Copyright
Sharone L Maital.

Economic behavior & social learning online

. (page 1 of 2)
Online LibrarySharone L MaitalEconomic behavior & social learning → online text (page 1 of 2)
Font size
QR-code for this ebook


HD28
.M414






WORKING PAPER
ALFRED P. SLOAN SCHOOL OF MANAGEMENT



ECONOMIC BEHAVIOR & SOCIAL LEARNING
August 1986



Sharone L. Maital
Shlomo Maital
Nava Pollak



WP#1807-86



MASSACHUSETTS

INSTITUTE OF TECHNOLOGY

50 MEMORIAL DRIVE

CAMBRIDGE. MASSACHUSETTS 02139



ECONOMIC BEHAVIOR ^ SOCIAL LEARNING

August 1986

Sharone L. Maital

Shlomo Maital

Nava Pollak WP#1807-86



ECONOMIC BEHAVIOR & SOCIAL LEARNING

Sharone L. Maital Shlomo Maital * Nava Pollak

Staff Psychologist Associate Professor Graduate Student

Faculty of Industrial Engineering &
Haifa School Management,

Psychological Services Technion - Israel Institute of Technology

HAIFA ISRAEL



The nature and evolution of the social
learning approach to human behavior is described and
summarized and its theoretical and empirical
application to economic behavior in general, and to
intertemporal choice in particular, is summarized. The
actual and potential contribution of social learning
theory to improving our understanding of economic
choice is discussed. By way of example, results from a
longitudinal panel study that began in 1975, focusing
on delay of gratification and its relation to
educational achievement, are described.



VohXhcomiiiQ, ktoix McFadijun S HzatkeA McFadf/eii, e.ditou,
ECOUOmC PSVCHOLOGV, Hofitii Holland, AmteAdam, 19S6



* currently Visiting Professor, Management of Technology Program,
Sloan School of Management, M.I.T.



M.I.T. LIBRARIES
APR 1 6 1987

RECEIVED



"There is no more powerful, pervasive influence on how
individuals think and cultures interact than our different
perspectives on time — the way we mentally partition time
into past, present and future. Every child learns a time
perspective appropriate to the values and needs of his
society."

-Alexander Gonzalez & Philip G. Zimbardo (1985)



INTRODDCTION

Social learning theory posits that the choices people make
depend on the subjective value of each choice, the expectancy that it
will indeed occur, and the ability to implement the choice decision;
emphasis is placed on the interaction of the nature of the choice
situation, how people perceive it, and what they know and have learned
about it and similar choices in the past. We contend that the social
learning model is a theoretically sensible and empirically viable
paradigm of economic behavior, being an integration of dynamic and
cognitive approaches (which interpret behavior mainly as a function of
individual nature and internal thought processes) and the stimulus-
response approach, which sees behavior as solely determined by the
consequences of actions. The purpose of this paper is to summarize the
social learning view of behavior, as it has evolved since its
inception in 1954, and to apply this model empirically to a
particularly important type of economic behavior: the choice of
immediate present rewards versus delayed-but-larger future ones.

The structure of this paper is as follows. The first section
defines social learning theory (SLT) as first conceived by Julian
Rotter in 1954 and expanded upon by Bandura (1977) , and discusses how
it has evolved to the present. Emphasis is placed on the SLT view of
motivation: WHY people act as they do. Section Two discusses how SLT
can explain important aspects of economic behavior in general, and in
particular, how SLT illuminates the economic theory of intertemporal
choice. Section Three describes a longitudinal panel study of time
preference (willingness to defer gratification) among a group of
Israeli teenagers, built on the social learning paradigm, aimed at
discovering the psychological and socio-economic determinants of
economic and educational achievements. The last section summarizes
and concludes.

THE NATURE AND EVOLUTION OF SOCIAL LEARNING THEORY

Social learning theory was first fully enunciated in Julian
Rotter's book Social Learning & Clinical Psychology (1954). The
objective. Rotter later wrote, was "to integrate two diverse but
significant trends in American psychology — the "S-R" or



n



reinforcement" theories, on the one hand, and the "cognitive" or
field" theories on the other" (Rotter, Chance & Phares, 1972, p. 2).



