----------------- HES POSTING -----------------
[Posted on behalf of Mohammed Gani. RBE]
RVW-Review of Rational Choice Theory (RCT)
Professor Tollison's review of Amartya Sen's _Rationality and Freedom_
clarifies the crucial issues regarding the use of Rational Choice Theory
(RCT). May I add a few 'constructive' comments to his review.
Tollison: "Second, many of the criticisms of RCT by Sen and others are equivalent to
letting the best become the enemy of the good. No one can reasonably argue that RCT is a
convincing theory of human behavior at all times and under all circumstances. It cannot
explain why the soldier throws himself on the grenade to save his compatriots. Maybe it
will in the future, but this is not the point. The point is that it provides a reasonable
explanation of choices in the vast majority of cases of ordinary behavior, and ordinary
behavior in market and non-market settings arguably constitutes the bulk of human
existence. Given this empirical regularity, we depart from the use of the model at our own
analytical peril. Why discard a useful theory because it does not explain everything, but
almost everything? And needless to say, with what do we replace homo economicus? Homo
Boobus?"
Gani: I believe that a sharp distinction between economic, political and cultural
rationality may help make RCT a general theory under which Economic Choice, Political
Choice and Cultural Choice would be special cases. Please allow me to attempt a brief
exposition of some concepts that may help. (A) Let an economic need be that of procuring
self-sustenance, a political need be that of assuring security (against enemies from
inside and outside) and a cultural need be that of assuring self-reproduction, both
biological and spiritual. (B) Let the actions devoted to meet the need be social in the
sense that the individual cannot meet the needs without the participation of others. A
right is
created when the others permit certain manners of meeting the need, but
prohibit certain other manners. Thus trading for profit is allowed, but
plunder and fraud are not allowed. Collective self-defense by a politically
organized group allows attacking enemies, but not the members of the group.
Again, one may make love to one's spouse, but not to the spouse of another.
(C) Let an institution be a set of rules that define agreed manners of
meeting the needs. The rules may include the creation and operation of an
enforcement agency or system, be it custom or law. (D) While the terms
(need, right, and institution) are empirical entities, the scientist
requires analytical concepts to explain the actions. In this regard, a
general notion of rational choice is built on the concept of motive. A
motive creates a motion (or manner of action with clear sense of direction)
which can be analytically expressed. The economic motive is that of profit
or gain. Economic analysis must be able to distinguish between what is
profitable and what is not and decide that the profitable direction is
taken. The political motive is that of protection (of one's self and
group). The cultural motive is acceptance. The economic action is based on
self-interest, while the political action may involve sacrifice. Political
action is chosen according to whether the recipient of action is an enemy
or friend. (Hopefully, someone will show how power to protect is
acquired/used in politics.) Cultural action distinguishes between good and
bad, howsoever defined. One might suppose that the good increases
acceptance, while the bad decreases it and the cultural motive is to
increase acceptance.(E) Analysts develop tools to show how economic,
political and cultural rationality may explain economic, political and
cultural actions in terms of their respective motives. While economists
have developed optimization subject to constraints to show how economic
rationality explains allocation, political science and anthropology have
yet to do the same for!
political and cultural actions. AS far as I know, economists have yet to develop the
concept of rationality suitable to explain exchange, as shown below.
Tollison: "Third, the charge from Sen and others that economics is about narrow self-
interest is entirely misplaced. When I was taught RCT as a graduate student, it was made
clear from the start that economics is silent on what people want. Rather, given what they
want, individuals pursue their goals efficiently. Hence, there is room for all types of
behavior in RCT. Individuals may be sinners or saints, angels or alligators, but self-
interest, scarcity, and the law of demand are the best predictors of their actions. RCT is
a big tent whose scope has expanded radically over the last half century, and it continues
to expand each day as scholars continue to find new areas of application. Narrow self-
interest is a canard that has unfortunately captured the interest of a few otherwise
thoughtful economists. And the fact that practitioners in other disciplines do not
understand RCT except as a caricature of narrow, greedy behavior should come as no
surprise; in some cases RCT threatens to overtake their disciplines and devalue their
human capital."
Gani: There is some danger that political and cultural actions, which do not abide by
economic motives, may be misread by economists.
Tollison: Fourth, I have not done the basic research on the issue, but I wonder what the
scorecard is in the anomalies business? My impression is that practically every anomaly
has been met with empirical evidence suggesting that RCT is the preferred explanation of
the postulated choice outcome or experimental evidence. In any event, it would be a useful
exercise to catalogue every proposed anomaly and what the evidence shows in each case. We
need to know if the set of verifiable anomalies is an empty economic box. And if an issue
is unresolved, this can be a separate category on the scorecard.
