems. The rules of solidarity (read ‘reciprocity’) based on
intentional actions to do good, if used aggressively to intervene in the market system to ‘improve’
its outcomes, will destroy it. But equally important, if the self-interested motivations in markets are
clumsily applied to families and small groups, it will crush them.
The convergent principles discussed above have emerged from theorizing, experimental testing,
neuroscience, anthropology, psychology and philosophy. They constitute a powerful understanding
of a world that has given us abundant life and enabled us to multiply and to replenish the earth to
an extent beyond all comprehension by our ancestors, and potentially to bequeath untold fortune to
our descendents.
English version of German translation appearing in Wirtschaft, 10, August 2001, p 21. |