The S-R (stimulus-response) model asserts the preeminence of the
environment in determining behavior: "A person does not act upon the
world, the world acts upon him" (Skinner, 1971, p. 211). In contrast,
humanist, existentialist and cognitive psychologists emphasize that
individuals have the capacity for purposive action and conscious
judgment, sufficient to act upon the world.

Somewhere along a continuum between these two poles lies the
"reciprocal determinism" (Bandura, 1978, p. 345) of social learning
theory, which asserts that complex interactions exist among behavior,
cognition, and the external environment, and that behavior can best be
understood in terms of these interactions. Rotter's fundamental SLT
equation states that behavior potential (BP) , the potentiality of any
behavior's occuring in a given situation, is a function of expectancy
(E) , the probability held by the individual that a particular
reinforcemment will occur as a result of a specific behavior on his
part, and reinforcement value (RV) , the degree of a person's
preference for that reinforcement if the possibilities of occurrence
of all alternatives were equal:

[1] BP = F(E, RV)

Rotter emphasizes that subjective reinforcement value is
systematically independent of expectancy.

To economists, equation [1] is instantly recognizable as a close
cousin to the definition of expected utility (the product of the
probability of an outcome and that outcome's utility, summed over all
possible outcomes) . The expected utility [EO] approach, pioneered by
von Neuman and Morgenstern (1944), remains the bricks-and-mortar of
the economic theory of behavior under uncertainty. However, SLT
(unlike EU) emphasizes the subjective nature of expectancy:

A variety of other factors [in addition to people's actuarial
experience with the event in the past] operate in specific
instances to influence one's probability estimates. Such factors
may include the nature or the categorization of a situation,
patterning and sequential considerations, uniqueness of events,
generalization, and the perception of causality. (Rotter, 1972,
p. 13).

In addition, SLT emphasizes the crucial importance of past learning
experience, which shapes both reinforcement values and expectancy.

Rotter emphasized that "the psychological situation is an
extremely important determinant of behavior ... in sharp contrast to
those positions that .. .assert that once the basic elements of
personality are identified, reliable prediction follows" (Rotter et
al, 1972, p. 37). Experiences with one particular situation,
particularly those involving social interaction with other people, are
often generalized to other, similar situations. SLT calls this
"generalized expectancies", which Rotter terms the basis for social



attitudes. The distinction between specific and generalized
expectancies is an important one in SLT. Specific expectancies result
from one's previous experiences with reinforcement in the same
situation. This expectancy is subject to change and manipulation.
Given a sufficient number of replications, changes in specific
situational expectancies can lead to changes in generalized
expectancies, which are part of one's personality. A major role of
socialization is the development of generalized expectancies.

According to SLT, behavior is in part learned because both

expectancies and reinforcement values are influenced by past

experiences. Behavior is in part socially determined because both

expectancies and reinforcement values are shaped by relations with
other persons and groups.

Rotter led efforts to develop empirical measures of "generalized
expectancies". His scale for the measurement of interpersonal trust
(Rotter, 1967), defined as "the expectancy held by an individual or a
group that the word, promise, verbal or written statement of another
individual or group can be relied upon", was found to have predictive
value for a wide range of behaviors. Another personality variable
with roots in social learning theory is "internal-external locus of
control", "the degree to which [people] accept personal responsibility
for what happens to them, in contrast to the attribution of
responsibility to forces outside their control (Battle & Rotter,
1963). I-E Locus of Control has been used in over a thousand studies
of behavior; see, for example, early studies by Liverant and Scodel
(1960) and Gore and Rotter (1963).

Theories, like people, change and evolve as they mature, and SLT
is no exception. Different adherents have chosen to emphasize
different aspects of SLT. One of SLT's most important contributions
to methods of psychological intervention has been its focus on
"observational learning" (a process whereby one learns new behaviors
simply by observing and imitating another person or "model") (Bandura
and Mischel, 1965). Modelling, which is most often an incidental
process explaining many social behaviors learned in the course of
socialization, has been used to alter behavior constructively and
permanently. More recently, Albert Bandura (1977, 1982) has stressed
the importance of perceived self-efficacy ("judgments of how well one
can execute courses of action required to deal with prospective
situations", Bandura, 1982, p. 122) as a key part of social learning
theory. An individual's feeling of capability plays an important role
in how we perceive reality, in motivating our actions and in how well
we carry them out, Bandura has argued. The important role of self-
efficacy in the social learning theory view of motivation will be
discussed later.