Gani: I fully agree that RCT has no viable alternative. But there are some serious
shortcomings of RCT as it stands now. (The so-called anomalies that arise when one tries
to explain non-economic action with economic rationality are not proper anomalies, because
economics is not even supposed to explain those actions.) Despite great many attempts,
social choice in economic actions remains a mystery. Here is my attempt to crack it.
(1)There are just two basic economic actions. Allocation is a strictly individual action
based on strictly individual choice while exchange is a social action based on social
choice. (2)In an allocation, an individual has full control over the use of something that
already belongs to him. Rational choice in that case involves what we may call linear
consistency of preference. An individual's preferences over the set of options can be
arranged in a transitive line so that if good x is preferred to good y and if y is
preferred to good z, than x must be preferred to z. The individual chooses the most
preferred point available within the budget or endowment. But in an exchange, one
individual owns something, but prefers another good belonging to somebody else. The
problem is to enable the individual to get what belongs to somebody else in exchange for
what belongs to the agent. Thus suppose that there are three agents (A, B, C) who
respectively own three goods (x, y, z). Suppose that A prefers y to x and x to z,(and
hence of course y to z). B prefers z to y and y to x, (and hence z to x). Lastly, C
prefers x to z, and z to y, and (hence x to y). Each agent's preference has linear
consistency. But the agents together have no linearly consistent preference order. In a
situation just like this, Arrow finds that it is impossible to formulate a social welfare
function or to discern social choice because of the linear inconsistency of preferences of
different individuals.
My answer is shocking. There is no need for interpersonal preference orders
to have any linear consistency. Exchange is possible if and only if linear
inconsistency is present. In the cited example, A's good x goes to C, whose
good z goes to B, whose good y goes to A, giving each agent the most
preferred good and thereby maximizing social welfare. The social choice is
also very clear: give to each agent the good most preferred by that agent.
The key thing is that over the exact same pair of goods, the buyer and the
seller must have precisely opposite direction of preference (otherwise
called double coincidence). Thus Adam may prefer 3 bananas that belong to
Beth to 2 apples that belong to him. Exchange is possible if and only if
Beth prefers 2 apples to 3 bananas. Thus compared to 3 bananas, 2 apples
are more preferred by its prospective buyer Beth and less preferred by its
prospective seller Adam. It leaves gains for both. If both agents preferred
apples to bananas, then nobody would sell apples and nobody would buy
bananas.
I believe that Arrow's impossibility theorem does not set up the social
choice problem correctly at all: because it is neither an allocation
problem nor an exchange problem. It is not a problem for society to choose
only one good when different people prefer different goods. Had it been an
allocation problem, one would specify the endowments of the agents and the
simple solution would be that each agent would use the endowment to produce
whatever good the agent preferred most and all three goods would be
produced according to the distribution of endowment. In an exchange
problem, if and only if circularly consistent preferences existed would it
be possible to arrange exchanges that would be mutually gainful. In neither
case would there be any need to chose only one good out of the three.
By the way, I think that Arrow's impossibility theorem is also inapplicable
to political choice. If a society has different voting blocks with
preference for different candidates, the solution is a parliament with many
leaders for many voting blocks, and not a single leader. If there is room
for only one leader, the problem is for the leader to learn how to attract
majority support, and not for the voters to choose from among several
morons. If no candidate wins clear majority, another election is required.
Tollison: Fifth, modern economic theory does not assume that all actors are equally
rational (whatever this may mean). People pursue their self-interests with different
degrees of skill, either naturally acquired or learned. It has always seemed to me that
this point is not fully appreciated by the critics of economics or by economists
themselves as they go about their business of explaining human behavior. And I am not
talking about bounded rationality, but rather innate differences in the ability of
individuals to pursue their goals effectively. Many anomalies seem to derive from this
basis or at least from the absence of repeated play formats wherein such differences would
be mitigated through learning.
Gani: I do plead guilty for not appreciating unequal rationality. My trouble is that I see
economics as a study of the kinds and quantities of traded goods and not as a study of
human behavior. I do use human motive to explain the behavior of goods (but not the
actions of humans), since the kinds and quantities are chosen by humans. But to me
rationality is an analytical idea, and not an empirical one. I just do not see how one can
be more or less rational than another. To me, rationality is simply the idea that whatever
people actually do, they do it because they prefer what they do to what they do not do but
could have done. I will never understand if a human does what he does not want to do, with
the clear freedom not to do it.
Mohammad Gani
------------ FOOTER TO HES POSTING ------------
For information, send the message "info HES" to [log in to unmask]
|