The two leading current proponents of social learning theory,
Albert Bandura and Walter Mischel, have each stressed an
"interactionist" position. Mischel asserts that "we continuously
influence the 'situations' of our lives as well as being affected by
them in a mutual, organic interaction. . .such interactions reflect not
only our reactions to conditions but also our active selection and
modification of conditions through our own choices. . .people



continuously select, change and generate conditions just as much as
they are affected by them." (Mischel, 1984, p. 350). Bandura has
stated this position even more forcefully, by defining what he terms
■reciprocal determinism": "a process [in which] ...behavior, internal
personal factors and environmental influences all operate as
interlocking determinants of each other... [in a] triadic reciprocal
interaction rather than a dyadic conjoint or a dyadic bidirectinal
one" (Bandura, 1982, p. 346. (See Table 1).



Model

1. Pure cognitive
and dynamic



2. Pure stimulus-
response



3. Unidirectional
interaction



4. Bidirectional
Interactionist



Equation
B = F(P)



B = F(E)
B = F(P,E)



B = F(P,E)
P=f(E), E=g(P)



5. Social learning B = F(P,E),

Interactionist P = G(B,E),
(reciprocal determinism) E = H(B,P)



Comment

Behavior is a function solely
of cognitive and other
internal events that affect
perceptions and actions.



Behavior is a function solely
of the external environment

Behavior is a function of both
personal and environmental
variables, where P and E are
independent of one another



Like the eclectic model, but
with P and E dependent on one
another

Internal personal factors,
environmental conditions and
behavior operate as mutually
reciprocal determinants of
one another



Table 1. Five Models of Behavior. Recent developments in social
learning theory have emphasized behavior itself as a variable that
influences personal factors and environment, as well as being
influenced by them. (Based on Bandura, 1978).



One of the standard definitions of economics, due to Frank Knight
(1933), is: the science that examines five basic questions: "what to
produce, where and how to produce it, who should get the output, and
when resources should be used." The question glaringly absent is the
"why" — why do we produce, or engage in any particular economic
behavior. The proximate answer of microeconomics to the "why" is: in
order to maximize "utility". The failure of the utility-maximization
postulate, and of the utility-function black box, to capture any



element of reality has been clear for many years.
Social learning theory in general, and self-efficacy in particular,
provide a useful and sensible motivational framework for economics.
What people do, according to SLT, depends on who they are, what their
tastes and abilities are, what they have experienced in their lives,
by whom they are influenced, to which groups they belong, and the
particular choices they face. According to SLT, "performance
differences in any situation depend on differences in expectancies,
...in the subjective values of the expected outcomes. .. [and in]
differences in the self-regulatory systems and plans that each
individual brings to the situation." (Mischel, 1984, p. 346). The
ability (or inability) of individuals to actually carry out actions
previously decided upon is often a crucial determinant of behavior.
With notable exceptions (Shefrin and Thaler, 19 ), economic theory
has not taken into account the possibility of incongruity between
desire and deed (Maital, 1986), an incongruity which SLT (and of
course many other psychological theories) incorporates in its paradigm
almost from the outset.

Motivation rests in part on whether individuals believe
themselves able to do what they want to do. As Bandura has noted:

In their daily lives people continuously make decisions
about what courses of action to pursue and how long to
continue those they have undertaken. Self-efficacy
judgments, whether accurate or faulty, influence choice of
activities and environmental settings. People avoid
activities that they believe exceed their coping
capabilities, but they undertake and perform assuredly those
that they judge themselves capable of managing. Judgments
of self-efficacy also determine how much effort people will
expend and how long they will persist in the face of
obstacles or aversive experiences. (Bandura, 1982, p. 123).

Perceived self-efficacy, measured empirically, has been able to
account for a wide range of behaviors, such as career and occupational
choice, success in reducing dependence on drugs and alcohol, reactions
to stress and even physical stamina.

SOCIAL LEARNING APPROACHES TO ECONOMIC BEHAVIOR

Economic choice falls into two categories: intratemporal choice
among a set of alternatives at a given point in time, and
intertemporal choice, among different alternatives available now and
in the future. Maital (1986) has argued that virtually all economic
choice is intertemporal in nature, because in addition to choosing
which good or service to consume, we almost invariably decide at what
rate to consume it — an inherently intertemporal problem.

We shall argue that social learning theory provides a powerful
tool for both broadening and deepening the economic theory of present-
future choice. This section provides an SLT-inspired model of inter-
temporal choice and cites empirical evidence in support of the model.
The following section describes preliminary results from our own
longitudinal study.



Consume now, or save and invest, in order to consume more later?
This is a fundamental economic question faced by individuals,
families, corporations, and entire nations. Individuals and families
build human and physical capital by saving part of their current
income and using it to accumulate claims on future consumption.
Corporations grow by retaining profits, instead of distributing them
to shareholders, and reinvesting those profits in enlarging the
productive capacity of the company. The rate at which the actual and
potential output of a nation grows depends on the intertemporal
choices of its individuals, families and corporations.

Economic theory expresses individual preferences for immediate
utility, as opposed to utility deferred in time, as an interest rate;
the higher the interest rate, the greater the premium attached to
immediate gratification. In general, it is assumed that these
interest rates, or "subjective rates of time preference", are
equalized across individuals by borrowing and lending in capital
markets. The argument is one based on arbitrage. If Smith has a
subjective interest rate of 10%, and Jones, 5%, it pays Jones to lend
to Smith at, say 8% interest. For Smith, this interest rate provides
current consumption at a cost (of reduced future consumption) less
than he would be willing to pay; and for Jones, this interest rate
provides an incentive to forego current consumption greater than the
minimum one necessary. In the course of Jones lending to Smith,
Smith's current consumption increases, and as it does, the marginal
value to Smith of an added dollar of current consumption declines; as
a result. Smith's subjective interest rate declines. At the same
time, as Jones' current consumption declines (in order to lend to
Smith) , the marginal value to Jones of an added dollar of current
consumption increases, and with it, Jones' interest rate. This
process will continue (because it is mutually beneficial to both
borrower and lender) until Smith's and Jones' interest rates are
equal. In reality, there are financial intermediaries (banks,
brokers, etc.) who mediate the process — but in principal, the result
is the same: subjective interest rates come into general equality.

Social learning theory, in contrast, posits that: subjective
time preference (or "willingness to defer gratification", as it is
known in the SLT literature) , is learned in the course of
socialization (especially from parents and peers) beginning at very
early ages; is influenced by such general expectancies as
interpersonal trust; becomes relatively stable in the early teens;
differs widely across persons, even (and especially) after it has
stabilized in late childhood ; may be situationally specific; and is a
two-part process, consisting of first the learning of the 'delay'
value or preference, and second and separately, the acquisition of the
ability to implement the chosen preference (Mischel, 1974, 1986;
Mischel and Ebbesen, 1970; Mischel, Ebbesen and Zeiss, 1972; Mischel
and Moore, 1973) .

Economic theory accepts ex ante differences in time preference,
and posits that ex post differences are smoothed out by 'buying' and
'selling' current consumption in free markets. In this view, saving
behavior can be explained by differences in time preference, up to the



point where such differences disappear. Social learning theory argues
that differences in time preference persist throughout life, and are
therefore important personal variables that help explain many types of
economic behavior.

Delay of gratification [DG] has abeen studied extensively as a
testable reward choice situation, in exploring and verifying SLT. All
of this research, especially its empirical content, is relevant to the
economic theory of intertemporal choice, asking as it does how time
preference is learned, how it changes with maturity, how it differs
across individuals, and how it can be changed or influenced. A summary
of the main findings is given in Table 2.



Table 2. Correlates of Preference for Deferred Rewards



Variable

1. social responsibility

2. tendency to cheat

3. delinquency

4. ego strength

5. father absence

6. socioeconomic status

7. IQ



8. school adjustment
and achievement



9. n-achievement

10. waiting-time
activity



11. trust



12. role models



Direction

positive

negative

negative

positive

negative

positive

positive

positive

positive

positive
positive

positive



Source
Mischel (1961a, b)
Lanese (1967)
Mischel (1961a)
Shybut (1970)
Mischel (1958)

Walls & Smith

(1970)
Mischel & Metzner

(1962)



Mendell (1967),
Davids & Sidman
(1962)

Mischel (1961b)



Shack & Massari

(1973)
Mahrer (1956),
Glass (1969),
Mischel & Staub
(1965)

Bandura & Mischel
(1965)



The view that each individual carries a distinctive subjective
rate of interest, as different from that of others as his fingerprint,
is a simple yet powerful idea that can illuminate many types of
behavior:

o why some persons invest dollars and years in education, while

others do not;
o why some people buy expensive, long-lasting durable goods,

while others buy cheaper, shorter-lived products;
o why some persons commit economic crimes, even when certain

they will ultimately be caught and punished,
o why entire nations set aside large portions of their national

income for investment and lending to other countries, while

other nations prefer to consume almost all of their income.

Fuchs (1980) conducted an empirical study designed "to
measure differences in time preference across individuals and to test
for relationships between time preference, health behaviors and
schooling". Of the 500 adults aged 25-64 surveyed by telephone, two-
thirds gave consistent responses to a series of six questions asking
the respondent to choose between a sum of money now and a larger sum
at a specific point in the future. Fuchs found that the implicit
interest rate revealed in their replies is weakly correlated with
years of schooling (negative) , cigarette smoking (positive) and health
status (negative) , thus confirming that present orientation is related
to behaviors that emphasize current (as opposed to future)
gratification, in education and health. He also found that family
background, especially religion, was an important determinant of time
preference.

We now turn to a longitudinal panel study of willingness to defer
gratification, as an illustration of how social learning theory can
serve as a guide for behavioral study of intertemporal choice.

A LONGITUDINAL EMPIRICAL STUDY OF TIME PREFERENCE

An overview: In 1975 Sharone L. Maital (1976) investigated a
group of 142 Grade 8 children, aged 13-14, in four different Israeli
schools. The main objective of the study was to test Mischel's two-
part process theory of delay of gratification, among children older
than those studied in previous research, and to determine whether the
'delay preference' and 'actual waiting' stages are empirically
distinguishable. Extensive socioeconomic, educational and
psychometric data were collected for each child, along with both
behavioral and stated time preference measures.

A decade later, efforts were made to locate as many of the
original 142 participants as possible and reinterview them, after they
had completed their high school educations and military service, and
were about to marry, take jobs, and begin careers, at ages 23-24. We
felt that even if the reinterviewed group was small, it would permit
us to test many interesting hypotheses about the antecedents of time
preference, its stability over time, its relation with educational,
social and economic achievements, and in general about the empirical



link between socioeconomic and psychological variables measured in
early adolescence and the same variables measured ten years later, in
early adulthood. A total of 41 persons from the original group were
located and interviewed.

What follows is a description of some of the main empirical
results, including those from the earlier 1975 study, none of which
have been previously published. The design of the study, and the
theoretical constructs used in it, were framed according to social
learning theory. The overall aim of the study is better described by
a question, rather than a clear-cut hypothesis, a question of interest
to economics and psychologists alike:

* Which personal and social characteristics of young adolescents
are most helpful in predicting their later educational, social and
economic behaviors?

In the study, social learning theory is directly applied to the

understanding and prediction of intertemporal choice, in a manner that

neither denies the importance of economic and social variables nor
asserts their preeminence over personal variables.

STUDY 1 (1975)

METHOD

Subjects: A total of 142 grade eight students from five classes in
four elementary schools in a suburban Tel Aviv district were tested
during the second trimester of the school year: 73 girls and 69 boys.

Design and procedure: The analysis of delay of gratification
behavior was based on questionnaire responses, plus a behavioral test
for the willingness to wait for a deferred reward. The questionnaires
measured preference for larger-but-delayed rewards, interpersonal


1

Online LibrarySharone L MaitalEconomic behavior & social learning → online text (page 1 of 2)